Worker Participation: Current Research and Future Trends, Volume 16 (Research in the Sociology of Work)

WORKER PARTICIPATION: CURRENT RESEARCH AND FUTURE TRENDS i RESEARCH IN THE SOCIOLOGY OF WORK Series Editor: Randy Ho...

Author: Vicki Smith | ed.

57 downloads 1062 Views 2MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

WORKER PARTICIPATION: CURRENT RESEARCH AND FUTURE TRENDS

i

RESEARCH IN THE SOCIOLOGY OF WORK Series Editor: Randy Hodson http://www.sociology.ohio-state.edu/work Volume 1: Volume 2:

Class Consciousness Peripheral Workers

Volume Volume Volume Volume

3: 4: 5: 6:

Unemployment High Tech Work The Meaning of Work The Globalization of Work

Volume 7: Volume 8: Volume 9:

Work and Family Deviance in the Workplace Marginal Employment

Volume 10: Volume 11: Volume 12:

Transformation of Work Labor Revitalization: Global Perspectives and New Initiatives The Sociology of Job Training

Volume 13: Volume 14: Volume 15:

Globalism/Localism at Work Diversity in the Workforce Entrepreneurship

ii

RESEARCH IN THE SOCIOLOGY OF WORK

VOLUME 16

WORKER PARTICIPATION: CURRENT RESEARCH AND FUTURE TRENDS EDITED BY

VICKI SMITH Department of Sociology, University of California, USA

Amsterdam – Boston – Heidelberg – London – New York – Oxford Paris – San Diego – San Francisco – Singapore – Sydney – Tokyo JAI Press is an imprint of Elsevier

iii

JAI Press is an imprint of Elsevier The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA First edition 2006 Copyright r 2006 Elsevier Ltd. All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent veriﬁcation of diagnoses and drug dosages should be made British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN-13: 978-0-7623-1202-3 ISBN-10: 0-7623-1202-5 ISSN: 0277-2833 (Series) For information on all JAI Press publications visit our website at books.elsevier.com Printed and bound in The Netherlands 06 07 08 09 10 10 9 8 7 6 5 4 3 2 1

iv

CONTENTS LIST OF CONTRIBUTORS

ix

INTRODUCTION : WORKER PARTICIPATION : CURRENT RESEARCH AND FUTURE TRENDS Vicki Smith

xi

PART I: WORKER PARTICIPATION IN CONTEMPORARY CORPORATE WORKPLACES THEORIZING TEAMWORK UNDER CONTEMPORARY CAPITALISM Steven Peter Vallas

3

ORGANIZATIONAL NARRATIVES AND THE CONSTRUCTION OF RESISTANCE: UNION ORGANIZING AT A ‘‘PROGRESSIVE’’ COMPANY Joshua L. Carreiro

25

THE EFFECT OF PARTICIPATIVE WORK SYSTEMS ON EMPLOYEE EARNINGS Michael J. Handel

55

GENDERED CONTRADICTIONS: MANAGERS AND WOMEN WORKERS IN SELF-MANAGED TEAMS Nancy Plankey Videla

85

v

vi

CONTENTS

YOU GET PAID FOR THAT?: JOB AND ESTABLISHMENT LEVEL VARIATIONS IN THE USE OF INCENTIVE COMPENSATION Jeremy Reynolds

117

SELECTIVE PARTICIPATION: CONTROLLING WORKERS’ INPUT AT GENERAL MOTORS Jeffrey S. Rothstein

151

PART II: RECENT TRENDS IN COOPERATIVES DEMOCRACY AT WORK WITHIN THE MARKET: RECONSIDERING THE POTENTIAL George Cheney

179

WORKPLACE DEMOCRACY COMES OF AGE: ECONOMIC STABILITY, GROWTH, AND WORKFORCE DIVERSITY Joan S.M. Meyers

205

CAN COLLECTIVIST-DEMOCRACY BRING GENDER EQUALITY? THE EFFORTS AT TWIN OAKS Joyce Rothschild and Amy Tomchin

239

DEMOGRAPHIC AND ENVIRONMENTAL INFLUENCES ON THE DIFFUSION OF CHANGES AMONG ISRAELI KIBBUTZIM, 1990–2001 Raymond Russell, Robert Hanneman and Shlomo Getz

263

COOPERATIVE ENTREPRENEURIALISM: RECONCILING DEMOCRATIC VALUES WITH BUSINESS DEMANDS AT A WORKER-OWNED FIRM Joel Schoening

293

Contents

vii

PART III: MEDIATING WORKER PARTICIPATION: THE LABOR MOVEMENT AND THE STATE OPPORTUNITIES AND CHALLENGES FACING NEW WORKFORCE INSTITUTIONS: A CLOSE-UP ANALYSIS OF AN ALTERNATIVE STAFFING SERVICE Esther B. Neuwirth

319

THE SPIRIT OF OUTSOURCING: CORPORATE AND STATE REGULATION OF LABOR UNDER H-1B VISA AND TANF POLICIES IN THE U.S. Frank Ridzi and Payal Banerjee

345

GETTING JOBS AND BUILDING CAREERS: REPRODUCING INEQUALITY IN STATESPONSORED JOB SEARCH ORGANIZATIONS Vicki Smith, Heather Kohler Flynn and Jonathan Isler

375

ABOUT THE AUTHORS

403

This page intentionally left blank

viii

LIST OF CONTRIBUTORS Payal Banerjee

Department of Sociology, Syracuse University, Syracuse, NY, USA

Joshua L. Carreiro

Department of Sociology, University of Massachusetts, Amherst, MA, USA

George Cheney

Department of Communication, The University of Utah, Salt Lake City, UT, USA

Heather Kohler Flynn

Department of Sociology, University of California, Davis, CA, USA

Shlomo Getz

The Institute for Research of the Kibbutz, University of Haifa, Haifa, Israel

Michael J. Handel

Department of Sociology, Northeastern University, Boston, MA, USA

Robert Hanneman

Department of Sociology, University of California, Riverside, CA, USA

Jonathan Isler

Department of Sociology, University of California, Davis, CA, USA

Joan S. M. Meyers

Department of Sociology, University of California, Davis, CA, USA

Esther B. Neuwirth

Care Management Institute, Kaiser Permanente, Oakland, CA, USA

Nancy Plankey Videla

Sociology Department, Texas A & M University, College Station, TX, USA

Jeremy Reynolds

Department of Sociology, University of Georgia, Athens, GA, USA

Frank Ridzi

Department of Sociology, Center for Regional and Urban Applied Research, Le Moyne College, Syracuse, NY, USA

ix

x

LIST OF CONTRIBUTORS

Joyce Rothschild

School of Public and International Affairs, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA

Jeffrey S. Rothstein

Douglas A. Fraser Center for Workplace Issues, Wayne State University, Detroit, MI, USA

Raymond Russell

Department of Sociology, University of California, Riverside, CA, USA

Joel Schoening

Department of Sociology, University of Oregon, Eugene, OR, USA

Vicki Smith

Department of Sociology, University of California, Davis, CA, USA

Amy Tomchin

School of Public and International Affairs, Virginia Polytechnic Institute and State University, Blacksburg, VA, USA

Steven Peter Vallas

Department of Sociology and Anthropology, George Mason University, Fairfax, VA, USA

WORKER PARTICIPATION: CURRENT RESEARCH AND FUTURE TRENDS Worker participation is the coin of the realm in twenty-ﬁrst century workplaces. It is an umbrella concept that encompasses a wide range of sociotechnical arrangements and institutional settings. Analyzed in the pages of sociological, industrial relations, and labor economics journals, and explored in numerous scholarly monographs, worker participation – as concrete organizational practice and as rhetoric – seems to have swept across the spectrum of American workplaces. The verdict on what precisely constitutes worker participation, on speciﬁc causes, consequences, or outcomes is inconclusive. Yet there is no avoiding the fact that worker participation has become a major paradigmatic approach of the late twentieth and early twenty-ﬁrst centuries. This is true even in workplaces that seem devoid of any commitment to workers or in jobs that, on the surface, seem not to require much thought or initiative from workers: in fast-food establishments, low-end retail stores, and the like. Some fast-food restaurants like Burger King have introduced teamwork and empowerment programs (Talwar, 2002, Chapter 6) and even WalMart, the embodiment of contemporary ‘‘low-road’’ labor relations, refers to its employees as ‘‘associates’’ who have ‘‘careers’’ within the company and are members of ‘‘teams.’’ Firms have been known to integrate temporary workers into participation programs, challenging intuitive assumptions that only attached, long-term employees have access to them (Smith, 2001a, Chapter. 4; Smith, 2001b). While it is easy to brush these occurrences off as mere window dressing, they are indicative of a perception on the part of business managers and employers that there is value in at least mimicking the trend of recognizing employees and encouraging their participation. American managers have been experimenting with different ways of ensuring worker participation and effort since the inception of industrial capitalism. Formal worker participation programs per se gained great xi

xii

INTRODUCTION

popularity in the 1970s and 1980s, when corporations began seeking ways to become more competitive, given increasing global competition, new technologies, the rule of Wall Street, and the pressure to develop new products quickly. The core premise of the notion is that when workers are enabled to make more decisions about how they do their work, about how they run their work groups, about the materials they use, and the products they develop they will work more effectively and be more invested in the fate of the ﬁrms they work for. The premise is inherently plausible: democratizing information, empowering workers, and encouraging their involvement can create greater job satisfaction and lead workers to develop a stake in the quality of the goods and services that they produce. Experiments with worker participation became part of the national psyche when, in the 1980s, General Motors (GM) and the United Auto Workers jointly developed a worker-centered system to produce the Saturn (Rubenstein & Kochan, 2001). In another highly visible experiment, GM teamed with Toyota in a joint venture, New United Motor Manufacturing Inc. (NUMMI) (CA), which also incorporated a high-trust, participative system in its plans to produce cars for both companies (Adler, 1993).1 In 1987, the U.S. Congress passed the Malcolm Baldridge National Quality Improvement Act (Public Law 100-107), which each year honors American businesses deemed to exemplify quality manufacturing and service processes. One criterion for receiving the Baldridge Award is how well companies have managed to enlist workers’ efforts in improving quality, further enshrining workers’ primacy in helping ﬁrms compete. By the early twentyﬁrst century, the conﬁdence in worker participation is so strong that it has taken on a ‘‘bandwagon’’ effect, often adopted by ﬁrms even in the absence of deﬁnitive data about its beneﬁts (Staw & Epstein, 2000; Zbaracki, 1998). What is worker participation, exactly? What forms does it take in diverse workplaces and in what conﬁgurations can it be found? The literature identiﬁes an extensive number of labels, arrangements, and techniques that fall under the rubric of worker participation. They include self-managed teams, quality circles, Total Quality Management, ﬂexible practices, Japanese production methods, lean production, employee involvement, just-intime production, co-governance, joint consultation, high-performance workplaces, and employee ownership. The depth of arrangements varies across business organizations, ranging from the shallow (a company uses one arrangement, such as quality circles) to deep (a company may implement multiple arrangements such as quality circles and self-managed teams, and put much effort into manufacturing participative cultures).

Introduction

xiii

OUTCOMES RELATED TO WORKER PARTICIPATION IN CAPITALIST FIRMS The ﬁrst cluster of papers in this volume studies the effect of worker participation on individuals, group processes, and organizations. This topic mirrors the predominant emphasis in the literature wherein worker participation, broadly deﬁned, has been regressed against nearly every conceivable outcome in diverse work settings. Quite reasonably, a driving question for social scientists is what happens when worker participation is introduced. What are the consequences of top-down participation schemes and are they meaningful? Do they change the distribution of rewards and opportunities, or reconﬁgure dynamics between workers? The study of outcomes is significant because it touches on whether worker participation programs genuinely change the nature of work, improve workers’ jobs, strengthen workers’ hand or merely perpetuate traditional power structures. Many have examined the effects of worker participation on a series of discrete outcomes: wages and other forms of compensation, productivity, organizational success, citizenship behaviors, workers’ attitudes, quit rates, job satisfaction and morale, autonomy, workplace safety and accident rates, job reduction, work intensiﬁcation, commitment, psychological and physical health. Handel’s (this volume) and Reynolds’s (this volume) research builds on this tradition, analyzing the effects of worker participation on employee earnings and incentive compensation, respectively. A subsection of the effects literature examines the factors which might explain outcomes: the presence or absence of unions in the workplace, strength or weakness of union leaders’ commitments, the manner in which management implements new programs, communication patterns, technological conditions under which worker participation is adopted, the degree of trust between labor and management, and gender and race status of the workers involved (although the number of studies analyzing the latter are few and far between).2 Several observations can be made about this literature. First, while there is little consensus about whether worker participation programs genuinely improve corporate performance and other organizational outcomes, there is modest consensus that they tend to solicit greater commitment and enthusiasm on the part of workers. Randy Hodson’s oeuvre represents the most thorough exposition of this point (for an excellent summary of his ongoing research on the topic, see Hodson, 2001a). This, it seems to me, is consistent with one of the most durable insights we have from human relations theory: workers respond positively when management pays attention to them as agents with their own needs and interests. Second, outcomes stemming from

xiv

INTRODUCTION

worker participation programs are strongly affected by existing organizational and social-relational conditions. The context in which programs are implemented matters enormously, determining whether worker participation programs will succeed or fail, be met with workers’ cynicism or engagement. As Vallas (this volume) shows, for example, identities that workers develop under more traditional managerial ‘‘regimes’’ can limit the potential for change when team systems are introduced into the workplace. Rothstein (this volume) makes a parallel claim, arguing that the different patterns of industrial relations at two General Motors’ plants significantly inﬂuenced the way managers introduced a new ‘‘global manufacturing system.’’ A third insight one gains from reading this literature is that generalizations about outcomes are difﬁcult because there is considerable variability in the way worker participation is deﬁned and measured. While social scientists may disagree, for example, about whether worker participation schemes have a positive effect on autonomy, it is hard to be conclusive because they may measure different phenomena (for example, measuring perceptions of job autonomy versus measuring actual autonomy on the job). Related to the problem of definition and measurement is the fact that managers and workers themselves may use terms that differ from prevailing definitions held by social scientists (Edwards & Collinson, 2002), which can mislead researchers about what they study. Hackman and Wageman (1995) found that companies that profess commitments to worker participation may have big gaps between what they say they do and their actual practices. Finally, it is important to remember that when researchers make claims for what constitutes the particular outcome, ‘‘organizational success,’’ the very definition of success is a social construct, often arbitrarily formulated by managers and business leaders themselves, then repeated uncritically by researchers.3 Field researchers who study effects are driven by the same questions about outcomes stemming from worker participation but tend to be more preoccupied with the nature of power and hierarchy. The ‘‘rhetoric vs. reality’’ debate has captured the attention of scholars who typically go into workplaces, participate in or closely observe some of the new schemes and practices, interview workers, and analyze corporate documents. In doing so, they attempt to determine whether workers are gaining the competencies and opportunities often hyped in optimistic rhetoric about worker participation, or whether these gains are compromised because workers are simply being asked to do more without gaining substantive power. Important to proponents of participative arrangements is the claim that worker participation programs erode hierarchical managerial control, redistributing power and decision making into the hands of workers. Yet

Introduction

xv

many who study these schemes argue that while the social organization of control indeed has shifted, with more responsibility for workplace activities being overseen collectively by workers, the scope and depth of managerial control has not diminished or disappeared. Rather, they argue, new schemes and rhetorics about empowerment have made it possible for management to exert hierarchical control over workers in novel ways. Concurrently, peerand self-management introduces new, coercive elements of lateral control into workgroups.4 Some critics argue that worker participation schemes lead to greater work intensiﬁcation without delivering the beneﬁts to workers that might accrue from their deepened engagement with their work. Others see more ambiguous effects and feel that worker participation programs have the potential to tip the balance in favor of workers, can introduce new tensions that challenge managerial control, or can afford new opportunities for skill acquisition and mobility.5 The rhetoric vs. reality debate, however (in which I situate my own previous work, Smith, 2001a, 2001b), is limited by dichotomizing research questions: are worker participation programs good or bad for workers; are they empowering or coercive; do they allow for genuine worker control or do new regimes of worker participation obscure deeper managerial control? While I agree with the fundamental concern that workers might lose ground when work innovations are introduced, it seems to me that at this point in time a different tack is needed, that the ‘‘good’’ vs. ‘‘bad,’’ ‘‘reality vs. rhetoric’’ debate is at a stalemate. A wide reading of studies strongly suggests that these arrangements are both good and bad, that they do inject new constraints but simultaneously new opportunities, and that in capitalist ﬁrms management control is rarely relinquished but that it is being modiﬁed. Attempting to characterize organizational innovations such as worker participation as either/or misses the dialectical and complex realities of power, skill, reward, opportunity, and individual agency and subjectivities. Two chapters in Part 1 suggest possible lines of alternate research. As Collom (2003), Edwards and Collinson (2002), and Hales (2000) point out elsewhere, managers have considerable ﬂexibility in the ways they use both the terms and the practices of empowerment and participation. Participation schemes often are ambiguous and open-ended, making it easy for managers to appropriate and deploy them in different ways for their own purposes. Carreiro (this volume) provides a vivid example of how top management of a progressive food chain usurped the language of empowerment and participation to counter and eventually defeat the efforts of workers to organize a union. If managers can selectively appropriate worker participation schemes and rhetorics, do workers do the same and how? Vallas (this

xvi

INTRODUCTION

volume) argues that workers (subaltern groups) can use these schemes to advance their own interests and calls for a ‘‘culturally nuanced’’ account to understand such outcomes. Other research programs seem urgently needed. How do gender, race, and ethnicity shape workers’ experiences with worker participation programs and conversely, how do participative programs reshape relations between diverse groups of workers? Plankey Videla (this volume) argues that managers themselves can thwart new work programs when they tout self-management schemes predicated on the concept of the gender-neutral and independent team member, all the while relegating women to sex-typed positions. Following another research angle: as new institutionalists might predict, a belief in the value of worker participation has diffused throughout the corporate world. What have been the precise institutional, cultural, and legal mechanisms by which this belief has been promulgated? What are the isomorphic processes through which corporations and their human resource managers have transformed their practices? Drawing on Barley and Kunda’s (1992) work on historical ‘‘surges’’ of rational and normative ideologies of control, we might also investigate how worker participation ﬁts into these cycles as well as the possible limits to the ideology and theory of participation in the future.

PARTICIPATION IN WORKER-OWNED ORGANIZATIONS The second cluster of articles in this volume analyzes worker participation in democratic organizations: worker coops, a commune, and kibbutzim. These chapters make a valuable contribution to our contemporary understanding of this organizational population. The social science literature on participation in worker-and-member-owned enterprises is thin in comparison to the number of studies conducted in for-proﬁt ﬁrms. But as we can see in this volume, there is some exciting research being done on cooperatives that have survived into the twenty-ﬁrst century. Whether worker-owned enterprises are productive and profitable, whether they are capable of surviving over the long run, and the nature of power and democracy are central concerns of these papers just as they are central to the research on worker participation in for-proﬁt workplaces. Small alternative, worker-run cooperatives proliferated in the 1960s and 1970s. A generation of social scientists studied these cooperatives up close, conveying a strong sense of the excitement, experimentation, and frustrations

Introduction

xvii

that people experienced when they tried to create alternatives to the rational, bureaucratic organizations in which the overwhelming majority of people worked. Rothschild-Whitt’s (1979) classic article outlined the dimensions of the holistic, substantive rational organization, contrasting it to the bureaucratic organization that atomized people, rationalized their actions, and alienated them from the process and products of their labor. We have compelling accounts from that era of how, to name a few, bread, plywood, trees, groceries, crisis support, and other human services were produced in those cooperatives.6 Some of these early generation co-operatives have survived by staying close to their original democratic goals, while adapting to modern contingencies. Rothschild and Tomchin (this volume) analyze the practices at Twin Oaks, a long-lived residential and producers’ commune that has been committed to gender equality nearly from its inception. Twin Oaks has found organizational and distributive mechanisms for minimizing a traditional sexual division of labor and maximizing women’s and men’s participation in economic production and child rearing. As Meyers (this volume) points out, most of the alternative organizations that were established in the 1960s and 1970s failed overtime. Many of them were weakened by a shifting, hostile political climate, while others were unable to reconcile their democratic and cooperativist principles with the pressure for economic growth. The most often-cited example of a larger, region-wide cooperative survivor – Mondrago`n – has had to centralize and standardize its practices in response to greater economic competition and globalization (Cheney, this volume). The kibbutzim in Israel, some of the oldest cooperative enterprises we know of, have had to innovate to survive and in doing so have become less democratic, less collective, and less egalitarian than they were historically (Russell, Hanneman, & Getz, this volume). One organizational type which seemed to hold hope for democratization at work – Employee Stock Ownership Plans (ESOPs) – has rarely lived up to its promise of enabling substantive worker control (Blasi, 1988). ESOPs, in which workers are the majority of the shareholders of a company, are the most common type of worker-owned ﬁrm in the United States. With passage of federal legislation that created tax advantages for ﬁrms with ESOPs in 1974, this ownership form has come into wider and wider use over the years. Logue and Yates (2001) estimate that in 1998 there were approximately 11,400 ESOP ﬁrms in the U.S., a number that includes much variation in the percentage of workers/owners. ESOPs have often been formed in the wake of economic crisis; workers face the possibility that their

xviii

INTRODUCTION

ﬁrm is going to close down or go bankrupt, band together, and buy it out (see Varano, 1999, for a case study of this process at Weirton Steel in West Virgina). Although some researchers and many community activists held high expectations for ESOPs to increase work participation (Whyte, Hammer, Meek, Nelson, & Stern, 1983), typically, ESOPs have not given workers/owners much voice in the governance of the ﬁrm, whether at the macro level of determining investment decisions or product development, or at the micro level of self-managing in the workplace (Blasi, 1988). Putting these somewhat discouraging trends alongside our profoundly individualistic culture, many people adamantly believe that designing democratic and participative work organizations is impossible. Several of the articles in this section suggest otherwise. They point us in directions that could be fruitfully explored by social scientists as well as utilized by social change activists who continue to struggle for alternatives to life in bureaucratic, privately held corporations. The articles here raise a set of questions that are important for theorizing alternatives to present-day, hierarchically controlled workplaces. In the twenty-ﬁrst century, can worker-owned and managed ﬁrms be as productive and competitive as those, which are owned and controlled by private employers? Is organizational and economic growth incompatible with democratic participation? Schoening (this volume) points out that workers in a cooperative that manufactures outdoor recreational equipment developed a new framework – cooperative entrepreuneurialism – that helped them reconcile the contradictions between economic growth and worker democracy. Meyers outlines a rich set of participative practices, which she calls densely clustered democracy, that have allowed an extremely diverse workforce in a worker-owned ﬁrm to both expand the business, yet keep their democratic traditions robust. The successful co-operative that Meyers analyzes reminds us that there truly are viable alternatives to alienated work in contemporary America.

CONDITIONING WORKPLACE AND LABOR FORCE PARTICIPATION: STATE AND LABOR MOVEMENT EFFECTS ON WORKERS The ﬁnal cluster of chapters examines the issue of worker participation from a very different angle, focusing on how external mediators, such as state agencies, state policies, and the labor movement, try to create particular conditions for people to participate in the labor force. While the chapters in Parts I and II analyze variables pertaining to the internal mechanisms and

Introduction

xix

social relations of economic organizations, the chapters in Part III take as their point of departure forces and barriers that workers encounter before entering the labor market. As numerous researchers have shown over the years, there are many mediating factors that affect people’s ability to ﬁnd jobs, to gain access to different workplaces, labor markets, and distinct types of employment contracts. Labor market intermediaries, such as training, job placement, job search, and staffing agencies, play a role in stratifying access to job opportunities and skill acquisition. State policy can determine who is available or able to work which jobs and can seriously limit workers’ job and occupational mobility, as the chapters by Ridzi and Banerjee, and Smith, Flynn, and Isler show. State policy can also promulgate dependent relations that render different groups of workers quite vulnerable to policy and institutional changes. If workers are to beneﬁt from new workplace innovations that have depth and offer opportunity – in workplaces where the commitment to organizational participation is more than rhetorical – they need to gain access to them in the ﬁrst place. Not all groups have access to information about how job and labor markets work; not all groups have access to networks that can link them to jobs with ‘‘good’’ employers; and not all groups know how to approach employers, to market themselves in their search for employment. Thus, when state organizations in particular make services available to achieve these ends, the quality and the content of these services matter enormously, with respect to shaping opportunity (Ridzi & Banerjee, this volume; Smith, Flynn, & Isler, this volume). In today’s economy, many workers are victims of the worst effects of temporary employment, such as the lack of beneﬁts, lower pay, and insecurity; at the same time, not all groups can easily ﬁnd secure jobs in today’s economy. Thus, institutional support intended to protect workers and to enable them to land permanent jobs matters enormously as well. Neuwirth (this volume) analyzes the efforts of a labor-movement-based staffing agency to remediate the deleterious conditions of temporary workers. Whether innovative organizations such as this, designed to offer that institutional support, can survive or not are pressing questions for social scientists and for policy makers and activists. The chapters in the ﬁnal section of this volume all shed light on these issues.

CONCLUSION Combined, the articles in this volume represent cutting-edge research conducted on a broadly deﬁned topic. Worker participation can be viewed from

xx

INTRODUCTION

many different angles; the papers included here illuminate three. They are suggestive of a vigorous research area in the sociology of work and I would hope that many readers will pursue the themes and questions they raise.

NOTES 1. Similar experimentations with the Japanese model of production were very well known in the 1980s. Japanese transplants include Toyota’s Camry plant in Kentucky (Besser, 1996), Subaru-Isuzu’s plant in Indiana (Graham, 1995); for other examples see Milkman (1991). 2. For a sampling of ‘‘effects’’ studies see Batt (2004), Be´langer, Edwards, and Wright (2003), Berg, Kalleberg, and Appelbaum (2003), Delbridge and Whitﬁeld (2001), Easton and Jarrell (1998), Gunn (1984, Appendix 2), Handel (2005), Hodson (2001a, 2001b), Hodson and Roscigno (2004), Jones (1984), Murray, Be´lanager, Giles, and Lapointe (2002), Pencavel (2001), Poole, Lansbury, and Wailes (2001), Preuss (2003), Shaiken, Lopez, and Mankita (1997), Siebold and Shea (2001), Skuratowicz and Hunter (2004), Smith (1993), and Townsend and Scott (2001). 3. See, for example, Accounting, Organizations and Society, a journal which analyzes the social and political nature of accounting and economic calculation. 4. For a sampling see Barker (1993), Essamel and Willmott (1998), Graham (1995), Grenier (1988), and Sewell (1998). 5. Besser (1996); Milkman (1997); Smith (1996, 2001); and Vallas (2003, forthcoming). 6. For a sampling see Gastil (1993), Gunn (1984), Jackall and Levin (1984), and Mansbridge (1983).

ACKNOWLEDGMENTS First, I thank Randy Hodson for providing the opportunity to assemble this volume. His own research on worker participation is the gold standard of scholarship on worker participation. I am very grateful to a number of people for assisting with the production of this volume. The following individuals took time from their busy schedules to review submissions: Carolina Apesoa-Varano, Tom Beamish, Beth Bechky, Fred Block, Teresa Casey, Julie Collins, Eric Grodsky, Matt Huffman, Jonathan Isler, Joan Meyers, Estee Neuwirth, Xiaoling Shu, Eva Skuratowicz, Clare Stacey, Steve Vallas, and ﬁve anonymous reviewers. I’d also like to thank Sarah Ovink for help with editing the manuscripts and assisting with ﬁnal assembly of the volume.

Introduction

xxi

REFERENCES Adler, P. (1993). The learning bureaucracy: New United Motors Manufacturing, Inc. Research in Organizational Behavior (Vol. 15, pp. 111–194). Greenwich, CT: JAI Press. Barker, J. (1993). Tightening the iron cage: Concertive control in self-managing teams. Administrative Science Quarterly, 38, 408–437. Barley, S., & Kunda, G. (1992). Design and devotion: Surges of rational and normative ideologies of control in managerial discourse. Administrative Science Quarterly, 37, 363–399. Batt, R. (2004). Who beneﬁts from teams? Comparing workers, supervisors, and managers. Special issue on new work practices. Industrial Relations, 43, 183–212. Be´langer, J., Edwards, P., & Wright, M. (2003). Commitment at work and independence from management: A study of advanced teamwork. Work and Occupations, 30, 234–252. Berg, P., Kalleberg, A., & Appelbaum, E. (2003). Balancing work and family: The role of highcommitment environments. Industrial Relations, 42, 168–188. Besser, T. (1996). Team Toyota: Transplanting the Toyota culture in the Camry plant in Kentucky. Albany: SUNY Press. Blasi, J. R. (1988). Employee ownership: Revolution or ripoff? Cambridge, MA: Ballinger Publishing Co. Collom, E. (2003). Two classes and one vision? Managers’ and workers’ attitudes toward workplace democracy. Work and Occupations, 30, 62–96. Delbridge, R., & Whitﬁeld, K. (2001). Employee perceptions of job inﬂuence and organizational participation. Industrial Relations, 40, 472–489. Easton, G., & Jarrell, S. (1998). The effects of total quality management on corporate performance: An empirical investigation. Journal of Business, 71, 253–306. Edwards, P., & Collinson, M. (2002). Empowerment and managerial labor strategies. Work and Occupations, 29, 272–299. Essamel, M., & Willmott, H. (1998). Accounting for teamwork: A critical study of group-based systems of organizational control. Administrative Science Quarterly, 43, 358–396. Gastil, J. (1993). Democracy in small groups: Participation, decision making and communication. Montpelier, VT: New Society Publishers. Graham, L. (1995). On the line at Subaru-Isuzu: The Japanese model and the American worker. Ithaca: Cornell/ILR Press. Grenier, G. (1988). Inhuman relations: Quality circles and anti-unionism in American industry Philadelphia: Temple University Press. Gunn, C. E. (1984). Workers’ self-management in the United States. Ithaca: Cornell University Press. Hackman, J. R., & Wageman, R. (1995). Total quality management: Empirical, conceptual, and practical issues. Administrative Science Quarterly, 40, 309–342. Hales, C. (2000). Management and empowerment programmes. Work, Employment, and Society, 14, 501–519. Handel, M. (2005). Trends in perceived job quality, 1989–1998. Work and Occupations, 32, 66–94. Hodson, R. (2001a). Dignity at work. Cambridge: Cambridge University Press. Hodson, R. (2001b). Disorganized, unilateral, and participative organizations: New insights from the ethnographic literature. Industrial Relations, 40, 204–230. Hodson, R., & Roscigno, V. (2004). Organizational success and worker dignity: Complementary or contradictory? American Journal of Sociology, 110, 672–708.

xxii

INTRODUCTION

Jackall, R., & Levin, H. (1984). Worker cooperatives in America. Berkeley: University of California Press. Jones, D. (1984). American producer cooperatives and employee-owned ﬁrms: A historical perspective. In: R. Jackall & H. Levin (Eds), Worker cooperatives in America (pp. 37–56). Berkeley: University of California Press. Logue, J., & Yates, J. (2001). The real world of employee ownership. Ithaca: Cornell/ILR Press. Mansbridge, J. (1983). Beyond adversary democracy. Chicago: University of Chicago Press. Milkman, R. (1991). Japan’s California factories: Labor relations and economic globalization. Los Angeles: University of California, Institute of Industrial Relations. Milkman, R. (1997). Farewell to the factory: Auto workers in the late twentieth century. Berkeley: University of California Press. Murray, G., Be´langer, J., Giles, A., & Lapointe, P.-A. (2002). Work and employment relations in the high-performance workplace. London: Continuum Press. Pencavel, J. (2001). Worker participation: Lessons from the worker co-ops of the Pacific Northwest. New York: Russell Sage Foundation. Poole, M., Lansbury, R., & Wailes, N. (2001). A comparative analysis of developments in industrial democracy. Industrial Relations, 40, 490–525. Preuss, G. (2003). High performance work systems and organizational outcomes: The mediating role of information quality. Industrial and Labor Relations Review, 56, 590–605. Rothschild-Whitt, J. (1979). Collectivist democracy: An alternative to rational-bureaucratic models. American Sociological Review, 44, 509–527. Rubenstein, S., & Kochan, T. (2001). Learning from Saturn: Possibilities for corporate governance and employee relations. Ithaca: Cornell/ILR Press. Sewell, G. (1998). The discipline of teams: The control of team-based industrial work through electronic and peer surveillance. Administrative Science Quarterly, 43, 397–428. Shaiken, H., Lopez, S., & Mankita, I. (1997). Two routes to team production: Saturn and Chrysler compared. Industrial Relations, 36, 17–45. Siebold, D., & Shea, B. C. (2001). Participation and decision making. In: F. Jablin & L. Putnam (Eds), The new handbook on organizational communication: Advances in theory, research, and methods (pp. 664–703). Thousand Oaks: Sage Publications. Skuratowicz, E., & Hunter, L. (2004). Where do women’s jobs come from? Job resegregation in an American bank. Work and Occupations, 31, 73–110. Smith, V. (1993). Flexibility in work and employment: The impact on women. Research in the Sociology of Organizations, 11, 195–216. Smith, V. (1996). Employee involvement, involved employees: Participative work arrangements in a white-collar service occupation. Social Problems, 43, 166–179. Smith, V. (2001a). Crossing the great divide: Worker risk and opportunity in the new economy. Ithaca: Cornell/ILR Press. Smith, V. (2001b). Teamwork vs. tempwork: Managers and the dualisms of workplace restructuring. In: D. B. Cornﬁeld, K. E. Campbell & H. J. McCammon (Eds), Working in restructured workplaces: Challenges and new directions for the sociology of work (pp. 7–28). Thousand Oaks: Sage Publications. Staw, B., & Epstein, L. (2000). What bandwagons bring: Effects of popular management techniques on corporate performance, reputation, and CEO pay. Administrative Science Quarterly, 45, 523–556. Talwar, J. P. (2002). Fast food, fast track: Immigrants, big business, and the American dream. Boulder, CO: Westview Press.

Introduction

xxiii

Townsend, A., & Scott, K. D. (2001). Team racial compositions, member attitudes, and performance: A ﬁeld study. Industrial Relations, 40, 317–337. Vallas, S. (2003). The adventures of managerial hegemony: Teamwork, ideology and worker resistance. Social Problems, 50, 204–225. Vallas, S. (Forthcoming). Empowerment redux: Structure, agency, and the re-making of managerial authority. American Journal of Sociology, 111(6). Varano, C. (1999). Forced choices: Class, community, and worker ownership. Albany: State University of New York Press. Whyte, W. F., Hammer, T., Meek, C., Nelson, R., & Stern, R. (1983). Worker participation and ownership: Cooperative strategies for strengthening local economies. Ithaca: ILR Press. Zbaracki, M. (1998). The rhetoric and reality of total quality management. Administrative Science Quarterly, 43, 602–636.

Vicki Smith Editor

This page intentionally left blank

xxiv

PART I: WORKER PARTICIPATION IN CONTEMPORARY CORPORATE WORKPLACES

1

This page intentionally left blank

2

THEORIZING TEAMWORK UNDER CONTEMPORARY CAPITALISM Steven Peter Vallas ABSTRACT Despite two decades of discussion and debate, key issues regarding the transformation of work have remained highly uncertain, largely because of the limitations in the theoretical frameworks with which team systems have been approached. In this paper, I draw loosely on my own fieldwork to articulate a series of propositions regarding features of the new managerial regimes that have remained only poorly explored. These center on the structural tensions and contradictions that team systems elicit, the contested and negotiated nature of the new regimes, and the complex interplay between the new work-systems and the identities that workers import from the old-managerial regimes. The paper suggests that the overarching concern with rates of diffusion and the performance effects of team systems should give way to more culturally nuanced approaches that can bring to light the divergent and often inherently contradictory forms the new regimes assume in an era marked by increasing demands for labor market flexibility.

INTRODUCTION For two decades now, analysis of work organizations has been awash in arguments concerning the spread of new work practices (Block, 1990; Piore Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 3–24 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16001-X

3

4

STEVEN PETER VALLAS

& Sabel, 1984; Smith, 1997; Vallas, 1999; Zuboff, 1988). Yet, in spite of the proliferation of articles and books on the ‘‘participative turn,’’ the nature of the new managerial regimes has remained highly uncertain. Although advocates of team systems have often argued that new technologies and market pressures tend to favor an expansion of worker autonomy and control (e.g., Adler, 1992a; Hirschhorn, 1984; Zuboff, 1988), the validity of these claims remains far from clear. Over against arguments that lean systems, TQM, and high-performance work organizations serve to enhance the work situations of front-line employees, remarkably little is known about how workers actually experience and respond to the new work practices (Appelbaum, Bailey, Berg, & Kalleberg, 2000; Berggren, 1992; Godard & Delaney, 2000). Even the geography of workplace change remains unclear: Although most U.S. analysts tend to assume that continental European ﬁrms and industries make much greater provisions for worker autonomy than do their counterparts in the U.K. or United States, this dichotomous view fails to acknowledge the existence of signiﬁcant variation among the European nations themselves (Benders, Huijgen, & Pekruhl, 2002; Poutsma, Hendrickx & Huijgen, 2003). The importance of these and kindred questions regarding the new managerial regimes is likely to grow increasingly pronounced, given mounting pressures for increased labor market ﬂexibility throughout the advanced capitalist world, from Germany to Japan and Australia. One conclusion at least seems safe: the theoretical frameworks that have been developed to explain the new work practices – most notably, the ‘empowerment’ and ‘labor control’ perspectives – seem unable to bear the weight so often assigned to them (Appelbaum et al., 2000; Godard & Delaney, 2000; Smith, 2001; Vallas, 2006). In this paper, I offer a set of critical observations that seek to identify features of the new managerial regimes that warrant closer attention than they have received. Drawing loosely on previous research – both my own and that of others – I argue that neither the empowerment nor the control perspective can provide an adequate conceptual basis with which to grasp workplace transformation in the contemporary setting. As I suggest below, despite their seemingly opposed arguments, these two perspectives are marked by several shared limitations, adherence to which has in fact been one source of our continuing uncertainty regarding the nature of the new regimes. I develop a series of propositions regarding workplace transformation that seek to advance an alternative approach toward the nature and consequences of workplace change. These propositions are not meant to serve as testable hypotheses, but rather to single out features of team systems that have not received the attention they deserve. As will become clear

Theorizing Teamwork under Contemporary Capitalism

5

in the following argument, I contend that team systems often manifest structural tensions and contradictions, as conﬂicting organizational logics compete for predominance. Such conﬂict and contention ought not to be understood as an exceptional condition, or as the mark of managerial failure; rather, such instability is a constitutive characteristic of the nascent post-bureaucratic regimes, which by their very nature open the way to the ongoing negotiation of work routines and even workers’ identities.

THE THEORETICAL IMPASSE In the United States, efforts to humanize work organizations have a long history, reaching back to producers’ cooperatives, to calls for employee councils, and (more recently) to managerial programs advocating the humanization of work. Such efforts remained marginal in their inﬂuence, however, until the late 1970s, when rising international competition began to erode the economic supremacy that U.S. manufacturing had established. It was at this time, when U.S. producers faced unprecedented threats from overseas competition, that advocates of new work practices began to ﬁnd sharply increasing demand for their services on the part of leading corporate ﬁrms. A wave of new managerial thinking began to radiate outward, from best-selling monographs and journal articles into professional conferences, training handbooks, and the curricula of the nation’s leading management schools. Drawing on diverse sources of inspiration – e.g., Japanese interest in total quality management and statistical process control, Western European notions of socio-technical systems and industrial districts – a loosely coupled movement arose that advocated far more participative or egalitarian forms of organizational culture than had been the norm under mass production (Adler, 1992a; Hirschhorn, 1984; Womack, Jones, & Roos, 1990). The managerial lexicon expanded to include such novel entries as ‘‘lean production,’’ the ‘‘learning organization,’’ ‘‘high performance work systems,’’ ‘‘self-directed teams,’’ and ‘‘ﬂexible specialization,’’ to name but a few. Darwinian thinking infused itself into these approaches, with advocates contending that ﬁrms which embraced the ‘participative turn’ were likely to survive and even ﬂourish; those refused to adapt were destined to grow extinct (Smith, 1997; Vallas, 1999). A key claim in this genre – often asserted but only occasionally substantiated (cf. Berggren, 1992) – held that the new managerial regimes did more than advance such managerial goals as enhanced productivity or quality. In addition, they stood to enhance the sense of meaning, autonomy, and

6

STEVEN PETER VALLAS

accomplishment that workers themselves derived from their jobs. Such symbolic job rewards, reinforced by material beneﬁts like proﬁt-sharing, enhanced forms of payment, and job security, would ﬁnally deliver the consent and commitment that a century of Fordism had failed to achieve (Appelbaum et al., 2000; Zuboff, 1988). An historical divide had been reached. Now, the success of the capitalist ﬁrm depended on its ability to empower its own employees. As interest in ‘empowerment’ theory grew, however, so too did skepticism about its core claims (Berggren, 1992; Dohse, Jurgens, & Malsch, 1985; Parker & Slaughter, 1994). Much of this skepticism portrayed lean production as a form of hyper-Taylorism, in that it used the normative pressures that team loyalty provides as a means with which to intensify the work process (hence, Parker & Slaughter’s depiction of lean production as ‘‘management by stress’’). Eventually, drawing on the logic of the labor-process school, critics developed an alternative conception of the new work practices that portrayed them as comprising a novel system of labor control that was if anything more subtle and pernicious than anything that industrial capitalism had known. This approach accepted the premise that the new work practices manifested substantial continuities with the old-managerial regimes, in that they, too, sought to limit the capacity for labor struggle and organization. Now, however, management sought to control not only workers’ outward behaviors but also their innermost convictions as well. Put differently, the new work practices sought more closely to align workers’ mentalities with the interests of the ﬁrm, in effect ‘‘controlling the underlying experiences, thoughts and feelings that guide [employee] actions’’ (Kunda, 1992 p. 11). Hence, the rise of such terms as ‘‘hegemonic despotism’’ (Burawoy, 1985), ‘‘coercive autonomy’’ (Smith, 1990), as well as ‘‘concertive,’’ ‘‘normative,’’ and ‘‘post-Fordist hegemonic’’ control (Barker, 1993, 1999; Graham, 1995; Kunda, 1992). For more than a decade now, the bulk of the research has been polarized between advocates of these two views (Frenkel, Korczynski, Shire, & Tam, 1999; Hodson, 2001; Smith, 2001). Within recent years, however, researchers have begun to chafe at the constraints that these perspectives impose on our efforts to make sense of the new work practices. Indeed, in another context (Vallas, 2006), I have pointed to the assumptions held in common by both the empowerment and the control approaches, arguing that beneath their obvious differences lie signiﬁcant limitations that are common to both points of view. Thus, both approaches adopt heavily deterministic outlooks that tend to marginalize any notion of human agency. Both focus singlemindedly on managerial structures, ignoring the social and organizational

Theorizing Teamwork under Contemporary Capitalism

7

conditions that mediate their effects. Both bring a teleological lens to bear on workplace transformation, as if the latter inevitably led toward stable and enduring ‘‘consensual’’ end states (Storey, 1985). And much of the intensity of this debate has stemmed from the fact that both views are tethered to highly prescriptive accounts of the capitalist ﬁrm that tend to be heavily value laden. It was with these sorts of theoretical constraints in mind that Hodson (2001) lamented the ‘‘theoretical straitjacket’’ in which the ﬁeld has found itself trapped. It was with precisely these considerations in mind that I constructed a multi-site comparative ethnography of the transformation of industrial work. In this research, I focused on ﬁve distinct manufacturing sites that were engaged in the production of forest products – mainly pulp and paper – for both commercial and consumer products (Vallas, 2003a, b, 2006). I will forego any formal introduction of these data; sufﬁce it to say that my research strategy sought to combine the richness of single-site case studies with the broader analytical power of multi-site research. Encountering a diverse set of production areas and manufacturing complexes, ranging from highly traditional, hide-bound settings to state-of-the art plants that employed ‘‘team concept’’ from their very design, I sought to understand the nature and evolution of workplace innovation, the features that characterize the trajectory of change initiatives, and the ways in which workers respond to such wholesale efforts to restructure their working lives. The results, on which I freely draw below, were instructive in that they began to point beyond the theoretical coordinates that have dominated the ﬁeld. The sections below are organized in terms of a series of critical propositions concerning the study of the new work practices that have emerged during the course of my research. As becomes clear in the conclusion, the premise that underlies these propositions is an argument on behalf of an alternative conception of workplace change that is attuned to the various institutional forces, organizational contradictions, and processes of negotiation within the ﬁrm that together operate to shape the nature and consequences of the new managerial regimes.

BEYOND THE IMPASSE: SOME ARGUMENTS ABOUT THE NATURE OF WORKPLACE CHANGE Proposition 1. The study of team systems and group work has fastened on a single strand of organizational change – its ‘‘normative’’ or participative element – in abstraction from rival currents of restructuring whose existence

8

STEVEN PETER VALLAS

greatly complicates the nature and consequences of workplace change. As they seek to enhance their competitive position, ﬁrms typically resort to a multiplicity of human resource strategies that tend to combine in contradictory and tension-ﬁlled ways, lending workplace change an over-determined character that cannot be reduced to a single strand of organizational change. Theory and research on team systems have highlighted efforts to strengthen the citizenship rights of many employees. In the terms of Derber and Schwartz (1983), this development constitutes an historic weakening of the boundary between the polity and the capitalist ﬁrm, making possible an infusion of democratic rights into the realm of private property. Yet, it is vital to acknowledge that this is not the only boundary that has been weakened. Apace with the growing emphasis on labor market ﬂexibility, the boundary between the ﬁrm and market pressures has also begun to break down, precipitating an implosion of market forces into the organization, often sweeping away the job structures that had previously sheltered workers against the vicissitudes of the labor market (Cappelli, 1997). This is especially clear in the rapid expansion of non-standard forms of employment in the United States, with temporary work, independent contractor arrangements, and part-time positions increasing far more rapidly than has the labor force as a whole (Kalleberg, Hudson, & Reskin, 2000; Kalleberg, Reynolds, & Marsden, 2003). This implosion of market relations into the innermost recesses of the ﬁrm sometimes ﬁnds reﬂection in the very language that managerial initiatives support, as when employees are expected to regard their co-workers in other departments as their ‘‘customers.’’ The result has introduced salient contradictions into the employment relationship, as two very different logics – one premised on citizenship and the other on commodity relations – compete for predominance. These contradictions have begun to outstrip even the most sophisticated theoretical interventions. Thus, in their inﬂuential effort to re-write the evolution of managerial paradigms in the United States, Barley & Kunda (1992) contended that managerial regimes have historically alternated between surges of ‘‘rational’’ and ‘‘normative’’ discourse. By the former they have in mind Taylorism and operations research; by the latter, welfare capitalism, human relations models, and, more recently, organizational cultures steeped in the ethos of participation (Kunda and Ailon-Souday, 2005). My argument is that these two forms of discourse do not only alternate between historical periods, but instead increasingly coincide at one and the same time, engendering structural tensions and contradictions that are often inherent

Theorizing Teamwork under Contemporary Capitalism

9

features of the new managerial regimes (McLoughlin, Badham, & Palmer, 2005).1 Indeed, it was the emergence of precisely such structural tensions that led Sewell (1998) to speak of hybrid or ‘‘chimerical’’ regimes (in that they combine surveillance with citizenship), and which prompted Stark (2001) to coin the term ‘‘heterarchy’’ (a term he used to characterize the tension between capitalist and socialist logics within Eastern European factories). This point arose in my own ﬁeldwork (e.g., Vallas, 2003a). Sophisticated team systems were most often introduced within new production areas, where massive new production complexes had been installed. Here, the members of production teams were led to expect an expansion of their control over machine operations (the normative strand of workplace change). Yet, what unfolded was instead a far more complicated and internally contradictory state of affairs. Because the new production complexes were so capital intensive, production managers and machine superintendents encountered performance pressures that made them understandably loathe to leave substantial areas of machine operations to the discretion of their hourly employees. Moreover, as the ranks of credentialed employees expanded, with process engineers enjoying a newfound centrality within the production process, standardized conceptions of process control were imposed in the form of best practices, grade recipes, and formalized rules that stood at odds with the premises of ‘‘team concept.’’ Often confronting even more severe limits on their control over machine operations than existed within traditionally organized production areas, operators at the new, team-based production areas came to chafe at the new constraints. Many complained of feeling ‘‘handcuffed’’ by the new production rules and of ‘‘having to work from the neck down.’’ They commonly resented being made to ‘‘feel like puppets on a string.’’ In such cases, the result was an accumulation of suspicion and distrust among hourly workers concerning management’s real priorities, as two conﬂicting logics – a technicist and a participative logic – were brought to bear on the labor process. Although it may be tempting to view these tensions and contradictions as anomalous events, or as simply the product of poor execution by these ﬁrms, an alternative reading sees them as an increasingly common property of workplaces undergoing the transition from one system of authority to another. Interestingly, analogies may be found in the literature on post-socialist manufacturing plants, which have at times combined both socialist and capitalist logics at one and the same time (Burawoy & Lukacs, 1992; Stark, 2001). In the advanced capitalist ﬁrms, the questions that need to be addressed are many. Under what economic and organizational conditions

10

STEVEN PETER VALLAS

are such structural tensions most pronounced? Are they rooted primarily within organizational conﬂicts internal to the ﬁrm (e.g., struggles between human resource and engineering divisions)? Or do they reﬂect the conﬂicting demands – the ‘‘multiple audience’’ problem – brought to bear on the ﬁrm from within its wider environment? Further, how do managers seek to cope with such contradictions? Arguably, if ﬁrms have used market uncertainties to exact greater compliance with team-based initiatives, as several analysts contend (e.g., Burawoy, 1985; DiTomaso, 2001; Fraser, 2002), then the notion of ‘‘citizenship’’ has begun to acquire an implicitly coercive character that seems likely to place sharp limits on the formation of worker consent. Proposition 2. Research has focused one-sidedly on managerial initiatives, as if employees passively conformed to such practices. Yet, in fact, subaltern groups of employees often selectively comply with, modify, negotiate, and sometimes even appropriate team systems, as their own interests require. In short, the introduction of new work practices often sets in motion organizational dynamics involving workers in various occupational groups, and which are likely to have material effects on the forms that workplace change assumes. Human resource theorists in the United States have been particularly prone to emphasize the determinative role played by particular types of work structures, seeing them as the decisive factors affecting the outcome of workplace change. Much of this thinking has suggested that signiﬁcant beneﬁts from new work practices require a signiﬁcant investment in ‘‘bundles’’ of change rather than piece-meal reform. For example, the combination of self-directed teams, cross-functional training, and enhanced systems for compensation are expected to produce a critical mass of workplace transformation (Pil & MacDufﬁe, 1996), while smaller, more delimited change might not (cf. Appelbaum et al., 2000). Similar claims have been advanced in other forms, giving rise to what might be called a form of ‘‘industrial botany,’’ in which different species of innovations are expected to produce differential results (Batt & Appelbaum, 1995). The common assumption here is that the outcome of workplace change is an unmediated reﬂection of the managerial initiatives that ﬁrms introduce (Hunter, MacDufﬁe, & Doucet, 2002). What disappears from such a view is any concern for the social and organizational processes that unfold during the transformation itself – that is, for the informal patterns of contention and negotiation that inevitably shape the outcome of any new work system (Thomas, 1994).

Theorizing Teamwork under Contemporary Capitalism

11

That workers cannot be presumed to be passive recipients of managerial initiatives emerges with great clarity in the recent studies developed by Smith (2001). Focusing on ofﬁce and factory workers in disparate settings, she found that workers often did embrace team systems, but not because of any implicit acceptance of managerial goals. Rather, their actions reﬂected workers’ hope that the new work practices would provide resources that held strategic signiﬁcance from their own point of view. Thus, by learning how to run team meetings or to coordinate tasks among team members, minority workers within routine ofﬁce jobs hoped to gain a foothold within the corporate world. Likewise, manual workers facing the shutdown of their wood products plant embraced a high-performance work system not because they shared management’s outlook, but because they believed that complying with the new work system would strengthen the ﬁnancial viability of their plant. Smith’s point is that workers actively respond to the new work systems, using them in ways that reﬂect their own material needs. Much the same point emerged in my own ﬁeldwork. Even workers who were relatively resigned to the hierarchical nature of the wage–labor relationship selectively chose which aspects of team concept they deemed acceptable. Thus, senior workers often chafed at the expectations of job rotation, for it seemed to signal a regression to less skilled tasks they had left behind them many years ago. Often, workers refused to assume responsibility for enforcing discipline or safety rules on their coworkers, since this undermined their most deeply held conceptions of worker solidarity. Especially revealing were instances in which workers engaged in concerted activities aimed at reshaping the contours of the team systems in which they were employed. In one such case, when a group of workers felt betrayed by the manager that had led their team initiative, they mobilized a determined campaign to force the manager to resign (Vallas, 2003b, 2006). Indeed, where team systems seemed to provide for the greatest degree of empowerment, the reasons often stemmed from the unanticipated consequences of managerial initiatives, which allowed workers informally to appropriate the new work practices and to use them for their own ends. Proposition 3. The tendency has been for theorists of team systems to adopt a teleological approach that views the new work practices as driven toward a stable and unproblematic ‘end’ (typically involving heightened consent and commitment). Yet, in fact, precisely because they often weaken or dismantle rigid, centralized bureaucracies, the new work practices can foster increased levels of contention, with informal negotiation becoming even more fundamental to outcomes than has been the case in the past.

12

STEVEN PETER VALLAS

That team systems often interject signiﬁcant elements of volatility into the work organization has been a repeated theme within much extant research. The ethnographic accounts developed by both Grenier (1988) and Graham (1995) drew attention to the organizational volatility that can ensue in the wake of team-based initiatives (in one case, the introduction of Quality Circles; in the other, lean production in a Japanese transplant). Research on new technology has likewise produced similar ﬁndings (Kelley, 1986; Noble, 1984; Thomas, 1994). An even clearer instance in which the new work practices served to undermine managerial legitimacy can be found in a recent ethnography of a British bank (Weeks, 2004). Senior management faced employees who almost uniformly held a derisive view of their employer. Rituals of derogation and complaint were among the most important sources of solidarity that employees enjoyed; ironically, despite its overtly negative content, this ‘‘culture of complaint’’ was critical to the ongoing integration of the ﬁrm. Yet, because management was convinced of the need for a new and more positive organizational culture, the ﬁrm repeatedly introduced consultant-designed systems for worker participation that seemed to pale in comparison to the celebrated instances of managerial innovation widely reported in the management journals. The result served only to reafﬁrm the cynical view that predominated among the ﬁrm’s employees – in turn spawning further efforts at organizational change. What emerges from Weeks’s account is an image of an organization that was locked in a self-perpetuating spiral of complaint and managerial reform. Even a merger and acquisition, followed by subsequent bouts of restructuring, did little to alter this conﬂicted and cynical state (cf. Kunda, 1992). My own research also found clear tendencies for team systems to reproduce or even to inﬂame the boundaries between managers and workers, disrupting the very legitimacy that management hoped to achieve. This point emerged from a series of paired comparisons between production areas that employed equivalent technologies and were engaged in the production of similar product grades, yet that relied on distinct systems of work relations – i.e., traditionally organized versus team-based systems (cf. Hodson, 1998, 2001; Vallas, 2003b). In this analysis, I identiﬁed and coded discrete statements and behaviors that reﬂected three different aspects of workplace culture: the legitimacy of managerial practices, the salience of the management/hourly boundary, and the frequency of behavioral deﬁance (see Vallas, 2003b for a full discussion). Contrary to both the empowerment and the control perspectives, team systems actually seemed to reduce the legitimacy that management enjoyed and to heighten class boundaries. I

Theorizing Teamwork under Contemporary Capitalism

13

found no evidence that behavioral deﬁance tended to decline within teambased production areas; indeed, it seemed to focus on intrinsic job rewards that were often difﬁcult for management to address. Even at the non-union plant I studied, where team systems were introduced on a plant-wide basis, salient boundaries appeared between management and workers, eventually becoming as pronounced as in older, unionized plants. Interpreting these patterns leads me to suggest that such unanticipated consequences were the result of developments that unfolded at three distinct levels of analysis. First, at the level of worker cognition, the introduction of team systems encouraged workers to adopt an increasingly demanding, intrinsic orientation toward their jobs. Yet, because such levels of job identiﬁcation often seemed to outpace management’s capacity to transform the work structures on which it relied, the result tended to fuel a sense of deception and hypocrisy among front-line employees. Second, at the structural level, team systems fostered patterns of solidarity and mutual support that often spilled over management control, enabling workers to use teambased ties to resist or reshape managerial initiatives. Third, and at the cultural level, by introducing the language of participation into the workplace, team systems provided workers with a legitimate rhetorical framework with which to claim decision-making powers they had previously been denied. My point is that when direct participation lags, management becomes the victim of its own success: instead of congealing into a system that secures worker commitment and consent, the cognitive, social and cultural dimensions of work instead begin to generate unanticipated consequences, providing workers with precisely those organizational resources they need to claim discretionary powers they had previously been denied. Again, it may be tempting to view such disruptive effects of team systems as exceptional states, or as reﬂecting inconsistent or incompetent managerial staffs. I contend that the reverse is true, that textbook exemplars are the exceptional instances, and that the bulk of the cases where team systems are introduced languish in a more precarious, unstable, and conﬂicted state. Indeed, precisely because they seek to loosen bureaucratic controls, the new forms of work organization may by their very nature be increasingly susceptible to or even invite such contestation (Kuhlmann & Schumann, 2001; Vallas, 2006). Conﬂict and instability, rather than an unproblematic commitment and consent, may be the normal outcomes of the new work practices, lending managerial regimes an inchoate, constantly negotiated character that is not well described in terms of commitment and consent. This may be one reason why American managers have seemed to back off from support of self-directed team systems, which they ﬁnd difﬁcult to

14

STEVEN PETER VALLAS

control (Osterman, 2000). It may also help explain the ephemeral nature of team systems, whose survival rates are often remarkably low over time. This phenomenon in particular warrants much further research than it has received.2 Proposition 4. Understanding the meanings that emerge in the wake of workplace change will require more systematic attention to the relationship between work organizations and the types of subjectivity they promote, for the identities and action-orientations that workers adopt are very likely implicated, both as determinants and as effects, in the outcome of workplace change. There is, by now, a growing accumulation of studies that have used survey methods to gauge employee attitudes and perceptions regarding workplace change. The literature here is, however, rife with inconsistencies. Some studies ﬁnd that formal provisions for participation demonstrably enhance the meaning that workers enjoy in their work situations, and do so without promoting the stress about which critics have forewarned (Appelbaum et al., 2000; Freeman & Rogers, 1999; Hunter et al., 2002). Hodson’s systematic quantitative analysis of all English-language ethnographies (2001) largely but not entirely supports this pattern, as does the study of U.S. autoworkers by Hunter et al. (2002). Yet other studies ﬁnd very different results. In a British study that used nationally representative survey data for 1998, team membership was found to have no signiﬁcant effects on the experience and perceptions of employees (Harley, 2001). Using data from a Canadian survey, Godard (2001) found a complex pattern in which autonomous work teams and total quality management had generally positive effects, which then decayed or even reversed once the highest levels of workplace change were reached. The pattern was one in which high-performance expectations provoked deep-running conﬂicts among employees about their work situations (cf. Milkman, 1997; Shaiken, Lopez, & Mankita, 1997; Weeks, 2004). Such research on worker attitudes has only begun to scratch the surface of the subjective responses to workplace change. Seeking to explore this terrain further, my own ﬁeldwork (2006) sought to identify the most salient types of response manifest among workers confronting organizational and technological change. Two of these types represented action-orientations in which workers exhibited deeply traditional conceptions of work and self that collided against the demands of the new work practices. In these instances, workers manifested occupational identities that lay at opposite ends of a continuum that ranged from passive accommodation to fervent deﬁance. In the former case, workers embraced self-concepts that reiﬁed their subordinate

Theorizing Teamwork under Contemporary Capitalism

15

positions within the ﬁrm, while in the latter, they adopted such ‘‘recalcitrant selves’’ as to reproduce managerial stereotypes about the untrustworthiness of manual employees. In both cases, occupationally induced selves inherited from the past operated to establish subtle, yet powerful limits on the possibility of workplace change.3 Yet, these were not the only forms of self that I encountered. In between the two poles of accommodation and resistance lay two other types that drew on egalitarian action-orientations that proved more favorable to the expansion of direct participation. In one such type, skilled craft workers felt emboldened to view themselves as the equal of the credentialed engineers with whom they worked, and thus to usurp the right to perform complex analytical work (mainly programming the new process control systems) much as did the engineers. Here, workers embraced conceptions of themselves as deserving of the right to be included in strategically signiﬁcant tasks. (Said one worker, ‘‘I’m not gonna back off and say ‘Oh, I can’t do that. That’s Process Control.’ I am Process Control! That’s my job!’’) Such workers left electrical engineering textbooks and schematic diagrams at their desks, serving as markers of their occupational identities. In another type of such response, workers formulated conceptions of themselves as citizens of the plant community, and thus as deserving a voice that was the equal of that allowed to salaried employees. Such workers engaged in concerted actions that blocked signiﬁcant personnel decisions that management would otherwise have made, in effect renegotiating the boundaries of managerial authority. My point here is that the success or failure of the new work practices is not merely a matter of organizational strategies; it reaches deep into the habitus of the workers involved. In an effort to understand the relation between work and self under newly emerging managerial regimes, a literature has begun to emerge on the survival strategies that workers and managers adopt and the conceptions of self which organizations increasingly require of employees. Kunda (1992) speaks of ‘‘engineered selves,’’ and Casey (1995) of ‘‘designer selves,’’ in each case referring to efforts through which corporations seek to articulate employees’ identities with the requirement of the ﬁrm (Hochschild, 1983; Schoenberger, 1997). Tempering such fears are recent studies that have begun to draw attention to the undercurrents of cynicism that often result from efforts to foster strong organizational cultures (see Fleming & Spicer, 2003; Weeks, 2004). This phenomenon again speaks to the limits of managerial control even under conditions where employees lack formal means of self-defense. In a recent paper on the subjective implications of economic insecurity, Collinson (2003) has proposed a three-fold typology that holds particular

16

STEVEN PETER VALLAS

relevance here. Two of the types he recognizes – the ‘‘deferential’’ and the ‘‘resistant’’ – are familiar, and roughly correspond to the types that emerged in my own research, as sketched out above. By ‘‘deferential’’ selves, Collinson refers to workers who become ‘‘preoccupied with themselves as valued objects in the eyes of those in authority’’ (2003, pp. 537, 536) and adopt a conformist response to the ﬁrm’s demands. At the opposite end of the continuum are ‘‘resistant’’ selves, in which workers sustain a positive sense of identity through opposition to the ﬁrm. They, thus engage in various forms of subversion, however furtive, that range from foot-dragging and indifference to more overtly deﬁant behaviors. The ﬁrst of these two types involves a collapse of the boundary between self and employer; the second type involves a fervent effort to keep that boundary in place. Collinson’s third type of self – the ‘‘dramaturgical’’ self – is more complex in that it fuses elements of the ﬁrst two types. Here, employees cope with the uncertainties of the contemporary economic context by maintaining the outward appearance of conformity even as they inwardly resist the ﬁrm’s demands. What is required here is the ability to convincingly perform the rites of commitment and consent, while maintaining a ‘back stage’ region that is immune to their effects. Thus, managers seek to foster the appearance of working especially long hours (e.g., by leaving their ofﬁce lights on, suit jackets on the backs of their chairs), even after they’ve gone home for the day (see Collinson & Collinson, 2004). The point here is that strategic forms of interaction begin to take the place of more authentic ones, in the process undermining the sense of achievement and integrity that workers have traditionally sought. This is of course the claim that Sennett (1998) has advanced concerning the ‘‘corrosion of character’’ under ﬂexible capitalism. His argument – that economic and organizational uncertainty, coupled with superﬁcial forms of solidarity, act to undermine the basis on which self and solidarity depend – has not received the empirical attention it deserves. Needed are nuanced, ﬁnely grained studies that can chart the various identities that workers’ jobs have encouraged them to adopt, and the ways in which these identities act back on the effort to restructure work. Equally important will be studies that can empirically ground our conceptual typologies, exploring the ways in which the ‘‘staging’’ of compliance and the prevalence of strategic forms of interaction induce workers to redeﬁne their selves. Proposition 5. Efforts to understand the outcome of workplace change will need to pay much closer attention to the linkages between instances of workplace change and the wider organizational ﬁelds in which they emerge. Especially needed are studies aimed at identifying ways in which

Theorizing Teamwork under Contemporary Capitalism

17

institutional structures impinge on the organizational capacities that disparate groups of managers enjoy in the effort to restructure work organizations. If the foregoing comments have much validity, then workplace change is indeed characterized by complexity, contention, and over-determination. As such, it represents a multi-layered and often contradictory phenomenon in which various occupational groups – executives, managers, engineers, technicians, and manual workers of varying skill levels – mobilize competing conceptions of how work might best be arranged. The question must therefore be addressed: what factors inﬂuence the relative capacities that each group brings to bear on the process of workplace change? One useful point of departure can be found in the conceptual approach developed by Fligstein (1987, 1990). In his approach, different groups of managerial employees invoke competing ‘‘conceptions of control,’’ or shared understandings of how work structures ought to be arranged. Thus, technical professionals, human-resource managers, ﬁnancial analysts, and legal professionals all bring distinct cognitive approaches to bear on the organization of work. The relative position of each group within the ﬁrm is not idiosyncratic, but is instead conditioned by the political, legal, and economic conditions that obtain at any given time. Pursuing a similar line of reasoning, Jacoby (1985) has shown how the surge of the workers’ movement after World War I dramatically altered the nature of personnel management, calling forth massive shifts in the structure of large ﬁrms and eventually giving rise to the profession of human-resource management (see Edelman, 1992). In a comparative study of the U.S., UK, Germany, and Spain, Guillen (1994) concludes that the differential position of management intellectuals (HR professionals, social scientists concerned with the workplace, and consultants of varying stripes) had material effects on the reception that Taylorism, human relations and structural analysis received (Poutsma et al., 2003). None of these authors was concerned with the contemporary post-Fordist era. Yet, their work suggests the need for closer attention to the position that various groups of managerial actors hold during the process of change, and to the ways in which the institutional structures found within particular nations vary in this regard. In the United States, for example, although team systems are dependent on the abstract systems and programs developed by management theorists, consultants, and business school curricula, the actual implementation of such systems often depends on the organizational positions that managerial groups have established for themselves within a

18

STEVEN PETER VALLAS

given nation-state. Often, the power of American human-resource managers is quite limited, their staffs far less numerous than those enjoyed by engineering departments, and their positions deemed something of a luxury (Prechel, 1994). Arguably, if the quality movement has been characterized by a greater emphasis on the instrumental than the normative components of lean production, the reason ought not to be sought in the essential features of the lean system itself (Hackman & Wageman, 1995; Klein, 1994), but rather in the relative balance of power established among distinct groups of managers in the United States, owing to the institutional arrangements established in this country. Arguably, the position of managers and executives within nations characterized by corporatist institutions (such as in Sweden or the Netherlands) will be sharply different, with outcomes that reﬂect this difference.

DISCUSSION AND CONCLUSIONS The foregoing analysis is premised on the belief that the realities of workplace transformation have taken forms that are more complex, contradictory, and contingent that existing theories have allowed. Neither the theory of empowerment (with its utopian constructs) nor the theory of intensiﬁcation and labor control (which substitutes a fatalist, dystopian outlook) properly equips us to grasp the contested nature of the team systems currently unfolding within the corporate world. Needed are models of workplace change and team systems that acknowledge the limits of managerial initiatives and the unanticipated consequences they so often entail. Rather than presuming that outcomes are a function of managerial designs alone, our thinking must begin to incorporate aspects of workplace life that have previously been conﬁned to the margins of theoretical discussion and debate: the existence of structural tensions and contradictions as constitutive features of workplaces undergoing far-reaching transformations; the informally negotiated nature of the outcomes that obtain, which often proceed down paths at odds with managerial plans; the fractious and often volatile states to which the new work practices are prone; and the lingering effects that inherited forms of subjectivity and identity are likely to have on the outcomes that unfold. Needed too is a more reﬂexive awareness that the new regimes are likely to reﬂect variations in the resources their advocates enjoy, not only within particular ﬁrms, but also within wider organizational ﬁelds. Equally important, discussion and debate need to acknowledge the ironies and dilemmas that often characterize the transformation of work. For

Theorizing Teamwork under Contemporary Capitalism

19

example, change requires a deep and abiding commitment on the part of whole organizations – but suffers grievously when the process itself acquires a centralized, hierarchical cast. Support for the new work practices is especially pronounced among large, well-positioned corporations that enjoy ample resources – but these ﬁrms are typically saddled with bureaucratic complexities and constraints that provide little encouragement for the spontaneity on which team systems depend. Moreover, the energy and momentum that often seem to fuel movements for self-direction rest on precisely those forms of charismatic authority that are so difﬁcult to institutionalize within work organizations. Indeed, the question of power – so basic to Weber’s theory of domination – has received remarkably little direct commentary in recent debates over workplace change (cf. Thomas, 1994). The list of unexplored themes remains long. As I have obliquely suggested, we know surprisingly little about the predominant trajectories (or careers) that characterize team initiatives, or about the conditions that account for their differential rates of survival or death. A disproportionate number of the most inﬂuential studies has been concentrated on manufacturing industries, with insufﬁcient attention to the fate of workplaces within the burgeoning service (and government) sectors. Suggestive ﬁndings have begun to emerge about how the gender, racial, and ethnic composition of team systems shapes the outcomes that unfold. Arguably, legal structures and constraints affect the course of workplace change in ways that have remained unexplored. These and many other themes await systematic research. I contend that addressing such questions in the most fruitful manner will require new research strategies that are by their very nature attuned to the social and cultural dynamics that are characteristic features of work organizations undergoing change. No reasonable scholar will question the value of large-scale, nationally representative, cross-sectional survey research designs. These are after all important methods with which to study the diffusion of team systems, the conditions that promote or retard their introduction, and their consequences for performance. Yet these sorts of research designs have often entailed certain hidden costs, in that they implicitly conﬁne our thinking within a ‘‘states and rates’’ outlook that renders contradiction and contention all but impossible to apprehend. It is of course also the case that single-site ethnographies carry their own inherent limitations, the most obvious of which are their uncertain generalizability and replicability.4 For this reason, a powerful case can therefore be made on behalf of research designs that chart an intermediate course between ethnographic case studies and nationally representative survey designs. One such approach would entail comparative, multi-site ethnographies – qualitative

20

STEVEN PETER VALLAS

research strategies that can unearth rich, ﬁnely textured accounts of organizational cultures, while at the same time providing enough analytic breadth for comparison. Such research strategies might easily select multiple sites within a given company or industry and develop deeply grounded interpretations of the nature and consequences of managerial initiatives. They might also explore the social conditions that explain the divergent paths that the new regimes pursue. A strong case can be made for applying such designs in ways that traverse national boundaries, as for example in designs that concentrate on work establishments maintained by the same corporation with branches in different countries (Gallie, 1978). Doing so may require that the organization of our research must itself conform to the principles of team systems, using collaborative methods of research that enable us to outgrow the traditions of the lone ethnographer. The challenge of such research should be obvious. Much as with the effort to achieve inter-coder reliability, substantial training and ongoing dialogue will be needed to achieve substantive consistency among the teams of ethnographers, without sacriﬁcing the conceptual insights generated by each (Barley, 1996). My point here is that the very organization of our research may need to conform to the contours of the phenomenon we seek to understand. The contemporary context in which the advanced capitalist workplace is lodged is rife with uncertainty, with massive shifts in the employment relationship, the structure of workers’ tasks, and the nature of managerial authority. Over against such change, I suggest that a long-established axiom in the study of work – the observation that even seemingly powerless workers have ways of acting back on organizational elites – may well have continuing relevance in the workplace today, even in an era notable for the globalization of capital, the decline of labor organizations, and the growing demands for labor market ﬂexibility. The participative turn may, I suggest, lead ﬁrms down paths that defy existing theories – and the designs of managers as well.

NOTES 1. Interestingly, Adler (1992b)’s interpretation of NUMMI, Inc., does view both the rational and the normative sides of lean production as co-existing at one and the same time. But Adler argues that workers can be induced to embrace Taylorist work methods and even to participate in the standardization of their own jobs (as his conception of ‘‘democratic Taylorism’’ contends). I suggest that the typical result is quite different from Adler’s conception, and instead pits two conﬂicting logics – the participative and the instrumental – against one another.

Theorizing Teamwork under Contemporary Capitalism

21

2. An important issue here concerns the linkage between workplace change and charismatic forms of authority. It may be that managers who champion the new forms of work organization are in fact relying on charismatic authority, which is inherently (pace Weber) difﬁcult to routinize within large-scale formal organizations. Indeed, the routinization of charismatic authority may itself deplete the momentum on which team systems rest. 3. Although space limitations intervene on this point, it seems likely that gender relations play a greater role in the transformation of work than the current literature allows. My own research suggests that occupational identities that obstruct the course of workplace change (whether through accommodation or resistance) are often hinged with deeply traditional gender conceptions (see Kalev, 2004; Ollilainen & Rothschild, 2001; Vallas, 2003a). 4. For a critique of the positivist view of case study designs, see Burawoy (1998).

ACKNOWLEDGEMENTS This paper was originally presented as the keynote address for the 9th Annual Meeting of the International Workshop on Teamwork (IWOT) in Lisbon, Portugal, September 8–9, 2005. I wish to thank the IWOT organizers, and especially Jos Benders, Tomaso Pardi, and Stephen Procter for their insightful commentary. Direct correspondence to Vallas, Department of Sociology and Anthropology, George Mason University, 4400 University Drive – MS 3G5, Fairfax, VA 22032, USA or [email protected].

REFERENCES Adler, P. (1992a). Introduction. In: P. Adler (Ed.), Technology and the future of work (pp. 3–14). New York: Oxford. Adler, P. (1992b). The learning bureaucracy: New United Motor Manufacturing, Inc. Research in Organizational Behavior, 15, 111–194. Appelbaum, E., Bailey, T., Berg, P., & Kalleberg, A. (2000). Manufacturing advantage. Ithaca, NY: Cornell. Barker, J. R. (1993). Tightening the iron cage: Concertive control in self-managing teams. Administrative Science Quarterly, 38(Sept.), 408–437. Barker, J. R. (1999). The discipline of teamwork: Participation and concertive control. Thousand Oaks, CA: Sage. Barley, S. R. (1996). Technicians in the workplace: Ethnographic evidence for bringing work into organization studies. Administrative Science Quarterly, 41, 404–441. Barley, S. R., & Kunda, G. (1992). Design and devotion: Surges of rational and normative ideologies of managerial control. Administrative Science Quarterly, 37, 363–399. Batt, R., & Appelbaum, E. (1995). Worker participation in diverse settings: Does the form affect the outcomes? British Journal of Industrial Relations, 33(3), 353–378.

22

STEVEN PETER VALLAS

Benders, J., Huijgen, F., & Pekruhl, U. (2002). What do we know about the incidence of group work (if anything)? Personnel Review, 31(3), 371–385. Berggren, C. (1992). Alternatives to lean production: Lessons from the Swedish automobile industry. Cornell: ILR. Block, F. (1990). Post-industrial possibilities: A critique of economic discourse. Berkeley: University of California Press. Burawoy, M. (1985). The politics of production. Berkeley: University of California Press. Burawoy, M. (1998). The extended case method. Sociological Theory, 16, 4–33. Burawoy, M., & Lukacs, J. (1992). The radiant past: Ideology and reality in Hungary’s road to capitalism. Chicago: University of Chicago Press. Cappelli, P. (1997). Change at work. NY: Oxford University Press. Casey, C. (1995). Work, self, and society: After industrialism. London: Sage. Collinson, D. (2003). Identities and insecurities: Selves at work. Organization, 10(3), 527–547. Collinson, D., & Collinson, M. (2004). The power of time: Leadership, management, and gender. In: C. F. Epstein & A. Kalleberg (Eds), Fighting for time: Shifting boundaries of work and social life (pp. 219–248). New York: Russell Sage Foundation. Derber, C., & Schwartz, W. (1983). Toward a theory of worker participation. Sociological Inquiry, 53(1), 61–78. DiTomaso, N. (2001). The loose coupling of jobs: The subcontracting of everyone? In: I. Berg & A. L. Kalleberg (Eds), Sourcebook on labor markets: Evolving structures and processes (pp. 247–270). NY: Plenum Press. Dohse, K., Jurgens, U., & Malsch, T. (1985). From Fordism to Toyotism? The social organization of the labor process in the Japanese automobile industry. Politics and Society, 14(2), 115–146. Edelman, L. (1992). Legal ambiguity and symbolic structures. Organizational mediation of civil rights law. American Journal of Sociology, 97, 1531–1576. Fleming, P., & Spicer, A. (2003). Working at a cynical distance: Implications for power, subjectivity and resistance. Organization, 10(1), 157–179. Fligstein, N. (1987). The intra-organizational power struggle: The rise of ﬁnance personnel to top leadership in large corporations, 1919–1979. American Sociological Review, 52(1), 44–59. Fligstein, N. (1990). The transformation of corporate control. Cambridge: Harvard University Press. Fraser, J. A. (2002). White collar sweatshop: The deterioration of work and its rewards in corporate America. NY: Norton. Freeman, R. B., & Rogers, J. (1999). What workers want. Ithaca, NY: ILR/Cornell and Russell Sage Foundation. Frenkel, S., Korczynski, M., Shire, K., & Tam, M. (1999). On the front line: Organization of work in the information economy. Ithaca, NY: ILR Press. Gallie, D. (1978). In search of the new working class: Politics and technology in the capitalist enterprise. Cambridge: Cambridge University Press. Godard, J. (2001). High performance and the transformation of work? The implications of alternative work practices for the experience and outcomes of work. Industrial and Labor Relations Review, 54(4), 776–805. Godard, J., & Delaney, J. T. (2000). Reﬂections on the high performance paradigm’s implications for industrial relations as a ﬁeld. Industrial and Labor Relations Review, 53(3), 482–502.

Theorizing Teamwork under Contemporary Capitalism

23

Graham, L. (1995). On the line at Subaru-Isuzu. Cornell: ILR. Grenier, G. (1988). Inhuman relations: Quality circles and anti-unionism in American industry. Philadelphia: Temple University Press. Guillen, M. (1994). Models of management: Work, authority, and organization in a comparative perspective. Chicago: University of Chicago Press. Hackman, J. R., & Wageman, R. (1995). Total quality management: Empirical, conceptual and practical Issues. Administrative Science Quarterly, 40, 309–342. Harley, B. (2001). Team membership and the experience of work in Britain: An analysis of the WERS98 data. Work, Employment & Society, 15(4), 721–742. Hirschhorn, L. (1984). Beyond mechanization. Cambridge, MA: MIT Press. Hochschild, A. (1983). The managed heart: The commercialization of human feeling. Berkeley: University of California. Hodson, R. (1998). Organizational ethnographies: An underutilized resource in the sociology of work. Social Forces, 76, 1173–1208. Hodson, R. (2001). Dignity at work. Cambridge: Cambridge University Press. Hunter, L., MacDufﬁe, J. P., & Doucet, L. (2002). What makes teams take? Employee reactions to work reforms. Industrial and Labor Relations Review, 55(3), 448–472. Jacoby, S. M. (1985). Employing bureaucracy: Managers, unions, and the transformation of work in American industry, 1900–1945. New York: Columbia University Press. Kalev, A. (2004, August 11–16). Cracking the glass cages? Work teams, cross-training and the sex and racial diversity of management. Paper presented at the annual meeting of the American Sociological Association, San Francisco. Kalleberg, A. L., Reskin, B. F., & Hudson, K. (2000). Bad jobs in America: Standard and nonstandard employment relations and job quality in the United States. American Sociological Review, 65(2), 256–278. Kalleberg, A. L., Reynolds, J., & Marsden, P. V. (2003). Externalizing employment: Flexible stafﬁng arrangements in U.S. organizations. Social Science Research, 32, 535–552. Kelley, M. R. (1986). Programmable automation and the skill question: A reinterpretation of the cross-national evidence. Human Systems Management, 6(3), 223–241. Klein, J. (1994). The paradox of quality management: Commitment, ownership, and control. In: C. Heckscher & A. Donnellon (Eds), The post-bureaucratic organization: New perspectives on organizational change (pp. 178–194). Thousand Oaks, CA: Sage. Kuhlmann, M., Schumann, M. (2001). What’s left of workers’ solidarity? Workplace innovation and workers’ attitudes toward the ﬁrm. In: S. P. Vallas (Ed.), Research in the Sociology of Work 10: The transformation of work. NY: JAI. Kunda, G. (1992). Engineering culture. Cambridge: MIT Press. Kunda, G., & Ailon-Souday, G. (2005). Managers, markets, and ideologies: Design and devotion revisited. In: S. Ackroyd, R. Batt, P. Thompson & P. S. Tolbert (Eds), The Oxford handbook of work and organization (pp. 200–219). New York: Oxford University Press. McLoughlin, I. P., Badham, R. J., & Palmer, G. (2005). Cultures of ambiguity: Design, emergence and ambivalence in the introduction of normative control. Work, Employment, and Society, 19(1), 67–90. Milkman, R. (1997). Farewell to the factory. Berkeley: University of California Press. Noble, D. (1984). Forces of production. New York: Knopf. Ollilainen, M., Rothschild, J. (2001). Can self-managing teams be truly cross-functional? Gender barriers to a ‘‘new’’ division of labor. In: S. P. Vallas (Ed.), Research in the Sociology of Work, 10. The transformation of work. NY: JAI.

24

STEVEN PETER VALLAS

Osterman, P. (2000). Work reorganization in an era of restructuring: Trends in diffusion and effects on employee welfare. Industrial and Labor Relations Review, 54(2), 179–196. Parker, M., & Slaughter, J. (1994). Working smart: A union guide to participation programs and re-engineering. Detroit: Labor Notes. Pil, F. K., & MacDufﬁe, J. P. (1996). The adoption of high involvement work practices. Industrial Relations, 35(3), 423–455. Piore, M., & Sabel, C. (1984). The second industrial divide. New York: Basic Books. Poutsma, E., Hendrickx, J., & Huijgen, F. (2003). Employee participation in Europe: In search of the participative workplace. Economic and Industrial Democracy, 24(1), 45–76. Prechel, H. (1994). Economic crisis and the centralization of control over the managerial process: Corporate restructuring and neo-fordist decision-making. American Sociological Review, 59(5), 723–745. Schoenberger, E. (1997). The cultural crisis of the firm. London: Blackwell. Sennett, R. (1998). The corrosion of character: The personal consequences of work in the new capitalism. NY: Norton. Sewell, G. (1998). The discipline of teams: The control of team-based industrial work through electronic and peer surveillance. Administrative Science Quarterly, 43(2), 397–428. Shaiken, H., Lopez, S., & Mankita, I. (1997). Two routes to team production: Saturn and Chrysler compared. Industrial Relations, 36(1), 17–45. Smith, V. (1990). Managing in the corporate interest: Control and resistance at an American bank. Berkeley: University of California. Smith, V. (1997). New forms of work organization. Annual Review of Sociology, 23, 315–339. Smith, V. (2001). Crossing the great divide: Worker risk and opportunity in the new economy. Ithaca, NY: ILR/Cornell University Press. Stark, D. (2001). Ambiguous assets for uncertain environments: Heterarchy in post-socialist ﬁrms. In: P. DiMaggio (Ed.), The twentieth century firm: Changing economic organization in international perspective (pp. 69–104). Princeton: Princeton University Press. Storey, J. (1985). The means of management control. Sociology, 192, 193–211. Thomas, R. J. (1994). What machines can’t do: Politics and technology in the industrial enterprise. Berkeley: University of California Press. Vallas, S. P. (1999). Rethinking post-Fordism: The meanings of workplace ﬂexibility. Sociological Theory, 17(1), 68–101. Vallas, S. P. (2003a). Why teamwork fails: Obstacles to workplace change in four manufacturing plants. American Sociological Review, 68(2), 223–250. Vallas, S. P. (2003b). The adventures of managerial hegemony: Teamwork, ideology, and worker resistance. Social Problems, 50(2), 204–225. Vallas, S. P. (2006). Empowerment redux: Structure, agency, and the re-making of managerial authority. American Journal of Sociology, 111(6). Weeks, J. (2004). Unpopular culture: The ritual of complaint in a British bank. Chicago: University of Chicago Press. Womack, J., Jones, P., & Roos, D. (1990). The machine that changed the world. New York: Rawson. Zuboff, S. (1988). In the age of the smart machine. New York: Basic.

ORGANIZATIONAL NARRATIVES AND THE CONSTRUCTION OF RESISTANCE: UNION ORGANIZING AT A ‘‘PROGRESSIVE’’ COMPANY Joshua L. Carreiro ABSTRACT This case study of a union organizing drive at a Whole Foods Market, the world’s largest natural and organic foods supermarket chain, considers the impact of the company’s employee participation scheme and the accompanying organizational narrative on the outcome of the unionization effort. By relying on the language and symbolism of the existing organizational narrative, union organizers were able to give meaning to their movement, but not without limiting the movement’s potential for significant change and success. Ultimately, their efforts served to reinforce the organizational narrative and the existing employee participation scheme, not transform it. Based on this case study, I argue that organizational narratives are an important location of organizational control.

Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 25–53 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16002-1

25

26

JOSHUA L. CARREIRO

INTRODUCTION Organizations tell stories. Company CEOs, members of management, employees, the news media, and others related internally and externally to the organization relay stories about organizational life. The stories, or narratives, serve to represent the organization to the external environment in order to gain legitimacy (Meyer & Rowan, 1977), and also to ensure social order and control within organizations (Wilkins, 1983). Additionally, stories give meaning to organizational behavior and involvement and are a significant means to replace individual memories and understanding with collective memory and comprehension (Boje, 1991). Stories people tell within organizational settings are often representative of ‘‘the values participants share, the social prescriptions concerning how things are to be done, and the consequences of compliance or deviance’’ (Wilkins, 1983, p. 82). Also, organizational stories ‘‘legitimate the power relations within the organization; they rationalize existing practices, traditions, and rituals; and they articulate through exemplars the philosophy of management and the policies which make the organization distinctive’’ (Martin & Powers, 1983, p. 97). This paper will explore the use of narratives at Whole Foods Market (WFM), the world’s largest natural and organic foods supermarket chain with over 170 stores and 32,000 employees. By analyzing a counter-movement within the organization (a union organizing drive that occurred between 2002 and 2003 at a WFM in Madison, Wisconsin) as well as the counter-narratives of the movement, I illustrate the power of organizational narratives in shaping organizational membership and meaning. Additionally, I argue that the union organizers’ reliance on existing and dominant organizational narratives limited the possibilities for organizational resistance. Despite being a counter-movement, the union organizing drive utilized the language and framework of the dominant organizational narrative, simultaneously making their efforts and intentions intelligible to coworkers, consumers, and the community while constraining their own movement narrative and structure. Davis (2002) argues that, ‘‘even when a movement narrative deﬁes a dominant social narrative it remains subject to other narrative expectations in the wider culture, which constrain and even foreclose movement understandings and courses of action’’ (pp. 25–26). And, as Polletta (2002) suggests, ‘‘activists’ very understandings of ‘strategy,’ ‘interest,’ ‘opportunity,’ and ‘obstacle’ may be structured by the oppositions and hierarchies that come from familiar stories’’ (p. 37). Ultimately, the union’s criticism of the company and its construction of meaning around

Organizational Narratives and the Construction of Resistance

27

union representation were weakened because it adhered to the company’s discourse and symbolism. While I cannot argue that the unionization effort failed because of the union organizers’ reliance on dominant organizational narratives – an argument that would require more varied data than I was able to obtain – I do argue that the union’s reliance on the WFM narrative contributed to the inability of unionization to take root at WFM. As such, the WFM unionization drive serves as an illustrative example of an organizing strategy that union organizers may want to avoid or at least reconsider. This case study illustrates a union’s inability to challenge the progressive rhetoric of a large, well-established company. At WFM, union organizers sought to align unionization with the dominant corporate narrative by arguing that union representation would strengthen the company and bring it closer to its professed progressive ideals. However, because the union’s critique of the company adhered to the framework of the dominant organizational narrative, company leadership was able to discount the union organizers’ claims and co-opt the counter-narrative and its accompanying critique. Perhaps the creation of a unique language and critique, which utilized a labor-based framework as opposed to a company-based framework, would not have been so easily co-opted by WFM leadership. This case study is also important because it documents and analyzes a union organizing drive in the interactive service economy (Leidner, 1993). If the United States labor movement is to expand its base, it must make gains in this growing sector of the economy. More specifically, however, this study documents a unionization effort in a company that professes a commitment to progressive principles and democratic practices, but remains antithetical to union representation for its employees. As such, it highlights strategies and responses that may be typical of other businesses that promote progressive or democratic ideals in the way the company is run or the products they sell. Because companies like WFM emphasize their afﬁnity with progressive values and causes as a way to promote their business, they are both vulnerable to criticism that they are not adhering to their own rhetoric and protected from criticism concerning workplace conditions. WFM’s claims about worker empowerment and support for environmental issues encourages critics to scrutinize the supermarket’s practices, but also grants the company considerable leverage in the public eye when WFM falls short of its own standards. For example, WFM management and supporters attempted to sidestep accusations of worker exploitation by arguing that despite the company’s shortcomings, it remained more progressive than

28

JOSHUA L. CARREIRO

comparable supermarkets. Thus, businesses like WFM present unique union organizing opportunities as well as challenges. After a brief discussion of research methods, I analyze the WFM organizational narrative and its relation to the team-based organization of the company. Next, I analyze the union’s counter-narratives and their relation to the WFM organizational narrative. I suggest that possibilities for organizational resistance were limited by the systems of meaning created through organizational storytelling. After that, I illustrate the various ways that the company’s organizational narrative, in addition to providing meaning, structured organizational order and control and I conclude with an analysis of the resolution of the unionization effort.

RESEARCH METHODS The analysis of the WFM organizational narrative that follows is based on the company’s detailed mission statement available publicly on the corporate website (http://www.wholefoods.com/company/philosophy.html) and in the company’s employee handbook, the General Information Guide. Additionally, information was gathered from newspaper sources, press releases, and letters sent by various members of the WFM corporate staff, including the CEO and founder John Mackey, to employees and consumers. I was able to gain access to corporate letters, as well as union letters, public statements, media interviews, and newspaper sources related to the union counter-narrative through the extensive archival database maintained by the Madison union organizing committee available via their website (http://www.wholeworkersunite.org) and through the Madison Independent Media Center (http://.madison.indymedia.org/info/display/whole_foods/index/php). Also, I conducted telephone interviews with three members of the Madison WFM union organizing committee who were also employed at WFM during the organizing drive. The interviews took place during the Fall of 2004, after the union organizing drive had ended. Unfortunately, I was not able to conduct interviews with employees who were also not members of the union organizing committee, nor was I able to gain access to members of WFM management for interviewing. My interview data come from two sources (personal interviews and archived media interviews) and I distinguish between the two types throughout the text. When referencing an interview that I conducted, I indicate the name of the respondent. However, when referencing an interview that I did not conduct, I cite the appropriate source of the interview. All radio station

Organizational Narratives and the Construction of Resistance

29

interviews, as well as press conferences are archived at the Madison Independent Media Center website.

WHOLE FOODS MARKET ORGANIZATIONAL NARRATIVE The philosophy and core values of WFM constitutes a narrative, or story, that the company tells to employees, customers, investors, and other organizations in its ﬁeld about what WFM is and does. The company’s dominant narrative is based on the often-recounted story of the origins of WFM. According to this tale, WFM began as the brainchild of CEO and founder John Mackey in 1978. Mackey opened his ﬁrst natural foods market, Safer Way, in a Victorian mansion near downtown Austin, Texas after dropping out of college and joining a vegetarian-housing cooperative. As the story goes, Safer Way sought to bring natural foods to average consumers by combining natural food products with more mainstream supermarket offerings. In addition, Mackey eschewed the ‘‘small is beautiful’’ ethos of the natural food cooperatives and grocers that surrounded him and eventually opened a second store, which was more than four times the size of his nearest competitor. WFM is a for-proﬁt business. And while Mackey never intended WFM to be a worker-owned cooperative, he does claim to promote democratic principles and progressive values through the structure of the company and the products sold. However, despite these claims, WFM more closely resembles a traditional hierarchical ﬁrm than a truly democratic worker-owned cooperative. For example, employees are provided avenues to make suggestions and provide feedback, but decision-making power is held by executives and managers. WFM has eventually grown, through a series of acquisitions and development, to more than 170 stores in the United States and the United Kingdom (as of 2005). However, the origination narrative of WFM roots the company in Mackey’s own story of perseverance and ingenuity, as well as the democratic and progressive values of the natural foods cooperatives and grocers that inspired WFM and have often been acquired or put out of business by the company. WFM’s organizational narrative is developed more formally in the company’s elaborate mission statement – perhaps the most direct presentation of the WFM story. The company’s ‘‘Whole Philosophy’’ mission statement is publicly available on the WFM website and within the handbook provided to all new hires during orientation. ‘‘The Whole Philosophy’’ contains the

30

JOSHUA L. CARREIRO

company’s ‘‘Declaration of Interdependence,’’ which lists the company’s ‘‘core values’’ that are the ‘‘underpinning ofycompany culture’’ which ‘‘transcendysize and our growth.’’ According to the employee handbook and company website, ‘‘These core values are the soul of our company’’ (emphasis added). The employee handbook, more generally, provides a narrative of the company. According to CEO and founder John Mackey, ‘‘It helps us give people a sense of the past, the present, and the futurey(and) how we tell them about the game we’re playing’’ (quoted in Posner, 1989, p. 84). The WFM motto, stated in the ‘‘Declaration of Interdependence,’’ deﬁnes the company by its relationship to the concerns of its customers, employees, and community, especially concerns over the environment: Our motto – Whole Foods, Whole People, Whole Planet – emphasizes that our vision reaches far beyond just being a food retailer. Our success in fulﬁlling our vision is measured by customer satisfaction, Team Member excellence and happiness, return on capital investment, improvement in the state of the environment, and local and larger community support.

The set of core values promoted by WFM signiﬁes the company’s purported essential nature or, in the company’s own words, its soul. The core values are divided into ﬁve subheadings that emphasize the quality and organic nature of the products sold at WFM, the company’s commitment to ‘‘satisfying and delighting’’ their customers, an empowering work environment, proﬁt and growth, and the company’s concerns and care for the community and environment. These values constitute the substance of the WFM organizational narrative and highlight the company’s commitment to abstract ideals of environmental improvement, employee empowerment, and community citizenship. The abstract nature of many of the values works to buffer the claims from attacks on their status as truth. Ewick and Silbey (1995) argue that narratives, because they make implicit rather than explicit claims regarding truth, ‘‘elude challenges, testing, or debate’’ (p. 214). The WFM narrative operates on at least three levels. First, the narrative is a commodity that is packaged and ‘‘sold’’ to WFM customers and employees. WFM’s image as an employer that values employee empowerment and participation and as a retailer that is environmentally responsible promotes the company to a certain population of workers and consumers. In addition to marketing speciﬁc products, WFM markets its philosophy, values, and a lifestyle as a product for consumption. Second, the narrative contributes to the construction of the organizational reality as employers and consumers experience it. In her research on

Organizational Narratives and the Construction of Resistance

31

organizations and narratives, Witten (1993) argues that recent organizational analyses have emphasized how, through talk and other symbolic behaviors, ‘‘people in organizations produce a shared, intersubjective understanding of the nature of their reality; provide themselves with a scheme for making sense of that reality; and, importantly, objectify these understandings so that the nature of their constructed environment appears as ‘real’ to them’’ (p. 100). However, not all organizational members contribute to the creation or the meaning of organizational narratives equally. Power differentials in the organization, based on position, number, access to information and the dissemination of information, certainly hinder or assist the construction of meaning through narratives in organizational settings, as the case of the Madison union drive illustrates. Finally, narratives function as a form of organizational control. Narratives provide ‘‘models of correct behavior and rules for the extension of the models to new situations’’ (Witten, 1993, p. 105), which supply organizational members with scripts to follow in absence of direct supervision and overt control. Narratives also ‘‘impart values that affect problem definition’’ (Witten, 1993, p. 105). In this sense, narratives are illustrative of what Lukes (1974) termed the ‘‘third dimension’’ of power, which accounts for the ability of institutions to shape the interests of individuals in such a way as to keep conﬂict and competing interests latent and limit what may be named as problematic. In addition, narratives can create an ‘‘invented tradition’’ of the organization that teaches members about the futility of resistance. This invented tradition controls by ‘‘illustrating the rule of anticipated reactions, promoting an avoidance of tests of strength’’ (Witten, 1993, p. 112; emphasis in the original).

NARRATIVES ABOUT EMPLOYMENT AND COMMUNITY CITIZENSHIP The company’s narrative of employment emphasizes that work at WFM is both fulﬁlling and empowering and thus more than just a paycheck. WFM is not necessarily unique among retail companies in its emphasis on extramonetary work incentives. Other major retailers, such as Wal-Mart, also emphasize the beneﬁts of their ‘‘unique’’ workplace culture to employees. However, the extent to which WFM publicizes its commitment to worker empowerment and participation is extraordinary. On the ﬁrst page of the company’s website, and within the pages of its employee handbook, WFM touts its empowerment of workers to ‘‘make their own decisions’’ through

32

JOSHUA L. CARREIRO

organization in self-directed work teams. In this way, WFM differs from countless other interactive service work employers that promote worker empowerment, but mirror Fordist production techniques in their routinization of the labor process. Given the high- turnover rates associated with retail work, routinization, and the interchangeability of workers tend to be the norm (Garson, 1988; Leidner, 1993). Contrary to Fordism, WFM stresses the incorporation of employees into the organizational culture of the company, not just its labor process. The company states in their ‘‘Declaration of Interdependence’’: Our success is also dependent upon the collective energy and intelligence of all our Team Members. In addition to receiving fair wages and beneﬁts, belief in the value of our work and ﬁnding fulﬁllment from our jobs is a key reason we are part of Whole Foods Market. Achieving unity of vision about the future of our company, and building trust between Team Members is a goal of Whole Foods MarketyWe aim to cultivate a strong sense of community and dedication to the company.

These goals further accentuate the intention of WFM management to make the more banal aspects of the employment relationship, like beneﬁts and wages, secondary concerns. Instead, attention is focused on extra-monetary motivations for membership in the corporation, which emphasize the culture of the organization, including its values as expressed in the dominant organizational narrative. The gap between the ideals of WFM and the reality of some of its employees is documented in a Washington Post expose´ on the introduction of a WFM to Washington D.C. neighborhood undergoing significant gentriﬁcation (Hull, 2001). The author interviews a WFM employee who lives in a homeless shelter yet reads in her employee handbook that ‘‘Drawing a paycheck is nice, but it’s not the whole meaning of life. We believe we offer each of our Team Members an opportunity to fulﬁll a higher purpose; helping to make the world a better place’’ (Hull, 2001, p. 19). WFM is not directly responsible for this employee’s lack of permanent housing, but this vignette highlights the disconnect between the company’s professed ideals and the material need of some WFM employees. The WFM culture is built upon a narrative that diminishes distinctions in status based on organizational position by using the metaphor of team membership to describe employment relations and stresses the company’s policy of open communication. WFM workers are considered ‘‘Team Members’’ instead of employees. From checkout clerks and baggers to the Administrative Team, all who work at WFM are labeled Team Members

Organizational Narratives and the Construction of Resistance

33

and are organized into ‘‘self-directed’’ work teams. Each store is organized around a Store Leadership Team, Team Leaders, and Team Members. This language allows WFM to avoid evoking divisions between management and employees and suggests a degree of equality and the existence of shared goals. In her analysis of the team concept at a US automobile plant, Graham (1995) found that the company tried to ‘‘connect the worker to something greater than simply holding down a job’’ and sought to motivate the workers to ‘‘involve themselves in the job as they would in volunteer activities’’ (p. 95). Companies using the team concept appropriate worker solidarity and support by formalizing work groups and circumventing the spontaneous formation of small, informal work groups (Graham, 1995). At WFM, the use of teams to organize workers effectively creates a sense of cohesion within particular work groups, but encourages competition and division between groups, which inhibits storewide commonality and solidarity. One employee and union organizer, Michael,1 describes teams at WFM as a ‘‘family within a family.’’ I think that’s what they were trying to do (create divisions between groups of workers). This was the function of our team, these are the goals of our teamyyou would have a team meeting and it would be just your team and your team leader would talk about things related to your team and you would sit with your team in the store meetings and cheer for your team. They really tried to create this ‘‘we’re this group; we’re this family within a family.’’

Team membership symbolically implies that employees of the company are not ‘‘working for’’ or ‘‘employed by’’ WFM, but equal members of the organization. Levels of authority and the differences in pay and power that accompany them are downplayed. According to the ‘‘Declaration of Interdependence’’: We strive to build positive and healthy relationships among Team Members. ‘‘Us versus them’’ thinking has no place in our company. We believe that the best way to do this is to encourage participation and involvement at all levels of our business.

WFM’s policy regarding pay states that no executive may earn, through salary and bonuses, more than 14 times what the average employee earns. In addition to salary and bonuses, however, executives receive stock-option grants every year, which are not capped. Between 1994 and 2003 executives received between 5,000 and 10,000 shares of stock/year (Blackman, 2004). The current salary and bonus cap is set at $409,000 (Blackman, 2004), which means that the average hourly wage of WFM non-executives working 40 h/ week is roughly $15 per hour, or about $29,000 per year. However, this average wage accounts for all non-executives, which includes supervisors

34

JOSHUA L. CARREIRO

and managers, who likely earn a higher wage, as well as frontline workers, who likely earn a lower wage. According to my informants, starting employees at WFM tend to earn between $7 and $8 per hour, which amounts to between $13,400 and just over $15,000 per year for an employee working 40 h/week. At the time of the organizing drive, WFM offered health beneﬁts to full-time workers, but with considerable premiums. Since the organizing drive, however, WFM has begun offering health beneﬁts to employees without high premiums after 800 work hours. WFM attempts to further diminish the appearance of hierarchy within the company by emphasizing open communication policies through Team Member Forums and Advisory Groups and what the employee handbook refers to as ‘‘open book, open door, and open people practices.’’ Team Member Advisory Groups, or TMAGs, are storewide committees made up of elected representatives from each individual Team (e.g., bakery, produce, etc.) in the store, which hold periodic meetings to discuss storewide issues and make recommendations based on Team Member input. Team Member Forums occur once annually and are an opportunity for all Team Members to address Store Team Leaders and Executive Team Members. In many ways these forums and advisory groups resemble the organization of cooperatively owned and collectively run businesses. However, unlike their democratic counterparts, WFM has no obligation to heed the concerns or recommendations of the TMAGs, making the open communication policies more of a symbolic gesture than anything else. In reality, TMAGs and Team Member Forums are more about feedback than democratic decision making. Initially, some workers at the Madison store were hesitant to support the union organizing drive because they felt all existing avenues for addressing worker concerns, like the TMAGs, had not been utilized. However, once workers learned that their coworkers had tried to use conventional communication structures to voice their concerns to management, they became increasingly skeptical of the WFM narrative and interested in unionization. Michael, a worker who initially shunned the union organizing campaign but later became an active member of the organizing committee, describes this transformation: As we started organizing and ﬁnding out that indeed people had tried to become members of TMAG or tried to make changes through TMAG or had contacted the regional president or were contacting the store team leader and nothing was changingythat was kind of disappointing because you realize maybe this wasn’t going to be taken as seriously as the company said. Also, it looks very good to have groups of workers come together to represent the workers, but when it comes down to it, the company chooses to listen to or not to listen to those workers. So you could ask over and over and over again

Organizational Narratives and the Construction of Resistance

35

to have your work environment change and if the company just decides because they don’t want to or if they just decide because it is not ﬁscally profitable to do that, then they’ll just simply choose not to.

Michael, like other Team Members, believed in the story WFM constructed until he recognized its inconsistencies. The organizational narrative emphasizes the similarity between work and play at WFM and highlights alternative rewards to employment, in addition to pay. According to the employee handbook, the company avows a ‘‘commitment to makey jobs more fun by combining work and play and through friendly competition to improveystores.’’ However, ‘‘friendly competition’’ and the combination of work and play, in tandem with the organization of work by teams, heighten the presence of coworker control. In a study of a company that utilizes self-directed work teams, Barker (1993) found that through competition and self-direction ‘‘teams had created an omnipresent ‘tutelary eye of the norm,’ with the team members themselves as the eye, that continually observed their actions, ready to either reward or, more importantly, punish’’ (p. 432). Under team organization, self-discipline and peer pressure among coworkers replace overt control by management. Direct control is still present, but it exists between Team Members as well as between employer and employee. WFM’s CEO and founder John Mackey describes the control operating at the store level as the antithesis of hierarchy and bureaucracy. Instead, he states, ‘‘We have lots of self-examination going on. Peer pressure substitutes for bureaucracy. Peer pressure enlists loyalty in ways that bureaucracy doesn’t’’ (quoted in Fishman, 1996, p. 2). The manager of a WFM in Massachusetts reiterates Mackey’s emphasis on peer pressure and illustrates the type of control described by Barker. In his own words: There are people who are really good about working when the manager is on the ﬂoorybut as soon as the manager disappears, they lose controlyI’m not the one you need to impress. It’s your fellow team members. And they will be as tough as they can be. (quoted in Fishman, 1996, p. 3)

Much like Foucault’s (1977) description of the Panopticon, self-directed work teams create a ‘‘state of conscious and permanent visibility that assures the automatic functioning of power,’’ makes surveillance ‘‘permanent in its effects, even if it is discontinuous in its action,’’ creates and sustains power relations ‘‘independent of the person who exercises it,’’ and places workers in a ‘‘power situation of which they are themselves the bearers’’ (p. 201). Like the Panopticon, management by self-directed work teams arranges ‘‘things in such a way that the exercise of power is not added on

36

JOSHUA L. CARREIRO

from the outside, like a rigid, heavy constraint, to the functions it invests, but is so subtly present in them as to increase their efﬁciency by itself increasing its own points of contact’’ (Foucault, 1977, p. 206). As power is dispersed, so is the ﬂow of narratives throughout the organization, leading to the concealment of the origination of the narrative and a ‘‘polyvocality that inoculates and protects the master narrative from critique,’’ which makes the hegemony appear varied and even contradictory (Ewick & Silbey, 1995, p. 212). WFM also constructs narratives about its relationship to the community and environment, which emphasize the company’s commitment to local and organic food suppliers, sustainable agriculture, environmentally sound practices, and community citizenship. Because WFM specializes in expensive products, the company markets an image of healthy living and environmental concern as much as the products themselves. Therefore, WFM maintains customer loyalty largely through the company’s espoused values and its relation to consumer identity. In many ways, WFM presents itself as being similar to democratic food cooperatives, despite being a for-proﬁt company. Because of this, consumer expectations of WFM may be higher than typical supermarkets. Consequently, the union organizing committee sought to take advantage of consumer expectations as part of their organizing strategy. The WFM narrative is thus constructed in the triadic relationship between WFM management, Team Members, and consumers. As the union organizing drive progressed, management attempted to depict the movement as contrary to both the company and consumers’ ideals. Through communications to Team Members, public statements to the media, and letters to customers directly WFM management hoped to maintain their public image in the face of the union’s challenge. Union organizers’ stories of the organizing drive, however, attempted to ﬁt the movement within the ideals of the company in an effort to make unionization intelligible to customers. According to one of the lead organizers and WFM employee, Kathryn, the decision to incorporate the WFM narrative was a deliberate attempt to sway those invested in the WFM narrative, workers and consumers alike: We really had to carefully and thoughtfully maneuver a point – and I don’t mean it in a manipulative way – but reaching those people whoybelieve in the WFM philosophy and the team environment. Those are the people you are going to have to reach in order to have an effective organizing drive. And so for us we wanted those people on our side and there was no better way to do that than to say, ‘‘these are Whole Foods Market’s very own stated philosophies and values and this is the reality so there is clearly a huge

Organizational Narratives and the Construction of Resistance

37

disconnect and the only way to solve that disconnect or to work on that disconnect is through unionizing.’’ And so that was the approach we took.

Unionization was promoted as a way to create narrative consistency out of the multitude of stories communicated by WFM management, Team Members, and consumers, not as a direct challenge to those stories.

ORIGINS OF THE MADISON UNION ORGANIZING DRIVE According to members of the union organizing committee, Team Members at the Madison WFM became interested in being represented by the United Food and Commercial Workers (UFCW) because of erratic raise schedules, unilateral changes in company policy, and diminishing opportunities for effective communication between Team Members and management. While seemingly disparate issues, the initial concerns that led to the union organizing campaign were all related to inconsistencies in the WFM narrative that structure the WFM employment relation. The company’s unpredictable raise schedules, its unilateral policy implementation, and a breakdown in worker–management communication channels signaled to workers that the stories of employment at WFM, that they themselves helped create, did not necessarily match many of the employees’ own experiences. Thus, the counter-narratives constructed by the union organizing campaign, as illustrated in the previous quote, emphasized the ability of union representation to create consistency between the stories told by WFM management and Team Members to coworkers and consumers. Team Members at the Madison store felt that raises were granted in an unfair and haphazard manner. According to union organizers, teams with high-labor costs were routinely denied raises regardless of individual Team Member performance or seniority. In addition, decisions about raises were too often made in a subjective manner by Team Leaders, WFM’s term for department managers. Michael, an employee and member of the union organizing committee, describes the process this way: After so many months you basically ﬁll out a form and then get down on your knees and try to convince your boss to give you a raise and then they kind of arbitrarily assign you anywhere between a quarter and a dollar depending on how much money your team is making or your performance. It’s very arbitrary. So I mean, if your Team Leader doesn’t like you, they can say you are not being very productive and they are not going to give you a raise at all. And there is no recourse.

38

JOSHUA L. CARREIRO

Union supporters attributed the lack of a consistent and transparent raise policy to management’s ability to implement store policy unilaterally and thus viewed unionization as a solution. Concern about management’s ability to implement store policy without seeking Team Member input was not limited to raises. A new dress code that prohibited facial piercings, artiﬁcially colored hair, or logos on clothing, put into practice without consultation of Team Members, irritated many employees and led them to question the Madison store leadership’s commitment to the WFM narrative concerning inclusion, democracy, and diffusion of power. For the minority directly impacted by the policy change, the new dress code was viewed as a personal attack. However, for many others, the policy change was viewed as a cautionary tale and cause for concern. Kathryn, an employee and member of the union organizing committee, explains: For some of usyit was really a direct assault on us because we were the few people remaining who did have piercings and weird hair. But for other people it was like some sympathy for us and also the realization that this was one thing that WFM was known for – not coming down on workers and being an inclusive place for people to look how they want and dress how they want and whatever. And so it was an all too real indication that the company was changing from its roots. There are some of us who don’t really believe they had those philosophical values really upheld in the ﬁrst place, but then people who did believe in those started to see through that and see that the company was changing.

Because WFM relies to such a great extent on narrative for organizational coherence, consistency is essential to its organizational structure. According to the union organizers, Team Members were sensitive to experiences that did not ﬁt within the story that they were being told and helped to create about their employment. Michael articulates this sensitivity well: They were changing the dress code policy and their whole sort of thing was like ‘‘we’re democratic and we like your input and we’re a team.’’ And so this (the new dress code) was all sprung on us without anyone notifying us that this was going to happen. And part of what they say is that ‘‘we’re a different company and we honor diversity’’ and so they were changing that y and that didn’t really affect a lot of people, but it just pissed people off because they made no effort to talk to us about it.

The dress code became a cautionary tale about the structure of WFM employment that Team Members told to each other. Later, this narrative inconsistency, like others, would be incorporated into the union narrative. Kathryn explains how union supporters built off of the WFM narrative when constructing their own story: It (telling stories of organizational change) was another way to show people that we didn’t have any power and if they come to us and implement policies like this they can

Organizational Narratives and the Construction of Resistance

39

really do anything and change anything they want to. We tried to use it like an educational tool to show people that this is what happens when workers aren’t unionized – you don’t have any negotiating power at all.

For many, targeting the company’s narrative inconsistencies seemed an obvious choice as a union strategy, especially after the union organizing campaign began and WFM management actively sought to disrupt the organizing efforts. Michael explains: This company, making these gigantic, public statements about how great they are and then turn around and spend a lot of money to try and stop their workers from organizing, (which) does not look good for the company. We just brought that to people’s attention – we were truthful, we didn’t create a situation that wasn’t really happening, we just brought it to people’s attention that this is what the company says and this is what the company does.

Union organizers also emphasized WFM’s narrative inconsistencies in the stories they told to the media and consumers about their efforts to unionize. According to David, another employee and member of the union organizing committee: It just makes for better media when you have a conﬂict, like a hypocrisy, it makes for betterystoriesy They were an institution in the part of the city they were in – natural food shoppers were like, ‘‘it’s WFM!’’ So when they started doing something bad it generates attention.

Because WFM markets an image as well as products, narrative consistency is important to the manipulation of consumers as well as employees. Union organizers thus sought to gain consumer support by emphasizing WFM narrative inconsistency and constructing narratives of unionization that emphasized the union’s ability to maintain WFM’s integrity as a socially responsible retailer. For example, the president of the UFCW local that aided WFM workers in their organizing effort referred to WFM as ‘‘the natural foods Wal-Mart’’ and suggested to consumers that they, like WFM employees, were ‘‘getting ripped off’’ (Gaus & White, 2003, p. 5). Because the union organizers’ strategy relied so heavily on creating narrative consistency, they were forced to engage with existing WFM discourse and symbolism. While most of the union organizers’ emphasis on the company’s dominant organizational narrative was critical – for example, they identiﬁed narrative inconsistencies – much of the organizers’ narrative focused on articulating the compatibility between unionization and WFM’s philosophy and core values. By borrowing WFM discourse and symbolism, the union organizers were able to tap into pre-established systems of meaning and convey unionism, which may have been foreign to many of the employees

40

JOSHUA L. CARREIRO

and consumers, in familiar terms. However, because the union organizers’ told a story of unionization that attempted to synthesize union membership and WFM’s philosophy and core values, the counter-narrative never successfully challenged the WFM narrative, but instead was absorbed by it.

TEAM MEMBER EMPOWERMENT, PARTICIPATION, AND COMMUNICATION Union organizers emphasized that Team Members at the Madison store wished to unionize because of, not despite, a collective commitment to the values of the company. The movement narrative stressed the company’s emphasis on creating an empowered workplace through team membership and participation. In an interview (July 7, 2002) with a Madison radio station, one Team Member expressed this sentiment well. So I think it goes back again to the info that we’re presented with when we’re hired at Whole Foods. We’re all given a packet that states the core values and what we can expect from working at Whole Foods and a lot of us like what we see on paper; we just want to make sure that it’s actually carried out in our employment at Whole Foods.

In a letter (May 23, 2002) written by the union organizing committee announcing its organizing effort, these feelings were echoed: Indeed, many of us enjoy our jobs and would like to continue working for the company. However, we want to make Whole Foods Market the kind of store it professes to be in its philosophy and core values. In our decision to unionize, we are upholding the shared company vision – not betraying it, as management might claim. Because this is the only option for true worker empowerment, we ask that our wishes to unionize be supported. Only with a legal contract will we be afforded and guaranteed a say in the decisions that affect us – and through unionizing we can obtain such a contract.

By afﬁrming that Team Members at WFM were ‘‘upholding the shared company vision’’ through their unionization effort, members of the organizing committee downplayed the notion that their efforts represented a challenge to managerial authority or the team-based organization of the company. Instead, they drew attention to the similarity between unionization and the WFM vision and suggested that union representation is a path to worker empowerment and greater access to participation in creating company policy. The union’s own account of the origins of the organizing effort (posted on http://www.wholeworkersunite.org) articulates this point well:

Organizational Narratives and the Construction of Resistance

41

A contract does not affect the atmosphere in the store – thus, open communication and the overall ‘‘team’’ feeling don’t change when workers unionize. Workers do, however, have the support of a union organization if they are treated unfairly.

Additionally, union organizers used the company’s claims of Team Member empowerment to question WFM contestation of the union election. From the beginning of the organization drive, regional as well as national management fought the entrance of a union into the Madison WFM store. From not accepting the union support cards signed by a majority of workers, to holding several mandatory all-worker meetings in which management criticized the union, and contesting the National Labor Relations Board (NLRB) union election results, WFM management tried to dissuade Team Members from voting union. According to Michael, an employee and member of the union organizing committee, WFM management tried to persuade Team Members from voting in favor of the union by holding captive meetings and deﬁning unions as counterproductive businesses with their own ﬁnancial interests: The company spent a lot of time and money making wonderful PowerPoint presentations and glossy handouts explaining how the union was going to screw you. ‘‘You start from zero; you are going to lose all your beneﬁts.’’ They had two main meetings – one of them was ‘‘here is everything you are going to lose if you vote for the union’’ and the second one was ‘‘unions are big business too, this is how much money they make, this is how much the president of the union makes, isn’t that unfair that this guy makes $300,000 a year.’’ Trying to get people riled up aboutyclass stuff.

In a narrative turn, the organizing committee attempted to portray management’s resistance to the unionization effort as discordant with the WFM values of Team Member empowerment and participation and instead as a form of typical managerial control. In an interview with a local radio station (July 2, 2002), a member of the union organizing committee articulated this perspective: One of the most obnoxious things about sitting through these mandatory meetings that Whole Foods is holding for every single employee of the store is to listen to them talk about how well the corporation treats us and how John Mackey (CEO and founder) really does care and what our core values are about – but we have absolutely no impact. If something is wrong in the store we can bring it up with our team leader and it will be ignored and will definitely not reach John Mackey.

The sentiment expressed in the story of the origins of the Madison unionizing effort suggests that unionization is not a threat to the team environment of WFM, but merely a safeguard against the betrayal of the ‘‘team feeling’’ and a contribution to the empowerment of Team Members to participate in their workplace.

42

JOSHUA L. CARREIRO

Union organizers also claimed that unionization is not only compatible with the team environment, but also contributes to the realization of the team ideal. During the same radio interview (July 2, 2002), an employee and member of the union organizing committee argued that the unionization effort has brought workers together for the common cause of improving the company: A lot of the employees that work in the store really believe in the Whole Foods philosophy and the core values that they profess. The effort is about making the store a better place to work for everyone so this is really something that we see as a positive change. A lot of the workers in the store have been getting to know each other for the ﬁrst time and are really uniting together to make a stand and make Whole Foods a better place to work for everyone.

WFM management had claimed, as many companies do in the midst of an organizing drive, that the union would only serve to drive a wedge between the various Team Members at the Madison store, something antithetical to the company’s commitment to a team environment. However, the movement’s narrative emphasized the cooperation and collaboration needed to organize the union and that characterizes union membership in general. A ﬁnal important aspect of the movement’s narrative that emphasized the congruence between unionization and the team environment was the assertion that union representation was an almost spontaneous outgrowth of team membership. This sentiment was expressed well by an employee and member of the union organizing committee during a radio interview (July 2, 2002): This has been a 100% internal organization of essentially doing everything that Whole Foods subscribes to in their mission statement and philosophy. That we formed together as a team, in a team environment and an environment of empowerment in order to collectively make our place of employment the best place it could be.

By connecting their organizing drive to the team environment, the Team Members were able to deﬂect management accusations that the movement was sponsored by outsiders, the UFCW, and therefore did not reﬂect the true desires of WFM employees. In a letter to customers explaining the union organizing drive, WFM management stated that ‘‘the interference of an outside party would disrupt our unique culture and (we) believe that we can do a better job if we all work together in the spirit of our core values to improve our work environment.’’ Therefore, union organizers built off of their association of unionization with the team philosophy in an attempt to strengthen their claim that the organizing effort was independent and internally driven.

Organizational Narratives and the Construction of Resistance

43

IMPROVING WHOLE FOODS MARKET AND MAINTAINING THE VALUES OF THE COMPANY Similar to the movement’s claims concerning Team Member empowerment, union organizers also claimed that unionization could help maintain the company’s values, especially in a time of growth. John Mackey, the CEO and founder of WFM, is open about his plans to expand the company worldwide, but tempers that reality by discussing profits and growth in the context of maintaining a commitment to the company’s core values. For example, in the ‘‘Declaration of Interdependence,’’ the company states, ‘‘Whole Foods Market intends to grow. We will grow at such a pace that our quality of work environment, Team Member productivity and excellence, customer satisfaction, and ﬁnancial health continue to prosper.’’ By emphasizing the union’s interest in assisting WFM achieve its corporate vision in the face of growth and change, the movement’s narratives avoided portraying unionization as an attack on the company or its values and reafﬁrmed its commitment to those values. In a radio interview (July 2, 2002), an employee and union organizing committee member expressed this sentiment: But once Whole Foods starts to have a goal of having 600 stores worldwide, in order to do that they are starting to make policy changes regarding just small things like dress codes and beneﬁts policies and discounts and things like that. In order, we feel, to become a worldwide leader in the grocery industry they’re going to continue to make changes to become more mainstream. And what that does is it takes away from everything that Whole Foods was built upon. I mean its foundation and its core values were something completely different than from what mainstream was 22 years ago and that’s what John Mackey (CEO and founder) wrote his mission statement about and that’s what Whole Foods, the soul of the company isy To make more money at the expense of your employees is where we feel Whole Foods has failed and we’re trying to get them to live up to everything that they proscribe to.

Not only does unchecked growth have a negative impact on WFM employees – in the form of policies that affect their everyday work life – but, according to the union narrative, it betrays the company’s core values and consumers. During a radio interview (July 2, 2002), a Madison employee told a story to illustrate the company’s movement away from its core values because of commitment to ‘‘the bottom line’’: Another thing to consider is that while Whole Foods looks great on papery they are not following through on what they are supposed to be doing. I was the produce buyer for a period of time and I was told not to buy from local farmers because it was too expensive and we could get the same kind of produce from our distribution center in

44

JOSHUA L. CARREIRO California for a lower cost. One of the core values is supporting local farmers and we weren’t doing that. Ask any of the buyers if they are encouraged to ﬁnd local farmers and support them. It’s kind of embarrassing to work there because Whole Foods is not beingyif they would do what they said they did it would be one thing, but they are not.

This story illustrates the company’s failure to live up to its commitment to consumers and fulﬁll its stated goal of community citizenship by bypassing local farmers for cheaper produce. Like other stories told by union supporters, this story seeks to align Team Member and consumers’ interests and render unionization a consumer issue. All of these stories, however, warn of the potential impact company growth could have on workers and consumers. For employees, unfettered growth means a shift from the company’s professed commitment to worker empowerment and participation toward a more hierarchical workplace organization that demands uniformity and routinization rather than innovation and diversity. For consumers, a considerably larger company means that WFM will no longer support local agriculture and producers, but will instead look nationally or internationally for the best price. In both cases, union supporters frame company growth as a shift from ‘‘the way Whole Foods Market used to be’’ toward the company’s alignment with its mainstream competitors. And, in both cases, the union is offered as a solution to this narrative inconsistency.

‘‘WHOLE FOODS SUPPORTS THE RIGHTS OF ITS WORKERS TO UNIONIZE’’ In 1996, the year that WFM opened in Madison, the company placed a banner on the front of the store in an effort to allay community criticism that the company was antiunion and therefore not a good ﬁt for the laborfriendly city of Madison, Wisconsin. The banner read, ‘‘Whole Foods Supports the Right of its Workers to Unionize Should They Choose to Do So.’’ Despite being the nation’s second largest non-union grocery chain (WalMart is the largest), and an outspoken critic of union representation (CEO and founder John Mackey has publicly referred to unions as being ‘‘like herpes’’), the banner communicated to Team Members and customers alike that WFM was at least amenable to unionization, if not supportive of organized labor. WFM later changed its position. A letter to customers stated that while the company ‘‘recognized the right of workers to organize and hold a union election’’ they felt the presence of an ‘‘outside party’’ would interfere with the company’s ‘‘unique culture,’’ thus creating a possible paradox for customers who supported WFM because of its unique products

Organizational Narratives and the Construction of Resistance

45

and stated values. Union organizers later utilized the inconsistency and incorporated the 1996 banner into their campaign, as the following quote from an organizing committee press conference illustrates: When the Madison store openedy Whole Foods publicly claimed to respect the right of its employees to unionize; however, thus far they have only discouraged employees from supporting our union. They have held off-premise meetings with supervisors and used mandatory all-store meeting last night to dissuade employees from unionizing. All these are steps that have been detailed in union-busting literature. This is why we are seeking the support of the community to assist in exercising our legal right to unionize.

The organizing committee attempted to align with consumers by relaying a narrative of mutual betrayal. Union representation, according to the organizing committee, could mend the company’s narrative inconsistencies.

‘‘A WORKPLACE BASED ON MUTUAL TRUST, RESPECT AND LOVE’’ In communications to all WFM employees, the executive management of WFM constructed its own story of the origins of the unionization effort and a cautionary tale of what unionization would mean. The communications were an effort by the company to provide a version of the Madison story to WFM workers nationwide and an attempt to illustrate to Madison and nonMadison WFM employees alike that unions do not ﬁt within the organizational narrative of the company. In a letter to all Team Members (May 17, 2002), WFM management reiterated its original organizational narrative in an effort to make union representation unintelligible or outside the framework of the WFM story: At Whole Foods Market, we have worked hard to create a workplace in which you are involved in the decisions that will impact you and empower you to work directly with leadership and with each other to resolve issues and make necessary changes. As you are probably aware, Whole Foods is the leader in the grocery industry for creating unique, innovative work environments based on respect for each other, opportunity for personal and ﬁnancial growth and a fair, fun and enjoyable work place where you are listened to and where you are included in the decisions that determine how this store is run and how it succeeds. You are what make Whole Foods Market great and we believe that a union’s ‘‘us’’ versus ‘‘them’’ environment is not what any of us need.

Not until after the Team Members at the Madison store voted in favor of union representation did management’s description of the union organizing drive begin to engage with the union organizer’s counter-narrative. During

46

JOSHUA L. CARREIRO

the union organizing effort, WFM management did not defend the company’s narrative, but instead focused their energy on persuading Team Members that union representation was a fraud. Kathryn, an employee and member of the union organizing committee, recalls: I think most of their efforts, at least in our captive audience meetings, were about how bad unions were in general, how bad the UFCW was, how it looses your voice, just typical blather you hear in those meetings I guess. In my memory I don’t remember them saying, ‘‘no, we really do live up to our philosophy, no we really do believe in these values.’’ It wasn’t really carried out in that way, but what I understand I think that is really a strategy on their part to not get caught on the defensive – it makes sense that they wouldn’t allow themselves to get put in that position.

Five days following the successful union election, which occurred on July 12, 2002, the executive management of WFM sent a letter to all Team Members nationwide providing their account of the organizing drive and the subsequent election. On the same day, CEO and founder John Mackey sent a letter to all WFM employees. Mackey’s letter reiterates the main points made in the corporate letter, which emphasize the isolated nature of the Madison union victory due not to a failure in the culture of the company, but failure on the part of that store’s leadership, and the incongruence between unionization and the WFM philosophy and values. The post-election corporate letter sent to all Team Members allocates a significant amount of space to countering the union organizer’s claims about the compatibility between WFM and unionization. Similar to the union organizers, WFM management believed that the Madison store’s motivation to unionize was based on narrative inconsistency between WFM’s stated policies and procedures and actual practice. In the postelection corporate letter, management claimed that the election ‘‘outcome was the direct result of the failure of the previous store and regional leadership to follow WFM policies and procedures at this store.’’ Further, management admitted that the ‘‘situation was a significant failure on the part of leadership – we were not walking our talk in the Madison store.’’ Mackey’s letter (July 17, 2002) decries that it is ‘‘unfortunate that the Madison store has only a few Team Members who have worked in other WFM stores – Team members who would have been able to share their positive experiences about working in a store where our culture is alive and well.’’ Further, the letter implies that unionization is incompatible with the ‘‘unique culture’’ of WFM. Mackey argues that if ‘‘we intend to hold on to our unique culture that is based on mutual trust and respect, this leadership failure is something that cannot and will not be repeated elsewhere.’’

Organizational Narratives and the Construction of Resistance

47

Like the union organizers, management at WFM acknowledged that the supporters of the union at the Madison store believed in the philosophy, values, and culture of WFM. However, unlike the union organizers, management did not agree that a union contract was the best way to guarantee that the policies and procedures of WFM would be upheld. According to WFM management, ‘‘being a union shop will mean a very different world from the one they are seeking; many of you with union workplace experience already in your career know this and have shared with us your own feelings.’’ The letter ends by asking all Team Members to ‘‘commit to afﬁrming more than ever the mission of creating together a workplace based on mutual trust, respect, and love.’’

‘‘THE BEST WAY TO RESPECT THE WISHES OF OUR MADISON TEAM MEMBERS IS TO WITHDRAW RECOGNITION FROM THE UNION’’ Following the successful election in July 2002 of the UFCW to represent the Team Members at the Madison WFM, store management reluctantly recognized the union. However, by late January 2003 a ﬁrst contract with the union had yet to be negotiated and WFM had cancelled negotiation talks for the next month citing scheduling conﬂicts (Nathans, 2003a). The union accused WFM of stalling the ﬁrst contract in hopes that the union would be decertiﬁed by an employee vote after the one-year waiting period required by the NLRB. WFM denied the allegation. However, a little more than one year following the union’s election and before a ﬁrst contract had been negotiated, WFM announced that the company would no longer recognize the union. Citing a petition that a majority of Madison Team Members submitted to the NLRB requesting that the union no longer be considered their ofﬁcial representative, the company withdrew recognition of the union, which ended contract negotiations and union representation in general. Union supporters argued that the petition was fraudulent because WFM management assisted antiunion Team Members with gathering signatures and allowed employees to solicit signatures during work hours, both violating existing labor law (Nathans, 2003b). The union ﬁled claims with the NLRB concerning these alleged abuses. However, resolution of the claims may take up to three or four years to resolve – almost guaranteeing that few original union supporters at the Madison WFM will be employed when the NLRB decides the case. Because of the lengthy adjudication process

48

JOSHUA L. CARREIRO

associated with NLRB charges, many employees break labor laws knowing that they will not be punished until after the union organizing effort has been stiﬂed (Hurd & Uehlein, 1994). Like other low-wage service sector employers, WFM experiences considerably high employee turnover in some of their stores. Thus, negotiating a ﬁrst contract proved as difﬁcult as the initial organizing effort. WFM management stalled in the hopes that union supporters would be replaced over time by anti-union Team Members or at least those indifferent to union representation. Union organizers charged that management specifically sought out new hires that would be less likely to support the union and thus be more easily convinced to favor decertiﬁcation. Kathryn, an employee and member of the union organizing committee, recalls: It was kind of funnyybecause we noticed that the people they tended to hire, for lack of a nicer way of putting it, were kind of socially awkward and really not wanting to be part of anything and just kind of (believed), ‘‘this is my job, I’m going to come here and work and then go back home.’’ Like those kind of people or, for instance, they hired a couple of older people, but they ended up being supportive, they were just trying to hire people that didn’t really ﬁt the typical WFM worker.

In addition to apathetic Team Members, union organizers noticed that after the election, management favored younger hires that would be less likely to view employment at WFM as a focal point in their lives. According to Michael, another employee and member of the union organizing committee: One of the big things I noticed was a huge demographic change. When I ﬁrst started working there, there were a number of people in their mid to late 20s. And then I noticed that there were a lot of ‘‘just out of high school’’ or ‘‘just old enough to work’’ (employees) when we were trying to have a kind of reorganization. And also there was almost a complete turnover of employees. It was almost like every single employee was brand new and so we were trying to not only negotiate a contract, but also a complete reeducation and reorganization of Team Members. The thing we kept on hearing over and over again was ‘‘I just don’t care.’’ I think that had a lot to do with the age of the workers. They didn’t have anything invested in the job; they knew they weren’t going to be there very long; they were like ‘‘it’s just a grocery store.’’

The hiring choices made by the Madison store management after the election reﬂects an acknowledgement of a correlation between team membership and unionization. By seeking out new hires that eschew the extra-monetary aspects of employment at WFM, management embraced employees who viewed working at WFM as ‘‘just a job’’ and WFM as ‘‘just a grocery store’’ in spite of the company narrative that characterizes employment at WFM as participatory and empowering and more than ‘‘just a paycheck.’’

Organizational Narratives and the Construction of Resistance

49

Despite continued narrative inconsistency, WFM management evoked the company’s organizational narrative in its explanation of the union decertiﬁcation. In a press release dated November 12, 2003, CEO and founder John Mackey announced, ‘‘the company fully expects to put this brief period of unionization behind us and move on together with a renewed sense of cooperation and shared vision.’’ The press release ends by restating the dominant organizational narrative, ‘‘Whole Foods Market was founded on the ideal of respect for the individual, shared fate, fairness in the workplace, and Team Member excellence and happiness.’’

CONCLUSION In the end, the Madison WFM was able to avoid unionization. WFM management accomplished this through a mixture of strategies. Like other employers adverse to union representation, WFM utilized traditional antiunion tactics such as mandatory employee meetings where management encouraged Team Members to vote ‘‘no’’ in the union election, the ﬁring of a few key union supporters to serve as an illustration and warning to other employees, and promises of positive organizational change if workers decided against union representation and negative change if they voted for representation. However, a tactic that proved somewhat unique to WFM was its reliance on the company’s organizational narrative to discipline the workforce and legitimate management’s antiunion tactics. Following the union election, an employee and union organizing committee member was asked in a radio interview (July 12, 2002) to comment on the union’s narrow victory (65 votes for the union, 54 against): I think there was a lot of buying into sort of the unique culture and special work environment that Whole Foods seems to think it has. So I think that given our feelings that the store is unique in its culture a lot of people felt that the union wouldn’t do well in the store if they think of traditional unions of the 60s and the steel mills and how that sort of would clash with what we’re trying to do here at Whole Foods.

In an interesting narrative turn, the Team Member made reference to WFM’s ‘‘unique culture and special work environment’’ to tell a story about why the union had only limited support among Team Members. Previously, union organizers had referred to the team-based organization and WFM’s claims of worker empowerment as a way to explain the resemblance between the union and team-based organization and to tell a story of the movement that characterized unionization as an almost spontaneous outgrowth of team membership. The company’s dominant organizational narrative thus

50

JOSHUA L. CARREIRO

provides the language for all of the stories told by the counter-movement; both those that explain potential success and those that account for impending failure. The dominant organizational narrative of WFM resembles a hegemonic tale (Ewick & Silbey, 1995) to the extent that the stories articulate and reproduce relations of power. The narratives told by WFM management about its relationship to employees and customers simultaneously conceal and strengthen the hierarchical control imposed on Team Members. By emphasizing the team-based organization of WFM, as well as the opportunities for worker empowerment and participation, the company attempts to downplay class divisions based in hierarchical distinctions between organizational positions, value the input as well as the output of employees, and foster a common and unique organizational culture. At WFM, the distribution of power supports the organizational narrative, while at the same time the organizational narrative reinforces the distribution of power. Because power is dispersed through team organization, the WFM narrative is detached from its organizational origin and achieves the polyvocal and hegemonic status described by Ewick and Silbey (1995). Davis (2002) argues that social movements ‘‘struggle against preexisting cultural and institutional narratives and the structures of meaning and power they convey’’ (p. 25). The unionization effort at the Madison WFM illustrates this point well. By aligning the interests of the union with those expressed in WFM’s organizational narratives and portraying the union as the development and fulﬁllment of a team-based organization and philosophy, union supporters hoped to allay fears among company management and Team Members alike that union representation contradicted the company’s ‘‘unique culture.’’ In addition, by recycling the company narrative, union supporters attempted to translate the foreign (union representation) into the familiar (team membership). Ultimately, however, the union stories reiterated and supported the organizational narrative. For example, union supporters did not openly challenge the team-based organization that characterizes WFM, but only claimed to want to reinforce it with other forms of worker empowerment. Likewise, union supporters’ stories about their movement did not associate WFM’s narrative inconsistencies with systemic failure, but instead emphasized the company’s failure to balance recent growth with the maintenance of WFM’s vision. The union movement did offer counter-narratives that criticized WFM directly. For example, union supporters argued that the company mislead

Organizational Narratives and the Construction of Resistance

51

Team Members and the Madison community when it pledged to allow unionization if desired. Yet, for the most part, union supporters remained within the framework of the WFM narrative, which ultimately shaped what was named as problematic. The focus of the union movement, much like the focus of the company narrative, was on the extra-monetary and abstract values of the WFM employment relation and avoided establishing a laborbased framework in which to ground its strategy. In addition to constraining the union effort, this strategy also created an opportunity for WFM management to tell a story of the unionization effort that highlights the movement’s commitment to the ideals of the company and downplay its criticism. The union organizing drive at the Madison WFM illustrates many of the problems associated with organizing in the low-wage service sector. Organizers had to contend with high employee turnover, co-workers with varying levels of dependence on their jobs, an employer that could choose to close an individual store rather than risk the spread of unionization throughout a chain, and a high level of dependence on consumers for support. In addition, WFM presented other challenges and opportunities because of the company’s public image as a progressive company with democratic principles. The union organizers at the Madison WFM sought to deal with these challenges in an innovative way – by focusing on the organization’s narrative inconsistency – but were ultimately unsuccessful in their efforts. The union organizers could have chosen to openly attack the superﬁcial vision of worker empowerment in place at WFM rather than attempt to build off of it. Because the union narrative remained close to the WFM narrative, it is possible that union organizers were unable to convince potential supporters that what the union had to offer was distinct enough from what already existed. The union organizing strategy, by relying so heavily on the framework of the WFM organizational narrative, ultimately backﬁred. Instead of making union representation intelligible to those unfamiliar with the concept, the strategy may have not drawn out what distinguishes union representation from team membership and therefore may not have offered real solutions to the problems that the union organizers identiﬁed and many Team Members experienced. An in-depth study that includes the employees of the Madison WFM is required for a more complete account of the failure of the unionization effort. The power of the organizational narrative, however, certainly contributed to the choice of strategy by the union organizers and to the ultimate resolution of the union organizing drive.

52

JOSHUA L. CARREIRO

NOTES 1. Names of informants have been changed for anonymity.

ACKNOWLEDGMENTS The author would like to thank Sarah Babb, Keith Brown, Dan Clawson, Vicki Smith, and Robert Zussman for their encouragement, helpful comments, and thoughtful criticism. A version of this paper received the Harry Braverman Award from the Society for the Study of Social Problems in 2005.

REFERENCES Barker, J. R. (1993). Tightening the iron cage: Concertive control in self-managing teams. Administrative Science Quarterly, 38, 408–437. Blackman, A. (2004). Whole Foods links executive pay to worker pay. Pittsburgh Post-Gazette, C–12 April 13. Boje, D. M. (1991). The storytelling organization: A study of story performance in an ofﬁcesupply ﬁrm. Administrative Science Quarterly, 36, 106–126. Davis, J. E. (2002). Narrative and social movements: The power of stories. In: J. E. Davis (Ed.), Stories of change: Narratives and social movements (pp. 3–31). Albany, NY: State University of New York Press. Ewick, P., & Silbey, S. S. (1995). Subversive stories and hegemonic tales: Toward a sociology of narratives. Law & Society Review, 29(2), 197–226. Fishman, C. (1996). Whole Foods is all teams. Fastcompany, from www.fastcompany.com/ magazine/02/team1.html Foucault, M. (1977). Discipline and punish: The birth of the prison. New York: Vintage Books. Garson, B. (1988). The electronic sweatshop: How computers are transforming the office of the future into the factory of the past. New York: Simon & Schuster. Gaus, M., & White J. (2003). Natural union busters? In These Times, February 17, p. 5. Graham, L. (1995). On the line at Subaru-Isuzu: The Japanese model and the American worker. Ithaca, NY: ILR Press. Hull, A. (2001). Divided feast: How a high-end consumer’s paradise is redrawing the lines in a formerly low-rent neighborhood. The Washington Post Magazine, April 1, p.W19. Hurd, R. W., & Uehlein, J. B. (1994). Patterned responses to organizing: Case studies of the union-busting convention. In: S. Friedman, R. W. Hurd, R. A. Oswald & R. L. Seeber (Eds), Restoring the promise of American labor law (pp. 61–74). Ithaca, NY: ILR Press. Leidner, R. (1993). Fast food, fast talk: Service work and the routinization of everyday life. Berkeley, CA: University of California Press. Lukes, S. (1974). Power: A radical view. New York: Macmillan. Martin, J., & Powers, M. E. (1983). Truth or corporate propaganda: The value of a good war story. In: L. R. Pondy, P. J. Frost, G. Morgan & T. C. Dandridge (Eds), Organizational symbolism (pp. 93–108). London: JAI Press, Inc.

Organizational Narratives and the Construction of Resistance

53

Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. Nathans, A. (2003a). Grocer gets rid of union: Whole Foods pulls recognition. The Capital Times, November 13. Nathans, A. (2003b). Whole Foods cancels talks: Union suspicious. The Capital Times, January 22. Polletta, F. (2002). Plotting protest: Mobilizing stories in the 1960 student sit-ins. In: J. E. Davis (Ed.), Stories of change: Narratives and social movements (pp. 31–52). Albany, NY: State University of New York Press. Posner, B. G. (1989). The best little handbook in Texas. Inc, 84, (February). Wilkins, A. L. (1983). Organizational stories as symbols which control the organization. In: L. R. Pondy, P. J. Frost, G. Morgan & T. C. Dandridge (Eds), Organizational symbolism (pp. 81–92). London: JAI Press, Inc. Witten, M. (1993). Narrative and the culture of obedience at the workplace. In: D. K. Mumby (Ed.), Narrative and social control: Critical perspectives (pp. 97–120). London: Sage.

This page intentionally left blank

54

THE EFFECT OF PARTICIPATIVE WORK SYSTEMS ON EMPLOYEE EARNINGS Michael J. Handel ABSTRACT In the last 20 years wage inequality in the United States has grown significantly. At the same time, the increased popularity of employee involvement or the so-called high-performance work practices seems to offer opportunities for more skilled, autonomous, and participatory work. For many, this is a positive alternative to low-wage jobs, though others suggest that such jobs may raise skill requirements sufficiently to leave many workers behind and thereby contribute to growing inequality. Yet others are more critical and view participatory work systems as merely a method of work intensification. This paper examines the impact of participative work systems on workers’ wages and generally finds modest evidence of significant effects.

It is widely recognized that post-war employment has changed signiﬁcantly in the last 25 years but there is continuing debate over the nature of the changes. In the ﬁrst half of the post-war period, the most notable development was the expansion of stable, bureaucratic hierarchies with well-deﬁned roles that

Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 55–84 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16003-3

55

56

MICHAEL J. HANDEL

offered relatively generous material rewards, even though job tasks, often shaped by scientiﬁc management, were not necessarily fulﬁlling (Edwards, 1979). In the 1970s and early 1980s, a series of problems – inﬂation, deep recessions, sharply declining productivity growth, and surging Japanese imports – threw this system into crisis and exerted pressure to alter many aspects of existing economic organization, including work and the employment relationship. Pressures continue today even though many of these problems have receded, partly because some sources of change remain important, such as information technology and globalization, and because ideas that became popular during the crisis period have become institutionalized. Changes in work and employment that most observers would consider generally negative include stagnant or declining real wages, rising inequality, increased workloads, declining job security, and the erosion of other institutional protections, such as unionization and the real value of the minimum wage. By contrast, other changes suggest improvements in job quality, such as the spread of new work practices that enhance skills, autonomy, and decision-making responsibilities. Ironically, both kinds of changes received impetus from the crises of the 1970s–1980s and it was not long before some were arguing that participatory work systems might serve as an antidote to the ‘‘low road’’ policies that sought to regain competitiveness by squeezing labor costs (Appelbaum & Batt, 1994; National Center on Education and the Economy, 1990). Interest in workplace participation among academics originated with concerns over job satisfaction during and immediately following the 1960s boom (e.g., United States Department of Health, Education, and Welfare, 1973; Walton, 1974; see also Hodson, 2001, 2002). However, adoption by employers on a wide scale did not occur until the surge of manufacturing imports from Japan in the late 1970s and 1980s lent credibility to these ideas. Thus, practices originally promoted in the United States for addressing non-pecuniary issues came to be seen as instrumental for improving productivity and wages in an increasingly competitive world. However, critics noted that participatory practices are not necessarily antidotes to low road strategies; they can be combined with task simpliﬁcation and increased monitoring (Adler, 1993; Barker, 1993; Berggren, 1994; Graham, 1993; Smith, 1997). Institutional theory argued that adoption of these practices tended to be more ritualistic than substantive (Abrahamson & Fairchild, 1999; DiMaggio & Powell, 1983; Zbaracki, 1998). Research on whether participatory work practices are actually associated with higher wages remains limited and the results are very mixed (Handel & Levine, 2004).

The Effect of Participative Work Systems on Employee Earnings

57

This paper uses various measures of autonomy, participation, and related concepts to examine whether participatory work practices are associated with higher wages. Results suggest more limited impacts than the more optimistic accounts suggest but they also do not support negative views of participation, at least when wage levels are the outcome variable.

1. INTRODUCTION Since the late 1970s, the wages of most workers have fallen and wage inequality has risen to its highest levels in the post-war United States (Katz & Murphy, 1992; Mishel, Bernstein, & Allegretto, 2005). The widely shared afﬂuence of the early post-war period rested in part on low- to mediumskilled jobs that paid relatively high wages despite their Taylorized character, often due to the direct or indirect effects of unions. However, by the early 1980s this ‘‘Fordist’’ employment model declined in response to economic crises. In the language of segmented labor markets theory, the number of jobs in the secondary labor market, such as those in low-wage services, grew. In addition, the logic of secondary labor markets, such as competition based on cheap labor costs, pressed its way into the primary labor market segments (Harrison & Bluestone, 1988; cf. Edwards, 1979). The result was described as a declining number of ‘‘good’’ jobs and a growing number of ‘‘bad’’ jobs (Loveman & Tilly, 1988). At the same time, a quite separate group of researchers argued that a new set of organizational pressures were tending toward the reorganization of work in ways that were favorable to employees, sometimes called ‘‘postFordist’’ or ‘‘high-performance work systems.’’ Increased competition, greater consumer demand for novelty, customization, and quality, and the spread of computers at work meant that it was both possible and necessary to break with Taylorist models of work and employment relations. In a fastchanging, information-rich environment, it made more sense to empower workers and give them the skills and authority to innovate, solve problems, and respond rapidly to changing demands. New work systems give non-management workers some higher-level tasks usually associated with craft, supervisory, and professional staff jobs and treat them more as full members of the ﬁrm. Workers share in any productivity gains they help achieve through various bonus, proﬁt-sharing, or stock ownership plans. Greater workplace cooperation and employee commitment replace previously adversarial relations, which presumably contributes to productivity growth, as was believed to be true for Japanese ﬁrms (Blasi et al., 2000;

58

MICHAEL J. HANDEL

Osterman, 2000; Piore & Sabel, 1984; Tausky & Chelte, 1988; Walton, 1985; Zuboff, 1988; for a summary see Smith, 1997). In light of Japanese success, many came to view these new practices as a possible source of high-paying jobs to replace the vanishing Fordist model and avoid the pitfalls of a ‘‘low road’’ employment strategy based on low skills and low wages (e.g., Bluestone & Bluestone, 1992; National Center on Education and the Economy, 1990). As one review argued, ‘‘Despite the obstacles some ﬁrms have succeeded in becoming high-performance workplaces y these companies are the best hope American workers have for middle-class opportunities – good wages, employment security, and interesting work’’ (Appelbaum & Batt, 1994, p. 13). Interestingly, others who also take a favorable view of these organizational changes have drawn a less optimistic inference regarding their consequences. They believe that greater workplace participation contributed to the growth wage inequality by increasing the returns to human capital, in contrast to the more institutional explanations offered by the original studies of inequality growth described above (Bresnahan, Brynjolfsson, & Hitt, 2002). Whether participatory work practices are viewed as contributing to or mitigating inequality growth, both views agree that such jobs involve greater skill and, consequently, higher wages for those who can secure them. However, the new participative management techniques soon attracted critics, as well. They argued that the changes in skill demands and decisionmaking responsibility were modest. New practices simply repackaged familiar human relations ideas with the goal of co-opting workers into identifying with ﬁrm goals and discouraging unionization. Employees are induced to share private expertise with management, participate in time study and speed-up of their own jobs, and submit to tighter control through peer pressure and computer monitoring (Barker, 1993; Fantasia, Clawson, & Graham, 1988; Graham, 1993; cf. Braverman, 1974, pp. 35ff.). High-performance work practices are simply another aspect of the drive to cheapen labor costs, in this view. Since this perspective describes these new developments as an intensiﬁcation of traditional capitalist principles, it is often referred to as a neoFordist interpretation of the new employment system. In summary, according to proponents, new employment systems involve: more varied and skilled (‘‘enriched’’) job tasks, as opposed to repetitive and deskilled work (e.g., job rotation, quality control tasks, record keeping); employee participation in decision-making (e.g., deciding task assignments, production scheduling, contributing to performance evaluations); autonomy from close supervision;

The Effect of Participative Work Systems on Employee Earnings

59

cooperative relations among workers and between management and labor; contingent compensation, such as gain sharing or performance bonuses. Critics would also include increased workload, faster work pace, closer monitoring, and increased job stress as part of their deﬁnition or description of these management initiatives.

2. THEORY AND RESEARCH Although there is great hope that participatory work practices offer a new model of high-wage, high-skill jobs for large numbers of workers, theory and research on the subject remain limited and ﬁndings are mixed (Handel & Levine, 2004). There are a number of reasons to expect that high-performance work practices might raise wages. Firms may reward workers for sharing the large stock of experiential knowledge they have accumulated over time (Walton, 1985; Zuboff, 1988). Since high-performance work practices also involve new and qualitatively more skilled tasks related to problem-solving and quality control, they are typically associated with higher levels of worker training (Frazis, Gittleman, & Joyce, 2000). In this way, high-performance practices may raise wages because they induce ﬁrms to enhance their employees’ human capital. Gift exchange versions of efﬁciency wage theory argue that ﬁrms valuing worker morale and loyalty will also pay higher wages, conditions that might well obtain in a system that requires greater cooperation, organizational commitment, and discretionary effort (Akerlof, 1982). Efﬁciency wages also reduce employee turnover, which is important for employers who invest signiﬁcantly in ﬁrm-speciﬁc training. In addition, both principal-agency theory and theories of the ﬁrm as a cooperative, ‘‘mutual gains’’ enterprise suggest that various forms of gain sharing and performance bonuses align the interests of employees and employers and contribute to increased work effort and rewards (Blasi, Conte, & Kruse, 1996; Kandel & Lazear, 1992; Kruse, 1992; Walton, 1985). The preceding assumes that the introduction of high-performance work practices increases the productivity of a ﬁrm’s workforce, which results in greater employee rewards. However, many high-performance organizations use relatively rigorous employee screening and selection procedures, such as pre-employment tests, extensive interviews, and team exercises (Florida & Kenney, 1991, p. 388; Murnane & Levy, 1996, pp. 22ff.). To the extent that high-performance workplaces skim the cream off the labor market, their employees might receive equally high wages even in traditional workplaces. In

60

MICHAEL J. HANDEL

this case a positive association between high-performance work and wages may reﬂect sorting or selection effects rather than a genuinely causal relationship. There are also reasons one would not expect to observe much of a wage payoff to high-performance work practices. The magnitude of the skill upgrading and increased decision-making responsibilities remains largely unknown and may be relatively modest and easily acquired by existing employees (Cappelli & Rogovsky, 1994). Even in cases where there are signiﬁcant changes in the workplace and associated productivity gains, workers may not have the power to force management to share those gains given the decline in labor’s power over the last 20 years. In these cases, the various forms of contingent pay, such as gain sharing and proﬁt sharing, may substitute for base pay rather than supplement it. Alternatively, insofar as employee involvement programs improve employee satisfaction, they may act as a compensating differential allowing employers to offer lower wages and still attract workers (Williamson, 1985, pp. 268ff.). Finally, survey indicators of high-performance work practices may contain random or other measurement error that biases coefﬁcients toward zero. As in the case of selection effects, the estimates in this case would reﬂect more a limitation of the data than the nature of the relationships that actually obtain in the world. Since selection effects would be expected to bias coefﬁcient estimates upward and measurement error biases them downward, the presence of both would result in each offsetting the other. Despite its importance for the sociology of work, the relationship between workplace participation and wages has not been researched extensively and coefﬁcient estimates vary widely, though tending to cluster in a range suggesting that employee involvement is associated with wages that are 0–5% higher than otherwise (Handel & Levine, 2004). Descriptions of a few of the better-known studies illustrate this diversity. A qualitative study of six retail ﬁrms found modest effects on skill levels and little pressure to pay higher wages after introducing innovative practices in this generally low-wage industry (Bailey & Bernhardt, 1997). Surveys in steel minimills, machined goods, and telecommunications industries found no relationship between participative work organization and wages (Arthur, 1992; Batt, 2001; Kelley, 1996). Cross-sectional and panel analyses from a nationally representative survey of establishments using a limited set of measures also found no relationship between high-performance work practices and wages or wage growth over time (Osterman, 1994, 2000). However, other surveys found positive effects. A study of participative work systems in steel, apparel, and medical electronics found that self-directed teams were associated with a 10% wage premium in apparel

The Effect of Participative Work Systems on Employee Earnings

61

and a standard deviation increase in an index of participation was associated with about a 2.75% wage premium in steel and apparel (author’s calculations from Appelbaum, Bailey, Berg, & Kalleberg, 2000, pp. 172, 214). A survey of retail bank branches found a 5% wage differential associated with the presence of quality circles, which rose to about 9.5% in more automated branches (Hunter & Lafkas, 2003). A national survey of employees found about a third of workers in employee involvement programs reported receiving a pay raise of an unspeciﬁed amount as a consequence of their participation (Freeman & Kleiner, 2000). Given the small number of studies on this subject, many with specialized samples and somewhat contradictory results, there is a need for further research. This paper uses three nationally representative datasets with multiple measures of variables commonly associated with high-performance systems as predictors of employee wages. Variables include measures of autonomy and control, participation, skill, performance-based compensation, and workplace cooperation and commitment. Existing theory and research suggests the following hypotheses to be tested: H1. Employees who have (a) greater autonomy, (b) more complex tasks, and (c) participate in decision-making earn more than otherwise similar employees. H2. Conversely, employees working under more Taylorist conditions earn less than otherwise similar employees. H3. Performance-based compensation is associated with higher earnings. H4. Cooperative organizational climates and individual organizational identiﬁcation are associated with higher earnings.

3. DATA The data used here are the General Social Survey (GSS) module on Work Organizations (1991), the Class Structure and Class Consciousness Study (1980), conducted by Erik Olin Wright, and the Labor Market Survey (1991), conducted by Wright and Michael Hout. These surveys contain unusually detailed measures of autonomy and participation at work. Unlike most surveys dealing with teams or quality circles, these contain behaviorally speciﬁc indicators of how people participate in decision-making at work, rather than simply indicating membership in a team whose actual responsibilities, inﬂuence, and meeting frequency may be highly variable, as suggested by the

62

MICHAEL J. HANDEL

skepticism of institutionalist research. Because the unit of analysis is the employee rather than the more commonly used establishment level (e.g., Appelbaum et al., 2000; Osterman, 1994, 2000), it is also possible to control more effectively for individual background variables. Although the data are not recent, the logic of the argument regarding the impact of participative practices is universal; human capital and efﬁciency wage theories imply that greater skill and participation are associated with higher wages. Other research also indicates that workplace participation was present during this period in both organizations that consciously implemented employee involvement programs and those for which it was an implicit organizational design principle (Appelbaum & Batt, 1994; Hodson, 2001, 2002; Walton, 1985). The strengths of these datasets are the detailed measures of workplace participation and broad labor force coverage that enable one to test the theory of ‘‘high road’’ employment systems. Nevertheless, it is possible that different conclusions might result if more recent national surveys were to replicate the items used here. Unfortunately, despite similarities in question wording, the two Wright surveys use different response options and the 1980 survey uses a ﬁlter question before a battery of items that is omitted in the 1991 survey, which prevents direct comparisons of trends between the 2 years despite generally similar survey items. Dependent variable. The dependent variable is (ln) hourly wages for all models using the Labor Market Survey (1991) and ln (annual earnings/ weekly hours worked) for models using the GSS (1991) and Class Structure Survey (1980). The annual earnings variables from the GSS and Class Structure Survey are categorical, requiring the use of interval midpoints, but fortunately the categories are relatively detailed so the loss of precision is not great. Likewise, in both surveys weekly hours worked refers to the current job and earnings refers to the previous year, but hours worked across adjacent years are known to be highly correlated, which makes the adjustment of earnings for work time preferable to not controlling for hours worked at all. Study variables. The independent variables are measures of relatively speciﬁc kinds of decision-making and non-traditional tasks and more general selfreports of the extent of autonomy or input employees have in the workplace. This combination of speciﬁc behavioral items and more global evaluations should perform well in capturing the dimensions of work that are of interest here. Research indicates that general measures of autonomy and participation are associated with more speciﬁc indicators of high-performance work practices (Appelbaum et al., 2000, pp. 181f.; Cappelli et al., 1997, pp. 99ff.).

The Effect of Participative Work Systems on Employee Earnings

63

The GSS also contains a wider set of variables on workplace skills, control, compensation policies, organizational climate, and employee commitment that are relevant to debates over innovative workplace organization. The items and scales used as independent variables are listed in brief form below and described more fully in Tables A1 and A2

General Social Survey Participation, skill, and task content 1. Autonomy/participation scale (a ¼ 0.80) 3. Spending decisions – participation (1 ¼ yes) 5. Importance of education as source of skills (omitted ¼ very important)

2. Five-task scale (a ¼ 0.75) 4. Level of expenditure ($19911/2/ 100) 6. Importance of training as source of skills (omitted ¼ very important)

Workplace Control 7. Machines control job (omitted ¼ not true)

8. Computer monitoring (omitted ¼ not true)

Compensation 9. Proﬁt sharing/stock options – eligible (1 ¼ yes) 11. Uniform raises (1 ¼ yes)

10. Performance bonus – eligible (1 ¼ yes)

Cooperation at work: Organizational climate 12. Labor–management relations (omitted ¼ good) 14. Flextime – eligible (1 ¼ yes)

13. Vote for union (1 ¼ yes) 15. Peer relations (omitted ¼ good)

Cooperation at work: Individual commitment 16. ted 18. ted

Voluntary extra effort (omit¼ agree) Likelihood of quitting (omit¼ not at all likely)

17. Actual effort (omitted ¼ somewhat 4 others) 19. Organizational identiﬁcation scale (a ¼ 0.72)

64

MICHAEL J. HANDEL

Class Structure Survey (1980) and Labor Market Survey (1991) Respondent’s autonomy/participation – R can decide to: 1. Introduce a new work task

2. Slow work pace for a day

Respondent’s policy inﬂuence – R personally involved in policy decisions regarding: 3. Basic methods or procedures of4. Signiﬁcant change in products/ work services 5. Routine work pace

6. Distribution of funds within overall budget

7. Increase/decrease number employed Labor Market Survey (1991) only Non-management inﬂuence on decision-making – how much influence nonmanagement employees have over: 8. Task assignments 10. Introducing new technology 12. Discipline policies

9. Introducing new ways of organizing work 11. Making changes in products/ services 13. Deciding pay levels for different jobs

14. Deciding lay-off policies Control variables. All models also control for a range of background variables that affect wages according to previous research: education, experience and its square, tenure, occupation (managers/professionals, clerical/sales, craft, blue-collar/service), one-digit industry, race, gender, union status, part-time work status, marital status and its interaction with gender, residence in a metropolitan area, and region of residence. All models also control for establishment size except those from the Class Structure Survey, which lacks this measure. Additional controls for usually unobserved worker quality are included to mitigate selection effects. These are employer screening methods, such as pre-employment testing, for GSS models and organizational position for models using the Class Structure and Labor Market Surveys. All samples are restricted to wage and salary workers.

The Effect of Participative Work Systems on Employee Earnings

65

Since the discussion of workplace participation often focuses on their potential value for transforming low paid, less-skilled work into more remunerative jobs, interactions of predictors with occupation are also reported in the text where signiﬁcant. No interaction effects were signiﬁcant in the GSS models and they were relatively infrequent in the other models, perhaps partly reﬂecting the relatively small sample sizes of these datasets.

4. RESULTS 4.1. General Social Survey Baseline models with only control variables are reported in Table A3 for all datasets. The models for the GSS give reasonable coefﬁcient estimates for education, experience, and gender, while the coefﬁcients for black, union, and part-time status are lower than expected. The coefﬁcients for the occupational dummies are rather large but in the expected direction, and have a limited effect on the size of the other coefﬁcients. The second baseline model attempts to control for selection effects by including measures of usually unobserved labor quality. The items asked respondents whether they had to take an intelligence test, skills test, drug test, or physical or submit letters of reference to obtain their current job. These variables represent some of the screening devices used by highperformance workplaces (Murnane & Levy, 1996) and one would expect that workers subjected to more exacting hiring procedures would be of higher quality. Unfortunately, these controls do not perform very well in practice, though at least one dummy for each screening variable except drug testing is signiﬁcant when they are interacted with occupation (not shown). All regression models in Table 1 augment this interactive model with individual study variables or small groups of predictors in the ﬁrst column. All study predictors are included in a single, joint model in the second column. The independent variables of interest are grouped into ﬁve categories, corresponding to different aspects of participatory systems discussed in the literature:

autonomy/participation/skill employer control techniques contingent compensation workplace cooperation – organizational climate workplace cooperation – individual commitment.

66

MICHAEL J. HANDEL

Table 1.

OLS Regression Estimates of Effects of Participation on Pay (GSS 1991). Scale/Variable

I. II.

Autonomy, participation, skill 1. Autonomy/participation Adj. R2 2. Five-task index 3. Participate in spending decisions (1 ¼ yes) 4. Level of spending pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ $1991 =100 Adj. R2

III. How important as source of job skills was: 5. Education (omitted ¼ very) Not at all important Somewhat important Important 6. Training (omitted ¼ very) Not at all important Somewhat important Important Adj. R2 Workplace control IV. 7. Job is controlled by machines (omitted ¼ not true) Somewhat true True Very true 8. Computers monitor job performance (omitted ¼ not true) Somewhat true True Very true Adj. R2 V.

Compensation 9. Proﬁt sharing (1 ¼ yes) 10. Performance bonus (1 ¼ yes) 11. Uniform raises (1 ¼ yes) Adj. R2

Individual Predictor Models 0.0763 0.3315 0.0054 0.1386 0.0292

Joint Model

0.0345 0.0431 0.1941 0.0032

0.3393

0.1400 0.1283 0.0187

0.0888 0.1288 0.0463

0.1919 0.1715 0.0074 0.0745 0.0038 0.0612 0.3427

0.0028 0.1046 0.0118

0.0807 0.1035 0.0690

0.0666 0.1126 0.0447 0.3318

0.2232 0.1246 0.0162

0.1670 0.1446 0.0738 0.3433

0.2238 0.0693 0.0440

The Effect of Participative Work Systems on Employee Earnings

67

Table 1. (Continued ) Scale/Variable

Cooperation: Organizational climate VI. 12. Labor–management relations (omitted ¼ good) Very bad Bad Neither good nor bad Very good 13. Vote for a union (1 ¼ yes) 14. Flextime (1 ¼ yes) 15. Relations among coworkers (omitted ¼ good) Very bad Bad Neither good nor bad Very good Adj. R2 Cooperation: Individual commitment VII. 16. R willing to work harder than required to help organization (omit ¼ agree) Strongly disagree Disagree Strongly agree 17. Compared to others, R works (omitted ¼ somewhat more) Much less Somewhat less About the same Much more 18. How likely R will leave organization in 12 months (omitted ¼ not at all likely) Somewhat likely Very likely 19. Organization identiﬁcation scale Adj. R2

Individual Predictor Models

Joint Model

0.2761 0.1117 0.1277 0.0276 0.0559 0.0866

0.5122 0.0474 0.1475 0.0001 0.0119 0.1041

0.1292 0.1375 0.0913 0.0510 0.3349

0.4578 0.1569 0.1463 0.0138

0.1018 0.1450 0.0106

0.0607 0.0823 0.0430

0.3197 0.0162 0.3931 0.5531 0.0505 0.0110 0.0863 0.1323

0.0274 0.1465 0.0310 0.3346

0.0269 0.1845 0.0524 0.3529

Note: Dependent variable is ln (annual earnings/weekly hours worked). All models include controls for education, experience, experience2, female, black, union, part-time, four occupational groups, whether the respondent faced four employer screens (intelligence test, skills test, physical, letters of reference), screens occupation interaction, one-digit industry, tenure, establishment size, married, married female, region, and resident of metropolitan area. po0.10. po0.05. po0.01.

68

MICHAEL J. HANDEL

Models I and II include different measures of autonomy, task complexity, and participation (Hypothesis 1). Model I uses a standardized scale constructed from four general statements regarding job autonomy and participation (a ¼ 0.80). Model II uses three behaviorally speciﬁc measures of participation. The ﬁrst is a standardized index of ﬁve speciﬁc tasks the respondent performs with employees he/she does not supervise, all of which are commonly found in high-performance work teams (e.g., organize schedules and tasks, evaluate others’ performance) (a ¼ 0.75). The others are a binary indicator for whether the respondent participated in spending decisions in the last year and, if so, the size of the largest spending decision, expressed as a square root to minimize skewness.1 Results indicate that a standard deviation increase in autonomy is associated with an 8% wage differential. The mean autonomy scores for managers/professionals and blue-collar/service workers differ by 0.62 standard deviations, so if the entire gap were closed blue-collar workers would earn 5% more on average, compared to the 40% earnings gap in the baseline models (Table A3). Even this effect disappears in the joint model, possibly because of the size of the correlation with whether the respondent participates in spending decisions (r ¼ 0.41). Participation in spending is associated with a signiﬁcant wage differential of approximately 15–20%. About 56% of managers/professionals and 12% blue-collar/service workers participated in spending decisions, though the average size of spending decisions does not vary by occupation as much as might be expected (not shown), which may account for the fact that this variable is not signiﬁcant. The ﬁve-task index also has no signiﬁcant effect on wages in either model. Model III includes two more general measures of job complexity (Hypothesis 1) asking how important formal education and formal on-the-job training from the current employer were as sources of skills that respondents use on their jobs. The importance of education is not signiﬁcant, though coefﬁcients are appropriately signed. Presumably the wage effects of having a job that uses skills taught in school are better captured by the personal education and occupation variables. This argument does not apply to employer-provided training, which is widely seen as a critical part of ‘‘high road’’ employment strategies for less-skilled workers and a way to reduce inequalities across education and occupation groups. Consistent with expectations, those for whom on-the-job training was not at all important as a source of skills earned nearly 20% less than others, but the effect slips below conventional levels of signiﬁcance in the joint model.

The Effect of Participative Work Systems on Employee Earnings

69

There is some debate over the extent to which participative work systems have actually raised human capital requirements and thereby contributed to inequality growth in the United States in the last 25 years (Bresnahan et al., 2002; Cappelli & Rogovsky, 1994). In fact, both the importance of education as a source of job skills and respondents’ own education are correlated rather modestly with the autonomy (r ¼ 0.21) and ﬁve-task (r E 0.15) scales and the importance of training is even more weakly associated with these measures (r ¼ 0.09). Participation in spending decisions, much less common among front-line workers, is somewhat more correlated with the education’s importance as a source of jobs skills (r ¼ 0.23) and the respondent’s own education (r ¼ 0.30). On balance this evidence does not suggest that the more common forms of workplace participation have strong effects on the demand for human capital. The GSS module is the only recent survey with information to test Hypothesis 2 on the wage effects of Taylorism. Respondents were asked if their work is controlled by machines or monitored by computers, illustrating Edwards’ (1979) concept of ‘‘technical control’’ (see also Garson, 1988; Graham, 1993; United States Congress, Ofﬁce of Technology Assessment, 1987; cf. Zuboff, 1988). About 21% of all workers and 30% of blue-collar/ service workers said it was ‘‘true’’ or ‘‘very true’’ that machines control how they do their job. About 11% of all workers and 18% of both clerical/sales and craft workers responded similarly when asked if computers monitored their job performance (not shown). However, Model IV indicates technical control is not signiﬁcant in either model and computer monitoring has no consistent effect, though for reasons not readily evident low levels of monitoring are associated with a wage premium in the joint model. Model V examines whether various forms of performance-based pay, gain sharing, and bonuses associated with participatory workplaces are associated with higher earnings (Hypothesis 3). About 29% of employees said they were eligible for proﬁt sharing and stock options and 24% were eligible for cash or stock performance bonuses. By contrast, about half of all respondents said that pay raises were given to all workers regardless of performance where they worked. The incidence of none of these varied greatly by occupational group. Model V indicates that the presence of proﬁt sharing has a large effect on wages (ca. 20%) and is one of the most robust effects of all the variables in this study. The effect of having a bonus system is large and signiﬁcant in the individual-predictor model (ca. 14.5%), which actually includes the proﬁt sharing and uniform pay raise variables, but is insigniﬁcant when other variables are entered in the joint model. The presence or absence of merit

70

MICHAEL J. HANDEL

pay raises is not signiﬁcantly related to wages in either model, despite the presumed incentive effects of such policies. There is no evidence that any of the bonus pay variables is negatively associated with average pay levels, as might be expected if employers were simply altering the weight of different forms of compensation to shift more risk onto employees, which is one criticism of contingent employment policies. Positive employer–employee relations are another key ingredient of recent innovative work systems (Hypothesis 4). The high-performance model takes for granted that a more cooperative workplace is more productive and, presumably, more remunerative as a result of a reciprocal exchange of greater effort for higher earnings between employees and management (Akerlof, 1982; Appelbaum et al., 2000). Models VI and VII examine the earnings effects of workplace cooperation using both the levels of general organizational climate and individual commitment. None of the measures of cooperative organizational climate are associated with higher earnings except for the highly anomalous wage premium (67%) associated with ‘‘very bad’’ employee–management relations in the joint model (Model VI). It would not be surprising if this result, based on 24 cases in the most negative response category, could not be replicated in other datasets. Model VII also ﬁnds few effects of various individual-level measures of workplace cooperation, such as willingness to work harder than required, actual effort relative to coworkers, quit likelihood, and a standardized scale measuring the extent to which workers identify with their employing organization (a ¼ 0.72). Needless to say, there are potential issues with measurement error with many of these variables. Most of the variation for the ﬁrst variable is limited to those who ‘‘agree’’ and ‘‘strongly agree’’ that they are willing to work harder than necessary. About 97% report that they do about as much or more work than others doing similar work, although, consistent with intuition, the few who acknowledge working less than others receive much lower wages (–40% to –55%) in both the individual and joint models. These results generally provide little or mixed support, at best, for the notion that cooperation improves productivity or wages. The modest effect of individual commitment on wages is consistent with analyses of the effect of commitment on self-reported job performance using the same data (Kalleberg & Marsden, 1995). In sum, except for participation in spending decisions, measures of autonomy and participation at work do not have large or consistent effects on wages, whether measured using global, subjective evaluations, or behaviorally speciﬁc indicators. However, in contrast to critics, different

The Effect of Participative Work Systems on Employee Earnings

71

measures of Tayloristic workplace control do not have negative effects on earnings. Various forms of performance-based and contingent pay are positively associated with earnings, consistent with the high-performance model and inconsistent with neo-Fordist expectations that they are used to substitute for base pay and shift business risk to employees.

4.2. Class Structure and Labor Market Surveys The Class Structure and Labor Market Surveys conducted by Erik Olin Wright and colleagues are another source of speciﬁc measures of workplace participation, though they do not have measures of other dimensions of high-performance work systems, such as contingent pay and workplace cooperation. Therefore, all models in this section are tests of Hypothesis 1. The baseline models (Table A3) show reasonable coefﬁcient estimates for most variables, though the estimate for blacks is somewhat low, the estimate for part-time status is unexpectedly signed for 1980, and some coefﬁcients for organizational position have inconsistent values. There are two series of items common to these two surveys and one unique to the 1990/1991 survey. Unfortunately, direct comparisons or pooling data across years is not possible for the common items because of changes in a ﬁlter question for some of them and differences in the number of response categories across years. In addition, the pay variable for 1980 uses interval midpoints for previous year’s earnings divided by weekly hours on current job, similar to the GSS, while the dependent variable for 1990/ 1991 is the calculated, current hourly wage. There are three groups of measures used as independent variables: respondent’s job autonomy/participation respondent’s policy inﬂuence decision-making inﬂuence of non-management employees at respondent’s workplace (1990/1991 only). The tables present results from models in which the predictors are entered individually; joint predictor models did not alter results substantively.2 To conserve space, interactions of participation variables with occupation are only noted in the text where they affect results. Since the decision-making inﬂuence of non-management employees would be expected to affect mostly the wages of non-managers, the models using this set of variables are estimated for a sample restricted to sales, clerical, craft, other blue-collar, and service workers who do not hold supervisory or managerial positions.

72

MICHAEL J. HANDEL

Though sample sizes are reduced considerably, inclusion of other workers might well bias these coefﬁcients downward since they would not necessarily be expected to beneﬁt from the ability of others to exercise decision-making authority. The independent variables are numbered consecutively across Tables 2 and 3 to facilitate comparisons to the variables as described in Table A2, i.e., the ﬁrst variable in Table 3 is number 8. Autonomy/participation. The surveys asked whether respondents can decide to introduce a new work task on their own or they can decide to signiﬁcantly slow their work pace for a day when desired, considered an indicator of workplace control here. The 1980 survey offered a binary choice of responses (yes/no), while the 1991 response choices had a different structure (someone else decides, joint decision, decide on own). The ability to decide to introduce a new work task is associated with a 7–9% wage premium in 1991 only, but interaction with occupation indicates that only managers and professionals beneﬁt from this form of participation and the wage premium is 13–16% (not shown). In contrast, control over work pace is associated with a 7% wage premium in 1980 but is unrelated to wages in 1991. For 1980, the interaction effect is again positive only for managers and professionals (b ¼ 0.17, po0.05) but the effect for blue-collar/service workers is actually negative (b ¼ 0.23, po0.05) (results not shown). Policy influence. The surveys also ask about the respondent’s personal inﬂuence over a number of workplace policy decisions. Answers to these items can be made consistent across years (do not participate, give advice, participate directly in decision-making), but only after collapsing three codes from the 1980 survey, so the results are not strictly comparable. Direct participation in policy decisions regarding changes in basic work methods or procedures is associated with a 11–13% wage premium in both years. Again, interaction models (not shown) suggest the effects apply mainly to managers and professionals for 1991, but a joint signiﬁcance test of all interaction coefﬁcients falls just short of conventional levels of signiﬁcance (p ¼ 0.11), so the possibility of uniform effects across occupations cannot be ruled out. Direct participation in decisions to signiﬁcantly change products/services offered is associated with a 15% wage premium in 1991 and the effect holds for all occupations. Direct participation in decisions regarding workplace employment levels is associated with a 15% wage premium in 1991 but, again, interaction models suggest that this result is driven mostly by effects for managers and professionals (not shown).

The Effect of Participative Work Systems on Employee Earnings

73

Table 2. OLS Regression Estimates of Effects of Respondent’s Autonomy/Participation and Policy Inﬂuence on Pay, Class Structure (1980) and Labor Market (1991) Surveys (Full Sample). Predictor Autonomy/participation Can R decide to (omitted ¼ no) 1. Introduce new task Joint decision (1991) Yes (1980)/decide on own (1991) Adj. R2 2. Slow work pace Joint decision (1991) Yes (1980)/decide on own (1991) Adj. R2 R’s policy inﬂuence over (omitted ¼ none) 3. Work methods Give advice Participate in decision-making Adj. R2 4. Change product/service Give advice Participate in decision-making Adj. R2 5. Work pace (1980 only) Give advice Participate in decision-making Adj. R2 6. Budget allocation Give advice Participate in decision-making Adj. R2 7. Stafﬁng levels Give advice Participate in decision-making Adj. R2

1980

NA 0.0057 0.4171 NA 0.0680 0.4163

1991

0.0664 0.0894 0.3824 0.0148 0.0615 0.3782

0.0347 0.1301 0.4187

0.0427 0.1051 0.3797

0.0137 0.0345 0.4149

0.0143 0.1471 0.3806

0.0393 0.0771 0.4158 0.1085 0.0475 0.4157 0.0725 0.0537 0.4154

0.0181 0.1115 0.3968 0.0621 0.1527 0.3794

Note: Dependent variables are ln (annual income/weekly hours worked) (1980) and ln (wage) (1990/1991). All models include controls for education, experience and its square, female, black, union, part-time, four occupational groups, whether R exercises supervisory duties, organizational position, whether employer is for-proﬁt, one-digit industry, tenure, and marital status and its interaction with female. Models for 1990/1991 also include controls for establishment size. Item 5 not asked in 1991. NA, not applicable. po0.10. po0.05.

74

MICHAEL J. HANDEL

Table 3. OLS Regression Estimates of Effects of Non-Management Decision-Making Inﬂuence on Pay, Labor Market Survey (1991) (NonManager Sample Only). Non-Managers’ Decision-Making Inﬂuence over (Omitted ¼ None) 8. Task assignment Very little Some A lot Adj. R2 N 9. Organizing work Very little Some A lot Adj. R2 N 10. New technology Very little Some A lot Adj. R2 N 11. Changes in products Very little Some A lot Adj. R2 N

0.1151 (0.0660) 0.0883 (0.0598) 0.0826 (0.0876) 0.2612 434 0.1014 (0.0787) 0.0398 (0.0703) 0.0899 (0.0875) 0.2639 438 0.0542 (0.0525) 0.0707 (0.0484) 0.0722 (0.0798) 0.2857 602

12. Discipline Very little Some A lot Adj. R2 N 13. Pay levels Very little Some A lot Adj. R2 N 14. Lay-off policies Very little Some A lot Adj. R2 N

0.0308 (0.0448) 0.0746 (0.0547) 0.1886 (0.1017) 0.2912 600 0.0490 (0.0512) 0.1022 (0.0703) 0.2497 (0.1150) 0.2896 604 0.0619 (0.0606) 0.0066 (0.1035) 0.0165 (0.1207) 0.2600 435

0.0015 (0.0508) 0.0291 (0.0503) 0.1414 (0.0801) 0.2861 603

Note: Dependent variable is ln (wage) (1990/1991). All models include controls for education, experience and its square, female, black, union, part-time, four occupational groups, whether R exercises supervisory duties, organizational position, establishment size, whether employer is for-proﬁt, one-digit industry, tenure, and marital status and its interaction with female. Differences in sample sizes reﬂect ballot structure of survey as well as slightly varying missing value rates. po0.10. po0.05.

The Effect of Participative Work Systems on Employee Earnings

75

Participation in policy decisions regarding work pace (1980 only) is not signiﬁcantly related to wages, nor is participation in policy decisions regarding the distribution of funds within the overall budget. This last ﬁnding, which appears inconsistent with results for a somewhat similar variable in the GSS, actually masks the fact that the interaction with occupational group indicates a substantial effect of budgetary inﬂuence on the earnings of managers and professionals (b ¼ 0.21, po0.05) in 1991. Combining all of the personal policy inﬂuence variables into a single standardized, additive scale (a ¼ 0.86 [1980], 0.76 [1990]) yields no signiﬁcant effects in 1980 and a signiﬁcant effect for only managers and professionals in 1991 (b ¼ 0.047, po0.05). Non-management decision-making influence. Respondents to the 1991 survey were asked if non-management employees have inﬂuence over various decisions in their workplace. Since the question refers to non-managers, the sample for these analyses is restricted to that group (Table 3). There is no strong evidence that non-managers’ inﬂuence over task assignments, the introduction of new ways of organizing work, the introduction of new technology, or lay-off policies improves their wages. Since these activities represent some of the distinctive features of high-performance work systems, the absence of effects is notable. Qualifying this conclusion is evidence that when non-managers have a lot of inﬂuence over changes in products/services and employee discipline they do receive a wage premium of 14–19% (po0.10). The strongest effect on wages is non-management inﬂuence over pay decisions (b ¼ 0.25, po0.05), an area in which non-management employees report the least inﬂuence; the mean response is between ‘‘no’’ and ‘‘very little’’ inﬂuence. Interaction models indicate that these results are driven mostly by the wage premium enjoyed by a small number of blue-collar/ service workers. When all of these variables are combined in a single standardized, additive scale (a ¼ 0.78), they have a slight positive impact on earnings (b ¼ 0.038, po0.10). These two datasets have the advantage of using a wide range of very speciﬁc measures of decision-making, rather than more general, subjective evaluations of job characteristics. They are some of the most direct indicators of workplace participation available in nationally representative datasets. However, the results are only modestly more favorable toward the view that high-performance work practices have the potential for reducing wage inequality, though some variables, such as respondent’s inﬂuence over changes in the product and non-management inﬂuence over pay

76

MICHAEL J. HANDEL

levels, are associated with a substantial wage differential for less-skilled occupational groups.

5. CONCLUSION Current theories of organizational and workplace transformation debate whether participative work systems represent a new model of high-skill/ high-wage employment or merely a disguised effort to intensify job demands. Even researchers who agree that participation raises skills and wages, researchers differ on its distributional effects. Some believe they rescue workers from a low wage fate while others believe they contribute to the growth in inequality by raising skill demands so high they leave many less skilled workers behind. It appears that all of these views exaggerate the consequences of highperformance work practices, at least with respect to possible wage effects. This paper has examined the effects of numerous measures of autonomy, job enrichment, participation in decision-making, contingent compensation, workplace cooperation and control, and other aspects of job design that have been the focus of recent debates. Very few measures have any consistent impact on wages, positive or negative, particularly for lower-paid occupational groups that are the main focus of concern. The results generally do not support the optimistic position that extension of these practices holds great potential for raising workers’ wages, nor do they suggest that high-performance work practices are likely to exacerbate what some believe is a skills mismatch that is responsible for the growth of wage inequality. Likewise, various employer control techniques do not seem to be associated with lower wages, despite predictions by segmented labor market theory and more recent critics of workplace participation. While recent innovations in human resource practices may be desirable for many reasons, such as increased job satisfaction, there is the strong possibility that skill upgrading, gift exchange, and other changes entailed by the use of high-performance work practices are simply too modest or sufﬁciently offset by workers’ weak bargaining power to inﬂuence wages signiﬁcantly in most cases. The general sparseness of signiﬁcant effects suggests that high-performance work practices may have limited potential for raising the wages of less-skilled workers. It is possible that more recent efforts have overcome past limitations, but the present results are consistent with a number of more recent studies using different measures discussed previously. However, issues of measurement error, unobserved heterogeneity, and limited sample sizes are problems in almost all studies and argue for caution.

The Effect of Participative Work Systems on Employee Earnings

77

In order to gain greater certainty with respect to these questions, future research will need better data to go beyond the methods available for this and other existing studies. An ideal dataset would include panel data from matched employer–employee surveys that would allow for measurement error correction for the indicators of job characteristics and to control for unobserved employee and organizational characteristics using ﬁxed effects models. Until such data are available, all conclusions in this area of the sociology of work will remain provisional.

NOTES 1. The root transformation is better than the more usual log transformation because of the large number of cases with zero values that would have undeﬁned values if logs were used. 2. All coefﬁcients are insigniﬁcant when entered jointly in 1991 and most results were little changed in the joint model for 1980.

ACKNOWLEDGMENTS I would like to thank Christopher Jencks, Michael Piore, Lee Rainwater, Christopher Winship, Erik Olin Wright, Jonathan Zeitlin, and participants in the University of Wisconsin Economic Sociology seminar for their helpful comments on earlier versions of this paper. I am responsible for any remaining problems.

REFERENCES Abrahamson, E., & Fairchild, G. (1999). Management fashion: Lifecycles, triggers, and collective learning processes. Administrative Science Quarterly, 44, 708–740. Adler, P. S. (1993). Time-and-motion regained. In: R. Howard (Ed.), The learning imperative (pp. 255–275). Cambridge, MA: Harvard Business Review Book. Akerlof, G. A. (1982). Labor contracts as partial gift exchange. Quarterly Journal of Economics, 97, 543–569. Appelbaum, E., Bailey, T., Berg, P., & Kalleberg, A. L. (2000). Manufacturing advantage: Why high performance work systems pay off. Ithaca, NY: Cornell University Press. Appelbaum, E., & Batt, R. (1994). The new American workplace. Ithaca, NY: ILR Press. Arthur, J. B. (1992). The link between business strategy and industrial relations systems in American steel minimills. Industrial and Labor Relations Review, 45, 488–506. Bailey, T. R., & Bernhardt, A. D. (1997). In search of the high road in a low-wage industry. Politics and Society, 25, 179–201. Barker, J. R. (1993). Tightening the iron cage: Concertive control in self-managing teams. Administrative Science Quarterly, 38, 408–437.

78

MICHAEL J. HANDEL

Batt, R. (2001). Explaining wage inequality in telecommunications services: Customer segmentation, human resource practices, and union decline. Industrial and Labor Relations Review, 54, 425–449. Berggren, C. (1994). NUMMI vs. Uddevalla. Sloan Management Review, Winter, 37–49. Blasi, J., Conte, M., & Kruse, D. (1996). Employee stock ownership and corporate performance among public companies. Industrial and Labor Relations Review, 50, 60–79. Blasi, J., Kruse, D., Sesil, J., Kroumova, M., & Carberry, E. (2000). Stock options, corporate performance, and organizational change. Oakland, CA: National Center for Employee Ownership. Bluestone, B., & Bluestone, I. (1992). Negotiating the future: A labor perspective on American business. New York: Basic Books. Braverman, H. (1974). Labor and monopoly capital. New York: Monthly Review. Bresnahan, T. F., Brynjolfsson, E., & Hitt, L. M. (2002). Information technology, workplace organization, and the demand for skilled labor: Firm-level evidence. Quarterly Journal of Economics, 117, 339–376. Cappelli, P., Bassi, L., Katz, H., Knoke, D., Osterman, P., & Useem, M. (1997). Change at work. Oxford University Press. Cappelli, P., & Rogovsky, N. (1994). New work systems and skill requirements. International Labour Review, 133, 205–220. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational ﬁelds. American Sociological Review, 48, 147–160. Edwards, R. (1979). Contested terrain: The transformation of the workplace in the twentieth century. New York: Basic Books. Fantasia, R., Clawson, D., & Graham, G. (1988). A critical view of worker participation in American industry. Work and Occupations, 15, 468–488. Florida, R., & Kenney, M. (1991). Transplanted organizations: The transfer of Japanese industrial organization to the U.S. American Sociological Review, 56, 381–398. Frazis, H., Gittleman, M., & Joyce, M. (2000). Correlates of training: An analysis using both employer and employee characteristics. Industrial and Labor Relations Review, 53, 443–462. Freeman, R. B., & Kleiner, M. M. (2000). Who beneﬁts most from employee involvement: Firms or workers? American Economic Review, 90, 219–224. Garson, B. (1988). The electronic sweatshop: How computers are transforming the office of the future into the factory of the past. New York: Simon and Schuster. Graham, L. (1993). Inside a Japanese transplant: A critical perspective. Work and Occupations, 20, 147–173. Handel, M. J., & Levine, D. I. (2004). The effects of new work practices on workers. Industrial Relations, 43, 1–43. Harrison, B., & Bluestone, B. (1988). The great u-turn: Corporate restructuring and the polarizing of America. New York: Basic Books. Hodson, R. (2001). Dignity at work. New York: Cambridge University Press. Hodson, R. (2002). Worker participation and teams: New evidence from analyzing organizational ethnographies. Economic and Industrial Democracy, 23, 491–528. Hunter, L. W., & Lafkas, J. J. (2003). Opening the box: Information technology, work practices, and wages. Industrial & Labor Relations Review, 56, 224–243. Kalleberg, A. L., & Marsden, P. V. (1995). Organizational commitment and job performance in the U.S. labor force. Research in the Sociology of Work, 5, 235–237.

The Effect of Participative Work Systems on Employee Earnings

79

Kandel, E., & Lazear, E. P. (1992). Peer pressure and partnerships. Journal of Political Economy, 100, 801–817. Katz, L. F., & Murphy, K. M. (1992). Changes in relative wages, 1963–1987: Supply and demand factors. Quarterly Journal of Economics, 107, 35–79. Kelley, M. R. (1996). Participative bureaucracy and productivity in the machined products sector. Industrial Relations, 35, 374–399. Kruse, D. L. (1992). Proﬁt sharing and productivity: Microeconomic evidence from the United States. The Economic Journal, 10, 24–36. Loveman, G. W., & Tilly, C. (1988). Good jobs or bad jobs: What does the evidence say? New England Economic Review, January/February, 46–65. Mishel, L., Bernstein, J., & Allegretto, S. (2005). State of working America 2004/2005. Ithaca, NY: Cornell University Press. Murnane, R. J., & Levy, F. (1996). Teaching the new basic skills. New York: Free Press. National Center on Education and the Economy. (1990). America’s choice: High skills or low wages!. Rochester, NY: National Center on Education and the Economy. Osterman, P. (1994). How common is workplace transformation and who adopts it? Industrial and Labor Relations Review, 47, 173–188. Osterman, P. (2000). Work reorganization in an era of restructuring: Trends in diffusion and effects on employee welfare. Industrial and Labor Relations Review, 53, 179–196. Piore, M. J., & Sabel, C. (1984). The second industrial divide. New York: Basic Books. Smith, V. (1997). New forms of work organization. Annual Review of Sociology, 23, 315–339. Tausky, C., & Chelte, A. F. (Eds). (1988). Workers’ participation. Work and Occupations, 15(special issue), 363–489. United States Congress, Ofﬁce of Technology Assessment. (1987). The electronic supervisor: New technology, new tensions. Washington, DC: Government Printing Ofﬁce. United States Department of Health, Education, and Welfare. (1973). Work in America: Report of a special task force to the secretary of health, education, and welfare. Cambridge, MA: MIT Press. Walton, R. E. (1974). Alienation and innovation in the workplace. In: J. O’Toole (Ed.), Work and the quality of life (pp. 227–245). Cambridge, MA: MIT Press. Walton, R. E. (1985). From control to commitment in the workplace. Harvard Business Review, 63, 77–84. Williamson, O. E. (1985). The economic institutions of capitalism. New York: Free Press. Zbaracki, M. J. (1998). The rhetoric and reality of total quality management. Administrative Science Quarterly, 43, 602–636. Zuboff, S. (1988). In the age of the smart machine: The future of work and power. New York: Basic Books.

80

MICHAEL J. HANDEL

APPENDIX The items and scales used as independent variables are listed in brief form (see Tables A1–A3) Table A1.

Indicators of Participative Work Systems from General Social Survey (GSS) (1991).

Scales and Itemsa Participation, skill, and task content 1. Autonomy/participation (standardized scale) (a ¼ 0.80) a. Who decides what you do on job and how you do itb b. I can work independently c. I have a lot to say over what happens on my job d. Job allows me to take part in making decisions that affect my work 2. Five-task scalec (standardized scale) (a ¼ 0.75) a. Organize schedules or tasks (1 ¼ yes) b. Devise new ways of doing tasks (1 ¼ yes) c. Check workers’ progress or attendance (1 ¼ yes) d. Evaluate job performance (1 ¼ yes) e. Recommend promotions or disciplinary actions (1 ¼ yes) 3. Spending decisions In last year, did you participate in decisions about spending any of your organization’s money as part of your job (1 ¼ yes) 4. Level of expenditure About how much money involved in the largest decision ($1991)d 5. Importance of education How important was formal education as source of skills you use on jobe 6. Importance of training How important was formal on-the-job training where you work now as source of skills you use on jobe Workplace control 7. Technical control How I do job is controlled by machines I work with 8. Computer monitoring Computers monitor my job performance Compensation 9. Profit sharing Eligible to receive proﬁt sharing/stock options (1 ¼ yes) 10. Performance bonus Eligible to receive cash or stock bonuses for performance or merit (1 ¼ yes) 11. Uniform pay raises Where you work, are raises given to all regardless of performance (1 ¼ yes)

The Effect of Participative Work Systems on Employee Earnings

81

Table A1. (Continued ) Scales and itemsb Cooperation at work: Organizational climate 12. Labor–management relations How would you describe relations in your workplace between management and employees?f 13. Vote for union In a secret ballot, would you vote for or against union representation (1 ¼ favor) 14. Flextime Eligible for ﬂextime (1 ¼ yes) 15. Peer relations How would you describe relations in your workplace between coworkers?f Cooperation at work: Individual commitment 16. Voluntary effort I am willing to work harder than I have to in order to help organization succeedg 17. Actual effort Compared to others doing similar work, how much work would you say you do? (emphasis original)h 18. Likelihood of quitting How likely you will try hard to ﬁnd new job with another organization within next 12 months (1 ¼ very likely)i 19. Organizational identification (standardized scale) (a ¼ 0.72)j a. I am proud to be working for this organization b. My values and the organization’s values are very similar c. I would turn down another job for more pay in order to stay with this organization d. I would take almost any job to keep working for this organization e. I feel very little loyalty to this organizationk f. Success of my organization depends a lot on how well I do my job a

Unless otherwise noted, all variables coded on a four-point scale (1 ¼ not true at all, 2 ¼ somewhat true, 3 ¼ true, 4 ¼ very true). b Responses coded: 1 ¼ someone else decides what I do and how I do it, 2 ¼ someone else decides what I do, but I decide how I do it, 3 ¼ I have some freedom in deciding what I do and how I do it, 4 ¼ I am basically my own boss. c Items ask: ‘‘Which work activities do you do with employees whom you do not ofﬁcially supervise?’’ (1 ¼ yes). d All ﬁgures calculated only for those reporting participation in spending decisions. e Coded: 1 ¼ not at all important, 2 ¼ somewhat important, 3 ¼ important, 4 ¼ very important. f Coded: 1 ¼ very bad, 2 ¼ quite bad, 3 ¼ neither good nor bad, 4 ¼ quite good, 5 ¼ very good. g Coded: 1 ¼ strongly disagree, 2 ¼ disagree, 3 ¼ agree, 4 ¼ strongly agree. h Coded: 1 ¼ much less, 2 ¼ somewhat less, 3 ¼ about the same, 4 ¼ somewhat more, 5 ¼ much more. i Coded: 1 ¼ very likely, 2 ¼ somewhat likely, 3 ¼ not at all likely. j All items coded: 1 ¼ strongly disagree, 2 ¼ disagree, 3 ¼ agree, 4 ¼ strongly agree. k Reverse coded when included in scale.

82

MICHAEL J. HANDEL

Table A2. Indicators of Participative Work Systems from the Class Structure Survey (1980) and Labor Market Survey (1991). Variables Respondent’s job autonomy/participation Can you decide to:a 1. Introduce a new work task 2. Slow work pace for a day Respondent’s policy inﬂuence Are you personally involved in policy decisions regarding:b 3. Basic methods or procedures of work 4. Signiﬁcant change in products/services 5. Routine work pace 6. Distribution of funds within overall budget 7. Increase/decrease number employed where you work Non-management inﬂuence on decision-making Where you work, how much influence do non-management employees have on:c 8. Task assignments 9. Introducing new ways of organizing work 10. Introducing new technology 11. Making changes in products/services 12. Discipline policies 13. Deciding pay levels for different jobs 14. Deciding lay-off policies a Codes for 1980 are 0 ¼ no and 1 ¼ yes. Codes for 1991 are 0 ¼ someone else decides, 1 ¼ joint decision, 2 ¼ decide on my own. b Codes for 1980 and 1991 are 0 ¼ do not participate, 1 ¼ give advice, 2 ¼ participate directly in decision-making, but for 1980 this collapses a ﬁve-category response set. Minor differences in question wording between years. Item 5 not asked in 1991. c These items not asked in 1980. Codes for 1991 are 1 ¼ no inﬂuence, 2 ¼ very little inﬂuence, 3 ¼ some inﬂuence, 4 ¼ a lot of inﬂuence.

The Effect of Participative Work Systems on Employee Earnings

Table A3.

83

OLS Regression Coefﬁcients of Effects of Baseline Variables on Wages.

Variable

General Social Survey (1991) 1

Education Education (years) oHigh school Some college BA Graduate Experience Experience2/100 Female Black Union Part-time Hourly worker Occupation (omit ¼ manager) Clerical Craft Blue-collar/service Hiring screens Intelligence test Skills test Drug test Physical Letters of reference Organizational position (omit ¼ nonmanager) Top management Upper management Middle management Lower management Supervisor Supervisory duties

0.0705

2 0.0715

Wright Surveys 1980

1991

0.1529 0.0880 0.2964 0.3543 0.0179 0.0228 0.2205 0.0809 0.1565 0.0105

0.0276 0.0404 0.1465 0.0388 0.0033 0.0053

0.0284 0.0409 0.1409 0.0485 0.0170 0.0293

0.2620 0.0123 0.3537 0.2427 0.0464 0.0744 0.2201 0.0232 0.2390 0.2332

0.2515 0.1596 0.3994

0.2455 0.1690 0.4100

0.0828 0.1260 0.2526

0.1021 0.0947 0.2793

0.2426 0.1382 0.2176 0.1460 0.0464 0.0825

0.1004 0.2128 0.1250 0.0724 0.0540 0.1015

0.0436 0.0609 0.0171 0.1217 0.0251

84

MICHAEL J. HANDEL

Table A3. (Continued ) Variable

General Social Survey (1991) 1

Adj. R2 N

0.3195 684

2 0.3195 681

Wright Surveys 1980

1991

0.4158 1,037

0.3797 1,313

Note: Part-time deﬁned as working less than 35 hours per week. Omitted occupational group is managers, professionals, and technical workers. GSS: Dependent variable is annual earnings (1990) divided by current weekly hours (1991). All models also include controls for tenure, one-digit industry, establishment size, part-time status, married, married female, region, and resident of metropolitan area. Wright Surveys: Dependent variables are ln (annual earnings/usual weekly hours worked) (1980), using midpoints of annual earnings categories, and ln (hourly wage) (1991). Both models control for tenure, one-digit industry, marital status and its interaction with gender, and whether employer is for-proﬁt. The 1991 model also controls for establishment size. The omitted category for organizational position is non-management. The variable for supervisory duties equals one for those answering afﬁrmatively: ‘‘As an ofﬁcial part of your job do you supervise the work of other employees or tell other employees what to do?’’ po0.10. po0.05. po0.01. po0.001.

GENDERED CONTRADICTIONS: MANAGERS AND WOMEN WORKERS IN SELF-MANAGED TEAMS Nancy Plankey Videla ABSTRACT Organizational literatures stress the empowering effects of worker participation programs. The case of a Mexican garment factory is used to examine the contradictory location of women in self-managed teams. While self-managed teams require independent and assertive workers, women workers are hired specifically for their docility. I argue that managers provide the tools and mechanisms for workers to be autonomous decision-makers, while at the same time they gender teams in ways that assure continued female disadvantage. Placed in this contradictory location, women workers both reproduce and resist gender subordination by carving out spaces of independent action, using the language of traditional womanhood.

Organizational literatures stress the empowering effects of worker participation programs, of which self-managed teams represent one of the most

Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 85–116 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16004-5

85

86

NANCY PLANKEY VIDELA

complete examples (Appelbaum, Thomas, Berg, & Kalleberg, 2000; Florida & Kenney, 1991; Yeatts & Hyten, 1998). Self-managed teams have the explicit goal of converting workers into independent thinkers and actors. The notion of the team member – although posed as gender-neutral – invariably implies a disembodied male individual who possesses the masculine qualities of assertiveness, leadership, and independence (Acker, 1990; Jensen, 1989). How do women workers – hired because they conform to management’s perception as docile workers – accommodate to the contradictory managerial directive to be assertive and independent workers? The case of selfmanaged teams in a Mexican garment factory, where 70% of the workers are women, provides an opportunity to examine the contradictory location of women in sex-segregated occupations organized into self-managed teams. This article examines how managers construct self-managed teams and women’s roles in them, as well as how workers respond to these new roles. I argue that both managers and workers are active subjects who shape and reshape the contours of self-managed teams. Although managers provide the tools and mechanisms for workers to be autonomous decision-makers, they also gender teams in ways that assure continued female disadvantage. Placed in this contradictory location, women workers both reproduce and resist gender subordination. By incorporating new attributes of assertiveness into their work roles, women carve out spaces of independent action, yet they do so in defense of, and using the language of, traditional womanhood. This study is based on the case study of Moctezuma,1 a Mexican garment ﬁrm in Central Mexico established in 1956. Using typical Taylorlist assembly lines, Moctezuma rose to the top of the high-end men’s suits segment in Mexico. Reorganized in 1996 under Japanese-inspired lean techniques to more fully participate in the global garment commodity chain, Moctezuma functioned around the key tenets of just-in-time production, quality-at-thestart, and self-managed teams. Although the shift to lean production did not include changes in the workforce or management, the 1400 workers – 70% of whom were women – were now organized into teams that produced a discrete portion of the garment (a semi-product) focusing on production with quality. An international consulting ﬁrm oversaw the transition to lean production, in collaboration with a cadre of high-level managers whom the consultants had trained. According to Hobday (1995), this is the most effective way to implement lean production at an already established factory, termed a brown-ﬁeld site. Crucial to the reorganization of production at Moctezuma was the ideology of ‘‘community of fate’’ (Cole, 1979) that helped workers internalize the ﬁrm’s interests as their own. I have elsewhere argued (Plankey Videla, 2006) that management constructed the ideology of

Gendered Contradictions: Managers and Women Workers

87

the ﬁrm as a ‘‘community of fate,’’ a manufactured belief that workers and managers have equal and intertwined interest in the ﬁrm’s success. High wages, expanded corporate welfare programs, and rituals of the ﬁrm as a family buttressed this ‘‘community of fate’’ ideology.2 This article focuses solely on the recent period of self-managed teams at Moctezuma, examining the ways in which management both empowered and curtailed women’s work and the ways in which workers maintained and subverted those gendered practices to respond to their own interests. The ﬁrst part of this article discusses the methods, the literature on the gendering of jobs and sex segregation, and the literature on the practice of teamwork. The second part of the article examines how management directed workers to take on more male characteristics of assertiveness and independence, while simultaneously sex-typing jobs as female through the recruitment, training, and job placement processes. The third part of the paper examines women’s participation in – and resistance to – this contradictory location, looking at team dynamics and labor–management relations. I conclude by drawing attention to wider implications of the study and fruitful areas for further research.

METHODS The data analyzed here are from an ethnographic research project at a successful Mexican garment ﬁrm in Central Mexico, where I labored as a worker for nine months between 2000 and 2001. For this research, I also interviewed 26 managers, 25 workers, and 3 policy-makers. Managers were interviewed at the ﬁrm during work hours. Workers on the other hand, were all interviewed off company premises at a location of their choice; sometimes in their home, other times at restaurants. On return visits in the summers of 2002 and 2004, I re-interviewed both workers and managers. Interviews with policy-makers were at carried out at their convenience, two at their homes and a third in government ofﬁces. All interviews, which I conducted in Spanish as I am a native Spanish-speaker, were tape-recorded and later transcribed. I secured permission to study and work at the factory from top-level managers who shared both my academic interests in Japanese-style lean techniques and connections to the University of Wisconsin at Madison where I was a graduate student. Given our shared professional networks and academic background, I was given open access to the factory and its personnel. I had the added advantage of being able to analyze approximately one thousand worker ﬁles.

88

NANCY PLANKEY VIDELA

I chose participant observation because I wanted to provide a worker’s perspective on teamwork, which is largely missing from the organizational literature. As a feminist ethnographer, my aim was to make links between macro- and micro-social processes and show how larger power dynamics affect workers’ – especially women workers’ – everyday lives. This required me to be overt about my study, reﬂexive about my knowledge and position, and straightforward about what I would do with the results. Furthermore, it meant allowing others to choose whether or not to participate in the study, although I recognize that power relations at times made such a choice nearly impossible. For example, it is possible that workers did not feel they could refuse my request to interview them. Despite being uncomfortable, being reﬂexive also meant naming the privileges of class, ethnicity and nationality that set me apart from workers and at times made me feel closer to management. Although my worldview and sympathies were with the workers, the fact that my son went to the same school and saw the same doctor as several of the children of management underscored my class privilege. Furthermore, I could leave the factory at any time, choose to not go to work when I was sick, and take a break when my partner could no longer stay home to care for our child. Within these limitations, I secured permission to work and do research at the factory from management, work teams, and the union, providing both myself and those interviewed the opportunity to dialogue, and analyze the processes of teamwork.

Gender and Development, Sex Segregation, and Gendered Organizations Theories Research on the international division of labor has long demonstrated that men and women experience manufacturing employment differently. Although high percentages of women are the primary or sole provider for their families, they are often paid lower wages than their male counterparts because they are constructed as supplemental earners (Collins, 2002, 2003; Elson & Pearson, 1986; Ferna´ndez-Kelly, 1983; Nash & Ferna´ndez-Kelly, 1983; Ong, 1987; Ward, 1990; Wolf, 1992). However, as Barajas Escamilla and Sotomayor Yala´n (1995) demonstrate for Mexico, wages earned by female heads of households and daughters are not supplementary but crucial for household survival. Tiano (1994) supports this ﬁnding. Although one third of the women workers in the Mexicali electronics and apparel maquilas in her study were heads of households, the wages of the

Gendered Contradictions: Managers and Women Workers

89

majority of women workers were also indispensable for household survival strategies. Researchers have found that employers further justify lower wages by denoting women as ‘‘naturally’’ suited for the apparel industry. Sewing, for example, has been traditionally deﬁned as women’s domestic work, a claim which is then used to mark women’s employment in apparel ﬁrms as an extension of the domestic sphere and to naturalize their skills. Higher wages for sewing are not warranted, the argument goes, since little training is required beyond what women have learned since they were little girls at home (Collins, 2002). Moreover, employers frame women as docile, undemanding, nimble-ﬁngered, and uninterested in participating in unions (Chow, 2002; Collins, 2003; Elson, 1983; Sklair, 1993). However, Stoddard (1987) in Ward, 1990) indicates that women score higher on the dexterity tests precisely because such tests are set up to measure gender-speciﬁc skills. Management and corporations utilize gender ideologies to justify lower wages and to justify relegating women to particular jobs. Milkman (1987) notes that during World War II, employers hired women in labor-intensive industries like electronics and garments, not for their special skills, but because they could pay women lower wages. Sklair argues that transnational corporations employ women to lower the wage levels of the entire population. By claiming to employ women because of their ‘‘feminine sensibilities’’ (read: docile, non-unionized) it forces men to take on such qualities to become employable: to become the ‘‘ideal worker’’ (Sklair, 1993). Leslie Salzinger (2003) goes even further, arguing that managerial tropes and strategies produce different gendered subjectivities in different factories to ﬁt productive imperatives and managerial ideologies. Ching Kwan Lee (1998), on the other hand, points to the importance of different gender ideologies in the ways that management set up production regimes when dealing with ‘‘daughters’’ versus ‘‘matron’’ workers. Management’s manipulation of gender ideologies thus plays an important role in keeping women at the bottom rung of the gendered international division of labor. Gender and work theorists have made a similar case for the sex-segregation of labor markets, occupations, and jobs. According to the sex segregation literature, the devaluation of women’s domestic role is imported into the workplace, with a concomitant devaluation of prestige and pay (England, Herbert, Kilbourne, Reid, & Magdal, 1994). Some researches try to explain how certain jobs are denoted as women’s work or men’s work and the effects of these assumptions on the wage gap (Beneria & Stimpson, 1987; Hartmann, 1983; Milkman, 1987; Morgan & Knights, 1991). These latter theorists pay special attention to the ways that gender ideologies of

90

NANCY PLANKEY VIDELA

employers, male workers, unions, and state agents mark and perpetuate gender inequalities at work. Milkman’s (1982, 1987) work on the electronics and auto industry in the U.S. during World War II is noteworthy for its historical look at the ideology of sex-typing, which ‘‘constructs women’s ‘primary’ commitment as devotion to home and family, whether or not they also work for pay’’ (1982, p. 340). Starting off with the assumption that women were less committed to work, employers deployed historical and industry-speciﬁc idioms of sex-typing that labeled certain jobs as male and others as female. Not surprisingly, in the industries which were labor intensive, like the electronics or garment industry, greater percentages of women were employed, justiﬁed through the sex-typing idiom of women as having more dexterity and manual ability. As Milkman (1987) demonstrates, sex-typing leads to jobs being segregated by sex, which has resulted in remarkably durable and stable gender inequalities and ideologies. More recently, gender theorists have explored how organizations perpetuate gender segregation by structuring gender differences into organizational practices and assumptions (Acker, 1990, 1992, 1998; Britton, 2000; Hall, 1993; Hearn & Parkin, 1992; Hochschild, 1983; Pierce, 1995; Skuratowicz & Hunter, 2004; Smith, 1993, 1997; Williams, 1995). This research is informed by the important work of Joan Acker (1990, 1992, 1998) who understands ‘‘gendered institution’’ to mean ‘‘y that gender is present in the processes, practices, images and ideologies, and distributions of power in the various sectors of social life’’ (1992, p. 567). Given the prominence of teamwork and self-management in the late twentieth and early twenty-ﬁrst centuries, it is important to understand their gendered causes, consequences, and processes. While gender theorists have examined teams and gender (Lamphere, Zavella, & Gonzalez with Evans, 1993; Salzinger, 2003; Smith, 1993), few studies have expressly focused on the special circumstances that self-managed teams present to women and men at work. My work specifically aims to build upon gender theorists’ work to contribute to the burgeoning debate on self-managed teams and gender. This understudied yet growing ﬁeld of inquiry includes the work of Ollilainen (1999), Metcalfe and Linstead (2003), and Salzinger (2003). The ﬁrst two works take on organizational theorists, disputing their claims of gender neutrality in teamwork. Ollilainen’s (1999) unpublished dissertation compared four self-managed teams of mixed-sex composition to unveil the salience of gender in teams and the ways in which teamwork is gendered. She found that, although using a language of equality, self-managed teams reverted to a gendered division of labor where women performed most of the emotional and clerical labor. Metcalfe and Linstead (2003), on the other

Gendered Contradictions: Managers and Women Workers

91

hand, examine teamwork manuals and theories, ﬁnding that teamwork is theorized not as gender-neutral but rather as embodied masculinity. Salzinger’s (2003) work on four maquilas in Ciudad Jua´rez presents the case study of auto-parts assembler ‘‘Particimex,’’ which is organized around teams. Although she does not explicitly denote them as self-managed, her description of their responsibilities and purview vis-a`-vis management would so designate them. Salzinger argues that following managerial tropes, managers hired young women explicitly for their docility and because they could pay them cheap wages. When production was organized around selfmanaged teams at Particimex, women workers’ identities were ﬂexible in that they unproblematically assumed the role of assertive workers at the factory, even though gender relations at home remained traditional. This article focuses on a case similar to that of Salzinger’s Particimex – a shop-ﬂoor consisting predominantly of women, hired for their purported docility, and reorganized into self-managed teams where they were expected to take on masculine traits of assertive and independent leadership. However, working on the shop-ﬂoor for an extended period of time,3 allowed me to ask more speciﬁc questions: How were self-managed teams gendered by management? And how did women workers respond to their contradictory location on the shop-ﬂoor? In other words, how did gender permeate selfmanaged teams at Moctezuma?

MANAGEMENT’S ROLE IN SELF-MANAGED TEAMS: GENDERING TEAMS Management theorists (Katzenbach & Smith, 1993; Womack, Jones, & Roos, 1990; Yeatts & Hyten, 1998) and sociologists (Appelbaum et al., 2000; Appelbaum & Batt, 1994; Barker, 1993; Besser, 1996; Florida & Kenney, 1991; Graham, 1995; MacDufﬁe, 1995) depict teamwork as a gender-neutral process. However, there are key decisions at each step establishing teamwork practices which are informed by managers’ and workers’ gender ideologies, denoting some jobs as women’s work and others as men’s work. Moreover, reorganizing a ﬁrm around teamwork has differential consequences for women and for men. I will illustrate these gendered processes by examining recruitment, worker training, and worker placement. At Moctezuma, self-managed teams had clearly demarcated obligations and rights. At the core of the team system, teams were responsible for producing a pre-set volume of garments with the requisite high-level quality.

92

NANCY PLANKEY VIDELA

Teams were organized by dividing the assembly line into semi-products, which were then designated to teams of 10–20 workers. Management hierarchies were ﬂattened, doing away with line supervisors and a portion of middle-managers, replacing them with departmental level ‘‘advisors.’’ Given that the model of self-managed teams assumed a highly skilled workforce, the role of advisors was to assist teams when needed, follow work-ﬂow, and communicate issues to team leaders. Departmental managers, with the assistance of industrial engineers, provided each team with the exact calculations of how many movements per operation (in standard allotted minutes) were required to reach the daily quota. Theoretically, it was then the responsibility of the team to ‘‘manage itself’’ to assure that work was done according to volume and quality speciﬁcations. In order to meet their obligations, self-managed teams had the ability to exercise rights provided under the new lean production system. First, teams were autonomous. This meant that unless invited by team members, advisors were not allowed to meddle in team decisions. Any problems or questions had to be addressed directly with the team leader, who then spoke with the team member in question. Second, teams had the right to call meetings to decide how to get work done, solve questions dealing with production or stemming from interpersonal problems, or respond to advisor’s requests. Decisions were made after a brief discussion by a show of hands. Third, permission slips to be tardy or miss work were given by the team. It was also the team which decided whether the worker had to repay work hours or if the team would cover that person’s work quota. Fourth, each team was responsible for training new workers. This included deciding what operation the new worker would train, and who did the training. Fifth, teams had the ability to expel problematic workers if they did not respond to the team’s exhortations to improve working habits or interpersonal communication styles. Self-managed teams relied heavily on the team leaders who were to be the communication channel between workers and managers. Team leaders rotated according to seniority, beginning with a term of six months; but after the ﬁrst year this was revised so that leaders rotated every two months. It was the team leader’s responsibility to oversee the production process, provide advisors with production tallies every two hours, call attention to an individual’s bad quality or low productivity, mediate conﬂict between members, and meet weekly with other team leaders. Furthermore, team leaders called daily meetings to facilitate team discussions on how to assign work, cover absenteeism, provide permits, and allocate multi-skill training. In effect, team leaders ﬁgured prominently in self-managed teams.

Gendered Contradictions: Managers and Women Workers

93

The rules that governed self-managed teams at Moctezuma called for workers to be active participants with management in the daily functioning of the ﬁrm. For the system to work, individual workers had to be wellinformed about production matters so as to intelligently discuss how to solve issues as they came up, assertive to be heard over the clatter of machines and other worker’s opinions, and independent so as to vote with productive imperatives rather than individual interests at heart. However, the majority of workers in teams were women of all ages that had been hired not for their independence but their docility. Moctezuma, like most other garment ﬁrms, has always hired predominantly women. Since its inception the ﬁrm has preferred women workers, with management characterizing them as nimble-ﬁngered, patient, and dexterous. This is not new. Collins (2002, 2003), Elson and Pearson (1986), Ferna´ndez-Kelly (1983), Kopinak (1995, 1996), Ong (1987), and Salzinger (2003) have all found that these idioms and gender ideologies, which translate into preferring women for their docility, underpin management views. Notably, when Moctezuma reorganized as a lean ﬁrm with self-managed teams, the proﬁle of the preferred worker changed but the ﬁrm’s hiring practices did not change. While Milkman (1987) notes that once established in a ﬁrm or industry, sex-typing is remarkably stable and resilient, this does not fully explain management practices at Moctezuma. I will demonstrate how managers purposefully recreated the nimble-ﬁngered and docile woman worker of days past through recruitment, training, and job placement practices while at the same time placing women in the role of active decision-maker.

Recruitment: Selecting the Ideal Worker Although the Human Resources managers claimed that productive imperatives guided the selection of workers, gender ideologies underlying management’s recruitment efforts marked women as the ideal worker. This is evidenced by the idioms of women’s naturalized skill. Moreover, as recruitment ofﬁcers discussed their work, they did so in gendered terms. They framed job requirements as characteristics that enhanced productivity on the shop-ﬂoor. However, these characteristics were gendered in nature. Instead of highlighting education levels and previous employment, they sought women who had a stable home life and had social interaction skills. It is to those characteristics that I now turn. A key factor underpinning gendered recruitment efforts at Moctezuma was management’s claim that constructed women’s skill as ‘‘natural.’’ This

94

NANCY PLANKEY VIDELA

idiom led recruitment ofﬁcers to frame the ideal worker as female. In fact, up until 1996 only women were allowed to work as sewing machine operators. At that point, international consultants who were hired to reorganize the ﬁrm insisted on also hiring men. According to the consultants, excluding men went against the philosophy of hiring the most qualiﬁed worker, regardless of sex. Even though men were technically allowed to apply for all jobs after this intervention, other recruitment, training and placement mechanisms relegated men mostly to the higher-paying piece-rate operations in Cutting and Final Pressing. Despite the consultants’ advice, the long-standing tradition of segregating jobs by gender, initiated by management in the 1950s, still reigned. The head of quality for all Jackets, who had worked at the company for over 15 years explained: ‘‘women are naturally suited for sewing.’’ Another high-level male manager, who had begun work only two years earlier noted: I don’t know if it is genetic, but it is statistically proven that women have greater skill and ability with their hands. In this ﬁrm, as a garment ﬁrm, great skill is required. There are women that have more skill than others, but on average, women have greater skill than men. Many of the operations require small movements that give the garment piece that special quality. Women have shown to have greater patience in learning the operations, they are more social. There are few men who have the will to sit at a machine because of what others will say, like that they are fags (sic) or that in their house they were taught that only women sew. In my case, I sit down at the machines. I have always been curious about any kind of machine. But [preferring women] is really because of their skills.

This manager constructs women’s sewing skill as natural because women learn sewing at home. Men who sew, on the other hand, are often cast as homosexuals. Note how this manager defended himself against being labeled female or homosexual by constructing his interests in machines as a matter of intellectual curiosity, not natural skill. This high-level manager enunciated gender ideologies underlining managers’ beliefs about what kinds of workers are needed: female workers. The worker proﬁle, updated assiduously every year by the Human Resources staff, consisted of the following characteristics: age, sex, educational attainment, experience, and compound measures of life-cycle stage and personality. The manager responsible for hiring described it this way: There are two proﬁles. We have to watch for characteristics that affect production and medical requirements. The proﬁle that they must ﬁt is to have the ideal age (that they be over 25 years old), and preferably that they be female. We have seen that women have greater dexterity when it comes to sewing, however we do accept men. [They must] have junior high school [education]. The civil status does not matter, as long as they are stable, whether they are single or married. [What matters is] that you have a steady place to live,

Gendered Contradictions: Managers and Women Workers

95

live no more than an hour away, that you have someone to take care of your children, y and that you not plan to have a child in the next year and a half – so that you can be integrated into productive [work] lifey and experience. If they have [sewing] experience it’s very easy for them to get in.

According to this description, productive imperatives required mature young women who knew how to sew and did not have life circumstances which would interfere with work. Interestingly, the company did not care if the women were married or if they had children. Their main concern was whether the applicant lived with a parent or extended family who could provide childcare when needed. This view assumed most women would eventually become mothers. The greatest problem with hiring women was not motherhood but lack of stable childcare. The medical requirements asked for speciﬁc heights so that they comfortably ﬁt at the sewing machine and cutting tables. However, transcription of an interview I conducted with a manager uncovered gendered word usage not observable in English. The manager used feminine pronouns to describe the machinists (las maquinistas) and male ones for the cutters (los cortadores). This corresponded with the speciﬁed heights: shorter (women) for sewing machines and taller (men) for cutting tables. In fact, there were women and men of all heights in all departments. Another medical requirement included a prohibition on any kind of tattoo, supposedly an indication of anti-social behavior. When I asked how managers made the correlation between tattoos and anti-social behavior, they told me that people who had tattoos were likely to belong to a gang, or at the very least, to have poor judgment. In fact, a good number of male applicants were rejected by the doctors for having tattoos that had not been previously disclosed. Since men were more likely to have tattoos, men were more often excluded from the pool of employable workers. The effect of both of these medical requirements – height and absence of tatoos – justiﬁed an ideological preference for women in teams. The last medical requirement, while not found anywhere in writing, was certainly ofﬁcial. Every female applicant who made it through the ﬁrst phase of tests, and returned for the medical exam, underwent a pregnancy test. While this practice is illegal under Mexican law, it is common practice in the export assembly and manufacturing industry and has been for years (Ferna´ndez-Kelly, 1983). When I ﬁrst arrived at the factory, I asked a Human Resources manager about this common business practice. She denied it, underscoring the illegality of the practice. Eight months later, after having reviewed hundreds of worker ﬁles with the pregnancy test results attached, I asked her again. By this time, not only did I have evidence to

96

NANCY PLANKEY VIDELA

rebuke her denial, but we also had established a friendly working relationship. She told me that the pregnancy exams were still performed, but were no longer reported in the ﬁle after a complaint was made against the ﬁrm for this practice. Thus, the medical worker proﬁle included physical and social parameters, with tattoos and pregnancy tests being measures of both. Under Mexican law, women are permitted twelve weeks of paid maternity leave (Colectivo Atabal A.C., 1994). Moreover, in an effort to expand corporate welfare beneﬁts to tie workers to the ﬁrm under the ‘‘community of fate’’ construct, Moctezuma allowed women to work at 70% efﬁciency after the sixth month of pregnancy and to take off an hour early from work while nursing. Thus, although managers predicted that women would inevitably experience motherhood, they frowned on hiring pregnant women. Women were the preferred worker; however, surveillance to determine where female workers were in their life cycle was used to discriminate against certain women. Management knew that women were likely to become pregnant, but their preference was to have the women working at the peak of their level of productivity before they took time off to have children. A third characteristic sought by recruitment ofﬁcers was the all-important ‘‘team attitude.’’ Company memoranda from the transformation period indicate a concern for ﬁnding just the right (female) worker who could work with others. In one particular memo, a high-level Human Resources manager on the implementation team observed that: This team is very motivated; as mentioned there is great integration and participation, with the exception of two people, [names], who are easily singled out by their attitudes. They resist changes proposed and seek ways to be the leaders of the group. However, the other workers don’t pay much attention to them. It is important to note that whenever we broke up into teams, they called others to them, instead of them seeking out their teammates as the [team learning] dynamic intended y . Regarding the feedback sessions and group dynamics to detect leadership qualities or apathy, the following people were found to be lacking in interest or apathetic [two names]. Another worker, [name], is insecure. I think she has problems adapting to teamwork.

Such information, culled from training sessions and initial teams, was used to ﬁlter out potential trouble-makers and construct a proﬁle for future team workers. Certain characteristics were avoided: workers with leadership qualities, aggressive personalities, or lacking in self-motivation. Team attitude, on the other hand, meant being conﬁdent (but not too conﬁdent, that would be seen as aggressive), following directions, and getting along with others. Such characteristics favor the social female over the male competitor.

Gendered Contradictions: Managers and Women Workers

97

This worker proﬁle was conﬁrmed by worker ﬁles and interviews with the Human Resources recruitment ofﬁcer.4 To ﬁnd just the right employee, applicants were required to pass four exams, entailing several visits to the factory. Although applicants took psychometric exams to measure intelligence, aptitude, manual dexterity, and ability to follow directions, the personality evaluation with written comments by the Human Resources interviewer carried the most weight. These comments reﬂected Human Resources’ subjective assessments of the person’s ability to get along with others. Having seen reports from many interviews while examining worker ﬁles, it is clear that the company sought accommodating, pleasant, and quiet workers. Some comments had warnings to the effect that the person interviewed appeared to be too aggressive or outgoing, thus cautioning managers to take care when hiring decisions were made. The worker proﬁle compiled through these different mechanisms – that of a social, submissive woman with networks for childcare – was then used to screen applicants. Situated in a central state in Mexico with a tight labor market, there was a steady stream of applicants who showed up at the factory gates looking for employment. At times when there were vacancies, workers also recommended family members to be interviewed, who then received preferential treatment. I was told that applicants recommended by family members within the factory were preferred because they felt more pressured to perform at work. According to the manager who oversaw the recruitment and examination of applicants, approximately 100 people applied monthly. Of those, only 35 or 40 matched the desired worker proﬁle. Of these, a large number were denied when they did not pass the medical, psychological, eyesight, or dexterity exams. In the end, approximately 10 of the 100 applicants began training every month at the training center.

Training: Filtering out Independence and Building up Family There were two main ways in which the training process was gendered. The ﬁrst entailed communicating the message of the ﬁrm as ‘‘family’’ on posters and through the ‘‘community of fate’’ ideology. The second was through the month-long training session, which also served as a screening process. My introduction to the garment factory was through the doors of the training center. The day began at 7:00 a.m., starting time for the ﬁrst shift. The classroom had two rows of tables arranged lengthwise in the middle. The tables were lined with old industrial sewing machines on all four sides. Metal and wood chairs that looked at least 20 years old were in front of each

98

NANCY PLANKEY VIDELA

sewing machine. Against one sidewall were two steam irons and across the room there was a drawing and a cutting table. Cubbies for odd-sized practice cloths were in the corner by the electric scissors. As in any other classroom, a desk in front of a blackboard was at the front of the room. We could keep our sweaters and purses in the lockers by the door. The only color in the gray room came from the posters extolling the virtues of teamwork and multi-skilled workers. Posters brandished messages such as: Teamwork: multi-skilled workers will grow more within Moctezuma (cartoon of a smiling girl); Recipe for a good service: make good suits; make the exact number requested by the client; deliver on time (cartoon of a girl smiling and baking a cake); Working in teams sounds better (image of a musical band made up of girls); and Working in teams we can ﬁnd the best solution to our problems (image of three female cartoon characters talking and socializing). The posters contained bright colors with buttons, scissors, pins, and tape measures that morphed into cartoon characters. While the presentation seemed harmless, the message was serious. From the moment workers entered the training room, or for that matter the shop-ﬂoor, they were bombarded with the message that the only way the ﬁrm was going to be profitable was for the workers to strictly follow the principles of lean production: discipline, teamwork, and quality. Moreover, this message was gendered. The cartoon characters were invariably childish, happy, and female. Some posters, such as the one written in the form of a cooking recipe, framed work as an extension of the home, and equated women’s central yet subordinated role in the home with their role in the ﬁrm. Company newsletters distributed to all workers and managers emphasized the factory as a family. Here are a few of the comments from the ﬁrst bulletin published in 1996: Aurelia from Team One tells us: the work in teams is good, even though I still have some doubts about how to deal with the bundles, but I feel that work is better in teams because we can be more interconnected. Lourdes from Team Two comments: the change is good because by improving the quality of our work we will have more clients. Besides we need more companies [in Mexico] who are open and willing to change in order to have more clients and create more jobs. As a team we know it’s a bit more work, but we know we can do it. Working in teams does away with selfishness between workmates because we become one since we have the same responsibility. Working like this is more family like, we get along better and we help each other.

Gendered Contradictions: Managers and Women Workers

99

The message was clear: teams were good for workers and the ﬁrm. On the one hand, workers were happier in a more ‘‘family-like’’ environment, where they communicated, suggested improvements, and worked together. Note that the chosen quotes emphasize the individual’s dual satisfaction with getting along with others and increasing production. Company bulletins published in 2001 contained similar messages with the added bonus of special sections on favorite recipes, games, and childrearing practices. In the end, both the posters and newsletters employed the image of family to describe the ﬁrm. The discourse of the ﬁrm as family was reinforced by portraying workers as happy girls, with management as the enlightened authority father ﬁgure providing direction. While most literature on work and organizations assumes the universal worker to be a male (who is ambitious and assertive) (Acker, 1990, 1992), the opposite was true at Moctezuma. At this garment factory ‘‘worker’’ meant female, submissive, and social. That is, women’s ascribed characteristic of being social, of getting along with others, was a key ingredient of teams, gendering teamwork as female. Moreover, the relationship between teams and management in practice was one of paternalistic familialism. The managers were the parents who knew best, workers in teams (who were overwhelmingly female) were treated as adolescent daughters who were given a limited amount of autonomy, and workers under piece-rate (of which the majority were men) played the role of adolescent males who were assumed to be trouble-makers but were left to make their own mistakes and play games. The second way that gendered processes were evident at the training center was in the instructor’s training placement and decisions to fail or pass workers, thus either ending their stay at the factory or signifying that they were employable and ready to move up to the shop-ﬂoor. On the ﬁrst day of training, the instructor asked each worker to verbally (and publicly) answer a personal and work history questionnaire. Of the ten people who started training that day, three were men. All three were married with children. Two of the young men had previously worked at the factory in the stock room, where inputs and materials were gathered to be distributed to the shop-ﬂoor. The older man, probably in his late 30s, was new to the city but had worked as a tailor and machinist in garment factories in a nearby state. The women ranged from 19 to 36 years of age, most of them single. Three of the single women lived at home. They shared their income with their families or planned to save their wages to start small businesses. The four other women were single mothers, two with small children that had to be dropped off at a government-run daycare center at 6:30 every morning. I

100

NANCY PLANKEY VIDELA

noticed that each time these two young women were absent, it was because their child was sick and they had no one to care for him or her. Although they lived with their families, most of their families worked, so childcare support for mothers or sisters was not possible. The other two women had older children who took care of themselves. One of these women had previously been employed at Moctezuma for six years, attaining certiﬁcation in two operations. However, when her mother – who took care of her children – suffered a stroke, she quit her job at Moctezuma to care for her. After a year and a half, her sister returned from the U.S. to take care of the mother and my coworker’s children, so my coworker could return to work. As in the case of the younger single mothers in the training course, childcare – and in this case also eldercare – was considered to be the purview of women, even when these women were also the main providers for the family. After the trainees presented their work histories on the second day of training, the instructor walked us through the operation of steam pressing, machine sewing, print-making, and cloth-cutting. Then we were all given an 8 in. 8 in. piece of cloth and told to learn how to sew. However, it was soon clear that not everyone was meant to learn to sew. One of the young men who had previously worked in the stock room was sent to ﬁll a recent job opening on the second day. The other young man only practiced pressing and cutting operations. Only the tailor joined the women as they practiced operations on the sewing machines. Thus, all of the women were slated as sewers, while two of the three men were set apart for other work. From the very start, even before workers could demonstrate particular aptitude for sewing, pressing, or cutting, we had been divided into men’s work and women’s work. Gendered processes also became evident during training when I tallied up who successfully completed the training module. During my time in the training center, the instructor failed ﬁve workers – three of them because they were ‘‘hard to get along with.’’ Having worked with these three young women for three weeks, I was struck by the fact that all three could be described as independent and free-spirited. In fact, two of them were excellent seamstresses and all three were progressing well in the training module. They did not exhibit difﬁculties relating to coworkers nor did they have trouble following the instructor’s directions. However, they were self-assured, conﬁdent young women. ‘‘Hard to get along with,’’ as the instructor described it, was a code for independent. The other two individuals who were ﬁred included the young man that had only practiced pressing and cutting, and a shy woman who was not progressing well on the machines. Interestingly, there was another woman

Gendered Contradictions: Managers and Women Workers

101

who was performing even worse and with terrible quality. However, she was, as the instructor described her, ‘‘a team player.’’ Contrary to management rhetoric and the oft-touted teamwork philosophy, women’s submissiveness – not skill – was the key marker of a good and employable worker. In this way, management’s gender ideologies along with the requisite for ‘‘team-attitude’’ (i.e., non-aggressiveness) denoted what work was appropriate for women before employees even entered the shop-ﬂoor.

Placement: Women’s Work and Men’s Work Given the differential treatment that men and women received at the training center, it came as no surprise that when workers graduated to the shopﬂoor, men were generally assigned to be bundle boys and cutters (at times they became pressers). Women invariably became either sewers or pressers. These jobs carried distinct responsibilities and privileges on the shop-ﬂoor. Before discussing the different job placements, it is important to understand how work was organized and jobs distributed on the shop-ﬂoor. Cutting was done exclusively in the Cutting Department, where workers laid down the cloth on long tables with the assistance of machines. Electric scissors, connected to an overhead system of wires and tubing, were used to cut the carefully stacked cloth. At the end of the long tables stood the ticket and labeling operation where each piece of cloth – cashmere and acetate lining – was given a ticket that indicated the client order, to which piece of clothing it belonged, size, color, and input details. After the cut cloth was bundled, it was taken to Pants or Jackets. The ﬁrst operation in every assembling department consisted of organizing the cut pieces into a series of bundles that would be distributed to particular teams for them to complete a sub-product. Most teams contained mainly sewing operations with pressing intermixed to facilitate further sewing. At the end of the department were the teams that conducted the ﬁnishing operations (such as buttons, brand labels, ﬁnishing stitches), cleaning of loose threads, and ﬁnal pressing of products. While production imperatives did not require a particular gender for each operation, jobs were indeed gendered. In the Cutting Department, women laid out the cloth but men cut the fabric with electric scissors. A few women were also cutters but they had entered the shop-ﬂoor with previous cutting experience. When asked why men were chosen for cutting, the Human Resources and Recruitment ofﬁcer responded that the cutting tables required a particular height usually found only in men. Upon visual inspection, cutters

102

NANCY PLANKEY VIDELA

came in all sizes but were mostly male. Another manager claimed that cutting was dangerous work and thus given to male workers. In fact, the work of cutters was highly skilled work: a minute deviation from the model could ruin large quantities of cloth. However, managers did not frame the work as skilled but rather as dangerous and necessitating a particular height. Had it been framed as skilled work, it would have been more difﬁcult to keep it as a male enclave since women were said to be ‘‘naturally’’ skilled and best suited for detailed work. While men did the cutting, women operated the ticket and labeling machines, located at the end of the same tables. This was not considered difﬁcult or dangerous, but rather detailed monotonous work. The gendered division of labor in the Cutting Department was thus propelled by management ideologies, not productive or physical imperatives. There was an added layer of gender inequality in the Cutting Department; it was the last full department that operated under piecework in addition to having a higher base wage. As piece-workers, those in Cutting could control their own work-pace (within reason) without having to be pushed by a team. This meant they could either self-exploit (working over 100% to earn bonuses) or work at a leisurely pace. Most cutters worked over 100%, earning higher wages as individuals than the rest of the shop-ﬂoor, which was organized in teams. In this way, those in the Cutting Department, who were mostly male, earned the highest wages at the factory.5 A similar gendered division of labor occurred in Jackets and Pants. Theoretically, men and women were able to work at any operation. However, in practice women were concentrated in sewing operations and men in bundling. While both performed pressing operations, there were fewer men on pressing machines than women. The bundling operations, usually located at the beginning of the production process in each department or in teams where sub-products reconnected to complete the clothing piece, were almost always done by young men. This work was described to me by a manager as ‘‘heavy work, where a man’s strength is needed to move bundles.’’ The ﬁrst job I did on the shop-ﬂoor was that of bundling. Indeed, it was heavy work, but no less than the pressing operations I later performed. For the fourth team in Pants, bundling consisted of checking the ticket on each sub-product to make sure that pant front sides (from team one) and back sides (from team two) matched by size, model, and client order. Then I would go to team three to pick up the completed waistband and belt loops for that particular model and size, place them in a cart together with the front and back panels, and pass the work to one of my teammates. The only ‘‘heavy’’ part of the work I encountered came when one of the three

Gendered Contradictions: Managers and Women Workers

103

previous teams delayed the collating of materials, resulting in stacks that later had to be moved. Bundles of ten pant panels quickly add up to a significant weight, but never one that was unmanageable. Indeed the bundling operation was considered the easiest (and partly the reason they had me doing it!). Moreover, in the ﬁrst teams in Jackets, where I later worked, the bundling operation was normally done by a man but had been recently given to a woman on disability for back problems. Thus, the rationale that bundling was a male job because of the heavy nature of the work does not stand up to empirical evidence. The other operation I observed men doing was that of pressing. This work consisted of standing up on either large pressers that are pulled down over a humped board (to make curved creases) or small overhead steam irons with thin boards to open seams for further sewing. While both men and women operated both types of presses, men were more likely to operate the large presses and women the small steam irons. During the nine months on the shop-ﬂoor I operated both types of presses, ﬁnding speciﬁc challenges with each but no different physical or skill requirement for either. Placement on pressing machines was decided by the instructor and the department head. Recall that the (female) instructor did not allow the male workers to train on the sewing machines, except for one who was already a tailor. Thus, when a presser was needed, men would be more likely to be ready for work. However, it was at the department level that it was decided which type of pressing machine to place a worker. In this way, the gender ideology of managers shaped the gendered division of labor at the pressing machines. Sewing, on the other hand, was clearly marked as woman’s work. In the Pants and Jackets department, a miniscule number of men sat at the sewing machines. I recall no more than 10 men, out of approximately 400 workers in these 2 departments, who sat at sewing machines. However, of these 10, 2 worked on new computerized state-of-the-art machines. In Pants, there were two of these of computerized sewing machines, one operated by a recent male hire, the other by a female worker with eight years seniority. Thus even within the sewing operations, the result of worker placement was the exclusion of women from the more technically advanced machinery. Interestingly, it was not only management’s but workers’ gender ideology that buttressed this gendered division of labor. In my interviews, I asked coworkers why so few men operated sewing machines. Erlinda, who had worked at the factory for ﬁve years responded ‘‘Sewing has always been women’s work.’’ Other women echoed this belief, adding an explanation – ‘‘women are more patient,’’ ‘‘our [women’s] hands are softer and gentler. Men are brutes with their hands. They cannot do the work.’’ The men

104

NANCY PLANKEY VIDELA

interviewed, however, had a different perspective. Juan, one of four men that I worked with (of whom two sewed) sheepishly responded, ‘‘Well, supposedly women have more skill, but it’s not true.’’ Alberto, who taught me the bundling operation said, ‘‘We men are brusquer with out hands, but with practice we can become more skilled.’’ Interestingly, it was the men – denying the feminization of their work – who claimed equal skill between the sexes. Only one woman interviewed seemed less certain of women’s natural superior dexterity. Marı´ a, a highly skilled worker, said ‘‘[Men] are ashamed to be seen at the machines; they say it’s for women. But if there is a job, just be thankful there is a job, it does not matter if a man or woman does it.’’ The quotes demonstrate how women did not question the gendered division of labor on the shop-ﬂoor. Women’s beliefs mirrored the larger patriarchal gender ideology of Mexican society, taking for granted that sewing was women’s work. While not surprising, it is important to note how workers participate in their own gender subordination; it is both management and workers who produce gender on the shop-ﬂoor. However, the structures and relations of power in the factory favor management’s interests over those of workers.

WORKER’S RESPONSE: ACCOMMODATION AND RESISTANCE From the beginning of the employment process – in the formulation of the worker proﬁle and during recruitment – management’s gender ideologies structured later team dynamics. Management continued to select women for the desired characteristics of docility, even though, under the new system of self-managed teams, assertiveness was required. Faced with this contradiction, workers combined assertiveness with their own gender ideologies and those of management in ways that resisted management control. At the beginning, workers overwhelmingly used the language of traditional femininity to carve out spaces of independent action. As self-managed teams matured, these spaces were claimed not only in the name of motherhood but also in terms of worker rights. In this section, I describe the ways in which workers participated in their own gender subordination by taking on the role of team member in ways that equated worker’s interests with that of the ﬁrm. In the following, I will demonstrate how resistance was enacted in favor of traditional femininity ﬁrst, and later in favor of worker rights. The contradictory location of workers resulted in incremental responses of accommodation combined with resistance.

Gendered Contradictions: Managers and Women Workers

105

The gradual transformation of Moctezuma women factory workers from piece-rate workers into full members of self-managed teams was described to me by one female coworker as being in someone else’s skin: it did not feel natural, it did not ﬁt well. For years she had been coming to work, minding her own business, keeping track of how many seams she had sewn together, continually racing between the clock and the piece-rate payment system. Now, under self-managed teams she was expected to understand and watch over the production process, especially as it pertained to her team’s work, seeking ways to cut costs and improve efﬁciency. At ﬁrst it was awkward, especially speaking up in groups and speaking to managers as (near) equals. Encouraged by higher wages, enhanced worker beneﬁts, and a growing camaraderie among workers, many women took on greater roles in team functioning. Some of the time, their participation was a form of accommodation and in other cases, an act of resistance.

Accommodation: Reinforcing Gender Ideologies through Increased Efficiency Self-managed teams gave workers new rights. As autonomous teams, they could hold back direct managerial control over their work. However, team philosophy also held individual workers accountable to their teammates, creating a powerful form of peer pressure termed concertive control by Barker (1993) and characterized as a control regime premised on surveillance and discipline by Sewell and Wilkinson (1992). In the context of pervasive team control, workers were empowered to take on more responsibility for production. And they did. An important way in which workers, and especially women workers, accommodated management’s contradictory directive to be both traditionally female (i.e. docile) and act typically male (i.e. assertive) was through performing emotional labor. The responsibilities of team leaders were many. Among the most important was to secure amicable working relationships between members. Interestingly, women and men team leaders performed this task differentially, with management’s tacit and sometimes explicit approval. In the two teams on which I worked, one team leader, Luisa, was a young single woman with seven years seniority. In the other team, Juan, a young married man with only three years at the factory, occupied the position of team leader. Since Juan’s team was smaller in size, his turn at the rotating position arrived sooner. An affable and seemingly shy young woman stationed at one of the few computerized sewing machines, Luisa carried out her duties in an

106

NANCY PLANKEY VIDELA

understated manner. She smiled often, greeted everyone cordially, and went to great lengths to make the peace between teammates who did not get along, gently coaxing and cajoling people to follow the rules of teamwork. Throughout the days, she carried out her duties with conﬁdence, although she often stopped by the workstation of the older female workers to ask for advice. She deferred to their opinion on most issues. In meetings, training and production-related decisions were made after she described the issue at hand, requested opinions, facilitated a quick discussion, and put the issue up for a vote. Rarely did Luisa put forth her own opinion or dominate discussion. Luisa carried out her leadership duties with feminine sociability. Juan, on the other hand, had a ﬂare for the dramatic. Although it was common knowledge that he was recently married and had a new-born son, Juan took advantage of his location at the pressing machine by the aisle to yell out catcalls as young women passed by. More often than not, he started meetings with his opinion of what should be done, calling for a quick vote. Negotiating interpersonal conﬂicts was not his strength, as he loudly and vulgarly reminded other team members when they asked him to solve a conﬂict. In fact, I witnessed several shouting matches ﬁlled with obscenities between him and two older women in the team. The fact that the older women demanded respect based on their seniority and expertise, and called attention to the ways in which he did not perform his duties as team leader, irritated him to no end. Since I was his assistant at the ironing press for several weeks, he conﬁded that he refused to be held accountable to those ‘old hags’ (viejas). Juan was unwilling to perform the emotional labor that came along with being team leader. Instead, he stuck to gender norms of male superiority even though his position called for equality. Management and other teammates excused Juan’s aggressive behavior, claiming that as a man, he could not help himself. The subtext was that men are brutes; they cannot control their aggression. On the other hand, Luisa was expected to conform to the role of peace-maker, acting as a mother of feuding children, putting their interests above her own. The fact that management allowed Juan’s belligerence and encouraged Luisa’s care-taking evidenced a continuation of traditional gender norms on the part of both management and workers. When I questioned coworkers about whether Luisa or Juan’s actions as leaders were typical, they responded afﬁrmatively. My own observations and interviews with others who had been team leaders conﬁrm this. Luisa’s peace-making role was not the only marker of having internalized the ﬁrm’s interests as one’s own. Both teams worked hard to reach and surpass production with quality quotas, noting the trust placed on them

Gendered Contradictions: Managers and Women Workers

107

through self-managed teams as a crucial motivator. A 21-year-old single woman, Eva, described self-management’s advantages in this way: It is to our advantage because you [the worker] can solve problems. You are responsible for your work. You become one with the factory; you have to do that and I have to do this, and you become more responsible with your own work and towards your female coworkers (compan˜eras).

Eva internalized the ﬁrm’s goal of cultivating committed and assertive workers. It is interesting to note that Eva was in Juan’s teams; although she was engaged with her role as empowered team player, she was one of those who excused Juan’s lack of solidarity as natural for males. Eva, like many other women, was combining the gender norms of the larger Mexican society (the same that marked her as cheap, docile labor) with the new role she was being asked to take on in self-managed teams. Workers at times combined strategies of accommodation and resistance. Again, in Juan’s teams but this time against his wishes, the women organized a special team lunch. Up until the time Juan had become leader, the team had carried out this group-building ritual every Friday. Several equated it with a family meal. However, the family like ritual entailed breaking company rules: the team gave one person a permit ‘‘to go for a doctor’s visit’’ before lunch time. This permit was then used to take a bus to a nearby open market to purchase tacos for everyone and sneak the food back into the factory. Since I did not require permission slips to go in and out of the factory, I was asked to go along. The idea was that in the future I could provide this service by myself. Marı´ a, one of the older women on the team, and I took everyone’s order and dispatched on the mission. When we returned, everyone was waiting for us on the steps of the side-entrance to the shop-ﬂoor. This area was hidden from managers’ views, providing a good space to hang-out. Although Juan did not support this informal teambuilding exercise, he had also placed an order. When his order was mixed up, he bitterly complained that Marı´ a had done so on purpose. Playing the feminine peace-making role, I gave him my tacos and took the blame for the mistake. When company and team allegiances are intertwined, it is difﬁcult to not internalize ﬁrm goals. The examples of differential team leadership performance between Luisa and Juan demonstrate both how traditional gender ideologies still governed shop-ﬂoor relations while concomitantly women were beginning to take on masculine leadership characteristics. Likewise, the taco incident shows how women, faced with the male team leader’s inaction, invoke traditional imagery of the family and take over relationship-building activities. My

108

NANCY PLANKEY VIDELA

coworkers had internalized the concept of team to the degree that they would break company rules.

Resistance: Marshalling Independence in the Name of Traditional Femininity Women workers, like Luisa, took on leadership roles to fulﬁll production directives and further the ﬁrm’s interests, as set forth by the team system. At other times, workers used the rules of self-managed teams to resist work regulations in favor of their own interests. However, at the beginning they did so using the language of motherhood and feminine vulnerability. Two speciﬁc examples – determining an earlier end to the day and giving permits to mothers – demonstrate this. As self-managed teams solidiﬁed, workers used the rules of autonomous teams to further worker rights using the language of both traditional femininity and human rights. When the Pants Department expelled a department manager, this case was made clear. When management reorganized production at Moctezuma, the workday for the ﬁrst shift started at 7 a.m. and ended at 4:36 p.m. Once self-managed teams had gathered momentum in their work-pace, however, team members decided to let themselves out at 4 pm if their production quota was complete. After all, they said, they were autonomous. The argument was made to management not only based on fulﬁlling production imperatives, but also on the need for mothers to get back to their domestic responsibilities. There were children to be picked up from school, homes to clean, laundry to wash, meals to cook, children to bathe. While it was not this gendered logic that convinced managers to acquiesce to the earlier departure time (more likely it was the imminent revolt if they did not), nonetheless women (with men’s support) utilized the discourse of autonomy and motherhood to frame their demand. In a similar vein, self-managed teams favored workers who were mothers when they distributed permits. Teams could give only one permit per day to workers requesting permission to be absent or arrive tardy. Single women were both sympathetic and frustrated by the fact that if they had a medical justiﬁcation and a mother had a sick child or appointment at school, teams repeatedly voted to give the mother the sought-after permit. This meant that the worker who missed work hours without a permit would not be paid for that day’s work. When I was present in team meetings where such issues were discussed, conversation invariably turned to demands of motherhood. For example, in one meeting the team was deciding who to give the permit

Gendered Contradictions: Managers and Women Workers

109

to: Berna, whose child had been sick and in the hospital or Angela, a young woman who had been to the doctor’s for an appointment. Angela had a note from the doctor excusing her from work; Berna had no proof that her child had been to the hospital or had even been sick. One team member who did not get along well with Berna spoke in her favor: ‘‘Well, our kids can’t tell us when they are going to get sick. When they get sick we have to put everything down to care for them.’’ In her defense of Berna, this woman tapped into norms of female solidarity based on their identities as mothers. Angela put the doctor’s note into her work apron pocket and signed with resignation. She looked at me, saying, ‘‘It’s always like this.’’ Managers expressed great displeasure and frustration at the ways in which workers were – in their view – abusing the power of self-managed teams to give excessive numbers of permits. Moreover, depending on the team, their sense of internal solidarity, and the reason for the absence, workers were granted permits without having to repay the hours they missed. If a woman’s absence or tardiness was caused by familial responsibilities, especially having to tend to a sick child, teams were more likely to pardon the missed work. Workers, of whom the majority were women, used the discretionary power given to them by the team system to recompense mothers for their dutiful familial labor. In doing so, women justiﬁed and used the language of motherhood to favor their interests instead of the ﬁrm’s. Both of these cases – changing the departure time and giving out permission slips – occurred early on in the implementation of self-managed teams. The last example of empowered workers is the case of the Pants Department’s refusal to work with a departmental manager who used vulgar language and aggression to control workers. Recall that under self-managed teams, managers could only speak directly to team leaders; teams controlled who came in and out of their space. While technically this rule did not cover the departmental manager, workers extended their discretion to criticize the Pants departmental manager due to his overly aggressive behavior. The workers in the Pants Department, around 150 in total, loudly and jointly protested what they described as inappropriate behavior for this top-level manager. They accused him of speaking to workers (most of them women) in vulgar and aggressive terms. The complaint to company leaders and the union6 was framed as an affront to women workers’ feminine sensibilities as well as their rights as workers. In the end, he was ﬁred. It is interesting to note that this was one of the few times that workers went to the union with a labor relations complaint. Women workers invoked the discourse of traditional femininity, together with workerist discourse, to protect themselves from an overly aggressive manager. Women

110

NANCY PLANKEY VIDELA

told and retold this story often on the shop-ﬂoor, a morality tale of sorts, an example of empowered women. A shy and quiet female single coworker told me how she liked self-managed teams. When I asked her why, she perked up and said: ‘‘You feel independent because no one is harassing (acosando) you to do your work. One knows how much production to give at the end of the day. There is no need for them [managers] to be all over the worker.’’ In interviews workers claimed to be initially excited about self-managed teams until they saw how management rolled back the rights given to teams. Laura, who had been at the company for twelve years, described her experiences this way: [At the beginning] self-managed teams meant that they [workers] could make decisions within the team, decisions regarding work, how to organize themselves y in order to get the production out. They could make decisions about everything, for work, to support themselves, to give each other permits. They made decisions about permits and then got the production out. That’s what it was supposed to be y But later, a permit had to be given by the department head. And that is not what the rules say. The team was supposed to give the permits. The boss was only supposed to sign it. But lately this is not respected anymore.

The more women workers demanded to make use of the rights accorded to autonomous teams, the more managers resisted losing control over the labor process. A struggle ensued between women workers and managers over the right to be self-managed teams and the role of workers on the shop-ﬂoor. By the sixth year of implementation, managers took away many of the characteristics that made teams autonomous. In the process, they lost the commitment of workers. The contradictory location of women – being hired for their docility but asked to be assertive and independent – backﬁred on management.

DISCUSSION Similar to the managerial tropes encountered by Ong (1987) in Malaysia, Ferna´ndez-Kelly (1983) and Salzinger (2003) on the U.S.–Mexico border, and Wolf (1992) in Java, managers at Moctezuma sought women workers for their docility and low wages. However, the implementation of self-managed teams at this garment factory, part of a radical reorganization of production under Japanese-inspired lean techniques, made Moctezuma different. Unlike Taylorist assembly lines with heavy supervisory control, selfmanaged teams required independent, multi-skilled workers who could supervise each other and production. In order for this to happen, workers

Gendered Contradictions: Managers and Women Workers

111

were given considerable control over their labor and the team’s labor. Therefore, women workers in self-managed teams face different constraints and possibilities than women workers in traditionally organized ﬁrms. The literatures on sex-segregation and gendered organizations provide analytical tools for understanding how self-managed teams affect workers. Milkman’s (1987) work on the electronic and auto-parts industry during World War II demonstrates how organizational inertia and management gender ideologies maintained the sex-typing of jobs, yet made them ﬂexible enough to respond to productive and historically contingent changes. Building on the concept of sex-typing of jobs, Britton (2000) shows how jobs can be categorized as masculine or feminine even if they are ﬁlled with the opposite sex (gender-typing). In this way, the men who performed women’s work of sewing and pressing at Moctezuma were marked as having female characteristics; as less-than-men they were invisible to management. Some men retreated into acting as adolescents; others, like Juan, acted out a hyper-sexualized role, challenging his gender-typing yet reinforcing larger gender ideologies. Meanwhile, women in self-managed teams were asked to take on masculine roles. Following Acker’s (1990, 1992) call to examine how organizations and organizational practices are gendered, however, it is clear that while women were asked to take on masculine roles, the organizational practices, images, and ideologies pressed women in the opposite direction. The posters and the messages all over the factory walls pictured women workers as happygo-lucky girls. Working in the factory was an extension of home; women were seen as supplemental income earners not as providers for their families. Jobs at Moctezuma were gender-typed female. This contradictory location is made evident in the literature on self-managed teams and gender. While Ollilainen (1999) and Metcalf and Linstead’s (2003) work shows that self-managed teams are indeed gendered in theory and practice, they do not ask how this process happens. Salzinger’s (2003) study of Particimex begins to tackle this issue by noting the effect of self-managed teams on young women. She suggests that there is a transformation of docile girls into assertive women at work, which does not extend to gender relations at home. It has been my goal to examine how this process occurred. In this examination of gender and self-managed teams at Moctezuma, I argue that it was the interaction of management’s and workers’ gender ideologies with the re-structured work place that called for masculine roles – what I call women’s contradictory location in self-managed teams – that resulted in a struggle to deﬁne women’s new roles. It was in this contradictory location that women workers combined gender norms of traditional

112

NANCY PLANKEY VIDELA

femininity with empowered workerist discourse to re-make self-managed teams and their role within them, even if only temporarily. When management lost control over women’s labor power, the experiment with selfmanagement ended; a diluted team concept followed. Under the new rules, both team leaders and team meetings disappeared. Management took over decision-making powers in teams. The result of women’s contradictory location in self-managed teams, then, was that women unevenly incorporated empowered attitudes within the dominant trope of traditional womanhood. However, this was a forging of a new identity that grew with time. Early decisions the teams made to invoke the language of traditional femininity contrasted with their later expulsion of the department manager using the combined discourse of traditional gender norms, women’s rights, and worker rights. Future research in the area of gender and self-managed teams might examine the stability of this change in women’s perception of self. How do women’s new experiences with leadership roles, independent action, and assertive behavior at work affect their gender identities away from the workplace, and under what conditions does this become an enduring transformation? Moreover, in this case women at different life-stages and circumstances have different interests. Therefore, an examination of the ways in which women – whether single young childless women, older single women, single women with children, or married women with children – experience team work and construct their identities and interests in similar and different ways, could be useful to better understand the possibilities for social movement organizing.

NOTES 1. The name of the company and individuals are pseudonyms to assure conﬁdentiality. 2. For more detail see Nancy Plankey Videla (2006), ‘‘It Cuts Both Ways: Workers, Management and the Construction of a ‘‘Community of Fate’’ on the Shop-Floor in a Mexican Garment Factory.’’ Social Forces, Forthcoming, June 2006. 3. While both Salzinger and I are ethnographers, her study was a comparison of four shop-ﬂoors and the ways management differentially constructed women as female subjects. Due to this research design, she worked at Particimex for six weeks (Salzinger 2003, p. 6). On the other hand, my ethnography focused on one ﬁrm and the ways that teamwork affected workers. I worked on the shop-ﬂoor at Moctezuma for nine months, with two different teams, allowing me to observe and question my co-workers’ response to teamwork.

Gendered Contradictions: Managers and Women Workers

113

4. Toward the end of my study, ﬁrm management granted me permission to examine all thousand plus worker ﬁles. 5. There were plans for the Cutting Department to be reorganized under teamwork but it never happened due to lack of funds stemming from the investments to reorganize the ﬁrm and the 2000 recession in the U.S. The reasons, mechanisms, and outcomes of industrial upgrading at Moctezuma are detailed in Plankey Videla (2005) ‘‘Following Suit: An Examination of Structural Constraints to Industrial Upgrading in Mexico.’’ Competition and Change, 9(4), 307–327, November 2005. 6. The union at Moctezuma was a ‘white’ or company-union, representing management’s interests. For more information on the role of the union in Moctezuma see note 2 above. For more information on white unions in Mexico see LaBotz (1992), Mask of Democracy: Labor Suppression in Mexico Today. Boston: South End Press.

ACKNOWLEDGEMENTS I am very thankful to Robert Mackin, Gay Seidman, Vicki Smith and an anonymous reviewer for their helpful comments and support. The research was funded in part by a John T. and Catherine D. MacArthur Global Studies Fellowship at the University of Wisconsin, Madison and an International Research Travel Grant from the Texas A&M University.

REFERENCES Acker, J. (1990). Hierarchies, jobs, bodies: A theory of gendered organization. Gender & Society, 4(2), 139–158. Acker, J. (1992). Gendered institutions: From gender roles to gendered institutions. Contemporary Sociology, 21(5), 565–569. Acker, J. (1998). The future of ‘gender and organizations’: Connections and boundaries. Gender, Work, and Organizations, 5(4), 195–206. Appelbaum, E., & Batt, R. (1994). The new American workplace: Transforming work systems in the United States. Ithaca, NY: ILR, Cornell University Press. Appelbaum, E., Thomas, B., Berg, P., & Kalleberg, A. L. (2000). Manufacturing advantage: Why high performance work systems pay off. Ithaca, NY: Cornell University Press. Barajas Escamilla, M. del R., & Sotomayor Yala´n, M. (1995). Rotacio´n de personal en la industria maquiladora de Tijuana: mujeres y condiciones de vida. In: S. Gonza´lez, O. Ruiz, L. Velasco & O. Woo (Eds), Mujeres, migracio´n y maquiladora en la frontera norte (pp. 189–214). Mexico City: Colegio de Me´xico. Barker, J. R. (1993). Tightening the iron cage: Concertive control in self-managing teams. Administrative Science Quarterly, 31, 561–586. Beneria, L., & Stimpson, C. (Eds) (1987). Women, households, and the new economy. New Brunswick, NJ: Rutgers University Press. Besser, T. L. (1996). Team Toyota: Transplanting the Toyota culture to the Camry plant in Kentucky. Berkeley, CA: University of California Press.

114

NANCY PLANKEY VIDELA

Britton, D. M. (2000). The epistemology of the gendered organization. Gender & Society, 14(3), 418–434. Chow, E. N. (2002). Transforming gender and development in East Asia. New York: Routledge. Cole, R. E. (1979). Work, mobility, and participation: A comparative study of American and Japanese industry. Berkeley, CA: University of California Press. Colectivo Atabal, A. C. (1994). El ABC de las trabajadoras dome´sticas: Cartilla sobre los derechos y las obligaciones laborales de las trabajadoras y trabajadores del servicio dome´stico. Mexico City: Colectivo Atabal, A.C. Collins, J. L. (2002). Mapping a global labor market: Gender and skill in the globalizing garment industry. Gender and Society, 16, 921–940. Collins, J. L. (2003). Threads: Gender, labor and power in the global apparel industry. Chicago, IL: The University of Chicago Press. Elson, D. (1983). Nimble ﬁngers and other fables. In: W. Chapkins & C. Enlow (Eds), Of common cloth: Women in the global textile industry (pp. 5–14). Amsterdam, The Netherlands: Transnational. Elson, D. & Pearson, R. (1986). Third world manufacturing. In: Feminist Review (Ed.), Waged Work: A Reader (pp. 67–92). London: Virago. England, P., Herbert, M. S., Kilbourne, B. S., Reid, L. L., & Magdal, L. (1994). The gendered valuation of occupations and skills: Earnings in 1980 census occupations. Social Forces, 73(1), 65–99. Ferna´ndez-Kelly, M. P. (1983). For we are sold, I and my people: Women and industry in Mexico’s frontier. Albany, NY: State University of New York. Florida, R., & Kenney, M. (1991). Transplanted organizations: The transfer of Japanese industrial organizations to the U.S.. American Sociological Review, 56(June), 381–398. Graham, L. (1995). On the line at Subaru-Isuzu: The Japanese model and the American worker. Ithaca, NY: ILR, Cornell University Press. Hall, E. J. (1993). Waitering and waitressing: Engendering the work of table servers. Gender & Society, 7(3), 329–346. Hartmann, H. (1983). Capitalism, patriarchy, and job segregation by sex. In: E. Abel & E. Abel (Eds), The signs reader: Women, gender and scholarship (pp. 193–225). Chicago: Chicago University Press. Hearn, J., & Parkin, W. P. (1992). Gender and organizations: A selective review and a critique of a neglected area. In: A. J. Mills & P. Tancred (Eds), Gendering organizational analysis (pp. 44–66). Newbury Park, CA: Sage. Hobday, M. (1995). Innovation in East Asia: The challenge to Japan. Cheltenham, UK: Edward Elgar. Hochschild, A. R. (1983). The Managed Heart. Berkeley, CA: University of California Press. Jensen, J. (1989). The talents of women, the skills of men: Flexible specialization and women. In: S. Wood (Ed.), The transformation of work (pp. 141–155). London: Unwin Hyman. Katzenbach, J. R., & Smith, D. K. (1993). The wisdom of teams: Creating the high-performance organization. Boston, MA: Harvard Business School Press. Kopinak, K. (1995). Gender as a vehicle for the subordination of women maquiladora workers in Mexico. Latin American Perspectives, 22(1), 30–49. Kopinak, K. (1996). Desert capitalism: Maquiladoras in North America’s Western industrial corridor. Tucson, AZ: The University of Arizona Press. LaBotz, D. (1992). Mask of democracy: Labor suppression in Mexico today. Boston, MA: South End Press.

Gendered Contradictions: Managers and Women Workers

115

Lamphere, L., Zavella, P., & Gonzalez, F. with Evans, P. (1993). Sunbelt working mothers: Reconciling family and factory. Ithaca, NY: Cornell University Press. Lee, C. K. (1998). Gender and the South China miracle: Two worlds of factory women. Berkeley, CA: University of California Press. MacDufﬁe, J. P. (1995). Workers’ role in lean production: The implications for worker representations. In: S. Babson (Ed.), Lean work: Empowerment and exploitation in the global auto industry (pp. 54–69). Detroit, MI: Wayne State University Press. Metcalfe, B., & Linstead, A. (2003). Gendering teamwork: Re-writing the feminine. Gender, work and organizations, 10(1), 94–119. Milkman, R. (1982). Redeﬁning ‘‘women’s work’’: The sexual division of labor in the auto industry during World War II. Feminist Studies, 8(2), 336–372. Milkman, R. (1987). Gender at work: The dynamics of job segregation by sex during World War II. Urbana, IL: University of Illinois. Morgan, G., & Knights, D. (1991). Gendering jobs: corporate strategy, managerial control and the dynamics of job segregation. Work, Employment and Society, 5, 181–200. Nash, J., & Ferna´ndez-Kelly, M. P. (1983). Women, men and the international division of labor. Albany, NY: State University of New York. Ollilainen, A. M. (1999). Gendered processes in self-managing teams: a multiple case study. Unpublished dissertation, Department of Sociology, Virginia Polytechnic Institute and State University. Ong, A. (1987). Spirits of resistance and capitalist discipline: Factory women in Malaysia. Albany, NY: State University of New York Press. Pierce, J. L. (1995). Gender trials: Emotional lives in contemporary law firms. Berkeley, CA: University of California Press. Plankey Videla, N. (2005). Following suit: An examination of structural constraints to industrial upgrading in Mexico. Competition and Change, 9(4), 307–327. Plankey Videla, N. (2006). It cuts both ways: Workers, management and the construction of a ‘‘community of fate’’ on the shop-ﬂoor in a Mexican garment factory. Social Forces, June, Forthcoming. Salzinger, L. (2003). Genders in production: Making workers in Mexico’s global factory. Los Angeles: University of California Press. Sewell, G., & Wilkinson, B. (1992). Someone to watch over me: Surveillance, discipline and the just-in-time labour process. Sociology, 26, 271–289. Sklair, L. (1993). Assembling for development: The Maquila industry in Mexico and the United States. San Diego, CA: Center for U.S.–Mexican Studies, University of California. Skuratowicz, E., & Hunter, L. W. (2004). Where do women’s jobs come from? Job resegregation in an American bank. Work and Occupations, 31(1), 73–110. Smith, V. (1993). Flexibility in work and employment: The impact on women. Research in the Sociology of Organizations, 11, 195–216. Smith, V. (1997). Employee involvement, involved employees: Participative work arrangements in a white-collar service occupation. Social Problems, 43(2), 166–179. Stoddard, E. R. (1987). Maquila assembly plants in Northern Mexico. EI Paso, TX: Texas Western Press. Tiano, S. (1994). Patriarchy on the line: Labor, gender and ideology in the Mexican maquila industry. Philadelphia, PA: Temple University Press. Ward, K. (Ed.) (1990). Introduction and overview. Women workers and global restructuring (pp. 1–22). Ithaca, NY: ILR, Cornell University Press.

116

NANCY PLANKEY VIDELA

Williams, C. L. (1995). Still a man’s world: Men who do ‘women’s work. Berkeley, CA: University of California Press. Wolf, D. L. (1992). Factory daughters: Gender, household dynamics, and rural industrialization in Java. Berkeley, CA: University of California Press. Womack, J. P., Jones, D. T., & Roos, D. (1990). The machine that changed the world: The story of lean production. New York: Rawson Associates. Yeatts, D. E., & Hyten, C. (1998). High performance self-managed work teams. Thousand Oaks, CA: Sage.

YOU GET PAID FOR THAT?: JOB AND ESTABLISHMENT LEVEL VARIATIONS IN THE USE OF INCENTIVE COMPENSATION$ Jeremy Reynolds ABSTRACT Most research on worker participation treats it as an establishment-level phenomenon even though it is seldom used on an establishment-wide basis. This paper, however, examines how three forms of incentive compensation are used at the job level, and it assesses the potential ramifications for inequality. I find that the use of incentive compensation reflects the gender composition, unionization, and functional role of jobs. Jobs with many full-time women, for instance, are less likely to use group incentives and profit sharing because they are less likely to play central or managerial roles in establishments. This suggests that incentive compensation may increase inequality. Recent efforts to make U.S. organizations more competitive have not affected all workers equally. Downsizing, for instance, helped some workers

$

The data used in this study were gathered with support from the National Science Foundation under awards SBR-9507964, SBR-9511715, and SBR-9507914. The data are now available through the Inter-University Consortium for Political and Social Research (ICPSR).

Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 117–149 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16005-7

117

118

JEREMY REYNOLDS

keep their jobs by allowing organizations to avoid bankruptcy, but it also endangered the psychological, physiological, familial, and economic wellbeing of workers who were laid off (Konovsky & Brockner, 1993). Likewise, the proliferation of temporary work and other nonstandard employment relationships helped organizations remain competitive despite ﬂuctuations in the business cycle. However, since nonstandard jobs typically provide substandard rewards (Kalleberg, Reskin, & Hudson, 2000), their use has also contributed to inequality, especially gender inequality (Smith, 1997). Even workplace participation, which was often intended to improve rather than degrade or eliminate jobs (Hackman & Wageman, 1995; Sirianni, 1995), has the potential to increase inequality (Handel & Levine, 2004; Rothschild & Ollilainen, 1999). Given the continued spread of worker participation (Osterman, 2000) and its ability to affect worker well-being (see Barker, 1993; Freeman & Kleiner, 2000), the need for more information about its connection to inequality is particularly great. In this paper, I provide new information about the connection between workplace participation and inequality by examining one form of worker participation (i.e., incentive compensation) and how its use varies among jobs. Researchers have known for many years that workplace participation is rarely introduced on an establishment-wide basis (Lawler, Mohrman, & Ledford, 1992; Osterman, 1994). However, since most existing studies examine worker participation at the establishment level, patterns of intraestablishment use and their ramiﬁcations have remained largely unexamined (for an exception see Reynolds, 2006). This study helps ﬁll that void by examining job level variations in the use of three incentive compensation systems (i.e., group incentives, pay-for-learning, and proﬁt sharing) and how those variations are likely to affect inequality. Below, I describe the three incentive compensation systems in more detail. Then I discuss how economic, discriminatory, and power-based motivations may inﬂuence which jobs rely on such systems. Finally, I use multi-level data from the second National Organizations Study (NOS-II) to identify job and establishment characteristics that are associated with the use of each compensation system, and I offer some reﬂections on how the observed patterns of use may affect inequality. I conclude by suggesting some avenues for future research.

BACKGROUND Worker participation programs come in many different forms (Lawler et al., 1992), and there is some disagreement regarding how to deﬁne them. Some

You Get Paid for that?

119

discussions of worker participation pay special attention to teams, job rotation, and other practices that alter the structure of work. Authors who take this perspective often argue that incentive compensation and training programs only support the reorganization of work and are thus fundamentally different from practices that actually change the structure of work (Gittleman, Horrigan, & Joyce, 1998; Osterman, 1994; Pil & MacDufﬁe, 1996). Other authors deﬁne worker participation more broadly and do not make such sharp distinctions between programs that change the structure of work and those that do not (Appelbaum, Bailey, Berg, & Kalleberg, 2000; Delaney & Huselid, 1996; Ichniowski, Shaw, & Prennushi, 1997). In this paper, I use the broad deﬁnition of worker participation developed by Appelbaum et al. (2000) that recognizes incentive compensation as a form of worker participation. I do this for two reasons. First, using a broad deﬁnition emphasizes that all worker participation programs serve one primary goal: they are designed to elicit discretionary effort. Different programs pursue that goal in different ways. Job rotation, for instance, focuses on providing workers with more opportunities to participate. Cross-training helps workers develop additional skills. Incentive compensation systems provide extra motivation. In each case, however, the ultimate goal is to elicit discretionary effort. Second, using a broad deﬁnition encourages a comprehensive examination of how worker participation affects workers. Organizations often use a combination of practices such as teams, training, and group incentives to encourage greater worker participation (Gittleman et al., 1998). By focusing only on practices that change the structure of work, researchers could obscure how efforts to promote participation can inﬂuence inequality. Incentive compensation systems, for instance, do not change the structure of work, but they have a direct and intentional effect on earnings. I focus here on three kinds of incentive compensation: group incentives, pay-for-learning, and proﬁt sharing (for information about other types of incentives see Dixon, Hayes, & Stack, 2003). Group incentive plans (e.g., gain sharing) tie rewards to the achievements of a group, but depending on the plan, the rewards may or may not be distributed evenly among the members of the group (Honeywell-Johnson & Dickinson, 1999). In many ways, these plans bear a close resemblance to more traditional reward systems, but they focus on the group rather than on the individual worker. Pay-for-learning (also called skill-based pay) is an incentive compensation system in which people are paid for what they are capable of doing. Organizations using this system of incentives identify the skills they value and then reward employees with higher compensation when they have learned those skills (Shaw,

120

JEREMY REYNOLDS

Gupta, Mitra, & Ledford, 2005). The actual use of the skills, which is emphasized by more traditional pay schemes, is less important. Pay-for-learning was introduced by Edward Lawler and ﬁrst implemented by General Mills (Cappelli & Neumark, 2001). Finally, profit sharing is one of the older incentive compensation systems, and although it can take different forms, one of the better-known variations is the Scanlon plan (Welbourne & Mejia Gomez, 1995). Under proﬁt sharing plans, workers are typically rewarded based on the proﬁts or earnings generated by the whole organization (Honeywell-Johnson & Dickinson, 1999). The reward, therefore, is often only weakly tied to the actions of any individual because the economic performance of the organization depends on factors beyond the control of any one person (Honeywell-Johnson & Dickinson, 1999). My goal is not to evaluate how incentive compensation systems affect worker or organizational performance, but rather to gauge the potential impact of such systems on workplace inequality. To affect inequality at all, incentive compensation must be able to inﬂuence worker outcomes, and although it is not entirely clear how incentive compensation affects workers, there are good reasons to believe that it does. Multi-pronged worker participation initiatives that combine several practices such as teams, crosstraining, and incentive compensation are associated with desirable outcomes. In particular, they are associated with greater training (Whitﬁeld, 2000), job satisfaction (Cotton, 1993; Freeman & Kleiner, 2000; but see Wagner, 1994), intrinsic rewards and trust (Appelbaum et al., 2000), pride (Hodson, 1996), and work–family balance (Berg, Kalleberg, & Appelbaum, 2003). Incentive compensation in particular has been associated with a wage premium (Barkume, 2004). Worker participation plans, however, can also lead to negative outcomes. Initiatives that involve teams, for instance, have been criticized on the grounds that they lead to work intensiﬁcation, heightened stress, and other undesirable outcomes (Barker, 1993; Ezzamel & Willmott, 1998; Sewell, 1998). Apparently, incentive compensation can lead to undesirable outcomes as well, including the heightening of peer pressure and peer monitoring (Knez & Simester, 2001). These ﬁndings show that incentive compensation can alter the economic and non-economic rewards jobs offer. In order to understand how incentive compensation inﬂuences workplace inequality, it is important to examine how the use of incentive compensation varies within and between establishments. If incentive compensation systems are distributed evenly among jobs, then their potential for creating inequality is limited because their desirable or undesirable effects will also be distributed equally. However, if incentive compensation systems are distributed unevenly, they are likely to increase or reduce inequality. Other

You Get Paid for that?

121

authors have made similar observations with respect to worker participation more generally. Worker participation, for example, may reduce inequality by providing unskilled workers with additional training and job opportunities (Appelbaum et al., 2000; Rothschild, 2000; Smith, 1996) as well as greater autonomy and job security (Batt, 2004). The uneven diffusion of worker participation could also increase inequality by improving the job rewards of workers who already have good jobs (Handel & Levine, 2004; Rothschild & Ollilainen, 1999). There is some evidence that this is happening with teams: jobs that require higher levels of human capital are more likely to use teams with decision-making powers (Reynolds, 2006). In short, how incentive compensation and other types of worker participation affect inequality will be determined to a great extent by who participates (see also Handel & Levine, 2004; Rothschild, 2000). Consequently, I take a different perspective on worker participation than many authors. Most existing studies examine worker participation as an establishment level phenomenon (see for example Appelbaum et al., 2000; Appelbaum & Batt, 1994; Gittleman et al., 1998; Osterman, 1994, 2000). These studies have yielded much important information about worker participation, and some even examine topics that are relevant for the study of inequality (see Heywood & Jirjahn, 2002). Establishment level studies, however, also tend to obscure the fact that worker participation is seldom introduced on an establishment-wide basis (Lawler et al., 1992; Osterman, 1994). For some lines of inquiry, this oversight may not matter so much, but if research on inequality were to focus only on the establishment level, it would overlook a major source of inequality. Jobs are tremendously important in generating inequality. Gender differences in wages, for instance, are closely tied to job level phenomenon (Huffman, Velasco, & Bielby, 1996), and research suggests that job boundaries also play a crucial role in the diffusion of worker participation plans (Finegold & Wagner, 1998; Manley, 2000; Vallas, 2003). Therefore, although I account for establishment level variations in the use of incentive compensation, I pay special attention to job level variations. Below, I discuss several reasons to expect that incentive compensation plans will be implemented unevenly among jobs. Then I examine the interand intra-establishment patterns of adoption and identify job and establishment characteristics that are associated with the use of group incentives, pay-for-learning, and proﬁt sharing. In doing so, I provide new information about the use of incentive compensation and move one step closer to the larger goal of determining how worker participation affects the distribution of rewards at work.

122

JEREMY REYNOLDS

THEORY AND HYPOTHESES Some discussions of workplace participation emphasize the humanistic goals of those who developed various forms of participation. Pioneers in the use of teams, for instance, argued that the reorganization of workers into teams should beneﬁt workers by contributing to their personal development (Sirianni, 1995). In particular, they wanted to develop an alternative to Taylorist models of production that alienate workers by ignoring their potential intellectual contributions to the production process (Lawler et al., 1992). The research of these pioneers reﬂects their concern for the common worker: they developed their ideas about teams by studying coal miners (Sirianni, 1995). Some authors still evaluate the success of worker participation programs according to how well they incorporate lower-level employees (see Vallas, 2003). Leading supporters of total quality management also argued that organizations should care about worker well-being (Hackman & Wageman, 1995). Nevertheless, there are good reasons to expect that establishments will use incentive compensation in ways that do not reﬂect a concern for personal development, the common worker, or worker well-being. First, incentive compensation is primarily designed to increase performance, and it is based on the idea that inequality is useful because it motivates workers. It would be surprising if establishments used incentive compensation to pursue humanistic goals. Second, it is naı¨ ve to expect that workplaces will be free from the racial, gender, power, and economic dynamics that create inequality in society at large. The Devaluation Hypothesis: Racial and Gender Discrimination The literature on work and stratiﬁcation suggests that incentive compensation may be implemented unevenly in ways that reﬂect racial and gender discrimination. It is well-known that racial and gender discrimination affect other worker outcomes. For instance, as the percentage of women or minorities in a job or occupation increases, wages decrease even after controlling for human capital and working conditions (Catanzarite, 2003; Huffman et al., 1996; Tomaskovic-Devey, 1993). Researchers are still trying to identify the mechanisms that account for these relationships, but the devaluation hypothesis provides one well-accepted explanation (Catanzarite, 2003; Huffman & Cohen, 2004; Kmec, 2003). According to this hypothesis, wage differences are partially caused by cultural norms that devalue women and minorities, the roles they hold, and the types of work they do. Gender devaluation, for example, can occur through ‘‘the cognitive error of not

You Get Paid for that?

123

seeing how much female occupations contribute to the ‘bottom line’ of proﬁt or other organizational goals’’ (Kilbourne, Farkas, Beron, Weir, & England, 1994, p. 695). Based on the logic of the devaluation hypothesis, I expect that the use of incentive compensation will be less common as the percentage of women or minorities in a job increases. Establishments tend to overlook the potential contributions of female and minority dominated jobs and will be less likely to devise special plans for rewarding the discretionary effort that workers in such jobs can provide. The Power Process Perspective and Unionization The use of incentive compensation may also be inﬂuenced by the distribution of power within an establishment. The power process perspective suggests that groups of workers negotiate their status within an establishment by exercising their power during periods of organizational change (Thomas, 1994, p. 178). Following this logic, we can expect that unions will wield their power to help determine whether and how an establishment introduces incentive compensation. Nevertheless, since workers use their power to both resist and co-opt worker participation initiatives (Batt, 2004; Vallas, 2003) and since union attitudes toward worker participation are ambivalent (Thomas, 1994), it is not clear what effect unions will have. Some research in the 1980s suggested that union members were supportive of worker participation (Fenwick & Olson, 1986). More recent research, however, shows that unions can either facilitate the adoption of worker participation (Appelbaum & Batt, 1994; Eaton & Voos, 1992) or make its adoption less likely (Lawler et al., 1992). Some studies even indicate that unions do not inﬂuence the adoption of worker participation at all (Collins, Hatcher, & Ross, 1993; Gittleman et al., 1998; Osterman, 1994). In light of these mixed results, I simply hypothesize that unionized and non-unionized jobs will have different probabilities of using incentive compensation. Evolutionary Economics: Costs and Benefits Finally, many contemporary discussions of workplace participation emphasize the economic motivations for its use (Cappelli & Rogovsky, 1994). Evolutionary economics, for instance, suggests that concerns about the potential costs and beneﬁts of worker participation will be the primary factor behind its adoption. Indeed, establishments do not always adhere to humanistic ideals when implementing participation plans (Rothschild & Ollilainen, 1999), and concerns for the bottom line are often revealed by a

124

JEREMY REYNOLDS

tendency to adopt innovations that are less costly owing to their compatibility with existing human resource practices (Pil & MacDufﬁe, 1996). The adoption of worker participation at the establishment level, for instance, is positively associated with the skill levels of front line employees (Hunter, 2000; Osterman, 1994). Furthermore, job level variations in the use of teams reﬂect the skills that jobs require (Reynolds, 2006). In short, organizations use worker participation practices in ways that alleviate the need for extra training. I suggest that concerns about costs and beneﬁts will inﬂuence the use of incentive compensation systems at the job level in several ways. To the extent that the use of incentive compensation is driven by concerns about costs, establishments will use them in jobs that require high levels of human capital. As noted above, worker participation is thought to be most effective when workers are motivated to work for the good of the organization, are allowed the opportunity to do so, and have adequate skills (Appelbaum et al., 2000). Admittedly, establishments do not always introduce worker participation plans in bundles so as to maximize opportunities, skills, and motivation simultaneously (Handel & Levine, 2004), but introducing incentive compensation in jobs that already require high skill levels is an easy and inexpensive way to ensure that workers have the skills and motivation they need to have a notable effect on performance. Therefore, I expect that establishments will be more likely to provide incentive compensation in jobs that require high levels of human capital. If establishments want to make sure that incentive compensation systems generate signiﬁcant performance beneﬁts, they will implement them in large and/or functionally central jobs. Automobile manufacturers, for instance, will use group incentives, proﬁt sharing, or pay-for-learning to motivate assembly line workers, who are both central to the production process and quite numerous. The incentive compensation system may not be extended to all assembly line workers, but automobile plants that use incentive compensation should be more likely to do so among assembly line workers than among janitors or mail clerks. In general, to the extent that incentive compensation systems are used to improve performance, I expect that they will be more common in jobs that employ many people or play a central role in generating an establishment’s main product or service.

DATA This analysis uses multilevel data from the 1996 National Organizations Study (NOS-II) (Kalleberg, Knoke, & Marsden, 1995) to model the use of

You Get Paid for that?

125

incentive compensation in jobs that are clustered within U.S. establishments. The NOS-II was created with the help of Dun and Bradstreet Information Services, which provided a probability sample of U.S. establishments that was stratiﬁed by establishment size. Sampling with probability proportional to size helped ensure that medium-sized and large establishments, which are far less numerous than small establishments, were not underrepresented (see Kalleberg, Reynolds, & Marsden, 2003). Using computer assisted telephone interviews and/or written questionnaires, the NOS-II completed data collection between June 1996 and June 1997 with 1002 of the 1835 sampled establishments thus achieving a response rate of 54.6%. The ﬁnal sample contains a wide range of proﬁt and non-proﬁt establishments from all industries that vary in size from those with no full-time employees to those with over 40,000 full-time employees. By using a large probability sample of U.S. establishments, these data avoid many common forms of bias (for information about biases, see Appelbaum & Batt, 1994; Godard & Delaney, 2000; Hunter, 2000). Furthermore, the NOS-II sample closely reﬂects the population of U.S. establishments in the percentage of establishments from different size categories and industries (see Table A.1). Respondents in each of the 1002 establishments in the NOS-II were asked to provide the titles for one ‘‘core,’’ one ‘‘administrative support,’’ and one ‘‘managerial’’ job at their establishment. The core job is deﬁned as the job that is directly involved in the main product or service of the establishment and involves the largest number of employees. The administrative support job is the secretarial, clerical, or ofﬁce-staff job that involves the greatest number of workers. The managerial job is the job that involves the largest number of managers, supervisors, or other administrators. Together, the core, administrative support, and managerial jobs in the NOS-II represent 254 different occupational categories as measured by the three-digit 1990 Census Occupational Classiﬁcation System. The core jobs are very diverse. They represent 219 different occupational categories. Administrative support and managerial jobs are less varied, but the administrative support jobs still represent 55 different three-digit occupational categories, and the managerial jobs represent 63 occupational categories. It is important to note that the content of a job is not enough to identify it as core, administrative support, or managerial. Rather, jobs were classiﬁed depending on their role in the division of labor.1 At many large stores, for example, ‘‘cashier’’ could be an administrative support job and ‘‘sales worker’’ could be the core job. At convenience stores, on the other hand, ‘‘cashier’’ could be a core job. In a similar fashion, the owner of a small business might say that he or she ﬁlls the core job, even if it is largely administrative and would be considered a

126

JEREMY REYNOLDS

managerial job at another establishment. Ultimately, 538 establishments provided information about a core, administrative support, and managerial job, 230 establishments provided information about two jobs, and 218 establishments provided information about one job. Therefore, the NOS-II contains information about (538 3)+(230 2)+(218 1) ¼ 2,292 jobs, each of which represents a core job (N ¼ 981), an administrative support job (N ¼ 676), or a managerial job (N ¼ 635). Owing to its multilevel structure, the NOS-II provides information that a simple representative sample of workers, jobs, or establishments cannot. In particular, it makes it possible to separate job- and establishment-level determinants of worker participation by examining jobs that are clustered within establishments. After eliminating cases with missing data on the variables examined here, there are 1,402 jobs in 715 establishments.2 Because the NOS-II does not contain detailed information about the human capital (i.e., experience, education, and training) required for the various core, administrative support, and managerial jobs, I have supplemented the NOS-II with occupational-level data from the Occupational Information Network (ONET) 3.1 database (see Employment and Training Administration, 1996). ONET serves as the replacement for the Dictionary of Occupational Titles, which had been criticized for several weaknesses including out-dated occupational classiﬁcations and gender bias (see Peterson et al., 2001; Spenner, 1990). ONET 3.1 was created in 1996 by analysts who ranked 900 different occupations on a wide variety of characteristics, including the combined amount of experience, education, and training they require (see Appendix D, Employment and Training Administration, 1998). By aggregating these data upward from the 900 ONET occupations to the 496 occupations in the 1990 Census Occupational Classiﬁcation System, it was possible to merge the NOS-II and ONET 3.1 data using a custom crosswalk provided by the National Crosswalk Service Center (National Crosswalk Service Center). Dependent Variable NOS-II respondents were asked to provide information about how the core, administrative support, and managerial jobs in their establishments are paid. In particular, they were asked to answer ‘‘yes’’ or ‘‘no’’ to three questions: Are any _______ paid using group incentives, such as gain sharing? Do any _______ receive pay for learning new skills? Do any _______ participate in a proﬁt sharing or bonus program?

You Get Paid for that?

127

Using these questions, I have constructed three dichotomous dependent variables that indicate whether jobs offer group incentives, pay-for-learning, or proﬁt sharing. Job Level Explanatory Variables In order to uncover signs of discrimination and the importance of worker power, I measure the racial and gender composition of jobs at each establishment and whether they are unionized. NOS-II respondents at each establishment reported the percent of full-time workers in the core, support, and managerial jobs that were white and the percent that were female. They also indicated whether or not each job was unionized. Since evolutionary economics suggests that the use of incentive compensation will be related to the human capital jobs require, I describe each NOSII job in terms of the overall preparation it requires. In particular, I used information from ONET 3.1 to calculate the average combined level of experience, education, and training required by each job in the NOS-II.3 Using occupational level data as a proxy for job level characteristics leads to some aggregation bias, but it also has the distinct advantage of clarifying the direction of causation. If the human capital measure came from the NOS-II, a relationship between it and a dependent variable could mean that incentive compensation is often accompanied by training programs. Using occupational level data from ONET provides much ﬁrmer evidence that pre-existing human capital demands of the job led to the use of incentive compensation. To allow for non-linear relationships, I measure preparation using indicator variables that place each job in one of ﬁve categories: those that require little or no preparation, some preparation, medium preparation, considerable preparation, and extensive preparation. (Coding information and descriptive statistics for all independent variables are available in Table A.2) Finally, I use data from the NOS-II to distinguish jobs by their relative size (number of full-timers in the job as a percent of full-timers at establishment), and their role in the division of labor (core, administrative support, and managerial). Because evolutionary economics suggests that establishments consider costs and beneﬁts when adopting worker participation plans, I expect that the size of a job and its role in the division of labor will be related to its use of incentive compensation. Establishment Level Controls Since adoption is a multilevel process and this analysis builds on research that examined the adoption of worker participation at the establishment

128

JEREMY REYNOLDS

level, I control for establishment characteristics that have been highlighted in previous studies. Large-scale studies of U.S. establishments have produced mixed results regarding the inﬂuence of establishment size (Gittleman et al., 1998; Lawler et al., 1992; Osterman, 1994) and age (Ichniowski & Shaw, 1995; Kelley, 1996) on the adoption of worker participation. Therefore, I do not make any predictions regarding their effects, but I control for both. Age is measured in years. Size is measured as the log of the number of full-time employees at an establishment. I also identify establishments that are afﬁliated with a larger, parent organization since such connections may have an effect similar to size. Since there is strong support for the notion that inter-organizational networks can serve as ‘‘mechanisms for the diffusion of innovative practices’’ (Haunschild, 1993, p. 564), I use a dummy variable to identify establishments that consult with other establishments when creating training programs. Indeed, the use of worker participation is associated with consultants (Sitkin, Sutcliffe, & Schroeder, 1994), partnerships with local universities (Hunter, 2000), and even simple contact with organizations already using worker participation (Pil & MacDufﬁe, 1996). To account for the possibility that the use of incentives will vary by industry, I use a dummy variable to identify establishments in manufacturing industries. Previous studies have found that while pay-for-learning is often more successful in manufacturing establishments than in service establishments (Shaw et al., 2005), gain sharing is more successful in service establishments (Kim, 1999). I also use a dummy variable to identify non-proﬁt establishments. I expect that they will be less likely than other establishments to use proﬁt sharing, because they will want employees to focus on goals other than proﬁt. Finally, since foreign competition encourages the adoption of worker participation (Appelbaum & Batt, 1994; Lawler et al., 1992; Osterman, 1994), I include a variable to measure how much foreign competition establishments face in their main market.

DESCRIPTIVES Table 1 shows the percentage of jobs using each type of incentive compensation system. Proﬁt sharing, which is used in 41% of the jobs, is the most commonly used compensation system. This ﬁnding is consistent with studies of incentive compensation at the organizational level, which also indicate that proﬁt sharing is the most widely used incentive plan (Honeywell-Johnson & Dickinson, 1999). Table 1, however, also provides information about how the use of incentive compensation varies from one type of job to

You Get Paid for that?

Table 1.

129

Percent of Jobs Using Three Forms of Incentive Compensation.

(N ¼ 1,402)

Group Incentives

Pay-for-Learning

Proﬁt Sharing

All jobs

15

26

41

Core jobs Administrative support jobs Managerial jobs

15 10 21

30 22 24

39 36 54

another. Proﬁt sharing and group incentives are more common in managerial jobs than in core or administrative support jobs. In fact, over half of managerial jobs use proﬁt sharing. Pay-for-learning, on the other hand, is most commonly used in core jobs: 30% of core jobs use pay-for-learning. Administrative support jobs also use incentive compensation, but they do so less frequently than core and managerial jobs. Table 2 provides another perspective on the use of incentive compensation by examining the percent of jobs in major occupational groups using gain sharing, pay-for-learning, and proﬁt sharing. Clearly, the nature of the job has an inﬂuence on the type of compensation system used. Jobs in professional specialty occupations, for example, are seldom rewarded on the basis of group performance, but they use pay-for-learning more frequently than jobs in nearly every other occupation. This ﬁnding is consistent with the notion that professionals often work independently and must maintain jurisdiction over an abstract and evolving body of knowledge (Abbott, 1988). Jobs in the sales industry, in contrast, are less frequently paid for learning. The notion of the ‘‘sales team,’’ however, seems to carry over into the reward system in the form of group incentives, and since success in sales is often equated with volume and proﬁt, it is no surprise that proﬁt sharing is common among sales jobs. In Table 3, I provide a brief overview of the bivariate relationships between job level characteristics and the use of group incentives, pay-forlearning, and proﬁt sharing. Overall, the results are consistent with the expectation that job characteristics would be related to the use of incentive compensation. Except for job size, all of the relationships are also in the expected direction. More importantly, except for percent white, all of the variables are related to the use of at least one type of incentive compensation. This indicates that the use of incentive compensation is not random or universal and thus likely to have an effect on inequality in the workplace. The next section examines these relationships in more detail while accounting for the characteristics of the establishments and the environments in which they are situated.

130

Table 2.

JEREMY REYNOLDS

Percent of Jobs Using Incentive Compensation by Occupation.

Occupation

Group Incentives Pay-for-Learning Proﬁt Sharing

Executive, administrative, managerial Professional specialty Technician Sales Administrative support Protective service Service occ except protect and private Farming, forestry, ﬁshing Precision production, craft, repair Machine operator, assembler, inspector Transportation, material moving Handler, equip, cleaner, helper, laborer

22 9 13 23 10 12 14 20 19 21 14 14

21 40 23 16 23 24 33 30 44 28 9 22

51 28 53 52 37 24 27 30 52 43 55 33

All jobs

15

26

41

MULTIVARIATE ANALYSIS Analytic Strategy The multilevel data in the NOS-II facilitate the examination of job-level effects because they make it possible to examine outcomes for multiple jobs in the same establishment, but multilevel data also require special analytical strategies. When observations are clustered together within larger units (in this case jobs within establishments) observations from the same cluster tend to be more like each other than observations from different clusters. Under these conditions, a normal logistic regression, which assumes that cases are independent, tends to produce biased parameter estimates and underestimate the standard errors (Guo & Zhao, 2000). In order to account for the clustering of jobs within establishments, I use a multilevel logistic regression model, which adjusts for clustering when estimating the standard errors and coefﬁcients (Guo & Zhao, 2000). The models also contain a disturbance term to account for all unobserved characteristics of the level-2 clusters (see Guo & Zhao, 2000). In this analysis, for instance, the disturbance term in each model controls for unobserved establishment characteristics such as differences in organizational cultures that could lead to inaccurate estimates of the job-level effects. Multi-level models, however, are not without difﬁculties. When using a dichotomous dependent variable, such models tend to underestimate the

You Get Paid for that?

Table 3.

131

Correlations between the Use of Incentive Compensation and Job Level Explanatory Variables.

Job Level Explanatory Variables

Group Incentives

Pay-for-Learning

Proﬁt Sharing

Devaluation hypothesis Percent female Percent white

–0.09 0.01

0.03 0.00

–0.10 0.00

Power process perspective Unionized

–0.08

0.04

–0.18

Evolutionary Economics Preparation required by job Job size Core job Administrative support job Managerial job

0.04 –0.08 0.01 –0.09 0.09

0.04 0.04 0.08 –0.06 –0.02

0.08 –0.13 –0.05 –0.08 0.14

ﬁxed and random effects, particularly when the cluster size is small or the observations in a cluster are highly correlated (Guo & Zhao, 2000). In the sample used here, approximately 20% of the clusters have only one observation, and the clusters are small (maximum cluster size ¼ 3; average cluster size ¼ 2.0). The correlation between observations in the same cluster on the dependent variables is also substantial (from 0.42 to 0.70). In order to minimize these difﬁculties, I estimated the models presented here using Stata’s xtlogit procedure, which calculates the parameters using Gauss–Hermite quadrature, a procedure that is particularly good at handling small clusters (for a discussion see Rabe-Hesketh, Skrondal, & Pickles, 2002). Nevertheless, the impact of the independent variables may still be underestimated, so the results presented here are conservative, and we can have conﬁdence in variables that are statistically signiﬁcant. Table 4 presents odds ratios from three sets of nested random effects logistic regressions, one set for each dependent variable. All the models control for establishment characteristics that are believed to affect the use of worker participation at the establishment level. As noted above, the disturbance term in each model controls for unobserved establishment characteristics. The ﬁrst model in each panel also includes measures that help identify the racial and gender differences in the use of incentives that are predicted by the devaluation thesis. The second model adds a measure of unionization to examine support for the power process perspective. The third and fourth models test predictions derived from evolutionary economics about efforts to control costs and maximize beneﬁts. Speciﬁcally, the

Odds Ratios from Random Effects Logistic Regressions of Incentive Compensation Use at the Job Level on Job and Establishment Characteristics. Group Incentives

Job level explanatory variables Devaluation hypothesis Percent female Percent white Power process perspective Unionized Evolutionary economics Some preparation required Medium preparation required Considerable preparation required

Job size Core job Managerial job

1

2

4

0.99 (0.00) 1.00 (0.01)

0.99 (0.00) 1.00 (0.01)

0.99 1.00 (0.00) (0.00) 1.00 1.00 (0.01) (0.01)

0.35 (0.22)

0.39 (0.24)

0.32 (0.21)

0.64 (0.25) 1.42 (0.53) 1.49 (0.58) 3.07 (2.69)

0.84 (0.36) 1.13 (0.48) 0.93 (0.44) 2.37 (2.11) 0.98 (0.01) 4.86 (2.35) 3.31 (1.78)

1

2

3

1.00 (0.00) 1.00 (0.00)

1.00 (0.00) 1.00 (0.00)

1.00 (0.00) 1.00 (0.00)

Proﬁt Sharing 4

1

2

1.01 0.99 0.99 (0.00) (0.00) (0.00) 1.00 100 1.00 (0.00) (0.00) (0.00)

3

4

0.99 (0.00) 1.00 (0.00)

0.99 (0.00) 0.99 (0.01)

2.79 2.82 2.13 (1.23) (1.25) (0.99)

0.06 0.06 0.06 (0.03) (0.04) (0.04)

0.61 (0.19) 0.77 (0.24) 1.19 (0.39) 1.76 (1.29)

1.61 (0.56) 4.75 (1.76) 3.34 (1.24) 6.64 (5.80)

1.02 (0.35) 1.00 (0.36) 1.45 (0.57) 1.49 (1.14) 1.00 (0.01) 3.21 (1.19) 1.54 (0.68)

1.89 (0.74) 3.39 (1.35) 1.89 (0.83) 5.51 (5.05) 0.98 (0.01) 3.58 (1.51) 3.19 (1.52)

JEREMY REYNOLDS

Extensive preparation required

3

Pay-for-Learning

132

Table 4.

Establishment age Log (number of FT emp.) Part of larger organization Consultation w/r training Manufacturing industry Non-proﬁt sector Foreign competition

w2 N Latent Intra-class Correlation

0.99 (0.01) 1.00 (0.00) 1.19 (0.13) 1.58 (0.65) 2.72 (1.11) 2.13 (1.07) 0.11 (0.06) 0.89 (0.17)

0.99 (0.01) 1.00 (0.00) 1.21 (0.13) 1.65 (0.67) 2.71 (1.10) 2.26 (1.13) 0.13 (0.08) 0.87 (0.17)

1.00 (0.01) 1.00 (0.00) 1.20 (0.13) 1.55 (0.63) 2.61 (1.08) 2.60 (1.34) 0.11 (0.07) 0.89 (0.17)

1.00 (0.01) 1.00 (0.00) 1.08 (0.13) 1.63 (0.68) 2.71 (1.16) 2.97 (1.54) 0.11 (0.07) 0.86 (0.17)

0.99 (0.01) 1.00 (0.00) 0.92 (0.08) 0.58 (0.20) 4.26 (1.45) 1.55 (0.65) 1.10 (0.44) 0.85 (0.15)

0.99 (0.01) 1.00 (0.00) 0.90 (0.08) 0.55 (0.20) 4.41 (1.54) 1.51 (0.64) 0.93 (0.38) 0.85 (0.15)

0.99 (0.01) 1.00 (0.00) 0.90 (0.08) 0.53 (0.19) 4.35 (1.54) 1.60 (0.68) 0.88 (0.37) 0.86 (0.15)

0.99 (0.01) 1.00 (0.00) 0.95 (0.10) 0.50 (0.20) 4.64 (1.75) 1.84 (0.85) 0.89 (0.39) 0.88 (0.16)

0.99 (0.01) 1.00 (0.00) 1.36 (0.14) 1.49 (0.61) 2.52 (1.01) 0.86 (0.40) 0.01 (0.01) 1.21 (0.22)

0.99 (0.01) 1.00 (0.00) 1.42 (0.15) 1.64 (0.65) 2.45 (0.99) 0.99 (0.46) 0.01 (0.01) 1.23 (0.24)

1.00 (0.01) 1.00 (0.00) 1.45 (0.16) 1.58 (0.65) 2.35 (0.97) 1.24 (0.61) 0.01 (0.01) 1.28 (0.28)

1.00 (0.01) 1.00 (0.00) 1.29 (0.15) 1.82 (0.78) 2.33 (1.00) 1.34 (0.65) 0.01 (0.01) 1.28 (0.27)

You Get Paid for that?

Establishment level controls Establishment age

34.1 36.1 40.3 47.39 21.08 24.31 28.84 41.5 75.88 85.6 93.71 89.81 1402 1402 1402 1402 1402 1402 1402 1402 1402 1402 1402 1402 0.72 0.71 0.72 0.73 0.70 0.70 0.71 0.73 0.78 0.78 0.80 0.81

po0.05. po0.01.

133

134

JEREMY REYNOLDS

third model adds measures of how much preparation jobs require to see if establishments use incentives in jobs where high levels of human capital may reduce the need for training. The fourth model adds measures of job size and function to see if incentives are more common in jobs that are likely to affect establishment performance because of their size or their role in the establishment. Multivariate Results In Table 4, the results reveal substantial establishment-level variance in the use of incentives. When multi-level models include a random intercept but no random slopes, the ICC represents the proportion of the unexplained variance that can be accounted for by unobserved characteristics of the level-two units. Therefore, the consistently large and signiﬁcant latent ICCs in Table 4 suggest that much of the residual variance in the use of incentives is due to establishment-level differences.4 The ICCs can also be interpreted as indicating that when establishments use incentives in one job they tend to do so in other jobs as well. As might be expected, the ICC is largest for proﬁt sharing, which is more closely linked to establishment level events than the other two incentive systems. Furthermore, judging from other analyses using these data, it appears that jobs within an establishment are more similar in the use of incentives than they are in the use of teams (see Reynolds, 2006). The ICCs reported in Table 4, however, should be interpreted with caution because they are based on the assumption that the dichotomous dependent variables reﬂect underlying, continuous variables. When that assumption is not appropriate, the ICC for a logistic regression should be estimated in other ways that account for the observed values of the covariates in the equation (Goldstein, Browne, & Rasbash, 2002). These other ways of estimating the ICC produce what are called manifest intra-class correlations because they are based on manifest or observed variables rather than latent variables (Rodrı´ guez & Elo, 2003). I calculated the manifest ICC for each of the full models in Table 4 using several values of the linear predictor and the same value of su used for calculating the latent ICC.5 As expected, the manifest ICCs were smaller than the latent ICCs, but they were still substantial. Calculated as Pearson correlations, the manifest ICCs for the values at the 25th, 50th, and 75th percentiles of the linear predictor were as follows: (group incentives .42 .47 .50) (pay-for-learning .49 .51 .52) (proﬁt sharing .59 .61 .61). Like the latent ICCs, these manifest ICCs indicate that much of the residual variance in the use of incentives is at the

You Get Paid for that?

135

establishment level. They also indicate that there is a moderate correlation between the use of incentives in jobs within the same establishment. If the use of incentives is truly a discrete function rather than a reﬂection of a latent continuous variable, the manifest ICCs should be a more accurate reﬂection of the importance of establishment characteristics than are the larger latent ICCs (Goldstein et al., 2002). Ultimately, however, both approaches to estimating the ICC indicate that establishment and job characteristics play an important role in explaining the use of incentives. The rest of the numbers in Table 4 show how speciﬁc job and establishment level characteristics affect the use of incentives, and I begin by brieﬂy summarizing the results for establishment characteristics that have been the focus of much previous research. Establishment age is not associated with the use of incentives of any kind, but jobs in larger establishments are more likely to use proﬁt sharing. The strongest ﬁnding is that jobs are more likely to involve incentives when they are located in establishments that consult with other establishments when developing training programs. For instance, jobs are 2.72 times as likely to involve group incentives when their establishment consults with others about training.6 Jobs in the non-proﬁt sector, on the other hand, are signiﬁcantly less likely than those in the for-proﬁt sector to use group incentives and proﬁt sharing. Other establishment characteristics are rarely or never signiﬁcant. In summary, incentive use is related to some characteristics that establishments can cultivate (e.g., consultation) and some characteristics they cannot change easily (e.g., sector). Perhaps the most interesting results in Table 4 are those in the ﬁrst row, which indicate that the use of group incentives and proﬁt sharing is negatively associated with the percent of full-time women in a job. More speciﬁcally, a one percent increase in the percent of full-time female workers in a job is associated with a one percent decrease in the odds of having those types of incentive compensation. A ten-point increase in the percent of fulltime female workers would result in a nine percent decrease in the odds of using group incentives and proﬁt sharing.7 These results echo the well-documented negative relationship between the feminization of jobs and access to other job rewards. Therefore, although they do not show that establishments actively discriminate against women, they do show a gender disparity in the use of incentives. The changes in the statistical signiﬁcance of percent female help explain why gender composition is associated with the use of group incentives and proﬁt sharing. The coefﬁcients for percent female are signiﬁcant (po0.01) until the last model, where the p-value for female increases to 0.51 in the model for group incentives and 0.21 in the model for proﬁt sharing. To

136

JEREMY REYNOLDS

clarify why they lost signiﬁcance, I conducted a supplementary analysis and found that the dummy variables for core and managerial jobs rather than the variable measuring job size account for the change. This indicates that jobs with many full-time women are not less likely to use incentives because they require different levels of human capital or because they are more or less likely to be unionized – percent female remained signiﬁcant after the introduction of those variables. Rather, feminized jobs are less likely to use group incentives and proﬁt sharing because they are less likely to be core or managerial jobs. If feminized jobs were just as likely as other jobs to be core and managerial jobs, they would be just as likely to use group incentives and proﬁt sharing. These results are not entirely consistent with the devaluation thesis because they suggest that the use of incentives is tied to the role of a job rather than to its gender composition. Nevertheless, the results highlight another well-known mechanism in the creation of gender inequality: occupational segregation. The results for pay-for-learning tell a similar story. The positive association between percent female and pay-for-learning was hidden until the models included a control for core status. Bivariate analyses help explain the origins of this suppressor effect: percent female is positively correlated with the use of pay-for-learning, but it is also negatively correlated with being a core job, a role that is associated with the use of pay-for-learning. Consequently, the positive association between percent female and pay-forlearning was only revealed after controlling for core status. Substantively, this suggests that if feminized jobs were as likely as other jobs to be core jobs, the likelihood of being paid for learning would actually increase with the percent of women in a job. The message, therefore, remains unchanged. Feminized jobs are less likely to offer incentives because of the roles they tend to play in establishments. The regressions also provide some support for the power process perspective. Unionization is not associated with the use of group incentives, but it is positively associated with the use of pay-for-learning and negatively associated with the use of proﬁt sharing. This does not prove that unions are steering the use of incentives to suit their tastes. It is possible, for instance, that establishments introduce pay-for-learning as a way to ﬁght unionization. Nevertheless, the results are consistent with the idea that unions exercise their power during times of organizational change in order to secure their own interests. Finally, the results provide mixed support for evolutionary economics and the argument that establishments use incentives in ways that minimize costs and maximize beneﬁts.

You Get Paid for that?

137

First, jobs that require high levels of education, training, and experience are not more likely than other jobs to use group incentives or pay-forlearning, and proﬁt sharing is only weakly related to the preparation jobs require. These results are not consistent with the argument that jobs requiring high levels of human capital will be especially likely to use incentive compensation because they do not require additional training. Nevertheless, if incentive compensation is good for workers, these results do suggest that incentive compensation is being used in ways that echo the devotion of worker participation pioneers to the well-being of workers with modest levels of human capital. Second, as job size increases, jobs become less likely to offer group incentives or proﬁt sharing. This could indicate that the size of a job is not a good measure of how important it is to establishment performance, but it could also indicate that establishments try to control costs and beneﬁts by using incentives on a limited basis. Third, group incentives and proﬁt sharing are much more common in core and managerial jobs than in administrative support jobs, and pay-forlearning is more common in core jobs than in administrative support jobs. These results are consistent with the notion that establishments use incentive compensation systems strategically as a way to elicit discretionary effort from workers who hold jobs that are likely to have a large inﬂuence on establishment performance.

CONCLUSION This paper offers new insights into the connection between worker participation and inequality by examining how and why some jobs are more likely than others to use incentive compensation. The results show considerable variation in the use of three incentive compensation systems (group incentives, pay-for-learning, and proﬁt sharing) at the job and establishment level. Furthermore, the results suggest that organizations use compensation plans in ways that reﬂect gender, power, and economic pressures. The observed patterns of incentive use at the job level are gendered, but they are not entirely consistent with the devaluation thesis, which suggests that jobs will be devalued and thus under-rewarded if they are dominated by female or minority workers. The racial composition of a job, for instance, is not related to the use of group incentives, pay-for-learning, or proﬁt sharing. Furthermore, the analysis seems to indicate that jobs with many full-time female workers are less likely to use group incentives and proﬁt sharing

138

JEREMY REYNOLDS

because feminized jobs are less likely to be managerial or core jobs. In other words, it may be that the role of a job rather than its gender composition guides the use of incentive compensation, and that gender inequality is reinforced by segregation rather than by devaluation. Nevertheless, the potential effect is the same: if incentive compensation is good for workers as some authors have found (Barkume, 2004), then incentive compensation plans are likely to exacerbate existing gender inequalities because they are less common in female dominated jobs. The results also indicate that power and economic concerns inﬂuence the use of incentive compensation plans at the job level. The power process perspective argues that unions use their power during times of establishment change to shape worker outcomes. My results indicate that unionized jobs may use their power to avoid the implementation of proﬁt sharing. The results are also partially consistent with the predictions of evolutionary economics. Incentive compensation is concentrated in core and managerial jobs where it is especially likely to have an impact on establishment performance. Furthermore, the likelihood of using group incentives and proﬁt sharing declines as job size (and thus the cost of using incentives) increases. Contrary to my expectations, however, the use of incentive compensation does not vary with the human capital required by a job. If establishments are acting rationally, this suggests that pre-existing levels of human capital may not do much to defray the costs or maximize the beneﬁts of introducing incentive compensation. From a theoretical perspective, these results are quite interesting. The use of incentive compensation at the job level reﬂects gender and power differences as well as economic factors. This means that the distribution of worker participation schemes is not random and that theories that have helped explain the distribution of other job characteristics can be useful in explaining the use of worker participation as well. Moreover, the use of incentive compensation depends on a number of factors and is not easily explained by one theoretical perspective. Just as researchers have found it helpful to examine the gender gap in wages from multiple perspectives, it is useful to examine worker participation from multiple perspectives as well. This analysis also has implications for the study of inequality. By examining the use of incentive compensation at the job level, I have shown that jobs in the same establishment vary considerably in their odds of offering group incentives, pay-for-learning, and proﬁt sharing. More research will be needed to determine how different incentive compensation plans affect workers. Nevertheless, since evidence suggests that incentive compensation can be associated with higher wage rates (Barkume, 2004) as well as heightened peer pressure and peer monitoring (Knez & Simester, 2001), the

You Get Paid for that?

139

observed non-random patterns of incentive compensation use have the potential to inﬂuence the distribution of economic and non-economic job rewards and thus inequality. In particular, this analysis indicates that incentive compensation will inﬂuence the gender gap in wages, the divide between managers and workers, the distinction between employees who play a central role in establishments and those who do not, and disparities between union and non-union workers. In the future, researchers should make an effort to understand the ﬁner patterns of intra-establishment adoption and their ramiﬁcations. In order to do this, we will need more data and research on worker participation among workers, jobs, occupations, departments, and other units of analysis. In particular, researchers should continue to examine how worker participation is related to gender, unionization, and the function workers and jobs play in their establishments. We also need more information about the forces that generate the patterns of adoption observed here and how those patterns of adoption affect inequality. We already know that other elements of the high performance workplace are related to inequality (see for example Smith, 1997). Now, we need to investigate how the use of worker participation inﬂuences workplace inequality. To do this, we will need to utilize data that contain information about the intra-establishment use of worker participation and its economic and social consequences for workers. Worker participation could eventually provide women and minorities with access to job rewards and training that they have not received in the past (Rothschild, 2000). At the moment, however, it seems that establishments are creating some undesirable outcomes. If incentive compensation is good for workers, as some authors suspect, then it is being used in ways that re-enforce existing gender inequalities. If incentives are bad for workers, then the use of incentives is eroding the quality of jobs held by men, nonunionized workers, managers, and workers who play central roles in the functioning of their establishments.

NOTES 1. If the core job provided administrative support or was managerial in nature, it was counted as a core job and no information was collected about an administrative support or managerial job. 2. Prompted by the comments of an anonymous reviewer, I used MICE, Patrick Royston’s Stata add-on (see Royston, 2004) to examine how the results of my analysis might differ if missing values were imputed so that all 2,292 jobs could be

140

JEREMY REYNOLDS

included in the analysis. The results of this supplementary analysis suggested that using imputed values would not lead to any changes in my substantive conclusions. More information about the imputation procedure and the results are available upon request. 3. More speciﬁcally, I calculated the human capital measures using what ONET calls ‘‘job zones.’’ For more information about job zones, see the appendix of this paper and Appendices F and G of the ONET 1998 data dictionary (Employment and Training Administration, 1998). 4. The latent ICCs reported at the bottom of Table 4 are part of the standard output provided by Stata’s xtlogit procedure. For logistic regressions, they are calculated using the same formula used for linear models except that the level-one variance is assigned a value of p2/3, the variance of the standard logistic distribution (see Rodrı´ guez & Elo, 2003). The formula for the latent ICC is given below, where su squared is the level-two variance. Latent ICC ¼ r ¼

s2u s2u þ p2 =3

5. The calculations were performed using xtrho, a Stata add-on developed by Rodrı´ guez and Elo (2003). 6. I interpret the exponentiated coefﬁcients as odds ratios (see Guo & Zhao, 2000; Long, 1997). In this case, for example, 100[exp b1] ¼ 100[2.721] ¼ 172% more likely (or 2.72 times as likely). An exponentiated coefﬁcient less than one indicates a negative relationship. 7. The effect of any size change (d) in an independent variable is computed by exponentiating the product of the change and the coefﬁcient (Long, 1997: 80). In this case, where the exponentiated coefﬁcient is .99, the un-exponentiated coefﬁcient is 0.01, and the effect of a 10 unit change is 100[exp(0.01*10)] ¼ 9.0%.

REFERENCES Abbott, A. (1988). The system of professions: An essay on the division of expert labor. Chicago: University of Chicago Press. Appelbaum, E., Bailey, T., Berg, P., & Kalleberg, A. L. (2000). Manufacturing advantage: Why high-performance work systems pay off. Ithaca, NY: Cornell University Press. Appelbaum, E., & Batt, R. (1994). The new American workplace: Transforming work systems in the United States. Ithaca, NY: ILR Press. Barker, J. R. (1993). Tightening the iron cage: Concertive control in self-managing teams. Administrative Science Quarterly, 38, 408–437. Barkume, A. J. (2004). Using incentive pay and providing pay supplements in U.S. Job markets. Industrial Relations, 43(3), 618–633. Batt, R. (2004). Who beneﬁts from teams? Comparing workers, supervisors, and managers. Industrial Relations, 43(1), 183–212. Berg, P., Kalleberg, A. L., & Appelbaum, E. (2003). Balancing work and family: The role of high-commitment environments. Industrial Relations, 42(2), 168–188.

You Get Paid for that?

141

Cappelli, P., & Neumark, D. (2001). Do ‘high-performance’ work practices improve establishment-level outcomes? Industrial & Labor Relations Review, 54(4), 737–775. Cappelli, P., & Rogovsky, N. (1994). New work systems and skill requirements. International Labour Review, 133(2), 205–220. Catanzarite, L. (2003). Race-gender composition and occupational pay degradation. Social Problems, 50(1), 14–37. Collins, D., Hatcher, L., & Ross, T. L. (1993). The decision to implement gainsharing – the role of work climate, expected outcomes, and union status. Personnel Psychology, 46(1), 77–104. Cotton, J. L. (1993). Employee involvement: Methods for improving performance and work attitudes. Newbury Park, CA: Sage. Delaney, J. T., & Huselid, M. A. (1996). The impact of human resource management practices on perceptions of organizational performance. Academy of Management Journal, 39(4), 949–969. Dixon, M. R., Hayes, L. J., & Stack, J. (2003). Changing conceptions of employee compensation. Journal of Organizational Behavior Management, 23(2–3), 95–116. Eaton, A. E., & Voos, P. B. (1992). Union and contemporary innovations in work organization, compensation, and employee participation. In: L. Mishel & P. B. Voos (Eds), Unions and economic competitiveness (pp. 175–215). Armonk, NY: M.E. Sharpe. Employment and Training Administration. (1996). ONET 3.1. Retrieved March, 2002, from http://www.onetcenter.org/database.html Employment and Training Administration. (1998). ONET 98 data dictionary release 1.0. Retrieved March, 2002, from http://www.bookstore.gpo.gov/market/o_net/datadict/ datadict.pdf Ezzamel, M., & Willmott, H. (1998). Accounting for teamwork: A critical study of group-based systems of organizational control. Administrative Science Quarterly, 43(2), 358–396. Fenwick, R., & Olson, J. (1986). Support for worker participation: Attitudes among union and non-union workers. American Sociological Review, 51(4), 505–522. Finegold, D., & Wagner, K. (1998). The search for ﬂexibility: Skills and workplace innovation in the German pump industry. British Journal of Industrial Relations, 36(3), 469–487. Freeman, R. B., & Kleiner, M. M. (2000). Who beneﬁts most from employee involvement: Firms or workers? American Economic Review, 90(2), 219–223. Gittleman, M., Horrigan, M., & Joyce, M. (1998). ‘Flexible’ workplace practices: Evidence from a nationally representative survey. Industrial and Labor Relations Review, 52(1), 99–115. Godard, J., & Delaney, J. T. (2000). Reﬂections on the ‘high performance’ paradigm’s implications for industrial relations as a ﬁeld. Industrial and Labor Relations Review, 53(3), 482–502. Goldstein, H., Browne, W., & Rasbash, J. (2002). Partitioning variance in multilevel models. Understanding Statistics, 1(4), 223–231. Guo, G., & Zhao, H. (2000). Multilevel modeling for binary data. Annual Review of Sociology, 26, 441–462. Hackman, J. R., & Wageman, R. (1995). Total quality management – empirical, conceptual, and practical issues. Administrative Science Quarterly, 40(2), 309–342. Handel, M. J., & Levine, D. I. (Eds). (2004). Editors’ introduction: The effects of new work practices on workers. Industrial Relations, 43(1), 1–43. Haunschild, P. R. (1993). Interorganizational imitation: The impact of interlocks on corporate acquisition activity. Administrative Science Quarterly, 38, 564–592.

142

JEREMY REYNOLDS

Heywood, J. S., & Jirjahn, U. (2002). Payment schemes and gender in Germany. Industrial and Labor Relations Review, 56(1), 44–64. Hodson, R. (1996). Dignity in the workplace under participative management: Alienation and freedom revisited. American Sociological Review, 61, 719–738. Honeywell-Johnson, J. A., & Dickinson, A. M. (1999). Small group incentives: A review of the literature. Journal of Organizational Behavior Management, 19(2), 89–120. Huffman, M. L., & Cohen, P. N. (2004). Racial wage inequality: Job segregation and devaluation across US labor markets’. American Journal of Sociology, 109(4), 902–936. Huffman, M. L., Velasco, S. C., & Bielby, W. T. (1996). Where sex composition matters most: Comparing the effects of job versus occupational sex composition on earnings. Sociological Focus, 29(3), 189–207. Hunter, L. W. (2000). The adoption of innovative work practices in service establishments. International Journal of Human Resource Management, 11(3), 477–496. Ichniowski, C., & Shaw, K. (1995). Old dogs and new tricks: Determinants of the adoption of productivity-enhancing work practices. Brookings Papers on Economic Activity: Microeconomics (Spring), 1–65. Ichniowski, C., Shaw, K., & Prennushi, G. (1997). The effects of human resource management practices on productivity: A study of steel ﬁnishing lines. American Economic Review, 87(3), 291–313. Kalleberg, A. L., Knoke, D., & Marsden, P. V. (1995). Interorganizational networks and the changing employment contract. Connections, 18(2), 32–49. Kalleberg, A. L., Reskin, B. F., & Hudson, K. (2000). Bad jobs in America: Standard and nonstandard employment relations and job quality in the United States. American Sociological Review, 65(2), 256–278. Kalleberg, A. L., Reynolds, J., & Marsden, P. V. (2003). Externalizing employment: Flexible stafﬁng arrangements in U.S. organizations. Social Science Research, 32(4), 525–552. Kelley, M. R. (1996). Participative bureaucracy and productivity in the machined products sector. Industrial Relations, 35(3), 374–399. Kilbourne, B. S., Farkas, G., Beron, K., Weir, D., & England, P. (1994). Returns to skill, compensating differentials, and gender bias – effects of occupational characteristics on the wages of white women and men. American Journal of Sociology, 100(3), 689–719. Kim, D. O. (1999). Determinants of the survival of gainsharing programs. Industrial and Labor Relations Review, 53(1), 21–42. Kmec, J. A. (2003). Minority job concentration and wages. Social Problems, 50(1), 38–59. Knez, M., & Simester, D. (2001). Firm-wide incentives and mutual monitoring at continental airlines. Journal of Labor Economics, 19(4), 743–772. Konovsky, M. A., & Brockner, J. (1993). Managing victim and survivor layoff reactions: A procedural justice perspective. In: R. Cropanzano (Ed.), Justice in the workplace: Approaching fairness in human resource management (Vol. 1). Hilsdale, NJ: Lawrence Erlbaum Associates. Lawler, E. E., Mohrman, S. A., & Ledford, G. E., Jr. (1992). Employee involvement and total quality management: Practices and results in Fortune 1000 companies. San Francisco: Jossey-Bass. Long, J. S. (1997). Regression models for categorical and limited dependent variables. Thousand Oaks, CA: Sage. Manley, J. E. (2000). Negotiating quality: Total quality management and the complexities of transforming professional organizations. Sociological Forum, 15(3), 457–484. National Crosswalk Service Center, from http://www.xwalkcenter.org/

You Get Paid for that?

143

Osterman, P. (1994). How common is workplace transformation and who adopts it? Industrial and Labor Relations Review, 47(2), 173–188. Osterman, P. (2000). Work reorganization in an era of restructuring: Trends in diffusion and effects on employee welfare. Industrial and Labor Relations Review, 53(2), 179–196. Peterson, N. G., Mumford, M. D., Borman, W. C., Jeanneret, P. R., Fleishman, E. A., Levin, K. Y., et al. (2001). Understanding work using the occupational information network (ONET): Implications for practice and research. Personnel Psychology, 54(2), 451–492. Pil, F. K., & MacDufﬁe, J. P. (1996). The adoption of high-involvement work practices. Industrial Relations, 35(3), 423–455. Rabe-Hesketh, S., Skrondal, A., & Pickles, A. (2002). Reliable estimation of generalized linear mixed models using adaptive quadrature. The Stata Journal, 2(1), 1–21. Reynolds, J. (2006). Teams, teams, everywhere?: Job and establishment level predictors of team use in the United States. Social Science Research, 35, 252–278. Rodrı´ guez, G., & Elo, I. (2003). Intra-class correlation in random-effects models for binary data. The Stata Journal, 3(1), 32–46. Rothschild, J. (2000). Creating a just and democratic workplace: More engagement, less hierarchy. Contemporary Sociology – a Journal of Reviews, 29(1), 195–213. Rothschild, J., & Ollilainen, M. (1999). Obscuring but not reducing managerial control: Does TQM measure up to democracy standards? Economic and Industrial Democracy, 20(4), 583–623. Royston, P. (2004). Multiple imputation of missing values. The Stata Journal, 4(3), 227–241. Sewell, G. (1998). The discipline of teams: The control of team-based industrial work through electronic and peer surveillance. Administrative Science Quarterly, 43(2), 397–428. Shaw, J. D., Gupta, N., Mitra, A., & Ledford, G. E., Jr. (2005). Success and survival of skillbased pay plans. Journal of Management, 31(1), 28–49. Sirianni, C. (1995). Tavistock Institute develops practices of contemporary work reform. Retrieved April 13, 2004, from http://www.cpn.org/topics/work/tavistock.html Sitkin, S. B., Sutcliffe, K. M., & Schroeder, R. G. (1994). Distinguishing control from learning in total quality management: A contingency perspective. Academy of Management Review, 19(3), 537–564. Smith, V. (1996). Employee involvement, involved employees: Participative work arrangements in a white-collar service occupation. Social Problems, 43(2), 166–179. Smith, V. (1997). New forms of work organization. Annual Review of Sociology, 23, 315–339. Spenner, K. I. (1990). Skill – meanings, methods, and measures. Work and Occupations, 17(4), 399–421. Thomas, R. J. (1994). What machines can’t do: Politics and technology in the industrial enterprise. Berkeley: University of California Press. Tomaskovic-Devey, D. (1993). The gender and race composition of jobs and the male-female, white black pay gaps. Social Forces, 72(1), 45–76. Vallas, S. P. (2003). Why teamwork fails: Obstacles to workplace change in four manufacturing plants. American Sociological Review, 68(2), 223–250. Wagner, J. A. I. (1994). Participations effects on performance and satisfaction – a reconsideration of research evidence. Academy of Management Review, 19(2), 312–330. Welbourne, T. M., & Mejia Gomez, L. R. (1995). Gainsharing: A critical-review and a future research agenda. Journal of Management, 21(3), 559–609. Whitﬁeld, K. (2000). High-performance workplaces, training, and the distribution of skills. Industrial Relations, 39(1), 1–25.

144

JEREMY REYNOLDS

APPENDIX ONET 3.1 Job Zones Combined levels of experience, education, and training required for a job. (Information excerpted from Appendices F and G of the 1998 ONET codebook.) Job Zone 1: Little or No Preparation Needed (SVP Range: Below 4.0) Overall Experience – No previous work-related skill, knowledge, or experience is needed for these occupations. Education – These occupations may require a high-school diploma or GED certiﬁcate. Some may require a formal training course to obtain a license. Job Training – Employees in these occupations need anywhere from a few days to a few months of training. Usually, an experienced worker could show you how to do the job. Job Zone 2: Some Preparation Needed (SVP Range: 4.0 too6.0) Overall Experience – Some previous work-related skill, knowledge, or experience may be helpful in these occupations, but usually it is not needed. Education – These occupations usually require a high-school diploma and may require some vocational training or job-related course work. In some cases, an associate’s or bachelor’s degree could be needed. Job Training – Employees in these occupations need anywhere from a few months to one year of working with experienced employees. Job Zone 3: Medium Preparation Needed (SVP Range: 6.0 too7.0) Overall Experience – Previous work-related skill, knowledge, or experience is required for these occupations. Education – Most occupations in this zone require training in vocational schools, related on-the-job experience, or an associate’s degree. Some may require a bachelor’s degree. Job Training – Employees in these occupations usually need one or two years of training involving both on-the-job experience and informal training with experienced workers.

You Get Paid for that?

145

Job Zone 4: Considerable Preparation Needed (SVP Range: 7.0 too8.0) Overall Experience – A minimum of two to four years of work-related skill, knowledge, or experience is needed for these occupations. Education – Most of these occupations require a four-year bachelor’s degree, but some do not. Job Training – Employees in these occupations usually need several years of work-related experience, on-the-job training, and/or vocational training.

Job Zone 5: Extensive Preparation Needed (SVP Range: 8.0 and above) Overall Experience – Extensive skill, knowledge, and experience are needed for these occupations. Many require more than ﬁve years of experience. Education – A bachelor’s degree is the minimum formal education required for these occupations. However, many also require graduate school. Job Training – Employees may need some on-the-job training, but most of these occupations assume that the person will already have the required skills, knowledge, work-related experience, and/or training.

Coding Information and Descriptive Statistics. Variable Deﬁnition

Dependent variables Group incentives Pay-for-learning Proﬁt sharing

146

Table A.1.

Job uses group incentives (l ¼ yes, 0 ¼ no) Job uses pay for learning (l ¼ yes, 0 ¼ no) Job uses proﬁt sharing (l ¼ yes, 0 ¼ no)

Max

Mean or percent

S.D.

0 0 0

1 1 1

15% 26% 41%

0.36 0.44 0.49

0

100

59%

41

0

100

83%

27

0

1

10%

0.30

1

5

2.85

1.06

0

1

22%

0.41

JEREMY REYNOLDS

Job level explanatory variables Devaluation hypothesis Percent female Percent of full-time employees in job who are women Percent white Percent of full-time employees in job who are white Power process perspective Unionization Are any (core/clerical workers) covered by a union contract? (l ¼ yes, 0 ¼ no) Evolutionary economics Preparation required by Average combined levels of experience, job education, and training required for the 1990 census occupation category of job. (See the appendix of this paper and Appendices F and G of the 1998 ONET codebook for more information.) Little or no preparation Average job zone is 1.01.9 required

Min

Core job

Managerial job

Administrative support

Establishment level controls Establishment age Number of FT employees Part of larger organization

Average job zone is 2.02.9 Average job zone is 3.03.9

0 0

1 1

29% 23%

0.45 0.42

Average job zone is 4.04 9

0

1

24%

0.43

Average job zone is 5.0

0

1

2%

0.15

Number of full-time employees in job as a percent of the full-time employees at the establishment The job with the largest number of people who are most directly involved with the establishment’s main product or service (l ¼ yes, 0 ¼ no) The job that involves the largest number of managers, supervisors, or other administrators (l ¼ yes, 0 ¼ no) The secretarial, clerical, or ofﬁce-staff job that involves the greatest number of workers (l ¼ yes, 0 ¼ no)

0.01

100

28%

0

1

44%

0.50

0

1

33%

0.47

0

1

23%

0.42

Age of establishment in years The number of full-time employees at the establishment Is (ORG) in any way part of a larger organization or is it completely independent? (0 ¼ independent, l ¼ part of larger org)

0 1

295 27,665

39 533

0

1

30%

29

You Get Paid for that?

Some preparation required Medium preparation required Considerable preparation required Extensive preparation required Job size

39 1563 0.46

147

(Continued ).

Variable Deﬁnition

Consultation w/r training

Manufacturing industry Non-proﬁt establishment Foreign competition

148

Table A.1.

Does (ORG) consult with an association of organizations in developing its training programs? (l ¼ yes, 0 ¼ no) Establishment manufactures durable or nondurable goods (l ¼ yes, 0 ¼ no) Is this a non-proﬁt or a for-proﬁt organization? (l ¼ non-proﬁt, 0 ¼ for-proﬁt) How much competition would you say there is in your main market or service area from foreign organizations y (1) None (2) very little (3) a moderate amount (4) a great deal

Min

Max

Mean or percent

S.D.

0

1

38%

0.49

0

1

20%

0.40

0

1

25%

0.43

1

4

1.80

1.01

JEREMY REYNOLDS

Table A.2.

Representativeness of NOS-II Data by Establishment Size and Industry. County Business Patterns 1997 Index of Dissimilarityb N

%

Weighted (%)

0–4 5–9 10–19 20–49 50–99 100–249 250–499 500–999 1000+

106 71 88 146 97 117 81 63 158

11.4 7.7 9.5 15.7 10.5 12.6 8.7 6.8 17.0

52.9 18.0 10.7 9.7 2.8 2.7 1.7 0.2 1.3

3,757,627 1,354,488 856,118 572,437 194,068 113,832 29,214 10,903 6,182

54.5 19.6 12.4 8.3 2.8 1.7 0.4 0.2 0.1

0.8 0.8 0.9 0.7 0.0 0.5 0.6 0.0 0.6

Total

927

100.0

100.0

6,894,869

100.0

5.0

Agriculture (0) Mining and construction (I) Manufacturing (2, 3) Transportation, comm., utilities (4) Wholesale and retail trade (5) Finance, insurance, real estate (6) Services (7, 8)

9 46 165 61 209 56 392

1.0 4.9 17.6 6.5 22.3 6.0 41.8

2.0 10.1 4.9 4.3 33.3 4.9 40.4

116,588 693,893 393,091 300,644 2,118,710 676,799 2,543,677

1.7 10.1 5.7 4.4 31.0 9.9 37.2

0.1 0.0 0.4 0.0 1.2 2.5 1.6

Total

938

100.0

100.0

68,434

100.0

5.9

N

%

Size (] FT+PT Employees)

You Get Paid for that?

NOS-II 1996–1997a

Industry (1-digit sic code)

a

149

Public administration and military establishments are not included in the NOS-II calculations because the CPB does not contain information on such establishments. b The index of dissimilarity is calculated as:|NOS-II—CBP|/2. It represents the percentage of establishments that would have to be reclassiﬁed to make the two distributions match.

This page intentionally left blank

150

SELECTIVE PARTICIPATION: CONTROLLING WORKERS’ INPUT AT GENERAL MOTORS Jeffrey S. Rothstein ABSTRACT Post-fordist production systems emphasize the need to tap workers’ knowledge to enhance productivity and quality. Often overlooked, however, is the potential conflict in expecting workers to contribute to processes that may make their jobs harder. This article compares employee participation schemes at two General Motors assembly plants to illustrate the potency of this dilemma and the range of ways managers focus or limit employee participation to achieve the company’s goals. In Silao, Mexico, General Motors carefully constructed a labor relations environment that cultivated broad employee participation. In Janesville, Wisconsin, local managers placed constraints on employee participation to ensure continuous production.

INTRODUCTION Among the commonalities shared by post-fordist production paradigms is the overwhelming consensus that shop ﬂoor workers possess important Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 151–175 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16006-9

151

152

JEFFREY S. ROTHSTEIN

information regarding the production processes in which they labor. Therefore, in sharp contrast to mass production systems based on the strict separation of conception and execution of work, modern manufacturing systems incorporate mechanisms that tap workers’ knowledge to improve production processes and enhance productivity and quality. Often overlooked, however, is that encouraging the worker agency inherent to these employee-participation schemes poses certain risks for ﬁrms. Ultimately, workers decide what knowledge to share with their employers and when to do so. Alternatively, workers may decide to withhold their knowledge altogether, or even to participate in ways that undermine the ﬁrm’s goals. For managers, the risk associated with policies intended to extract greater employee participation increases as the scope of the involvement sought from employees expands. Merely soliciting workers’ ideas offers employers far fewer potential beneﬁts than empowering their employees to make decisions on the shop ﬂoor, such as by enabling them to stop the assembly process if a defect is detected. However, entrusting workers with such authority also exposes employers to the possibility that workers will stiﬂe production by unnecessarily halting production. This dilemma is especially acute in the auto industry, where new manufacturing systems based on the precepts of ‘‘lean production’’ (Womack, Jones, & Roos, 1990) include countervailing pressures and policies. On the one hand, productivity is maximized by maintaining a constant ﬂow of production along an assembly line. On the other hand, quality is supposed to be enhanced by empowering workers to stop the assembly line to call attention to, and address, manufacturing defects. Overall efﬁciency depends, therefore, on management’s capacity to strike a balance between the dual demands of productivity and quality, and ensure that workers do not needlessly stop the assembly line under the false pretense of quality concerns. Therefore, for employers, these employee-participation schemes require more than simply opening the ﬂoodgates of worker involvement to include channeling and ﬁltering that participation to achieve management’s goals of improved productivity and enhanced quality. This article compares the ways that managers ﬁltered employee participation at two of General Motors’ assembly plants producing Chevy Suburbans – a greenﬁeld site in Silao, Mexico and an 80-year-old facility in Janesville, Wisconsin. Both plants were supposed to be operating an identical lean production system that the automaker called its ‘‘Global Manufacturing System,’’ which included programs designed to engage employees in the drive for improved productivity and quality. Each had installed systems that enabled workers to stop the assembly line to address defects in

Selective Participation

153

product quality, and each included ways for workers to contribute their ideas for improving the production process. In reality, contrasting industrial and labor relations led to broad differences in the scope and nature of employee participation at the two plants. In Silao, where the collective bargaining agreement with the local labor union granted General Motors near complete control over personnel decisions and the organization of work, managers carefully screened and trained prospective employees. Once hired, work teams were granted broad discretion over their use of the Andon system by which they could stop the assembly line to address problems to guarantee quality. Additionally, these workers were encouraged to apply the knowledge they gained on the assembly line to recommend changes to the production process. In Janesville, on the other hand, General Motors’ contractual obligations to the United Auto Workers union left them with little control over personnel decisions, and managers were wary of their employees’ motives and reluctant to cede control over the assembly line. Instead, they ﬁltered participation on the shop ﬂoor. Supervisors restricted the use of the Andon system, lest workers hamper productivity through its use, and workers’ suggestions were viewed with skepticism and subjected to lengthy review. Unlike previous studies of lean production and the comparisons they facilitate, these two case studies offer an opportunity to explicitly identify the impact of industrial labor relations on employee participation because so many other variables have been controlled. Whereas the variation found between studies of employee participation schemes to date could only be attributable to a broad range of variables, in this study the industry, company, product, and even the corporate vision of the manufacturing system are identical. Both plants were operating the Global Manufacturing System to assemble Chevy Suburbans and other full-size sport utility vehicles. What differs are the local workers, their unions, and the national economic, industrial relations, and historical contexts into which the Global Manufacturing System and its employee participation schemes were being introduced. Therefore, this study illustrates the impact the social relations of production have on employee-participation programs. Data regarding labor relations and the operation of the Global Manufacturing System at the two plants were collected through intensive ﬁeld research at both sites. I spent a total of six months in Silao, Mexico, split between two visits in 2002 and the beginning of 2003. My research in Janesville began during the summer of 2003, and included several visits a week for about six months. Fieldwork tapered off to once a week through much of 2004, with more infrequent visits continuing into 2005.

154

JEFFREY S. ROTHSTEIN

I was fortunate to gain access to the shop ﬂoor at each plant and observe assembly in process. Through a combination of ofﬁcial plant tours open to the public and private visits as well, I was able to trace the assembly process in both Silao and Janesville from its beginning in the body shop to its culmination in ﬁnal assembly. Paint departments were off-limits to all visitors to protect the ﬁnish on the vehicles. Personal guides at each plant explained the overall organization of production, the roles assumed by line operators and their supervisors, and speciﬁc activities being observed, in addition to responding to whatever edifying questions I asked. Details of the way the Global Manufacturing System operated at the two plants, as well as their respective industrial and labor relations practices, was culled from interviews with managers, union ofﬁcials, and workers at each plant. Managers responsible for human resource and personnel decisions, as well as those involved in the implementation of the Global Manufacturing System were interviewed at both plants, as were elected union ofﬁcials representing the workers in each plant. In addition, workers at each location described their experiences in the factories through both semi-structured interviews and informal discussions. In Janesville, I interviewed several dozen line operators, some repeatedly over the course of the investigation. I was able to interview roughly a dozen Mexican workers, several on more than one occasion. The workers in Silao were more difﬁcult to approach than those in Janesville, in part because they were bused in and out of the plant. In addition, I speak Spanish with a non-native accent, and this may have led the workers to exercise caution while speaking about their experiences working for General Motors. Contacts at both sites also provided company documents that shed light on human resource policies and plant policies more broadly. Regardless of the asymmetries in the manner data was collected in Janesville and Silao, managers, union ofﬁcials, and workers in each of the plants provided overwhelmingly consistent portrayals of the factory in which they worked. The article proceeds as follows. The next section reviews the literature on employee participation schemes as they apply to the shift from mass to lean production in the automobile industry, highlighting the tenuous balance that automakers must maintain as they encourage employee input while attempting to achieve ever high rates of productivity. I then describe General Motors’ vision for its Global Manufacturing System as a lean production system, emphasizing the role envisioned for employee participation within production process. The precise implementations of the participative components of the Global Manufacturing System in Silao and Janesville are then compared, with a focus on the link between each plant’s human

Selective Participation

155

resource policies and the point at which the management exercised control over line operators’ participation to ensure that it did not negatively affect productivity. In the conclusion, I return to a discussion of employee-participation schemes at the two plants and in new manufacturing systems more broadly, ultimately suggesting that they may be as much about industrial and labor relations as productivity and quality.

WORKER PARTICIPATION AND LEAN PRODUCTION In companies such as General Motors, the adoption of lean production involves an implicit renunciation of a fundamental precept of the mass production system by which the automaker grew into one of the world’s largest companies. As widely recounted, but in best detail by Hounshell (1984) and Rubenstein (1992), until the early 1980s, automobile manufacturing among the Big Three automakers (General Motors, Ford, and Chrysler) in the United States was organized in close adherence to Taylor’s (1911) principle of ‘‘scientiﬁc management’’ separating the conception and execution of work. For the most part, workers labored on assembly lines, repeatedly performing tasks in a manner designated to them by management. Little was expected of them intellectually, and few avenues existed for them to contribute their ideas for improving the production process. In fact, workers had no formal role in addressing product defects they spotted in the course of their own work, even if systemic in nature. In the 1970s, the fortunes of U.S. automakers began to change as OPEC’s oil embargo against the United States sent gas prices souring, creating an instant demand for the smaller, more fuel efﬁcient cars produced by Japanese automakers, just as those companies had saturated their own domestic market (Kenney & Florida, 1993). Though the early competitiveness of Japanese vehicles in the United States was widely attributed to their small size and fuel efﬁciency, their continued popularity after the oil crisis was due to the reputation they had developed for engineering, quality, and reliability. Before long, and especially as Japanese ﬁrms began operating assembly plants in the United States, the way in which Japanese automakers built cars came to be viewed as the secret to their success. By the early 1980s, in North America and in much of the rest of the world as well, manufacturing systems pioneered in Japan had exposed the inefﬁciencies and competitive inferiority of mass production, and the Big Three began scrambling to catch up by adopting elements of what became known as ‘‘lean production,’’ the term

156

JEFFREY S. ROTHSTEIN

popularized by Womack et al. (1990) in their description of the Toyota manufacturing system, The Machine That Changed the World. The impact of lean production on auto workers is twofold. First, shifting from mass to lean production intensiﬁes the pace of work along the assembly line. Unlike the ﬂexible specialization (Piore & Sabel, 1984) model in which manufacturers gain a competitive advantage due to the versatility of their skilled workers, lean production rededicates ﬁrms to the standardization of work prescribed by Scientiﬁc Management. Whereas mass production includes the intentional placement of buffers within the manufacturing system as security to ensure the continuous movement of the assembly line, under lean production such buffers are viewed as wasted time, labor, parts, money, etc. that should be eliminated. Rather than employing extra workers and spacing work to allow downtime between each job, under lean production jobs are engineered to keep workers continuously engaged in the production process. Doing so allows a factory to produce more cars with fewer workers. Second, in contrast to Scientiﬁc Management’s fundamental principle separating the conception and execution of work, lean production seeks to engage workers in the drive for improved productivity and quality, transferring ‘‘the maximum number of tasks and responsibilities to those workers actually adding value to the car on the line’’ (Womack et al., 1990, p. 99). As described in nearly all the accounts of Toyota’s manufacturing system (Adler, 1995, 1999a; Besser, 1996; Mishina, 1998) and similar lean production systems (Liker, Fruin, & Adler, 1999; Rinehart, Huxley, & Robertson, 1997; Rubenstein, 2001), work teams at these plants are responsible for quality as well as production. Rather than repairing defective products after assembly is complete, which was the norm under mass production, workers in lean systems are expected to build ‘‘in-station.’’ They are encouraged to treat the next work team in the assembly process as their own ‘‘customer’’ and to never pass a faulty product down the line. To facilitate this, workers have the capacity to stop the assembly line in order to address defects and collectively problem solve if need be. Furthermore, nearly all plants practicing lean production include a version of the Japanese ‘‘kaizen,’’ or continuous improvement system, whereby workers contribute their ideas for improving the production process, quality, safety, or the factory overall. The point is to simultaneously engage everyone involved in the production process in a drive to continuously cut waste – downtime, personnel, parts, scrap, etc. – in order to challenge the manufacturing system to become leaner, while encouraging workers to take responsibility for ensuring product quality. Early proponents (Kenney & Florida, 1993; Womack et al., 1990) of lean production denied any inherent conﬂict between a system that simultaneously

Selective Participation

157

increases the pace and intensity of work while expecting workers to contribute to that process. Instead, they described lean production as a universally beneﬁcial system that workers would embrace because while ‘‘the mass-production plant is often ﬁlled with mind-numbing stress, as workers struggle to assemble unmanufacturable products and have no way to improve their working environment, lean production offers a creative tension in which workers have many ways to address challenges’’ (Womack et al., 1990, pp. 101–102). In his study of New United Motor Manufacturing, Inc. (NUMMI), the Toyota–GM joint venture, Paul Adler (1995) described the combined routinization of work and solicitation of workers’ participation as ‘‘democratic taylorism,’’ implying that equality had been brought to the traditionally hierarchical auto industry. Labor scholars critical of lean production (Babson, 1995; Graham, 1995; Milkman, 1997; Parker & Slaughter, 1995) however, see in lean production not ‘‘a radical break with Fordism but merely a more advanced form’’ in which ‘‘labor control occurs through a set of far more subtle and carefully contrived strategies than in Fordist factories’’ (Dassbach, 1996, p. 21). These researchers labeled lean production ‘‘management-by-stress’’ (Parker & Slaughter, 1988, 1995) and criticized research that espoused the beneﬁts of lean production for workers but ‘‘did not include any direct examination of the impact of this system on workers’’ (Milkman, 1997, p. 141). Furthermore the reorganization of line operators into work teams that is nearly universal under lean production is seen as an effort to squeeze greater output from workers through peer pressure. When combined with a duty to assist in the process of enhancing productivity, critics see a system in which workers are compelled to contribute to the degradation of their own labor standards and work environment. This debate between business and labor scholars focuses on the perceived enhancement or degradation of working conditions associated with lean production rather than any real disagreements over how the system functions. It mostly revolves around the rhetorical excesses of those enthusiastic about new manufacturing systems. However, to the extent the debate reﬂects actual differences between management and labor, it raises important issues affecting the implementation of lean production systems, especially with regard to employee participation programs. If, as lean production’s critics suggest, workers view the schemes as enlisting them in the deterioration of their own working conditions, might they be less likely to participate as their employers intend? Understanding this, how would shop ﬂoor supervisors under pressure to maintain productivity respond? Would they fully implement employee-participation programs that require they hand over control of production to their workers?

158

JEFFREY S. ROTHSTEIN

Current research of lean production in the auto industry does not quite address this dynamic, or fully recognize the potential risk to productivity associated with increased employee participation. Instead, studies have more broadly shown that work practices including a high degree of employee participation are most successful when part of a more comprehensive human resource strategy (MacDufﬁe, 1995; Pil & MacDufﬁe, 1996). In fact, each of what Vallas (2003) refers to as ‘‘celebrity organizations’’ for their well-documented success in implementing lean production include carefully orchestrated-industrial relations that support the lean production system. At NUMMI, Toyota and General Motors negotiated with the United Autoworkers Union (UAW) an untraditional labor agreement that provided a role for the union in governing the shop ﬂoor, as well as worker participation in determining their team leaders, monitoring quality, and improving the production process. Furthermore, former workers at the plant were rehired by a joint labor-management committee that selected workers for their willingness to embrace the new manufacturing system (Adler, 1999a, b; Adler, Goldoftas, & Levine, 1998; Milkman, 1991; Parker & Slaughter, 1988, 1995). Likewise, GM involved the UAW in the planning of its Saturn plant in Spring Hill, Tennessee, adopting a version of lean production with an emphasis on democracy (Adler, Kochan, MacDufﬁe, Pil, & Rubinstein, 1997; Cornette, 1999; Rubinstein & Kochan, 2001). Unlike the automakers contractually obligated to include the UAW, Japanese automakers have universally avoided having to deal with unions in the workplaces they have established in the United States. Instead, accounts of lean production at Honda (Shook, 1988), Subaru-Isuzu (Graham, 1995), and Toyota’s plant in Georgetown, Kentucky (Adler, 1999a; Besser, 1996; Mishina, 1998) demonstrate an intense effort by the companies to establish a shop ﬂoor environment within their factories that minimizes overt conﬂict through combinations of human resource policies, from selective hiring to establishing alternative dispute resolution procedures. These cases also show successful implementations of lean production, but the links between the human resource practices and employee participation remain vague. At other plants, attempts to nurture labor relations to support lean production, or make preemptive adaptations to lean production to accommodate local industrial relations have proven less sufﬁcient, requiring further changes in, and retreats from, lean production. At CAMI, a GM–Suzuki joint venture in Ingersoll, Ontario, workers went on strike for ﬁve weeks in 1992 to demand alterations in the lean production system to improve labor conditions (Rinehart et al., 1997). Likewise, at the plant Ford and Mazda opened in 1987 in Flat Rock, Michigan, employee participation was scaled

Selective Participation

159

back, with improvements to the production process becoming more management driven, reﬂecting a more typical U.S. managerial style. Eventually, the labor contract evolved to include key provisions of the UAW–Ford national contract (Babson, 1998; Fucini & Fucini, 1992). Ford’s attempt to implement Mazda’s production model in Hermosillo, Mexico followed a similar pattern. Opened in 1986, the half billion dollar facility gained quick praise for matching the productivity of its sister plant in the U.S. However, high turnover and labor discontent required the system of lean production originally envisioned be scaled back to ‘‘return to more traditional forms of work organization’’ (Carrillo & Montiel, 1998, p. 314). Like the cases of successful implementations of lean production, these examples point to the need to maintain social relations of production that support employee participation, but the precise manner in which those break down is obfuscated by other factors. The two plants in this study provide a clearer picture of the extent to which a factory regime supporting employee participation can be cultivated and the dynamic and immediacy with which such policies can break down. In Silao, the participation schemes were coupled with a comprehensive human resource strategy that painstakingly prepared workers to contribute to improving the productivity of the plant and quality of its products. In Janesville, shop ﬂoor supervisors with little conﬁdence in their ability to extract employee participation in a manner commensurate with the ﬁrm’s goals of high productivity and quality maintained productivity by restraining workers’ input on the shop ﬂoor. Because both the Janesville and Silao plants were supposed to be implementing the same manufacturing system in assembling the same product, this study is able to highlight the precise success and failure of the system, as well as the sources of each, in a way that cannot be fully extrapolated from broad comparisons of plants manufacturing an assortment of products with different conceptions of lean production and employee participation.

GENERAL MOTORS’ ‘‘LEAN’’ GLOBAL MANUFACTURING SYSTEM General Motors’ Global Manufacturing System was based on the Toyota Manufacturing System that GM learned from the Japanese automaker at NUMMI, their joint venture in Fremont, California. According to the Personnel Director in Janesville, ‘‘GMS is basically the Toyota production system with a General Motors tweak to it, though not a huge tweak. It is

160

JEFFREY S. ROTHSTEIN

basically the old Toyota production system.’’ Janesville’s Plant Planner concurred that ‘‘it is equivalent to the Toyota production system’’ but that ‘‘to put some ownership in it, its called GMS – Global Manufacturing System. And that’s with every plant in General Motors. That is the manufacturing philosophy and principles and guidelines all wrapped into one.’’ In preparing to introduce the Global Manufacturing Systems in their plants, managers from both Silao and Janesville visited NUMMI to observe the manufacturing system in action. As a manufacturing philosophy, General Motors’ described its Global Manufacturing System as based on the ‘‘consistent adoption of ﬁve principles – people involvement, standardization, built in quality, short lead time, and continuous improvement. The principles are interrelated and implemented as a complete system.’’ These ﬁve principles embodied the attitude asserted by proponents of lean production that the manufacturing system can be a boon for management and labor alike. Through ‘‘people involvement,’’ GM claimed to recognize ‘‘its employees as the most valuable resource,’’ who it would ‘‘provide the support necessary to y work in a motivated, empowered and participative way. In other words, the people actually doing the work are typically the best equipped to suggest and implement improvements.’’ This meant that workers would take responsibility for ‘‘standardization’’ of work, the ‘‘dynamic process by which we document, follow and perform our work according to the core standards, terminology, principles, methods and processes to achieve a common base from which to improve.’’ In addition, through ‘‘built-in quality’’ ensuring ‘‘that defects are not passed’’ down the assembly line, workers would be expected to ‘‘ask yourself what you are doing to make your work processes ‘fool-proof’ and ensure the output of the process is without defects.’’ Ultimately, this would foster ‘‘continuous improvement,’’ which GM deﬁned as an ‘‘attitude’’ that ‘‘supports all employees in improving their own jobs and environment for the continuous improvement of the company.’’ Of course, General Motors was not naı¨ ve to the fact that implementation of these ﬁve principles would vary from plant to plant. As the company explained, ‘‘the system is dynamic. With each new plant or renovation of a current plant, it is further reﬁned and implemented with regional variation, based on the individual plant environment, supplier capability, vehicle architecture and cultural factors.’’ There is, perhaps, no better illustration of the range of variation in implementations of Global Manufacturing System than the differences in the ways Silao and Janesville practiced employee participation.

Selective Participation

161

As implemented in both the assembly plants in Silao and Janesville, the Global Manufacturing System included two basic means by which workers were supposed to participate. First, at each plant, workers could submit ideas for improving the production process. Those ideas would be reviewed by management for possible implementation. And at each plant, an incentive program earned workers bonuses if their idea led to cost savings for the automaker or one of their suppliers. Second, each plant had installed an Andon system by which workers could stop the assembly line, the primary tool facilitating ‘‘built-in quality.’’ Of course, the actual ways in which these participation programs functioned differed fundamentally. In Silao, carefully selected and trained workers were given broad discretion over their use of the Andon cord, and their ideas for improving the system were integral to the design and improvement of the production process. However, in Janesville, where most of the employees’ careers predated the Global Manufacturing System, and many objected to the intensiﬁcation of their work that the new manufacturing system required, workers received little comprehensive training around the new manufacturing process or the participative elements to which they were supposed to be the key factors. Instead, control over participation was exercised on the shop ﬂoor, where supervisors concerned with maintaining production closely monitored and limited workers’ involvement. Both plants in this study were under pressure to maintain both high levels of productivity and quality. GM’s full-size sport utility vehicles were among the automaker’s most popular and proﬁtable vehicles at the time this research was being conducted. Indeed, both plants had responded, progressively raising their productivity in recent years. In fact, at least as measured by the industry standard Hours Per Vehicle (HPV), the plants were quite comparable. According to Harbour Consulting, which publishes an annual study of automotive manufacturing performance throughout North America, in 2002, Silao required 27.66 labor hours to produce one sport-utility vehicle (SUV), while Janesville needed 27.73 (The Harbour Report North America, 2003). Likewise, though Harbour had reported in 1998 that in Silao, ‘‘quality beats all other GM C/K plants whether measured by J.D. Power or by warranty’’ (The Harbour Report North America, 1998), by 2004 the gap had been closed. JD Power and Associates ranked the plants virtually equal, with purchasers of vehicles assembled in Janesville and Silao reporting 108 and 109 problems per hundred vehicles (PPH) respectively (Leute, 2005).

162

JEFFREY S. ROTHSTEIN

WORKER PARTICIPATION IN SILAO In Silao, the employee participation aspects of the Global Manufacturing System were integrated, with workers’ ideas generated, at least in part, through their experiences using the Andon system. The Andon system itself was the primary quality control mechanism facilitating the goal of building ‘‘in-station.’’ Each workstation was equipped with an Andon cord that a worker could pull if they experienced a problem, be it a defect that had been passed down the line, a malfunctioning tool, or a faulty part. Pulling the cord would stop the assembly line within the work team’s area, or ‘‘zone’’ of the assembly line, as they were called. Simultaneously, the work team’s assigned alphanumeric code began blinking on overhead LCD signs throughout the plant, and their designated musical jingle would begin playing, thereby alerting all the members of the work team that their coworker was experiencing a problem. The team would then converge on the area to resolve the problem. Ideally, this was done quickly, and production restarted before causing a backup further up the assembly line and a lack of work further down it. Once the immediate problem had been resolved, workers were encouraged to use the experience, and their broader knowledge of the production process, to recommend changes to increase productivity, quality, or safety and health. These ranged from simple ideas, such as relocating stocks of parts to make them easier to reach, to alterations requiring the redesign of the vehicle, its parts, tools, or the installation process. Recommendations accepted by plant management for adoption immediately established a new standard for all to follow. These were documented on a large display in the center of the plant, including copies of the original submission and the names of the workers who made it, a description of the outcome, and before and after photos. The Personnel Relations Manager claimed an 85% participation rate in the plant’s programs to solicit employees’ ideas, which ranged in signiﬁcance both to the workers and General Motors. Some workers reported making simple suggestions that made their work easier. One reported requesting a stool for a job that could be more comfortably performed sitting down. Another praised the system for ‘‘promoting ergonomics’’ after GM accepted his recommendation that a supply of parts be raised to reduce the amount he had to bend to get them. Other recommendations resulted in substantial cost savings for GM or their suppliers, and in these cases workers were rewarded with a percentage. For instance, one worker was awarded nearly $8,000 for suggesting the redesign of a shipping container that reduced the transportation costs for the supplier shipping those parts.

Selective Participation

163

In addition to these chance opportunities, the ‘‘Ideas and Improvements’’ program was a systematic component of improving the production process. According to one production engineer, line operators’ suggestions were a regular and fundamental part of working out the kinks associated with annual model changes. Each year, even small modiﬁcations made to the vehicles so that consumers could distinguish one model year from the next required engineers to alter and standardize the production process at each workstation. As the changes were introduced, waves of recommendations for improvements would ﬂood in from the workers attempting to implement the engineers’ designs. Over the course of a couple of months, the number of recommendations would decrease as the teams settled into what appeared to be the most logical and efﬁcient work processes. The high level of employee participation General Motors achieved in Silao was supported by a comprehensive human resource strategy designed to manufacture consent (Burawoy, 1979) on the shop ﬂoor. It began with General Motors selection and recognition of the union that would represent their employees and continued with the recruitment, training, and socialization of new employees. As was common practice in the region, upon determining to build a new assembly plant in Silao, General Motors selected a union to represent their workers. Human resource managers among GM’s local parts suppliers indicated that doing so was a preventive measure that allowed ﬁrms to avoid organizing drives by ‘‘radical’’ unions that might encourage workers to strike or cause other ‘‘problems.’’ Although a few of the local factories used the opportunity to establish ‘‘sindicato blancos’’ – sham unions common in Mexico that provide employers security by preventing union organizing among workers on the grounds that a union already exists, even if only on paper – GM and most of its suppliers recognized a regional union that maintained a high proﬁle in representing its members. SITIMM, the union’s Spanish acronym, was selected for its non-militant philosophy of embracing globalization and the employment opportunities economic change brought to the region. Renouncing industrial actions in favor of consultation and cooperation with employers, SITIMM nonetheless provided the workers a voice within GM’s assembly plant commensurate with the spirit of teamwork fundamental to local management’s implementation of the Global Manufacturing System. In fact, besides negotiating wages and beneﬁts, SITIMM mostly provided services to their members to help them adapt to work in a factory and ensure they did not lose what were the best paid blue-collar jobs in the region. These goals complemented GM’s own need to maintain low labor turnover to ensure the

164

JEFFREY S. ROTHSTEIN

stability of the work teams upon which the Global Manufacturing System was based. In fact, maintaining low turnover was one justiﬁcation behind the meticulous recruitment and training process General Motors had established to select and socialize their new employees. As the premier employer in the region offering relatively well-compensated, stable employment, GM found itself with no shortage of applicants. Plant employees wearing their embroidered work shirts were even stopped on the street by strangers inquiring as to how they might gain employment at the plant. The challenge for General Motors was to select and train those workers who would best function within the structure of the Global Manufacturing System. In Silao, General Motors recruited mostly men with a ninth grade education. Hiring workers with a ninth grade education was intended to attract applicants with the basic reading and mathematical skills necessary to perform their work, but avoid those whose higher educational achievement might raise their expectations for interesting work or career opportunities, a problem that caused high labor turnover at Ford’s plant in Hermosillo, Mexico (Carrillo, 1995; Carrillo & Montiel, 1998; Shaiken, 1990, 1994). Furthermore, though the Personnel Relations Manager insisted that the automaker had no gender preference, women comprised only 5% of the workforce. This may have been a reﬂection of the prevailing attitude, directly expressed by the human resource manager at one of GM’s suppliers, that women disrupt the ‘‘stability’’ and ‘‘smooth functioning’’ of work teams by taking advantage of the paid maternity leave mandated by Mexican labor law. Regardless, either directly due to their gender, or indirectly through the company’s screening process, women were offered employment at far lower rates than their male counterparts. Applicants that met the basic requirements of being more than 18-years old, with a ninth grade education, and a military release card underwent a lengthy screening process that began with written and practical examinations, was followed by an extensive-paid-training period, and ultimately ended with successful integration into one of the plant’s work teams. First, applicants sat for written tests in basic reading and math, a personality test, and perhaps an IQ test. Then, they were observed in mock work experiences, as the Personal Relations Manager illustrated: I’ll give you an example: We put them to assembly. We give them wheel rims. We give them instructions. We give them safety equipment. And we put them to work in a team to assemble rims. And they give us very valuable information regarding how they react to particular circumstances, other workers. This is very important.

Selective Participation

165

Applicants who passed these initial evaluations and a subsequent physical exam were then offered conditional employment in the form of a 12-week training and assessment period that one union ofﬁcial referred to as the ‘‘coco wash’’ (brainwashing.) The ﬁrst eight weeks were classroom training to prepare new hires to join the workforce as line operators. According to the Personnel Relations Manager: They spend eight weeks in induction getting to know the systems, what we call the ‘‘Silao Production System,’’ that involves all the systems that they operate: the comparison system, the safety system, the quality system, the personnel system. Everything is perfectly deﬁned.

Although, as a rule, GM did not hire workers with industrial experience so as to avoid having to rid them of ‘‘bad’’ habits, the automaker included in its training speciﬁc comparisons of ‘‘traditional’’ and ‘‘transformed’’ manufacturing processes and ‘‘traditional’’ and ‘‘modern’’ organizations. These extensive training materials outlined the type of company that GM hoped to be and, by implication, the company it used to be. Unlike traditional organizations, in modern organizations like the Silao factory, managers at ‘‘the high levels are not omnipotent,’’ ‘‘information moves laterally rather than hierarchically,’’ supervisors are ‘‘facilitators’’ of work, not ‘‘trackers,’’ there is ‘‘continual redeﬁnition of roles,’’ work is done with ‘‘use of all the abilities’’ as a ‘‘comprehensive task,’’ and there is ‘‘compromise and knowledge of goals’’ and a ‘‘congenial work environment.’’ To prepare for working in the modern organization, and in the Silao Production System speciﬁcally, the new employees spent the bulk of their eight-weeks-classroom training learning the interpersonal skills they would need to work and problem solve as members of teams. The 450 page training text distributed to the workers was divided into thirteen sections, with a unit on each of: ‘‘Working in Teams,’’ ‘‘Communicating with Others,’’ ‘‘Inﬂuencing Others,’’ ‘‘Resolving Conﬂicts,’’ ‘‘Reaching Agreement in Teams,’’ ‘‘Rescuing Difﬁcult Meetings,’’ ‘‘Participating in Meetings,’’ ‘‘Valuing Differences,’’ ‘‘Making the Difference,’’ ‘‘Toward Empowerment,’’ ‘‘Empowerment and the Leader,’’ as well as two on ‘‘Supervising Effective Meetings.’’ These were intended to emphasize the role of the workers both in building the cars and conceiving the manner in which those vehicles should be built. Upon completion of the eight-week program, the new employees were then ready to begin the ﬁnal phase of training and evaluation – actually joining a work team. As the Personnel Relations Manager emphasized, theory is important, but ‘‘above all, they are workers in a team. Our

166

JEFFREY S. ROTHSTEIN

operating system here is based on workers in teams.’’ So, the next test was to determine whether they could apply what they had learned, whether they could be ‘‘perfectly integrated into the system.’’ Only after successfully integrating into a work team over a four-week period, under the guidance of a team leader, did the assessment period end and the worker gain the status of an employee. However, during the ‘‘12 weeks containing the principle training,’’ the Personnel Relations Manager insisted that ‘‘they can be rejected at any moment.’’ In the end, roughly one in four applicants who met the ﬁrm’s basic requirements eventually joined the workforce. The high level of employee participation found in Silao, therefore, was the result of more than just the successful implementation of effective policies to solicit and workers’ input. Taking advantage of its status as a major employer in a Mexican greenﬁeld site, General Motors carefully cultivated labor relations that encouraged and supported the employee participation, and provided an avenue for them to address problems outside the production system by turning to their union. Then, only those applicants deemed likely to thrive in, and not challenge, such a system were offered employment.

WORKER PARTICIPATION IN JANESVILLE In contrast to the meticulous screening and training of job applicants in Silao, which was designed to place on the shop ﬂoor a workforce prepared to embrace the participative elements of their jobs, in Janesville, General Motors had little control over the hiring of workers at all. Like the majority of General Motors’ plants in the United States, the average worker in Janesville was in their mid-40s and had more than 20 years seniority in the plant. These workers, a mixture of Janesville natives who had worked in the factory their entire career, and transplants who had moved to the area after plant closings in other parts of the country, started their careers in the mass production era, when employee participation was not integral to the production process. In fact, many of them expressed dismay at the notion that they should tell General Motors how to make them work harder. Even the plant’s younger workers, most of whom were offered jobs during a wave of hiring in 1995, were not subjected to a lengthy screening process like in Silao. Rather, they were hired by lottery. Each GM employee was given a single application they could pass on to anyone they wanted. Those applications were kept on ﬁle and selected randomly as workers were needed. For some applicants, the wait was very short. Others claimed to have received calls to report for an interview after several years had passed.

Selective Participation

167

Regardless, once called, applicants were hired with little or no screening beyond a basic physical exam. Then they went to work without receiving any of the extensive training given to their counterparts in Mexico. As a result, line operators in Janesville expressed little understanding of the philosophy or different components of the Global Manufacturing System. Furthermore, unlike the union General Motors hand-picked to represent their workers in Silao, line operators in Janesville were represented by Local 95 of the UAW, which traced its history to the 1937 sit-down strikes by which the union gained recognition from GM. Like all afﬁliates of the UAW, Local 95’s primary responsibility was to negotiate and enforce the terms of local working conditions, while wages and beneﬁts were negotiated on a nationwide basis. Over the previous two decades, maintaining employment in the plant had become Local 95’s primary focus, if not obsession. For several reasons, the plant employed roughly half the workers it had at its peak in the early 1980s. In the mid-1980s, GM removed an entire assembly line, leaving the plant with only one. In addition, increasing outsourcing had transferred the manufacture of whole sub-assemblies to parts suppliers. Moreover, the adoption of lean techniques designed to keep line operators working 55 seconds out of each minute had gradually whittled away at employment levels as the automaker sought to build more cars with fewer workers. In fact, according to the union’s Shop Chairman, the highest elected union ofﬁcial in the plant, employee participation was intended to engage the workforce in the drive to reduce their own numbers. He indicated that GM was asking the union and the workers to cross a line between participating in the improvement of product quality and plant safety into a realm in which workers recommended to the automaker, ways in which they could cut jobs. He didn’t object to workers participating per se, but suggested that the plant management was going too far. As he explained: What’s involved is giving the people on the ﬂoor more input in to the daily runnings of the plant. If they [GM] make each area into a little bitty team, that team, their job responsibility is a certain amount of jobs in that area. How can they make them better? How can we resolve quality issues? But management wants to come out and say ‘‘we need to take a person out. How do you think we can take a person out?’’ That’s when we put the ﬂag up. No, that ain’t our job, to tell you how to eliminate people. They [GM] got their ulterior motives that are tied in there. They want the people to agree on how they’re going to work harder.

It was into this overall labor relations environment that General Motors introduced the same components of the Global Manufacturing System intended to harness worker agency and initiative that they had in Silao: the

168

JEFFREY S. ROTHSTEIN

Andon System and programs for soliciting employees’ ideas and recommendations for improving the work process. According to the plant’s Personnel Manager, these were supposed to operate much as they did in Silao to ensure quality production. He expressed his opinion that: The advantage of the Andon system, or GMS is that they [the line operators] determine whether they can ﬁx something in station. So, they are always shifting good work – always shifting a good truck. If they’ve got problems, they’ve got a chord that they can pull, and somebody will immediately respond.

However, workers on the assembly line complained that the Andon system did not operate that way. Numerous line operators spoke of supervisors more concerned about the number of vehicles built than their quality, begging, exhorting, and demanding them not to pull the cord. ‘‘Don’t shut me down’’ was the refrain. One woman working in the paint department summed up the commonly expressed contention that management’s dedication to quality was more theoretical than practical, especially when it came to a choice between quality and production numbers: They can come right through and the foreman will tell you that ‘‘we’re here for quality. Quality counts. You see dirt on that job, you stop the line, you sand out that dirt. You polish that job. You make sure that you get every piece of dirt out of that job because quality is our number one priority.’’ And then we’ll have a breakdown, and all of a sudden ‘‘don’t stop that line for nothing. If you can’t get it, that’s alright. Just let it go. They’ll get it in the next area. Don’t worry about it. Don’t shut down the line for any reason.’’ All of sudden when it comes down to the numbers, and the numbers are low, we don’t want quality anymore, you get that truck out that door.

One of her coworkers concurred, drawing the conclusion that, ‘‘management likes to preach that the people down there are the most important aspect they have. Safety is above anything else. Quality means more than numbers. I guess they talk the talk but they don’t walk the walk.’’ Moreover, workers complained that their supervisors were disinclined to investigate the root cause of systemic problems, which really did frustrate workers as much as proponents of lean production (Kenney & Florida, 1993; Womack et al., 1990) claim in stressing the systems superiority over mass production. But line operators in Janesville indicated that, as one put it, management’s attitude was more likely to be ‘‘what can we do to get you to shut up right now? We’ll zip your lip for today and when it [the problem] comes back in two days, we’ll do the band-aid again without getting to the real issue.’’ One team coordinator working the night shift in the body shop recalled that when he worked troubleshooting problems as a repairman on the assembly line, he was speciﬁcally instructed to ‘‘never shut the line down on Friday’’ because the entire day was mandatory overtime paid at

Selective Participation

169

time-and-a-half, and so stoppages disproportionately drove up the labor costs of those vehicles. However, even as they complained about the restrictions shop ﬂoor supervisors placed on their use of the Andon system, line operators acknowledged that some of their coworkers could not be trusted with the authority, and almost every worker interviewed told stories of their coworkers abusing the Andon system, pulling the cord for no reason, or when they could easily ﬁx a problem themselves. One worker with more than 30 years seniority explained that in his new job he pulled the Andon cord because ‘‘the line moves too fast. I can’t do the work. I don’t know how to use the tools. I can pull the [Andon] cord or I can send product out back [to the ﬂoat to be repaired]. So, I pull the cord.’’ A worker in the body shop said he knew a coworker who intentionally allowed his supply of parts to run out so that he could pull the Andon cord and take a break. One team coordinator proudly offered that her team led the plant in Andon pulls per shift, and that she encouraged her team to pull the cord to demonstrate to GM that their work load was too intense. For workers, the perceived disinterest by management to their input extended beyond the Andon system to the programs in place to solicit their ideas for improving the system. Like in Silao, workers in Janesville could submit to management their recommendations for improving the production process, and like with Silao’s ‘‘Ideas and Improvements’’ program, cost-saving ideas could earn bonuses for workers. In fact, the Personnel Manager envisioned the Janesville system by which line operators could inﬂuence the standardized work process as mimicking that in Silao, where workers perform their jobs as directed until their recommendations were accepted and become the new standard. He said: We’re going down the same path. We’re not quite there yet. The processes for us are very similar to what the processes will be for Silao. When I speak about the team members being able to decide how they are going to do the work, there is a scroll or a standardized way that they are going to have to do that, but they are afforded the opportunity to make the suggestions and make the changes, but once they do that it becomes locked in and that becomes the standardized way that we do that work on both shifts.

Team members, however, reported that their ideas and advice went largely ignored, or languished. One reported raising an issue repeatedly over a 10-month period, and that ‘‘nothing has happened, and it won’t.’’ Another line operator claimed he eventually asked the union to ﬁle a grievance on his behalf over management’s failure to address his long-term concern that a team further up the line was regularly installing faulty clamps into vehicles rather than pulling the Andon cord. Rather than build in-station, they were

170

JEFFREY S. ROTHSTEIN

passing the problem down the line to his team, whose work was disrupted because the bad clamps would not receive the parts they were installing. Although one high ranking union ofﬁcial reported that when he worked on the assembly line he received a substantial bonus for recommending a design change to a particular part, he expressed exasperation that it took almost an entire year before the automaker acknowledged and implemented his recommendation. A former manager in Janesville, who also spent some time at the plant in Silao, conﬁrmed that workers’ ideas were met with far more indifference in the Wisconsin plant. He indicated that the prevailing attitude in Janesville was that designing the production process was the role of engineers. In improving the system, therefore, the norm was not to look to workers for suggestions, but to return to the engineers. As he explained, ‘‘everything is done from the outside [of the plant]. It’s basically, ‘ergonomics – well we’ve got ergonomics engineers out there.’ If you have a problem you get a hold of an ergonomics engineer and study the problem independent of the person actually doing the work.’’ Furthermore, he explained, management was generally skeptical of workers’ ideas precisely because supervisors knew that workers tended to link the suggestion program to the other aspects of the Global Manufacturing System that resulted in a faster and more intense assembly line than they had experienced under the mass production system. Therefore, managers ﬁgured that workers would withhold their best ideas, even those that might make their work easier, ﬁguring that ‘‘man, if I make my job easier, I could probably do things faster, and that means they’ll put more work content into my 57 seconds. So, I don’t want any part of that.’’ This assumption was bolstered by the large number of recommendations that were simply profane. Workers also used the system to air complaints. ‘‘People will put in a suggestion that says ‘put a sign by the drinking fountain that says ‘no spitting’. And then they’ll make 50 more of them for every drinking fountain in the plant. So they just clog it with all sorts of stuff’’ that made sifting through the recommendations more of a chore than it was usually worth’. Unlike in Silao, where GM’s human resource strategy maintained a labor relations climate in which supervisors entrusted workers with the ability to stop the assembly line and relied on their input to improve the production process, Janesville was caught in a vicious cycle that prevented the participation schemes from taking hold. Under pressure to maintain productivity and distrustful of their line operators, supervisors prevented workers from stopping the assembly line and paid little attention to their suggestions. For line operators already skeptical of management’s professed interest in input

Selective Participation

171

from workers, their supervisors’ actions reinforced the consensus among the workforce that the Global Manufacturing System was nothing more than a disingenuous attempt by GM to make them work harder. Most learned to ignore the Andon cord and withhold their suggestions for improving the production process. A minority misused the system, thereby reinforcing their supervisors’ concerns that, if allowed, workers would participate with the intent of slowing production.

CONCLUSION In their zeal to highlight the potential beneﬁts of employee participation in the workplace, proponents tend to avoid addressing the tension inherent to asking workers to contribute to the process of intensifying their own work. Instead, they have generally attempted to argue themselves out of the dilemma by claiming that the enhanced intellectual challenge the workers enjoy overrides any physical stresses the systems might cause. Evidence suggests, however, that whether in a unionized setting or not, engaging workers in such activities requires cultivating a shop ﬂoor environment in which both employees and their managers expect participation to facilitate the ﬁrm’s goals. Otherwise, the risk of losing productivity to employee participation may outweigh the potential gains. The cases presented here illustrate this point. In both cases, General Motors was attempting to extract employee participation as part of their Global Manufacturing System. In Silao, the automaker was able to integrate the participation schemes as an integral part of the lean production system by indoctrinating carefully selected employees into the system and placing them onto the assembly line. In Janesville, however, where the new manufacturing system was being introduced to workers accustomed to their role in mass production, the employee participation aspects became marginalized within the overall manufacturing system almost from the outset. Shop ﬂoor supervisors placed constraints on workers’ participation to forestall workers from using the Andon system to slow production. This reinforced workers’ suspicions that the Global Manufacturing System was designed to extract their labor rather than their ideas, lessening the likelihood of their full participation, and thereby fulﬁlling their supervisors’ expectations regarding the level and type of participation they should expect. In both cases, however, the key for managers was to exert sufﬁcient control to keep employee participation on track to achieve management’s goals, or at least not to allow workers to hinder productivity through the systems.

172

JEFFREY S. ROTHSTEIN

If participation schemes were universally beneﬁcial, and workers’ embrace of them a given, General Motors would not have gone to such lengths to screen their applicants in Silao, eliminating women, anyone with previous industrial experience, or perhaps too strong an independent streak. Likewise, had GM’s managers in Janesville sincerely intended to risk delegating some control over the assembly line to workers, they probably have provided line operators training similar to that received by workers in Silao, so the workers would have a better understanding of the Global Manufacturing System. It appears more likely, however, that local managers never believed the lean production system devised in Detroit could be fully implemented with the workers they had. But General Motors is in the business of producing automobiles, not innovative manufacturing systems. And though the implementation of GM’s lean production system in Janesville lacked the employee participation seen in Silao, the best available measures of productivity and quality suggest that the plants were equally efﬁcient in assembling Chevy Suburbans. Apparently, the management-led improvements emphasizing the work of engineers in Janesville were equally as effective as the worker-inspired changes that became routine in Silao. Perhaps employee participation is just one tool to enhanced productivity and quality that General Motors saw as advantageous in Silao but not Janesville, in much the same way that the automaker responded to the wage gap between the United States and Mexico by automating the Janesville plant more thoroughly than their Silao facility. Alternatively, it is possible that effective employee participation is more fundamentally a component, rather than a result, of manufacturing consent on the shop ﬂoor. Allowing workers in Silao to stop the assembly line and encouraging them to contribute their ideas for improving an assembly process that has already been engineered to be highly synchronized and routinized might be part of the same human resource strategy that led General Motors to recognize a local union and adopt an applicant screening process that includes eight-weeks training for what are essentially unskilled assembly jobs. While not necessarily a means of co-opting workers to contribute to their own exploitation – as some critics of lean production claim, employee participation might provide an additional avenue for workers to redress problems that would otherwise fester and escalate. An absence of genuine employee-participation programs in Janesville might have been a simple recognition that the dynamics of local labor relations precluded their effectiveness, and an understanding that consent was not so much manufactured as coerced through the very real threat that General Motors could decide to close their oldest assembly facility.

Selective Participation

173

This is not to imply that employee-participation schemes are nothing more than labor relations’ manipulations under the guise of enhancing productivity and quality. Within some industries and ﬂexible manufacturing systems, they may very well be a crucial element to remaining competitive. However, in the auto industry and others like it, in which the manufacture of nearly identical, complicated products proceeds along a carefully synchronized assembly line on which workers’ tasks have been meticulously calculated to keep them in near constant motion, the impact of employee participation on the overall system is suspect, and the risk of sacriﬁcing productivity to work stoppages signiﬁcant. Otherwise, the performance of the plant in Silao would have exceeded that of Janesville, perhaps encouraging the supervisors in Janesville to risk implementing the participative elements of the Global Manufacturing System more thoroughly in their own plant.

REFERENCES Adler, P. S. (1995). ‘‘Democratic taylorism’’: The Toyota production system at NUMMI. In: S. Babson (Ed.), Lean work: Empowerment and exploitation in the global auto industry (pp. 207–219). Detroit: Wayne State University Press. Adler, P. S. (1999a). Hybridization: Human resource management at two Toyota transplants. In: J. K. Liker, P. M. Fruin & P. S. Adler (Eds), Remade in America: Transplanting and transforming Japanese management systems (pp. 75–116). New York: Oxford University Press. Adler, P. S. (1999b). Teams at NUMMI. In: J.-P. Durand, P. Stewart & J. J. Castillo (Eds), Teamwork in the automobile industry: Radical change or passing fashion? (pp. 126–150). London: MacMillan Press. Adler, P. S., Goldoftas, B., & Levine, D. I. (1998). Stability and change at NUMMI. In: R. Boyer, E. Charron, U. Jurgens & S. Tolliday (Eds), Between imitation and innovation: The transfer and hybridization of productive models in the international automobile industry (pp. 128–160). Oxford: Oxford University Press. Adler, P. S., Kochan, T. A., MacDufﬁe, J. P., Pil, F. K., & Rubinstein, S. (1997). United States: Variations on a theme. In: T. Kochan, R. D. Lansbury & J. P. MacDufﬁe (Eds), After lean production: Evolving employment practices in the world auto industry (pp. 61–83). Ithaca: Cornell University Press. Babson, S. (1995). Lean production and labor: Empowerment and exploitation. In: S. Babson (Ed.), Lean work: Empowerment and exploitation in the global auto industry (pp. 1–37). Detroit: Wayne State University Press. Babson, S. (1998). Mazda and Ford at ﬂat rock: Transfer and hybridization of the Japanese model. In: R. Boyer, E. Charron, U. Jurgens & S. Tolliday (Eds), Between imitation and innovation: The transfer and hypbridization of productive models in the international automobile industry (pp. 161–188). New York: Oxford University Press. Besser, T. L. (1996). Team Toyota: Transplanting the Toyota culture to the Camry plant in Kentucky. Albany: State University of New York Press.

174

JEFFREY S. ROTHSTEIN

Burawoy, M. (1979). Manufacturing consent: Changes in the labor process under monopoly capitalism. Chicago: University of Chicago Press. Carrillo, J. (1995). Flexible production in the auto sector: Industrial reorganization at FordMexico. World Development, 23(1), 87–101. Carrillo, J., & Montiel, Y. (1998). Ford’s Hermosillo plant: The trajectory of development of a hybrid model. In: R. Boyer, E. Charron, U. Jurgens & S. Tolliday (Eds), Between imitation and innovation: The transfer and hypbridization of productive models in the international automobile industry (pp. 295–318). New York: Oxford University Press. Cornette, G. (1999). Saturn: Re-engineering the new industrial relations. In: J.-P. Durand, P. Stewart & J. J. Castillo (Eds), Teamwork in the automobile industry: Radical change or passing fashion? (pp. 85–106). London: MacMillan Press. Dassbach, C. H. A. (1996). Lean production, labor control, and Post-fordism in the Japanese automobile industry. In: W. C. Green & E. J. Yanarella (Eds), North American Auto Unions in crisis (pp. 19–40). Albany: State University of New York Press. Fucini, J. J., & Fucini, S. (1992). Working for the Japanese: Inside Mazda’s American auto plant. New York: The Free Press. Graham, L. (1995). On the line at Subaru-Isuzu: The Japanese model and the American worker. Ithaca: Cornell University Press. Hounshell, D. A. (1984). From the American system to mass production 1800–1932: The development of manufacturing technology in the United States. Baltimore: Johns Hopkins University Press. Kenney, M., & Florida, R. (1993). Beyond mass production: The Japanese system and its transfer to the U.S.. New York: Oxford University Press. Leute, J. (2005). Rankings give GM good news. Janesville Gazette (May 29). Liker, J. K., Fruin, W. M., & Adler, P. S. (Eds) (1999). Remade in America: Transplanting and transforming Japanese management systems. New York: Oxford University Press. MacDufﬁe, J. P. (1995). Human resource bundles and manufacturing performance: Organizational logic and ﬂexible production systems in the world auto industry. Industrial and Labor Relations Review, 48(2), 197–221. Milkman, R. (1991). Labor and management in uncertain times: Renegotiating the social contract. In: A. Wolfe (Ed.), American at century’s end (pp. 131–151). Berkeley: University of California Press. Milkman, R. (1997). Farewell to the factory: Auto workers in the late twentieth century. Berkeley: University of California Press. Mishina, K. (1998). Making Toyota in America: Evidence from the Kentucky transplant, 1986–1994. In: R. Boyer, E. Charron, U. Ju¨rgens & S. Tolliday (Eds), Between imitation and innovation: The transfer and hypbridization of productive models in the international automobile industry (pp. 99–127). New York: Oxford University Press. Parker, M., & Slaughter, J. (1988). Choosing sides: Unions and the team concept. Boston: South End Press. Parker, M., & Slaughter, J. (1995). Unions and management by stress. In: S. Babson (Ed.), Lean work: Empowerment and exploitation in the global auto industry (pp. 41–53). Detroit: Wayne State University Press. Pil, F. K., & MacDufﬁe, J. P. (1996). The adoption of high-involvement work practices. Industrial Relations, 35(3), 423–455. Piore, M. J., & Sabel, C. (1984). The second industrial divide: Possibilities for prosperity. New York: Basic Books.

Selective Participation

175

Rinehart, J., Huxley, C., & Robertson, D. (1997). Just another car factory? Lean production and its discontents. Ithaca: Cornell University Press. Rubenstein, J. M. (1992). The changing U.S. auto industry: A geographical analysis. New York: Routledge. Rubenstein, J. M. (2001). Making and selling cars. Baltimore: The Johns Hopkins University Press. Rubinstein, S., & Kochan, T. (2001). Learning from Saturn: Possibilities for corporate governance and employee relations. Ithaca: ILR Press. Shaiken, H. (1990). Mexico in the global economy: High technology and work organization in export industries. Unpublished manuscript, San Diego. Shaiken, H. (1994). Advanced manufacturing and Mexico: A new international division of labor? Latin American Research Review, 29(2), 39–71. Shook, R. L. (1988). Honda: An American success story. New York: Prentice-Hall Press. Taylor, F. W. (1911). Principles of scientific management. New York: Harper & Brothers. The Harbour Report North America 1998. (1998). The Harbour Report North America 1998. Troy: Harbour and Associates, Inc. The Harbour Report North America 2003. (2003). The Harbour Report North America 2003. Troy: Harbour and Associates, Inc. Vallas, S. P. (2003). Why teamwork fails: Obstacles to workplace change in four manufacturing plants. American Sociological Review, 68(2). Womack, J. P., Jones, D. T., & Roos, D. (1990). The machine that changed the world. New York: Rawson Associates.

This page intentionally left blank

176

PART II: RECENT TRENDS IN COOPERATIVES

177

This page intentionally left blank

178

DEMOCRACY AT WORK WITHIN THE MARKET: RECONSIDERING THE POTENTIAL George Cheney ABSTRACT This essay treats both democracy and the market. The essay assesses the condition of modern workplace democracy and reconsiders the potential for genuine and thoroughgoing democratic practices within corporations that find themselves ‘‘globalizing’’ and responding to market forces. To ground my analysis, I draw upon the case of the Mondrago´n cooperatives in the Basque Country, Spain, a system of worker-owned-and-managed cooperatives that began to engage the European and global markets in the years immediately preceding European Union unification in 1992. I wish to update and widen the scholarly discussions of Mondrago´n while also using the case to identify some of the most important contours of the intersection of democracy and the market today.

INTRODUCTION Democracy is at once a tradition with a presumed unbroken lineage; a mythology; a common point of reference; an instrument of political will; and Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 179–203 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16007-0

179

180

GEORGE CHENEY

a set of speciﬁc practices related to access, voice, decision, and efﬁcacy. Above all, ‘‘democracy’’ is a contested term. Nowhere is this fact more evident than in the varied appeals to democracy in national political cultures as diverse as those of China, Mexico, Cuba, Venezuela, Iraq, Russia, and the U.S. From the standpoint of neoliberal economics, which now holds sway in North America, Europe, and many other parts of the world, the best assurance of democratic political institutions is unrestrained free trade (e.g., Friedman, 1999). In this way, a particular form of corporate-consumer capitalism is de facto equated with the democratization of states and societies (see the analysis of this and related arguments in Almond, 1991). This is only one view, of course, however dominant it may be on the world stage and however much it may rely upon a mythic portrayal of industrialization of nations that denies the real protectionism that occurred in perhaps every case (Korten, 1995). This essay treats both democracy and the market. While it ranges broadly in terms of ideas and practices under examination, the present account focuses attention on the potential for democratic work practices within the context of market globalization. Speciﬁcally, the essay assesses the condition of modern workplace democracy and reconsiders the potential for genuine and thoroughgoing democratic practices within corporations that ﬁnd themselves ‘‘globalizing’’ and responding to market forces. Especially, it is important to reconsider what ‘‘participation’’ means – or can mean – within the corporation as it responds to and itself fashions processes of globalization, including the privileging of the consumer. That is, given market globalization, what can we expect, or hope for, in terms of democracy ‘‘inside’’ the ﬁrm – for the employees at all levels of the organization? This is a reasonable assessment to make, especially given the assertion of contemporary neoliberal advocates that a globalized market is the best route to individual participation as well as prosperity (Halal, 1996). And, while workplace participation and broader socioeconomic participation are not always joined, there is a strong case to be made for the relatively consistent application of democratic principles across these spheres (Pateman, 1970). To ground my analysis, I draw upon the case of the Mondrago´n cooperatives in the Basque Country, Spain, a system (or rather two systems, including the ULMA Group, which broke away from the Mondrago´n Cooperative Corporation [MCC] in 1992 over issues of local autonomy and control) of worker-owned-and-managed cooperatives that began to engage the European and global markets in the years immediately preceding European Union uniﬁcation in 1992. In its 50-year history, this set of worker cooperatives have moved to a prominent position in the Basque, Spanish and even European economies, while still attempting to hold onto

Democracy at Work within the Market: Reconsidering the Potential

181

foundational values that include equality, solidarity, and participation. This celebrated case still begs the question, ‘‘Has Mondrago´n sold out?’’ To some extent, I will address that question through an analysis of both the cooperatives’ internal dynamics and their changing relations to the exterior. My observations come from a seven-year study of Mondrago´n from 1992 to 1999 (Cheney, 1995, 1997, 1998, 1999, 2001, 2002, 2004), and periodic contact with the cooperatives since that time. With this essay, I wish to update and widen the scholarly discussions of Mondrago´n while also using the case to identify some of the most important contours of the intersection of democracy and the market today.

MEANINGS AND PRACTICES OF DEMOCRACY Of all the pithy and provocative ideas about democracy that I could quote or paraphrase here, this one jumps to mind: that the ancient Greeks had absolutely no doubt as to the possibility of democracy, yet they questioned its ultimate worth because of their ongoing reﬂection about multiple options in political organization. By contrast, ‘‘we’’ have no doubt whatsoever of the ultimate worth of democracy, yet we remain unsure of its practical possibility (Corcoran, 1983). In other words, we are unshaken in holding out a certain vision of democracy as a point of orientation, but less sure of how to get there. When we talk about work, the room divides on both the question of goal and the one of possibility. In the domain of work, democracy has always had an uneasy status, even in the Western world with its mythological allegiance to Greco–Roman ideals. In fact, the ancient Greeks relegated work – particularly physical labor – to a realm outside of the public sphere (Ciulla, 2000). After all, the citizens of ancient Athens were landed white males who sought to liberate themselves from work. Work had absolutely nothing to do with democracy, from the citizen’s standpoint. In a slavebased economy, the idea of democratizing work would have been anathema to these privileged men who were most concerned with managing estates, determining public policy, and ensuring the supremacy of their city-state. A multitude of reviews can be found on the subject of the historical and contemporary forms of workplace democratization. These studies represent the disciplines of sociology, political science, communication, management, economics, and anthropology. Still, the most famous formulation of participation at work in relation to broader democratic ideals is Carole Pateman’s (1970) Participation and Democratic Theory, in which she argues that in the cases of both organizational and national-level participation, it is

182

GEORGE CHENEY

important to get concrete about what is practiced and what is possible, lest myth overpower analysis. Pateman’s broad-ranging examination of democracy at work relied upon systems then in place in Yugoslavia for primary examples, and her study became the touchstone for further work on efforts to democratize work. I will not attempt to summarize the various participative patterns and models in democratically inclined work organizations, but I will observe that an overwhelming majority of such cases has been more interested in productivity at work than in genuine democratic practice (Dachler & Wilpert, 1978; Greenberg, 1986; Heller, Pusic´, Strauss, & Wilpert, 1998; Lichtenstein & Harris, 1993; Stohl & Cheney, 2001; Strauss, 1982). In other words, democratic goals per se have usually been subordinated in the structures and practices of participative systems (Cloud, 2005). Still, there are important cases from the domains of worker cooperatives, feminist organizing, and social movements that have successfully maintained their commitment to democratic goals and values (Iannello, 1992; Lindenfeld & Rothschild-Whitt, 1982; Rothschild & Whitt, 1986). One of the most celebrated and scrutinized (and sometimes dismissed) cases of workplace democracy is that of the Mondrago´n cooperatives. In the 50-year history of these cooperatives we ﬁnd multiple dimensions of democracy at work: political, social, and economic senses of employee participation and voice. And interpretations of this experience range from unrestrained idealism (Morrison, 1991) to cynical suspicion (Kasmir, 1996). Most studies of the Mondrago´n cooperatives fall somewhere between these poles, usually emphasizing one or a few aspects of the enterprises – including emergence (Bradley & Gelb, 1983), structure (Whyte & Whyte, 1991), ﬁnancial performance (Thomas & Logan, 1982), leadership (Azurmendi, 1992), and organizational culture (Greenwood & Gonza´lez, 1989). My own longitudinal case study of Mondrago´n began with a focus on the workings of participation in several representative cooperatives (as parts of the two corporations of MCC and ULMA) but gradually broadened to consider economic, cultural, and rhetorical issues related to the cooperatives’ place in a global market.

COOPERATIVISM VIS-A`-VIS POLITICAL, SOCIAL, AND ECONOMIC CONCEPTIONS OF DEMOCRACY A close examination of worker cooperatives in general leads us to ask important questions about democracy’s multiple dimensions of political

Democracy at Work within the Market: Reconsidering the Potential

183

involvement, social engagement, and economic equality. As much as any other work since Marx’s (1978) Capital, Dahl’s (1985) A Preface to Economic Democracy asked us to take seriously the idea that democratizing economic activities and relationships – and by implication, worklife – must be part of a genuine and embracing vision of democratic potential in contemporary industrial society. In effect, Dahl asks whether the ideals we hold highest and closest apply to domains like work, and the economy where we spend so much time and upon whose vitality our livelihood depends. Dahl then answers this overarching question in the afﬁrmative, while also acknowledging the limitations to democratic practices at work: for example, in terms of size of organization, transparency of information, and level of individual motivation to engage. However, he refutes the common strawman argument against self-governing enterprises that ‘‘you can’t have the monkeys running the zoo’’: a popular managerialist view that negates the work-based knowledge and expertise of people at most levels of a ﬁrm. Dahl advocates worker-owned-and-governed enterprises on the grounds of principles of equality, liberty, and consistency. The third principle refers to the fact that democracy should not be upheld in one domain of public life and simultaneously suppressed in another. Dahl’s perspective is important not only because it advances from an a priori rights perspective but also because it developed over time in response to what Dahl observed as basic injustices in the structuring of worklife. In that regard, worker cooperatives have been featured as hopeful possibilities in a variety of economic, political, and social theories (compare Ellerman, 1992; Harris, 1997; Vanek, 1977). On the practical level, scholars, activists, and practitioners will identify a broad movement of ‘‘cooperativism,’’ which features worker-owned-andmanaged cooperatives alongside producer and consumer cooperatives (e.g., Fowler, 1947; Krimerman & Lindenfeld, 1992; Zwerdling, 1978). In fact, there is an international association dedicated to these ideals and practices (International Cooperative Alliance, 1996), along with a number of similarly dedicated national-level and regional associations. This movement traces its inspiration and examples to the so-called Utopian communities of Britain and the U.S. during the 19th century, and especially to the enterprises of Scottish industrialist Robert Owen. This lineage is important not only because of the values that are invoked (such as collective ownership at the level of the ﬁrm) but also because it makes clear the complexities of any particular case. That is, despite the celebrated ‘‘autonomy’’ of the cooperatives founded by Owen and his allies, these worker cooperatives were neither autonomous with respect to the ﬁnancial vicissitudes of the market nor in the sense of the individual employee being

184

GEORGE CHENEY

relatively free of over-the-shoulder monitoring at work (Wren, 1987). Thus, even in the history and mythos of cooperativism, we ﬁnd important tensions and contradictions that are still being debated today (Uribe-echebarria, 1981). Another important aspect of cooperativism is that it probably does not qualify as a social movement (certainly not a global one) today in terms of scale, coherence, identity, or inﬂuence. What may be fairly said instead is that cooperativism represents a set of ideals about economic and social participation that are applied to varying degrees within diffuse networks of worker-owned-and-managed enterprises (Adams & Hansen, 1992; Cornforth, Thomas, Lewis, & Spear, 1988; Logue, Glass, Patton, Teodosio, & Thomas, 1998; Rosen, Klein, & Young, 1986). Many of those enterprises, including various examples in the U.S. and Canada, make direct reference to the Mondrago´n experience as one of the best examples of economic success with a democratic organizational structure. A few examples of Mondrago´n-inspired worker cooperatives are the Burley Design Cooperative of Eugene, Oregon (a leading manufacturer of bicycle carts) (see Cheney, 1995), the Arizmendi Bakery of Oakland, California (T. Huet, personal communication, June 1999), and New Dawn Enterprises of Cape Breton, Nova Scotia (a community economic development corporation) (see MacLeod, 1997). And, there are advocates of Mondrago´nlike systems in Australia, Israel, Japan, and Mexico, among other countries (Dauncey, 1998; Mathews, 1999). These and other cases naturally tend to select certain aspects of the Mondrago´n cooperatives, although what they share are: (1) some commitment to equal ownership by employee-members, (2) some commitment to democratic decision-making by employee-members, and (3) a foundational belief in the viability of Mondrago´n-like experiments outside of the Basque cultural context and Spanish historical experience. In other words, demo cracy is conceived in both economic and decision-making terms, and Mondrago´n’s governing structure is used in part to launch the other cooperatives. In the case of Mondrago´n itself there is the nagging question of commitment to the cooperative movement more generally, particularly as the MCC has largely ceased to advertise/market itself as a set of employeeowned-and-managed businesses and while it has acquired numerous noncooperative ﬁrms around the world (Cheney, 1999, 2002). The philosophy of cooperativism, particularly as manifest in worker cooperatives, joins economic and democratic rights – although in a very different manner than contemporary neoliberal economic rhetoric. In the typical worker cooperative, common control over the equity of the ﬁrm is

Democracy at Work within the Market: Reconsidering the Potential

185

closely linked to political voice in terms of self-determination. Thus, ‘‘ownership’’ has both strict economic and sociopolitical senses. At the same time, we must acknowledge that these two forms of participation are both analytically and in practice independent. That is, a worker cooperative may be based on worker ownership of all of the equity or ﬁnancial value of the ﬁrm while not allowing for much participation at the level of policy decision-making. In fact, this is probably the modal arrangement in large worker cooperatives around the world.1 The distinctions between types of participation are important in the case of Mondrago´n and for cooperativism in general. The original structure and principles of the cooperatives emphasized both economic equality – in terms of a comparatively ﬂat wage scale (at 1:3) – and strong rights of participation, with respect to the ‘‘political’’ exercise of voice in one-person, onevote general assemblies. Each cooperative, in fact, maintains a general assembly as well as an elected president, an appointed general manager, an elected social council (intended to counterbalance management’s focus on productivity with more worker-centered concerns), and a watchdog council concerned with accounting and integrity (Thomas & Logan, 1982; Whyte & Whyte, 1991). Economic equality at Mondrago´n is still a very real presence, although the wage ratio has widened beyond 1:7, with a cadre of high-level managers enjoying the beneﬁts of having their salaries pegged more closely to the larger market. Indeed, with this ‘‘loosening’’ of the internal wage structure in 1992, the market was used as the principal argument in favor of the change. To ensure successful recruitment of top-level managers, engineers, and analysts, it was asserted that the cooperatives had to tie a group of salaries to the market, allowing them to rise to at least about 75% of the ongoing salary rates. At the same time, the actual structuring of work, especially in the largest industrial cooperatives of the Mondrago´n system, was usually far from enlightened. Until the late 1980s, for instance, relations on the shop ﬂoor of FAGOR, the giant home appliance manufacturer, could fairly be described as Taylorist (see, e.g., Kasmir, 1996). In fact, assembly line technologies prevailed in some of the largest industrial cooperatives until about 1990. While many of the cooperatives sent new members, or ‘‘socios,’’ to training that included the philosophy of cooperativism, many were primarily concerned with the ‘‘technical formation’’ of the new employee. For many years, large cooperatives have held preparatory groups, or ‘‘charlas,’’ in advance of general assemblies, but over time, these meetings became somewhat ritualized in response to increasingly

186

GEORGE CHENEY

scripted annual assembly sessions. As I have observed elsewhere, it is ironic indeed that a system so imbued with democratic principles, and situated in a culture where political voice is regularly exercised in a number of ways, does not take fuller advantage of local wisdom in organizing participation (Cheney, 1999, 2002). In fact, I discovered that by the mid-1990s, local wisdom was frequently subordinated to prevailing models of total quality management, lean production, and just-in-time inventory systems as they were imported from the U.S., Japan, and Germany. These other economies and their dominant corporations took on an aura of supremacy for policymakers in the MCC, especially, such that the leaders seemed to slavishly follow fashionable models. To be sure, much of the change at Mondrago´n may be attributed to growth, bureaucratization, and institutionalization, especially as the cooperatives (that is, MCC) went from being minor players in the Basque and Spanish economies to ﬁnancial powerhouses. This is true even for the breakaway ULMA Group. Also, there is a certain hard reality to globalization for Mondrago´n as European markets consolidated and opened up. However, it is also important to examine the pressures brought to bear through the globalization of consumerism as well as the choices top managers make to be ‘‘more like other multinational corporations.’’ In these two arenas of goals, pressures and responses, we ﬁnd the heart of the analysis of today’s cooperativism. Let us ﬁrst consider the role of consumerism and consumer culture in the Mondrago´n’s case and then take a closer look at managerial assumptions and regimes.

THE TRANSFORMATION OF THE CITIZEN TO THE CONSUMER ‘‘Consumerism was the twentieth century’s winning ‘ism’’’ (Gopnik, 1997, p. 80). With this bold and deliberately ironic statement, a writer for The New Yorker magazine made the point that whatever other movements rose and fell in the past 100 years or so, the force of consumerism is indeed with us. Indeed, it is now so common to speak of ‘‘consumer society,’’ to substitute the term ‘‘consumer’’ for ‘‘citizen,’’ and to speak of nations like China as ‘‘emerging markets of consumer power,’’ that in everyday talk consumption has ceased to be an object of attention. It is as taken for granted as breathing. For people in industrialized societies, regardless of their position on the political spectrum, consumption is not just a means to live but a way of life (Miles, 1998; Schor, 2004). In fact, increased consumption has become so intertwined with work and work goals that many people willingly take on

Democracy at Work within the Market: Reconsidering the Potential

187

extra work hours and job stresses to leverage a higher level of lifestyle, thought to be necessary (Schor, 1992). And, nowhere is this truer than in the contemporary U.S., where the ‘‘gimme’’ form of consumerism has swamped the politically responsible form – at least so far (Cheney, 2005). Nevertheless, consumer culture is both multidimensional and wide-ranging: it includes within its reach examples of conspicuous consumption, the commodiﬁcation of aspects of society (such as values) that are not readily translated into the material or that would not have lent themselves easily to packaging in the past (Desmond, 1995). Consumerist society goes hand in glove with the broader marketization of institutions and sectors, ranging from education, health care, news production, and even religion (Fairclough, 1993). By ‘‘marketization,’’ I mean the tendency to use the ‘‘free market’’ as a common reference point for modeling not only business structures and relations but also nearly every other domain of society (Lair, Sullivan, & Cheney, 2005; Simpson, 2006). Running alongside this trend has been the triumph of marketing, reaching a peak in the 1990s. By contrast with the logics of production, sales, advertising, and public relations, marketing casts itself in a democratic ethos: discover what a person wants and give it to him (Cheney & Christensen, 2001). Even better, from the ﬁrm’s perspective, is to ‘‘help’’ the consumer to see what she may want and then provide exactly that. To the extent that marketing has infused not only all departments of the ﬁrm but also all sectors of society, it presumes to stand in for political participation in reimaging society as a supermarket. While there are many expressions of consumer ‘‘voice’’ – and some are indeed rebellious or counter-hegemonic (Gabriel & Lang, 1995) – the overall tendency of marketing is to co-opt, to devour even oppositional symbols in a manner not unlike what Marx foresaw as a dynamic of capitalism itself with its ever-widening capacity for commodiﬁcation. What is important about consumerism for the present analysis is how it transforms social relations, including those at work and in the wider public sphere. Social theorists du Gay and Salaman (1992) have described this often overlooked set of implications in their essay ‘‘The Cult(ure) of the Customer.’’ There they explain how the consumerist logic turns work relations into mere customer–supplier chains and presumes that what is marketed toward desire is necessarily best to produce. As a result, enormous pressure is dealt to the person in her role as an employee, while she may have the opportunity to exercise demand in her other principal role – that of shopper. Sitting at the top of the pyramid, if you will, is the consumer – or, at least that is the point of reference in practically all the operations of the ﬁrm.

188

GEORGE CHENEY

In fact, we witness these trends at Mondrago´n, and these are precisely the reasons that the scope of my study widened considerably through the 1990s. While I was originally concerned with the state of participative systems within the cooperatives, as they were undergoing market globalization, I gradually realized that I would need to look at the several ways in which consumer culture from ‘‘outside’’ and managerialism from ‘‘inside’’ were reshaping the nature of democracy within the cooperatives. Quite dramatically in the 1990s, members of the cooperatives began speaking of themselves more as ‘‘consumers’’ than as ‘‘citizens’’ in the public sphere beyond work and as ‘‘servants’’ of consumers while at work. As one ‘‘socio’’ put it to me: ‘‘When I’m at the mall, I’m king. But, when I’m at work, I serve the king.’’ This change in outlook both fueled and followed changes in managerial policy that elevated ‘‘customer satisfaction’’ as the preeminent goal of the cooperatives and consequently made marketing a concern of all parts of the organization. In fact in 1997, one of the surviving founders of the cooperatives, Alfonso Gorron˜ogoitia, expressed concern to me for what he saw as a general deterioration of the culture of sacriﬁce, participation, and solidarity, with people becoming more self-interested and self-indulgent. Within the cooperatives, a type of market realism was regularly invoked by the early 1990s, where worker ‘‘participation’’ became reinterpreted vis-a`-vis market globalization and detached from most of its political content. General assemblies and governing organs were less emphasized. Managers began speaking of the need for all employees to be market-andconsumer-conscious, and this was combined with the importation of some just-in-time techniques for warehousing as well as the widespread adoption of Kaizen, or Constant Improvement (Hobekuntza, in Euskara, the Basque language). What is extraordinary in the case of the Mondrago´n cooperatives is that this shift was propelled in the late 1980s and early 1990s in part by two of the surviving founders – members of a group of ﬁve engineers who, together with a Basque priest, founded the MCC in 1956. One of them, Jesu´s Larran˜aga, told me in 1994 that: We have to participate for reasons of competitiveness and the expansion of the market. And, the kind of participation we need is not something up in the clouds, dealing with abstract issues, but something continuous and concentrated in one’s job.

In this instance, the same value-based term was being involved, with a signiﬁcantly different meaning. In effect, the political dimension of cooperative practice was being entirely subordinated to the concerns of productivity and the global customer. More than perhaps any other change

Democracy at Work within the Market: Reconsidering the Potential

189

that I observed at Mondrago´n, this shows how entangled are ‘‘external’’ and ‘‘internal’’ causes for transformation. The Mondrago´n case helps clarify the relations of outside forces to internal conﬁgurations of participation and other programs – to consider how presumably external pressures affected internal commitments to core cooperative values. As my ﬁeld research progressed, especially during a six-month stay at Mondrago´n in 1994, I became playfully provocative in my interviews. For example, following their common claim that ‘‘The market demands that we do this,’’ I would frequently ask managers, ‘‘So, what is the market?’’ ‘‘How do we know the market when we see it?’’ ‘‘What indicators best represent the ‘voice’ of the market?’’ And, ‘‘Aren’t you a big player in the market?’’ These questions would often yield puzzled looks. Occasionally the queries would elicit acknowledgments of the cooperatives’ roles in constructing and reifying the market, granting it a type of supra-agency beyond its contributors. For example, cooperatives chose to focus on industries and services (such as auto parts) that necessarily placed the ﬁrms in the most internationalized and volatile parts of the global marketplace. In the case of MCC, in particular, the market is reinforced through predictable choices of products and services, by the unreﬂective adoption of common managerial regimes, through the soft-pedaling of social values in the face of neoliberal economic globalization, and by simply surrendering the idea of ‘‘doing business differently.’’ One of the key questions about market globalization in terms of its impact on a particular value-based organization (or set of organizations) is how deeply it affects the way business is conducted. There are basically three possibilities, or logics, to the effects of globalization in this regard: (1) The terms of the multinational corporate world and consumerist society are adopted, but practices within the organizations remain essentially unchanged. In this way, the discourses of market globalization, and their accompanying managerial systems/practices, simply ‘‘wash over’’ a particular organization or industry. Thus, an organization may ‘‘talk’’ lean production while maintaining its own traditional practices of participative decision-making. (2) The intermediate level of inﬂuence ﬁnds that managerial and consumer practices from without are appropriated and adapted in distinctively regional, local, or even organization-specific ways. In the second case, as in fact we ﬁnd in many parts of the world, the managerial discourse of a particular system or regime is adopted by the organization, but the policymakers take a ‘‘cafeteria’’ approach toward the implementation of only those parts of the system that are understood, preferred, or feasible (see Appelbaum & Batt, 1994). (3) At the level of greatest outside impact, the organization or organizations are fundamentally transformed by

190

GEORGE CHENEY

external inﬂuences. In this case, we would ﬁnd that a value-based organization looks substantially different from its earlier form, especially at the level of work decisions and other practices (Cheney, 2004). Ascertaining how deeply globalization affects organizations is what I mean by ‘‘getting concrete’’ about the meanings of market globalization. This tripartite scheme is relevant to all types of organizations, including worker cooperatives.

LESSONS FROM THE MONDRAGO´N COOPERATIVE EXPERIENCE I have already argued that the international cooperative movement is both less and more than the term suggests. As we have seen, the case of the Mondrago´n worker cooperatives is signiﬁcant not only because of its size, or even its longevity, but more importantly because of what may be inferred from its development over time. In fact, this case is so signiﬁcant in terms of its multiple features that a number of lessons can be drawn, which apply not only to literal worker cooperatives but also to a wider array of value-based democratic organizations. What began for me as an investigation into particularism – in the sense of understanding Mondrago´n’s unique or at least distinctive aspects – gradually assumed a broader applicability. Indeed, there are speciﬁc historical, cultural, economic, and organizational reasons for the cooperatives’ success. However, even when taken as a whole these do not fully account for either the development of the cooperatives into the forms we ﬁnd today or for the ways that these cooperative corporations embody struggles that are seen far beyond their quarters. In this way, the Mondrago´n experience speaks to enduring issues in theory and practice with respect to organizations – in particular, the capacity for value maintenance over time (compare Rothschild-Whitt, 1979; Satow, 1975; Weber, 1978). In short, we must conclude that Mondrago´n both is and is not special, and from its 50-year history we can derive the following practical themes for other organizations that would hold to core social values. Each theme is treated brieﬂy below. Fostering a Consensus around Certain Values Weber’s (1978) famously underdeveloped ‘‘fourth type’’ of authority and organizational rationality is presumed by some scholars to have been a decidedly value-based ideal type (Rothschild-Whitt, 1979; Satow, 1975).

Democracy at Work within the Market: Reconsidering the Potential

191

Whether this interpretive projection adequately represents Weber’s view remains in question. But what is certain is his ambivalence toward modern forms of rationality that would come to shroud or push out of view altogether genuine foundational social commitments. Maintaining core value commitments is difﬁcult not only because times change or even that the organization changes but also because the people change. Thus, the value commitments of the founding generation of cooperative employee-members may come to be seen as outdated by younger recruits. In fact, this is part of what has happened at Mondrago´n, where more individualistic conceptions of the career have arisen among 40-, 30-, and 20-somethings.

Maintaining a Simultaneously Open and Closed System For a cooperative or any organization to maintain its core values and practices, some self-protection is necessary. This much was clear from the experiences of the Owenite cooperatives and communities whose distinctiveness was lost once they became dependent upon outside capital and therefore absentee ownership. The zone of protection for a cooperative can include buffering from the vicissitudes of the market (as in the case of an internal banking system); a training and educational system that runs at least somewhat counter to popular, outside views of how to do business; or, an autonomy that allows the cooperative to shift its production and service emphases so as to preserve itself. The exact type of engagement of the market should be conscious, grounded in common experience, and with key goals in mind (as is discussed further below).

Seeking (and Finding) Leadership and Inspiration Inspired, even charismatic leadership is perhaps essential to the founding and maintenance of value-based organizations such as cooperatives. While I would have stated this position a bit less strongly during my studies at Mondrago´n in the 1990s, I would now take a more certain position that reﬂects the empirical realities of cases throughout the world in the arena of ‘‘alternative organizations.’’ Very few succeed with a purely group form of leadership, as can be seen in the record of socially responsible businesses and social movement organizations, in addition to cooperatives. The Body Shop, Ben and Jerry’s Ice Cream, and Patagonia are a few examples from the business world. The Grameen (or ‘‘People’s’’) Bank of Bangladesh is an

192

GEORGE CHENEY

example from the domain of NGOs and social movements. Some small feminist and radical egalitarian collectives do attest otherwise, but these seldom extend beyond a relatively local reach (Iannello, 1992). The problem harks back to Weber’s (1978) discussion of the ‘‘routinization of charisma’’: how best to preserve charisma institutionally after the founder is gone (see also Grant, 2004; Roper & Cheney, 2005). The option of simply watching and waiting for the new organizational messiah, as some of the older socios are still doing at Mondrago´n, is not a hopeful sign for the revival of democratic practice.

Conserving a Common Mission For cooperative or any other type of alternative organization to keep going with a degree of organizational integrity, values must be revisited – in terms of the labels used, the practices represented, and the preferences of the majority or the whole. For example, an ambiguous but emotionally charged value such as ‘‘solidarity’’ should be considered periodically for what it means in terms of current policies and practices. As the contemporary situation is compared with the founding context, members should ask, how much deviation from ‘‘the path’’ can be allowed? In part this depends on what the organizational members see as the constitutive or defining values of the ﬁrm and therefore how much adaptation or modiﬁcation is deemed permissible. The very understanding of democracy within the ﬁrm is at issue, as is discussed below.

Recognizing the Interdependence of the Social and the Economic This issue has already been addressed at various points throughout the essay. However, I would add here that very few managers in any type of organizations appreciate fully just how much these two sets of motivations work together. The typical position on democratic and enlightened human resource practices in tough ﬁnancial times is ‘‘not now.’’ Ironically, a common position on the social side of enterprise during ﬂush times is ‘‘not needed.’’ The fact is that socio-symbolic aspects of employment continue to be relegated to a lesser role than the one deserved, despite the accumulated evidence that employees crave genuine autonomy and opportunities for participation in the direction of their own work and the ﬁrm’s (cf. Freeman & Rogens, 1999; Weisbord, 1991; Wilkinson & Willmott, 1995).

Democracy at Work within the Market: Reconsidering the Potential

193

Enacting Democracy as a Process Throughout this essay, we have considered a variety of understandings and forms of ‘‘participation.’’ These are as myriad as interpretations of democracy itself. And, no one can presume an omniscient position to decide what ‘‘real’’ democracy or employee participation is. What we can do is urge any committed organization to engage employee-members in a systematic consideration of their own system of participation (Mygind, 1992), to recognize the practical limitations associated with any system, and to revisit the functioning of participative programs in light of their own goals and new information (cf. Stohl, 1993). Being Mindful of the Market I mean this caution not only in the sense of awareness of market forces but also such that the organization consciously adopts a point of view with respect to its market. No value-based organization should consider itself as completely passive in the face of market forces, if only because decisions about product and service specializations or niches may allow the ﬁrm more ﬂexibility than initially presumed. To surrender completely to The Market is in effect to renounce innovation, creativity, and the possibilities for ‘‘localized’’ difference. This strategy also makes for a self-fulﬁlling prophecy. Such exercises in autonomy, in turn, may allow the organization to establish and maintain the very types of participatory practices that it ﬁnds ideal and consistent with core value commitments. In addition, cooperatives and other so-called alternative organizations can expand their own infrastructure, indirectly, by forming partnerships with other related enterprises and ﬁnding allies throughout the community. This is one of the stated goals of the International Cooperative Alliance (1996), yet it is the exception rather than the rule in practice. By actively building networks of collaboration, cooperatives can expand their knowledge base, ﬁnd creative solutions to common problems, and encourage their own employee-members to participate more widely in what might someday be fairly deemed a social movement.

THE DECLINE AND POSSIBLE REVIVAL OF ORGANIZED LABOR In the philosophy, lore, and practices of ‘‘cooperativism,’’ organized labor is usually cast in an outsider’s role because of its reliance on an adversarial

194

GEORGE CHENEY

rather than a unitary form of democracy at work. That is, worker-ownedand-managed cooperatives are cast as beyond the division of labor and capital that characterizes the typical capitalist ﬁrm, and union activity is therefore seen as antiquated and superﬂuous. Experiences in some of the largest worker cooperatives, however, tell us that this often unquestioned assessment is mistaken or at least premature, for three reasons: (1) very few cooperatives, and no large ones (that is, with several thousand or more employees) achieve anything close to economic equality in terms of salaries and beneﬁts; (2) growth and bureaucratization (almost) inevitably lead toward great disparities in decisional power between upper-level managers (whatever they may be called) and rank-and-ﬁle workers; and (3) large cooperative systems tend to develop quasi-unions even where organized labor is proscribed or seen as irrelevant. All three of these reasons are easily observed in the Mondrago´n experience. I have already mentioned the gradual widening of the wage scale at Mondrago´n, away from the rather ﬂat and fairly radical 1:3 ratio and the linking of select upper-level managers’ salaries to the regional employment market. Second, processes of decision-making have become more controlled by managers, directly, and by systems or regimes of production imported from the outside, indirectly. During my time at Mondrago´n, this change was evident in three ways: (1) More managers without speciﬁc cooperative experience were being hired from outside. (2) New systems of production, including certain types of self-directed work teams, were being structured and implemented in a fairly top-down manner. (3) The ‘‘political’’ or policy-oriented aspects of employee input into decisions were being minimized, both in everyday talk and in the ritualization of general assemblies and preparatory discussions, as discussed earlier (see Cheney, 1999, 2002). Combine these trends with the imitation of non-cooperative multinational ﬁrms, and you have a situation that is ripe for the argument that unions should hold a place at the table. In fact, union activity has surfaced at various points in Mondrago´n’s history. Two periods are noteworthy here: the 1974 strike by representatives of Basque and Spanish labor unions who happened to work for the cooperatives; and the early 1990s to the present, where informal ‘‘cooperative groups’’ (as they call themselves in Euskara) have been asserting themselves as the ‘‘true’’ keepers of cooperative values. Because of the harsh reaction by management to the 1974 strike, in which a number of employees were ﬁred and later rehired, today’s unions must retain an informal status. That is, they are outlawed by MCC, just as they are ofﬁcially seen as irrelevant in an organizational and business arrangement where capital and labor are presumably joined (Kasmir, 1996).

Democracy at Work within the Market: Reconsidering the Potential

195

In the mid-1990s, I found that these groups continued to be strongest within the massive industrial cooperatives, such as FAGOR. From our discussions, I could see that their perspective was both legitimate and romantic. They objected to the increasing ‘‘managerialization’’ of MCC, whereby hierarchical distance greatly increased, and they were especially critical of the growing demands for production that fell upon the shoulders of employees as a result of accelerated work processes and diminished worker inﬂuence at the level of policy. At the same time, there was a clear yearning for a simpler time, when the cooperatives were not industrial and ﬁnancial giants, when relations throughout the cooperatives were more intimate, when the pace of life and work was slower, and when there were fewer outside economic inﬂuences. The containment and, in some cases, repression of organized labor in nations ranging from the U.S. to Indonesia is well documented elsewhere (e.g., Moody, 1997; Moore, 1996; Sustar, 2004). Deindustrialization, outright union busting, reductions in beneﬁts, and offshoring have all contributed to the weakening of labor’s position over the past three decades. In the U.S., labor’s decline in sheer numbers as well as popular inﬂuence is often traced to the Reagan Administration’s corporatist reaction to the strike of air trafﬁc controllers in 1981. After that, unions in the U.S. were considerably less visible, vocal, and potent politically. Even so, the decline of unions in the U.S. and Western Europe has a broader cultural basis that can be traced back even to the apparent golden era of the 1930–1950s. Here I refer to a set of images of labor vis-a`-vis individual ambitions, consumer democracy, and folk ideas of bureaucracy and effectiveness. Without going into a great deal of depth on this issue here, because of space limitations, I would say that the ‘‘positioning’’ or ‘‘framing’’ of organized labor as rigid, corrupt, and backward-looking has had a great deal to do with its decline (Martin, 2004). While these images or stereotypes can be supported by some empirical examples, they have risen to the level of a monolithic picture that is unfair and itself unresponsive to differences across cases. The academic and popular literatures on cooperatives seldom speak about organized labor, in large part because cooperativism styles itself as a ‘‘third way’’ between unbridled capitalist enterprise, where equality or wealth is held by a few owners/managers, and state-centered or state-sponsored socialism. In fact, the founder of the Mondrago´n cooperatives, Fr. Jose´ Marı´ a Arizmendiarrieta, saw the launch of the ﬁrst cooperative in just these terms. From such a standpoint, the often-adversarial posture of unions toward management would seem inappropriate or even inconceivable. However, organizations change and, as we have seen, their ‘‘success’’ in terms of growth and consolidation of power, not to mention impersonality

196

GEORGE CHENEY

of social relations, may provide fertile ground as well as principled justiﬁcation for labor union advocacy. Organized labor can respond to a real or perceived ‘‘value deﬁcit’’ in organizations, where professed standards of equality, solidarity, and democracy are no longer the true measure of practice and progress. In the case of Mondrago´n, the cooperative groups’ insistence on multiple dimensions of solidarity – and not only at the intercooperative or corporate level – represents a legitimate stance toward the loss of ‘‘value’’ in what has come to be called in more conventional ﬁnancial terms ‘‘a value-added enterprise’’ (Cheney, 1997). For all these reasons, both general and speciﬁc, it is no longer appropriate to keep the literatures on worker cooperatives and labor unions separate from one another as if they were different foods on a child’s plate. The more fundamental questions about democracy at work – in the full social, political, and economic senses – demand a case-by-case approach to the evaluation of any particular business structure, set of labor relations, or connection to the larger community. After all, many unions, in the U.S. and elsewhere, are now taking a more expansive approach to organizing, whereby engagement is seen on multiple levels, from the local to the community to the nation to the crossing of borders (Fantasia & Voss, 2004).

OTHER FORMS OF DEMOCRATIC ORGANIZING TODAY When we speak of ‘‘employee participation’’ or of its stronger form, ‘‘workplace democracy,’’ we must look both to explicitly designed cooperatives and to other forms of organizing that include networked and networks of organizations (Cheney, Christensen, Zorn, & Ganesh, 2004). Such organizations are not always grounded expressly in democratic values, yet many of them embody informal democratic practice. In addition, most of them testify to the capacity of new communication technologies for enabling freer forms of autonomy and communication ﬂow just as they sometimes allow for increased surveillance and centralization. In fact, the Internet is now a (perhaps the) key sphere for observing tensions between centralization and autocracy on the one hand and decentralization and radical democratization, on the other, because of the Internet’s capacity to reduce hierarchical distance and yet allow for surveillance (Harrison & Falvey, 2001). A now-famous case of organizing along these lines is the ‘‘multicephalous’’ (or multi-headed) movement that successfully derailed the MAI (Multilateral Agreement on Investment) in 1998. This movement is

Democracy at Work within the Market: Reconsidering the Potential

197

important both because of its organizational features and its market (or anti-market) focus (Tilly, 2004). Remarkably, this multilateral organizing produced few visible leaders, and it proceeded in a creative and expansive manner. This is but one prominent example.

CONCLUSION: COMMITMENTS, OBSTACLES, CAPITULATION, AND HOPES While I spent six months at Mondrago´n in 1994, many visitors from around the world came and went. Most of them, of course, had time only for an ofﬁcial tour and a couple of strolls around town and the countryside. Not surprisingly, they all had one question in common: ‘‘Has Mondrago´n sold out?’’ They did not want an equivocal answer, but that is precisely what I felt I had to give them. My own position on the cooperatives is that they have struggled to maintain core values of equality, solidarity, and participation in the face of their own ambitions, rising consumerism, and the realities of market globalization. Many employees, including a whole segment of managers, remained inspired and driven by a commitment of doing business differently. With respect to the impact on employee participation, perhaps the single most important factor has been the way dominant regimes of managerialism and productivity have been imported and applied: for example, with the decreased emphasis on social values and policy-oriented participation. This is consistent with Irene Goll’s (1991) provocative analysis that the ways that outside pressures are perceived and handled by organizations are as important as the empirical realities of those pressures themselves (cf. Kochan, Katz, & McKersie, 1994). Indeed, this is exactly what economist Melissa Moye (1993) found as her own analysis broadened to incorporate the speciﬁc strategic choices being made by MCC policymakers. Today, however, I would place an even greater emphasis on ideology than I did in the ﬁnal formulations of my book (Cheney, 1999, 2002). I would stress both the growing inﬂuence of consumerism in the culture and of neoliberal economic thought within and beyond policy circles of government and business. Both of these trends work against the multidimensional understanding of social solidarity that was a hallmark of the Mondrago´n cooperatives. That is, I had identiﬁed at least seven levels and applications of solidarity there, ranging from the interpersonal type (e.g., helping out a coworker in distress) to the cooperatives’ bonds with the communities.2 Between those micro and macro levels, we ﬁnd relations within and between the ﬁrms.

198

GEORGE CHENEY

The original value base of the cooperatives, aligned with Catholic social justice teaching and associated with the quiet persistence of a priest, now is challenged both internally and externally. Like the Church itself, the statement of ten principles of the Mondrago´n cooperatives, which includes a call for societal transformation, now is cast as antiquated and unrealistic by most of the leaders/managers of the cooperatives. This change is somewhat less true for the ULMA Group, where more of the corporate identity is derived from a stance of ‘‘being different’’ than its larger and dominant neighbor, MCC. Returning to employee participation speciﬁcally, what can be fairly observed about the mix of old and new methods at Mondrago´n? Here I ﬁnd Paul Bernstein’s (1976) three criteria for assessing employee involvement to be particularly useful, goading us to get as concrete as possible in our judgments. Bernstein argues that any claim to participation ought to be examined along the lines of: (1) the degree of control held by employees, (2) the types and range of issues over which that control is exerted, and (3) the level of the hierarchy which can be inﬂuenced by employees. At Mondrago´n, there is no question of a gradual decline in the power of participation according to these three criteria. And, this has especially been true since 1992, when the ﬁrst big step in European economic integration occurred. In fact, that year was used as a justiﬁcation for the consolidation of decision-making power at the top of MCC’s hierarchy well before integration was a fact: it was explained in terms of the need ‘‘to speak with one voice to the market’’ (as one of the founders put it to me; see Ormaechea, 1991). Again, we cannot separate the internal employee participation programs of Mondrago´n from the structure and positioning of the ﬁrm(s) with respect to the market. The capitulation of a great deal of local knowledge, along with strategic decisions to endorse conventional market strategy and prevailing managerial regimes, suggest to me a much more passive posture than was actually necessary for continuing ﬁnancial success (see also Taylor, 1994). In addition, the distancing of the cooperatives, including the giant consumer cooperative, EROSKI, from the images and commitments of ‘‘cooperativism,’’ represents a lost opportunity to exert moral leadership. The extent to which such adaptations are inevitable given the dominance of the corporate form in the global economy is an important question for debate (Ritz & POCLAD, 2001). All that said, the Mondrago´n system, or systems, is still far more socially inspired and far more vibrant at the level of employee relations than most capitalist ﬁrms. Moreover, I have never visited an organization outside of those with explicit social-movement objectives where the discussion of

Democracy at Work within the Market: Reconsidering the Potential

199

values and principles is so prominent. How long these observations will apply is in question, however. In the global market, Mondrago´n remains a beacon of creativity and hope within the cooperative ‘‘movement,’’ alongside feminist collectives, local economies, alternative trade organizations, decentralized action networks, multi-union ‘‘corporate campaigns,’’ people’s alliances across class and national boundaries, and resistance by the poor against the privatization of water. As breaks continue to develop in what had appeared until 1999 to be an unstoppable tide of ‘‘free trade,’’ these various forms of social and economic experimentation should be better understood, nurtured, and sustained. The contest over the meanings and practices of democracy vis-a`-vis the market will continue in multiple ways and in multiple sites.

NOTES 1. Conversely, of course, many employing organizations (including some employee stock ownership programs, or ESOPs, as well as some capitalist enterprises – either publicly or privately held) offer little in the way of shared ownership yet may provide certain forums for what might be called ‘‘social’’ participation – participation in some range of decisions about work. These vehicles come in the form of programs such as participative management (popular in the 1940–1960s, and including socio-technical approaches), quality circles (popular in the 1970–1980s), and self-directed work teams (which have been widely used since the late 1980s) (see the overviews in Cheney, Straub, Speirs-Glebe, Stohl, DeGooyer, Whalen, GarvinDoxas, & Carlone, 1998; Seibold & Shea, 2001; Stohl, 1987). In the case of teams, the actual work is restructured, where quality circles and participative management usually involve ‘‘councils’’ or committees that are layered over the work (if you will). These are groups with representation from various departments and/or levels of the ﬁrm, whose typical goal is to improve productivity. Participation is thus a vehicle rather than a goal in itself. In large organizations of any sort, rarely does employees voice rise to the level of political power, in terms of collective possibility or actuality on the part of workers/ employees/associates. Thus, we usually ﬁnd substantial limitations on employee voice through these traditional programs, regardless of their assertions to the contrary (see, e.g., McLagan & Nel, 1995). In fact, there is strong empirical evidence in many ﬁrms to show that programs such as quality circles have been used to suppress employee participation, and especially to undermine union organizing (Grenier, 1988). 2. These meanings refer to: interpersonal camaraderie and ﬁnancial support; the ‘‘ﬂat’’ wage scale; ‘‘intercooperation among individual co-ops,’’ connection to the local community; commitment to Basque identity; and identiﬁcation with cooperativism and its mission of economic and social justice (see Cheney, 1997).

200

GEORGE CHENEY

ACKNOWLEDGMENTS I would like to thank everyone at Mondrago´n who made this research possible, the universities of Colorado and Montana for research support, and Dana Cloud and Dean Ritz for enriching conversations on democracy and the market in the past few years.

REFERENCES Adams, F. T., & Hansen, G. B. (1992). Putting democracy to work. San Francisco: BerrettKoehler. Almond, G. (1991). Capitalism and democracy. Political Science and Politics, 24, 467–474. Appelbaum, E., & Batt, R. (1994). The new American workplace: Transforming work systems in the United States. Ithaca and London: Cornell University Press. Azurmendi, J. (1992). El hombre cooperative: Pensamiento de Arizmendiarrieta. Mondrago´n, Basque Country, Spain: Otalora. Bernstein, P. (1976). Necessary elements for effective worker participation in decision making. Journal of Economic Issues, 10, 490–522. Bradley, K., & Gelb, A. (1983). Cooperation at work: The Mondrago´n experience. London: Heinemann. Cheney, G. (1995). Democracy in the workplace: Theory and practice from the perspective of communication. Journal of Applied Communication Research, 23, 167–200. Cheney, G. (1997). The many meanings of ‘‘solidarity’’: The negotiation of values in the Mondrago´n worker-cooperative complex under pressure. In: B. D. Sypher (Ed.), Case studies in organizational communication, 2 (pp. 68–83). New York: Guilford. Cheney, G. (1998). ‘‘It’s the Economy, Stupid!’’ A rhetorical-critical perspective on today’s market. Australian Journal of Communication, 25, 24–44. Cheney, G. (1999). Values at work. Ithaca and London: Cornell University Press. Cheney, G. (2001). Forms of connection and severance in and around the Mondrago´n workercooperative complex. In: G. Shepherd & E. Rothenbuhler (Eds), Communication and community (pp. 135–156). Mahwah, NJ: Erlbaum. Cheney, G. (2002). Values at work, rev. and updated ed. Ithaca and London: Cornell University Press. Cheney, G. (2004). Arguing about the place of values and ethics in market-oriented discourses of today. In: P. Sullivan & S. Goldzwig (Eds), New approaches to rhetoric (pp. 61–88). Thousand Oaks, CA: Sage. Cheney, G. (2005). The United Consumers of America. Or, is there a citizen in the house? Second annual Lecture in the Humanities. University of Utah, Salt Lake City. Cheney, G., & Christensen, L. T. (2001). Organizational identity: Linkages between internal and external communication. In: F. M. Jablin & L. L. Putnam (Eds), The new handbook of organizational communication (pp. 231–269). Thousand Oaks, CA: Sage. Cheney, G., Christensen, L. T., Zorn, T. E., Jr., & Ganesh, S. (2004). Organizational communication in an age of globalization: Issues, reflections, practices. Prospect Heights, IL: Waveland Press.

Democracy at Work within the Market: Reconsidering the Potential

201

Cheney, G., Straub, J., Speirs-Glebe, L., Stohl, C., DeGooyer, D., Whalen, S., Garvin-Doxas, K., & Carlone, D. (1998). Democracy and communication at work: A multi-disciplinary review. In: M. Roloff (Ed.), Communication yearbook, 21 (pp. 35–91). Thousand Oaks, CA: Sage. Ciulla, J. (2000). The working life: The promise and betrayal of modern work. New York: Three Rivers Press. Cloud, D. (2005). Fighting words: Labor and the limits of communication at Staley, 1993 to 1996. Management communication quarterly, 18, 509–542. Corcoran, P. (1983). The limits of democratic theory. In: G. Duncan (Ed.), Democratic theory and practice (pp. 13–24). Cambridge: Cambridge University Press. Cornforth, C., Thomas, A., Lewis, J., & Spear, R. (1988). Developing successful worker cooperatives. London: Sage. Dachler, H. P., & Wilpert, B. (1978). Conceptual dimensions and boundaries of participation in organizations: A critical evaluation. Administrative Science Quarterly, 23, 1–39. Dahl, R. (1985). A preface to economic democracy. Berkeley: University of California Press. Dauncey, G. (1988). After the crash: The emergence of the rainbow economy. Rendlesham, UK: Green Print. Desmond, J. (1995). Reclaiming the subject: Decommodifying market knowledge. Journal of Marketing Management, 11, 721–746. du Gay, P., & Salaman, G. (1992). The cult(ure) of the customer. Journal of Management Studies, 29, 615–633. Ellerman, D. (1992). Property and contract in economics: The case for economic democracy. Cambridge, MA: Blackwell. Fairclough, N. (1993). Critical discourse analysis and the marketization of public discourse: The universities. Discourse and Society, 4, 133–168. Fantasia, R., & Voss, K. (2004). Hard work: Remaking the American labor movement. Berkeley: University of California Press. Fowler, B. (1947). The co-operative challenge. Boston: Little, Brown & Company. Friedman, T. L. (1999). The lexus and the olive tree. New York: Farrar, Straus, & Giroux. Gabriel, Y., & Lang, T. (1995). The unmanageable consumer. London: Sage. Gopnik, A. (1997). Trouble at the tower. The New Yorker, 4 August, 80. Goll, I. (1991). Environment, corporate ideology, and employee involvement programs. Industrial Relations, 30, 138–149. Grant, S. (2004). Narrating the body shop: A story about corporate identity. Unpublished dissertation. University of Waikato, Hamilton, New Zealand. Greenberg, E. S. (1986). Workplace democracy: The political effects of participation. Ithaca: Cornell University Press. Greenwood, D., & Gonza´lez, J. L. (1989). Culturas de FAGOR. San Sebastia´n, Basque Country, Spain: Editorial Txertoa. Grenier, G. (1988). Inhuman relations: Quality circles and anti-unionism in American industry. Philadelphia: Temple University Press. Halal, W. (1996). The new management: Democracy and enterprise are transforming organizations. San Francisco: Berrett-Koehler. Harris, Z. (1997). The transformation of capitalist society. Lanham, MD: Rowman & Littleﬁeld. Harrison, T. M., & Falvey, L. (2001). Democracy and new communication technologies. Communication yearbook, 25, 1–43.

202

GEORGE CHENEY

Heller, F., Pusic´, E., Strauss, G., & Wilpert, B. (1998). Organizational participation: Myth and reality. Oxford: Oxford University Press. Iannello, K. P. (1992). Decisions without hierarchy: Feminist interventions in organizational theory and practice. New York: Routledge. International Cooperative Alliance. (1996). Statement of cooperative identity. Madrid: ICA. Kasmir, S. (1996). The myth of Mondrago´n. Albany: SUNY Press. Kochan, T. A., Katz, H. C., & McKersie, R. B. (1994). The transformation of American industrial relations. Ithaca and London: Cornell University Press. Korten, D. C. (1995). When corporations rule the world. San Francisco: Berrett-Koehler. Krimerman, L., & Lindenfeld, F. (Eds) (1992). When workers decide: Workplace democracy takes root in North America. Philadelphia: New Society Publishers. Lair, D. J., Sullivan, K., & Cheney, G. (2005). Marketization and the recasting of the professional self: The rhetoric and ethics of personal branding. Management Communication Quarterly, 18, 307–343. Lichtenstein, N., & Harris, H. J. (Eds) (1993). Industrial democracy in America. Cambridge: Cambridge University Press. Lindenfeld, F., & Rothschild-Whitt, J. (1982). Workplace democracy and social change. Boston: Porter Sargent Publishers. Logue, J., Glass, R., Patton, W., Teodosio, A., & Thomas, K. (1998). Participatory employee ownership: How it works. Kent, OH: The Worker Ownership Institute. MacLeod, G. (1997). From Mondrago´n to America: Experiments in community economic development. Sydney, Nova Scotia: University College of Cape Breton Press. Martin, C. R. (2004). Framed! labor and the corporate media. Ithaca and London: ILR Press of Cornell University Press. Marx, K. (1978). In: [B. Fowkes (Trans.)], Capital (Vol. 1). New York: Vintage. Mathews, R. (1999). Jobs of our own: Building a stake-holder society. London: Central Books. McLagan, P., & Nel, C. (1995). The age of participation. San Francisco: Berrett-Koehler. Miles, S. (1998). Consumerism as a way of life. London: Sage. Moody, K. (1997). Workers in a lean world. London: Verso. Moore, T. S. (1996). The disposable work force. New York: de Gruyter. Morrison, R. (1991). We build the road as we travel. Philadelphia: New Society Publishers. Moye, M. A. (1993). Mondrago´n: Adapting co-operative structures to meet the needs of a changing environment. Economic and Industrial Democracy, 14, 251–276. Mygind, N. (1992). The choice of ownership structure. Economic and Industrial Democracy, 13, 359–399. Ormaechea, J. M. (1991). La experiencia cooporativa de Mondrago´n. Mondrago´n, The Basque Country, Spain: MCC. Pateman, C. (1970). Participation and democratic theory. Cambridge: Cambridge University Press. Ritz, D., POCLAD. (Eds). (2001). Defying corporations, deﬁning democracy: A book of history and strategy. New York: Apex Press and the Project on Corporations, Law and Democracy. Roper, J., & Cheney, G. (2005). The meanings of social entrepreneurship. Corporate Governance, 5, 95–104. Rosen, C., Klein, K. J., & Young, K. M. (1986). Employee ownership in America: The equity solution. Lexington, MA: Lexington Books. Rothschild, J., & Whitt, J. A. (1986). The cooperative workplace. Cambridge: Cambridge University Press.

Democracy at Work within the Market: Reconsidering the Potential

203

Rothschild-Whitt, J. (1979). The collectivist organization: An alternative to rationalbureaucratic models. American Sociological Review, 44, 509–527. Satow, R. L. (1975). Value-rational authority and professional organizations: Weber’s missing type. Administrative Science Quarterly, 20, 526–531. Schor, J. (1992). The overworked American. New York: Basic Books. Schor, J. (2004). Born to buy. New York: Scribners. Seibold, D. R., & Shea, B. C. (2001). Participation and decision making. In: F. M. Jablin & L. L. Putnam (Eds), The new handbook of organizational communication (pp. 664–703). Thousand Oaks, CA: Sage. Simpson, M. (2006). Marketisation of ageing: The case of retirement communities in New Zealand. Diss., The University of Waikato, Hamilton, New Zealand. Stohl, C. (1987). Bridging the parallel organization: A study of quality circle effectiveness. In: M. L. McLaughlin (Ed.), Communication yearbook 10. Thousand Oaks, CA: Sage. Stohl, C. (1993). European managers’ interpretations of participation. Human Communication Research, 20, 97–131. Stohl, C., & Cheney, G. (2001). Participatory processes/paradoxical practices. Management Communication Quarterly, 14, 349–407. Strauss, G. (1982). Workers participation in management: An international perspective. Research in Organizational Behavior, 4, 173–265. Sustar, L. (2004). The labor movement: State of emergency, signs of renewal. International Socialist Review, 34, 1–13. Taylor, P. L. (1994). The rhetorical construction of efﬁciency: Restructuring and industrial democracy in Mondrago´n, Spain. Sociological Forum, 9, 459–489. Thomas, H., & Logan, C. (1982). Mondrago´n: An economic analysis. London: Allen & Unwin. Tilly, C. (2004). Social movements: 1768–2004. Boulder, CO: Paradigm Books. Uribe-echebarria, A. (1981). Burocracı´a o participacio´n. San Sebastia´n, Basque Country, Spain: Editorial Txertoa. Vanek, J. (1977). The labor-managed economy. Ithaca, NY: Cornell University Press. Weber, M. (1978). Economy and Society, Vol. 2. Berkeley: University of California Press. Weisbord, M. (1991). Productive workplaces. San Francisco: Jossey-Bass. Whyte, W. F., & Whyte, K. K. (1991). Making Mondrago´n: The growth and dynamics of the worker-cooperative complex. Ithaca: ILR Press of Cornell University Press. Wilkinson, A., & Willmott, H. (Eds) (1995). Making quality critical. London: Routledge. Wren, D. (1987) (3rd ed.). The evolution of management thought. New York: Wiley. Zwerdling, D. (1978). Workplace democracy. New York: Harper & Row.

This page intentionally left blank

204

WORKPLACE DEMOCRACY COMES OF AGE: ECONOMIC STABILITY, GROWTH, AND WORKFORCE DIVERSITY Joan S. M. Meyers ABSTRACT Previous studies of worker-owned firms claimed that participatory democracy only thrives in small, homogeneous groups. This article focuses on a successful 30-year-old worker-owned company with more than 200 employees to explain how broad and deep democratic control, a large workforce, and member diversity are brought together. Drawing attention to its combination of training, infrastructure, compensation for management functions, and workplace culture, I argue that an equitable distribution of power and resources does not require hierarchical management, friendship relations, size limits, or member homogeneity. The article highlights the need for greater scholarly attention to worker ownership possibilities for the current multiracial and multicultural working population.

The problem of work under capitalist control has long been a primary object of critique in sociology as well as in popular movements for social change.

Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 205–237 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16008-2

205

206

JOAN S. M. MEYERS

Socialist revolution was proposed and enacted as one solution; others turned their attention to the possibility of worker control within capitalist economies. Experiments in collective ownership and control of the workplace increased signiﬁcantly in the U.S. in the 1960s and 1970s. The hope was that democratic workplaces – owned and managed by labor – could preserve human autonomy and creativity, create pockets of worker empowerment, and model an alternative to workplace alienation. Yet these democratic workplaces were mostly small companies serving a niche counterculture market. As the counterculture faded, so did its alternative organizations. Given their small size and lack of market share, democratic workplaces’ ability to empower a wide swath of labor was highly questionable. The decline in such organizations was echoed in scholarly studies. U.S. scholars of workplace democracy have refocused on degrees of worker participation and control in traditional corporations (i.e., Rothschild & Russell, 1986). While some studies ﬁnd a potential for workplace democracy, there is at least as much scholarly pessimism about the ability of such organizations to create jobs with true autonomy, creativity, and dignity. Although many of the worker-owned businesses described in the literature of the 1970s and 1980s no longer exist, it may be too soon to dismiss that earlier vision of the potential of worker ownership. This study asks if worker-owned democratic businesses could be viable economic institutions for an increasingly diverse1 U.S. workforce by examining a ﬁrm that is entirely owned and democratically controlled by its workers. This organization has grown from a largely homogeneous handful of people in the 1970s into a multicultural membership of over 200 today. Alternative organizations have been faulted for an inability to extend democratic processes equitably to large, diverse populations while also providing economic stability. Nevertheless, many worker-owned businesses that survived the cooperative boom and bust offer important insights. This single case study of ‘‘One World Natural Grocery’’2 advances an argument that economic growth, workplace diversity, and broad participatory democratic management are not incompatible.

POSSIBILITIES AND LIMITS OF WORKER OWNERSHIP: LITERATURE REVIEW Unlike an investor-owned ﬁrm, in which day-to-day and long-term planning are in the hands of a small group of owners or ownership proxies,

Workplace Democracy Comes of Age

207

democratic workplaces are planned and managed by their workers. All (or a majority of) elected representatives are members of and accountable to the organization. Decades of research show that worker ownership can either be partial – cooperatives that do not require employees to become members, or Employee Stock Option Plans (ESOPs) with minority or majority stock ownership by employees – or total, with employee membership required. Areas of decision making can run from shop ﬂoor arrangement and safety procedures, to decisions about the products made and marketing plans, to allocation of proﬁts and long-term capital projects (Bernstein, 1982). Thus, worker ownership can occur in different entities and great differences are possible in the very deployment of such control. Worker-owned cooperatives in North America and Western Europe in the 1960s and 1970s, however, followed a common pattern: they were almost entirely communitarian in their critique of the modern separation of working and non-working self, aimed at ‘‘right livelihood,’’3 were entirely worker owned and governed, prioritized the redistribution of control to the group as a whole rather than to a managerial strata, and were sized to allow face-to-face decision making. Rothschild-Whitt (1982), one of the most inﬂuential scholars of that cooperative movement, labeled this form ‘‘collectivist-democratic,’’ and her characterization of collectivist-democratic workplaces profoundly shaped both scholarly and popular conceptions of worker ownership. She found rewards to be more idealistic than ﬁnancial, and that authority in such organizations is derived not from ofﬁce or expert status but from the collective process of generating agreement. Such consensus creates accountability between members as to how work is conducted, removing the need for external management. Eschewing the formal rationality Weber (1946) described as typifying capitalist modernity and rejecting ﬁnancial incentives, bureaucratically conferred status, and formal rules required a shared morality to create consensus-based accountability. Rothschild-Whitt found that such a shared morality came from emotional bonds of friendship; Mansbridge (1980) similarly found friendship at the base of collectivist unitary democracy. Friendship as a basis for organizational control and consent, however, has two preconditions: each member must believe that the others already have similar enough ideals for this working friendship to develop, and organizations must be small enough for each member to develop emotional bonds with the others. Indeed, whether out of choice or external constraint, almost all alternative North American organizations previously studied had 40 or fewer members (e.g., Ferguson, 1991; Jackall, 1984; Mansbridge, 1980; Rothschild-Whitt, 1982). Yet, although the ﬁnancial rewards of such enterprises were not great,

208

JOAN S. M. MEYERS

they were particularly popular among a large middle-class population unimpressed by the homogenizing materialism of the 1950s and 1960s (Breines, 1989). Members of such collectivist democracies described their rewards as the experience of equality of power (Mansbridge, 1980), working toward a righteous and morally self-enhancing goal (Kleinman, 1996), and non-alienated relations to the product, the production process, one’s coworkers, and oneself (Ferguson, 1991). Nevertheless, these collectivist workplaces did not last. By the mid-1970s, researchers found upwards of 5,000 such organizations (Rothschild-Whitt, 1982), but at present there are less than a tenth of that number.4 Almost none of the businesses previously studied now exist. While the globalized market has been hard on small businesses in general (see Greider, 1997), there are particular reasons for the decline in alternative workplaces. On one hand, corporations have exploited niches developed by alternative organizations (natural foods, women-centered health services, youth-oriented publications, etc.); on the other, small cooperatives had difﬁculty attracting capitalization for expansion due to a mix of non-proﬁt-oriented goals, unfamiliar organizational structures, and member disconnection from local business communities. Some small collectives continue as boutique responses to elite demand or by serving a market too small to be worth corporate investment (i.e., delivering heating fuel in an isolated rural community). However, the scale of such companies means an even smaller number of workers are involved in these organizations in today’s labor market. There may be beautiful things about the small-scale enterprise (Schumacher, 1973), but there are also serious drawbacks to the lack of economic growth in the previous generation of democratic workplaces. Beyond the struggle to compete against big business entry into specialized markets, economic marginality creates problems of instability. People without outside support from investments or family wealth are less likely to feel that idealistic rewards can offset the costs of maintaining membership in an unstable cooperative. As Mansbridge (1980) found, cooperatives’ inability to fully support its members leads them to be comprised of more white and middle-class collectivists, as the latter group has resources to supplement earnings. Kleinman (1996) also found that while women – particularly those with higher-earning male partners – might initially ﬁnd involvement rewarding, their donations of time were less valued than those of men, who were seen to be sacriﬁcing greater dominant cultural and ﬁnancial rewards. The lack of organizational value placed on these women’s work – as expressed in low wages – eventually caused women to leave the organization.

Workplace Democracy Comes of Age

209

Such studies indicate that respect cannot fully substitute for material valuation of worth at work. Inability to provide substantial economic reward, however, does not fully explain the homogeneity widely reported in these earlier cooperatives. Relying on shared beliefs to create consensus meant that most collectives recruited new members out of preexisting friendship networks (RothschildWhitt, 1982). Because informal personal networks are deeply segregated by race/ethnicity and class, friendship-based cohesion tends to reproduce homogeneity over time. Thus, there are historical and material reasons for the homogeneity of those earlier democratic workplaces. Yet the correlation of member homogeneity with a refusal of workplace hierarchy bears further examination. The assumption of a link between member homogeneity and the functionality of non-hierarchical control is widely posited in studies of contemporary cooperatives (e.g., Hacker, 1989; Loe, 1999; Pencavel, 2001), yet for the most part, this linkage relies on observed correlations between homogeneity and a lack of hierarchy in contemporary settings, and transposition of theory based on collectives that were formed in different historical and material contexts. These recent studies assume the necessity of organizational homogeneity to the democratic practices observed there, rather than questioning this homogeneity as perhaps impeding a more profound practice of democracy across a broader spectrum of members. That is, social science posits workplace democracy as (perhaps problematically) reliant on, rather than merely historically correlated with, a set of shared interests, beliefs, and backgrounds among its members. This social science frame then has an effect on workplace democracies, particularly those where the workforce is highly educated and has read such studies. Framing their perception of their experience within social science assertions of required homogeneity, collective members themselves accept intergroup similarity – both demographic and outlook – as a necessary condition for their collectivist democracy (i.e., Ferguson, 1991) and do not question it when they look for ways to improve their organization’s ﬁnancial health. Member homogeneity, however, is not merely an attribute but also a process of minimizing the salience of some social characteristics and highlighting others shared by group members. Kasmir’s (1996) study of Mondrago´n argues that democratic workplaces may repress internal class diversity in order to create the basis of consent to be governed representatively by one’s presumptive peers. Freeman (1984) criticized the U.S. women’s ‘‘structureless’’ movement of 30 years ago for positing equal power among ‘‘sisters’’ and thus ignoring the classed differences in cultural capital

210

JOAN S. M. MEYERS

that reproduced social stratiﬁcation of power within collectivist movement organizations. Sirianni (1993) similarly ﬁnds that structurelessness in alternative non-proﬁt groups with collectivist orientations reproduces racial social stratiﬁcation. Identifying the informality of structureless organizations as critically problematic, Sirianni argues that formal, hierarchical, and representative democracy is better able to foster diversity and protect subaltern interests. Yet is the abandonment of direct democracy the best solution? Maintaining a truly democratic distribution of power is difﬁcult in a hierarchy, as theorists have noted since the rise of the modern hierarchical bureaucracy (Michels, 1959; Weber, 1946). Authority from top-down leadership is too easily naturalized as attached to persons, not positions. Accountability is too easily lost when authority is positionally stratiﬁed (as opposed to the temporary delegation of authority in ‘‘structureless’’ or informal organizations). In addition, as New Institutionalist theorists have argued, the omnipresence of hierarchy is not necessarily due to extraordinary efﬁciency or productiveness in hierarchical organizations, but rather to the ability of the organizational form to maintain a perception of tight coupling among efﬁciency, productivity, and hierarchy. DiMaggio and Powell’s (1983) analysis of mimetic institutional isomorphism delineates the ease with which power distributions settle into hegemonic patterns: challenging the boss of a labormanaged ﬁrm would also challenge the logic of having a boss in the ﬁrst place. In short, organizational forms have their own structural and cultural momentum, and the history of hierarchical bureaucracy seems too resistant to meet the goals of broad empowerment. Nevertheless, the attention to the advantages of the formality of hierarchical bureaucracy is astute. Informal organizational structures may provide for greater ﬂexibility and ease of entry, but they have very poor mechanisms for accountability beyond those invoked through friendship or what Barrett and McIntosh (1982) call ‘‘familialism.’’ Freeman’s (1984) attack on structurelessness in the democratic social movements linked to the workplace democracies of the 1960s and 1970s was not, however, merely a call for accountability in an era of political organizing that had swung far from the rigidly bureaucratic model of the Communist-dominated Old Left, but also a call to combine formality with broad empowerment. While past social movement eras were dominated by either formal and representative democratic practices or informal and direct practices, contemporary social movements are much more likely now to employ direct participation within formalized democratic processes (Polletta, 2002). Given the parallels and interconnections Rothschild-Whitt (1982) found that between the last wave

Workplace Democracy Comes of Age

211

of worker-owned businesses and social movement organizations, much more formality might be expected in contemporary worker-owned enterprises. What is at stake in examining today’s democratic workplaces is to tease formality apart from its tight coupling with hierarchy – a task that might be even more critical in work organizations than in social movements because accountability in the former is not only about political power but about dollars and cents, and is thus needed to ensure that its members have economic means of support. This close, empirical analysis of the speciﬁcities of democratic practice are critical for creating genuine economic democracy: that is, inclusive, broadbased, and economically viable. As Rothschild and Russell (1986) point out, most scholarships on worker-owned cooperatives propose models of workplace democracies that are not grounded in the experiences and outcomes of existing organizations. Without examples, it is too easy to attack democratic workplaces on grounds of managerial and economic infeasibility and perceived barriers of race/ethnicity, gender, class, and sexuality diversity. By drawing heavily on empirical data, this article demonstrates how one organization attempts to resolve these issues.

ONE WORLD NATURAL GROCERY: A CASE HISTORY Since the beginning of the twentieth century, the West Coast of the U.S. has been notable for a predominance of worker cooperatives. These have become particularly robust in the past 30 years (Berman, 1982; Gunn, 1984), unlike the Midwest or the Northeast where consumer and producer cooperatives were most common. What I call the ‘‘Golden Valley’’ region of California has been a West Coast center of worker ownership during this period, in part drawing on a strong and independent labor community. One World Natural Grocery was founded in the medium-sized Golden Valley city of ‘‘Edgecliff,’’ which has become whiter and wealthier in the past 15 years. While these changes occurred in One World’s immediate neighborhood as well, the store is still surrounded by an ethnoracially and class-diverse population, and customers span a continuum from afﬂuent, mostly white ‘‘foodies’’ purchasing $40 bottles of imported vinegar to poor, mostly non-white immigrants attracted to low prices for less processed or organic food.

212

JOAN S. M. MEYERS

Originally, One World exempliﬁed Rothschild-Whitt’s (1982) small, homogeneous, friendship-driven collectivist democracy. Started in the early 1970s as a natural food-buying club of a non-traditional spiritual group, its founders typiﬁed the low-paid, white, dropout, educated population she found in similar businesses. A scandal rocked the community shortly after they opened a storefront to the public in 1975, causing many original members to leave and the organization to secularize rapidly. Owing to their cultural rejection of hierarchy and a desire to integrate the interests of old and new members, an organizational hybrid of collectivist and representative democratic practices resulted, as discussed below. As the business grew, various obstacles were innovatively overcome: to shield those averse to state interaction, a separate store and membership was set up and later dissolved to sell taxable items; undercapitalization for a move to a larger storefront was resolved through literally hundreds of small loans from the community of customers; and a change in legal status 10 years later helped attract necessary bank backing purchase their current building. Success through creative solutions generated organizational trust in non-traditional methods. The relocations in the early 1980s and the mid-1990s each necessitated a rapid burst of hiring, again calling for integration of many new workers into an atypical business culture. Nevertheless, One World did not move toward the hierarchy and job specialization becoming common in other expanding cooperatives in the Golden Valley region, including an employee-owned ﬁrm that hired an outside managerial force and another that increased its pay ratio from 3:1 to 4:1 after a fruitless search for a CEO. Instead, One World divided its departments to maintain small internal collectivist units and added more levels of representative bodies to coordinate relations between departments and the membership as a whole. This strategy seems to have paid off. Unlike the disappearance of almost all other worker-owned cooperative food stores in the region, One World doubled its workforce over the past 10 years, has annual sales in the multimillions, and donates close to $100,000 yearly in grants to local community organizations. To date, it still has no managerial strata. Furthermore, its diverse member composition is unlike the cooperatives depicted in earlier sociological literature. Contrary to the earliest criticisms of cooperatives, it has not degenerated into a corporation controlled by a small group of private owners nor stagnated as a small, struggling enterprise (Webb & Webb, 1897), and neither does it rely on a homogeneous friendship network for recruitment and retention. As the only 100% employee-owned business of its size to eschew positional management, it offers a unique view of the

Workplace Democracy Comes of Age

213

attempt to combine heterogeneity and broad-based worker empowerment. By extending theories of democracy, diversity, and growth, this study turns an exceptional case into an exemplar of the viability of workplace democratic processes.

RESEARCH METHODS Data for this study was collected using mixed methods: survey, archival, and participant/non-participant observation. Archival documents included current and past advertising, promotional and customer relations’ materials, meeting agendas and minutes, handouts and notices to employees, and internal ﬁnancial documents. While member surveys and the organization’s ﬁnancial records alone could reveal the presence or absence of economic growth and workforce heterogeneity, ethnographic study is particularly well suited to teasing out how people make meaning of their actions in speciﬁc contexts (Burawoy, 1991), and how practices of difference, democracy, and growth affect each other in a workplace. To go beyond simple assertion in this case, my immersion in the organization over a sustained period helps me explain how One World democratically mediates between different interests while remaining competitive in the marketplace. The majority of my observation research took place over ﬁve weeks in the summer of 2003, ﬁve to six days per week, eight to ten hours per day. I took extensive notes on the ﬂoor and wrote them out in greater detail at the end of each shift. Data was collected on daily activities and meetings of the four departments chosen for intensive study (one heavily production oriented, one primarily customer oriented, one primarily administrative and nonrevenue producing, and a customer-oriented department widely considered the most fractious in the store), meetings of the most powerful committees, membership-wide meetings, ‘‘Cooperative Concerns’’ sessions (discussed below), and orientations for new workers. These settings revealed key similarities and differences in expectations and outcomes across the organization. While shadowing members within each department, I informally interviewed workers about their lives and paths to One World. I collected demographic information in a more systematic manner on all members of each department, including job title and responsibilities, tenure at One World, job before One World, age, ethnoracial identiﬁcation, gender identiﬁcation, sexual identiﬁcation, education, where they grew up, how they would identify their family’s class, parents’ education and occupations, ﬁrst language, and whether they were U.S.-born or born elsewhere.

214

JOAN S. M. MEYERS

Research for this case study is part of a larger project on worker cooperatives. Thus, data from One World is contextualized within formal and informal interviews I conducted with members of all the larger (100+ employees) West Coast cooperatives, participation in three West Coast worker-ownership conferences, attendance at several membership and board of directors meetings of One World and other large cooperatives between the winter of 2002 and the summer of 2004, and interactions at numerous formal and informal social gatherings of worker owners in the Golden Valley region of California from 2001 through 2004.

DEMOCRACY, GROWTH, AND DIVERSITY Densely Clustered Democracy As noted above, workplace democracy can have different manifestations and therefore different problems can arise from a lack of voice and control in highly representative structures to the potential for deadlock or coercion among ‘‘friendly’’ small groups. One World, however, avoids many of these problems by combining both forms. One World has two primary representative bodies: a board of directors elected from the membership and responsible for long-term planning and ﬁnancial decisions; and an interdepartmental committee, also elected from the membership, primarily adjudicating grievances, settling conﬂicts between departments, and coordinating storewide concerns such as security.5 The authority of either body may be (and often is) challenged by decisions made at the monthly membership-wide meeting. Both these and other organizational bodies retain organizational – and entrepreneurial – dynamism through yearly elections. Almost all elections are contested and involve public candidate meetings and vigorous informal debate about issues and candidate suitability. Such organizational actions might seem less-actual struggles over power than exercises in creating legitimacy (as Kleinman, 1996 found talk of money to serve in an alternative organization), but they in fact result in regular turnover in elected positions, providing renewed and often refocused attention to internal affairs. Members report deciding to run for a speciﬁc committee in order to insure fulﬁllment of a task allocated to it by the membership – for example, creating an exercise program, purchasing new property, or ensuring more prompt translation of a wider set of internal documents. Beyond these manifestations of vertical power, One World also uses the horizontal power of collectivist democracy by allocating almost all

Workplace Democracy Comes of Age

215

day-to-day decision making to each of its 14 departments. Departments have autonomy in policies regarding purchasing, hiring and ﬁring, scheduling, and even decision making itself (e.g., majority rule, two-thirds majority, or consensus). No departments have managerial positions, although all have elected one- to ﬁve-member coordinating committees charged with different duties (some check time cards and initiate disciplinary actions, while others merely set schedules) and some have a single person choosing products to be sold. However, these roles are not more highly compensated than others: pay differences are primarily based on tenure, and poor buying or coordinating can result in the department voting in new personnel. In traditional ﬁrms, a successful manager will match employees’ skills with necessary tasks, and assign universally disliked tasks to those least likely to complain or resist. Such a manager must be highly astute and powerful; a less successful manager may make ineffective matches or create resentment and resistance over task allocation. In the collectivist-democratic departments of One World, members discuss concerns with their own and each other’s work on a regular basis. They then take on and trade off tasks that match skills and desires,6 often changing over time. While unpleasant chores tend to fall to newer employees, far fewer tasks are seen as such than in other workplaces. The increased ability of members to choose their work reduces the number of responsibilities seen as undesirable. Thus, the horizontal practice of democracy makes quick and effective use of workers’ skills, while also allowing members to take on new tasks and develop new skills. In short, the culture of One World is highly formal but non-hierarchical, with democratic processes densely clustered and widely distributed. It is therefore almost impossible for an employee not to engage with the democratic organization of the workplace. Such democratic participation is also seen as work, and explicitly valued as such by One World: members receive the same rate as their other hours for meetings. Thus, One World ensures that democratic participation relies far less on an implicit social contract than it did in earlier generations of collectivist democracies by paying workers to do the planning and coordinating traditionally done by managers.7

Economic Growth One World’s focus on internal process contrasts sharply with the ideal hierarchical organization in which decisions are made quickly by one or a few managers and implemented by an obedient workforce. A typical corporate

216

JOAN S. M. MEYERS

manager can instantly ﬁre a worker for theft; in contrast, One World’s process requires a collective body (usually the department, but occasionally the Interdepartmental Coordinating Committee [ICC] or the board of directors) to terminate a member. This body then calls for temporary suspension of pending ICC investigation, and there is potential for appeal by the accused worker all the way up to the monthly membership meeting. Expending resources on time-consuming and non-revenue-generating procedures would seem to handicap an alternative organization competing with more efﬁcient corporations, particularly those with stripped-down layers of management. Yet if One World’s democratic process is inefﬁcient and non-competitive, how can its marked economic growth be explained? One World has developed from a handful of volunteer members to a workforce of over 200 employees whose median total annual compensation8 averages more than $40,000 – about 10 percent higher than the per capita median income recorded by the U.S. Census for the city of Edgecliff (see Fig. 1). In comparison, as shown in Fig. 2, the average hourly wage of unionized grocery workers in California is 25 percent less than the state average, and nearly 40 percent less than One World’s.9 One World members work an average of 27.35 h/week over the year – or, more typically, 32 h/week with almost six and a half weeks of paid holiday, vacation, and personal time off. While threats to strip away healthcare $45,000 $40,624 $40,000 $35,000

$35,000

$30,000 $25,000 $20,000 $15,000 $10,000 $5,000 $0 Edgecliff

Fig. 1.

One World Natural Grocery

Average Per Capita Income of Inhabitants of the City of Edgecliff and Members of One Word Natural Grocery.

Workplace Democracy Comes of Age

217

$30 $25.04

$25 $20.44 $20 $15

$14.97

$15.31

Large supermarket workers

Unionized supermarket workers

$10 $5 $0 Working Californians

One World members

Fig. 2. Average Hourly Earnings of Working Californians, Large Supermarket Workers, Unionized Supermarket Workers, and Members of One World Natural Grocery in 2004.

beneﬁts have mobilized unionized grocery workers nationwide, One World’s healthcare package covers a choice of doctors and alternative practitioners, weekly mental health visits, and applies to all members averaging at least 24 hours per week.10 One World moved from successively larger leased storefronts to its current ownership of a quarter of a city block, occupying a 40,000 square-foot building and slowly buying smaller properties around it. In addition to making navigation of aisles easier, increased space also allowed the business to offer a greater variety of items, and a greater range of high- and low-end items within each sales category. One World’s annual sales were $28 million in 2003, and their yearly growth ﬁgures seem to show an ability to hold their own against encroachment by national health food chains like the non-union Whole Foods.11 One World’s democracy is not unconnected to its growth. In the 1970s, members proposed that One World should respond to customer demand for paper products and other taxable items, but were met with ﬁerce opposition from other members. Some argued that selling taxable non-food items would increase problematic interaction with the state, and others wanted to preserve face-to-face democracy and the wide rotation of tasks possible only with a small workforce that could not handle more products. Furthermore, it would be impossible to add these items without reducing the stock of

218

JOAN S. M. MEYERS

basic, edible items – a reduction that some members opposed. Their solution was to form a non-proﬁt umbrella organization administering two separate stores, each with a small and interdependent staff, and only one of which paid sales tax. This solution also extended into quite literal growth into a storefront a few doors away. The success of the material and organizational growth was widely seen as a winning and creative response to external challenges, and gave One World practical proof of possibilities outside a small circle of intimates. Conﬁdence in growth encouraged One World to seek a single, larger storefront in the early 1980s, while maintaining its twostore innovation with a dividing wall. As each store’s business grew, however, maintaining separate books became onerous and the two stores ﬁrst merged ﬁnancially in the 1980s and then spatially reunited in the mid-1990s in a retail space on par with all but the biggest supermarkets in Edgecliff. One World’s willingness to grow stands in sharp contrast to the dozen or more other worker-owned and consumer food cooperatives of that period that explicitly chose to stay small. With one exception, those smaller Edgecliff food cooperatives are no longer in business. Despite the ostensible inefﬁciency of decentralized control, One World’s democratic process in fact promotes economic growth. In addition to generating and implementing such creative solutions, this relationship is illustrated in the series of events surrounding a member termination. The ﬁred member appealed after a three-month ICC investigation. To avoid potential litigation, One World’s lawyers advised distributing copies of conﬁdential records for members to read during the monthly membership meeting, and then collecting and destroying the copies after the appeal vote. All members attending the meeting were paid for the time it took the slowest reader to ﬁnish the documents. While this incident is remembered with no small degree of frustration, it was nonetheless an extremely effective method. Not only did the membership uphold the ICC’s decision, but also thereafter appeals to the membership meeting diminished sharply as trust grew in the ICC and One World’s other decision-making bodies. Given the regular turnover in committees, this trust must be seen as invested in the democratic processes themselves. Such trust allows both representative and intercoordinating departmental committees to make decisions quickly and implement them more smoothly than is possible in a company characterized by employee distrust of, and resistance to, top-down workplace direction. Indeed, the four months of member labor involved in adjudicating this matter is hardly different from the months of formal documenting processes typical in current corporate settings, which consumes countless numbers of hours and introduces new layers of bureaucracy.

Workplace Democracy Comes of Age

219

Furthermore, such participatory democratic processes are also efficient.12 As Womack, Jones, and Roos (1990) argue in their transnational study of the auto industry, traditional hierarchical companies often suffer from the reluctance of lower-level managers to pass bad news up the chain of command, leading to bad planning based on inaccurate data. In contrast, they show, a ‘‘lean production’’ model overcomes such problems by reducing layers of management and spreading decision-making responsibility more widely across hierarchical levels so that, for instance, the design team can make use of expressed customer desire garnered from retail sales management in each new product. Similarly, One World’s day-to-day ICC coordination and board’s long-range planning is tightly linked to the immediacies of consumer demands, space and function issues in the physical plant, and supplier issues for the simple reason that such concerns are already the job of the members of these bodies when they are not meeting as the ICC or board. Eliminating a layer of managers that might at best poorly convey the realities of the ﬂoor – and at worst might intentionally sugarcoat a sour report to appear more in control – allows essential information to be quickly and directly incorporated. The interlinked accountability structures that result from the density of democratic processes produce another form of efﬁciency: reduction of employee theft and sabotage. More than a few members described arriving at One World in search of ways to skim time or actual cash, but quickly coming to feel they would be stealing from themselves and their community – in marked contrast to how they described their feelings and actions at previous jobs. Because members have regular, direct venues for expressing concerns and materializing solutions, and because each member sees other members’ actions as potentially affecting their mutually held business, there is almost none of the costly sabotage that is common where workers have no formal channels to control their working conditions. Democratic processes are supreme vehicles for ‘‘manufacturing consent’’ (Burawoy, 1979), but here it is consent to mutually proﬁt rather than consent to turn over one’s labor power to others. In a hierarchy, the only way to achieve status and power is to be promoted up the chain of command. Failure to be promoted causes ambitious employees to seek advancement in other companies, resulting in the problems of turnover described above. Because compensation is good and autonomy greater than at other jobs, and because all members start at the same rate and receive raises based primarily on seniority rather than their ability or failure to learn new tasks and increase responsibility, One World members have no push to leave.13 This might be viewed by those who see

220

JOAN S. M. MEYERS

advantages in the relatively high turnover of the current U.S. labor market as creating a stagnant employee pool and losing fresh visions. Democratic workplaces like One World, however, counter this in two ways. First, while burnout is common on One World’s ICC, and many members step down after particularly emotionally intense grievances have been adjudicated, they do not leave the organization itself. Instead, they pick up more hours in their home department and are available for formal and informal consultation and interaction with new ICC members, thus preserving institutional knowledge. In addition, One World does generate new ideas. While One World members work primarily in a single department, many have shifts in two or more. This allows creative sales strategies to be shared between departments while maintaining each department’s autonomy. Similar to the advantages researchers ﬁnd among the interlinked teams in ‘‘lean production’’ companies (Womack et al., 1990), overlapping departments help mesh together company processes and strategies. A member who feels stuck or stiﬂed in one department can apply to work in another, and it is fairly common for members with 10 or more years of tenure (approximately one quarter of the total workforce) to work in a department different from where they started. Retraining costs are outweighed by the transferal of knowledge across departments – computer skills, contacts with suppliers or customers, or retail proﬁciency. However, it is only in the density of these democratic processes – departmental and interdepartmental meetings – that such knowledge is efﬁciently brought out and transformed into sales advantages. Retention of workers in a retail business has yet another effect: customer loyalty. One World ﬂoor workers regularly greet customers by name, know many customers’ preferences, and are able to respond to requests with new goods or well-received explanations of the inability to provide them. These personalized interactions in an era of disposable and scripted service work (Leidner, 1993) themselves add value to One World’s ‘‘product’’: service in sales. Even beyond that, members’ tacit knowledge of consumer desires is rapidly transformed into increased sales through the collectivist-democratic processes of ordering, displaying, and pricing items. Finally, while One World assigns multimember committees to do tasks typically undertaken in corporations by single managers, it often saves money by doing so. Because there is a 3:1 pay ratio between the highest- and lowest-paid members, it would take a committee composed of more than a 100 of the highest-paid workers – devoting their full hours to the committee – to cost the organization as much as hiring a CEO at the U.S.’s average 370:1 pay ratio. One World’s density of worker input – including the occasional long, expensive meeting – thus appears to be cost effective.

Workplace Democracy Comes of Age

221

A mutually reinforcing relationship exists between economic growth and democratic process. Among a small labor-managed membership, burnout can result in loss similar to that in corporations because the few coworkers who might take up the slack already have their own set of extra responsibilities. As many One World members pointed out, in their larger organization there are always others able and willing to take on greater responsibility. That is, growth beyond a small circle of intimates allows for burdens to be distributed among a larger number of people, which funnels more energy toward managing problems and promoting the business, which creates an even larger membership to share burdens. Thus, One World’s sustained economic growth can be seen as not merely a ﬂuke, but an outcome of its densely clustered democratic processes.

Producing Diversity One World’s heterogeneity is noticeable on the ﬂoor, in heated discussions at meetings, and in its personnel records. Earlier studies of worker cooperatives highlighted two forms of homogeneity: demographic and ideological. Rothschild-Whitt (1982) describes recruitment in these earlier cooperatives as relying on preexisting shared beliefs about the value of decentralization and job rotation. In contrast, One World hiring committees more typically seek speciﬁc food industry skills and experience, explaining, ‘‘We can teach co-ops to anyone.’’ Indeed, newer members describe deep skepticism about – if not actual resistance to – the viability of ‘‘no bosses’’ upon being hired. In 1996, a rumor that One World was secretly owned by a small group of older members spread quickly among a large group of recent hires. Later, some of them explained that such ownership made more sense than the information in orientations, that everyone there owned it, because they had never heard of such a thing. However, differences in beliefs in or ideological commitment to collectivism are not only due to a lack of information. A small but vocal minority of members is at pains to detail problems they perceive in One World’s decentralized structure and describe various attempts they or other members made to institute permanent management positions – all of which failed at the departmental or storewide levels. That is, unlike the earlier generation of cooperatives, One World members not only start with different views on workplace democracy, but also many continue to differ from the majority, some for a long duration of their membership.

222

JOAN S. M. MEYERS

In addition to the differences in opinion over workplace control, One World differs from the previous model of a community of like-minded individuals in political commitment as well. While few are politically conservative, political differences between democratic centrists, communists, and anarchists produce periodic clashes. Former combatants from opposing El Salvadoran government and guerilla armed forces were simultaneously One World members. Some members describe their freedom to be openly queer at work, while others formally disapprove of displays of such openness and propose no longer closing the store in recognition of the city’s LGBT holiday. In short, the organization does not transform member opinion into a homogeneous whole, but instead coordinates between heterogeneous ideological positions. An equally pressing concern for democratic workplaces, however, is transforming a history of recruiting and beneﬁting only members of dominant social groups. One World is now demographically comparable in some respects to the population of other large supermarket workers, but differs in others. Its race and education proﬁles are strongly similar, while its gender proﬁle differs, as shown in Figs. 3–5.14 Speciﬁcally, while the class background of One World members strongly resembles that of workers in large supermarkets15 – measured in the similar rate of formal higher education (Fig. 5) – and One World’s racial makeup is similar to both large supermarkets and all California workers (with some overrepresentation of whites), One World has an overrepresentation of women when compared to working Californians and to workers in large supermarkets. Since One World began to track its ethnoracial demographics in 1983 as required by the state, whites have diminished from 85 to 61 percent of the membership. More than half of the non-white population is Latino – an internally diverse group of Chicano and ﬁrst- and second-generation North, Central, and South American members. In addition to whites and Latinos, membership is Asian, Asian American, and Southeast Asian; African American and Caribbean black; indigenous; and mixed race. Bilingual and primary non-English speakers interact with each other and their customers on the ﬂoor – mostly in Spanish, but also in Vietnamese, French, Cantonese, and Portuguese. Public and private signs and documents are translated into Spanish or English as needed by the in-house translation committee, and all large meetings automatically have professional simultaneous Spanish interpretation on wireless headsets. In a reversal of the demography of large supermarket workers and the working Californian population in general, there are somewhat more women than men as well as two members identifying as transgendered. More than 10 percent of the members described

Workplace Democracy Comes of Age

223 1%

100% 90% 80%

43%

43% 56%

70%

Transgender 60%

Female

50%

Male

40% 30%

57%

57% 43%

20% 10% 0% Working Californians

Fig. 3.

Large supermarket workers

One World members

Gender Demographics of Working Californians, Large Supermarket Workers, and Members of One World Natural Grocery.

themselves as non-heterosexual. Signiﬁcantly, jobs and departments are not particularly gendered or heteronormative. Less visible but no less signiﬁcant is the class shift that has taken place in the past 25 years. As Bettie (2003) convincingly argues, class is as much a cultural as material production, and ‘‘class comes to be known equally by markers that exist outside of discovering one’s position in paid labor, as an identity lived out in private life and personal relations’’ (42). Due to a lack of class language in the U.S. very few current One World members identify as ‘‘working class,’’ but private and personal markers of the working class permeate the lives of current One World members: few have college-educated parents; most members themselves have only a high-school education or maybe some college courses (Fig. 5); their family rented homes or, if they owned them, came from rural areas; many had taken jobs in high school for economic reasons; few could rely on family money to help out with extras like car payments or vacation. Having few family resources, working-class people must exist on their own job earnings and can less easily choose low wages in exchange for ‘‘empowered’’ job conditions. As compensation rises

224

JOAN S. M. MEYERS 1%

100%

6% 90%

1% 4%

1%

9%

7%

13%

6%

80% 20% 70% 28%

35%

60%

Native American African American

50%

Asian American/Pacific Islander

40%

Latino

62%

30% 53%

White

51%

20% 10% 0% Working Californians Large supermarket One World members workers

Fig. 4.

Ethnoracial Demographics of working Californians, Large Supermarket Workers, and Members of One World Natural Grocery in 2004.

at One World – not only in wages, but also proﬁt sharing and the rich beneﬁts’ package – membership becomes an increasingly attractive option. Unlike some worker collectives where members derive moral worth from earning less than their middle-class peers in the corporate world (Kleinman, 1996), One World members identiﬁed above-average compensation as an important part of their job satisfaction. Nevertheless, organizations that value expressions of dominant cultural capital (being able to write a clear argument in standard English for an agenda, speak calmly and succinctly before large meetings, or make sense of ﬁnancial documents) run the risk of replicating social stratiﬁcation by rewarding those with such cultural capital (Bourdieu, 1986). One World is not free of this dynamic. Several members with a family or educational background that cultivated middle-class cultural capital are perceived by others as ‘‘naturally’’ wiser, more talented, or otherwise more deserving of respect and power than those whose speech is less grammatically correct, who have

Workplace Democracy Comes of Age

225

100% 9%

10%

90% 80%

34%

70%

College degree or more No college degree

60% 50% 91%

90%

Large supermarket workers

One World members

40% 30%

66%

20% 10% 0% Working Californians

Fig. 5.

Levels of Higher Education of Working Californians, Large Supermarket Workers, and Members of One World Natural Grocery in 2004.

not been trained to express a coherent argument, or who do not know how high ﬁnance works. Several One World practices, therefore, become signiﬁcant in countering preexisting advantages in cultural capital. Many forms of cultural capital can be learned in workplace organizations if participants are not shamed out of engaging due to a lack of prior skill in areas like public speaking or ﬁnances. The practice of paying members to attend meetings and read the minutes of a variety of organizational bodies subsidizes practice in One World’s key skills. Although attendance at membership meetings is voluntary, becoming a member requires observing meetings with a ‘‘buddy’’ paid to promote new worker acculturation.16 To limit monthly membership meetings to their ofﬁcial three hours, all agenda items must include the historical context and business objective, describe projected outcomes, and outline implementation. This requirement tends to prevent frivolous or confusing items from reaching the stage of member discussion. The requirement also intimidates new workers with little research and writing

226

JOAN S. M. MEYERS

experience. To remove potentially classed barriers (and therefore raced, given the disproportionate U.S. racialization of class), members of the elected ‘‘Successful Participation Committee’’ are paid to work one-on-one with members to write and research proposals. Members thus gain skills and are invested with new forms of cultural capital. The density of committees and collectivist-democratic decision-making processes block the accumulation of personal power by those with dominant cultural capital. For instance, when one long-term white male board member was found to have gone around the backs of the membership – even to do something that was widely seen as beneficial to the organization and indeed the majority’s choice – his bid for board reelection was rejected by the membership. This was due both to a violation of the cultural trust in shared information and to the redundancy of his seemingly specialized knowledge in the board of directors, who did not need his presence to accomplish the organization’s goals. Of course, One World is not perfect and members make formal complaints – products carried or not, appropriateness of other departments’ displays, member behavior, allocation of space, potential expansion, and so on – through membership meeting items, appeals to the Interdepartmental Coordinating Committee (ICC) or board of directors, items placed on other departments’ agendas, or Cooperative Concerns sessions. The Cooperative Concerns, or ‘‘CoCo,’’ process was implemented as a way to shorten monthly membership meetings by displacing debate on contentious items. Arranged by either the board or the ICC at the request of at least three members, it runs exactly one hour, is scheduled no less than two weeks from the date requested, and attendance is – as with all meetings – paid at the same rate as members’ other tasks. In 2003, a CoCo meeting focused on African American underrepresentation was convened by the ICC at the initial petition of a long-term African American worker. In deploring the situation, a white female member described a black male member as a ‘‘unicorn,’’ inferring he was left to fend for himself as the only one of his kind. While there were at that time four other black male members of One World, and three African American women worked in his department, members were accurate in their overall self-criticism that One World failed to mirror the city’s eight percent African American population (see Fig. 4). Other members compared One World unfavorably to nearby unionized supermarkets where it seemed a larger number of employees are black.17 As people discussed the problem, the secretary outlined and tape-recorded what would later be distributed as verbatim minutes. At the end of the allotted hour, she abruptly ended the

Workplace Democracy Comes of Age

227

session, people thanked each other for their participation, and everyone left without proposing further action – as per the CoCo protocol.18 To an outsider, the CoCo format seems chaotic and ineffective. Can discussion be useful without formal afﬁrmative action, centralized hiring policies, or organizational analysis of why hires of people of color do not translate into retention within whiter departments? To an insider, however, the CoCo meeting is one in a series of interlocking ofﬁcial and unofﬁcial processes. CoCos formalize informal talk typical of any institutional setting. The debate is transcribed, translated into Spanish, and posted publicly in both languages. Discussions about the transcript occur in the break room where people read the logs, in hallways, and on the ﬂoor. By turning member complaint into documents all members are expected – and paid – to read, a shared discourse emerges. The salient points of this particular discourse were that underrepresentation is real and a legitimate organizational concern, and that internal departmental policies contribute positively and negatively to this problem. Crucially, this particular discussion occurred within an organizational context where economic resources are devoted to combating inequality through the ongoing efforts of an ‘‘anti-oppression task force.’’ This selfrenewing committee, initially selected from a pool of volunteers by the ICC and trained by consultants in 2001, holds an intensive multi-day workshop each year for up to 10 members who apply to take part and whose time is paid as their normal hours. Trainees describe bringing task force concepts and strategies into departmental and interdepartmental meetings where they sometimes meet resistance from other members, but have a robust language with which to advocate policy changes. Together these discussions and workshop outcomes frame a set of informal processes through which member demographics are politicized at the local level of departments, the site of day-to-day control. In her study of conﬂict resolution during the transformation of a private British coal mine to worker ownership, Hoffmann (2001) ﬁnds that increased democratic process leads to greater resolution – more informal but no less effective. In fact, while One World members correctly identify problems of underrepresentation, it is no small matter that One World has become more ethnoracially heterogeneous, not less, over the years.19 Thus, a ‘‘culture of complaint’’ (Weeks, 2004) may be how organizations become cognizant of problems to address. Over the year following this CoCo session, without any formal or centralized policies being put into place, a larger proportion than usual of new hires was African American and most of them are still working a year later.20

228

JOAN S. M. MEYERS

To sum up, One World effectively solves many of the earlier cooperatives’ problems of homogeneity with structural solutions that employ informal processes. By placing as high a monetary value on public discussion and complaint forums as on revenue-generating activities, by developing organizational bodies to identify pockets of privilege and inequality, and by devoting organizational resources to leveling the effects of cultural capital, One World creates the conditions that promote and coordinate heterogeneity and diminish inequalities among its membership.

ANALYSIS OF FINDINGS: PROMOTING DIVERSITY THROUGH GROWTH AND DEMOCRATIC PROCESSES One World’s organizational culture and achievements have important implications for studies of work and organizations. As the ﬁndings show, growth at One World did not occur in spite of loyalty to democratic practices, but was nourished by them. This is due in part to reducing the costs of management, but also to the creation of workplace mechanisms to harness tacit and institutional knowledge within the organization and deploy it for market advantage. That such mechanisms are inextricably connected to a democratic workplace culture can be seen very clearly in the example of the two-store innovation, where both One World’s democratic commitment to preserving majority and minority interests, and its ability to generate solutions from diverse member input, created a viable solution to a problem of internal dissent. Two Hundred Bosses Critics of democratic workplaces commonly claim that such organizations are too hampered by the slowness of multiple decision makers and the decentralization of organization goals to compete in the market. Nevertheless, it is perhaps more useful to compare worker ownership not to ideal manifestos for controlling labor, but to capitalism as it actually exists. Work ethnographies such as those by Gouldner (1954), Blauner (1966), Hochschild (1983), and Kunda (1992) reveal the ways in which considerable company resources are focused on controlling the movements, relationships, emotions, and attachments of employees within capitalist workplaces. While democratic processes are time consuming, they may waste less of the

Workplace Democracy Comes of Age

229

working day than the plotting and resistance between various levels of management and workers typical of a non-democratic workplace. Although Burawoy (1979) shows how management can recuperate some worker resistance for its own purpose, worker sabotage and disengagement is obviously both a material and moral threat to management, causing business schools to devote whole courses to the issue. The workplace studies cited above and others (e.g., Milkman, 1987; Moss & Tilly, 1996; Woods & Lucas, 1993) also make clear that hierarchical management as it actually exists is not the pinnacle of rationalized efﬁciency, but is itself hampered by racialized, gendered, and sexualized hiring and employment practices that impede maximum exploitation of available labor power by a particular ﬁrm. Keeping in mind the shortcomings of mainstream business, it becomes more apparent how One World’s participatory democratic practices drive and are regenerated by a thriving business. The managerial dynamism of traditional corporations is not only maintained at One World through voluntary turnover in committees, but is expanded by connecting nodes of tacit knowledge to long-term planning. The combination of decent income and autonomy encourages employee retention, which not only develops and maintains institutional knowledge but also facilitates customer loyalty. Accountability and ﬁscal and task-based empowerment built into the formality of One World’s processes decrease employee sabotage and theft. It is not that there is no boss at One World; rather, as many members explain, it is like the pressure of 200 bosses. By becoming a member, they simultaneously become target and practitioner of a multi-sited panoptical system of control. Yet because this system can be identiﬁed and addressed through extensive organizational channels, it is rarely experienced as a ‘‘Big Brother’’ or some other detached and external form of oppressive control.

Growth And Stability Some collectivist-democratic workplaces still attribute their survival as democratic bodies to their intimate membership. In contrast to One World’s strategy of increasing the size of a single membership to address diverse member and customer needs and interests, the San Francisco Bay Areabased Arizmendi cooperatives have created a system of daisy-chained collectives that offer freshly baked goods at several locations throughout the region. Branching out of the venerable Cheese Board in Berkeley, California, and coordinated by a parent foundation, each collective has committed

230

JOAN S. M. MEYERS

to training and sharing recipes with a new collective that, once established, goes on to spawn another (Cheese Board Collective, 2003). In developing small collectives, Arizmendi sees itself as preserving the necessary conditions for workplace democracy: face-to-face organizational control and consent. Yet One World shows that increasing member size does not have to mean giving up democratic practices and outcomes. One World both enjoys the beneﬁts of face-to-face collectivism and minimizes the disenfranchisement and alienation typical of large and representative bodies, but without inherently limiting organizational size.21 By connecting smaller units through multiple democratic channels, One World reduces the potential for accumulations of ‘‘star’’ or personal power found to be common in small groups (Freeman, 1984). Multiple and tightly linked levels of power allow for both stability and an exciting mix of ideas. This combination produces innovations in workplace practice that have helped One World develop further. The willingness to expand, however, is key to One World’s economic stability. It is not simply that One World was able to maintain collectivist democracy while growing, but that it did not see growth as incompatible with workplace democracy. Exceptional in its openness to expansion in its region, this growth allowed the organization’s democratic practice to be proﬁtable: by diversifying products and by increasing sales volume, One World was able to achieve economies of scale and to exert pressure on suppliers to provide higher-quality merchandise. The combination of expanded shelf space, market leverage, a close knowledge of consumer wants, and the collectivist-democratic organizational mechanisms to quickly transform such knowledge into sales created One World’s economic stability, which in turn allowed One World to extend power and resources to a heterogeneous population.

Affording Heterogeneity Indeed, a key ﬁnding of this research is that member diversity is largely predicated on economic growth and stability. Becoming a more substantial business allowed One World to more fully meet its members’ needs. Although no one is growing rich, no one is poor, either – One World’s mean total compensation of $40,852 provides an above-average living for its majority working-class membership. Because race and class are so deeply linked, above-average compensation allows more recruitment and retention of people of color than did the earlier and much more poorly paid cooperatives.

Workplace Democracy Comes of Age

231

Financial support is a key aspect of the ‘‘right livelihood’’ goal of earlier cooperatives: reduced alienation, greater dignity, and economic stability. One World’s ability to maintain and enlarge its ethnoracial diversity is also facilitated by the sheer numbers of members. Although there are, for instance, few African American men working in the store, in fact no one of them is a token. The larger the organization, the less likely it is that any one person is called upon to be the minority representative, a stressful burden that has been identiﬁed as a critical factor in the lack of diversity in alternative (Moraga & Anzaldu´a, 1983) and traditional business (Kanter, 1977) organizations. In developing a diverse workforce, therefore, it appears that a medium organization is more likely to be ‘‘beautiful’’ than a small one. Yet member diversity should not simply be regarded as an end in itself. In a retail enterprise like One World, located in a neighborhood that is also heterogeneous by class, race and ethnicity, and sexuality, a heterogeneous workforce is an asset to proﬁtability. The ability of One World to address its customers in multiple languages on a daily basis not only creates a comfortable shopping experience for a greater number of Edgeﬁeld shoppers, but also maintains conduits to discover and satisfy consumer requests. Focused displays and sales on non-mainstream holidays (Passover, Tet, Gay Freedom Day) encourage and seem to create expanding customer loyalty. Commitment to preserving shelf space for cheaper items with a lower proﬁt margin – items that members cite as desirable to their own families and friends – helps maintain a strong base of less-afﬂuent but nevertheless regular shoppers. Thus, a heterogeneous workforce is itself a marketing tool, reproducing economic growth and furthering economic stability.

CONCLUSION While few of the earlier generation of cooperatives survive today, their legacy remains in the increased willingness of lenders to ﬁnance cooperative expansion (Pencavel, 2001). In part due to these changes, several North American cooperatives that have lasted more than 20 years have found the means to grow beyond their original face-to-face collectivist democracy and now employ a variety of democratic strategies to manage their heterogeneity. Examples are as varied as Open Enterprises, an adult entertainment retailer with a workforce of 200; the Union Cab Company in the Midwest with 150; and Alvarado Street Bakery, a national producer of sprouted wheat baked goods with 120. Growth as a key to sustained existence suggests a countervailing trend in contemporary industrial

232

JOAN S. M. MEYERS

democracy. Rothschild and Russell (1986) note that labor-management theorists suggested that the more worker-owned ﬁrms succeed as businesses, the more they tend to lose their democratic character and degenerate into conventional capitalist organizations. Growth here, however, is not a repudiation of the smaller collectivist democracies, but rather an organizational expansion of the critique of hierarchical bureaucracy. This study makes a strong case for a return in scholarly attention to the ways in which worker ownership can provide economic growth for a diverse and democratic population. Clearly, investigation of other, similar cases would deepen and expand the argument presented here. Analyzing a large, mainly working-class democratic workplace with high levels of member homogeneity in a region different from Edgeﬁeld could reveal other barriers to ethnoracial diversity. Is the commitment to the density of formal democratic structures as strong in more homogeneous groups where there might be a greater assumption of similarity and thus a greater reliance on what Mansbridge (1980) describes as the ‘‘unitary democracy’’ of people who feel they have most interests in common? What effects might that have on dayto-day workplace control by the employees? A study of a neighboring large democratic workplace in Edgeﬁeld with contentious management battles and high numbers of college-educated members could illustrate the ‘‘tipping point’’ at which dominant groups accrue or retain more power and resources than subordinate groups. Investigating how smaller worker-owned companies incorporate One World’s density and formality of democratic structures could shed light on the extent to which workplace democracy is a tool for creating workplace diversity – even without growth in scale or available resources. Importantly, a comparison of One World with a hierarchically managed worker-owned company of similar size and with similar overall company diversity could help disentangle the relationship of democratic control to the distribution of power, autonomy, and wealth across ethnicity and race, gender, class, and sexuality; the improvement of job quality in terms of occupational safety and job satisfaction; and the development of transferable skills and knowledge across a diverse population. Such further studies would clarify the relationships between economic growth, democracy, and diversity in the workplace advanced here. The presence of such sites of future research reveals something very important: it is not necessary for advocates of social change in the workplace to seek hope only in employee participation schemes in capitalist organizations (e.g., Rothschild, 2000). In a job market ever more dominated by transnational corporations ﬁxed solely on the bottom line, it is time to

Workplace Democracy Comes of Age

233

refocus research on worker ownership and control, and to analyze its speciﬁc promises and rewards.

NOTES 1. ‘‘Diversity’’ is used broadly here to encompass race/ethnicity, gender, sexuality, and class. 2. One World Natural Grocery is a pseudonym. All names of organizations, places, and people that could identify One World have been changed. 3. This Buddhist tenet was loosely interpreted to mean work that did not harm people (in opposition to the arms industry, for instance) or the earth (as with chemical sales), but still allowed people to provide themselves with the basics of food, shelter, and community engagement. 4. According to the National Cooperative Business Association (NCBA), a U.S. federal agency, there are currently 300 worker-owned cooperatives (http:// www.ncba.coop/abcoop_work.cfm, n.d.). The NCBA also records 500 consumer food cooperatives, which often share the history, goals, and organizational structure of worker-owned cooperatives. Employees of a consumer cooperative, however, are far less empowered than those of a worker-owned business, and in several cases have unionized to protect their interests against community-owned management. While it is therefore difﬁcult to precisely assess the decline of these alternative organizations, there has clearly been a signiﬁcant decrease in such organizations from the thousands in the 1970s. 5. In addition, there are numerous smaller regular and ad hoc committees elected to facilitate the ﬂow of information (i.e., a committee to help members craft acceptable agenda items), goods (i.e., grants and donations committees), or services (e.g., an ecology committee creating efﬁcient and ecological lighting, design, and sanitation), or address member concerns (a design task force to reallocate space after new construction). 6. As members themselves note, such self-allocation can potentially reproduce socioeconomic patterns of occupational stratiﬁcation, even in a worker-controlled environment. This is an important area of future research, although such stratiﬁcation is not apparent at One World. 7. I thank Dina Biscotti for her comments about how meeting time is valued as participatory managerial work at One World. 8. Total compensation here includes wages, shares of the proﬁt paid in cash and paper shares, and interest on shares invested in One World. It does not include health insurance beneﬁts, childcare beneﬁts claimed by 39 members, paid gym memberships, a 20 percent discount on store items, and other non-cash beneﬁts. The average is based on data from the 186 members who worked at least 14 hours in the 2003– 2004 ﬁscal year, but excludes another 34 employees who had not yet achieved membership, whose employment terminated in 2003–2004, or who were on leave for the entire year. 9. All data regarding state workers and both unionized and non-unionized large supermarket workers come from Dube and Lantsberg (2004). One World average hourly salary reﬂects both wages and the yearly proﬁt-sharing lump sum. However, it

234

JOAN S. M. MEYERS

somewhat underestimates total compensation as it excludes dividends paid on noncash shares (an average annual payout of $1,496), and the additional childcare beneﬁts paid to 39 parent members. In comparison, Dube and Lantsberg caution that their 2004 data are likely a slight overestimate. 10. Of the working members, 122 average 25–40 h/week. Twelve members work more than this. There are 44 members who work 8–25 h/week. Seven members average less than 8 h/week by special permission of their department. (My ﬁgures exclude members on medical or other leave, and employees whose length of employment has not yet allowed them to become members.) 11. Whole Foods’ corporate strategy in the U.S. has typically been to move into niche markets developed by independent cooperatives. With 160 North American locations, it has been criticized for over a decade by community and union organizations for ﬁercely resisting unionization and paying sub-union wages. Its compensation data are not easily available, but websites created by workers report wages similar to those of other non-unionized grocery workers in California, and proﬁt sharing that is almost non-existent. 12. Polletta (2002) reminds us that the labor movement in the early twentieth century urged participatory-democratic structure over what was then seen as the hopeless inefﬁciency of non-responsive hierarchical bureaucracy; it was not until mid-century that cultural claims about hierarchy shifted to favor the presumption of efﬁciency in expert decision making. 13. In 2004, the average length of tenure at One World is eight years – three more than the average time U.S. workers remain with a particular employer. This ﬁgure must also be read in the context of the doubling of the membership in 1994. That is, at least half of One World could only have worked a maximum of 10 years. 14. Data on large supermarket workers and working Californians from Dube and Lantsberg (2004). Figs. 3 and 4 use 2004 One World personnel data. Fig. 5 is based on 2003 surveys of the four representative One World departments with 90 percent completion rates. 15. Education marks and measures class imprecisely, but most viably. 16. All workers must apply for membership after working roughly nine months or they are terminated. They must be voted in by their department, and must have completed ﬁve orientations and attended four speciﬁc non-departmental meetings. Unlike most other worker-owned cooperatives that have buy-in rates that can run as high as $5,000, the initial share purchase of membership is $10 – less than an hour’s pay. 17. While this perception may be accurate and due to a clustering of African American workers in one store, One World’s black membership is actually identical with the overall proportion of California black workers, and indeed 30 percent greater than the proportion of large supermarket workers in California (see Fig. 4). 18. Members report that these meetings sometimes end with a quick agreement to form a further workgroup (pending approval of funds by the board), but I never observed such occurrences. 19. That One World’s black membership does in fact mirror that of large supermarket workers in the state in no way undermines the legitimacy of One World’s desire to reﬂect Edgecliff’s black population – itself smaller than the state’s black population.

Workplace Democracy Comes of Age

235

20. As commonly noted by members and conﬁrmed by One World’s personnel records, ﬁrings are most common within the ﬁrst year of employment. One World’s termination rate is 25 percent in the ﬁrst year (both voluntary and involuntary), 10.5 percent in the second year, and drops off sharply to 3.3 percent thereafter – with no terminations of third and fourth year workers at all. This study cannot determine whether the reduced likelihood of being ﬁred later is due to a weeding out of bad or ‘‘bad ﬁt’’ workers, increased social bonds making ﬁring more emotionally fraught for department members, or members learning the tricks to continued employment. 21. Some members believe One World could not expand much more, but others who witnessed growth over the past 15 years disagree, as long as they can secure a physical space large enough to accommodate membership meetings. A membership of 1,000 would certainly also require even more interorganizational coordinating bodies and processes than are now in place, but the unfolding density of democratic structures suggests feasibility.

ACKNOWLEDGMENTS The University of California Institute for Labor and Employment provided funds that allowed me to begin researching contemporary democratic workplaces. Many thanks to Sea´n O´ Riain, Vicki Smith, Fred Block, Ming-cheng Lo, and Miriam Wells for their intellectual support and critique. Valuable suggestions in earlier stages of my project came from Jack Goldstone and Kim Voss. This article is particularly indebted to conscientious and insightful feedback from Vicki Smith, Julie Collins, and Fred Block. I continue to be beholden to the members of One World and People’s, whose knowledge, patience, and humor is far greater than I can convey.

REFERENCES Barrett, M., & McIntosh, M. (1982). The anti-social family. London: Verso. Berman, K. V. (1982). The worker-owned plywood collectives. In: F. Lindenfeld & J. Rothschild-Whitt (Eds), Workplace democracy and social change. Boston: Porter Sargeant Publishers Inc. Bernstein, P. (1982). Necessary elements for effective worker participation in decision-making. In: F. Lindenfeld & J. Rothschild-Whitt (Eds), Workplace democracy and social change. Boston: Porter Sargeant Publishers Inc. Bettie, J. (2003). Women without class: Girls, race and identity. Berkeley: University of California Press. Blauner, R. (1966). Alienation and freedom: The factory worker and his industry. Chicago: University of Chicago Press.

236

JOAN S. M. MEYERS

Bourdieu, P. (1986). The forms of capital. In: J. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241–258). New York: Greenwood Press. Breines, W. (1989). Community and organization in the New Left, 1962–1968: The great refusal (new ed.). New Brunswick, NJ: Rutgers University Press. Burawoy, M. (1979). Manufacturing consent: Changes in the labor process under monopoly capitalism. Chicago, London: University of Chicago Press. Burawoy, M. (1991). Reconstructing social theories. In: M. Burawoy (Ed.), Ethnography unbound: Power and resistance in the modern metropolis (pp. 8–27). Berkeley, Los Angeles, London: University of California Press. Cheese Board Collective. (2003). The Cheese Board: Collective works: Bread, pastry, cheese, pizza. Berkeley: Ten Speed Press. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational ﬁelds. American Sociological Review, 48(2), 147–160. Dube, A., & Lantsberg, A. (2004). Wage and health benefit restructuring in California’s grocery industry: Public costs and policy implications. Berkeley: UC Berkeley Center for Labor Research and Education. Ferguson, A. A. (1991). Managing without managers: Crisis and resolution in a collective bakery. In: M. Burawoy (Ed.), Ethnography unbound: Power and resistance in the modern metropolis (pp. 108–132). Berkeley, Los Angeles, London: University of California Press. Freeman, J. (1984). Untying the knot: Feminism, anarchism, and organization. London: Dark Star/Rebel Press. Gouldner, A. (1954). Patterns of industrial bureaucracy: A case study of modern factory administration. New York: Free Press. Greider, W. (1997). One world, ready or not. New York: Simon & Schuster. Gunn, C. E. (1984). Workers’ self-management in the United States. Ithaca, NY: Cornell University Press. Hacker, S. (1989). Pleasure, power and technology. London: Unwin Hyman. Hochschild, A. R. (1983). The managed heart: Commercialization of human feeling. Berkeley: University of California Press. Hoffmann, E. A. (2001). Confrontations and compromise: Dispute resolution at a worker cooperative coal mine. Law & Social Inquiry, 26(3), 555–596. Jackall, R. (1984). Paradoxes of collective work: A study of the Cheese Board, Berkeley, California. In: R. Jackall & H. M. Levin (Eds), Worker cooperatives in America (pp. 109–135). Berkeley: University of California Press. Kanter, R. M. (1977). Men and women of the corporation. New York: Basic Books. Kasmir, S. (1996). The myth of Mondrago´n: Cooperatives, politics, and working-class life in a Basque town. Albany, NY: State University of New York Press. Kleinman, S. (1996). Opposing ambitions: Gender and identity in an alternative organization. Chicago, London: University of Chicago Press. Kunda, G. (1992). Engineering culture: Control and commitment in a high-tech culture. Philadelphia: Temple University Press. Leidner, R. (1993). Fast food, fast talk: Service work and the routinization of everyday life. Berkeley: University of California Press. Loe, M. (1999). Feminism for sale: Case study of a pro-sex feminist business. Gender & Society, 13(6), 705–732. Mansbridge, J. J. (1980). Beyond adversary democracy. New York: Basic Books, Inc.

Workplace Democracy Comes of Age

237

Michels, R. (1959). Political parties. New York: Dover Publications. Milkman, R. (1987). Gender at work: The dynamics of job segregation by sex during World War II. Urbana: University of Illinois Press. Moraga, C., & Anzaldu´a, G. (Eds) (1983). This bridge called my back: Writings by radical women of color, (2nd ed.). Latham, NY: Kitchen Table Press. Moss, P., & Tilly, C. (1996). ‘‘Soft’’ skills and race: An investigation of black men’s employment problems. Work and Occupations, 23(1), 252–276. Pencavel, J. (2001). Worker participation: Lessons from the worker co-ops of the Pacific Northwest. New York: Russell Sage Foundation. Polletta, F. (2002). Freedom is an endless meeting: Democracy in American social movements. Chicago, London: University of Chicago Press. Rothschild, J. (2000). Creating a just and democratic workplace: More engagement, less hierarchy. Contemporary Sociology, 29(1), 195–213. Rothschild, J., & Russell, R. (1986). Alternatives to bureaucracy: Democratic participation in the economy. Annual Review of Sociology, 12, 307–328. Rothschild-Whitt, J. (1982). The collectivist organization: An alternative to bureaucratic models. In: F. Lindenfeld & J. Rothschild-Whitt (Eds), Workplace democracy and social change (pp. 23–49). Boston: Porter Sargeant Publishers, Inc. Schumacher, E. F. (1973). Small is beautiful: Economics as if people mattered. New York: Harper & Row. Sirianni, C. (1993). Learning pluralism: Democracy and diversity in feminist organizations. In: F. Fischer & C. Sirianni (Eds), Critical studies in organization and bureaucracy (pp. 555–576). Philadelphia: Temple University Press. Webb, S., & Webb, B. P. (1897). Industrial democracy. London and New York: Longmans, Green and Co. Weber, M. (1946). Essays in sociology. New York and Oxford: Oxford University Press. Weeks, J. (2004). Unpopular culture: The ritual of complaint in a British bank. Chicago and London: University of Chicago Press. Womack, J. P., Jones, D. T., & Roos, D. (1990). The machine that changed the world. New York: Rawson Associates. Woods, J. D., & Lucas, W. J. (1993). The corporate closet: The professional lives of gay men in America. New York: The Free Press. Worker-owned cooperatives. (n.d.). Retrieved February 20, 2005 from http://www.ncba.coop/ abcoop_work.cfm.

This page intentionally left blank

238

CAN COLLECTIVIST-DEMOCRACY BRING GENDER EQUALITY? THE EFFORTS AT TWIN OAKS Joyce Rothschild and Amy Tomchin ABSTRACT A division of labor that segregates household labor from capitalist employment and that de-values women’s work is ubiquitous in our society. This article examines Twin Oaks, a long-standing intentional community that is intensely focused on overcoming the gender-based inequalities they see in U.S. society. Specifically, they have tried to create a comprehensive alternative to capitalist work relations by developing a work system that values equally all forms of labor – from childcare to income-producing types of labor to pregnancy itself. We describe in this article the specific system they have developed for translating all forms of work into ‘labor credits’ on a one hour equals one credit basis and for encouraging men and women to perform work that in the surrounding society is often assigned to the opposite gender. We consider how they have accomplished this in some detail, and in our conclusion, we draw out some of the tensions or downsides this can create as well.

Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 239–262 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16009-4

239

240

JOYCE ROTHSCHILD AND AMY TOMCHIN

That modern bureaucratic work organizations are ‘gendered’ has been well conceptualized (Acker, 1990) and amply demonstrated. In making the case that gender-based inequities are everywhere in the capitalist work world and in not-for-proﬁt, for proﬁt and public organizations alike, research scholars have shown that jobs in our society are amazingly sex segregated (TomaskovicDevey, 1993), that the division of labor itself bears the burden of excessive gender typing (Harstock, 1997) and that this gender differentiation in the work world is reinforced and made possible by continuing differentiation in who does household labor and in who cares for children (Bunch & Carillo, 1990). Feminist social and political theorists have long argued that capitalist structures exploit women by relegating them to labor, often invisible or underpaid, that both sustains and reproduces households while simultaneously freeing men of household obligations and enabling them to contribute more time and energy to capitalist pursuits. Additionally, bureaucratic organizations, whether capitalist, not-for-proﬁt or in the public sector, have been found to engage in practices that limit women’s ability to participate and earn as men’s equals (Acker, 1990; Ferguson, 1984). Even alternative organizations that are premised on notions of equality and participation have found it difﬁcult to transcend gendered inequality. Kleinman (1996), for example, found that in an alternative holistic health center, gender inequalities continued despite members’ belief that everyone, male and female alike, should be equally valued in the organization. In an effort to learn how gender equality can be accomplished we analyze Twin Oaks, an intentional community and complex work organization founded in 1967 that takes gender equity as one of its core goals. Twin Oaks represents a concrete example of a sustained effort to overcome genderbased inequalities. Theirs is a purposeful effort to develop an economy, work relations and gender relations that provide a comprehensive alternative to capitalist relations. Although Twin Oaks exists within a larger capitalist society, and this may present special pressures as we will discuss, it remains devoted to creating a truly collectivist and egalitarian organization. Located in Louisa, Virginia, Twin Oaks is one of the longest standing of the modern intentional communities in the U.S. Founded on Skinnerian principles, it has jettisoned those early behaviorist concepts but has retained and even intensiﬁed its effort to establish an egalitarian and cooperative way of life, both in its economic and its social structures. Drawing on interviews with Twin Oaks members, our reading of documents members have produced, and ﬁeld work, we explore and assess how they have done at achieving their goal of gender equality in all realms of life, and particularly in work life.

Can Collectivist-Democracy bring Gender Equality?

241

METHODS The data on which this paper is based derives from interviews with members as well as ﬁeldwork that was conducted at Twin Oaks by the second author during January 2005. As a participant in a ‘‘Visitor’s Group,’’ the researcher was part of a 10-member group. Members of this group were considering joining Twin Oaks, and as such, were being offered an introduction to Twin Oaks’ policies, practices, members and philosophy. During her stay at Twin Oaks, the author interacted with members both socially (at meals and events) and through work. Some members were familiar with her research interests, having read a letter of introduction that she provided to the community prior to her arrival, and many made a point of approaching the researcher with information and insights. Others learned of her academic interest once she was there, and they too seemed more than willing to share with the researcher their impressions of some of the issues that we had hoped to address. None of the conversations with Twin Oaks members were recorded at the time they took place. Rather, the content of interactions and as close to verbatim quotes as possible were recorded in a personal journal during free time in the day and every evening, where every attempt was made to capture the relevant material that had been encountered. The author tried to write down observed episodes in their entirety. We chose the journaling method of data recollection because we judged it to be the least intrusive way to observe and learn from members as they went about their daily tasks. Although the presence of the researcher was clear, we believe that the attempt to be less conspicuous, through lack of visible recording or note taking devices, allowed us to capture more spontaneous and unedited responses (Denzin & Lincoln, 2003, pp. 47–49). During orientation sessions, however, when it was appropriate to take notes as members spoke, the author was able to record direct quotations. The names and distinguishing details of informants have been changed in this paper in an effort to protect members’ privacy. Although as researchers we did not get to spend as much time in the ﬁeld as we would have liked, going through the orientation for potential members allowed us the best access we could arrange. We integrated and followed our short time at Twin Oaks with many inquiries we directed to various members. Surely, our data are limited in which we had an opportunity to observe, ﬁrst-hand, for a longer period of time, we could have observed more of the nitty-gritty of group processes, conﬂict and decision-making. However, short of becoming actual members of Twin Oaks, we did everything we could to deepen our understanding of Twin Oaks, and in this paper, we

242

JOYCE ROTHSCHILD AND AMY TOMCHIN

make every attempt to summarize what we did come to understand about Twin Oaks from our ﬁeld observations, interviews and document review.

OVERVIEW OF TWIN OAKS As a feminist culture, at Twin Oaks we tend to disregard traditional gender roles and behavior. Women and men choose their roles in the community based on their interests, strengths, and passions, not on anyone else’s preconceived notions. Women and men operate chainsaws and drive tractors, and men and women cook and care for children y . Our commitment to fostering a supportive and joyful environment for all people – women, men, lesbians, gays, and children – is an integral part of Twin Oaks Community (statement from http://www.twinoaks.org).

Twin Oaks describes itself as an intentional community, which, according to the Fellowship of Intentional Communities, of which Twin Oaks is a member, is a community formed ‘‘to share resources, to create great family neighborhoods, to live ecologically sustainable lifestyles, or to live with others who hold similar values’’ (Fellowship for Intentional Communities). Rosabeth Moss Kanter, who examined communal life in her ground breaking work, Commitment and Community (1972), described the status and function of such communities ‘‘[a]s experimental groups’’ that ‘‘innovate with new forms of social organization, they imaginatively construct their own kind of collective being, and they strive for different and closer forms of human relationships’’ (Kanter, 1972, p. 213). Located in Louisa, Virginia, Twin Oaks is one of the United States’ longest lived intentional communities. The vision behind its founding was inspired by Walden Two. Twin Oaks has since shed its afﬁliation with Walden Two and today is a membership-based community centered on values of cooperation, sharing, nonviolence, equality and ecology (http://www.twinoaks.org). Twin Oaks, as a member of the Federation of Egalitarian Communities (2004), shares a similar vision of community with ﬁve other communities throughout the United States. Their aim is to help one another and to ‘‘help more people discover the advantages of a communal alternative, and to promote the evolution of a more egalitarian world’’ (http://www.thefec.org/about/). Membership is open to all individuals who meet membership requirements and who complete their membership process. The process begins with a three-week Visitor’s Program, where potential members live and work at Twin Oaks, meet current members and participate in educational lectures (called orientations, or ‘‘oreos’’) that introduce Twin Oaks’ policies and philosophy. At the end of the three-week visit those who want to apply for

Can Collectivist-Democracy bring Gender Equality?

243

provisional membership, the type of membership initially offered, arrange an interview with the membership team. Prospective members are asked to tell the team their life story, and answer questions about ‘‘how one deals with various aspects of community living like conﬂict, anger, people with different values, etc.’’ (http://www.twinoaks.org/FAQ.html#member). Each member of the community, if they so choose, is then able to post responses to the question of whether membership should be offered. The membership team considers these responses as it makes the ultimate decision about whether to offer provisional membership. If the applicant is accepted for provisional membership, he or she must leave Twin Oaks for 30 days before returning to the community. After six months of living at Twin Oaks, a provisional member is able to become a full member if given ﬁnal approval by the Membership Team. Twin Oaks is currently home to 93 adult members and their 15 children in seven residences. The residences are named after other communities throughout the world and each is referred to as a single living group, or SLG, and houses 10–20 members. Each member has her own private bedroom (although small children share with a parent, or parents, depending on a family’s individual arrangement) but all bathrooms, kitchens and living rooms are considered communal space and shared with the other members of the SLG and the community at large. Twin Oaks has a rule that there can never be more than a 60–40% split by gender, in either direction. If they ever have a gender imbalance greater than that, it is their policy and practice to ‘‘stop inviting as new members people from the dominant gender until we’ve come more into balance.’’ In other words, they pay attention at the invited visitor stage to the gender composition. Right now they are at a gender ratio of 50/50 in terms of their adult members. About a year ago, they report that they had 57% men for some months before they were able to re-balance. The Twin Oaks bylaws call for ‘‘a participatory form of government in which the voting members have either a direct vote or the right of impeachment or overrule’’ (Twin Oaks Bylaws, 1989). Although all members are involved in decision-making, a ‘Board of Community Planners’ makes most of the ﬁnal decisions. Board membership is rotated so that power does not become consolidated in the hands of a few individuals and planners can be impeached through a community vote. Each planner serves an 18-month term, which is staggered at six-month intervals. Current planners choose their replacements; however, community members can veto the board’s choice through individual votes. All adult members of Twin Oaks are eligible to serve in the position of planner.

244

JOYCE ROTHSCHILD AND AMY TOMCHIN

Twin Oaks strives to be as self sufﬁcient as possible and its members grow and produce a signiﬁcant portion of their own food, build and maintain their own living and working space, and collectively support each other ﬁnancially and emotionally. However, since Twin Oaks exists within the greater capitalist structure of the United States and must generate cash in order to secure goods and services from the outside society, the community also operates several revenue-generating businesses. One of these is a hammock-making enterprise, which, for the majority of the life of the community, has been its biggest moneymaker. However, this past autumn, Twin Oaks lost a lucrative contract with Pier One Imports, a nation-wide retailer, when the company decided to discontinue the sale of outdoor furniture. The community’s hammock sales were cut abruptly by 50% as a result. Twin Oaks has dealt with the loss of revenue by adding to their soyfoods business and expanding their book indexing business. They entered discussions with LL Bean to replace the lost hammock (rope) business, but this did not pan out. Now, they have formed an ‘‘economics team’’ to consider other options. As is their way, these option considerations were spearheaded by the Planners, who organized meetings, options and several avenues by which to solicit member input. Reportedly, large disagreements have not grown around the thorny question of what to do. Some of the implications of their pressed economic circumstances will be discussed in the next subsection. Working life and non-working life are well integrated at Twin Oaks, particularly because they often occupy the same physical and temporal space, and as we will show, community policies promote equality between the sexes in both the workplace and the household.

CHILDCARE WORK AND GENDER EQUALITY AT TWIN OAKS Since the devaluation of women’s labor in the household preceded and has been key to the historical devaluation of women’s labor in the public sphere (Bryson, 1996; Beckwith, 1986), we begin with an examination of how childcare and household labor are handled at Twin Oaks. We then go on to show how this forms the basis for a re-evaluation of work and personal value at Twin Oaks. The community’s treatment of childcare is quite unlike anything we see in the surrounding society. Where in mainstream households in our society parents take on the childcare role as an unremunerated but necessary task, at Twin Oaks childcare is seen as a job, and is awarded labor credits,

Can Collectivist-Democracy bring Gender Equality?

245

comparable to work hours, like any other job performed. All job tasks performed at Twin Oaks are awarded ‘labor credits,’ and those given for childcare work are the same as those given for other types of work tasks. This is an important point. Moreover, childcare is not prescribed to, nor performed exclusively by women, but rather, is a voluntary and communal activity open to any member of the community who is so inclined. At present, it is estimated by members that women and men do approximately the same amount of childcare. Currently, Twin Oaks has set a temporary maximum capacity of 15 children, meaning that as a group they have decided not to accept more than 15 children overall. This cap on the number of children, of course, inﬂuences membership decisions of the group. In its bylaws, Twin Oaks set a maximum of 1 to 5 ratio for children to adults. In other words, Twin Oaks bylaws maintain that 95 adults should be able to support 19 children; by consensus they have decided to lower this ceiling for now. This is an important result of feeling economically strained. According to Ralph, who gave an orientation session on children at Twin Oaks, this limit is based on the maximum number of children that Twin Oaks members feel they can support both ﬁnancially and emotionally. Since children do not work and are not expected to work, they are supported by all of the members of Twin Oaks, collectively, and not just by their parents as in mainstream society. This has important implications. Speciﬁcally, during periods of ﬁnancial hardship, like the one that Twin Oaks currently faces, there is a tension between those members who see children as ‘‘freeloaders’’ who live at Twin Oaks until they are 18, beneﬁting from other’s work without economically contributing to the community, and those who value the children in themselves and who see them as an integral part of their communal society. In the past, according to member reports, tensions between these two views have periodically erupted, and while we have not had the beneﬁt of observing ﬁrst-hand the terms and intensity of this conﬂict, we can say that such disagreements have been reconciled with group agreement on 15 children (under present circumstances). We know too, that some members speak of how ‘‘great’’ the kids are, that non-parent adults of both sexes volunteer to serve as ‘‘primaries’’ for other people’s kids on a daily basis (as we discuss later) and that the community has set aside funds to pay for the birth process and even for adoptions. On balance, it looks like the community likes having kids, but that this is not without dispute and negotiation. Common ideas are in evidence at Twin Oaks on certain aspects of raising children. For example, most members have home-births, with a midwife

246

JOYCE ROTHSCHILD AND AMY TOMCHIN

attending. Many practice extended nursing, where children breastfeed longer than most children in mainstream society, often until the age of three or more. Families usually choose to sleep in a family bed when children are babies. This means that the parents and the child/children sleep together. Most of the children, at least initially, are home-schooled. Parents who home-school their children are able to count a portion of the hours spent on that activity as part of their work fulﬁllment. For parents in mainstream society, these labor-intensive methods, usually unremunerated and extremely time-consuming, are often unrealistic for their children. The way this works is this: All adult members are expected to commit to 42 h per week of work at Twin Oaks. The speciﬁc tasks done are, with few exceptions, self-chosen. A new parent may put all of their work commitment into caring for their infant. Hours spent on childcare (called ‘Care Hours’) count, one for one, the same as hours spent on any other type of work at Twin Oaks. As the baby gets older, fewer hours would be expected to go to child care, but home schooling hours (called ‘Education Hours’) also count the same as hours spent on making hammocks or book indexing or any other kind of work activity. As it was explained to us, a certain number of labor credits are ‘‘attached’’ to each child, and anyone can take them. Often it is the parents who devote many of their hours to their child’s care, but others (called ‘‘primaries’’) may also devote some of their work hours to children’s care and education (private correspondence, September 30, 2005). We learned that in the ﬁrst few years of Twin Oaks’ existence there were no children. Most of the community’s founding members were single, young-adults. After several years some members identiﬁed themselves as wanting to have children and asked the group what they thought of this idea. This request, as with any request, was discussed by all of the members, and after ﬁnancial, social and logistical consideration, group approval was given to have children. Nowadays, the process for having a child has grown more involved. Those who want to have a child must ﬁrst publicly express their desire to have a child to the community, they must have already lived at Twin Oaks for at least two years, and they must talk to other Twin Oaks parents about their experiences and work out logistical considerations (like use of the limited number of strollers). Once they have fulﬁlled these obligations they are given permission to have or adopt a child, which means that the community will meet the ﬁnancial obligations that follow. This practice of seeking community ‘‘approval’’ for a decision that to many in a capitalist society would seem to be private is accepted at Twin Oaks, and gets little ‘‘push back.’’ Most members appear to accept this policy as economically warranted, reasoning that since the community

Can Collectivist-Democracy bring Gender Equality?

247

collectively provides all of the ﬁnancial resources for having a child, the issue and planning for a child are within its justiﬁed purview. Although this might appear to many readers as a disturbing group incursion into a private decision, it is important to view this issue from their lens as well: Twin Oaks does seek to support the families in question. They established an adoption/ pregnancy fund to provide money for a woman’s special needs while she is pregnant (vitamins, special foods), to pay for a mid-wife, and for medical or adoption expenses for birthing or adopting a child. The fact that Twin Oaks has been willing, on occasion, to cover adoption expenses (which can run into the tens of thousands of dollars), even within their modest budget, suggests that as a group they do value the presence of children. Lara, a single mother with two adopted children, emphasized this point when she indicated how difﬁcult it would have been for her to cover those expenses on her own were she still in the ‘outside world.’ Twin Oaks explicitly considers pregnancy, itself, as productive work and working hours for pregnant women are shortened at different stages of their pregnancy. Speciﬁcally, Twin Oaks provides 300 h of labor credits for a pregnancy (private correspondence, September 30, 2005). This time can be utilized at the parents’ discretion. For example, a relatively new mother at Twin Oaks said that by the end of her pregnancy she was taking about two hours a day (14 h a week). This 300 h of labor credit for a pregnancy is treated as the exact equivalent to 300 h spent on some other type of productive activity. Placing this in the context of paid leave from work for pregnancy in U.S. society, we know that the 1993 Family and Medical Leave Act provides for 12 weeks of unpaid leave and zero paid leave for the birth or adoption of a child. When we consider that Twin Oaks is operating at a much more modest income level than the average in the outside society, their provision of 300 h of labor credit for pregnancy, along with additional labor credits later for childcare and home schooling, reveal a very different view of what is ‘productive labor’ and a tearing down of the conventional wall between private (family) labor and public (paid) labor. At Twin Oaks, one hour of work equals one hour of labor credit, no matter what type of work one does. Unplanned pregnancies do occur and those families are allowed to stay at Twin Oaks. However, if the birth has not been pre-approved by the community, then the pregnancy and delivery costs for the child are not funded by the community, but are presented as a loan to the parents, which are considered ‘‘paid off’’ if the parents stay at the community for another ﬁve years. The community expects repayment on the loan if the family chooses to leave before ﬁve years. The community’s current moratorium on new members with children points to its struggle to economically sustain its members.

248

JOYCE ROTHSCHILD AND AMY TOMCHIN

By way of illustrating childcare at Twin Oaks, Ralph, the community’s childcare liaison, described how his wife, Claire, and he, raised both his son and stepson. He indicated that this was similar to the way other children at Twin Oaks were raised. ‘‘At ﬁrst,’’ explained Ralph, ‘‘Claire was the primary caretaker for the boys. She did the natural childbirth with a midwife, we shared a family bed (a bed in which all family members sleep together) and she did the extended nursing.’’ When asked about the family bed, Ralph explained that it allows parents of an infant to get more sleep since when the baby cries in the night ‘‘the mother just rolls over, nurses the baby, and everyone goes back to sleep.’’ After years of Claire taking the lead in caring for and teaching the boys, ‘‘She was done, she had had enough, and so I became their primary caretaker.’’ Ralph is now the primary caretaker and teacher for his two boys and Claire works full-time at other jobs at Twin Oaks. Ralph seemed to have an easy relationship with the children, and kids seemed to ﬂock to him for attention and guidance during our visit. With calm and patience, he responded to their individual demands and interests. For example, Shashi, a girl of ten, came up to us when we were working with hammer and nails, grabbed a hammer and asked if she could do what we were doing. Ralph carefully explained exactly how she was to hammer out the nails and discard them in the designated receptacle, and for the rest of the afternoon she was as productive as the rest of our team and without injury. During our short stay at Twin Oaks, we noticed that wherever Ralph went, children followed. From Ralph’s sense of pride and ease as the primary caretaker/educator for some of the children we inferred that he rejected gendered social norms found in mainstream society that relegate men to a peripheral role in childrearing. In our observation, there were plenty of men, probably as numerous as women, serving as the primary caregivers to the children at Twin Oaks, but we were unable to get an exact gender ratio on this. We do know that childcare work was viewed as the equivalent of any other labor activity at Twin Oaks, and it was accorded the same value in their system of ‘labor credits.’

TURNING PRIVATE HOUSEHOLD LABOR INTO PUBLICLY APPRECIATED WORK When the ﬁrst children were born at Twin Oaks, they were housed in a building called Deganya. All children under a certain age lived at Deganya (until young adolescence when they were given their own rooms). These

Can Collectivist-Democracy bring Gender Equality?

249

children were primarily cared for by ‘‘metas’’ (short for the Hebrew word, ‘‘metapelet,’’ which means nursemaid or nanny). Metas worked in shifts of four hours at a time, with someone always spending the night with the children. Metas were Twin Oaks’ designated members of either sex who chose to be specially trained to care for children and were considered well suited to the task. Reports from two members indicate that ‘‘there were somewhat more women than men’’ who served as ‘metas,’ but both men and women served (private correspondence, August 1, 2005). In the early days, metas were the only individuals who could be alone with the children. If neither biological parent was a trained meta, and if they wanted to take their child to visit a relative away from Twin Oaks, a meta would have to accompany them. In order to reinforce the relationship between metas and the children for whom they cared, all books with references to ‘‘mom’’ and ‘‘dad’’ were edited so that those titles were replaced with the word ‘‘meta.’’ Surely this says something about Twin Oaks’ self-conscious and extensive efforts to dismantle of the notion of the traditional nuclear family. Nevertheless, the Deganya and the meta system did not work for the parents, who demanded, and were given a bigger role in their children’s lives. Parents directly overturned the ‘meta’ system that Twin Oaks had developed, replacing it with a system more consistent with their concept of how to create close parent/ child bonds. After the meta system was disbanded in the early 1980s, children began sleeping in their parents’ rooms. Interestingly, the children’s house at Twin Oaks, Deganya, is named after the ﬁrst kibbutz founded in Israel. Since the childcare system was originally modeled after the kibbutz childcare system, this naming reinforced the style of childrearing chosen by Twin Oaks. The childcare system at the kibbutz Deganya developed as a solution for women so that they would not need to give up their work in order to raise their children. In this system, the community as a whole would be responsible (at least partially) for childcare. In practice, this meant that all women of the kibbutz would participate in caring for children. This developed into the ‘‘collective care’’ model that inspired collective education for children on the kibbutz (Fogiel-Bijaoui, 1992, p. 217). However, in Twin Oaks’ rendition, the idea was to eliminate the gendered division of labor, enabling men and women to participate equally in childcare. Twin Oaks’ decision to make childcare a non-gender speciﬁc job, and a publicly assigned job like any other, reinforced their egalitarian ideals and helped to diminish a gendered division of labor. This was not true of the kibbutz where it was women who were expected to work in the area of childcare (Fogiel-Bijaoui, 1992).

250

JOYCE ROTHSCHILD AND AMY TOMCHIN

Fogiel-Bijaoui, in her writings about the development of kibbutz life and gender roles, describes the process that took place on the kibbutz that created and reinforced a gendered division of labor pertaining to childcare. By the time that a generation of children was born on the ﬁrst kibbutzim, ‘‘the unequal division of gender roles created from the very beginning by the principle of economic proﬁtability, was legitimated by the deﬁnition of parental roles instituted within the communal settlement. From that time on the economic logic of proﬁtability and the social deﬁnition of parental roles were to reinforce each other’’ (Fogiel-Bijaoui, 1992, pp. 217–218). Since men were seen as more economically productive than women, women were given the role of childcare, which was understood by Kibbutz members as essential to the survival and growth of the community, but ﬁnancially unproﬁtable. Twin Oaks, by contrast, made a conscious decision both to view childcare as productive work similar to any other type of work at Twin Oaks and to encourage both male and female members to participate equally in it. However, once the system of metas was disbanded, parents were ultimately responsible for choosing who would participate in their children’s care, and participation of both men and women was at their discretion. A childcare system of ‘‘primaries’’ is currently in place at Twin Oaks. Individuals who want to spend time with a child on an individual and regular basis are considered the child’s ‘‘primaries.’’ One Twin Oaks member, Karin, who broke her tailbone while giving birth to her son, Pine, said that the help from other members, who became ‘‘primaries’’ for her son, was indispensable. Karin did not see how she would have been able to care for her newborn son on her own, which she would have had to do in ‘‘mainstream society.’’ ‘‘I could not be in a seated position for a long time. Others had to help me with everything,’’ Karin noted. Because of Pine’s early interaction with members other than his biological parents, now, at three years old, he still has 10 primaries who spend time with him each week. In the words of another female member of Twin Oaks, ‘‘I ﬁnd the quality of fathering here pretty phenomenal’’ (private correspondence, September 30, 2005). There are some parents who do not choose to have any primaries for their children, and instead spend the majority of their time with their children, teaching and supervising them themselves. ‘‘There are some kids that have many primaries,’’ Ralph told us at a childcare orientation. ‘‘Claire and I chose not to have any for our boys. We wanted them all to ourselves,’’ he said, almost defensively, closing his arms into his chest. Interestingly, Ralph felt that he had to defend his family’s decision not to have others involved. Despite the above example of Ralph’s close-knit family, consistently we

Can Collectivist-Democracy bring Gender Equality?

251

were told that nuclear families should not over emphasize their relationship or develop exclusionary tendencies. Another element of life at Twin Oaks that has proven important in removing gender inequality has to do with members’ living arrangements. All members live in single living groups and each adult has his or her own room. Even married people do not share rooms. Nina, who has lived at Twin Oaks for 15 years and is married to Jim, explained how their living arrangement works. They have single rooms (as do all adults) that are right next door to one another. ‘‘It is better this way,’’ she told us, since in the outside world men and women, in her opinion, do not have equal arrangements in married life. This system, according to Nina, helps to do away with some of those normative household inequalities, such as ‘‘who is expected to clean and keep things tidy.’’ It gives each person their own space, ‘‘allowing them to be together or apart as much as they like.’’ Nuclear families at Twin Oaks vary greatly in their makeup. Heterosexual and homosexual, as well as polyamourous arrangements are all present and appear to be accepted at Twin Oaks.

THE DIVISION OF LABOR, DEMYSTIFICATION OF SKILLS AND GENDER EQUALITY Twin Oaks goes out of its way, even at considerable expense, to involve women in all domains of work, even when it costs them extra time and training to extend to women members the necessary skill-sets to complete certain tasks. This conscious effort to demystify and disseminate knowledge (Rothschild & Whitt, 1986) affects the types of money-making businesses that Twin Oaks chooses to pursue, as all members, regardless of previous educational background, should, in their view, be able to acquire the knowledge and necessary skill-sets to work in all areas relatively quickly. Members are encouraged to work in each of the Twin Oaks industries and are trained as needed to perform the tasks involved. Currently, Twin Oaks’ main income generating businesses include hammocks, soyfoods, book indexing, and herbs. Besides Twin Oaks’ revenue generating businesses, members also perform most of the work necessary to maintain the community. This includes car and bicycle maintenance, constructing and maintaining buildings and growing foods for their own consumption. It also includes the education and care of their children. Samantha, a Twin Oaks member in her sixties, discussed how ‘‘the tradition and thinking at Twin Oaks has been to sacriﬁce efﬁciency (if need be) for equality.’’ She described circumstances where Twin Oaks ‘‘took the

252

JOYCE ROTHSCHILD AND AMY TOMCHIN

initiative to ﬁght gender gaps in all work areas – even at considerable expense’’ to the ﬁnancially struggling community. In one example, women were asked to consider volunteering to lay a concrete ﬂoor for a new dairy barn. None of the women knew how to lay a concrete ﬂoor, but they were offered training and guidance by members with experience in laying concrete ﬂoors, and according to Samantha, ‘‘the group ended up being entirely made up of women with no previous experience.’’ People say that the ﬂoor was laid perfectly. Though this occurred some time ago, it was evident that Samantha still takes pride in this achievement. In another example conveyed by Samantha, some of the women at Twin Oaks took over the auto-repair shop. ‘‘Initially, women tried apprenticing with men in the shop but found that they were not learning much more than the basics, such as how to change the oil or repair a ﬂat tire.’’ The community decided that it was more beneﬁcial if these women studied a particular car model intensively so they would know exactly how to repair it. The community agreed to purchase one car model, selling others and buying this one type of car for their ﬂeet. This plan was carried out despite the considerable cost to the community. For a good amount of time, according to Samantha, ‘‘women, exclusively, ran the auto shop and serviced the car ﬂeet. A while later Christina joined Twin Oaks. Since she had worked as a mechanic back in Denmark, and knew more about cars than most people at Twin Oaks, including the men, she became the manager of the auto repair shop.’’ While Twin Oaks’ 18 car ﬂeet is no longer maintained solely by women, the stigmatization that women could not ﬁx cars was effectively removed by women’s tenure in the auto repair shop, which Christina reinforces to this day. As stated earlier, members may choose to work on whatever work tasks they prefer. The one exception to this is the mandatory ‘‘K’’ shift. All adult members at Twin Oaks must do a ‘‘K’’ shift, or a kitchen shift, once a week. This means that they must clean the large, industrial style kitchen that feeds over a hundred people two meals a day, unless they have a health exemption. No one appears to be particularly fond of this work, but both male and female members alike realize its necessity and participate in the cleaning. Besides the required kitchen work, work is left to preference. Each adult member of Twin Oaks is expected to work 42 h week. Individual members create their own work plan: their speciﬁc combination of, for example, hammock work, childcare work, indexing work, kitchen work, and say, garden work. Not one type of work activity counts more than any other. All adult members of Twin Oaks, on a weekly basis, prepare a ‘‘labor sheet,’’ indicating what combination of work activities they wish to pursue in the coming

Can Collectivist-Democracy bring Gender Equality?

253

week. This sheet goes to the ‘‘Labor Assigner,’’ a job that is rotated among four members, each of them doing it once a month. It is the responsibility of the labor assigner to make sure that individuals’ preferred work duties ‘‘match up with the work that needs to be done.’’ If a necessary job does not get enough takers, then it is the job of the labor assigner to ﬁnd additional people to do that task, often drawing from a list they keep of people willing to do certain jobs on an ‘‘as needed’’ basis. If a certain work activity is consistently over-chosen by members, that area of work or business may expand to utilize all of the interested people. Some work may be chronically under-chosen, particularly we are told, work that is physically demanding like milking the cows and tofu production. In those cases, the manager of the work area simply works ‘‘extra hard’’ to attract more members to work in that area. Shirking, we are told, is unusual at Twin Oaks, but if an individual is not contributing quality work in their work activities, they would likely get feedback from the manager of the work area on how they might improve their work. ‘‘In an extreme case, the manager would ask the person to stop working in their area’’ (private correspondence, September 30, 2005). So long as the member’s work activity adds to 42 h, the individual is considered to have met their economic responsibilities. Some members perform the same job everyday and others constantly change their tasks and schedules, something that members told us was ‘‘a big plus’’ of work life at Twin Oaks that is not possible in most mainstream working environments. Philosophically and as an expression of their common values, everyone’s work accomplishments are formally viewed as equally contributing to the community. Members claim that everyone, regardless of educational background, race or gender, is to be treated as equally productive and valuable, and any efforts to raise one person while putting down another would be discouraged. Work is credited through what members term the ‘‘communist system,’’ which means that work activities are freely chosen on the basis of preference and that job performance is assumed and based on workers’ honor. The system they have worked out credits each type of work equally with every other. Invidious comparisons between different individual’s outputs are discouraged. All members are assumed to be contributing as much as they can. One hour of work is awarded one labor credit, regardless of how much or what is actually produced in that hour. For instance, an hour of car repair is awarded the same as one hour of childcare or one hour of book indexing. Different types of work are not differentially credited or valued, and consumption is virtually entirely collectivized at Twin Oaks, meaning that no ‘‘wages’’ or ‘‘salaries’’ are paid to anyone, but all members receive housing, food, clothing, health care and other communal purchases for their needs.

254

JOYCE ROTHSCHILD AND AMY TOMCHIN

While work assignments are, in the main, based on personal preference, the critical reader will want to know in what proportions men and women end up doing the various work tasks at Twin Oaks or to what extent the sexes sort themselves into a conventional division of labor. We have already noted by way of example that many women at Twin Oaks have been involved in auto repair and in laying a concrete ﬂoor, that everyone does an equal shift in the kitchen, and that many men at Twin Oaks are involved in child care. We could not determine the exact gender proportions in each activity at Twin Oaks, partly because these could shift every week. Overall, however, it was our observation that many members of Twin Oaks are sensitive to the gender segregation that persists in mainstream society and wish to avoid this in their own division of labor. What we were told and what we observed is that, by and large, both men and women are involved in virtually all of the work activities at Twin Oaks. In the words of one of our informants: Generally we are much more ‘gender-blind’ in work than the mainstream y . Many jobs here break down fairly well, with both genders being involved and active, more according to personality and interests than gender. The good news is we do pay attention to gender balance in our various work areas. The less-good news is that we all suffer from ‘cultural hangovers,’ and there are certain areas like plumbing, electric, etc, that most often are done by men. However, our auto mechanic and manager for the last 12 years is a woman. In terms of childcare, fathers are generally equally involved in their kids’ care. y To summarize, women and men share various work roles relatively well and evenly, and at the same time there are some cases in which members revert to more mainstream patterns of work choices. (private correspondence, September 30, 2005)

The bottom line seems to be this: while Twin Oaks has not achieved perfect gender equality with its division of labor, and while members of Twin Oaks do openly acknowledge and discuss their need to shed gender-based conditioning from the outside environment, they have come a very long way toward integrating both women and men into all work activities and toward valuing all work activities equally. There is a collective assumption that members will, in fact, perform their various jobs in a credible fashion. However, members acknowledge, when questioned, that certain individuals may take advantage of the system on occasion, but they believe that the majority of members respect the system and make it work. According to one of the founders of Twin Oaks (Susan) ‘‘If 85% of the people do 85% of the work, then 85% of the time the system works.’’ This represents their philosophy, and in the main, the members expressed the view that their system is fair. This is a long distance obviously from the unfairness and indignities perceived daily in the conventional

Can Collectivist-Democracy bring Gender Equality?

255

capitalist workplace (Hodson, 2001). Now and again, a member may make a case for seeking an exception to the system. For example, Susan, one of the founding members of Twin Oaks and the last one living there, believed strongly that in light of her age and seniority, she should be relieved of the work requirement at Twin Oaks. The group did not agree. Instead, the group decided to reduce her work obligation from the 42 h expected of all adult members to 30 h week. Dissatisﬁed with this outcome, Susan left the community, but in the process of making this decision, strong emotions were aroused on all sides and unanimity could not be reached. Some accused Susan of trying to use her age and history at Twin Oaks to get special privileges and to avoid compliance with policies that were designed to be borne by all. Susan continued to feel justiﬁed in her claim. In other words, not all differences have been resolved, and the question of how elderly members will be treated with respect to work obligations remains a tension. While Twin Oaks has rejected the notion of traditional ‘‘women’s work,’’ just as importantly, it has tried to remove the stigma that men in mainstream society may face if they wish to perform ‘feminine tasks.’ Twin Oaks has worked on both sides of the equation as they have sought to eliminate all gendering in the division of labor. This was evident at a visitors’ social when members were being introduced to the visitors. Some of the members had baked treats for the occasion. Robert, a man in his mid-60s, apologized to the group for his freshly baked lemon bars, stressing how much the taste had suffered from the absence of fresh lemons. Cooking is a sought after position at Twin Oaks and men take pride in their meal preparation. Gray, a man in his mid to late 50s, told our group to be sure to come early to Tuesday night’s dinner for the ﬁsh and chips that he has been preparing once a week for many years. Sam, an artist and manager of the bicycle repair shop at Twin Oaks, talked openly to us about his personal struggles with ‘sexism’ that were with him when he ﬁrst joined Twin Oaks. He talked particularly about his difﬁculties in being passed up for positions in the garden and on the farm for women. Sam emphasized that these experiences contributed to his own ‘‘personal growth and an understanding of women’s situation in mainstream society.’’ Sam is now a ‘‘primary’’ for a two-year-old girl, Fireﬂy, which means that he spends several hours a week with her in a one-on-one interaction – something that he never thought he could do. There is one regular men’s discussion group at Twin Oaks that has been continuing, it is estimated, for some 10 years now. This group meets on Sundays to talk about ‘‘personal issues, joys and struggles.’’

256

JOYCE ROTHSCHILD AND AMY TOMCHIN

Sam was interested in teaching the researcher all about repairing bicycles, even though the researcher confessed that she had no ability in this regard. Sam persisted, taking time to explain details on the bike he was repairing. To the researcher this seemed like a big expenditure of time on a person who might have little future use for this newly gained knowledge, but Sam assumed that of course she would want to know about bike repair and of course it was his duty to explain the process. Twin Oaks members are encouraged to share knowledge of their job or skill set with anyone who is interested in learning. Both the teaching and the learning side appear normative at Twin Oaks. The demystiﬁcation and dissemination of knowledge is one of the keys to equality, and a deﬁning characteristic of the collectivist-democratic organization, even if it slows present production (Rothschild, 1979), and Twin Oaks members were clearly willing to sacriﬁce present production in order to achieve a demystiﬁcation of the knowledge needed to perform their work tasks. Presumably, this level of cross training adds to the future productivity and ﬂexibility of the organization, but no members mentioned this. In another observed work experience of interest, Twin Oaks member, Sarah, manager of the dairy processing venture, was eager to explain everything to us—the processes used for making the different kinds of cheese, the different tools and machines used in the process, what types of cheese are made at Twin Oaks and how often, even though this was to be the only day that the researcher would work in this particular area. She asked the researcher to read through the binder that contained all the recipes used for the different cheeses, even though we were scheduled to make only one type of cheese on this day. Sarah encouraged the researcher to carry out most of the cheese-making process on her own which meant that for several hours her arms were submerged in a giant vat, stirring very warm milk and breaking up lumps as it curdled into cheese. The minimization of differentiation is an important dimension of Twin Oaks’ work culture and structure. At Twin Oaks there is a strong emphasis on task sharing, demystiﬁcation of skills and job rotation, as one would expect in any collectivist democratic organization. At Twin Oaks, the position of manager is voluntary. It is not a direct function of age, experience, education or social status. Indeed, most members manage at least one area where they oversee projects, budget, and make work process decisions. According to Sam, women at Twin Oaks can achieve more than in mainstream society because they are ‘‘not held back by typical political or social limitations and expectations of them.’’ Sam illustrated this point by using the example of the

Can Collectivist-Democracy bring Gender Equality?

257

dairy manager, a 21year-old woman who began to manage the dairy at the age of 19. He noted that she had never been to college. The position of manager for all areas of operation is strictly voluntary and does not carry the same prerequisite qualiﬁcations, rewards or prestige associated with the same position in a capitalist organization. A Twin Oaks manager will oversee the work schedule, the operational budget, and make sure that operations are running smoothly by addressing any problems that arise. Because managerial positions offer no extra beneﬁts but do require greater responsibility, those who ﬁnd them too challenging or are uncommitted to the role will generally step aside within a short time. As such, competition for managerial positions is not intense. Any conﬂict that could potentially arise over who gets to fulﬁll such positions would, we are told, be resolved by the current manager in a way consistent with Twin Oaks’ egalitarian values, but we did not observe any conﬂicts of this sort. Since there is no ofﬁcial hierarchy within the work structure at Twin Oaks, there is very little personal advancement in the way that most people measure it. The ‘‘career ladder’’ in most collective-democratic organizations is ﬂat, and cannot typically be used to further one’s career as it can in a bureaucratic organization (Rothschild & Whitt, 1986). Incentives to work are obviously not related to pay at Twin Oaks, and could not be, because the full product of labor and all living expenses have been collectivized. Nevertheless, incentives to work are both ideal and material, such as the feeling of pride one gets from a job well done, control over one’s work situation, a feeling of contributing to the collective well being of the community, job security, as well as beneﬁts that can be measured in material gain, like trips to conferences, invitations to events, and the capacity to help earn a livelihood for the whole community. Because members can directly see their contribution to the community, this is sustaining for most. At Twin Oaks, many members speak with appreciation of their ability to set their own work pace and hours. During the researchers’ stay at Twin Oaks, numerous people seemed to be just sitting around talking during regular work hours. To an outsider indoctrinated into what are usually viewed as ‘‘normal business hours,’’ this can look suspicious. However, members explained that the spacing out of their workweek over seven days allows people more free time during each day. Thus, people can and do choose different hours in which to do their work. Members plainly like creating their own work schedules, and one can imagine that this sometimes could add to the efﬁciency of their operations. At other times, given the mainstream hours of their suppliers and customers on the outside, they have

258

JOYCE ROTHSCHILD AND AMY TOMCHIN

to adapt to those conventional hours. We were told that many individuals do follow a workday routine that conforms to that of mainstream society, and because members are almost always close by, even when not at work, if their presence at their job is required they can usually be reached. Language too plays a role in equalizing relations at Twin Oaks. The Twin Oaks bylaws explicitly state that members ‘‘will not use titles of any kind among us’’ (Kanter, 1972, p. 22) and all members are referred to by their ﬁrst names. Members may choose to take on new names when they join Twin Oaks, as their joining is seen as a rebirth of sorts. Renaming is used both to help people to fashion the identity they seek as well as to discourage duplicate names among members. At Twin Oaks, at any given time, there is only one member called by a certain name. In addition, most Twin Oaks members use the preposition ‘‘co’’ to replace ‘‘him’’ and ‘‘her’’ (he/she) as a gender-neutral alternative for gender speciﬁc pronouns. A non-gender speciﬁc style of dressing is also evident at Twin Oaks, where each member wears what he or she would like regardless of traditional gendered categorizations. Since some clothing is held communally, in a room referred to as ‘‘Commie Clothes,’’ gender-neutral standards for dressing further minimizes gender differentiation. For example, during her stay the researcher noticed a man who always appeared in skirts and what looked like women’s clothing. No one else seemed to ﬁnd his dress in the least way odd. However, since Twin Oaks members often venture into mainstream society, those who dress in non-traditional clothing could be met with suspicion or harassment by outsiders, and they are aware of this. This presents yet another area in which members end up adapting to outside standards. The community recognizes the tension between Twin Oaks’ clothing ideology and mainstream expectations of gender speciﬁc dress, as their website states, ‘‘We all wear clothing we ﬁnd comfortable – pants for working in, skirts for coolness or festivity, heedless of mainstream ideas of ‘‘acceptable’’ fashion choices’’ (http://www.twinoaks.org).

CONCLUSION: IMPLICATIONS OF GENDER EQUALITY IN COLLECTIVIST-DEMOCRATIC ORGANIZATIONS This paper discusses the story of one collectivist democratic organization that has been concentrated and comprehensive in its efforts to create gender equality. While this case has the limitations inherent in any single case study, we think the reader will ﬁnd our investigation into Twin Oaks instructive

Can Collectivist-Democracy bring Gender Equality?

259

precisely because it is so comprehensive in its approach to building a genderequal organization that is decidedly anti-capitalist and anti-bureaucratic in its division of labor. Twin Oaks is a collectivist democratic community that contains quite a number of productive activities – from hand weaving hammocks to making cheeses, from indexing books for various publishers to repairing cars for themselves – all of which they formally value equally. On each of the eight characteristics of collectivist democracy set forth by the ﬁrst author in 1979 (Rothschild, 1979), Twin Oaks is a collectivist democratic organization. In its ideology and its practices, it manifests and in some ways exempliﬁes these eight ideal typical features. We began this article by pointing to the considerable scholarly literature that has found feminist organizations that have developed collectivist, consensus-oriented forms of decision-making and that has suggested an afﬁnity between feminist beliefs and participatory-democratic organizational forms (Iannello, 1992; Ferguson, 1984; Acker, 1990; Leidner, 1991; Staggenborg, 1989; Reinelt, 1994; Matthews, 1994; Mansbridge, 1980). Other researchers have suggested that gender equality does not necessarily arise even in feminist organizations (Kleinman, 1996) and that a feminist ideology may be unlikely to produce highly egalitarian, ﬂat, and democratized organizations (Bordt, 1997). In this study of Twin Oaks, we chose to focus our attention on the division of labor rather than on decisional processes. By turning our attention to an understudied aspect of these organizations, their work system or divisions of labor, we have found in Twin Oaks a case where feminist and collectivist ideology very much do support the development of an innovative work system that is radically egalitarian in the way it values people and their work. Most important for our purposes in this paper is the recognition that by employing a collectivist democratic form of organization, Twin Oaks has achieved an impressive level of gender equality. Men at Twin Oaks perform, on a daily basis, jobs that in the surrounding society would be considered ‘women’s work,’ while women at Twin Oaks perform many jobs that in the surrounding society would be considered ‘men’s work’. They do this without stigma, without resistance, indeed, with group encouragement and support. In stark contrast with the society that surrounds them, all of their work activities are formally valued equally and practically all of their consumption has been collectivized in a work system of ‘labor credits’ that they ﬁrst developed 40 years ago and that is still in use today. This is the key to the pronounced equality they have achieved inside their community.

260

JOYCE ROTHSCHILD AND AMY TOMCHIN

Precisely because Twin Oaks has gone so far to eliminate class and gender inequities, we can glean from their experience some of the tensions, or downsides, that come to the fore in radically egalitarian organizations such as theirs and that are not visible in the radically polarized context in which most of us live. Consider the following three tensions, evident at Twin Oaks, but rarely seen in the capitalist society that surrounds them. First, there has been much discussion in feminist scholarship of the devaluation of women and of women’s work that has taken place, as a result of shunting women to a ‘private sphere’ of household and childcare, while leaving men to work in the ‘public sphere.’ As we have seen, Twin Oaks has successfully combated this institution by altering family, childcare and living arrangements such that men are just as likely to be involved in household and childcare work as women and such that childcare is seen as a nongendered job of equal value to all other jobs in the community. This has brought with it, however, a tension not even considered in the feminist scholarship on this matter. Namely, as household and childcare work have been pushed into the public realm at Twin Oaks, enabling them to publicly recognize this work as work that is equal in value to every other type of work and to commit public resources to the household, the decision of what resources to commit to a household inevitably became a public issue. Speciﬁcally, the question of whether to approve the birth or adoption of a child became a community and very public question as soon as Twin Oaks decided that bringing up children was indeed a community responsibility. Twin Oaks recognized this from the beginning and included such provision in their bylaws. People in capitalist contexts where household and childcare labor are bourn privately might well consider such a community decision process quite an infringement on their autonomy. Many people in our society would not want to submit such private decisions for group approval. But, as the story of Twin Oaks illustrates, where the responsibility for child rearing becomes publicly bourn and collectively ﬁnanced, this not only raises the social value of the work that goes into raising a child, but it also moves the decision of child birth/adoption to the public realm. Second, a community or organization that puts equal relations ﬁrst and foremost cannot easily accommodate individual differences. Recall the case of Susan who pressed for, but did not receive, a reduction in her work obligation to zero. She ended up leaving Twin Oaks over this conﬂict. Some of the people at Twin Oaks felt badly about this outcome, but they could not see a way out of the bind they were in. The whole ethos and logic of their labor credit system is that all work activities will be regarded as equal in value to the community, and all members will do the same amount of labor.

Can Collectivist-Democracy bring Gender Equality?

261

To make exceptions is to get into the business of granting special privileges and this would defeat the logic of their whole system. If a system is going to be truly committed to the equal standing of all, it can make exceptions for whole categories of persons (like children who are not expected to work at Twin Oaks) but not easily for individuals. Third, Twin Oaks exists in an environment that is plainly not committed to egalitarian values or institutions. As described earlier in this paper, they have to make accommodations to their environment in many ways, small and large. The largest pressure from the environment is that Twin Oaks must sell their goods and services into a market outside the organization in order to survive economically. Inside the community, they can and do make decisions that advance gender and class equality (like taking the time to demystify and share the knowledge needed to perform the organization’s many work tasks, like fully crediting the childcare work that people do and so forth) but these very practices may lessen the product that Twin Oaks has produced to sell to the outside world. In other words, some of the practices that advance collectivist democracy and equality detract from the material standard of living at Twin Oaks. The result is that they live at a much more meager level of income than the average in our society. This is a fundamental dilemma for any collectivist democratic community in a capitalist context. Twin Oaks must interact all the time with an outside market and it must depend on this market for part of its sustenance. Although Twin Oaks, like any organization, cannot entirely insulate itself from the capitalist society in which it exists and from which most of its members arrive, these outside inﬂuences have not overwhelmed or distracted Twin Oaks from its oppositional purpose. While contradictions with outside institutions carry the potential to distort or undermine Twin Oaks’ ideology, Twin Oaks presents us with a worthy example of a collectivist-democratic organization that has been able to sustain its communalistic and egalitarian practices and ideology for almost forty years. In the main, Twin Oaks has succeeded in establishing an alternative to hierarchy where men and women are equal in the work they perform and in the social value accorded to their labor. Even what the outside world regards as ‘private labor,’ reproduction and childcare, has moved into the public realm at Twin Oaks, at once raising its social value and subjecting it to public discussion and decision making. In describing what the effort to build gender equality at Twin Oaks has entailed, we also hope we have encouraged the reader to consider some of the tensions this has brought forward and how very different they are from the structural tensions implied in the gargantuan inequalities wrought by capitalism.

262

JOYCE ROTHSCHILD AND AMY TOMCHIN

ACKNOWLEDGEMENTS The authors gratefully acknowledge the valuable comments and suggestions of our colleague, Max Stephenson and of the Editor, Vicki Smith, in improving our paper.

REFERENCES Acker, J. (1990). Hierarchies, jobs, bodies: A theory of gendered organizations. Gender and Society, 4(2), 139–158. Beckwith, K. (1986). American women and political participation: The impacts of work, generation, and feminism. New York: Greenwood Press. Bordt, R. (1997). The structure of women’s nonprofit organizations. Bloomington: Indiana University Press. Bryson, L. (1996). Revaluing the household economy. Women’s Studies International Forum, 19(3), 207–219. Bunch, C., & Carillo, R. (1990). Feminist perspectives on women and development. In: I. Tinker (Ed.), Persistent inequalities (pp. 312–321). Oxford: Oxford University Press. Denzin, N., & Lincoln, Y. (Eds) (2003). Collecting and interpreting qualitative materials, (2nd ed.). Thousand Oaks, CA: Sage Publications. Ferguson, K. (1984). The feminist case against bureaucracy. Philadelphia: Temple University Press. Fogiel-Bijaoui, S. (1992). From revolution to motherhood: The case of women in the Kibbutz, 1910–1948. In: D. Bernstein (Ed.), Pioneers and homemakers: Jewish women in pre-state Israel (pp. 211–233). Albany, New York: State University of New York Press. Hartsock, N. C. M. (1997). The feminist standpoint. In: L. Nicholson (Ed.), The second wave: A reader in feminist theory (pp. 216–240). New York: Routledge. Hodson, R. (2001). Dignity at work. Cambridge, MA: Cambridge University Press. Iannello, K. (1992). Decisions without hierarchy: Feminist interventions in organizational theory and practice. London: Routledge. Kanter, R. M. (1972). Commitment and community. Cambridge, MA: Harvard University Press. Kleinman, S. (1996). Opposing ambitions. Chicago: University of Chicago Press. Leidner, R. (1991). Stretching the boundaries of liberalism: Democratic innovation in a feminist organization. SIGNS, 16(2), 263–289. Mansbridge, J. (1980). Beyond adversary democracy. New York: Basic Books. Matthews, N. (1994). Confronting rape: The feminist anti-rape movement and the state. New York: Routledge. Reinelt, C. (1994). Fostering empowerment, building community: The challenge for statefunded feminist organizations. Human Relations, 47(6), 685–705. Rothschild, J., & Whitt, J. A. (1986). The cooperative workplace: Potentials and dilemmas of organizational democracy and participation. Cambridge, MA: Cambridge University Press. Rothschild-Whitt, J. (1979). The collectivist organization: An alternative to rational bureaucratic models. American Sociological Review, 44(August), 509–527. Staggenborg, S. (1989). Stability and innovation in the women’s movement: A comparison of two movement organizations. Social Problems, 36(1), 75–92. The Federation of Egalitarian Communities website. http://thefec.org/. Accessed November 5, 2004. Tomaskovic-Devey, D. (1993). Gender and racial inequality at work. Ithaca, NY: ILR Press.

DEMOGRAPHIC AND ENVIRONMENTAL INFLUENCES ON THE DIFFUSION OF CHANGES AMONG ISRAELI KIBBUTZIM, 1990–2001 Raymond Russell, Robert Hanneman and Shlomo Getz ABSTRACT This study examines the diffusion of 34 innovations among Israel’s 240 nonreligious kibbutzim from 1990 through 2001. The changes involve transfers of the authority of the general assembly to independent boards of directors and specialized committees or experts, privatization of consumption, and increasing inequality in compensation. We track year-to-year transitions among six relationships toward each innovation: not considering, rejected, discussing, decided to adopt, implementing, and using. Single-year transitions from ‘‘not considering’’ to ‘‘using’’ are relatively rare. Most innovations go through periods of discussion or implementation before being adopted. Innovations face substantial risks of being rejected at every stage. At each stage, acceptance of innovations by other organizations increases the likelihood of acceptance, implementation, and Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 263–291 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16010-0

263

264

RAYMOND RUSSELL ET AL.

retention. The effects of organizational size and age on innovations are not what classic theories of the ‘‘degeneration’’ of democratic workplaces predict. Recent changes in the kibbutzim appear instead to be an institutionalized response to market shocks.

INTRODUCTION Since the nineteenth century, social reformers have criticized the hierarchy and inequality of capitalist workplaces, and have called for the formation of democratic organizations of various kinds. This impulse has fed the formation of producers’ cooperatives, agricultural communes, and similar experiments in many countries. The Israeli kibbutz has long stood out as one of the most far-reaching and most successful of these experiments in the democratization of social life. Legally, Israel’s kibbutzim are registered as cooperatives. Like other cooperative organizations, kibbutzim aspire to be democratic in decisionmaking, and egalitarian in the sharing of income. Kibbutzim differ from most other cooperatives, however, in several important ways. Most cooperative organizations engage in either production or consumption. The kibbutzim combine these two activities into what Buber (1958) referred to as a ‘‘whole cooperative:’’ kibbutz members both live and work in their kibbutz. Many other cooperative organizations distribute ownership, income, and decision-making authority to individual members. Kibbutzim have preferred to share income and make decisions collectively. Both production and consumption have traditionally been under the control of a general assembly, in which all members have an equal voice, and leadership positions have been rotated. Finally, most other cooperatives link the income of members to the number of hours worked or units of output produced. The kibbutzim have rejected the use of material incentives as rewards for work, and have instead shared incomes on the basis of the socialist ideal of ‘‘from each according to his ability, to each according to need.’’ Since they ﬁrst appeared in 1910, the kibbutzim served as one of the most popular ways to settle Jews on the land of Palestine and later Israel. The collective ownership of the kibbutzim appealed to the Zionist and socialist ideals of many immigrants, while Zionist philanthropists like Arthur Ruppin saw the kibbutzim as an economical way to distribute resources. By May 1948, when the state of Israel was declared, the number of kibbutzim had grown to 177, and the kibbutzim and their members ﬁgured prominently in the new state’s economy, its government, and its army. Numerically, the

Demographic and Environmental Influences

265

kibbutz population has never been very signiﬁcant; the total number of kibbutz members and their children grew from 49,140, or 7.6% of Israel’s Jewish population, in 1948 (Near, 1997, Appendix 2) to 115,600, or 2.1% of Israel’s Jewish population, in 2002 (Pavin, 2005b). Economically, the kibbutzim have continued to play prominent roles. In 2001, they were responsible for 46.9% of Israel’s agricultural production, up from 38.9% in 1990, and 6.8% of its industrial production, up from 6.7% in 1990 (Pavin, 2005a). Like other cooperative workplaces, the Israeli kibbutzim have often been accused of making increasing use of nonmember labor (Simons & Ingram, 1997), and of increasingly oligarchical patterns of decision-making (Etzioni, 1958). As recently as 1990, however, the proportion of nonmembers in kibbutz factories remained below a third of the labor force (Russell, 1996), and few kibbutzim were willing to give up the norm of rotating managerial positions among the members (Getz, 1994). Although the democratic structures of most kibbutzim survived into the 1990s, by that decade a number of inﬂuences had begun to create pressures to modify many traditional practices. The growing salience of industrial and postindustrial activities made the economies of many kibbutzim increasingly complex, and increasingly dependent on specialized skills. Rising afﬂuence and individualism increased the importance of private households on kibbutzim (Ben-Ner, 1987), a development epitomized by the transfer of children’s sleeping quarters from centralized dormitories to their parents’ enlarged apartments. The signiﬁcance of these developments went unnoticed until 1985, when a sudden shift in governmental economic policy from expansion to price stabilization left many kibbutzim with too much debt, leading to the collective bankruptcy of the entire kibbutz movement. As the kibbutzim negotiated with the Likud-dominated government over new terms for their debt, they encountered little sympathy or respect for their traditions. Simons and Ingram (2003, 2004) have recently shown that even Labor-led Israeli governments treated the kibbutzim more like rivals than like allies since the 1950s. In the 1980s, under the inﬂuence of leaders like Margaret Thatcher, Ronald Reagan, and Mikhail Gorbachev, collective ownership was being subjected to calls for market reforms and privatization throughout the world. In Israel, the Likud acted as the chief spokesperson for such views, and promoted its own plans to privatize both state-owned and labor-owned alternatives to capitalist ﬁrms. In the late 1980s, thousands of kibbutz members themselves began to share the widespread impression that the kibbutzim were no longer economically viable in their traditional form. Many left their kibbutzim

266

RAYMOND RUSSELL ET AL.

(Ben-Rafael, 1997; Maron, 1998; Mort & Brenner, 2003), while others called for reforms from within (e.g., Harel, 1988). By the start of the 1990s, a consensus had emerged among the kibbutzim that they needed to change, and discussions shifted to the pros and cons of speciﬁc reforms. Before the decade was over, many kibbutzim had departed from their traditional practices in a number of signiﬁcant ways. For example, most kibbutzim transferred control of their industrial ventures from the general assembly to independent boards of directors. These boards, in turn, often hire professional managers from outside the kibbutz to run kibbutz ventures. Consumption on kibbutzim has been increasingly ‘‘privatized.’’ Some kibbutzim have closed their dining halls, while most others now charge members for meals they eat in them. As work and consumption become more differentiated and individuated, many kibbutzim are also experimenting with new forms of compensation that link individual effort with individual reward. As early as the nineteenth century, organizations that sought to provide more democratic or egalitarian alternatives to capitalism often stood accused of gradually transforming themselves back into capitalist hierarchies. J. S. Mill (1909, p. 790) complained, for example, that producer cooperatives have a tendency to ‘‘degenerate’’ over time. Marx (1967, p. 431) noted similarly that while these organizations represent ‘‘within the old form the ﬁrst sprouts of the new,’’ they ‘‘naturally reproduce, and must reproduce, everywhere in their actual organization all the shortcomings of the prevailing system.’’ In the twentieth century, the transformation of democracy into oligarchy and of communal settlements into more individualistic forms came to be viewed as inexorable consequences of universal laws. These include Michels’ (1962) ‘‘iron law of oligarchy,’’ the writings of Weber (1978) on direct democracy, of Meister (1984) on democracy in associations, and of Kanter (1968) and others on utopian communities. After having served for many decades as one of the major exceptions to these pessimistic laws, are the kibbutzim at long last succumbing to them? Parallels between the recent changes in the kibbutzim and classic discussions of degeneration, decommunalization, and oligarchy have often been noted (e.g., Ben-Rafael, 1997; Getz, 1994). While granting that the changes ‘‘have the potential, if implemented, to change the basic communal traits of the kibbutz’’ (Getz, 1994, p. 81), as of 1994, Getz was not yet ready to acknowledge that such a change in identity had already occurred. In this paper, we examine the recent changes in the kibbutzim from the perspective of theories of degeneration, but look for evidence of other processes as well. We consider the possibility that the recent diffusion of

Demographic and Environmental Influences

267

innovations among kibbutzim may be driven by processes quite different from those contemplated by classical theories of degeneration, but familiar to contemporary sociologists. For example, to what extent can the recent changes in the kibbutzim be explained as responses by individual kibbutzim to changes in their resource environments, or in their institutional environments, or to differences in the incidence among them of organizational inertia? This paper attempts to answer questions of this sort, by examining processes that affect the diffusion of these recent innovations among the Israeli kibbutzim. It begins by providing detailed information about the changes, and about the ways in which they are being diffused. The next section reviews theories that have the potential to shed light on the adoption of these changes. Later sections describe our methods and ﬁndings, and discuss their signiﬁcance.

DIFFUSION OF INNOVATIONS AMONG ISRAELI KIBBUTZIM, 1990–2001 Table 1 presents data on the spread of 34 changes among Israeli kibbutzim over the period from 1990 to 2001. The percentage of kibbutzim reporting that they had adopted each of these 34 changes in each year from 1990 to 2001 is shown in the right-hand columns of Table 1. The frequencies of acceptance shown in this Table are based on responses to a mail survey of all Israeli kibbutzim conducted annually by the University of Haifa’s Institute for Kibbutz Studies (Getz, 1994, 1998). Israel had 266 kibbutzim during this period, of which an average of more than 200 responded to the survey in each year. Between 11 and 16 responding kibbutzim in each year were from the 18 kibbutzim afﬁliated with religious movements, which differ so greatly from the other kibbutzim that they are not included in this analysis. Changes are sorted into three categories, depending on whether they involve decision-making, consumption, or compensation. Of the changes in decision-making shown in Table 1, the ﬁrst three involve the delegation of some portion of the authority previously vested in the kibbutz general assembly to smaller and more specialized organs: an internal control committee, a representative council, or a board of directors. By 2001, each of these reforms had been adopted by between a ﬁfth and a third of all kibbutzim. The next ﬁve changes listed create new structures to administer kibbutz-owned economic ventures. The most popular of these changes have

268

Table 1.

Percentage of Responding Kibbutzim Reporting Adoption of Innovations by Year. 1990

Privatization of Consumption Pay for travel Pay for electricity Pay for mealsa Comprehensive budget OK to own or use private cara Pay for recreation Increase size of house at own expensea Canceling breakfast in dining hall

13.8 0.0 47.2 23.7

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

1.0 14.9

1.6 23.5

2.1 22.6

3.7 28.9

3.9 27.4

25.3 5.7

43.1 8.0

45.5 11.1 23.1

57.6 15.8 26.5

61.7 22.9 32.0

7.4 27.0 13.1 61.7 20.8 25.9

12.4 29.3 17.1 64.2 30.7 35.9

16.9 29.9 21.4 70.0 34.8 51.7

9.4

27.2

27.6

25.4 27.1 22.8 65.8 20.9 65.1 31.2 41.9 7.5 37.8 49.2

32.3 23.4 20.9 60.2 19.2 73.9 41.8 39.5 14.9 51.3 47.6

12.5

16.2

25.8

28.0

34.0

50.5

56.4

15.6 27.0 17.2 74.6 37.1 64.0 42.8 32.2 14.1 40.7 59.8

46.3 43.7 6.2 61.7 12.1 28.2

58.0 58.2 18.1 69.7 30.2 30.8

69.6 66.5 41.1

77.8 71.6 56.8

84.0 72.9 63.1

79.2 77.3 67.6

86.9 79.3 70.2

8.3

8.3

7.8

56.4 55.2 11.9 59.3 19.2 31.7 14.7 11.4

70.9 62.9 30.2

23.3

42.2 41.8 3.4 60.0 9.4 30.7

37.0 37.3 16.9 14.4

46.1 38.7 22.1 20.8

53.8 46.2 35.4 26.4

60.0 48.8 35.6 32.7

63.7 56.9 45.8 36.3

67.6 58.9 48.3 47.0

29.3 22.3 47.4

14.2

RAYMOND RUSSELL ET AL.

Differentiation in Structure and Human Resources Internal control committee 0.8 7.3 Representative councilsa Board of directors Board of directors in industry 18.8 Board of directors in agriculture 6.3 Human resources division Members working outside as a branch Hiring ofﬁce holdersa Nonmember as chair of board of community Nonmember as chair of board of business Hiring managers for businesses

1991

Inequality in Compensation Pension plan Payment for work on Shabat Pay for overtime or additional worka Payment for additional work in services Pay for ofﬁceholders in difﬁcult jobsa Payment for seasonal work in ‘‘gius’’ Additional budget due to seniority Payment of differential salarya Connection between work and budgeta Budget with differential salary componenta Number of kibbutzim responding to the survey a

9.2

11.7 1.2

14.1 1.6

20.7 3.6

27.2 4.5

35.4 5.1 3.0

45.9 6.4 4.8

57.8 9.0 8.5 19.6

62.6 9.5 7.5 23.5 6.9

62.1 12.7 19.9 30.8 9.1

68.3 19.8 25.9 38.2 9.0

22.1

35.2

42.9

52.4

64.6

71.4

72.1

68.1

0.8

2.6

6.0

2.6

5.7

10.6

10.8

19.9

25.7

73.9 27.4 24.2 19.1

82.4 37.2 30.8 28.9

84.7 43.4 35.6 33.2

0.8

1.0

2.2

0

3.1

2.8

24.6 27.3 7.0

30.8 31.6 16.5

32.0 30.7 23.9

12.2

14.9

14.9

2.8 135

210

191

2.1 203

1.0 4.1 204

0 6.1 196

3.5 11.9 1.0 8.4 3.5 221

15.8 4.4 10.7 7.3 225

21.7 5.7 16.7 8.7 225

203

213

Demographic and Environmental Influences

Canceling evening meal in dining hall Dining operated by a contractor Privatization of health servicesa Privatizing the laundry Close dining halla

207

Coded as a ‘‘radical’’ innovation.

269

270

RAYMOND RUSSELL ET AL.

been the formation of boards of directors to govern industrial ventures (60.2% in 2001), and the establishment of human resource divisions (73.9%). The last four changes listed involve hiring experts from outside the kibbutz to manage kibbutz-owned economic ventures or to chair their boards of directors. Most of these changes had been adopted by about half of all kibbutzim by 2001. The second category of changes shown in Table 1 involves privatization of consumption. A number of items that kibbutzim previously treated as collective goods, such as meals, electricity, and recreation, are now being treated as personal expenses that individual members must pay for. Kibbutz members are now free to own their own automobiles and to make improvements to their homes at their own expense. Simultaneous with this expansion of the role of private households in the kibbutz economy, many previously collectively provided services are being curtailed. Half or more of all kibbutzim now close their dining halls in the morning and/or evening, and smaller proportions have privatized laundry services and health care. In acknowledgement of the parallels between these reforms and similar developments in other countries, the kibbutz literature explicitly describes this set of changes as constituting ‘‘privatization’’ (Ben-Rafael, 1997; Getz, 1994, 1998; Near, 1997). The last set of changes shown in Table 1 creates new forms of inequality in compensation. The most widely adopted change has been the opening of personal pension accounts for each individual member. The purpose of these accounts is to assure members of a secure retirement even if the kibbutz they belong to should be forced to close its doors. More controversial and less widely accepted are proposals to provide extra payments for additional days or hours of work, or to provide higher compensation for more difﬁcult or more highly skilled work. Several aspects of the entire set of changes are worthy of note. First, these are not merely ‘‘changes’’ in the kibbutzim, they are ‘‘innovations.’’ An organization can ‘‘change’’ merely by growing older or larger, or by reverting to a previous state, but as Hage (1999, p. 599) notes, ‘‘organizational innovation has been consistently deﬁned as the adoption of an idea or behavior that is new to the organization.’’ While all of the changes listed involve behavior that is new to the organizations that adopt it, we label as ‘‘radical’’ those innovations that strike us as departing most signiﬁcantly from the traditional model of the kibbutz, and as having the greatest potential to alter the identity of the kibbutz. Second, because the kibbutzim and their traditions have become institutionalized, successful innovation among them requires both the

Demographic and Environmental Influences

271

‘‘deinstitutionalization’’ of a previous practice, and the ‘‘reinstitutionalization’’ of a new practice to replace it (Jepperson, 1991). In this regard it is worth noting that by 1990 or a few years thereafter, all of these changes had acquired many of what Strang and Meyer (1993) describe as ‘‘institutional conditions for diffusion.’’ Individual changes had acquired labels and advocates, and received prominent attention in publications of the kibbutz federations. Rosner and Getz (1994, p. 42) note that in the past changes arose in speciﬁc areas of speciﬁc kibbutzim, and ‘‘were the result of ‘grassroots’ processes independently initiated by kibbutz members.’’ In the 1990s, changes began to be introduced in clusters, with the help of consultants from the federations and ‘‘innovation teams’’ on individual kibbutzim (Near, 1997, p. 352). Rosner and Getz (1994, p. 42) conclude, ‘‘The treatment of change has become institutionalized. Many kibbutzim have set up committees whose job is to deal with changes, while in other kibbutzim existing committees perform this function. These bodies are not concerned with solving problems as they arise, as in the past, but with suggesting and initiating changes to the kibbutz.’’ Third, this set of changes includes both innovations that had by 2001 become widely accepted, and those that had not. In a review of research on diffusion, Strang and Soule (1998, p. 285) observe, ‘‘There is a strong selection bias in diffusion research, where investigators choose ultimately popular practices as appropriate candidates for study. Investigation of practices that few adopt would provide a more balanced picture.’’ In this study, we hope to learn as much from innovations that have not been widely adopted as we do from those that have. To shed greater light on the processes by which these innovations were diffused, the General Secretary of every kibbutz was asked to report the stance of the kibbutz toward each innovation in each year using the following response categories: not considering it, discussing it, rejected it, decided to adopt it, implementing it, using it. Table 1 shows the percentage of responding kibbutzim reporting that they were currently ‘‘using’’ the innovation. Table 2 goes beyond Table 1 in two ways. First, it notes not merely the presence or absence of innovations, but also the incidence of intermediate stages, such as periods of discussion or implementation. Second, it compares the response of a kibbutz regarding a speciﬁc innovation in one year to its response to the same question in the following year. With six response categories in each year, there are 36 possible combinations of responses from year to year. Of these, most previous studies of change in organizations have recognized only three: absence of the innovation in both periods (upper left corner), presence in both periods (lower

272

RAYMOND RUSSELL ET AL.

Table 2.

Transition Probabilities from Time One to Time Two Adoption Status.

At Time One

Rejected Not considered Discussion Decided to adopt Implementing Currently using

At Time Two Rejected

Not considering

Discussion

Decided to adopt

Implementing

Currently using

Number of cases

0.215 0.015

0.554 0.752

0.149 0.141

0.005 0.012

0.009 0.014

0.067 0.067

549 17,184

0.022 0.007

0.326 0.172

0.336 0.124

0.057 0.153

0.073 0.113

0.186 0.432

4,620 825

0.010 0.005

0.139 0.100

0.108 0.028

0.044 0.018

0.142 0.021

0.558 0.829

1,017 9,734

right), and transitions from absence to presence (upper right). In Table 2, the most frequent responses indeed are either lack of interest in both periods or use of the innovation in both years, as the earlier literature contemplates. But for organizations making the transition from ‘‘not considering’’ to ‘‘using,’’ single-year transitions are the exception, not the rule. Insofar as organizations reporting ‘‘no consideration’’ of an innovation in one year do not simply repeat this response in the following year, they are much more likely to report that they are discussing the innovation (14.1%) than that they have already adopted it (6.7%). Of the 33,929 pairs of observations reﬂected in Table 2, 6,462 originate in the intermediate statuses ‘‘discussing’’ (4,620), ‘‘decided’’ (825), and ‘‘implementing’’ (1,017). Reports of these intermediate statuses in one year by no means assure adoption in the next. Organizations that report discussing an innovation in one year are equally likely to report in the next year either that they are no longer considering the innovation (32.6%), or have accepted the innovation (31.6%), or are still discussing it (33.6%). Reports that an organization has decided to adopt an innovation or is in the process of implementing it lead to reports of actual use of the innovation in the following year only about half of the time. Even after innovations have been reported as adopted, they suffer high mortality, with only 82.9% reported as still in place in the following year. The pattern of growing acceptance of innovations shown in Table 1 is thus a net effect of a diverse set of contributory and contradictory processes. Adoption of innovations is frequently preceded by uncertain periods of deliberation and implementation, and may be followed by either retention

Demographic and Environmental Influences

273

or abandonment. The remainder of this paper is an effort to identify the principal inﬂuences that shape these processes.

RELEVANT THEORIES Degeneration, Regeneration, and Organizational Inertia A wide range of literature suggests that kibbutzim should become more likely to embrace the changes under consideration here, the older and larger they become. These include Weber’s theories of direct democracy and of bureaucracy (1978), theories of the relationship between size and differentiation such as those of Durkheim (1933) and Blau (1970), and classic analyses of the structural transformation of the kibbutz by Talmon (1972) and Cohen (1983). Within the literature on producer cooperatives, this is also the old idea that cooperatives ‘‘degenerate’’ over time (Blumberg, 1968). Batstone (1983) argues that periods of degeneration in democratic workplaces are often followed by periods of ‘‘regeneration,’’ during which members perceive their democratic traditions to be in decline, and make efforts to re-assert and formalize them. Degeneration followed by regeneration would produce a curvilinear effect for age on changes that take the workplaces away from their traditions, positive for low values of age, but negative at higher levels. Two studies of the use of nonmember labor in producer cooperatives have reported effects for age of this type. One of these studies dealt with French worker cooperatives (Estrin & Jones, 1992), the other with Israeli kibbutzim during 1951–1965 (Simons & Ingram, 1997). In contradiction to theories both of degeneration and of regeneration is Hannan and Freeman’s theory of ‘‘organizational inertia’’ (1984), which predicts that organizations grow less likely to change, the older and larger they become. In this study, it may be possible to reconcile the apparent contradictions among these theories, if we examine the consequences of size and age for the initial consideration and acceptance of innovations separately from their consequences for implementation and retention. Hannan and Freeman (1984) argue that organizations avoid change as they grow and age, because change becomes more difﬁcult and potentially disruptive. Organizational inertia theory may therefore be more relevant to the implementation of change than to initial consideration. Theories of democratic degeneration and regeneration, on the other hand, address the motivation for organizations to introduce change, and may therefore be more relevant to consideration and acceptance.

274

RAYMOND RUSSELL ET AL.

Resource Scarcity and Organizational Change Common to many treatments of change in organizations is the notion that organizations are most likely to abandon time-honored institutions and to consider illegitimate changes when economic hardships or other crises compel them to consider reforms (Hage, 1999; Kraatz & Zajac, 1996; Oliver, 1992; Tolbert & Zucker, 1996). The most common argument for this is that resource scarcities make time-consuming or costly traditions more difﬁcult to afford. In democratic organizations like the kibbutzim, an added stimulus for change is that democratic practices may be seen as having caused the distress. Two crises shook the kibbutzim in the 1980s. First, the collective bankruptcy of the entire kibbutz movement in 1985 left all kibbutzim increasingly dependent on their own self-generated resources to survive. Second, defections of disillusioned members and children in the years after 1985 left many kibbutzim unable to attract and retain a sufﬁcient number of members to keep their communities viable (Ben-Rafael, 1997; Maron, 1998; Mort & Brenner, 2003). In keeping with the previous literature, we expect both economic crises and demographic crises to make kibbutzim more likely both to consider and to accept proposed reforms. But economic and demographic crises may have different effects on the capacity of organizations to implement and retain innovations than on their readiness to consider and accept them. Previous studies indicate that organizations with ‘‘slack’’ resources are in a better position to bear the costs of innovations than are organizations in which material and/or human resources are scarce (Bourgeois, 1981; Burns & Wholey, 1993; Thompson, 1967). For such reasons, we expect measures of economic and demographic crises to be associated with higher frequencies of consideration and acceptance of innovations, but lower frequencies of successful implementation and retention. Effects of Adoption of Innovations by Similar Organizations We see the literature on organizations as supporting two predictions about the relationship between the proportion of all organizations using an innovation in a given year and the relative frequencies of the transitions shown in Table 2 over the following year. First, higher proportions of organizations using an innovation should reduce the frequency with which the remaining organizations give it no consideration, while increasing the frequency of all other transitions. At low rates of adoption, ‘‘knowledge of the

Demographic and Environmental Influences

275

structures among nonadopters y will be extremely limited y ’’ (Tolbert & Zucker, 1996, p. 182). Many actors will be uninterested in an innovation at this stage, because they do not know what it is, or cannot imagine how it might be applied in their own organization. When rates of adoption are high, an innovation can no longer be ignored, and an organization that still does not wish to use it should become more likely to make its negative assessment explicit by voting to reject. Higher frequencies of adoption should thus promote more talk about those innovations in other organizations, while also making it possible for organizations to accept without prolonged controversy innovations that they have not discussed in previous years. Even rejection and abandonment are likely to increase in frequency, because more votes are being held and more organizations that previously adopted the innovation have had a chance to have second thoughts. Our second hypothesis regarding this inﬂuence predicts that the percent of organizations reporting adoption of an innovation stimulates increases in the frequency of some transitions more strongly than others. For organizations that have not yet considered it, use without harmful consequences in other organizations argues in favor of the viability of the innovation. For organizations that have reached the stage of trying to implement the innovation, the example set by earlier adopters may encourage them to persevere, as well as providing solutions to common problems that can be exported to other organizations. Thus, higher acceptance of innovations not only publicizes innovations, it also legitimates and facilitates them. We therefore expect to ﬁnd that the percent of organizations currently using an innovation has positive effects on the frequency of transitions to adoption or in its direction, while having negative effects on the frequency of rejection and other transitions away from adoption. Hypotheses On the basis of these theoretical considerations, we offer the seven hypotheses listed below. The ﬁrst three reﬂect three views of the effects of size and age: classical theories of degeneration (Hypothesis 1), Batstone’s (1983) theory of regeneration (Hypothesis 2), and the theory of organizational inertia of Hannan and Freeman (1984) (Hypothesis 3). The next two posit contradictory effects for economic and demographic crises, predicting that they make kibbutzim more likely to consider and adopt innovations (Hypothesis 4), but less likely to complete their implementation or to retain them (Hypothesis 5).

276

RAYMOND RUSSELL ET AL.

The last two hypotheses are alike in predicting positive effects for the percentage of similar organizations currently using an innovation on transitions toward current use. They differ in what they expect for transitions away from use. In Hypothesis 6, adoption by other organizations makes organizations more aware of innovations, and therefore more likely to consider them, but its effect is only to call attention to innovations, and does not bias consideration in favor of acceptance of innovations. In the terminology of Strang and Soule (1998), Hypothesis 6 sees diffusion as due merely to mimicry, while Hypothesis 7 attributes diffusion to social learning. If Hypothesis 6 is correct, adoption will stimulate transitions away from adoption as strongly as it stimulates transitions toward implementation or current use. If Hypothesis 7 is correct, adoption by others reduces the frequency of rejections and other transitions away from adoption. Hypothesis 1. Organizations become more likely to consider and to accept innovations that increase differentiation and inequality, the older and larger they become. Hypothesis 2. The age of democratic organizations has a curvilinear effect on the consideration and acceptance of changes, positive over low ranges of age, but negative at higher values. Hypothesis 3. The older and larger the organization, the greater the likelihood that the organization will fail to implement or to retain innovations that were previously accepted or adopted. Hypothesis 4. Economic and demographic crisis make organizations more likely to consider and accept innovations. Hypothesis 5. Among organizations that have previously accepted innovations, economic and demographic distress increase the likelihood that innovations will fail to be implemented or retained. Hypothesis 6. The higher the percentage of similar organizations that are currently using an innovation, the lower the proportion of the remaining organizations that give it ‘‘no consideration’’ in the following year. Rising percentages of adoption thus increase the frequency of all transitions. Hypothesis 7. The percentage of similar organizations currently using an innovation has more strongly positive effects on the incidence of transitions toward adoption, and has more weakly positive or negative effects on transitions toward rejection.

Demographic and Environmental Influences

277

METHODS AND MODELS Detailed information about our measures and data is provided in the appendix. We note here only a few things that all readers need to know before attempting to interpret our results. Control Variables In addition to the ﬁve independent variables whose effects are featured in the seven hypotheses just presented, our models also include controls for a number of inﬂuences that on the basis of previous literature also seem likely to affect the consideration and implementation of these changes among Israeli kibbutzim. An important institutional constraint on the response of a kibbutz to each innovation is the federation it belongs to. In previous studies (Rosner & Tannenbaum, 1987; Simons & Ingram, 1997), kibbutzim afﬁliated with the more ideologically coherent Artzi federation have been found to be more faithful to kibbutz traditions in such areas as the avoidance of hired labor than the larger and more permissive Takam. Kibbutzim that adopt changes not permitted by their federations risk expulsion. We expect afﬁliation with the more traditional Artzi federation to have a discouraging effect on innovation at all phases. Kanter (1968) demonstrated that the longevity of past utopian communities in the United States depended in part on the isolation of those communities from contact with the external economy and society. Ben-Ner (1987) asserted that kibbutzim that are located near cities become increasingly integrated into the urban market economy, and adopt increasingly individualistic patterns of consumption in response. We interpret these works as suggesting that kibbutzim located at greater distances from cities are less likely to adopt innovations, because they are less likely to hear about them, and therefore are less likely to give innovations any consideration at all. Once they decide to adopt innovations, on the other hand, rural kibbutzim may be no less capable of implementing innovations than urban kibbutzim. Recent research on organizations has sought to differentiate ‘‘radical’’ or ‘‘divergent’’ changes from ‘‘convergent’’ ones. Unlike convergent change, which merely ﬁne tunes an existing organizational orientation, ‘‘Radical organizational change y involves the busting loose from an existing ‘orientation’ y and the transformation of the organization’’ (Greenwood & Hinings, 1996). ‘‘Divergent’’ changes, similarly, are changes that diverge

278

RAYMOND RUSSELL ET AL.

‘‘from accepted models, or abandon an institutionalized template for arranging y core activities and replace this template with a substantially different one’’ (D’Aunno, Succi, & Alexander, 2000). Radical or divergent changes, then, are changes that alter the identity of the organization that adopts them. This should make organizations more reluctant to consider and accept these changes, and should also make organizations that have accepted these innovations more likely to encounter difﬁculties in implementation, and less likely to retain them. In addition to the radical/nonradical dichotomy, innovations were also categorized by whether they dealt with decision-making, consumption, or compensation. These two measures accounted for a substantial portion of the differences among innovations. Since they have the advantage of being interpretable, we decided to retain these two categorizations in the ﬁnal model, and to use them instead of the full vector of dummy variables representing each innovation. Descriptive statististics for all independent and control variables included in this study are presented in Table 3. Modeling Transitions Below we present a series of models that analyze the effects of these variables on the probability that a kibbutz at a particular stage in the adoption process in a given year moves toward or away from implementing the innovation in the subsequent year. Models are estimated separately for each initial status. The models presented are multinomial logistic regressions, in which the generalized log odds of movement toward or away from implementation are expressed relative to the outcome of no change. Estimates by maximum likelihood are obtained from SPSS version 11.0 for Windows. For our analyses, data are formed as a ‘‘pool’’ of observations across years, innovations, and kibbutzim. Not all innovations were observed in all years for all kibbutzim. Some innovations were dropped in later years after they were widely diffused, while other innovations were added in later years. In addition, not all kibbutzim responded each year. As a result, it is not possible to apply standard mixed-model methods to control for nonindependence of observations. Additional information about the implications of this pooling of consequences is provided in the appendix. If all 240 nonreligious kibbutzim had responded to every question in every year in which it was asked, the cumulative total of observations would have reached 64,320. The actual total of observations available for these analyses is 33,983, a response rate of 52.8%. The loss of data due to each instance of

1 1. Size ( 100) 2. Age 3. Age squared ( 1000) 4. Economic crisis 5. Demographic crisis 6. Proportion accepting 7. Federation 8. Rural 9. Radical 10. Decision making 11. Consumption 12. Compensation

2

3

Descriptive Statistics for Covariates. 4

1.00 0.58 1.00 0.56 0.98 1.00 1.00 0.42 0.30 0.27 0.22 0.03 0.04 0.06 0.05 0.06 0.01 0.01 0.14 0.08 0.06 0.03 0.28 0.42 0.37 0.04 0.00 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 0.00 0.00 0.00 0.01 0.00

5

1.00 0.00 0.01 0.01 0.00 0.00 0.00 0.00

6

7

8

1.00 0.01 1.00 0.00 0.17 1.00 0.32 0.00 0.00 0.00 0.00 0.03 0.18 0.00 0.01 0.00 0.00 0.18

9

10

11

Mean

S.D.

2.34 1.33 51.44 14.39 2.85 1.43 3.08 1.41 0.00 0.22 0.29 0.22 0.33 0.47 75.24 56.06 1.00 0.34 0.47 0.19 1.00 0.33 0.47 0.11 0.60 1.00 0.25 0.43 0.34 0.40 0.50 0.43 0.49

Demographic and Environmental Influences

Table 3.

Note: Size is the number of members at the beginning of the spell of observation. Age is the number of years from the founding of the kibbutz to the beginning of the spell of observation. Economic crisis is measured at two time points on a scale from one to six. Demographic crisis is the change in membership in the year prior to the spell observed, expressed as a proportion of the value at the beginning of that year (direction changed) Federation is scored 1 ¼ Artzi, 0 ¼ otherwise. Proportion accepting is the percentage of all kibbutzim reporting an adoption status of ‘‘currently using’’ at the beginning of the spell of observation. Rural is the distance, in kilometers, from the kibbutz to the nearest place classiﬁed as urban. Radical change is coded as a dichotomy (0 ¼ incremental, 1 ¼ radical), (see text for discussion). Decision-making is coded as a dichotomy (1 ¼ innovation deals with differentiation of structure and human resources) (see text for discussion). Consumption is coded as a dichotomy (1 ¼ innovation deals with consumption practices) (see text for discussion). Compensation is coded as a dichotomy (1 ¼ innovation deals with consumption practices) (see text for discussion). po0.05 level (one-tailed). N ¼ 29,086 transitions.

279

280

RAYMOND RUSSELL ET AL.

nonresponse is high, because two observations are lost for each year that is missing, due to our need to organize observations into two-year pairs.

RESULTS Six possible responses in the ﬁrst year and six more in the second create a total of 36 possible transitions to model. Fortunately, our hypotheses do not hinge on the details of all 36 transitions. At each of the six possible initial conditions, our hypotheses ask only to what extent a given inﬂuence promotes transitions toward or away from adoption. Tables 4 and 5 present estimates of the effects of each inﬂuence on transitions in each of these directions, from each of these six starting points. Estimates of the effects of size are shown on the ﬁrst line of Tables 4 and 5. There is no evidence here that size promotes the consideration of these innovations, as in Hypothesis 1, or that it makes innovations less likely to be retained, as in Hypothesis 3. For kibbutzim that are already using an innovation, size signiﬁcantly reduces the probability that the innovation will be abandoned, rather than increasing it as predicted by inertia theory. The effects of age and age squared are shown on the second and third lines of Tables 4 and 5. Once again, the results for these variables do not match any of our hypotheses regarding them. Contrary to the predictions of both Hypothesis 1 and Hypothesis 2, neither age nor age squared has any Table 4.

Effects on the Probability of Moving Toward Use of Innovations. Status at Time 1

Size (100s) Age Age squared Economic crisis Demographic crisis Proportion using Federation Radical Decision making Compensation

Rejected

Not considered

Discussing

Decided

In process

In use

0.14 0.09 0.72 0.09 1.15 1.20 0.08 0.13 0.04 0.40

0.01 0.01 0.02 0.17 0.03 2.49 0.25 0.16 0.42 0.63

0.03 0.01 0.05 0.07 0.07 2.06 0.27 0.02 0.19 0.57

0.04 0.06 0.50 0.02 1.61 1.35 0.47 1.20 0.57 0.22

0.07 0.02 0.14 0.01 3.73 1.39 0.61 0.65 0.65 0.13

——— ——— ——— ——— ——— ——— ——— ——— ——— ———

Note: Standard errors available on request. Intercept and year effects not shown. po0.05 one-tailed.

Demographic and Environmental Influences

Table 5.

281

Effects on the Probability of Moving Away from Use of Innovations. Status at Time 1

Size (100s) Age Age squared Economic crisis Demographic crisis Proportion using Federation Rural Radical Decision making Compensation Nagelkerke R2

Rejected

Not considered

Discussing

Decided

In process

In use

——— ——— ——— ——— ——— ——— ——— ——— ——— ——— ——— 0.07

0.10 0.07 0.65 0.10 0.05 0.40 0.11 0.00 0.18 0.98 0.26 0.10

0.04 0.04 0.27 0.08 1.18 1.36 0.09 0.00 0.05 0.24 0.79 0.12

0.09 0.07 0.48 0.15 4.09 0.44 0.16 0.00 0.43 0.19 0.57 0.20

0.16 0.04 0.33 0.04 2.93 0.98 0.67 0.00 0.74 0.62 0.03 0.13

0.06 0.05 0.43 0.04 0.14 1.68 0.04 0.00 0.28 0.88 0.33 0.10

Note: Standard errors available on request. Intercept and year effects not shown. po0.05 one-tailed.

signiﬁcant effects on the probability of transitions toward use of the innovation. For kibbutzim that are already using an innovation, age makes organizations less likely to abandon the innovation, not more likely, as the theory of organizational inertia and Hypothesis 3 predict. Table 5 indicates that it is age squared rather than age that makes organizations more likely to abandon innovations that they have already put into use. This implies that in so far as organizational age may inhibit the retention of innovations, this occurs at high levels of age rather than low levels. The effects of economic and demographic crisis are shown in the fourth and ﬁfth lines of Tables 4 and 5. The measure of economic crisis has a number of effects that are generally consistent with the predictions of Hypothesis 4. For kibbutzim that have not previously considered an innovation, economic crisis signiﬁcantly increases the frequency of transitions toward use of the innovation (0.17), but also increases the probability of rejection (0.10). We interpret this to mean that economic crisis makes it more difﬁcult for organizations to ignore innovations and instead makes organizations more likely to consider each innovation and to vote it either up or down. For kibbutzim that were discussing an innovation in the initial year, economic crisis increases the probability of transitions toward use of the innovation (0.07), while reducing the probability of transitions away from use (0.08). Both of these ﬁndings are consistent with Hypothesis 4.

282

RAYMOND RUSSELL ET AL.

There is no sign in the tables of the negative effects of crisis on implementation and retention that were predicted by Hypothesis 5. The effects of economic crisis come entirely during the early stages of consideration and discussion. Our measure of demographic crisis shows no effects at these stages, but among kibbutzim that are in the process of implementing an innovation, demographic crisis promotes transitions toward adoption, not away as Hypothesis 5 predicts. The effects of the percentage of kibbutzim reporting that they had previously adopted an innovation at the beginning of each two-year observation are shown in the sixth lines of Tables 4 and 5. Insofar as adoption by other organizations does no more than to call attention to innovations, as Hypothesis 6 predicts, the percent previously adopting the innovation should have positive effects on transitions in both directions. Insofar as acceptance by other organizations also legitimates innovations, as Hypothesis 7 predicts, the effects of the percent adopting should be positive in Table 4, and either negative or not signiﬁcant in Table 5. The results ﬁt the pattern predicted by Hypothesis 7, not Hypothesis 6. The effect of the proportion of organizations already using the innovation on transitions toward adoption is statistically signiﬁcant in four of the ﬁve stages (not considering, discussing, decided, and implementing), and are positive in every instance. Its effect on transitions away from adoption attain signiﬁcance at three stages (discussing, implementing, and using), and are negative in each case. The seventh lines in Tables 4 and 5 show the effects of federation. As in previous studies, we ﬁnd that kibbutzim belonging to the Artzi federation adhere more closely to kibbutz tradition than those belonging to Takam. Among kibbutzim that had not previously considered an innovation, Artzi kibbutzim are less likely than others to take steps leading toward adoption. Among kibbutzim that are discussing innovations, have decided upon innovations, or are in the process of implementing innovations, Artzi kibbutzim are again less likely than others to take steps toward use. Once innovations have been put into use, however, Artzi kibbutzim are no less likely than others to retain them. Geographic isolation in rural locations has negative effects on innovation, as expected, but its effects are limited to the earliest stage. Among kibbutzim that have not previously considered innovations, distance from cities lowers the probability of transitions toward use. Once kibbutzim begin to discuss innovations or to implement them, location has no signiﬁcant effects. For organizations reporting in the ﬁrst year that they have not previously considered an innovation, radical changes behave as expected, being less

Demographic and Environmental Influences

283

likely than others to record transitions toward adoption in the following year. Once organizations begin to discuss or to implement or to use innovations, however, radical innovations are no less likely to be accepted or retained than others. Among organizations previously reporting that they had decided upon an innovation but not yet implemented it, radical innovations were more likely than others to record progress toward adoption in the following year (1.20). Once innovations have been put into use, radical innovations are less likely than others to be abandoned. It is tempting to interpret these ﬁndings as meaning that the decision to embark on a radical course of action somehow strengthens an organization’s commitment to completing the action. For organizations in the process of implementing decisions, the effects of radicalness are more ambiguous. At that stage, radical changes are both more likely to advance toward use, and more likely to be abandoned, which means that they are less likely than other changes to be at the stage of implementation for two years in a row. The ﬁnal two lines of Tables 4 and 5 estimate how changes involving decision-making or compensation differ from changes in consumption, which serve here as the contrast category. Where the effects of decisionmaking and compensation are of the same sign, this makes changes in consumption stand out as signiﬁcantly different from the other two categories. This occurs in three places. Changes involving consumption are less likely than others to be considered, but are also less likely to be abandoned after inital discussion, and are more likely to be retained. Among changes that are under discussion, changes involving compensation are more likely than others to record transitions both toward adoption, and away from adoption. This means that the principal effect of changes involving compensation at this stage is to cause organizations to move quickly through it, whether the direction of movement is forward or backward. Among changes that have are in the process of implementation, similarly, those involving decision-making are more likely than others to make transitions both toward adoption and away from it. Among changes that have recently been decided upon, those involving decision-making are least likely to make progress toward use.

DISCUSSION There is little support here for any of the theories that view the probability that a kibbutz will adopt the innovations discussed in this study as an effect of age or size. Theories of degeneration, of regeneration, and of organizational

284

RAYMOND RUSSELL ET AL.

inertia, have shown themselves to be equally unhelpful in explaining the incidence of these innovations within this population. Rather than being caused by size or age, we ﬁnd these transformations to be stimulated by economic crisis. If there is support here for any previous views of the fate of democracy in cooperative workplaces, it is for those of authors who attribute the decline of democracy in organizations not to internal factors like size or age, but to external market forces. Simons and Ingram (1997) have shown, for example, that the use of hired labor in Israeli kibbutzim has in the past risen or fallen in response to levels of dependency on banks. In the work of Sidney Webb and Beatrice Webb (1920), democratic workplaces are seen as abandoning their democratic practices, when they are persuaded through a process of organizational learning that they must do so in order to survive. They asserted, ‘‘All such associations of producers that start as alternatives to the capitalist system either fail or cease to be democracies of producers’’ (p. 29). Other accounts of the kibbutz crisis are consistent with this view that Israeli kibbutzim are introducing these innovations not out of any great enthusiasm for them, but out of a sense that they are compelled to make these or other changes if they wish to survive. Near (1997) complains, for example, of ‘‘a small group activists’’ in the federations who have ‘‘used the movement press and research institutions to advantage’’ and ‘‘have constantly reiterated the same basic theme: the crisis proved that without fundamental change the kibbutz could not survive’’ (p. 352). Regarding the effects of adoption by similar organizations, Hypothesis 7 predicts that adoption by others stimulates transitions toward adoption while having negative effects on transitions away, while Hypothesis 6 predicts that higher proportions adopting an innovation increase the frequency of all transitions. In Tables 4 and 5, the effects of the percentage of other kibbutzim previously accepting an innovation show the pattern predicted by Hypothesis 7 rather than that predicted by Hypothesis 6. The proportion using an innovation promotes transitions toward use at four out of ﬁve stages tested, and reduces transitions away from use at three out of ﬁve. This implies that acceptance of the innovation by similar organizations does not merely call attention to an innovation, but also increases its perceived attractiveness or legitimacy. While the proportion of other kibbutzim accepting an innovation operationalizes DiMaggio and Powell’s (1983) concept of ‘‘mimetic isomorphism,’’ the kibbutz federations illustrate their concept of ‘‘normative isomorphism,’’ with some elements of ‘‘coercive isomorphism.’’ Kibbutzim afﬁliate with federations that they perceive as sharing their values, but are

Demographic and Environmental Influences

285

thereafter required to follow their rules. Kibbutzim afﬁliated with the Artzi federation are signiﬁcantly less likely than other kibbutzim to make transitions toward the use of innovations at virtually every stage. In Tables 4 and 5, afﬁliations with federations are second only to the proportion of kibbutzim using each innovation in the pervasiveness of their inﬂuence. This means that institutional processes are indeed the predominant inﬂuences, as Rosner and Getz (1994) noted. But what kinds of institutional inﬂuences predominate? If examples set by other kibbutzim are more important than the federations, this means that mimetic isomorphism is more inﬂuential than normative. In the terminology of Greenleigh, Robert, MacFarlane, Bate, and Kyriakidou (2004), these innovations are being ‘‘diffused’’ from kibbutz to kibbutz more than they are being ‘‘disseminated’’ by the federations. Rural locations are like economic crises, inﬂuencing the initial consideration of changes, but having no inﬂuence on their later implementation or retention. Effects of characteristics of changes appear to differ from one stage to another. Radical changes are less likely to be considered, but are more likely to be put into use once they have been decided upon, and are less likely to be abandoned. Differences among changes involving decision-making, consumption, and compensation also differ from stage to stage.

CONCLUSIONS The classic Weberian view that direct democracy transforms itself into representative democracy in response to growing size and advancing age receives no support in this study. The sources of the current wave of innovations on Israeli kibbutzim are not internal, as in Weber’s model, but external. The decisive inﬂuences come not from the individual organization, but from the economy, and from the organizational ﬁeld. As in Stearns and Allan (1996) and Kraatz and Zajac (1996), it is not the most prosperous and most respected organizations at the core of this organizational ﬁeld that initiate innovations, but those that are most economically marginal. As growing numbers of such organizations adopt innovations, the example they set inspires other organizations to imitate them. Over the time period covered by this study, most innovations made by the kibbutzim were rather modest and were introduced in cooperation with the federations rather than in deﬁance of them. Although the kibbutzim were

286

RAYMOND RUSSELL ET AL.

changing, they were doing so in ways that maintained their coherence and structuration as an organizational ﬁeld. But because kibbutz-to-kibbutz diffusion was more inﬂuential over this period than dissemination, there appeared to be little to prevent the future spread from one kibbutz to another of ‘‘illegitimate’’ or ‘‘radical’’ changes. A case in point is the payment of differential salaries. After spreading slowly in the 1990s, this innovation began to be adopted rapidly in the new century. By 2004, 39% of kibbutzim had adopted differential reward systems, 12% were using combined systems, and only 49% were still adhering to their traditional systems of remuneration (Pavin, 2005b). As a result of these recent changes, the kibbutzim are now in danger of losing not only their identity, but also their unity. Contemporary descriptions of the kibbutzim (e.g., Gavron, 2000; Mort & Brenner, 2003) now portray them as conforming to multiple models rather than being of one type. Less than half of all Israeli kibbutzim currently match the characterization of a traditional kibbutz with which we opened this paper. About the rest we are forced either to conclude that in many cases, a kibbutz is no longer a kibbutz, or to expand our conception of the nature of the kibbutz to include the current changes. If we take the former course, we are left with a traditional kibbutz movement that has been greatly diminished in both numbers and inﬂuence. If the latter, we can assert that the kibbutzim are currently more numerous, more populous, and more economically productive than they have ever been, but must also acknowledge that they are also signiﬁcantly less democratic, less collective, and less egalitarian than they were in the past.

REFERENCES Batstone, E. (1983). Organizations and orientation: A life cycle model of French co-operatives. Economic and Industrial Democracy, 4, 139–161. Ben-Ner, A. (1987). Preferences in a communal economic system. Economica, 54, 207–221. Ben-Rafael, E. (1997). Crisis and transformation: The kibbutz at century’s end. Albany: State University of New York Press. Blau, P. M. (1970). A formal theory of differentiation in organizations. American Sociological Review, 35, 201–218. Blumberg, P. (1968). Industrial democracy: The sociology of participation. New York: Schocken. Bourgeois, L. (1981). On the measurement of organizational slack. Academy of Management Review, 6, 29–39. Buber, M. (1958). Paths in Utopia (R. F. C. Mull, Trans.). Boston: Beacon Press. Burns, L. R., & Wholey, D. R. (1993). Adoption and abandonment of matrix management programs: Effects of organizational characteristics and interorganizational networks. Academy of Managment Journal, 36, 106–138.

Demographic and Environmental Influences

287

Cohen, E. (1983). The structural transformation of the kibbutz. In: E. Krause (Ed.), The sociology of the kibbutz (pp. 75–114). New Brunswick, NJ: Transaction Books. D’Aunno, T., Succi, M., & Alexander, J. A. (2000). The role of institutional and market forces in divergent organizational change. Administrative Science Quarterly, 45, 679–703. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational ﬁelds. American Sociological Review, 48, 147–160. Durkheim, E. (1933). The division of labor in society. (G. Simpson, Trans.). New York: The Free Press. Estrin, S., & Jones, D. C. (1992). The viability of employee-owned ﬁrms: Evidence from France. Industrial and Labor Relations Review, 45, 323–338. Etzioni, A. (1958). The functional differentiation of elites in the kibbutz. American Journal of Sociology, 64, 476–487. Gavron, D. (2000). The kibbutz: Awakening from Utopia. Lanham: Rowman & Littleﬁeld. Getz, S. (1994). Implementation of changes in the kibbutz. Journal of Rural Cooperation, 22(1–2), 79–92. Getz, S. (1998). Winds of change. In: U. Leviatan, J. Quarter & H. Oliver (Eds), Crisis in the Israeli kibbutz (pp. 13–25). Westport, CT: Praeger. Greenleigh, T., Robert, G., MacFarlane, F., Bate, P., & Kyriakidou, O. (2004). Diffusion of innovations in service organizations: Systematic review and recommendations. The Milbank Quarterly, 82(4), 581–629. Greenwood, R., & Hinings, C. R. (1996). Understanding radical organizational change: Bringing together the old and the new institutionalism. Academy of Management Review, 21, 1022–1054. Hage, J. T. (1999). Organizational innovation and organizational change. Annual Review of Sociology, 25, 597–622. Hannan, M. T., & Freeman, J. (1984). Structural inertia and organizational change. American Sociological Review, 49, 149–164. Harel, Y. (1988). The new kibbutz: An outline. Kibbutz Currents, 2(August), 2–5. Jepperson, R. (1991). Institutions, institutional effects, and institutionalism. In: W. Powell & P. DiMaggio (Eds), The new institutionalism in organizational analysis (pp. 143–163). Chicago: University of Chicago Press. Kanter, R. M. (1968). Commitment and social organization: A study of commitment mechanisms in utopian communities. American Sociological Review, 33, 499–517. Kraatz, M. S., & Zajac, E. J. (1996). Exploring the limits of the new institutionalism: The causes and consequences of illegitimate organizational change. American Sociological Review, 61, 812–836. Littell, R. C., Milliken, G. A., Stroup, W. W., & Wolﬁnger, R. D. (1996). SAS system for mixed models. Cary, NC: SAS Institute, Inc. Maron, S. (1998). Kibbutz demography. In: U. Leviatan, J. Quarter & H. Oliver (Eds), Crisis in the Israeli kibbutz (pp. 1–12). Westport, CT: Praeger. Marx, K. (1967). Capital: A critique of political economy, Vol. III. New York: International Publishers. Meister, A. (1984). Participation and democracy in associations. In: A. Meister (Ed.), Participation, Associations, Development, and Change (J. C. Ross, Ed. and Trans.). New Brunswick, Transaction Books. Michels, R. (1962). Political parties: A sociological study of the oligarchical tendencies of modern democracy (E. Paul, & C. Paul, Trans.). New York: The Free Press.

288

RAYMOND RUSSELL ET AL.

Mill, J. S. (1909). Principles of political economy (W. J. Ashley, Ed.). London: Longmans, Green, and Co. Mort, J., & Brenner, G. (2003). Our hearts invented a place: Can kibbutzim survive in today’s Israel? Ithace & London: Cornell University Press. Near, H. (1997). The kibbutz movement: A history. Volume 2, Crisis and achievement 1939–1995. London: Vallentine Mitchell. Oliver, C. (1992). The antecedents of deinstitutionalization. Organization Studies, 13, 563–588. Pavin, A. (2005a). The kibbutz movement – facts and figs. 2004. Ramat Efal: Yad Tabenkin (Hebrew). Pavin, A. (2005b). The kibbutz movement – facts and figs. 2004. Ramat Efal: Yad Tabenkin (English). Rosner, M., & Getz, S. (1994). Towards a theory of changes in the kibbutz. Journal of Rural Cooperation, 22(1–2), 41–61. Rosner, M., & Tannenbaum, A. S. (1987). Organizational efﬁciency and egalitarian democracy in an intentional communal society: The kibbutz. British Journal of Sociology, 38, 521–545. Russell, R. (1996). Individual vs collective forms of sharing ownership in Israel. Journal of Rural Cooperation, 24(1), 67–86. Simons, T., & Ingram, P. (1997). Organization and ideology: Kibbutzim and hired labor, 1951–1965. Administrative Science Quarterly, 42, 784–813. Simons, T., & Ingram, P. (2003). Enemies of the state: The interdependence of institutional forms and the ecology of the kibbutz, 1910–1997. Administrative Science Quarterly, 48, 592–621. Simons, T., & Ingram, P. (2004). An ecology of ideology: Theory and evidence from four populations. Industrial and Corporate Change, 13, 33–59. Stearns, L. B., & Allan, K. D. (1996). Economic behavior in institutional environments: The corporate merger wave of the 1980s. American Sociological Review, 61, 699–718. Strang, D., & Meyer, J. W. (1993). Institutional conditions for diffusion. Theory and Society, 22, 487–511. Strang, D., & Soule, S. A. (1998). Diffusion in organizations and social movements: From hybrid corn to poison pills. Annual Review of Sociology, 24, 265–290. Talmon, Y. (1972). Family and community in the kibbutz. Cambridge, MA: Harvard University Press. Thompson, J. (1967). Organizations in action. New York: McGraw-Hill. Tolbert, P. S., & Zucker, L. G. (1996). The institutionalization of institutional theory. In: S. R. Clegg, C. Hardy & W. R. Nord (Eds), Handbook of organization studies (pp. 175–190). London: Sage Publications. Twisk, J. W. R. (2003). Applied longitudinal data analysis for epidemiology: A practical guide. Cambridge: Cambridge University Press. Webb, S., & Webb, B. (1920). A constitution for the socialist commonwealth of Great Britain. London: Longmans. Weber, M. (1978). Economy and society. (G. Roth, & C. Wittich, Eds). Berkeley: University of California.

Demographic and Environmental Influences

289

APPENDIX ON MEASURES AND POOLING OF OBSERVATIONS Measures Age. Age is the number of years that have passed since the kibbutz was founded. Size. We measure the size of a kibbutz as the number of members at the start of each two-year observation. This information is available in the statistical annuals of the kibbutz federations. Economic crisis. As part of the restructuring of kibbutz ﬁnances that occurred in the late 1980s, an evaluation of the economic condition of all kibbutzim was conducted in 1988. The index of economic crisis produced then begins at 1 for kibbutzim ‘‘in very good economic condition’’ and rises to 5 for kibbutzim ‘‘in very bad economic condition’’ or 6 for ‘‘young kibbutzim.’’ Demographic crisis. We test the effects of demographic crisis by calculating the average annual percentage change in the number of members from 1988 until the year of the observation. This measure of growth was transformed into a measure of ‘‘demographic crisis’’ by inverse coding. Percent adopting. To determine the percentage of kibbutzim that have adopted an innovation, we count only the kibbutzim reporting that they are currently using the innovation, and divide this by the total of all kibbutzim responding regarding that innovation in the given year. Federation. Kibbutzim afﬁliated with the more ideologically coherent Artzi federation are coded 1. Those afﬁliated with the larger and more permissive Takam federation are coded 0. The 18 kibbutzim afﬁliated with the religious kibbutz movement, as previously noted, are excluded from this analysis. Distance from cities. We measure distance from cities as the distance in kilometers from Tel Aviv or Haifa, whichever is nearer, following the standard practice of Israel’s Central Bureau of Statistics. The omission of Jerusalem has little impact in the case of the kibbutzim, because no more than a dozen kibbutzim are in closer contact with Jerusalem than they are with either Tel Aviv or Haifa. Radical change. Here we treat as radical any change that is seen as threatening a basic kibbutz value, such as equality or democracy, or as contradicting fundamental notions about the identity of the kibbutz, such as that it is a cooperative, and is collectively owned. In judging whether a given change was radical or not, the U.S.-based researchers did the initial coding,

290

RAYMOND RUSSELL ET AL.

but Getz exercised ﬁnal say on the grounds that he is a kibbutz member himself, conducts this survey every year, and is the only one of us who knows what is being said in Hebrew about these changes. We acknowledge that we face a risk that our measure of radicalism becomes simply a dichotomization of our measure of levels of acceptance. It is easy to be biased by the presumption that what is rare must be radical and what is common cannot be. Table 3 shows a correlation between the perceived radicalness of an innovation and the percent of kibbutzim accepting an innovation of 0.32. Because of this moderate association between radicalism and percent acceptance, we are particularly interested in seeing what unique effect is attributed to this variable, when the degree of acceptance of innovations is also controlled. The size of that unique effect is a measure of our success in capturing qualitative nuances of radicalism that are not just synonyms for infrequency of use. Note on Pooling of Observations The pooling of observations over years, kibbutzim, and innovations raises some statistical issues. The estimates of standard errors of effects (and, hence, tests of signiﬁcance) in the software used assume independent random sampling. Two observations from the same kibbutz are likely to be more similar than two observations from kibbutzim drawn at random; two observations from the same year are likely to be more similar than observations from two years drawn at random; and two observations of the same innovation are likely to be more similar than observations of two innovations drawn at random. Failure to account for all these forms of nonindependence of observations may result in estimates of standard errors that are too small. The preferred approach for dealing with the nonindependence of pooled data is the use of mixed effects models (Littell, Milliken, Stroup, & Wolﬁnger, 1996; Twisk, 2003). Use of mixed-models methods with effects for kibbutz, year, and innovation is not possible in the current case because of the unbalanced observational design and missing data. Instead, we have adopted a number of strategies that provide partial solutions. The effects of pooling across years are addressed in two ways. First, the models that we present include a vector of ﬁxed intercepts for year. This adjustment corrects for year-to-year changes in the innovations included in the survey, differences in who failed to respond in a given year, and any random time-speciﬁc shocks. Second, we include the proportion of observations that have already adopted each innovation in the year prior as a

Demographic and Environmental Influences

291

predictor of current transition probabilities. The inclusion of this predictor should remove auto-regression in the error term. The effects of pooling across innovations and across kibbutzim are similarly addressed through the inclusion of predictors that capture the main sources of commonality across them. In the case of innovations, models include whether an innovation is radical or not, and whether it is concerned with decision-making, consumption, or compensation. In the case of kibbutzim, models include such attributes as size, age, crisis, federation, and location. These approaches probably do not provide a complete solution to the pooling problem, and caution is needed in the interpretation of levels of statistical signiﬁcance, which are probably somewhat overstated in the tables.

This page intentionally left blank

292

COOPERATIVE ENTREPRENEURIALISM: RECONCILING DEMOCRATIC VALUES WITH BUSINESS DEMANDS AT A WORKER-OWNED FIRM Joel Schoening ABSTRACT Existing research on businesses that are both owned and managed by their workers suggests that these firms have one of two kinds of effects for their participants. They either learn to be better citizens of democratic society through daily democratic practice, or they become better capitalists through the daily practice of business ownership. Drawing on data collected through in-depth interviews and participant observation, I argue that cooperative participants learn both things. Furthermore, participants in cooperatives develop a spirit of Cooperative Entrepreneurialism that allows them to engage in free enterprise, while also adhering to the cooperative values of equality and democracy.

Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 293–315 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16011-2

293

294

JOEL SCHOENING

From downtown, there are two ways to get to Community Recreation Cooperative (CRC).1 One way is to drive along one of the city’s main thoroughfares. This route will lead you past some of the city’s oldest homes, then several big box stores, a mix of light and heavy industry, several gas stations, a few fast food restaurants, until ﬁnally a short trip down a side street takes you to the parking lot entrance. The other way is the Twisty Creek Bike Path, which follows a green belt along one of the city’s ﬂood control canals. Travelling along the bike path, a cyclist, skater, or pedestrian can wind past backyards, parks, and apartment complexes, slip behind the big boxes and under the main thoroughfare before turning off on a path built by CRC that leads directly to the back door of their facility. These two routes are representative of a tension in CRC, and in many cooperatives founded in the United States after 1970. This tension is between the social ideals that the cooperative and democratic workplace was intended to foster, and the pressure to conform to traditional business models in order to survive in a competitive market that valorizes proﬁt as the single most important goal. The thoroughfare route represents the economic pressures of the market in which CRC must survive. As a manufacturer, they must be closely connected to their suppliers and customers via the existing infrastructure. They must ﬁnd production space that leaves room for future expansion, and they must do so as affordably as possible. In contrast, the bike path is representative of CRC’s socially conscious side. The side that values the environment, honors the individual contributions and needs of workers, and is committed to beneﬁting the local community. However, adhering to these principles while also maintaining an organization based on proﬁt means that the cooperative not only has to face the normal struggles of doing business in a competitive market, but it must do so using a set of values that is at odds with the very market in which it operates. All worker-owned cooperatives – those organizations that are both owned and, at least to some degree, directed by their workers – face a similar set of tensions regardless of their principles. Conﬂicts such as those between democratic decision-making and efﬁciency (Russell, 1985), between the individual and the group (Jackall & Levin, 1984), between autonomy and authority (Rothschild & Whitt, 1979), just to name a few, have been identiﬁed by scholars as central to struggles that are inherent in organizations that are collectively owned and democratically managed organizations. Furthermore, research has shown that these same tensions are a signiﬁcant reason that cooperatives are relatively infrequent when compared to traditional organizations (Howard, 2000). There is a subset of ‘‘values based’’ (Cheney, 1999, p. 12) cooperatives like CRC however, that face another signiﬁcant

Cooperative Entrepreneurialism

295

contradiction. That is, how to reconcile values of equality, individual voice, respect for the average worker, and a concern for the environment with the demands of a business. Not only must organizations like CRC ﬁgure out how to avoid the dilemmas typical of the cooperative organization, but they must also try doing so while adhering to a set of values that make their task more difﬁcult. CRC, now over 25 years old, has managed to survive and to keep a grip, though not as tight as it once was, on the values that have made it a model of collective enterprise. This chapter will explore how CRC has managed to navigate the tension between the interests of creating a successful business and its interest in remaining committed to social values of equality and worker autonomy by relying on members who demonstrate a commitment to both the business needs of the cooperative and what they see as its social values. Furthermore, my research at CRC shows that this combination of interests, what I will call Cooperative Entrepreneurialism,2 overcomes an existing dichotomy found in current research that suggests cooperatives are either schools for democracy or schools for business. The emergent identity of Cooperative Entrepreneurialism contributes to our understanding of how cooperatives can succeed, how they affect their members, and how they maintain the loyalty of their workforce by offering a type of workplace that is more fulﬁlling than the traditional wage labor market. I collected the data for this chapter during 20 in-depth interviews with members and employees of CRC, a worker-owned and democratically managed manufacturer of outdoor recreational gear in a metropolitan area of approximately 250,000 people.3 Founded in the late 1970s by a few workers looking for an alternative to the traditional wage labor market, CRC now has sales in the neighborhood of $9 million and upwards of 90 workers, making it the largest cooperative of its kind in the United States.4 Like many of its contemporaries, CRC modeled itself after a style of producer cooperative that dates back to nineteenth century England. It is based on the idea that a group of workers can collectively own their factory and make democratic decisions about management and the allocation of proﬁt.

CASE HISTORY The tension between the social goals of the cooperative and the demands of a competitive market is evident in the cooperative’s history. For the ﬁrst decade of its existence CRC remained true to the ideal of equality. Though there were some charismatic leaders, all members earned the same wage and

296

JOEL SCHOENING

proﬁts were distributed at year’s end on the basis of hours worked. The few non-member employees needed to accommodate the seasonal nature of the business were given the same hourly pay and a beneﬁts package nearly identical to that of members. Furthermore, during this time it supported its social agenda outside company walls by sending volunteers and money to local environmental and anti-poverty campaigns. The cooperative at this time could be considered a ‘‘unitary’’ democracy (Mansbridge, 1983) in which the formal organization was a reﬂection of an existing system of shared values and friendship networks. As a result, decisions could successfully be made through face-to-face discussions that typically resulted in something very close to total consensus. The highly unitary nature of the cooperative, however, would not last forever. As the cooperative grew to over 30 members it became too time consuming to have all members participating in every decision. Meetings were drawing out late into the night over issues that were marginal to the functioning of the cooperative. The story that best exempliﬁes this change, and that I heard repeatedly in interviews, was that of a now-infamous meeting that focused on a single issue: a new dumpster. Although this dumpster was relatively irrelevant to the speed or efﬁciency of production, it took several hours of a weeknight meeting to sort through the information about the potential price, size, and manufacturer of a new dumpster, and then reach consensus on a decision that one empowered individual could have made in a matter of minutes. This event has taken on mythological proportions in the lore of CRC. Everyone I spoke with – even non-member employees who have been there for a relatively short time – was capable of using the story as an allegory for the cooperative’s past, present, and future. Some members tell the story with a hint of nostalgia for a past era. Others however, tell the story as a way to demonstrate how the cooperative was overly narcissistic during its early years, wasting valuable production time wallowing in hours of unnecessary democratic debate over every imaginable issue. Everyone however, views the story as representative of the moment when members realized that the cooperative could not continue to grow and at the same time maintain a system in which all members participated in every business decision. Ultimately, the meeting has become so mythological because it represents the ﬁrst major concession that the cooperative had to make, sacriﬁcing some equality in order to continue to grow and be successful in a market that demands speedy decision-making. The result was the development of a management structure based on principles of representative democracy. For the time being, CRC maintained

Cooperative Entrepreneurialism

297

the ﬂat pay scale, but the cooperative was divided into task teams, according to the type and location of work. Each team of members was then responsible for electing its own manager to a management team that had responsibility for handling the day-to-day business decisions. Additionally, a board of directors was created. The board consisted of eight members, four of whom were elected each year for two-year terms. The board was responsible for hiring a general manager and overseeing the long-term planning of the cooperative. In general, this is the structure of the cooperative today. Using this new structure the cooperative took the daily business decisions out of the hands of the cooperative at large and delegated the responsibility to elected ofﬁcials. Further evidence of the tension between the cooperative’s social ideals and market demands can also be found in the cooperative’s more recent history. Responding to changing market demographics and increased competition from brands making similar products overseas, the cooperative has ‘‘sourced’’ an increasing number of minor products – work that was formerly done in-house – to overseas factories. It has adopted Lean Production methods, which have been highly regarded in management circles (see esp. Kenney & Florida, 1993) as a system for increasing efﬁciency and quality while reducing overhead, but has also been criticized for increasing workloads without increasing wages (Parker & Slaughter, 1988). In addition, CRC has moved toward a more centralized power structure by taking some decisions out of the hands of the membership and the teams. Signiﬁcantly, team managers are now appointed or hired by the general manager rather than elected from the ranks of task teams. Also signiﬁcant in this regard, the cooperative has moved to a market-based pay scale, and it has taken on an increasing number of non-member, low-wage employees to supplement its labor needs. These changes were not taken lightly and have resulted in both a high level of turnover in the cooperative and some lasting tensions between those who feel that the cooperative has lost its soul by concentrating power and wealth in management, and those who feel that the cooperative’s ‘‘hippy element’’ was getting in the way of real ﬁnancial progress. In between these poles, however, sits an increasingly large number of members who see the developments I described above as the necessary means to keep the cooperative and the values that it represents alive. As Pete, an intense, vocal, gray haired man who has been a member for over ten years described it, ‘‘Democratic participation in management is all well and good, but people around here don’t always realize: no proﬁt, no cooperative. Just like that, your social values are irrelevant.’’

298

JOEL SCHOENING

The tacit statement here is that the cooperative and its social values are important and need to be protected. What Pete is emphasizing is a sentiment that is common in the cooperative: social values do not exist in a vacuum. CRC is not an island, and if the cooperative wants to protect its values, it must remain in business, which requires attention to the bottom line as well. This suggests, as I will demonstrate in the remainder of this chapter, that this group of workers represents something new in the cooperative literature: cooperatives and cooperative workers must not be either socially minded or business-minded, but they in fact can be, must be, both. Having re-conceptualized itself from an organization attempting to create an equitable work environment regardless of the business climate, to one which is attempting to create a successful business and that is respectful of its workers, CRC has made a major shift in its mission. However, in making this shift the cooperative has also facilitated a collective understanding of both individual needs and collective commitment that I call Cooperative Entrepreneurialism. While potentially sustaining the cooperative for the long term, Cooperative Entrepreneurialism may also serve as an example of how cooperatives can overcome the tension between the individualism so prevalent in contemporary United Sates society and the collectivism necessary for the long-term survival of a cooperative. More importantly, if cooperatives such as CRC can foster more collectivist sentiments among their workers, while at the same time providing them with a modest standard of living, it may indicate a crack in the hegemonic ideology of selfmaximization, and offer an example of an organization that can remain viable without necessarily having to fully submit its social agenda to the demands of the market. This paper will shed new light on this tension by moving beyond some of the existing dichotomies in the literature on cooperatives and demonstrating that cooperative members are not committed to either business values or social values. What is unique is that CRC workers, new and old, are developing a set of social values that retains a hint of counter-hegemonic force, even as the cooperative barrels down the path of neoliberal business practices.

LITERATURE REVIEW According to existing theory, cooperative workers are expected to derive one of two kinds of lessons from their experience with a cooperative. One theory suggests that daily democratic participation in the management of the company will inﬂuence cooperative members to become more civically

Cooperative Entrepreneurialism

299

engaged (Dahl, 1985; Pateman, 1970). The alternative point of view suggests that cooperatives, particularly in the context of capitalism, will imbue members with a ‘‘shopkeeper spirit’’ (Malatesta, 1922), in which cooperators come to view themselves as owners, entrepreneurs, and capitalists. Moreover, unless the cooperatives are rigorous in their commitment to goals other than the accumulation of proﬁt, participation in a cooperative has the effect of socializing workers into a capitalist mindset. Though cooperatives are democratic, this approach asserts that workers, by gaining access to capital, become concerned with protecting their investment, thus losing the solidarity of the cooperative experience. Therefore, cooperatives should not be seen as automatic agents for a more equitable economy, especially in a capitalist system. Rather they should be seen as effective means of creating good capitalists. In a society that places a high value on the acquisition of property, creating more business owners can be considered a great success. Yet this approach fails to measure the other potential beneﬁts of cooperatives, especially those value-based cooperatives such as CRC that have other signiﬁcant goals in mind. The main problem has been that many researchers have not considered the intent of the cooperatives they study. Though unintended consequences are also worthy of study, it would be wrong to conclude that cooperatives in general are incapable of imbuing their members with further knowledge and passion for democratic participation, when the case under study never had that as one of its goals to begin with. Similarly, it would be wrong to conclude that cooperatives always contribute to a shopkeeper spirit, when other goals of the cooperative under study have been left unexamined. Another problem has been that, due to the relative infrequency of functioning cooperatives, the great majority of empirical research in this area has focused on some part of the problem of proliferation. For example, work by Rothschild and Whitt (1986), Whyte and Whyte (1988), Jackall and Levin (1984), and many others have analyzed the various organizational responses to the tensions inherent to cooperative organization. Scholars have suggested that cooperatives, especially those with social goals, must have a high degree of commitment and/or homogeneity among their members in order to successfully navigate these tensions (Perry, 1978; Russell, 1985; Stryjan, 1989). In addition, in his study of Northwest plywood cooperatives, John Pencavel (2001) found that culture can have determining effects for cooperatives. Cooperatives in which a majority of members were drawn from homogeneous cultural backgrounds had a better chance of long-term success due to a more uniﬁed ideology than cooperatives with workers from more diverse backgrounds.

300

JOEL SCHOENING

This concept, known as ‘‘cultural homogeneity’’ (Greenberg, 1986), is a useful building block for understanding how members resolve the tensions in cooperatives. Cultural homogeneity can be deﬁned as the degree of similarity across the class or ethnic backgrounds of cooperative members. Where there is a high level of cultural homogeneity when the cooperative is formed, the cooperative will likely have an existing network to draw from for new members, and can more easily develop a shared concept of the cooperative. This, in turn, greases the wheels of future decision-making, which allows them to operate with less friction. Additionally, a strong shared sense of purpose in the cooperative is thought to protect against the concentration of ownership predicted by Webb and Webb’s (1920) degeneration thesis. Edward S. Greenberg (1986) is one of the few scholars to directly address the role of cultural homogeneity in the ability of cooperatives to overcome their inherent tensions. Greenberg studied the plywood mills in the Paciﬁc Northwest primarily through a survey of workers from cooperative mills, union mills, and traditional hierarchical non-union mills. On the one hand, the cooperative members that he surveyed were more likely than noncooperative workers in the same industry to have positive attitudes about cooperation in the workplace and were also slightly more active in local politics. On the other, he found that cooperatives are at least as likely to imbue members with an entrepreneurial spirit as they are with a spirit of participatory democracy. In his study the cooperative mill workers were signiﬁcantly less likely over time to believe they had political efﬁcacy, or that others could replicate their achievements. According to Greenberg, cooperative members became increasingly likely to adhere to an entrepreneurial ideology that supported the belief that they had achieved their success through their individual efforts and talent. Ironically, they were less likely to understand their success as the result of shared goals and cooperative ownership. Greenberg’s results led him to believe that in a capitalist economy, cooperatives should be expected to be no more than cooperative capitalism. That is, cooperatives should be understood as collections of shareholders, cooperating in order for each to increase the value of her/his shares, not as organizations working to forward any social values other than the protection of the business. Greenberg, however, does not sufﬁciently account for the original culture of his cooperative participants, nor does he evaluate the goals of the cooperative itself. The plywood cooperatives were not started with explicitly social goals such as expanding the democratic rights of workers or protecting the environment. These cooperatives were started with the goals of job

Cooperative Entrepreneurialism

301

creation and the enrichment of a local culture and community. Greenberg more likely measured the level of entrepreneurialism that existed in the local culture than the impact of the cooperative on workers, or the depth of their civic engagement. Thus, one could indeed conclude this – it is really that the ‘‘failure’’ is of a goal the cooperative never set for themselves, and that they operate within a larger culture, one of whose major tropes vis-a`-vis business is individual striving and entrepreneurialism, and decidedly not democracy. The issue of an organization’s intended goals is of particular relevance for cooperatives like CRC that have professed a commitment to values in addition to the accumulation of proﬁt for their members. This is especially the case for alternative organizations who are reliant on a high level of commitment on the part of individuals in an era when that commitment is becoming more difﬁcult to ﬁnd and maintain (Sirianni, 1987). The commitment factor may be increasingly important to the survival of organizations like CRC, who still claim to have a mission that is larger than simple ﬁnancial success but has neither the ﬁnancial resources to compete at the highest levels in the labor market, nor a homogeneous labor pool to draw from to ﬁll its ranks. More recent research on value-based cooperatives, such as Cheney’s (1999) and Kasmir’s (1996) studies of the Mondrago´n cooperative network, provide an insightful look into the ways in which cooperatives negotiate global market pressures and their social agendas. Cheney looks speciﬁcally at the issue of whether or not a large and multifaceted network of cooperatives like the Mondrago´n Corporacio´n Cooperativa (MCC) can continue to grow without losing its soul. What he ﬁnds is that MCC is going through a complex redeﬁnition of its values, and especially the way those values are put into practice. The bad news is that certain kinds of direct participation in management are being lost in favor of more traditional management structures such as Lean Production. The level of ﬁnancial inequality between the lowest paid workers and the highest paid workers has increased and the general workload of individuals is on the rise. On the good side, workers at many Mondrago´n facilities reported a resurgence of worker participation on the shop ﬂoor. At the level of everyday production, workers are retaining a measure of autonomy and discretion in their control of the labor process. In sum, MCC has not lost its soul in terms of worker’s control over some of their daily labor processes, but its values have been compromised in terms of worker voice in the direction of their company, and in the provision of equal rewards for equal effort. Sharon Kasmir (1996) approaches MCC from an angle similar to that of Malatesta. She asks if, by giving working class Basques a ﬁnancial stake in

302

JOEL SCHOENING

their economy, MCC has the long-term effect of weakening the working class solidarity of the Basque region and replacing it with middle class values of consumerism and property protection. In short, does the Mondrago´n system result in the development of the shopkeeper spirit? She goes on to argue that the project of the cooperatives can be considered an hegemonic project. The founder, father Jose´ Marı´ a Arizmendiarrietta, was profoundly interested in education and sought to use the cooperative to create a middle class life for the families of the region. Indeed, the cooperatives can, in many ways, be considered company towns with a factory, school, and grocery store all run through the same network. After the cooperatives were established, women were discouraged from working in the factories, and were instead encouraged to become homemakers. Likewise, cooperative members were encouraged to develop into entrepreneurs. These examples demonstrate the hegemonic project that Arizmendiarrietta had in mind for the cooperatives. Kasmir presents a compelling story about the inﬂuence of the cooperatives on social life in the Basque region, and unveils some long hidden truths about the Mondrago´n cooperative system. Indeed, her research provides persuasive evidence that cooperatives are reliant on a core belief system and that individuals’ social values can be transformed through their experience with a cooperative. Furthermore, Kasmir provides support for the idea that cooperatives can create a shopkeeper spirit. It should be noted however, that the great myth about which she speaks in her book, The Myth of Mondrago´n, is that of the idealization of the cooperative as an alternative to capitalist enterprise, which in her well-reasoned and documented view, it is not. In this respect, her research provides a model for the work presented in this chapter. That is to say, when analyzing alternative organizations for their ability to create changes in their members, researchers must be sure to distinguish between intended and unintended consequences. To a large degree, the ability of a cooperative to ‘‘succeed’’ – and similarly, what it fails to accomplish – must be measured against its intended consequences.

FINDINGS Contrary to existing research which presupposes that cooperatives either encourage members to become better practitioners of democracy or foster the shopkeeper spirit, my research suggests that the division may not be so clear and that, in fact, a cooperative may contribute to both. Furthermore,

Cooperative Entrepreneurialism

303

the combination of the shopkeeper spirit and the adherence to values of equality and democracy, what I call ‘‘Cooperative Entrepreneurialism,’’ may actually be the more common outcome, and an outcome that can be beneﬁcial to the success of the cooperative at large as well as for individual workers. Cooperative Entrepreneurialism arose in my interviews as CRC members consistently described the dualistic nature of the cooperative. These statements, as I will show below, nearly always include a rationalization for business decisions that draws on the values of the cooperative. Cooperative Entrepreneurialist statements are those that place the reason for doing business beyond proﬁt, and emphasize the collective interests of all members such as ‘‘We are in business to make a living wage for our members.’’ It would not be uncommon to hear a similar statement coming from a CEO or the public relations department of a large corporation, but you would not likely hear such a statement about the same company from a rank and ﬁle employee. At CRC, Brad described this sentiment to me in this way: Just the idea of bringing on partners instead of employees, I would never want to be to a point where I am getting comfortable making x amount per year and business goes up so I bring in two employees. Business turns down and I can let them go. Knowing the fact that their lives, their family’s livesy. If you take those same three people, ﬂatten the pay scale, if business turns down we all move down a little bit. Its like everyone is affected a little. It’s like when you get married. It’s the same kind of thing. You don’t say, ‘‘Hey baby, you got laid off, too bad for you, where is the mortgage this month?’’ You obviously help each other. In that way [the cooperative] really affected my outlook on business.

The reason for being in business as a cooperative is to protect co-members, and to attempt to distribute both burdens and rewards equally. The development of business savvy among cooperative members, and the combination of that savvy with the long-standing adherence to values of equality and democracy is the root of Cooperative Entrepreneurialism. Though the Cooperative Entrepreneurialism that developed at CRC did so as a result of CRC’s history, it should be noted that CRC is not completely unique in its historical roots and thus the ﬁndings here should not be considered idiosyncratic. There was a wave of cooperative start-ups in the late 1970s of which CRC was one. The majority of these cooperatives were founded, like CRC, for reasons other than the pursuit of proﬁt. Often, they were started by progressive individuals looking for a place where they could make a living without compromising their ideals (Jackall & Levin, 1984). CRC’s history is unique, however, in that it can be broken into three distinct periods – corresponding with product market and organizational

304

JOEL SCHOENING

changes – these I will call The Early Years, The Boom Years, and The Reorganization Years. The following sections elaborate on all three of these periods and detail how each era contributed to the development of Cooperative Entrepreneurialism. The Early Years The Early Years of the cooperative run from the founding of the cooperative in the late 1970s up to the beginning of the boom era in the mid-1980s. This period of time is distinguishable from the later eras due to the cooperative’s directly participatory and highly unitary democratic (Mansbridge, 1983) character during this time. As I will demonstrate, during the Early Years the cooperative was constituted of individuals who shared a progressive ideology – based in classical liberal principles of individual rights – that supported equality among workers and the democratic process. The Early Years are important to the development of Cooperative Entrepreneurialism because the initial members laid the foundation for what would later become codiﬁed as the values of the cooperative. Although members during the Early Years shared a progressive individualist ideology, CRC was not founded with a speciﬁc social agenda or value system in place. In fact, CRC only became a cooperative because the original owners – and the inventors of their most popular products – decided to sell the company as a result of their discomfort acting as owners and managers. According to River: The founders of the company were tired of feeling responsible for the livelihood of their employees and back then they weren’t making any money, a buck here a buck there, so they sold it. They felt as though, well, even if they weren’t hugely successful, that they had brought products onto the planet that when it came time that the world really needed them, they would be there, even if the company disappeared.

Having decided to sell, the founders offered their ﬁrst three employees the opportunity to take over the business. The original employees took the opportunity and created the cooperative so that they could all continue to provide themselves with a livelihood without creating a hierarchy among them, or writing a strict policy about how work would be accomplished. The original members were also not expressly interested in contributing to some larger movement, to changing the minds of others, or to democratizing the economy at large. On the contrary, they were interested in paying the bills while also retaining a high degree of control over their own labor process. Warren, another long-time member recalls:

Cooperative Entrepreneurialism

305

Our original beginnings weren’t especially noble or socially redeeming. We just simply wanted a job. We didn’t want to work any more than we had to, and most of the early members were more interested in trying to ﬁgure out how to spend less money, so they could earn less and therefore work 30 hours instead of 40 hours a week.

This is a common characterization of the early period in the cooperative’s history. Often, early participants described themselves in a way that connotes a lackadaisical attitude toward business. Many members communicated a sort of disbelief that the cooperative ever ended up successful at all. Warren’s humble depiction of the cooperative’s birth masks the underlying importance of CRC member’s commitments. Whether or not they thought so then, or think so now, the original members’ commitment to environmentally sustainable products crafted by workers who had near total autonomy in controlling their work and shared proﬁt and loss equally was quite radical. Though their commitments were not yet codiﬁed into a coherent set of policies, they were no less important as the cooperative began to grow and take on more members. Owing to the slow pace of business growth during the Early Years, it was easy for the cooperative to retain its unitary democratic organization and recruit new members from existing friendship networks and word of mouth about the organizational culture of the new business. The uncodiﬁed work ethic comes through in the following statement from Frank, another long time member who had formerly worked for a military contractor, as he explained his reasons for joining the cooperative: I was looking to do something cleaner, quieter, and less deadly than what I used to do. Working for Big Steel, a huge machine shop, I don’t know if you are familiar with engine lathes but we had one down there that could swing 20 feet. This industry to me seemed more um, socially and environmentally benign and actually positive, more so than most industries.

A similar and even more emblematic story comes from Roger: And I had been getting the counter-cultural newspaper [from the region] and they were doing a test report on CRC products. I noticed CRC was a coop, and I thought, ‘‘Well lets see, coop, outdoor products, two of my main interests so y’’ [laughs] That’s why I came. I came out in ‘81 and looked ‘em over. Went back and got all my remodeling done on my house and drove back out, in the middle of a blizzard. It was January.

Roger went on to tell me that when he got to CRC, he worked for free for several weeks until the cooperative ﬁnally took him on as a member. Many of the early members of CRC relocated themselves, and often their families, to be a part of the democratic atmosphere and for the prospect of escaping the drudgery and hierarchy of the traditional labor market.

306

JOEL SCHOENING

During this time the products being made at CRC were totally unique in their industry and were becoming renowned for their great quality. The combination of such an innovative product with the reputation of worker ownership cemented CRC as a brand name in the industry. This reputation helped CRC to grow, and by the early 1980s the cooperative had over twenty members, but still remained committed to total equality. As Jon describes that time: Everybody did everything. Everybody was on the Board [of Directors]. We all made all the decisions. At the time it worked because we were just on the verge of the explosion. It was great. Everybody did everything. Small tasks too. We would all eat lunch together. When we started making money we had CRC Free Lunch. Every other week CRC bought lunch for everybody. It was one of those things like, as soon as you came on, you were equal to everybody in every way. There was no seniority; there was no pay differential. Everybody was equal.

It would not be much longer, however, before the cooperative would realize that total equality and participation were irreconcilable with further growth. Nonetheless, the unitary commitment of early members to equality and democracy permanently framed the organization’s culture. The creation of what one member termed ‘‘the cooperative ethos’’ has been critical to the long-term survival of the organization’s social values, and therefore, to the development of Cooperative Entrepreneurialism.

The Boom Years The beginning of the Boom Years is marked by two separate but interrelated events. First, there was a huge surge in the sales of CRC products. Second, there was a corresponding surge in membership that forced the cooperative to increase its division of labor and create more speciﬁc job descriptions. These events contributed to the formation of Cooperative Entrepreneurialism because members came to believe that the cooperative could be a successful business. Furthermore, because the surge in membership forced the cooperative into a representative, rather than direct, style of democratic participation, the cooperative chose to put its values into writing, codifying what was previously implicit. In the mid-1980s CRC’s industry was on the verge of a major renaissance that would result in an explosion in sales and therefore in the potential market for CRC. The company found itself surprisingly well positioned. Though CRC had not gone the traditional route of creating a strategic marketing plan it had, by this time, established a strong reputation for high

Cooperative Entrepreneurialism

307

quality products. When the boom hit the industry, CRC rode the wave. As Gerry described it; We did so well, by such luck. It was one of those things, right place, right time, right product, and bing! We didn’t have to try. We had a three person sales and admin. team. We were like six thousand products backordered for three months. It was like they were ﬂying out the door as fast as we could make them. We didn’t have to try and sell, all we had to do was answer the phone.

This business rush meant that the cooperative had to bring on more workers quickly which led to the second change that marks the boom years, a considerably larger membership. In order to keep pace with the expansion in demand for its products, CRC chose to expand its labor force, a decision that would have a ripple effect on the cooperative and the development of Cooperative Entrepreneurialism. One effect was that it became immediately apparent that the cooperative would need to delineate a division of labor if it wanted to keep up with demand. No longer could ‘‘everyone do everything.’’ The cooperative was now employing industrial machinists, technicians, engineers, administrators, and myriad other specialists who had speciﬁc training and who were ﬁlling very speciﬁc needs in the cooperative. The rapid expansion in membership also brought a wider diversity of motivations and commitments among new members of the cooperative. Some of the new members still came because they were attracted to the principles of democracy and equality, but others were drawn in because they liked the industry, and many were attracted to it because it was rapidly becoming the best wage and beneﬁts package available for a manufacturing worker in the region. People who would otherwise be making $8 an hour, working nine to ﬁve at a company where they had to ask to go to the bathroom, could come to CRC and after six months they could be a member. All of a sudden they were making $50,000 a year with a great beneﬁts package, stock, and a vote in the direction of their company. Unfortunately, this diversity led to the kinds of moments typiﬁed in the story about the dumpster. With so many cooks in so many kitchens, decisions were nearly impossible to make within a timeline that was meaningful to CRC’s customers and suppliers. As Patrick explained: People thought being a business owner meant that they got to decide what color to paint the products. I was chairing the board when there were 25 to 30 people and everybody was trying to decide everything. It became ridiculous. [We need to decide] what color paint are we gonna use. [One person says,] ‘‘I think we oughtay’’

308

JOEL SCHOENING [Another says,]’’I think we oughtay’’ [I say,] ‘‘How about having the sales and marketing team decide that stuff? [They reply,] ‘‘But we’re all owners.’’ It just got ridiculous. That was the point where we had to change.

In this quote Patrick gives an example of members’ belief that ownership meant that they had a right to participate in all decisions. However, the increased size and division of labor in the cooperative meant that it was nearly impossible for any one individual to be up to date and an expert on all aspects of the cooperative. The cooperative needed some structure to delegate decision-making, and a committee was formed to draft a new structure for the cooperative. After 31 versions were written, distributed, debated, and rewritten, a new set of bylaws was ﬁnally accepted through a membership vote. Version 32, as it is known, laid the basis for the general organization of the cooperative to this day. The most important effect of the rapid expansion in the size and diversity of the membership, and of Version 32, was that it signaled an end to unitary and direct democracy. It was the ﬁrst ofﬁcial move toward a style of democracy that was more representative and adversarial (Mansbridge, 1983) in its characteristics. Though Version 32 represented a turn away from the cooperative’s direct democracy roots, ‘‘the cooperative ethos’’ that was built in the early years remained strong in the face of these new challenges. Indeed, the strength of members’ commitment to their social values could still be seen in several ways. In the early 1990s the cooperative organized a retreat where members spent the weekend writing a new mission statement. The mission statement was ratiﬁed by a vote of the entire cooperative and codiﬁed the cooperative’s decade-long commitment to respecting the contributions of all workers, creating environmentally benign products, and maintaining democratic decision-making. Furthermore, as the cooperative became more proﬁtable, it started taking a more active role in supporting its values by offering ﬁnancial support and volunteer time to organizations that represented its causes, such as civil rights, anti-poverty and forest protection campaigns. Roger put forth these terms: As we became more prominent, more social organizations would come looking for ﬁnancial assistance. And then we would try to ﬁgure out how do we, where do we draw the line between organizations that we assist vs. those that we don’t assist? We used to support the gay rights stuff. We used to have a sticker on our van, which I think is still there, and lawn signs. We got in trouble with some of the wood products industry for our support of a sustainable forestry organization. We got this letter that people were going to boycott us, but it doesn’t make any difference to us because not very many loggers use our products, and 95% of our products are sold out of state anyway. I took it as an honor.

Cooperative Entrepreneurialism

309

The Boom Years are important to the development of Cooperative Entrepreneurialism because they forced the cooperative to grow, and at the same time adapt their commitments to their new growth. Though the Boom Years forced CRC to grow beyond its unitary roots, it also forced its members to codify the cooperative ethos into a mission statement. In addition, the Boom Years were important because it made cooperative members believe that they could be successful ﬁnancially, and still hold on to their values. The Reorganization Years The Reorganization Years began in the late 1990s and are still under way at the time of this writing. They are marked, most predominantly, by a major contraction in CRC’s industry combined with increased competition, particularly in the form of major name brands that have created a similar product line and use less expensive materials and manufacturers from overseas. These changes have forced the cooperative to make some adaptations – such as the institution of a market-based wage scale and a policy governance system – which threaten to undermine the very core of the business: cooperative ownership and democratic control. However, the increased competition and resulting hard times have forced the development of a real business consciousness among members that, in combination with the longstanding commitment to values of equality and democracy, have furthered the creation of Cooperative Entrepreneurialism among members and continues to be a resource for the cooperative as it adapts to the current market conditions. The two most challenging changes for the cooperative thus far in the Reorganization Years have been the institution of a Policy Governance system and the change to differential pay. Many members had seen the need for these changes for a long time, but the majority of members were convinced by the cooperative’s history, that they could continue to be successful in their current form. Finally, the cooperative found itself in ﬁnancial difﬁculty, and the majority of the membership saw that they had to make some changes. In Susan’s words: Being on the Board [of Directors] was frustrating because as our proﬁtability started to go down, I was arguing that we had to start running this place more like a business and less like a work club. ‘‘Membership has its privileges’’ we always used to say as a joke. Well, nobody would listen until we lost money. Then all of the sudden people are saying, begrudgingly, ‘‘I guess I am going to start having to run this like a business.’’

The ﬁrst major step toward running the cooperative like a business was the implementation of Policy Governance. Policy Governance is a bureaucratic

310

JOEL SCHOENING

template that many companies use to set the chain of command and responsibility in their organizations. Before Policy Governance at CRC, as Brad puts it, ‘‘You could walk into the General Manager’s ofﬁce and say, ‘This policy is crap we need to change it.’ It may not get changed but you could go in and say that.’’ Although this kind of open door communication has a long tradition at CRC and is based in the cooperative’s value of equality, it has never been conducive to efﬁciency or even to effective communication. Many members were aware of this contradiction and were ready to accept Policy Governance because they could see its beneﬁts. Warren described it this way: Policy Governance has the beneﬁt of not micromanaging the general manager and the management team, and allows decisions to be made more quickly. That’s probably a desirable thing, provided that the membership still feels like there is some meaning to being a member other than simply owning an economic share. But the kinds of issues that come up from within the general membership are different. It’s more top down now than it used to be. But it’s a big space between that and being disenfranchised.

Warren here describes the balance he sees between not having a say in everything, yet still having some agency in the cooperative. Policy governance effectively creates a more clear system of accountability, which makes it easier to see if the board is doing their job, without overriding the existing democratic structure of the cooperative. Warren’s statement is based in a Cooperative Entrepreneurialist approach to the organization. He walks the line between what is necessary to keep the business alive and what would completely end the democratic participation of members in the control of their company. The next major change, and one that is potentially more threatening to the cooperative’s values, has been the implementation of a differential or ‘‘market-based’’ wage scale. For over 25 years the cooperative paid all members the same hourly wage. The equal wage had been one of the most important symbols of the cooperative’s commitment to equality. Gerry, a member who often represented CRC at a regional cooperative conference, described it this way: People [at the conference] were astounded that we had a ﬂat wage set up. ‘‘How do you do that?’’ And that’s the thing see, there are some people at CRC, who would act like, since CRC did it, that’s what comprised coop. Flat pay, well that’s part of being a coop. No it’s not. Coop, is one member one vote. The rest of the stuff you can pile on, but that doesn’t have to deﬁne us.

Members’ commitment to a ﬂat pay system was eventually over-ridden by pressures from the labor market. Falling sales numbers equated to smaller

Cooperative Entrepreneurialism

311

dividends, which meant that the cooperative could no longer keep some of its high skill positions ﬁlled. Pete described the situation this way: We ﬁnally got differential wage in here. Part of the reason is, it got to a point where for the same wage, we couldn’t attract any of these administrative or creative type people, the dividend is not there. So you can’t lure them in, you have to have something so that they will put up with the day-to-day work. We ﬁnally paid our general manager, it’s low by GM standards, but it’s huge by our standards and, it’s decent for the area. It’s two and a half times what the average person here is making in wages.

Though the cooperative’s hand has been forced in this decision, it has been able to bend without breaking. When the top person in an organization is making two and a half times the people at the bottom, and is also subject to their approval through a democratic system, much equality remains relative to just about any other kind of business organization. Regardless of these major changes, there remains among members a strong commitment to the democratic control of the business. When I asked Chelsea if she had learned about running a business as a result of her participation in the cooperative she had this to say: Yes, deﬁnitely, two terms on the Board really opened up my eyes to business. That’s not what my passion is though. I’ve got a side business that I run on the weekends so, okay, I’m learning about business now, and a lot of what I learned at CRC I’m applying. But to me that’s not as interesting. Business is pretty simple. The basics are anyway. Yeah, I learned that. So what? If we lose democracy at CRC, then it’s just another job to me.

The importance of the democracy, in Chelsea’s case, is essential to her connection to the cooperative. What is great about working there, for her, is that workers have a voice in decisions about the future of their job. Even when the question concerned what she had learned about business, democracy is important enough to her that it still comes through as the best part of her job. Furthermore, at CRC there is an increasing awareness of business demands at all levels of the cooperative. The reason for making changes to a more traditional looking business remains larger than simple proﬁt motive, and CRC retains a level of consciousness that is reﬂected in both the rewards that individual members report that they derive from their workplace, and in their desire to see the business succeed. For example, Warren had this to say about recent changes in the cooperative: We can’t be so consumed by our internal world that we end up going out of business. So we could spend a lot of time improving our communication, but is that going to make us any money? Is that going to keep us from going under? It’s a delicate balance because communication has to be maintained or else we are going to eat ourselves alive from the inside out.

312

JOEL SCHOENING

Changes in both the demographics of their market, as well as in the level of competition, have forced CRC to alter its course. As a result, the reorganization of the cooperative has meant an increased focus on proﬁts and from this perspective the goals of the cooperative have changed. However, in speaking with members it is clear to me that though the emphasis may have shifted, the commitment to democracy and respect for the average worker has survived. Furthermore, it is the combination of these commitments and business savvy that has both facilitated the emergence of Cooperative Entrepreneurialism, and allowed the company to survive the reorganization thus far. In the words of Frank: Whether we like these little inroads to our democracy or not, there is still tons [of participation]. If you go down on to that ﬂoor and you see the level of participation that is accessible, it’s unparalleled. There is no other place where there is more access. But its not like you can do any fucking thing you want anymore. So if we can survive the shortterm cash problem we have, this could turn out to be really all it could be.

The Reorganization Years are important because they have acted as a balance on the overly idealistic Early Years, and the easy proﬁts of the Boom Years. The realization that the cooperative is a business, and therefore must at least operate in ways that its customers and suppliers can rely on has forced CRC to become business savvy and has really brought the entrepreneurialism to the cooperative. It is during the Reorganization Years that Cooperative Entrepreneurialism really emerges. As Brad told me: I know a long time ago someone said that we should be doing more to help the poor, you know, that we should charge less for our products, that we should give away more money or something. And one of the former membersya good ol’ farm girl, a really straight up person said, ‘‘I don’t think we could do much to help the poor by being them. Ya know?

CONCLUSION CRC, it turns out, shows us that in a capitalist economic context, workerowned and democratically managed businesses can be incubators of both democratic and business ideologies. Two quotes exemplify this point. Brad: Already being [politically] left of center, looking at people and saying, ‘‘well okay, maybe I think their ideas are crack [baseless].’’ But after working in a democratic workplace, I know that they still deserve equal consideration. Not that they need to be argued and argued and argued, but they still need to be respected. You [have to] listen to the crackpot, who stands up and wants to spout off about something.

Cooperative Entrepreneurialism

313

Brad’s words attest to the fact that the tradition of equality and democracy at CRC survives and remains strong enough to impact new members. Through his participation in democracy at CRC Brad has learned to be patient with other’s perspectives and to listen, even though at ﬁrst he may think that they are useless. The following statement by Jason’s is also illustrative: This company can continue to serve us in 90–95% of the ways it does now, but we’ve got to tweak that other 5 or 10% so it can do the rest of the good things that it can do. It can’t do 100% of ‘em anymore. But gosh, if we can change the 10%, I guarantee you the results will get us back to the dividends of the early 90’s.

In this quote he is demonstrating how most of what is important to members about the cooperative, that 90% that includes the great beneﬁts, democratic participation and worker autonomy, can be protected, but there needs to be some minor changes made, the 10%, in order to make the cooperative a viable business. Jason’s sentiments sum up CRC’s current situation in Cooperative Entrepreneurialist terms by balancing the business demands with the beneﬁts of cooperative organization. The development of Cooperative Entrepreneurialism should be neither surprising nor disappointing. If we return to the metaphor of the two paths, the tension between social values and market pressures at CRC look more like a dialectic than two opposing monoliths. The cooperative is constantly adapting to the changes in the larger economy, and to do so, it relies on the energy and input of its members. This cooperative is creating change, though incremental, by creating people who have acquired new business skills while at the same time remaining committed to principles of equality and democratic participation. Worker-owned and democratically managed organizations might turn out to be a more stabilizing, rather than revolutionary force in a capitalist democratic society than some scholars would like to think. However, they also offer workers the power to be selfemployed and to be able to disengage from the wage labor market. A more powerful skill is hard to imagine.

NOTES 1. The name, location, and other identifying characteristics of the cooperative, the members and the employees have been changed to protect the anonymity of the cooperative and its members. 2. I thank a reviewer of an earlier version of this paper for providing me with ‘‘cooperative entrepreneurialism’’ as a moniker for my discoveries at CRC.

314

JOEL SCHOENING

3. Members are those workers who own a share of the company. Workers who do not own a share of the business were referred to by both owner and non-owner respondents, as employees, or sometimes ‘‘seasonal workers.’’ 4. The cooperative has always been mainly white in racial makeup, though not disproportionately so considering its region. It is also predominantly male, but this is not particularly surprising given that that CRC is a manufacturing company and many of the jobs, such as welding, fall into categories traditionally dominated by men.

ACKNOWLEDGMENTS I would like to brieﬂy thank those people whose contributions have made this work possible. First, I would like to thank my mother, whose recent death has reminded me that her many years of work have served as the inspiration for my own. I would also like to thank my partner, Dawn Borgardt. Her reading skills, support, and encouragement have been invaluable. I would also like to thank many other friends and family members whose patience and support has been so wonderful. Finally, I would like to thank the members and employees of CRC, who volunteered their time to speak to me, and whose insights, honesty, and hard work continues to provide an example for the rest of us.

REFERENCES Cheney, G. (1999). Values at work: Employee participation meets market pressure at Mondrago´n. Ithaca: Cornell University Press. Dahl, R. A. (1985). A preface to economic democracy. Berkeley: University of California Press. Greenberg, E. S. (1986). Workplace democracy: The political effects of participation. Ithaca: Cornell University Press. Howard, M. W. (2000). Self-management and the crisis of socialism. New York: Rowman & Littleﬁeld. Jackall, R., & Levin, H. M. (1984). Worker cooperatives in America. Berkeley: University of California Press. Kasmir, S. (1996). The myth of Mondrago´n. Albany: State University of New York Press. Kenney, M., & Florida, R. (1993). Beyond mass production. New York: Oxford University Press. Malatesta, E. (1922). Umanita Nova. Mansbridge, J. J. (1983). Beyond adversary democracy. Chicago: University of Chicago Press. Parker, M., & Slaughter, J. (1988). Choosing sides. Boston: South End Press. Pateman, C. (1970). Participation and democratic theory. Cambridge, England: University Press. Pencavel, J. (2001). Worker participation: Lessons from the worker co-ops of the Pacific Northwest. New York: Russell Sage Foundation.

Cooperative Entrepreneurialism

315

Perry, S. E. (1978). San Francisco scavengers: Dirty work and the pride of ownership. Berkeley: University of California Press. Rothschild, J., & Whitt, J. A. (1979). The collectivist organization: An alternative to rationalbureaucratic models. American Sociological Review, 44, 509–527. Rothschild, J., & Whitt, J. A. (1986). The cooperative workplace: Potentials and dilemmas of organizational democracy and participation. Cambridge, Cambridgeshire: Cambridge University Press. Russell, R. (1985). Sharing ownership in the workplace. Albany: State University of New York Press. Sirianni, C. (1987). Worker participation and the politics of reform. Philidelphia: Temple University Press. Stryjan, Y. (1989). Impossible organizations. Westport, CT: Greenwood Press. Webb, S., & Webb, B. (1920). A constitution for the socialist commonwealth of Great Britain. London: Longmans. Whyte, W. F., & Whyte, K. K. (1988). Making Mondrago´n: The growth and dynamics of the worker cooperative complex. Ithaca, NY: ILR Press.

This page intentionally left blank

316

PART III: MEDIATING WORKER PARTICIPATION: THE LABOR MOVEMENT AND THE STATE

317

This page intentionally left blank

318

OPPORTUNITIES AND CHALLENGES FACING NEW WORKFORCE INSTITUTIONS: A CLOSE-UP ANALYSIS OF AN ALTERNATIVE STAFFING SERVICE Esther B. Neuwirth ABSTRACT Temporary, part-time, and contract workers face a myriad of challenges as they seek to navigate the complex labor market landscape. Working Partnerships Staffing Service (WPSS), a project initiated by one of the most prominent labor councils in the U.S., sought to create a new type of labor market institution – one that could empower contingent workers by innovatively linking job placement with training, benefits, and membership-based services. However, like other social movement organizations that endeavor to combine advocacy and income generation, structural pressures led WPSS to conform in important ways to the dominant private-sector staffing-industry model. I argue that WPSS’s response to these pressures ultimately constrained their ability to successfully innovate. Analyzing the challenges facing new worker-centered institutions, this case study presents important insights on ‘‘next generation’’ union initiatives aimed at better positioning workers in the flexible economy. Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 319–343 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16012-4

319

320

ESTHER B. NEUWIRTH

To mitigate barriers to workplace participation for today’s growing contingent workforce, a host of alternative stafﬁng agencies has emerged, providing disadvantaged temporary workers with opportunities for job placement, training, portable beneﬁts, community, and political power. Private-sector stafﬁng agencies (formerly referred to as temporary-help agencies) are the largest and most visible type of Labor Market Intermediary (LMI), with a growing number of employers and workers turning to these agencies for help navigating the complex and uncertain conditions of contemporary labor markets (Benner, 2002; Smith, 2001). But while private-sector agencies primarily orient themselves to meet employers’ needs, sometimes to the detriment of their own workforce, alternative stafﬁng agencies are typically worker-centered institutions mitigating barriers to workplace participation for their employees (Benner, 2002). Alternative stafﬁng agencies strive to provide temporary workers with access to quality jobs, helping them obtain living wages and equitable terms of employment. However, efforts to create these worker-centered agencies face signiﬁcant challenges. This article presents a case study of the emergence and eventual closure of one alternative-stafﬁng agency, Working Partnerships Stafﬁng Service (WPSS), illustrating and analyzing the multiple challenges new institutions face when they try to better position workers in the contingent economy. WPSS emerged in the late 1990s as part of the national effort of the AFLCIO and Central Labor Councils to revitalize organized labor’s role across the country (Meyerson, 1998). Restructuring labor markets in the Silicon Valley region throughout the 1980s and 1990s led ﬁrms to rely increasingly on contingent workers and stafﬁng agencies (Barley & Kunda, 2004; Benner, 2002). By 1998, approximately one-third of the workers in Silicon Valley’s Santa Clara County had some form of non-standard employment, and stafﬁng agencies employed approximately 10 percent of the region’s workforce (Benner, 2002, pp. 44–45). Silicon Valley ﬁrms’ rapid expansion of their outsourcing practices during the economic growth period of the 1990s is a trend toward temporary work arrangements that has since spread through much of the U.S. (Mishel, Bernstein, & Boushey, 2003; Mishel, Bernstein, & Allegretto, 2005; Barley & Kunda, 2004). In response to this market restructuring, the South Bay AFL-CIO Labor Council – one of the most prominent of the 600 Central Labor Councils in the U.S.1 – sought to develop a way to protect workers in the Silicon Valley region from the ﬂexible economy’s uncertainties. As of 2001, the most recent year in which such data were available, the alternative stafﬁng sector comprised nearly 90 different organizations, most of which were established after 1995 (Carre et al., 2003). The majority of

Opportunities and Challenges Facing New Workforce Institutions

321

these alternative stafﬁng agencies primarily provided services to improve the labor-market experience of low-wage, disadvantaged, temporary workers (Carre et al., 2003), and like these organizations, WPSS set out to develop an organizational model that could address ﬁrms’ need for ﬂexible workers while balancing workers’ needs for beneﬁts, training, and representation in the marketplace.2 But WPSS also set out to forge a new path – combining delivery of services for both low-wage and higher-wage temporary workers with business proﬁtability and a union-oriented advocacy agenda. The WPSS model, based on an explicit critique of new corporate practices that disadvantage workers (Benner, 2002; Benner & Dean, 2000), tried to upgrade conditions for a wide range of California’s Silicon Valley temporary and contract workers by creating a ‘‘high road’’ stafﬁng agency; its ultimate goal was to change stafﬁng industry and ﬁrm hiring practices. The founders of WPSS sought to challenge the oppositional structure between labor and capital by forging a new model that could combine proﬁt making with worker advocacy and improved conditions for workers. However, powerful forces stood in their way. Five years after its formation, despite becoming self-sustaining and gaining national recognition among progressive-minded leaders and social entrepreneurs, the leaders of Working Partnerships (WP) and the South Bay AFL-CIO Labor Council (SBLC) chose to close down their stafﬁng service organization. WPSS’s efforts to mobilize a temporary and occupationally diverse constituency across multiple worksites were courageous. Similarly, its goals of providing comprehensive (and costly) training and beneﬁts to workers while maintaining low fees to client ﬁrms – expecting that ﬁrms would have fewer obstacles to permanently hiring WPSS workers after a brief three-month (temp-to-perm) period – broke new ground for alternative stafﬁng service agencies. Yet the end result of these efforts presents a cautionary tale. Analyzing the WPSS experience provides insights about the potential for forging new institutions that can mitigate barriers to more equitable workplace participation for the contingent labor force.

CONTINGENT EMPLOYMENT AND ITS DISADVANTAGES Temporary workers face a host of challenges. They consistently receive lower pay than regular full-time workers and they typically lack health insurance and pension beneﬁts (Benner, 2002; Kalleberg, Reskin, &

322

ESTHER B. NEUWIRTH

Hudson, 2000; Mishel, Bernstein, & Boushey, 2003; Vosko, 2000). Temporary workers also have fewer opportunities for upward mobility, and often perform devalued or deskilled work. Many of these workers lack meaningful training opportunities, and they are often excluded from collective bargaining rights or seniority privileges (Autor, 2000; Benner, 2002; Gonos, 1994; Gonos, 2001a, b; Rogers, 2000). Research has shown that even though employee-involvement programs may integrate some temporary workers, these same workers face very real employment disadvantages, including their positions’ temporary nature and their ineligibility for beneﬁts packages afforded to permanent workers doing similar work (Smith, 2001). And unlike most independent contractors who report relative contentment with their employment status, workers employed by temporary help agencies or in jobs of limited duration overwhelmingly report a preference for standard employment (Kalleberg et al., 2000; Osterman, 1999). The reasons for this preference are easily understood given the context of their labor: temporary workers earn, on average, 67 percent of standard employees’ wages, and less than half of temporary-help agency workers are covered by health insurance (Osterman, Kochan, Locke, & Piore, 2001, pp. 84–89). Although only 3 percent of the labor force today is employed by temporary-help agencies, the total number of workers employed by temporary help agencies has increased dramatically, from approximately 0.5 million in 1979 to over 3.5 million in 1999 (Osterman et al., 2001). Temporary employment as a whole has grown by approximately 11 percent per year since 1972 (Kalleberg et al., 2000; Segal & Sullivan, 1997). In California, the site of this study, the stafﬁng/personnel services industry was responsible for 10 percent of the new jobs created in the state between 1997 and 2000 (Pastor et al., 2003, p. 4). With the recession that began in 2000, the stafﬁng industry as a whole, and agencies in Silicon Valley in particular, suffered signiﬁcantly, though not in the long term. Between the end of the ﬁscal year 2000 and the ﬁrst quarter of 2002 national employment in the stafﬁng industry fell by 28 percent (Barley & Kunda, 2004, p. 317; Berchem, 2003), as many ﬁrms responded to the recession by reducing the number of contractors they employed. However, a survey conducted during this recessionary period by Kelly Services, a national stafﬁng ﬁrm, found that the proportion of the labor force working on a contingent basis was still 6 percent larger in 2002 than it was in 1998, reﬂecting the exceptional growth of the years preceding the recession. In addition, between March 2002 and March 2003, placements by stafﬁng ﬁrms increased by 10 percent and the number of agency

Opportunities and Challenges Facing New Workforce Institutions

323

employees grew by 11.7 percent (Berchem, 2003). By 2003, the stafﬁng industry was reporting gains in market share even while U.S. unemployment rose (Stafﬁng Industry Report, 2003, p. 2). These ﬁgures indicate that although the economic downturn that began in 2000 may have adversely affected individual stafﬁng agencies, particularly in its early stages, the industry as a whole recovered quickly and forged successfully ahead.

EXPERIMENTING IN NEW FORMS OF EMPLOYMENT – OVERVIEW OF THE SUCCESSES AND CHALLENGES OF WPSS By the late 1990s, the private-sector stafﬁng industry employed a great many temporary workers in the Silicon Valley region, and the disadvantages that stafﬁng agency employment held for workers had become widely known. For these reasons, WP – USA, a non-proﬁt policy institute afﬁliated with the SBLC,3 decided to launch a non-proﬁt business with a social mission. In spite of the serious obstacles the organization faced as it tried to compete with the well-entrenched private-sector stafﬁng industry, including the challenge of covering its costs in a difﬁcult economic climate, WPSS effectively mobilized the traditional stafﬁng service organizational model in novel ways. WPSS linked advocacy with comprehensive services such as job placement, training, and beneﬁt provision. By mobilizing a hybrid organizational model built around a self-conscious integration of more traditional modes of service and resource provision with political advocacy (Minkoff, 2002), WPSS was able to leverage multiple sources of support and funding. WPSS drew creatively on developments in the quite different and sometimes competing ﬁelds of organized labor, non-proﬁt/public sector workforce development, and private industry. Operating at the intersection of these ﬁelds allowed the organization to actualize four central goals, which minimized barriers to workplace participation and improved the conditions for some temporary workers in the region. These goals included: (1) placement of workers in jobs that met living wage standards; (2) delivery of comprehensive training programs; (3) delivery of beneﬁts that went beyond what was being provided by private stafﬁng services (including health insurance, paid holiday, vacation, and sick leave); and (4) high placement rates for workers in temp-to-perm jobs. By the fall of 2001, three years after the organization was formed, WPSS had successfully developed a strong organizational infrastructure and was

324

ESTHER B. NEUWIRTH

well on its way to generating enough sales to become self-sustaining. However, as it struggled to both legitimize its services and generate proﬁts, WPSS reproduced many stafﬁng-industry marketplace norms, losing some of its unique qualities and ﬁnding itself constrained by the contradictions inherent in working across different ﬁelds. In particular, WPSS regularly grappled with the entrenched dichotomy between organized labor and capital. Its primary goals, to advocate for improved rights and employment conditions for its temporary workers and to develop a self-sustaining and proﬁt-generating stafﬁng service, often conﬂicted. Contradictions and challenges arose for WPSS that eventually contributed to the decision to close the organization.

RESEARCH METHODS AND DATA COLLECTION Although there is an increasing interest in the role that new types of hybrid organizations can play in sustaining social movements and in instituting real political and economic change (Alperovitz, 2005; Greider, 2003; Minkoff, 2002), few systematic ethnographic case studies exist to illuminate the daily challenges and opportunities facing new forms of social entrepreneurialism. Most studies of these types of organizations typically rely on interview data or site visits, and thus lack the depth that ethnographic data can provide. In contrast, this study employs ethnographic as well as interview and archival data. Triangulating methods in this way allowed me to cross-check information gathered on organizational practices, organizational mission/goals, program implementation, staff roles, relationships among staff and members, and WPSS’s relationships with other institutions. The study’s fundamental research questions included: (1) Under what conditions did WPSS emerge? (2) What activities, strategies, and practices does WPSS conduct in order to meet its goal of improving access to employment, upgrading the quality of temporary assignments, and changing stafﬁng-industry practices? (3) How does WPSS identify its market niche and make connections with workers and customer businesses? (4) What are the organizational, policy, and sociological lessons of WPSS’s successes and challenges? From July through October 2001, WPSS gave me broad access to the organization’s resources, permitting me to conduct research as a participant observer of many of its activities.4 At WPSS, I held a part-time unpaid position as the coordinator of the health insurance beneﬁts program and was able to observe ﬁrst-hand the organization’s inner workings.

Opportunities and Challenges Facing New Workforce Institutions

325

I conducted over 500 hours of participant observation at WPSS. Throughout a four-month-period spent working at the organization, I observed WPSS’s employee placement, recruitment, and orientation practices. I also observed its approach to client sales and marketing. I worked closely with WP’s membership and training program staff, regularly attending internal staff meetings, temporary-worker membership-association meetings, and worker trainings. I also observed a daylong WP staff retreat, where I gained insights about how the stafﬁng service and the temporary-worker project as a whole ﬁt into the larger political and organizational agendas of WP and the SBLC. Besides allowing me access for participant observation, WPSS leaders helped me arrange interviews with their staff, client ﬁrms, members, and temporary-worker employees. I conducted over 33 hours of semi-structured interviews, nearly all of them in 2001; the last 3 hours of interview data, gathered in order to obtain follow-up information on the closure of WPSS, were collected in 2005. Interviews were transcribed and transcriptions were coded to identify the most salient issues (Berg, 2001; Loﬂand & Loﬂand, 1995). Throughout my analysis, for the purpose of protecting my informants, I maintained the conﬁdentiality of individuals I interviewed and observed. Finally, WPSS also allowed me to conduct archival research, granting me access to many of its internal organizational documents, which included organizational planning and marketing documents, foundation proposals and progress reports, business plans, and three years’ worth of minutes from staff meetings. I had full access to its library of information on the stafﬁng industry, which included valuable marketing material and stafﬁng-industry association newsletters. Drawing on these archival data, I further crosschecked information gathered in interviews. Analyzing this archival material gave me great insight into the historical trajectory of the organization and the issues that arose as it changed and adapted to new circumstances. The case study data presented here reveal a unique organization that tried to combine a political advocacy agenda with the provision of services. Therefore, the ﬁndings reported cannot be generalized to other alternative stafﬁng agencies or private-sector stafﬁng agencies. However, components of WPSS’s approach mirror the practices and missions of other alternative stafﬁng agencies, and to some extent private-sector stafﬁng agencies. Thus, there is much to learn from the WPSS experience. Studying the efforts of WPSS may be particularly instructive for understanding the challenges facing the contingent workforce, for whom the promise of a better future appears to lie in the creation of similar alternative institutions. Furthermore,

326

ESTHER B. NEUWIRTH

the WPSS case provides new insights on more generalizable obstacles to organizational innovation and social change.

GETTING OFF THE GROUND: DEVELOPING ORGANIZATIONAL CAPACITY AND A STRONG INFRASTRUCTURE In 1999, when WPSS ﬁrst began recruiting and placing temporary workers, it focused its efforts on reaching out to clerical workers, the most prominent group of temporary workers. One early goal of WP’s temporary-worker project was to create solidarity among these low-wage clerical workers, a particular population of disadvantaged job seekers. To accomplish this, it provided workers with a vehicle to obtain employment through WPSS, the stafﬁng component of the organization. The organization also endeavored to provide workers with training opportunities, through Working Partnerships Training Program (WPTP) in order to improve their job prospects. The organization attempted to facilitate communication and solidarity among clerical workers through the Working Partnerships Membership Association (WPMA), the organizing and advocacy arm of the organization that provided health care beneﬁts and avenues for collective action (Benner, 2002). Unlike their private-sector stafﬁng agency counterparts, WPSS tried to set a new standard in the industry by following the guidelines set by the Code of Conduct,5 an agreement developed by a diverse group of community-based organizations and unions, in collaboration with WP. The code speciﬁed the need for temporary workers to earn living wages,6 and to have access to health beneﬁts, training, and fair administrative treatment. Approximately three years after its inception, WPSS had approximately 750 employees registered with its stafﬁng service, a signiﬁcant volume of worker placements across different types of job placements,7 and ﬁrst-quarter growth for ﬁscal year 2001 of 45 percent. A signiﬁcant percentage, over 50 percent, of WPSS’s temporary workers was converted to permanent workers at client ﬁrms after an initial three-month temp-to-perm period. Modeling itself after dominant private-sector stafﬁng services, WPSS changed its strategy and broadened the scope of its placement services beyond the occupationally speciﬁc pool of clerical workers. For a combination of reasons, including the desires to reach out to a larger segment of the contingent workforce, to gain signiﬁcant market share, to generate greater proﬁts, and to meet a wider scope of client ﬁrm needs, WPSS began recruiting and placing a range of different salaried (low-to-high) workers in

Opportunities and Challenges Facing New Workforce Institutions

327

multiple occupations and professions. The stafﬁng service ﬁlled both temporary and temp-to-perm job openings for employees in the areas of accounting, administrative and clerical work, light industrial, customer service, information technology, and various other technical positions.8 Through WPTP, the training component of WP, WPSS could offer its employees different types of training programs. With the help of the Black Chamber of Commerce and several foundation grants, WPTP had acquired a group of computer stations and a sophisticated facility for training that focused on IT and clerical/word-processing skills upgrading. Over a twoyear period, WPTP trained over 100 workers, approximately 15 percent of whom were placed in jobs through WPSS. Through WPMA,9 the membership-based component of the organization, WPSS could offer its employees opportunities to network, engage in advocacy work, and access a hotline that provided information regarding legal matters that frequently arise from problems associated with temporary employment. This combination of employment-related services far exceeded the types of service offered by private-sector stafﬁng agencies10 (see Table 1). By the fall of 2001, WPSS had successfully developed a strong organizational infrastructure that included a professional ofﬁce environment in a prominent downtown area. WPSS drew upon long-standing ties between its founders and the City Council to arrange for low-cost ofﬁce space designated for the incubation of new start-up businesses. Reducing rent helped WPSS minimize costs, and having its own ofﬁces in an accessible downtown area of the city helped WPSS more effectively reach potential clients and employees. In addition to the professional ofﬁce space, WPSS had successfully hired and trained a professional team of ﬁve staff members all of whom had stafﬁng-industry experience and past records of placement and client sales that they could draw upon in order to help build WPSS’s capacity. By modeling itself after traditional stafﬁng agencies and hiring a group of personnel with private-sector stafﬁng experience, WPSS successfully built its stafﬁng component of the organization. For example, by hiring a branch manager with extensive stafﬁng-industry experience and – perhaps more important – business-leadership skills, WPSS operationalized its services: this manager led meticulous efforts to set up and effectively use a stafﬁng service-speciﬁc database; developed systems for tracking temporary employees and maintaining computerized records of their vacation, sick leave and holiday beneﬁts so that employees could be encouraged to take advantage of these beneﬁts; and generated new clients and conducted more extensive sales. Previously, WPSS staff had tried to coordinate these tasks, but only haphazardly.

328

Table 1.

ESTHER B. NEUWIRTH

Working Partnerships Workforce Development Programs.

Program WPSS Placement

WPTP Training and Career Counseling

WPMA Membership and Advocacy

Description WPSS provided employees with temporary, temp-to-perm and permanent placements that paid living wages ($10 an hour and above). They also provided employees with sick leave with pay, accrued at a rate of 4 hours/80 hours worked, with a maximum accumulation of 48 hours. Accrual began when the ﬁrst assignment commenced. Temporary employees received paid holidays that included six recognized/observed holidays in a given year; eligibility for holiday pay began after completion of 80 hours on assignment with WPSS. A WPSS employee was eligible for one week of paid vacation upon reaching 1,200 hours on assignment with WPSS. And employees had access to a special portable and low-cost health insurance program, accessible with the assistance of WPSS staff WPTP provided free, comprehensive, informationtechnology and clerical/administrative training opportunities (including an A+ Certiﬁcation course and a Cisco Academy training course). These courses were hands-on, classroom-based, instructor-led courses. Training programs were developed in partnership with local community colleges, employers, and unions to improve workers’ career prospects by increasing their skills and providing widely recognized certiﬁcation documentation. Coupled with training, WPTP also provided workers with assessment tests, individual career counseling, seminars, and follow-up sessions to assist workers in developing career action plans WPMA organized regular monthly meetings that facilitated networking, access to temporary worker advocacy programs (such as the Code-of-Conduct campaign to improve working conditions throughout the temporary help stafﬁng industry), and subscription to a member newsletter. WPMA also operated a temporary worker hotline that provided at-large workers with information on the rights and responsibilities of their employer of record, the process for ﬁling for unemployment insurance, and legal matters arising from problems associated with temporary employment

Opportunities and Challenges Facing New Workforce Institutions

329

Developing this strong infrastructure was possible in part because WP had received two signiﬁcant foundation-supported start-up grants, giving the organization time and ﬁscal ﬂexibility to establish its proﬁt-generating components. The organization effectively used its non-proﬁt status to apply for and receive foundation and state funding to run its programs and to obtain funds that partially insulated it from market uncertainties. To survive over the long term, the WPSS business model projected generation of sales revenues from its placement service, so that proﬁts earned from the stafﬁng services could be funneled into beneﬁts, training, and expansion of its worker-centered programs and services. By the fall of 2001, WPSS was successful in generating enough sales to cover the expenses of running the stafﬁng service. However, it was not generating signiﬁcant proﬁts, and the staff felt tremendous pressure to do so because the foundation grants supporting them would eventually run out. With organizational infrastructure in place, WPSS began to concentrate on becoming self-sustaining, with the goal of channeling revenues from placement into ﬁxed costs and improved services for its employees, as well as an advocacy campaign to change industry practices.

CONSTRAINTS TO INNOVATION: STRUCTURAL AND ORGANIZATIONAL CHALLENGES With its organizational infrastructure, WPSS, although an organization initiated by organized labor, modeled itself after successful private-sector stafﬁng services. Doing so tied it to a traditional model that led to both internal and external pressures to conform. In addition, WPSS struggled to ﬁnd a way to generate proﬁts, while simultaneously delivering comprehensive beneﬁts to its employees. Modeling or Innovating? As economic sociologists would predict, WPSS faced structural pressures that often led it to conform to the dominant stafﬁng-industry model rather than to innovate. In contrast to neo-liberal economic theorists, whose arguments depend on maximization of economic efﬁciency to explain why one form of organization predominates over others (Williamson, 1994), economic sociologists argue that structural forces tend to lead organizations to homogenize in spite of individual efforts to innovate (Granovetter &

330

ESTHER B. NEUWIRTH

Swedberg, 1992). In particular, the sociological theory of isomorphism suggests that organizations that seek to innovate face both powerfully coercive and subtle pressures that often limit variability. These pressures typically stem from some combination of the following three factors: (1) political inﬂuence and problems of legitimacy, or coercive isomorphism; (2) standard responses to uncertainty, or mimetic isomorphism; and (3) pressures associated with professionalization, or normative isomorphism (Powell & DiMaggio, 1991, pp. 66–67). The theory of isomorphism suggests that after disparate organizations in the same line of business coalesce into an actual field – a group of organizations that constitute a recognized area of institutional life – powerful forces may lead them to become similar to one another. Organizations may change their goals or develop new practices, and new organizations may enter the ﬁeld, but organizational actors construct around themselves an environment that constrains their ability to change (Powell & DiMaggio, 1991). This concept of isomorphism sheds light on why WPSS, although seeking to create a new type of alternative stafﬁng service, sometimes conformed to the dominant stafﬁng-industry model rather than attempting new approaches. For example, WPSS initially lacked legitimacy in its relationship with both client ﬁrms and prospective workers because it was a new organization with minimal name recognition and no substantial track record of client sales or employee recruitment and placement. This led WPSS to follow the practices of successful private-sector stafﬁng agencies by developing intake forms and processes for workers that were similar to those they encountered at other stafﬁng agencies. WPSS also developed outreach material and practices for clients that mimicked accepted stafﬁng-industry practices. In addition, WPSS staff, most of whom were hired to work at WPSS after having had extensive experience with private stafﬁng services, had a similar professional and ideological orientation that partially contributed to organizational conformity rather than to innovation. Particularly notable at the time of my observation was that WP staff had recently developed a new strategy to disentangle WPSS from the other WP and SBLC programs. The logic behind this strategy was that in order for the WPSS staff to do their job effectively, they needed to focus on making the stafﬁng ‘‘business’’ successful. Being involved in the other WP programs and participating in WP meetings and decision-making processes, the staff told me, took them away from their work and made them less effective, limiting their ability to focus on the enormous task of creating and managing the business side of the alternative stafﬁng service. Indeed, as research has

Opportunities and Challenges Facing New Workforce Institutions

331

shown for similar non-proﬁt organizations that provide training and placement, one of the biggest challenges is remaining focused and continually innovating in creative ways (Harrison & Weiss, 1998, p. 5). From 1999 to 2001, the initial start-up phase of the organization, WPSS staff were integrated into other WP activities, thereby maintaining a strong connection between WPSS and the political-advocacy arm of the organization. The WPSS ofﬁce was located amidst the other WP and labor council staff ofﬁces, and although the programs were distinct, they often shared staff and resources, had access to one another, and communicated easily with one another. However, in the fall of 2001, in order to provide WPSS with the ability to focus on enhancing the professional ‘‘look and feel’’ of the stafﬁng service, and to make the stafﬁng service more accessible to employees and clients, WPSS moved its ofﬁces to a separate building located several miles from the other WP ofﬁces. This move enabled WPSS to differentiate its services from the labor council and WP programs and to insulate the WPSS staff so they could focus their efforts. However, the move effectively separated WPSS from the other WP and labor council staff and programs, distancing WPSS from its labor roots and political commitments. Furthermore, a signiﬁcant challenge that arose out of the move was that WPSS was not able to effectively communicate and work collaboratively with the other WP programs. As one staff member explained, ‘‘our biggest challenge now because of the move is probably the communication, because they’re over there, we’re over here. We don’t interact very often.’’ In addition to the ofﬁce move, drawing individuals from the privatesector stafﬁng industry to staff WPSS created unique problems.11 Research has shown that achieving a dual bottom line, on the one hand generating revenues, while on the other hand advancing a political/social mission, demands staff who possess a combination of strong marketing skills, social/ political sensitivities, and business acumen. Finding, attracting, and retaining staff with these attributes have proven difﬁcult for other social-purpose stafﬁng services (Eisenberg, 2003, p. 42). WPSS’s strategy to hire staff with stafﬁng-industry experience had advantages, because such hires provided the organization with experienced professionals who knew about the stafﬁng industry. But the disadvantage was that the WPSS staff exhibited an orientation that encouraged WPSS’s organizational behavior to mimic the standard industry model. Because WPSS operated in an uncertain environment and presented ambiguous organizational goals (such as the goal to improve employment conditions for temporary workers), staff often ended up modeling their approach on the legitimate and successful stafﬁng companies they had

332

ESTHER B. NEUWIRTH

worked for in the past, where the primary goal was proﬁt generation. One WPSS staff person explained: This is very hard for me right now, because I have always been very successful, and now I’m not successful like I’m used to. And it’s for a variety of reasons y . So, I think that I will pattern this after my other successes, because I’m not changing the way I approach things, and I’m really not changing my belief in the industry y . It’s a little more exciting to be able to talk about the extraordinary things that we do for our employees, but I don’t think that I’m patterning it any differently than I have in the past, because I can only pull from my own experience from the past to do what I’m doing now.

That experience meant that rather than draw on the strengths of WP and the SBLC – their reputations in the region and their political and social networks – WPSS staff confronted the enormous challenges of creating an alternative stafﬁng agency primarily by pursuing the traditional stafﬁngindustry practice of ‘‘cold calling’’ clients, approaching client ﬁrms/organizations that had no prior contact with WPSS or its founders. Generally, they deﬁned their jobs and daily tasks at WPSS, including their approach to employee recruitment and placement, and sales to client ﬁrms, based on their private-sector experiences. The relocation of their ofﬁces, weakening ties to the more politically engaged WP programs, whose goals they insufﬁciently understood, exacerbated this disinclination to innovate. Also producing strong effects were coercive isomorphic pressures – both formal and informal pressures exerted on WPSS by client corporations upon which they depended, and by the cultural expectations in the environment where they operated. These types of pressures led WPSS to downplay and sometimes even conceal its non-proﬁt status and afﬁliation with unions. For example, at times, WPSS’s connection to unions or the very mention of its non-proﬁt status created problems for WPSS staff in their sales outreach to corporate clients. It was common knowledge that the IT sector in the region, the most dominant sector of the economy at the time of this study, was not supportive of unions. This anti-union sentiment exerted a great deal of pressure on WPSS. WPSS’s success depended in large part upon its ability to generate new clients, and to do so, WPSS sometimes had to downplay its alternative mission in order to conform to more traditional stafﬁng service procedures and practices. As one WPSS staff member explained: You can’t talk to everybody about unions, and that our whole premise of our beneﬁts is the result of [our connection] to the AFL-CIO. It’s all a derivative of the union premise of taking care of their employees. I can’t sit and articulate that to many people y . You have to be keenly aware of who you are talking to and what the situation is in their organization.

Opportunities and Challenges Facing New Workforce Institutions

333

In interviews with client ﬁrms, I learned that many of WPSS’s corporate clients were not aware of WPSS’s non-proﬁt status or union connections, and they were in fact not interested in or concerned about that aspect of the organization. What seemed to matter most to clients were the services that WPSS provided them. One hiring manager, when asked about WPSS’s mission explained, ‘‘Yes, well, a while back they sent me their brochure y but I don’t really remember what it says. No, I didn’t realize they were a non-proﬁt.’’ Although one of WPSS’s goals was to inﬂuence the beliefs and practices of hiring managers, client ﬁrms were often oblivious to the labor orientation and worker-centered approach of WPSS. WPSS’s desire to focus on sales (in order to generate proﬁts) meant that they had less time and/or motivation to educate hiring managers about their innovative agenda to meet ﬁrm demands while also improving conditions for temporary workers.

Generating Profits or Advocating for Workers? Trying to both improve the employment conditions of temporary workers and to generate proﬁts constantly challenged WPSS. The WPSS staff, who were primarily oriented toward getting new clients (sales with ﬁrms) and making placements, did not always share the goal with the other WP staff of organizing temporary workers. For the most part, the WPSS staff did not implement this latter goal in their daily work. The challenge of building an organization that operates across different ﬁelds, particularly the ﬁelds of organized labor and the stafﬁng industry, where staff may have different orientations and goals, proved difﬁcult. This combination of goals also runs counter to the normative organizational behavior of a traditional stafﬁng service, making it that much harder for the WPSS staff to manage. Trying to bring together the divergent goals of income generation and worker mobilization into one organization sometimes provoked tension among the staff and made it difﬁcult for them to move forward in their work. One WPSS staff member explained: We’re servicing two people. We’re servicing employees and clients. We’ve got to be fair to both of them. You know what I mean? Even though our main thing is to make sure our employees get treated fairly and they get what they deserve, we also have to treat our customers with respect and give them quality employees and do the things that we have to do to do that. So, we can’t get political, we can’t get into the advocacy part like they [referring to WP and WPMA staff] do because it would hurt us y . We have to be neutral y because if companies start seeing that you’re politically active it’s going to turn them away from us and that’s going to hurt us in implementing our mission y . The job of the Labor Council and Working Partnerships, USA – they’re the ones who try and

334

ESTHER B. NEUWIRTH

get legislation passed. Where we’re going to make an impact is by volume, by servicing companies, by having a lot of employees working for us y .

WPSS was caught in the bind of trying to accommodate both capital and labor, and simultaneously meet its own need to generate proﬁts to survive. For example, WPSS’s temp-to-perm conversion policy encouraged ﬁrms, after a 90-day waiting period, to hire workers on a permanent basis because WPSS ‘‘waived’’ the conversion fees. But this policy, which beneﬁted both capital and labor, undermined WPSS’s ability to generate proﬁts. Typically, stafﬁng services charge a conversion fee that earns proﬁts for the agency. For WPSS, their ‘‘no conversion fee’’ policy encouraged ﬁrms to make employees permanent. And yet, WPSS staff explained, this policy was ﬁscally risky for them, because every month their placement hours and proﬁts declined as their temporary workers became permanent. With many WPSS workers becoming permanent after just three months, WPSS lost revenues and members. While only approximately 24 percent of private-sector stafﬁng agency temporary workers is hired on a permanent basis by client ﬁrms (Osterman, 1999, p. 104), over 50 percent of WPSS temporary workers became permanent workers at WPSS client ﬁrms after approximately three months. WPSS staff explained that they were very proud of this practice because it supported upward mobility for their employees, however, it also undermined their organizational viability. Maximizing Profits or Delivering Services and Mobilizing Members? Although WPSS’s goal was to create programs that would provide advantages to workers, their programs for health insurance beneﬁts, training, and membership all faced tremendous challenges. For the health insurance beneﬁts program, the biggest challenges were developing a cost-effective program to cover all of WPSS’s employees, and attracting temporary workers to the program they had available at the time. As the literature on contingent workers would predict, the short-term nature of temporary work deterred workers from signing up for WPSS’s health care beneﬁts (Middleton, 1996, p. 568). Many of the WPSS temporary workers were ﬁrst and foremost focused on getting jobs, only after which might they consider securing beneﬁts. The health care plan WPSS adopted (a program administered through the Kaiser Family Foundation’s Kaiser Permanente Steps Plan) presented intrinsic challenges. Although it was much less costly than equivalent plans available on the private market, workers themselves had to pay for the full

Opportunities and Challenges Facing New Workforce Institutions

335

cost of coverage. Only low-wage temporary workers could qualify for the program, and because WPSS had broadened the scope of its employee recruitment and placement to provide services to a range of different-salaried contingent workers, a signiﬁcant portion of WPSS employees did not qualify for the plan. Moreover, it took two to three months for the health care plan provider to process a worker’s application, and employees had to reapply each year to qualify for the program. Length of time to coverage, difﬁculties in paperwork submission, and having to reapply annually were all factors that deterred employees from applying for WPSS’s health insurance beneﬁts program. WPSS staff were well aware of these problems and they were working with the WP research team to develop a new program that could meet the needs of a larger number of WPSS employees. However, although WPSS had hoped to provide health insurance to its employees like other alternative stafﬁng agencies, it was limited in its efforts because of the administrative and ﬁnancial challenges associated with providing coverage to a mobile and occupationally diverse workforce.12 Although the WPMA attracted many members, its challenge was to create a viable agenda around which to mobilize members and thus change practices in the stafﬁng industry that disadvantage workers. As labor researchers have found, often the most successful organizing campaigns combine both top-down leadership approaches (informed by knowledge of the issues and industry strategy) with bottom-up rank-and-ﬁle support (Milkman, 2000). WPMA had the leadership support mechanisms in place, but it struggled to develop rank-and-ﬁle support for its agenda. Efforts like WPMA aimed at mobilizing temporary workers face signiﬁcant challenges because of the ‘‘temporary’’ nature of workers’ jobs and the fact that these workers are fractured across worksites and occupational groups, making it difﬁcult to forge solidarity. In addition, one WP staff member explained, the temporary workers coming to WP were most concerned with ﬁnding work and earning a living:

I think, quite simply, the reason why [WPMA] is struggling is because it seems that the goals of the association are to create this union of temporary workers who want to advocate for workers’ rights. And the problem is that, what we’re ﬁnding is, in the membership association, that it is composed of a lot of people who are barely making it. And if you are barely making it, the last thing on your mind is advocating for anything that doesn’t have to do with putting food on the table, and making the rent, and that kind of thing y . I think that it’s really difﬁcult to y look beyond their immediate needs.

336

ESTHER B. NEUWIRTH

For WPTP, the challenge was to connect new trainees with WPSS jobs. At the time of this study, only 15 percent of new trainees (out of a group of approximately 100 trainees) had been placed by WPSS. The remaining WPTP trainees, whose cost for training the organization absorbed, often found jobs through other placement agencies. WPSS struggled to place the new WPTP trainees for a variety of reasons, including the fact that WP’s training program did not always focus on the job readiness skills applicable to WPSS jobs. In addition, WPSS’s policy to only place workers in jobs that paid $10 an hour and above, to abide by the Code of Conduct, and to meet the living wage ordinance, also contributed to this challenge, because the higher-wage WPSS jobs were not always suitable for low-skilled or newly trained WPTP job-seekers.

LEGAL, STRUCTURAL, AND ORGANIZATIONAL CHALLENGES ASSOCIATED WITH SERVICING AND MOBILIZING THE CONTINGENT WORKFORCE WPSS took on enormous challenges. On the labor side, they were charting new territory by trying to recruit, train, place, and eventually mobilize a temporary workforce from a wide range of occupational sectors across multiple workplaces. On the capital/business side, they tried to become a self-sustaining business while providing costly training and beneﬁts for a workforce that experienced high rates of turnover – in part due to their successful temp-to-perm conversion practice. WPSS also faced signiﬁcant legislative obstacles in its efforts to mobilize temporary workers. The organization formed in large part because traditional union strategies for organizing temporary workers were blocked. Legislation makes it very difﬁcult to organize workers across multiple worksites, as organizers seeking to mobilize temporary workers must make the legal case that temporary workers constitute a ‘‘community of interest.’’ This is especially difﬁcult when workers are temporary for a short period of time, when they move from one job site to another, or when they are assigned on a seasonal basis (Mehta & Theodore, 2000–2001). Although a recent ruling has facilitated mobilization of long-term temporary workers, known as permatemps, it remains hard to organize shortterm temporary workers in the light of legislative restrictions (Mehta & Theodore, 2000–2001).13 Moreover, the predominant models for union organizing in conjunction with the contemporary labor legislation framework

Opportunities and Challenges Facing New Workforce Institutions

337

are heavily weighted toward bargaining and organizing at the worksite or ﬁrm level rather than at multiple ﬁrms or across industries. Union initiatives and labor rulings have produced some changes to this model, but the old model still predominates (Mehta & Theodore, 2000–2001). Researchers have also identiﬁed generalizable structural obstacles that arise when trying to create a shared identity and mobilize the temporary workforce. WPSS, grappled with many of these. Given that temporary workers often work in multiple location, their contact with one another can be quite limited (Gottfried, 1992; Smith, 1998; Vosko, 2000). In addition, creating a collective identity among these workers is a great challenge, because the lines between temporary and full-time salaried workers in a given workplace are increasingly blurring, and temporary workers span a wide range of occupational groupings (Osterman, 1999, pp. 170–171; Smith, 2001). Further, because job tenure is often short for temporary workers, their commitment to organizing may be somewhat limited and efforts to organize them that much harder (Osterman, 1999). Generalizable organizational and business development factors also pose challenges for alternative stafﬁng agencies like WPSS. These organizations lack the networks and information-sharing institutions that private-sector stafﬁng agencies have with clients and with other private agencies, making it difﬁcult to get the large contracts that many private agencies obtain (Seavey, 1998). The demands of achieving a dual bottom line make the branch or general manager position difﬁcult to ﬁll, creating a serious point of vulnerability for the business (Eisenberg, 2003, p. 40). Cost-cutting competition in local markets can also create a strong downward pressure on hourly billing rates, gross margins for a stafﬁng business, and hourly wages of its workers (Carre et al., 2003; Mehta & Theodore, 2000–2001). Although many alternative stafﬁng agencies have developed responses to cost cutting, this issue remains of great concern. For WPSS, the most effective strategy for resisting rate-cutting approaches was leveraging partnerships with unions, the City Council, foundations, and other institutions to insulate themselves against market pressures. However, WPSS found it difﬁcult to maintain these partnerships and linkages across organizations in order to actualize their mission and goals.

CONCLUSION WPSS’s experiences suggest that it remains to be seen whether it is possible to bridge the divide between labor and capital through an alternative type of

338

ESTHER B. NEUWIRTH

stafﬁng agency. One important lesson learned from this study of WPSS is that working across different ﬁelds has the potential to transform obstacles into strategies for social change. Despite its decision to close down, WPSS did, in its brief existence, successfully deploy a familiar model of the stafﬁng service organization in unfamiliar ways (Clemens & Cook, 1999) by helping over half of its workforce become permanent employees through its temp-to perm conversion process, and by linking job placement with extensive training, beneﬁts, and membership-based services. For WPSS, strategic action involved leveraging multiple resources for support and funding to create a complex, interdependent set of organizations. WPSS achieved success in building an organizational structure with the capacity to bridge the worlds of labor and capital. However, like other social movement organizations that endeavor to combine advocacy and income generation, its money-mobilizing strategies, over time, reduced the resources it devoted to advocacy (Edwards & McCarthy, 2004, p. 139). Trying to reach ﬁscal viability, WPSS simpliﬁed organizational goals rather than pursue the more challenging and complex strategy of integrating the different organizational missions and goals. Although WPSS created a hybrid organization that had the potential to meld advocacy with services, it did not in fact implement its vision or fully actualize its hybrid nature on a daily basis. Perhaps a strategy that challenged the short-term market-driven approach to income generation (which is often associated with the type of foundation support WPSS received) with a long-term vision that could integrate the organization’s proﬁt-making and advocacy goals, might have minimized the isomorphic pressures on WPSS. The tendency to push autonomy can be dangerous but attractive, and WPSS isolated itself in this way. Innovative social ventures need support, both ﬁnancial and visionary, from multiple sources, in order to succeed. WPSS needed the leadership and political imagination of the larger organization and union movement to redirect the staff it hired. Thus, an additional cautionary note from this case study is that funding agencies working with social entrepreneurial organizations like WPSS must understand the necessity of supporting a long-term process. Another important insight gained from this analysis is that the strategy WPSS pursued of servicing an occupationally diverse workforce was inadequate to the challenges of multi-worksite mobilization. A different organizational strategy might have proven more viable – perhaps one that focused on just one occupation or industrial sector, or an occupationally diverse workforce in just one workplace setting. In addition, mobilization of a more stable sector of the contingent workforce, such as perma-temps,

Opportunities and Challenges Facing New Workforce Institutions

339

freelancers, or contract workers might have led to greater continuity in WP’s membership base. Unlike some temporary workers who hope to become permanent workers after a brief temp-to-perm period, these groups of workers maintain their temporary status for longer periods of time, which could facilitate continuity of membership and advocacy efforts. Indeed, Working Today, a national non-proﬁt organization, has been effective in developing a stable membership base of freelancers and consultants, advocating on their behalf and providing them with access to services previously available only to the traditional workforce of permanent employees.14 Similarly, WashTech, the group of Microsoft contract employees that ﬁrst began mobilizing in the late 1990s, is now afﬁliated with the national Communications Workers of America Union, providing its high-tech worker members with community, services, and collective bargaining rights.15 Across the country, there are also a number of occupation-speciﬁc groups of day laborers, health care and home care workers, domestic workers, and landscapers who are developing worker centers or co-operatives to train, place, and/or provide beneﬁts and fair-labor practices to contingent workers.16 WP, the non-proﬁt policy institute that founded WPSS, continues to operate, strategizing about how to best meet the needs of disadvantaged temporary workers. With expertise in the policy arena, it now devotes more resources to the work it began in the area of access to health care, developing and advocating for a new publicly funded adult health insurance program to cover uninsured workers who do not qualify for publicly subsidized health care and cannot afford the monthly premiums of group health plans at alternative stafﬁng agencies. Analysis of WPSS’s efforts should not be mistaken for an evaluation of the organization’s efﬁcacy or the value of its goals. The purpose of this case study of an alternative-stafﬁng agency has been to provide insights into the potential challenges and opportunities facing ‘‘next generation’’ union initiatives aimed at upgrading conditions for the contingent workforce. Further research into other models, including worker-owned co-operative placement agencies and contractor’s-union hiring halls, may shed additional light on how to move beyond the traditional labor/capital divide to safeguard workers from the more deleterious trends in contemporary labor markets.

NOTES 1. For an overview of the changing role of Central Labor Councils, see Gapasin and Wial (1998) and Ness and Eimer (2001).

340

ESTHER B. NEUWIRTH

2. See Carre et al. (2003) on alternative stafﬁng agencies and Eisenberg (2003) on social purpose stafﬁng companies. 3. Both WP and the SBLC are organizations that continue to operate today. 4. This research was part of a larger project that included ethnographic data collection at a for-proﬁt stafﬁng service in the summer and fall of 2000. 5. The goal of the Code-of-Conduct Campaign was to get other temporary agencies (private-sector agencies) to sign the code, thus agreeing to abide by the guidelines that the code established. In addition, employers and employees were encouraged to preferentially do business with those agencies that signed the code. Public institutions were also asked to recognize the code and extend the same preferential treatment to code signatories. 6. Typically, many stafﬁng service jobs pay low wages (minimum wage or just above) and offer few or no beneﬁts; see Kalleberg et al. (2000). At the time of this study, WPSS paid all its workers according to the living wage standard set at $10 an hour or above. Over time, with increases in the cost of living and changes in the hourly wage advocated by the Living Wage Campaign, that wage was raised. There has been a great deal written about Living Wage Campaigns; for a summary, see Osterman et al. (2001, pp. 135–138). 7. At the time of this study, WPSS had approximately 1,000 ‘‘hours billed’’ per week, a number that captures the volume of placements across all types of job placements. Younger stafﬁng agencies tend to have fewer hours billed than older agencies. WPSS’s hours billed were comparable to other alternative stafﬁng agencies of its size and tenure. For more on alternative stafﬁng agencies and ‘‘hours billed,’’ see Carre et al. (2003, pp. 14–15). Private-sector stafﬁng agencies typically have a much higher number of hours billed. For one private stafﬁng agency I observed, their hours billed averaged approximately 8,000 hours a week. Their goal was to average at least 1,000 hours billed each week for each staff person they employed at the branch ofﬁce (Neuwirth, 2004). 8. Although this article is not focused on WPSS’s impact on individual workers or worker outcomes in general, the majority of the 20 WPSS workers, members, and prospective workers I interviewed found that WPSS’s combination of beneﬁts and services were more advantageous to them than those provided by private-sector stafﬁng agencies. The majority of interviewees also found WPSS to be more helpful and more personable than other stafﬁng agencies they had encountered. 9. By 2001, WPMA claimed to have 500 members in the association. However, staff reported that only 10–15 of these members were active members who helped with the advocacy programs and only a handful of these active members came to the regular monthly membership meetings. 10. For more on the range of services provided by private-sector stafﬁng agencies, see Benner (2002). 11. When WPSS was launched in 1999, only personnel with non-proﬁt or union experience staffed the organization. Over time, WP came to realize that in order to develop organizational capacity, they needed to staff WPSS with personnel with private-sector stafﬁng-industry experience. At the time of this study, the WPSS staff had previously worked for private-sector stafﬁng agencies, but had very limited experience working with organized labor and non-proﬁt agencies. 12. For more information on challenges alternative stafﬁng agencies confront in trying to offer health insurance, see Carre et al. (2003).

Opportunities and Challenges Facing New Workforce Institutions

341

13. A detailed discussion of current labor laws and National Labor Relations Board rulings goes beyond the scope of this article. For examples of research on labor laws and how they impact temporary and contingent worker organizing efforts, see Mehta and Theodore (2000–2001); Herzenberg, Alic, and Wial (1998); DuRivage (1998). 14. See http://www.workingtoday.org. 15. See http://www.washtech.org. 16. See Fine (2006) and http://www.fairjobs.org.

ACKNOWLEDGMENTS Data collection and analysis for this article were conducted with support from the Institute for Labor and Employment and a University of California at Berkeley, Agency for Healthcare Research and Quality postdoctoral fellowship. Much gratitude goes to Working Partnerships – USA and the South Bay AFL-CIO Labor Council for allowing me to study their work. Special thanks to Vicki Smith, Fred Block, Chris Benner, and Sean O’Riain for invaluable guidance throughout the duration of my research. And I thank the following individuals for their insights and/or support: Karen Albright, Richard Cunningham, Phaedra Ellis-Lamkins, Kim Foltz, George Gonos, Larry Greer, Patricia Keller, Eileen Otis, Preston Rudy, Bindi Shah, Richard Schefﬂer, Eva Skuratovich, Neil Smelser, Leon Sompolinsky, Michael Stein, and Carol Zabin.

REFERENCES Alperovitz, G. (2005). America beyond capitalism: Reclaiming our wealth, liberty and our democracy. Hoboken, NJ: Wiley. Autor, D. (2000). Why do temporary help firms provide free general skills training? Unpublished manuscript. Working Paper 7637. National Bureau of Economic Research, Cambridge, MA. Barley, S., & Kunda, G. (2004). Gurus, hired guns, and warm bodies. Itinerant experts in a knowledge economy. Princeton and Oxford: Princeton University Press. Benner, C. (2002). Work in the new economy: Flexible labor markets in Silicon Valley. Oxford: Blackwell Publishing. Benner, C., & Dean, A. (2000). Labor in the new economy: Lessons from labor organizing in Silicon Valley. In: F. Carre, M. Ferber, L. Golden & S. Herzenberg (Eds), Non standard work: The nature and challenges of changing employment arrangements (pp. 361–375). Madison, WI: Industrial Relations Research Association. Berchem, S. (2003). ASA’s annual economic analysis of the staffing industry. American Stafﬁng Association. http://www.stafﬁngtoday.net

342

ESTHER B. NEUWIRTH

Berg, B. (2001). Qualitative research methods for the social sciences. Boston, MA: Allyn and Bacon. Carre, F., Herranz, J., Seavey, D., Vickers, C., Aull, A., & Keegan, R. (2003). Alternative job brokering: Addressing labor market disadvantages, improving the temp experience, and enhancing job opportunities – a report of the national study of alternative staffing services. Boston, MA: University of Massachusetts. Clemens, E. S., & Cook, J. (1999). Politics and institutionalism: Explaining durability and change. Annual Review of Sociology, 25, 441–466. DuRivage, V. (1998). CWA’s organizing strategies: Transforming contract work into union jobs. In: K. Barker & K. Christensen (Eds), Contingent work: American employment relations in transition (pp. 377–392). Ithaca: Cornell University/ILR Press. Edwards, B., & McCarthy, J. D. (2004). Resources and social movement mobilization. In: D. Snow, S. Soule & H. Kriesi (Eds), The Blackwell companion to social movements (pp. 116–152). Malden, MA: Blackwell. Eisenberg, S. (2003). Getting to work: ICA’s social purpose staffing companies. Brookline, MA: The ICA Group. Fine, J. (2006). Worker centers: Organizing communities at the edge of the dream. Ithaca: Cornell University/ILR Press. Gapasin, F., & Wial, H. (1998). The role of central labor councils in union organizing in the 1990s. In: K. Bronfenbrenner, S. Friedman, R. Hurd, R. Oswald & R. Seeber (Eds), Organizing to win (pp. 54–68). Ithaca: Cornell University/IRL Press. Gonos, G. (1994). A sociology of the temporary employment relationship. Unpublished doctoral dissertation. Rutgers State University, New Brunswick, NJ. Gonos, G. (2001a). ‘Never a fee!’: The miracle of postmodern temporary help. Working USA, 4, 3–36. Gonos, G. (2001b). Fee-splitting revisited: Concealing surplus value in the temporary employment relationship. Politics and Society, 29, 589–611. Gottfried, H. (1992). In the margins: Flexibility as a mode of regulation in the temporary help service industry. Work, Employment and Society, 6, 443–460. Granovetter, M., & Swedberg, R. (Eds) (1992). The sociology of economic life. San Francisco, CA: Westview Press. Greider, W. (2003). The soul of capitalism. New York: Simon & Schuster. Harrison, B., & Weiss, M. (1998). Workforce development networks: Community-based organizations and regional alliances. Thousand Oaks, CA: Sage Publications. Herzenberg, S., Alic, J., & Wial, H. (1998). New rules for a new economy: Employment and opportunity in postindustrial America. Ithaca: Cornell University/ILR Press. http://www.fairjobs.org. http://www.washtech.org. http://www.workingtoday.org. Kalleberg, A. L., Reskin, B., & Hudson, K. (2000). Bad jobs in America: Standard and nonstandard employment relations and job quality in the U.S.. American Sociological Review, 65, 256–278. Loﬂand, J., & Loﬂand, L. (1995). Analyzing social settings: A guide to qualitative observation and analysis (3rd ed.). Belmont, CA: Wadsworth Publishing Company. Mehta, C., & Theodore, N. (2000–2001). Winning union representation for temps. Working USA, 4, 37–58.

Opportunities and Challenges Facing New Workforce Institutions

343

Meyerson, H. (1998). A second change: The new AFL-CIO and the prospective revival of American labor. In: J. Mort (Ed.), Not your father’s union movement: Inside the AFLCIO (pp. 1–26). New York: Verso Books. Middleton, J. (1996). Contingent workers in a changing economy: Endure, adapt, or organize. New York University Review of Law and Social Change, XXII, 557–622. Milkman, R. (Ed.) (2000). Organizing immigrants: The challenge for unions in contemporary California. Ithaca: Cornell University/ILR Press. Minkoff, D. (2002). The emergence of hybrid organizational forms: Combining identity-based service provision and political action. Nonprofit and Voluntary Sector Quarterly, 31, 377– 401. Mishel, L., Bernstein, J., & Allegretto, S. (2005). The state of working America 2004/2005. Ithaca: Cornell University/ILR Press. Mishel, L., Bernstein, J., & Boushey, H. (2003). The state of working America 2002/2003. Ithaca: Cornell University/ILR Press. Ness, I., & Eimer, S. (Eds) (2001). Central labor councils and the revival of American unionism: Organizing for justice in our communities. Armonk, NY: M.E. Sharpe. Neuwirth, E. (2004). Blurring corporate boundaries: Staffing agencies, human resource practices and unions in the new employment relationship. Unpublished dissertation manuscript. University of California, Davis. Osterman, P. (1999). Securing prosperity. The American labor market: How it has changed and what to do about it. Princeton, NJ: Princeton University Press. Osterman, P., Kochan, T., Locke, R., & Piore, M. (2001). Working in America: A blueprint for the new labor market. Cambridge, MA: MIT Press. Pastor, M., Leete, L., Dresser, L., Benner, C., Bernhardt, A., Brownstein, B., & Zimmerman, S. (2003). Economic opportunity in a volatile economy: Understanding the role of labor market intermediaries in two regions. San Jose, CA and Madison, WI: Center on Wisconsin Strategy and Working Partnerships. Powell, W., & DiMaggio, P. (Eds) (1991). The New institutionalism in organizational analysis. Chicago: University of Chicago Press. Rogers, J. (2000). Temps: The many faces of the changing workplace. Ithaca: Cornell University/ ILR Press. Seavey, D. (1998). New avenues into jobs: Early lessons from non-profit temp agencies and employment brokers. Washington, DC: Center for Community Change. Segal, L., & Sullivan, D. (1997). Temporary services employment durations: Evidence from state UI data. Rep WP-97–23, Federal Reserve Bank of Chicago, Chicago. Smith, V. (1998). The fractured world of the temporary worker: Power, participation, and fragmentation in the contemporary workplace. Social Problems, 45, 411–430. Smith, V. (2001). Across the great divide: Risk, uncertainty, and opportunity in the new economy. Ithaca: Cornell University/ILR Press. Stafﬁng Industry Report. (2003). May temporary help shows gains while U.S. employment climbs to record level. Los Altos, CA, XIV, 2–3. Vosko, L. (2000). Temporary work: The gendered rise of a precarious employment relationship. Toronto: University of Toronto Press. Williamson, O. (1994). Transaction cost economics and organization theory. In: N. Smelser & R. Swedberg (Eds), The handbook of economic sociology (pp. 77–107). Princeton and NY: Princeton University Press and the Russell Sage Foundation.

This page intentionally left blank

344

THE SPIRIT OF OUTSOURCING: CORPORATE AND STATE REGULATION OF LABOR UNDER H-1B VISA AND TANF POLICIES IN THE U.S. Frank Ridzi and Payal Banerjee ABSTRACT This paper examines the experiences of welfare clients on Temporary Assistance for Needy Families (TANF) and Indian immigrant information technology (IT) workers on the H-1B visa to analyze how public– private collaborations in the spirit and practice of outsourcing, i.e. systematic fragmentation and decentralization of both corporate and state activities, function as mechanisms for disciplining labor. Through an analysis of these groups’ parallel experiences with exploitative work and employers in the U.S., this paper identifies how outsourcing is not merely a business model for cross-border trade, but also a key principle, component, and outcome of policy-based neo-liberal economic restructuring.

Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 345–373 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16013-6

345

346

FRANK RIDZI AND PAYAL BANERJEE

INTRODUCTION: TANF AND THE H-1B VISA IN THE CONTEXT OF THE GLOBAL ECONOMY AND THE LOGIC OF OUTSOURCING The mainstream debate on outsourcing, notable in the popular media and press, has raised widespread concerns about outsourcing’s impact on the U.S. workforce employed in skilled and semi-skilled white-collar occupations. Some interpret outsourcing as having severe negative consequences for the U.S. economy and attribute job loss, unemployment, and economic problems to this phenomenon (Dobbs, 2004; Marks, 2004; Thottam, 2004). Yet others claim that outsourcing contributes to job and economic growth not only in the U.S. (Drezner, 2004), but also in the countries where work is being outsourced (Weidenbaum, 2004) – leading to an overall improvement in standards of living. Research on outsourcing has recently been taken up by various constituencies, such as academics, policy analysts, the state, and politicians who have started to investigate this aspect of the economy to determine whether and/or how the domestic labor market has been affected (see, for example, Bhagwati, Panagariya, & Srinivasan, 2004; Koehler & Hagigh, 2004; U.S. GAO, 2004). The popular media, as well as the above-mentioned research-based sources, however, have privileged an understanding of outsourcing which focuses on international trade and the offshoring of white-collar occupations to developing countries. In the process, outsourcing, and by extension the explanation for workers’ agony in the U.S., has been narrowly relegated and associated with the corporate sector and trans-border trade activities. Believed to be a result of corporate economic agendas alone, the idea of outsourcing has accumulated a narrow connotation and has been conceptually severed from the larger and more immediate context of systematic fragmentation and decentralization of both corporate and state activities and methods of operation under neo-liberalism, which have critical significance for workers in all segments of the labor force. Instead of seeing outsourcing merely as a business model for trade in services across national boundaries, we regard it as one of the most significant aspects of global economic restructuring primarily because it can be identiﬁed as a key logic of operation permeating both state and corporate practices within the larger context of globalization. Outsourcing, notable in both the public and private sectors, includes and manifests itself through: increasing decentralization, contractual and temporary work, ﬂexibility, its retreat from assuming direct responsibilities toward workers, and its

The Spirit of Outsourcing

347

emphasis on individualization. In this paper we expand the analytical boundaries of the concept of outsourcing by examining state and corporate policies that shape the experiences of marginalized minorities, such as Indian immigrant IT workers on work visas and low-income welfare clients in the U.S. Taking the lived experiences of welfare clients and Indian immigrant IT workers on H-1B visas in the U.S. as the point of entry, this paper demonstrates how the mechanism of disciplining and regulating the labor of these two seemingly divergent labor groups has come to rely on the logic of outsourcing. This logic enables public–private collaborations regarding labor policies, where the state’s policies reinforce the ability of private capital or employers to manage and control labor for proﬁt maximizing. At ﬁrst glance, these two groups seem to represent antipodal social and economic circumstances. Welfare clientele are predominantly citizens, women, and as a result of institutional inequality, low skilled and provisionally employed in low wage and informal sectors of the service economy. On the other hand, Indian immigrant IT professionals on work visas are overwhelmingly male, highly skilled, and employed in the relatively highwage segment of the service economy. However, both groups are periodically targeted as scapegoats by other labor groups in the U.S. One is chastised for causing increased government expenditures because it is stereotyped as being ‘‘lazy,’’ while the other is critiqued for being too industrious and ‘‘stealing our jobs by doing them for a cheaper wage.’’ We argue, however, that an examination of the lived experiences of these two groups reveals unexpected similarities in terms of their relations to employers (i.e. capital) given how state policies tip the balance of power in the favor of employers. In particular, their disparate, yet in many ways parallel, experiences with exploitative work and employers in the U.S. suggest that the spirit of outsourcing is critical for private–public approaches to labor deployment in global economic restructuring. Both groups ﬁnd their everyday work experiences shaped, to a certain degree, by a restructuring of government policy, which reﬂects, reinforces, and relies upon outsourcing’s ideology of fragmenting and decentralizing processes involved in hiring and in producing goods and services to achieve cost–efﬁciency. This paper will show how, in the cases of welfare clients and immigrants on H-1B visas employed in IT occupations, the state and private capital are operating on a comparable, if not identical, principle. The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) (Public Law No. 104–193, 1996) permitted an unprecedented level of outsourcing of human service provision. It also created the Temporary Assistance for Needy Families (TANF) program and ended welfare’s entitlement

348

FRANK RIDZI AND PAYAL BANERJEE

status by placing a ﬁve-year limit on aid and requiring work in exchange for beneﬁts. This act induced low-wage workers to accept contingent employment in a ﬁeld increasingly dominated by temporary, unstable, and outsourced jobs. The two H-1B visa-related acts, namely the American Competitiveness and Workforce Improvement Act of 1998 (Public Law No. 105–277, 1998) and the American Competitiveness in the 21st Century Act of 2000 (Public Law No. 106–313, 2000) increased the annual quota of H-1B visas for foreign workers in response to corporate demands for high-tech professionals. Despite overall efforts to reduce immigration, H-1B provisions have been used extensively to recruit foreign workers, who serve as the key constituency for subcontracted labor on IT projects outsourced to both U.S.- and India-based consulting ﬁrms that are operating within the U.S. Together, these legislative acts serve to sustain an increased economic reliance on the principle of outsourcing by regulating the labor pools on which it depends. In the process, we argue, the state’s policies help to consolidate the power of private capital over labor. In the case of welfare recipients, the state’s reform policies authorize employers to claim an authoritative position in the issuing of welfare beneﬁts. Likewise, in the case of the H-1B visa this connection is manifested in the ways in which government policy empowers employers to exploit and control immigrant workers. This consolidation of power occurs within a wider context of changing employment relations within the U.S.

THE CHANGING NATURE OF WORK AND EMPLOYMENT RELATIONS The U.S. economy has, since the late 1960s and early 1970s, experienced severe declines in union activity and bargaining power, inﬂationary erosion of the minimum wage, a drop in higher-wage manufacturing jobs, corresponding increases in low-wage, service sector employment, and the globalization of the economy through immigration and trade (Heery & Salmon, 2000; Mishel, Bernstein, & Schmitt, 1999). These changes have resulted from a shift in focus from regional or national to international markets, which has disrupted the previous domestic balances of an ‘‘industrial dualism’’ that included a core economy of large manufacturers and a periphery of small manufacturers and service organizations (Noyelle, 1987). In many developing and advanced capitalist nations, such as the U.S., the

The Spirit of Outsourcing

349

manufacturing core has been replaced by growth in the service-oriented economic sector. These economic shifts have had differential impacts on each of the submarkets identiﬁed in dual labor market theories. While growth in the service sector has been instrumental in producing both relatively well-paid jobs requiring high skills and poorly paid jobs requiring low skills (Sassen, 1998; Smith, 2001), there has been a noted decline in the Primary Labor Market, which ‘‘provides ‘good’ entry jobs that, via internal promotion, ultimately provide higher wages, year round work, fringe beneﬁts (e.g. health beneﬁts), on-the-job training, opportunities for advancement, and generous retirement beneﬁts’’ (Wetzel, 1995, p. 72). Simultaneously, there has been a dramatic growth of the Secondary Labor Market, characterized by ‘‘ ‘bad’ jobs with low wages, episodic work, few or no beneﬁts, no training, and no real chance for advancement’’ (Wetzel, 1995, p. 72). Structural unemployment and underemployment have also corresponded with these economic and labor market shifts, as employers have adopted outsourcing techniques to increase profits and achieve ﬂexibility in hiring and production processes and therefore have subcontracted basic business functions to either domestic (e.g. subcontractors or temp agencies) or foreign contractors and offshore subsidiaries (Hira & Hira, 2005). Although recent scholarship has focused on the widespread use of contingent workers and contractors in the high-skilled IT sectors (see, for example, Barley & Kunda, 2004) and in low-skilled sectors (Heinrich, 2005; Polivka, 1996), and provided insightful details about the practices of staffing agencies, it has not attended to the role of the state, such as through the H-1B and welfare policies, in facilitating ﬂexible hiring. Furthermore, such work has not provided a comparative analysis of the significance of contemporary state policies in shaping contingent work in high- and low-wage sectors. A focus on the terms of immigration and welfare and their intersections with the practices of ﬂexible hiring and subcontracting is necessary to bring the nexus between state policies and interests of capital under scrutiny. During recent decades, and increasingly since the end of the 20th century, the state has assumed a corporate attitude by striving for efﬁciency and ﬂexibility through downsizing and by adopting mechanisms of decentralization and privatization to restructure itself and its various functions (Kettl, 1988; Smith & Lipsky, 1993). Given these trends in the restructuring of the U.S. government, which gained momentum since the 1950s, recent decades have actually been described as ‘‘the era of the contracting regime’’ (Smith & Lipsky, 1993, p. 43). This conceptualization provides a lens

350

FRANK RIDZI AND PAYAL BANERJEE

through which to view the dynamic relationships that have come to exist between the U.S. government and the nonprofit agencies with whom it is increasingly contracting in order to provide services. As Kettl (1988, p. 3) has pointed out, ‘‘While [government programs] have grown enormously in size – more than two and three-fourths times, even allowing for inﬂation, from 1955 to 1985 – the federal bureaucracy has not.’’ As he explains, what accounts for this paradox of larger federal programs without a larger federal bureaucracy is ‘‘the growth of government by proxy.’’ In the time since the end of World War II, the U.S. federal government has acted upon an imperative to do less of the rowing and more of the steering in the proverbial ship of governmental services. This public and private sector restructuring, associated with terms such as deindustrialization, post-Fordism or postindustrial capitalism, the New Economy, and neo-liberalism, and has been based on privatization and on the decentralization and fragmentation of hiring and production on a transnational scale to create ﬂexible conditions of proﬁt maximization and efﬁciency (Bonacich & Appelbaum, 2000; Harvey, 1989; Heery & Salmon, 2000; Noyelle, 1987; Ong, 1991; Persuad & Lusane, 2000). Such changes in the labor market, however, are not automatic and selfemergent, but are enabled and steered by institutional forces of both private and public sectors. Labor markets, as with any markets, are created by selfinterested individuals, market participants themselves, or third-party intermediaries (Rosenbloom, 2002). To better understand how labor markets are thus constructed in the new economy, which is known to demand unprecedented ﬂexibility of workers (Auer & Cazes, 2003; Heery & Salmon, 2000), it is helpful to probe the distinction between the work itself and the terms of work. Benner (2002) suggests a conceptual distinction between ﬂexible work (pertaining to actual activities and skills used in market production) and ﬂexible employment (referring to the contractual relationships of management and compensation) and proposes that although ﬂexible work is essential for competitive success within the new economy, ﬂexible employment is only partly shaped by the importance of ﬂexible work. Factors that engender and shape the dynamics of ﬂexible employment include legal, institutional, and organizational entities, and interests such as employers’ focus on reducing costs and shifting risk to workers. Flexible employment is further mediated by the rising importance of intermediaries, such as job-training programs and employment services, whether they be private sector, membership based, or public sector, that are increasingly called upon by prospective employees and employers to broker and shape the characteristics of employment relationships within a rapidly changing economic

The Spirit of Outsourcing

351

landscape (Benner, 2002; see also Barley & Kunda, 2004; Giloth, 2004). The conditions of ﬂexible employment, to the degree they are normalized by labor market intermediaries, and the general difﬁculties of ﬁnding ‘‘good jobs’’ in a dynamic economy have a consequential impact on the nature and extent of worker participation in the workforce. State-sponsored workforce systems such as welfare-to-work programs and the H-1B visa program have evolved in recent years partly in response to policy failures in the past and partly because of drastic economic changes on a global level (Kazis, 2004). Welfare-to-work and the H-1B visa are both public sector efforts to mediate and regulate the interface between employers and laborers with regard to work that is done in the U.S. They are governmental initiatives amid a wide range of labor market intermediaries that include temp agencies, job-training programs, labor market exchanges, and employment services (such as monster.com). In general, these intermediaries do the work of moderating connections between employers and employees, and vice versa, a service that has seen increasing demand in recent years as length of average job tenure decreases (Osterman, 2004). Although workforce development strategies are broadly conceptualized as serving ‘‘dual customers’’ (employers and workers/job seekers), the relative beneﬁt to each customer is not necessarily commensurate (Giloth, 2004). These arrangements of state involvement with the new economy have critical ramiﬁcations for worker participation as well. The goal of worker participation, or the degree to which decision-making power is extended to workers within their workplace (Poole, 1978; Tsiganou, 1991), is being severely challenged by processes of outsourcing. Typically viewed as Employee Stock Option Plans (ESOP), worker feedback mechanisms, union-related structures and employee-owned companies (Fairﬁeld, 1974; Kaufman & Kleiner, 1993; Kochan, 1984; Logue & Yates, 2001; Pencavel, 2001; Whyte, 1983), ‘‘genuine worker participation entails power redistribution not only at an enterprise, but also at the societal level to involve workers in the decision-making process to extend their citizenship rights in the workplace and thereby enhance their power’’ (Tsiganou, 1991, p. 13). This paper will demonstrate how state labor market intermediaries, as manifested through the state’s welfare-to-work program and the H-1B visa program, can and often do undermine the potential for power redistribution and expression of voice by regulating labor relations in ways that increase individual worker responsibility, decrease institutional accountability to labor, and increase corporate power over labor. As this research will begin to demonstrate, the state is as much an active player in these relations of power as are corporate interests.

352

FRANK RIDZI AND PAYAL BANERJEE

METHODOLOGY AND DATA Recent theoretical work has emphasized the enduring impact of macrosocial forces such as globalization and advanced transnational capitalism on peoples’ local day-to-day lived experiences (Acker, 2003; Alexander & Mohanty, 1997; Burawoy, 1991; Smith, 1990, 1999). This paper seeks to identify such connections in the case of TANF clients and Indian immigrant IT workers on H-1B visas. Our data were collected independently over the period of 1999–2004 in ethnographic and qualitative fashion (DeVault & McCoy, 2002). We each use a combination of interviews, observations, and textual analysis to trace the lineaments of state involvement in employer– worker relations. The body of TANF-related insights was gathered through an institutional ethnographic investigation (DeVault & McCoy, 2002; Smith, 1987, 1990) of daily work practices in the welfare ofﬁces of Mundial County, New York.1 This approach emphasizes the analysis of the ways in which the actions of multiple actors within an institution are interrelated. Data collection included approximately 38 interviews and observations of over 100 instances of client processing in multiple welfare ofﬁce locations and TANF fair hearings proceedings. Mundial County was selected because it includes a mid-sized city (approximately 500,000 residents) and adheres to a nationally publicized best-practices model for implementing the 1996 welfare reforms (Brown, 1997). For this reason, Mundial County was judged to be a suitable site for studying the local implementation of TANF in an urban context. Since client processing in welfare ofﬁces comprises a delicate setting, permission to conduct qualitative research was requested only after the completion of an on-site quantitative research project that was designed to answer questions presented by the agency. Once the quantitative phase of the research was completed, sufﬁcient rapport was developed with administrators and staff, who granted permission to research with the provision that client names and identities remain conﬁdential. Insights on the impact of H-1B visas on immigrant workers in IT occupations are based on approximately 40 in-depth semi-structured interviews with Indian immigrant IT workers in the U.S. Interviews and participant observation at select work sites were conducted in ﬁve states in the U.S., four located on the east coast and one in the south. Contrary to popular belief, these immigrants’ services are used not only by software and computer ﬁrms, but also by the entire gamut of corporations in the U.S. Clients for whom interviewees had worked represented insurance ﬁrms, hotel chains, ﬁnancial institutions, banks, and web-based companies located across the country. Therefore, interviews were conducted at different sites to

The Spirit of Outsourcing

353

analyze common trends across a wide sample. Interviewees were recruited through snowball sampling and were selected based on their status as Indian immigrants working in IT occupations. The majority of the participants were on the H-1B at the time of research. All interviewees had at least a bachelor degree and over 50% had postgraduate degrees in engineering or management. Participants’ ages ranged from 27 to 40 years. The majority, however, were in their mid-twenties to early thirties.

FINDINGS Through the provisions codiﬁed in TANF welfare policies and the H-1B visa program, the government extends and restructures privileges and thereby alters the nature of the employment contract. Welfare, no longer considered to be a citizen’s entitlement as of the reform of 1996, has been redeﬁned as a privilege that may be granted to unemployed and low-wage workers if they are judged to be deserving following a battery of diagnostic tests and conditions. In the case of welfare clients in Mundial County, the state’s reform policies authorize employers to assume a central role in this process. They are granted certain gatekeeping authorities and come to hold the power of granting various forms of endorsement which welfare clients must obtain in order to be eligible for aid. In the case of the H-1B visa, foreign workers are able to live and work in the U.S. if an employer offers them a job. Foreign workers employed on the H-1B are tied to their employment and employer and have no independent claim to this visa. Their legal standing in the U.S. is essentially determined by their employment and is incumbent upon the prerogatives of their employers. This suggests that the privileges granted by the state, either in the form of welfare beneﬁts (TANF) or work permits for foreign workers (H-1B visa), have come to be managed and executed by employers to a large extent. The following section will demonstrate how this relationship acquires form under terms determined by the state, and provide the empirical grounding to raise questions about the implications of these formations on the everyday lives of the workers affected. The ﬁndings presented below concentrate on the interaction of three complex and interrelated trends (see Table 1). The ﬁrst is a shifting of responsibility for work to individuals as workers on the H-1B are pressured to search for their own projects on behalf of their employers, and TANF clients are given ultimatums to achieve economic self-sufﬁciency. This shift of responsibility to individuals is accompanied by decreasing institutional accountability to laborers, where employers have implemented

354

FRANK RIDZI AND PAYAL BANERJEE

Table 1. The interaction of three complex and interrelated trends in H1B visa and TANF. Trend Increasing individual responsibility Decreasing institutional accountability to labor

Increasing employer power over labor

H1-B Visa

TANF

Responsible for searching for own projects Flexible doctrine of hire and ﬁre; relegating employers’ responsibility to the workers (ﬁnding work, insurance, beneﬁts, etc.) Control over visas and work and immigration status

Responsible for selfsufﬁciency Removal of welfare ‘‘entitlement’’ status; decreased protection from labor market Employer granted authority as source of evidence and credibility for applicants’ claims

ﬂexible hire and ﬁre policies and the state has terminated welfare’s entitlement status (which had hitherto been a governmental guarantee of protection from labor market exploitation). Ultimately, we assert that the state’s TANF and H-1B visa policies have resulted in the consolidation of private capital’s power over the labor groups that these policies regulate by explicitly granting authority to employers and by institutionalizing a pro-outsourcing stance.

Outsourcing and Welfare Reform When considered in terms of global economic restructuring, the welfare reform of 1996 enacted by the PRWORA is dually linked with what we outline as the spirit of outsourcing. It facilitates the wider trend of tentative work relations in which capital beneﬁts by employing workers on an exceedingly temporary basis (i.e. only when they are needed and for short periods of capital accumulation, see Campaign on Contingent Work, 2000) and, as will be discussed later in this paper, is carried out under the rubric of outsourcing. The welfare policies of 1996 have facilitated outsourcing through supporting private sector attempts to subcontract jobs in service and manufacturing by converting them to contingent and temporary (temp) positions. They do this by promising employers that new waves of former welfare recipients will be moved into the workforce (and potentially outsourced transient jobs) through welfare-to-work programs and by following through

The Spirit of Outsourcing

355

on this promise with strict work policies. Since 1996, the majority, approximately 60%, of former welfare recipients have entered the labor force (Boushey, 2001; also, see Hays, 2003). This trend has led to the estimation that approximately one to two million such additional workers will enter the workforce between 1993 and 2008 (Bartik, 2001, p. 72). Despite initial worries regarding the labor market’s inability to absorb all of these workers (Nightingale & Haveman, 1995), the low-wage economy generated a 40% job growth ﬁgure to accommodate this increase in labor in the two years immediately following the reform (1996–1998) (Lerman & Ratcliffe, 2001). This, however, is only one phase of the employment cycle. These former welfare recipients are ﬁlling increasingly temporary and contingent positions (Heinrich, 2005). While four of the top nine industries hiring former welfare recipients outpaced growth in overall national employment in the ﬁve years following the reform, they were also the hardest hit following the 2001 decline, with ﬁelds such as personnel supply services (i.e. temp services), which expanded by more than 5% between 1996 and 2001, losing as much as 20% of its jobs (Boushey, 2001). Overall, approximately 60.2% of welfare leavers have been concentrated in service and manufacturing industries, almost all of them at near-poverty wages (Loprest, 1999, 2003). Jobs in administrative support (i.e. clerical), service, manufacturing and construction labor, which collect the majority of welfare clients, together comprise close to 56% of all contingent work (Polivka, 1996). Given these conditions, Hays (2003, p. 59) estimates that, overall, only 10–20% of welfare recipients have attained ‘‘relatively permanent, above poverty stability’’ despite the fact that approximately 60% have left welfare. This state-endorsed inﬂux of welfare clients into the contingent labor market has critical implications for their relations with employers.

Welfare Clients and Employers With the reform of 1996, entitlement to welfare was ended and access to aid became theoretically contingent on compliance with a code of behavior that is in part policed by employers. Welfare restructuring demands that clientele assume individual responsibility for economic self-sufﬁciency. Consequently, 60% of states require job searching or other work-related activities as a prerequisite for receiving aid (Maloy, Pavetti, Shin, Darnell, & Scarpulla-Nolan, 1998). Those who do not meet these requirements are rejected. On the surface, this is a social policy that requires universal workforce participation among the poor and enforces it through verifying that its

356

FRANK RIDZI AND PAYAL BANERJEE

demands are met. The ideological implications reach much further. Universal workforce participation sends a clear message that poor families cannot avoid the vagaries of the current market. The government will continue to provide certain social welfare functions if, and only if, employers vouch for compliance with state-prescribed codes of behavior. Thus, the role of eligibility determination is in part delegated to employers. This fundamental realignment of the welfare state manifests itself on the local level in several ways. For instance, in the case of recently unemployed individuals, TANF staff members seek to ensure that individuals do not leave otherwise gainful employment to rely on welfare. To enforce this, staff, such as in the following excerpt, ask both for client versions of their work history and require that past employers verify this. Staff: Why are you no longer working; what happened, did you quit or were you ﬁred? Welfare Client: I had a problem with my manager. She was always yelling at me for things that were not my fault and I said that I’m not going to take it anymore and she told me to leave, I was ﬁred. Staff: Oh, I need this ﬁlled out by the employer y

This insistence on employer veriﬁcation is routine and required. While retrospective in its efforts to seek employer input, it is regularly deployed in conjunction with future-oriented state–employer interactions such as efforts to verify willingness of TANF clients to work. In 60% of states, applicants seeking aid must demonstrate willingness and commitment to work through job searching activities. Applicants in Mundial County, for instance, must apply for ten jobs each week, while waiting the typical 30 days for their aid application to be processed. These welfare hopefuls begin relationships with individual employers by ﬁling job applications and retrieving the necessary contact information such as manager name and phone number. Staff members complete this circuit by using the contact information that clients must submit to verify that job applications were ﬁled. One worker shares her template for this practice in a mock veriﬁcation phone conversation with an employer to whom a TANF applicant has applied for work: Hello. Can I speak with the manager? Oh not there, well supervisor, etc.? Well I’m calling from the Department of Social Services and I was wondering if you could tell me if Malika has placed an application for employment with you y

As she continues to explain, staff follow a similar script when verifying the word of clients that indicate that they have found employment.

The Spirit of Outsourcing

357

Hello, I am Alice from DSS and Joanna has informed me that she has found work with you. That’s great y I was wondering if you could help me verify some information. How many hours a week is she working/going to be working? y Well on average? y And the salary is? y

Although this is behind the scenes (i.e. not conducted in front of clients) it is prominent in clients’ understanding of what goes on during the application process. Staff routinely stress that potential employer information must be ﬁlled out clearly and completely so that the legitimacy of the application and hires can be veriﬁed. In addition to making connections to employers this way, welfare-to-work staff, under the guidance of staff dedicated to ‘‘job development,’’ invite employers to visit the ‘‘learn-to-work’’ ofﬁces, offer to post and advertise job ads, and maintain records of where clients are hired. In accord with the federal 1998 Workforce Investment Act (Public Law No. 105–220, 1998), staff members endeavor to cooperate with private employers as allies with many mutually shared goals (also see, Giloth, 2004). The veriﬁcation of employment history, job applications and hires, in conjunction with the business community outreach efforts of job developers, establishes the state and employers as partners in the welfare-to-work enterprise (McMenamin, 2001). It also places welfare clientele in a precarious position with relation to the labor contract. Employers familiar with the new welfare system come to realize that welfare clientele are required to accept any legal employment offered in order to remain eligible for aid. They also know that leaving a job of one’s own volition is grounds for TANF ineligibility. The TANF clientele, or at least the 26.4% of recipients (Ofﬁce of Planning, Research & Evaluation, 2001) and an undocumented number of applicants nationwide that are or will soon become employed, are disadvantaged by this arrangement. Their bargaining position for employment, wages, beneﬁts, and advancement is compromised. One former recipient explains the predicament this places on TANF clients who feel mistreated or exploited by employers: Damned if you do, damned if you don’t. Take the abuse because if you don’t, public assistance will kick you off because you quit the job. It’s called a voluntary quit.

In addition to weakening bargaining power, the surveillance networks established between the state and employers, through the efforts of welfare staff, make it difﬁcult if not impossible for welfare clientele to successfully conduct impression management, because their presentation of self must be reconciled with the impressions that employers share with welfare staff. Consequently, it also becomes difﬁcult to ensure that their version of events (rather than that of employers) steers the outcome of dealings with TANF

358

FRANK RIDZI AND PAYAL BANERJEE

gatekeepers. This becomes evident in the routine processing of paperwork in which the textualized voice of employers, no matter how carefully or carelessly recorded, assumes authority that can trump the word of clients. In the following example, a caseworker, when faced with conﬂicting data, opts to honor that of the employer over the data submitted by the client. This decision, which becomes visible because the caseworker decides to narrate to a coworker standing nearby, occurs almost seamlessly and, importantly, without consulting the client that is standing in front of them. Caseworker: This [form completed by the employer] says she is working 22 hours a week, but this [form completed by the client] says 30. According to this [employer form] she still has to do job search.

In the above instance, the consequences of yielding authority to the voice of the employer over that of the client results in an additional compliance requirement that would have been waived if the report of 30 hours had been accepted. In other cases, the results are more serious, and they often involve more substantial reliance on the word of employers. As in the following example of a staff member’s investigative work, once established, networks between employers and welfare staff can be called upon when deemed appropriate. Staff: I had a person the other day who provided me with veriﬁcation in a medical statement verifying that she had to stop work based on a doctor’s recommendation. So I called the employer and the employer indicated to me that that wasn’t true at all, that the person actually stopped working so that she could go back to school.

Such state-employer exchanges, as in the case above, routinely result in the termination or denial of TANF beneﬁts. Ultimately, even TANF clientele who ﬁle formal appeals to the state government cannot avoid state-employer liaisons. Their communications become evidence that it is submitted in fair hearings and is routinely accepted by hearing ofﬁcers (judges) as authoritative. The following is one such example: Job Coach: Well I called [the potential employer] and asked y ‘‘Are the applications written in the same handwriting?’’ ‘‘Yeah.’’ ‘‘O.K., thanks very much.’’ y Interviewer: Is that valid? Does that count as an application? Job Coach: [He deﬁnitively states,] No. I had to go to a fair hearing to testify that they didn’t do an independent job search. And you know that was y they lost.

In instances such as this, the adversarial nature of the formation and utilization of these networks of communication are made explicit. They position capital and the state as allies to whom impoverished women must

The Spirit of Outsourcing

359

prove their credibility and deservingness and by whom clients are jointly judged. Beyond this explicit alliance, collusion between the state and capital can be viewed in their collective reinforcement of the logic and practice that enables outsourcing. Welfare-to-work staff members speak about themselves as preparing clients for ‘‘the real world of work.’’ This involves insisting upon punctuality, professional dress, and above all, worker submissiveness to employer requirements. Clients who refuse to conform are labeled ‘‘hard to serve’’ and are treated in punitive fashion. One administrator describes the ‘‘tough love’’ approach that some coaches choose to take as being modeled after a nationally recognized program called ‘‘Strive’’ (also see Hays, 2003), but her overall description of welfareto-work does not stray far from this approach. [When you are] on public assistance (welfare) and if you don’t do what or go where you are assigned you can be sanctioned. The ﬁrst sanction is 30 days, the second is [more] y Orientation is the ﬁrst formal activity of the program. You have to show up or call and reschedule. If you fail to do either of these, you will be sanctioned. You do however; always have a chance to explain yourself in front of the judge.

Job coaches are quick to point out that their stern approach is for the good of their clients, but in confrontations with clients it becomes evident that employers also beneﬁt by receiving a conditioned workforce. In the following excerpt, a job coach reﬂects on a processing interaction in which routine training practices are questioned by a client. In the process, the coach reasserts her perceived role as a labor market intermediary whose goal is to prepare workers to meet the speciﬁcations of employers within a certain low-wage niche of the economy. Job Coach: I had one lady ask me yesterday. You know she came in well after twelve noon [This is the deadline.] and she said to me, ‘‘Haven’t you ever been late to work’’ or you know and the answer is ‘‘yes.’’ ‘‘I have been late to work, you know, but there are also jobs where if you arrive late for work you won’t have a job.’’ You know? So yes, some of them get kind of bent out of shape over the fact that um, if they arrive late we don’t see them. But the whole message that we’re trying to send is, ‘‘You have to be on time because some employers will not tolerate you being late’’ y You know, at all. You know? So you have to get into the habit of doing whatever is expected of you on time. So that’s the message that we’re trying to send.

Embedded within this message is an unquestioned assumption of an easily replaceable worker with little leeway and input into their working conditions. This type of worker is furthermore being trained to enter a job that is qualitatively different from that of the job coach, who admittedly possesses the latitude within her job to be late from time to time. This worker is not

360

FRANK RIDZI AND PAYAL BANERJEE

conceptualized as a stakeholder with leverage to negotiate a desirable working arrangement. Rather, in the words of the caseworker, this type of worker would do well to ‘‘get into the habit of doing whatever is expected of you on time.’’ The pro-outsourcing stance of this approach to welfare aid provisioning is brought into focus by a contrast with previous welfare state strategies that focused on developing human capital for career development rather than enforcing a work-ﬁrst approach that insists upon immediate and ongoing employment at a certain quota of hours per week, regardless of how many part-time and temporary jobs must be pieced together to achieve this. Despite evidence indicating the importance of investing in skill upgrading for career advancement, in 2000, less than 1% of TANF federal funds were spent on education and training and only 5% of TANF recipients participated in them (Poppe, Strawn, & Martinson, 2004). This contrast in approaches is not lost on clients, such as the following former welfare recipient who laments that lack of human capital persists as a barrier to the stated welfare goal of long-term self-sufﬁciency. Bring back more school aid. If you are stuck without a speciﬁc training, it’s slave labor. If you want people to be independent of the system, they need a livable wage, need an education. Count education as a work activity!

Such pleas, however, are not likely to gain support among staff who tend to view the present ‘‘work ﬁrst’’ approach as a reasonable remedy to past human capital investment inefﬁciencies (Ridzi, 2004). The staff is also not likely to be perceived by clients as allies in challenging the dominant discourse of work, which is based on a pro-outsourcing paradigm that is designed to facilitate a docile and ﬂexible labor force. The ‘‘work-ﬁrst’’ approach admits that what is demanded of low-wage workers in the new outsourcing economy is not preparing to climb a career ladder but rather scrambling to submit to rapidly changing employer demands. In addition to conditioning workers to embody the deference and obedience expected of menial laborers (Hays, 2003, pp. 43, 60), welfare-to-work also serves to convince them that they are expendable within the new economy. This approach to low-wage laborers is infused with the logic of outsourcing, which became clear when welfare ofﬁcials in Mundial County began brokering a deal to replace striking unionized food service workers with workfare placements. Though the union eventually compromised on a contract under this pressure, the message was clear; welfare-to-work disciplines not only welfare clients but also all other low-wage workers whose work can be easily transferred to alternative labor pools. In many ways, this type of

The Spirit of Outsourcing

361

‘‘conditioned’’ workforce, that recognizes that the power to wield jobs is increasingly in the hands of employers due to the repertoire of outsourcing, is what welfare-to-work is training clients to be a part of by, in the words of a client, forcing them to ‘‘tuck their tail between their legs to get [aid].’’ In these welfare activities a reconﬁgured relationship emerges among the state, employers, and welfare applicants. The complexities of these relations play out along two lines or faults within welfare ofﬁces. On the one side stand the state and employers, uniﬁed in their joint efforts to validate or debunk applicants’ claims to legitimacy as welfare recipients and committed jobseekers. On the other stands the welfare client, unable to ﬁnd refuge with the state without the assent of employers themselves. The 1996 welfare reform thus emerges as a break from past generous government provisions that capitalists had criticized for their tendency to ‘‘undermine labor discipline’’ (Akard, 1992, 1995). Instead, it can be seen as government’s attempt to aid capital’s efforts to reestablish control over workers and an end to the welfare state’s viability as an alternative to worker exploitation, or a place to ﬁnd sustenance while organizing as a labor collective to demand better working conditions (Piven & Cloward, 1993).

Outsourcing and the H-1B Visa The relationship between the H-1B visa and outsourcing is apparent for two interrelated reasons: one, the IT ﬁeld, which the H-1B serves extensively, relies on subcontracting and outsourcing to access labor; and, two, the majority of immigrants on the H-1B work as temporary contract labor for IT projects outsourced by large clients to consulting and labor-supplying ﬁrms located both within and outside the U.S. The H-1B is a temporary work visa that enables employers in the U.S. to hire skilled foreign workers for specialty occupations. The Immigration and Naturalization Services report of 2000 identiﬁed a close connection between labor needs in IT and the extensive use of the H-1B and unequivocally stated the critical importance of the visa for employers hiring immigrant workers for IT positions (Ayers & Syfert, 2002). Hailed as the ‘‘workhorse of the IT industry’’ (Ayers & Syfert, 2002, p. 540), the H-1B has been fundamentally tied to U.S. labor demands in IT and Indian immigrants have constituted an overwhelming majority in this foreign-worker category (Lowell, 2000). The number of Indians on the H-1B, historically high in the IT workforce, has been increasing steadily since the early 1990s (Lowell, 2000). In 1999, 60% of all H-1B visas went to immigrants hired for the IT ﬁeld and about 75% of all

362

FRANK RIDZI AND PAYAL BANERJEE

immigrant IT workers on H-1B visas were from India (U.S. GAO, 2000). The H-1B population was estimated at 360,000 in 1999 and at 425,000 in 2000 (Lowell, 2000). The majority of companies in the U.S. rely on a system of subcontracting, where their IT work and labor needs are outsourced to consulting companies. These consulting companies, in turn, execute these projects by recruiting large numbers of IT workers as temporary ‘‘contract workers’’ or ‘‘consultants’’ from different labor-supplying companies. Data from this research on Indian IT workers indicate that it is typical for this immigrant group on the H-1B visa to be positioned as temporary contract workers in the IT ﬁeld’s labor subcontracting system. Rajesh, an Indian software engineer who has been working as a contract worker on this visa since 1997, provided the following explanation for why corporations prefer ﬂexible hiring: The large American companies are not willing to hire directly, for they want the ﬂexibility and freedom that comes with consulting or subcontracting. They want to hire and ﬁre based on what projects they are doing. They want to get the projects done with contract workers y Through consulting [subcontracting] you can get any kind of people with any technical skills. But if you want to hire a person as a permanent employee, you have to give that person [a] bonus, a lot of beneﬁts, health insurance, whatever. But if you get the job done through consultants, through contractors, like the Indians, you can hire them when you need to and then ﬁre them anytime. Because they are not your employees, right. If the client does not like your work, or they run out of money, or they decide to stop the project, they can just get rid of these people in one day! You can ﬁre them anytime. With employees, there are some commitments y So that’s why people hire contract workers on short-term project basis.

Immigrants’ tenuous status under the terms of the H-1B visa increases their vulnerability and facilitates their deployment as an easily exploitable and ﬂexible labor force, ideal for subcontracted work. Despite recent trends in offshore outsourcing of IT projects, large proportions of outsourced work are still carried out onsite in the U.S. by immigrant IT workers whose entry into the U.S. labor market is necessitated by demands for unencumbered workers and facilitated by work visas. In many cases, the employment options available to TANF clientele demonstrate similar patterns of outsourcing in the low-wage contingent work sector.

Workers on H-1B Visas and Employers This research reveals that the terms of the H-1B visa, as determined by the state, empower employers to exploit and control immigrant workers. The

The Spirit of Outsourcing

363

H-1B visa is employment dependent and allows a foreign worker to work only for the speciﬁc company that processes the visa. Those under the H-1B have no independent claim over their visa and cannot work for a company if it declines to process their visa. As a result, these workers’ employment choices are limited to companies willing to apply for a visa on their behalf. Since most companies prefer to have a lean workforce and get the bulk of their IT projects done with contract workers, they do not hire immigrants as direct employees. Immigrants’ employment options, therefore, are severely restricted to small labor vendors or subcontracting ﬁrms that are willing to hire foreign workers and subcontract them to corporate clients. As contract workers, Indian immigrants on the H-1B are considered direct employees of their subcontracting ﬁrms, not the numerous clients where the subcontracting ﬁrms assign them. If H-1B workers quit working for the employer that has processed their visa to join another, their visas have to be reprocessed by the new employer. Thus, the validity of the visa and these workers’ status are determined by the duration of employment with the speciﬁc company that holds the visa. Under these circumstances, termination of employment with the visaholding company, when no other company has hired them, translates into loss of legal standing for foreign workers. In the context of employment in the IT industry, jobs for immigrants are primarily short-term, contractual, and are generally available only on a ‘‘project-to-project’’ basis, in contrast to relatively steadier employment opportunities with well-established consulting ﬁrms that are reserved for U.S. citizens or permanent residents. These tenuous terms confer an inordinate amount of leverage upon companies with respect to their H-1B employees. The workers, in acknowledgement of this power, harbor anxiety about maintaining their livelihood and legal status. Anil, an immigrant from India working as a contract worker on the H-1B, explained what it means to work on this visa in the subcontracting or outsourcing scenario. It was a pretty bad situation being on H-1B and getting laid off. It is much worse now, it was still pretty bad then. Because you fall out of status and which is technically illegal for no fault of yours. So any government agency can, the immigration agency can throw you out of the country without any reason, for any reason whatsoever. So it is a pretty bad situation being laid off.

Confronted with the fear that they might lose their legal status if ﬁred, immigrants on H-1B visas are forced to accept a range of exploitative practices and make many compromises to stay employed and in status. Ajit, another engineer on the H-1B working on a project as a contract

364

FRANK RIDZI AND PAYAL BANERJEE

worker with one of the leading telecommunications companies in the nation, said: y most of the H-1B people are getting exploited by their potential employers or current employers, because the employers know that you cannot go anywhere. And they maybe, if the employers are not good, they will not give you a raise. In the perfect world, the relation between employer and employee is symbiotic. But in the real world, in this particular scenario of the Bodyshoppers2 using H-1B and Green Card, the relation between the employee and employer is parasitic.

Cognizant of the fact that the foreign workers’ work and legal standings are dependent on their employers, companies routinely capitalize on this power imbalance to take advantage of their immigrant employees. Vivek explained this point as follows: Actually, I think being on H-1B is a handicap. Because, if you lose your job and are unable to ﬁnd another one, you have no status. Also, when your employers ﬁle for your Green Card when you are on your H-1B, they know you cannot leave them, so they refuse to increase your salary or entertain any requests, because they know you will stick to them for not only are they processing your Green Card, they are also holding your visa till your Green Card comes along. So the visa binds you.

This shows that the stipulations of the H-1B not only make immigrant workers dependent on their employers, but also subject them to a range of unfair practices given the imbalance of power dynamics in favor of the employers or private capital. The increasing dominance of the larger logic of outsourcing or fragmentation of both employment and production becomes clear if one analyzes the details of this speciﬁc type of immigration policy as they contribute to the exploitation of H-1B workers by compelling immigrants to accept additional responsibilities. Since labor-supplying ﬁrms catering to subcontracting and outsourcing are the primary employers of these immigrants, these workers’ employment is secure so long as they are placed in contract positions at clients’ projects. The subcontractors’ revenue generation is dependent upon their immigrant employees’ placement on temporary contract-based projects. If these employees are not placed with a client, the companies do not make any profits and proceed to lay off the immigrants. Many interviewees explained that the two-pronged fear of unemployment and loss of legal standing leads them to take on the responsibility of ﬁnding clients and projects on their own to ensure the continuity of their employment status. Harish, an Indian immigrant IT professional, who has been employed in the U.S. as a contract worker on the H-1B since the beginning of his career in the late 1990s, discussed his experience on this matter by saying:

The Spirit of Outsourcing

365

I know that to keep my job with the subcontractor, I need to keep working on some client’s project to generate revenues for my employer, so that he keeps me on his books, and I get to stay in status. So, when a project is nearing its end, I feel this pressure. In theory, my employer is supposed to ﬁnd me my next project. But most of the times they don’t or cannot. And I cannot sit quietly, waiting and depending on them to come up with something for me, because if they don’t, then I am the one that gets into trouble. My job and visa status are on the line, which is a big worry. The pressure is more on me than on anyone else. So, I have to take charge and start looking for clients or consulting companies who are doing projects and need contract workers. In my case, of the six projects that I have worked on, I found ﬁve! My employer found only one for me.

This shows how representatives of capital, ranging from corporate clients that indirectly engage these workers to the subcontractors that hire these workers directly, have systematically managed to transfer activities and responsibilities previously associated with employers onto the workers themselves. Under the circumstances of ﬂexible hiring in IT, many immigrant workers on the H-1B visa have been transformed into self-reliant and individualized agents responsible for ensuring the continuation of their work and employment despite being designated as employees. Their employers, in this case, are often little more than nominal entities that only perform the task of granting these workers a legal work status (for a proﬁt), so that the corporate sector may access this pool of labor. Deeply aware of immigrants’ desperation to keep their jobs to maintain their status under the H-1B, subcontracting employers have conveniently shifted the responsibility of ﬁnding potential work or ‘‘projects’’ onto immigrant workers. In the case of welfare clients, we see a parallel in terms of how individuals seek out new jobs to remain eligible for aid. Thus, both TANF clients and workers on the H-1B must actively assume a range of responsibilities to secure privileges and to maintain their status as deﬁned by the state. When immigrants concede to take on projects that pay substandard hourly wages to remain employed and in status, they subsidize labor costs and make the system of outsourcing and subcontracting further conducive to proﬁt maximization. This marginalization ﬁnds its counterpart in the instance of welfare clientele, who accept otherwise substandard job positions to maintain TANF eligibility and thereby subsidize the low wage, contingent labor market with government beneﬁts such as child care and supplementary cash. These examples from the research participants’ experiences show how the terms of TANF and the H-1B visa have engendered a set of outsourcing relations among citizens, immigrants, and their employers which compromise workers’ bargaining power by subjugating them to positions of dependence and deference, and, in contrast, empower employers to manipulate their power. These relations, however, extend beyond those

366

FRANK RIDZI AND PAYAL BANERJEE

that are being regulated and shaped by the state and include how the state itself operates via the principles of outsourcing.

THE STATE AS AN AGENT AND CLIENT OF OUTSOURCING The previous sections explore the ways in which the federal government, through welfare and immigration policies, has enabled the private sector to implement principles of outsourcing in their methods of operation. However, this picture is incomplete without the recognition that the state, by embracing the same model, turns outsourcing on itself as well. This is visible in the fact that considerable aspects of the administration of welfare-towork are outsourced as well as the fact that some of the work that the state outsources to prominent consulting ﬁrms, ironically, is delegated to subcontracted labor that tends to be immigrants on H-1B visas. Outsourcing in the case of TANF human services implementation refers to the government contracting with public, private, and nonprofit entities to provide speciﬁc services such as caseload management (in 40% of states) and job placement (43% of states) (Bandoh, 2003; U.S. GAO, 2002). As of 2001, 13% of all Federal TANF expenditures was spent on contracts with nongovernmental entities (U.S. GAO, 2002). Occurring in all states but South Dakota, 66% of these contracts (5,277 contracts in total nationally) are with secular nonprofit organizations and 27% with for-proﬁt organizations (U.S. GAO, 2002). By embodying outsourcing, this unprecedented involvement of nongovernmental entities in providing human services, even including temp agencies to process as high as 42% of the welfare population in some counties (Hanbey, 2003; Pawasarat, 1997), sets a tone for a new demeanor of partnership between the state and the market. This tone is reﬂected in the comments of welfare-to-work staff in Mundial County who admit that, though they do not see welfare reform’s innovations disappearing anytime soon, they in essence have temporary jobs which are based on the indefinite ﬂow of block grant monies. Joan, a welfare-towork administrator, explains how her job was created. It came about because of a contract and a partnership between the Department of Social Services and the community college that indicated that we [the new welfare-to-work agency] would handle all of the federal, state, and local tax money that was funneled toward welfare reform job training etcetera. Thus we are paid by the [agency] but the money is federal, state, and local y Our focus is to get the welfare rolls down to a manageable number, try to keep it to the federal ﬁve year limit, since that is when the

The Spirit of Outsourcing

367

federal funds cease and the state of New York y must pick up the tab. We do not know what type of aid will be provided [then], however. The state lawmakers in Albany have been quiet about this so far.

As Samantha, an administrator, also explains, this outsourcing mentality and practice furthermore extends beyond the ofﬁcial welfare-to-work ofﬁce that was established. Even our post-employment, which has just begun within the past few months, is contracted out to groups like the Urban League, the North Side Community Center, Disability Accommodations Association, and the Organization for Better Living.

Thus, it is not only the work that is to be done by welfare clients, but also the work of many welfare employees who were hired in the midst of the 1996 reform that is mediated by the logic of outsourcing. This structure is in keeping with the experimental nature of the reform and the fragmenting and economic qualities of outsourcing. If these workers succeed in reducing caseloads and ushering recipients from welfare to work, they may be continually funded. However, if they fail, or if they do their job so successfully that they are no longer needed, then the entire regime of job coaches and employment counselors can be dispensed with by the simple termination of grant funding. There will be no civil service structure that would protect the jobs of employees, as there would be in the case of employees of the welfare ofﬁces that predate the welfare reform. The outsourcing mode of state operation also extends beyond the realm of welfare-to-work to intersect, perhaps unexpectedly, with the domain of the H-1B. In fact some Indian immigrant IT workers that were interviewed for this research indicated that they had worked, as contract labor, on different welfare-related government and other projects that were being executed by leading consulting ﬁrms, which in turn gained access to immigrant workers through subcontracting companies. The projects these participants mentioned were with the welfare ofﬁce, the justice system, the department of motor vehicles, and the department of human services. An interview with Richard, an Indian immigrant who owns and runs a subcontracting ﬁrm, revealed that his company, which employs a large number of Indian immigrant IT workers on the H-1B visa to contract out on clients’ projects, placed some of its workers on an assignment with the department of labor. He said: The Department of Labor has a training program for unemployed citizens and we run the training program. So the Department of Labor is our direct client y we are providing the services directly. But for some others, there are more companies in layers. So there will be one more company between us and the client y with SharpImage

368

FRANK RIDZI AND PAYAL BANERJEE

Photocopying we have one more company in between that we are going through. This company is using our employees for their work with SharpImage.

In this case, Richard’s ﬁrm was dealing with the labor department directly to conduct a computer skills training program for unemployed citizens on welfare. In other cases, such as with SharpImage, several layers of companies were present between Richard’s ﬁrm and the ﬁnal client. Another interviewee, an upper-level manager at a leading consulting ﬁrm in the U.S., described how his company undertakes projects for the government and executes them by using labor supplied by multiple subcontractors and, in the majority of cases, contract workers thus accessed are Indian immigrant IT professionals. Although the ﬁnal placement of contract workers may occur via one or more subcontracted ﬁrms, the critical point to bear in mind is that the experiences of people on welfare and H-1B attest to the state’s role as both an agent and a client of outsourcing.

CONCLUSION In the sections above, we have explicated some of the crucial ramiﬁcations that TANF and H-1B policies hold for the workers they regulate. Through our analysis of the reorganization of public–private partnerships in terms of increasing decentralization, a retreat from assuming responsibilities for workers, and individualization, we have identiﬁed how outsourcing is not merely a profitable business model for cross-border trade, but that it is a key principle, component, and outcome of the process of neo-liberal global economic restructuring for both corporations and the state. Based on our comparative analysis of how the intersections of economic restructuring and state policies of the U.S. marginalize welfare recipients in low-wage temporary manufacturing and service employment and Indian immigrant workers on the H-1B visa in IT occupations, we propose a reconceptualization and extension of the discursive term ‘‘outsourcing’’ to better capture its connotation and domicile beyond the narrow ambit of trans-border offshoring of professional jobs. We extend two critical observations based on the above. First, we suggest that the constitutive elements of H-1B and welfare laws have aided the deployment of speciﬁc groups of people in ways that have allowed employers to access them as labor under terms that facilitate the implementation of the larger logic or spirit of outsourcing. Having access to these two pools of tractable labor has enabled and encouraged employers to alter the terms of employment, and institute fragmented and decentralized means of

The Spirit of Outsourcing

369

producing services and goods for maximizing proﬁt margins and/or achieving ﬂexibility. For TANF, the law lets the employers know that many lowwage laborers will move from welfare to work. The employers are also made aware that they will serve as legitimate corroborators of this workforce’s claims. The law then enforces the transition from welfare to work by encouraging, cajoling, and pushing people into this low wage, contingent work sector (Besharov & Germanis, 2002). Furthermore, the employers are given the prerogative, under the auspices of the state, to manipulate and mediate these conditions that are placed on labor to meet their own terms. In the case of the H-1B, the increase of annual quotas, again through the state’s mandate, informs employers that they will be able to access a large pool of immigrant workers who can be hired as temporary contract workers in the U.S. through outsourcing or subcontracting. The terms of the H-1B make immigrants dependent on their employers. This vulnerability serves to make the subcontracting system operate efﬁciently, as immigrants take it upon themselves to ﬁnd projects and to accept lower wages and other exploitative conditions of work. To a large extent TANF and H-1B laws set constraints under which their constituencies must enter labor market relations and thus determine how these people are to offer their labor as workers. These policies, therefore, socially construct individuals as labor of a speciﬁc category of marginalized status. Second, we contend that the mainstream discourse on outsourcing has privileged the notion of ‘‘boundaries’’ within the realm of the national economy (of the U.S., India, or China for instance) to examine the socioeconomic impact of global trade in goods and services. Outsourcing is primarily conceptualized as the economic activity of offshoring processes of producing goods and services across national economic boundaries. This also sets in motion an ‘‘us versus them’’ response, because only an adversarial relationship between workforces from respective countries can be imagined through this framework. However, our research and analysis show that we need to complicate the understanding of outsourcing and broaden the parameters of the concept. We show that outsourcing takes place both within and across national boundaries simultaneously and also across citizen and noncitizen populations, as we see in the case of welfare clients (a group that is not readily seen as being part of the outsourcing debate). More importantly, we show how the rationale of outsourcing is contained not only in the sphere of export and import and private capital, but that it has become a fundamental organizing principle through which public institutions and the state operate or formulate policies. This point challenges scholars to engage with the idea of ‘‘boundaries’’ in new ways. We suggest that it is critical to ground the

370

FRANK RIDZI AND PAYAL BANERJEE

concept of outsourcing in perspectives that examine how corporations and state institutions have forged alliances in the interest of capital to deploy labor under ﬂexible and exploitable circumstances.

NOTES 1. This and all proper names are pseudonyms. 2. In the vocabulary of immigrant IT workers, a Bodyshop is a subcontracting company in the business of supplying workers to large ﬁrms and consulting companies for outsourced IT projects on a contract basis.

REFERENCES Acker, J. (2003). Gender, capitalism, and globalization. Critical Sociology, 30, 1–26. Akard, P. J. (1992). Corporate mobilization and political power: The transformation of U.S. economic policy in the 1970s. American Sociological Review, 57, 597–615. Akard, P. (1995). The return of the market? Reﬂections on the real ‘conservative tradition’ in U.S. policy. Sociological Inquiry, 65, 286–309. Alexander, M. J., & Mohanty, C. T. (Eds) (1997). Feminist genealogies, colonial legacies democratic ftures. New York: Routledge. Auer, P., & Cazes, S. (Eds) (2003). Employment stability in an age of flexibility: Evidence from industrialized countries. Geneva: International Labour Ofﬁce, Employment Strategy Department. Ayers, K. N., & Syfert, S. D. (2002). U.S. visa options and strategies for the information technology industry. International Quarterly, 14, 535–565. Bandoh, E. (2003). Outsourcing the delivery of human services. (Welfare Information Network Issue Notes 7). Retrieved June 22, 2004, from http://www.ﬁnanceprojectinfo.org/Publications/outsourcinghumanservicesIN.htm Barley, S., & Kunda, G. (2004). Gurus, hired guns, and warm bodies: Itinerant experts in a knowledge economy. New Jersey: Princeton University Press. Bartik, T. J. (2001). Jobs for the poor: Can labor demand policies help? New York, NY: Russell Sage Foundation. Benner, C. (2002). Work in the new economy: Flexible labor markets in Silicon Valley. Malden, MA: Blackwell. Besharov, D.J., & Germanis, P. (2002). Family well-being after welfare reform. Retrieved April 21, 2003, from http://www.welfareacademy.org/pubs/familywellbeing/ Bhagwati, J., Panagariya, A., & Srinivasan, T. N. (2004). The muddles over outsourcing. The Journal of Economic Perspectives, 18, 93–114. Bonacich, E., & Appelbaum, R. P. (2000). Behind the label: Inequality in the Los Angeles apparel industry. Berkeley: University of California Press. Boushey, H. (2001). Last hired, first fired: Job losses plague former TANF recipients (EPI Issue Brief # 171). Washington, DC: The Economic Policy Institute. Brown, A. (1997). Work first: How to implement an employment-focused approach to welfare reform – A how-to guide. New York: Manpower Demonstration Research Corporation.

The Spirit of Outsourcing

371

Burawoy, M. (1991). Ethnography unbound: Power and resistance in the modern metropolis. Berkeley: University of California Press. Campaign on Contingent Work. (2000). What’s wrong with temp work? A report on the temp industry in Massachusetts. Boston, MA: Campaign on Contingent Work. DeVault, M., & McCoy, L. (2002). Institutional ethnography: Using interviews to investigate ruling relations. In: J. Gubrium & J. Holstein (Eds), Handbook of interview research (pp. 751–776). Thousand Oaks, CA: Sage. Dobbs, L. (2004). The myth of ‘insourcing’. U.S. News & World Report, 136(May), 56–57. Drezner, D. W. (2004). The outsourcing bogeyman. Foreign Affairs, 83, 22–24. Fairﬁeld, R. P. (1974). Humanizing the workplace. Buffalo, NY: Prometheus Books. Giloth, R. P. (Ed.) (2004). Workforce intermediaries for the twenty-first century. Philadelphia, PA: Temple University Press. Hanbey, M. (2003). TANF leavers and temp work in Washington state, 1998–2000. Unpublished master’s thesis, University of Washington. Harvey, D. (1989). The condition of postmodernity: An enquiry into the origins of cultural change. Cambridge, MA: Blackwell. Hays, S. (2003). Flat broke with children: Women in the age of welfare reform. New York: Oxford University Press. Heery, E., & Salmon, J. (Eds) (2000). The insecure workforce. New York: Routledge. Heinrich, C. J. (2005). Temporary employment experiences of women on welfare. Journal of Labor Research, 26, 335–350. Hira, R., & Hira, A. (2005). Outsourcing America: What’s behind our national crisis and how we can reclaim American jobs. New York: American Management Association. Kaufman, B. E., & Kleiner, M. M. (Eds) (1993). Employee representation: Alternatives and future directions. Madison, WI: Industrial Relations Research Association. Kazis, R. (2004). What do workforce intermediaries do? In: R. P. Giloth (Ed.), Workforce intermediaries for the twenty-first century (pp. 73–93). Philadelphia, PA: Temple University Press. Kettl, D. F. (1988). Government by proxy: Mis?Managing federal programs. Washington, DC: Congressional Quarterly Press. Kochan, T. A. (1984). Worker participation and American unions: Threat or opportunity? Kalamazoo, MI: W.E. Upjohn Institute for Employment Research. Koehler, E., & Hagigh, S. (2004). Offshore outsourcing and America’s competitive edge: Losing out in the high technology R&D and services sectors. Ofﬁce of Senator Joseph. I. Lieberman. Washington, DC. Retrieved August 24, 2005, from http://lieberman.senate.gov/ newsroom/whitepapers/Offshoring.pdf Lerman, R. I., & Ratcliffe, C. (2001). Are single mothers ﬁnding jobs without displacing other workers? Monthly Labor Review, 124, 3–12. Logue, J., & Yates, J. (2001). The real world of employee ownership. Ithaca, NY: ILR Press of Cornell University Press. Loprest, P. J. (1999). Families who left welfare: Who are they and how are they doing? Washington, DC: The Urban Institute. Loprest, P. J. (2003). Fewer welfare leavers employed in weak economy. Washington, DC: The Urban Institute. Lowell, B. L. (2000). H-1B temporary workers: Estimating the population. Working Paper No. 12. University of California, The Center for Comparative Immigration Studies, San Diego, CA.

372

FRANK RIDZI AND PAYAL BANERJEE

Maloy, K., Pavetti, L., Shin, P., Darnell, J., & Scarpulla-Nolan, A. (1998). A description and assessment of state approaches to diversion programs and activities under welfare reform. Retrieved April 9, 2004, from www.gwu.edu/~chsrp/wr/cover.html Marks, J. (2004). The outcry over outsourcing. State Legislatures, 30(May), 30–31. McMenamin, B. (2001). Payroll paternalism. Forbes, 167(April), 114–120. Mishel, L., Bernstein, J., & Schmitt, J. (1999). The state of working America 1998–1999. New York: Economic Policy Institute, Cornell University Press. Nightingale, D. S., & Haveman, R. H. (Eds) (1995). The work alternative: Welfare reform and the realities of the job market. Washington, DC: Urban Institute Press. Noyelle, T. J. (1987). Beyond industrial dualism: Market and job segmentation in the new economy. Boulder, CO: Westview Press. Ofﬁce of Planning, Research and Evaluation. (2001). Characteristics and financial circumstances of TANF recipients: October 1999–September 2000. Retrieved June 2, 2004, from http:// www.acf.dhhs.gov/programs/opre/characteristics/fy2000/128.htm Ong, A. (1991). The gender and labor politics of postmodernity. Annual Review of Anthropology, 20, 279–309. Osterman, P. (2004). Labor market intermediaries in the modern labor market. In: R. P. Giloth (Ed.), Workforce intermediaries for the twenty-first century (pp. 155–170). Philadelphia, PA: Temple University Press. Pawasarat, J. (1997). The employer perspective: Jobs held by the Milwaukee County AFDC single parent population (January 1996–March 1997). Milwaukee, WI: University of Wisconsin, Milwaukee Employment and Training Institute. Pencavel, J. (2001). Worker participation: Lessons from the worker co-ops of the Pacific Northwest. New York: Russell Sage Foundation. Persuad, R. B., & Lusane, C. (2000). The new economy, globalization and the impact on African Americans. Race and Class, 42, 21–34. Piven, F. F., & Cloward, R. (1993). Regulating the poor: The functions of public welfare. New York: Vintage Books. (Original work published 1971) Polivka, A. E. (1996). A proﬁle of contingent workers. Monthly Labor Review, 119(October), 10–20. Poole, M. (1978). Workers’ participation in industry. Boston: Routledge & Kegan Paul. Poppe, N., Strawn, J., & Martinson, K. (2004). Whose job is it? Creating opportunities for advancement. In: R. P. Giloth (Ed.), Workforce intermediaries for the twenty-first century (pp. 31–73). Philadelphia, PA: Temple University Press. Public Law No. 104–193. (1996). The Personal Responsibility and Work Opportunity Reconciliation Act of 1996. Public Law No. 105–220. (1998). Workforce Investment Act of 1998. Public Law No. 105–277. (1998). The American Competitiveness and Workforce Improvement Act of 1998. Public Law No. 106–313. (2000). The American Competitiveness in the 21st Century Act of 2000. Ridzi, F. (2004). Making TANF work: Organizational restructuring, staff buy-in, and performance monitoring in local implementation. Journal of Sociology & Social Welfare, 31, 27–48. Rosenbloom, J. L. (2002). Looking for work, searching for workers: American labor markets during industrialization. New York: Cambridge University Press. Sassen, S. (1998). Globalization and its discontents: Essays on the new mobility of people and money. New York: The New Press.

The Spirit of Outsourcing

373

Smith, D. E. (1987). The everyday world as problematic: A feminist sociology. Boston: Northeastern University Press. Smith, D. E. (1990). The conceptual practices of power: A feminist sociology of knowledge. Boston: Northeastern University Press. Smith, D. E. (1999). Writing the social: Critique, theory, and investigations. Toronto, Canada: University of Toronto Press. Smith, S. R., & Lipsky, M. (1993). Nonprofits for hire: The welfare state in the age of contracting. Cambridge, MA: Harvard University Press. Smith, V. (2001). Crossing the great divide: Worker risk and opportunity in the new economy. Ithaca, NY: ILR Press of Cornell University Press. Thottam, J. (2004). The fuss only fuels the outsourcing. Time, 163(24), 20–21. Tsiganou, H. A. (1991). Workers’ participative schemes: The experience of capitalist and planbased societies. New York: Greenwood Press. U.S. General Accounting Ofﬁce. (2000). H-1B foreign workers: Better controls needed to help employers and protect workers. (Publication No. GAO/HEHS-00-157). Retrieved October 6, 2005, from http://www.gao.gov/archive/2000/he00157.pdf U.S. General Accounting Ofﬁce. (2002). Welfare reform: Federal oversight of state and local contracting can be strengthened. (Publication No. GAO-02-661). Retrieved October 6, 2005, from http://www.gao.gov/new.items/d02661.pdf U.S. General Accounting Ofﬁce. (2004). International trade: Current government data provide limited insight into offshoring of services. (Publication No. GAO-04-932). Retrieved October 6, 2005, from http://www.gao.gov/new.items/d04932.pdf Weidenbaum, M. (2004). Outsourcing is a good thing – mostly. USA Today Magazine, 132(May), 20–22. Wetzel, J. R. (1995). Labor force, unemployment, and earnings. In: R. Farley (Ed.), State of the union: America in the 1990s (pp. 59–105). New York: The Russell Sage Foundation. Whyte, W. F. (1983). Worker participation and ownership: Cooperative strategies for strengthening local economies. Ithaca, NY: ILR Press of Cornell University Press.

This page intentionally left blank

374

GETTING JOBS AND BUILDING CAREERS: REPRODUCING INEQUALITY IN STATE-SPONSORED JOB SEARCH ORGANIZATIONS Vicki Smith, Heather Kohler Flynn and Jonathan Isler ABSTRACT Hunting for jobs is a socially, historically, and institutionally constructed process. Workers must learn how to find jobs and build careers but the ways in which they learn and the tools they have for doing so vary by class, race, gender, and cultural capital. In this paper, we analyze two organizations that teach clients job market behaviors that purportedly enable them to search for work. We argue that job search organizations (JSOs) can reinforce occupational and career inequalities. Focusing on the job market information and skills given to different occupational groups, messages about opportunity and mobility, and resources made available to clients, we show how JSOs prepare people to look for jobs along class-specific lines. These organizations discipline clients’ aspirations and shape their understandings of their occupational competencies Worker Participation: Current Research and Future Trends Research in the Sociology of Work, Volume 16, 375–402 Copyright r 2006 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1016/S0277-2833(06)16014-8

375

376

VICKI SMITH ET AL.

and weaknesses. The study highlights the importance of this understudied type of labor market intermediary.

Workplaces are sites for the reproduction of social and economic inequality. Sociologists have documented the organizational mechanisms and processes, and the actions of managers, employers, and workers themselves that contribute to asymmetrical positions and outcomes at work. The corpus of research conducted over the last three decades illuminates activities from the front- to the back-end of work organizations: barriers to entry evidenced in recruitment, interviewing, and hiring decisions (Kennelly, 1999; Moss & Tilly, 2001; Neckerman & Kirschenman, 1991; Shih, 2002; Waldinger & Lichter, 2003); processes of sex and race segregation (Bielby & Baron, 1986; Blau, Ferber, & Winkler, 1998; Braddock & McPartland, 1987; Cohn, 2000; Jacobs, 1995; Thomas, 2003); small-group processes that routinely exclude some groups from positions of authority and status (McGuire, 2002; Ridgeway, 1997; Smith & Elliott, 2002); and differential mobility, promotion, and training patterns (Knoke & Ishio, 1998; Reskin & Padavic, 2002). Combined, these studies paint a compelling picture of workplaces as arguably the most central vehicles of social stratiﬁcation in American society. But how do people ﬁnd their jobs in the ﬁrst place? How do job seekers gain the tools to participate in the labor force? A considerable literature exists on job search practices: techniques job seekers use (employment agencies, cold calls, newspaper ads, the internet, and networks), the effectiveness of such techniques, and racial, ethnic, and gender differences in job search practices (Bradshaw, 1973; Elliott & Sims, 2001; Granovetter, 1995; Hanson & Pratt, 1995; Holzer, 1988; Montgomery, 1992; Ports, 1993). But recently, sociologists and other social scientists have begun to assess the mediating role of formal organizations in the job search, examining labor market institutions in which people are enmeshed prior to applying for a job. Speciﬁcally, the role that labor market intermediaries – job training, job placement, and job search organizations (JSOs) – play in connecting people to occupational positions and labor markets has come under greater scrutiny. Such organizations directly and indirectly map job seekers onto a stratiﬁed occupational system. These organizations communicate to participants the jobs that are available to them and those that are not. They shape the quality, ﬂow, and utility of information about how to ﬁnd jobs.1 In theory, labor market intermediaries can give job seekers analytical and practical tools for taking some control over their careers, enabling greater self-determination, and indeed making it possible to participate on the best

Getting Jobs and Building Careers

377

terms possible in the labor force. For these reasons, it is important to understand whether and how they limit or enhance this outcome. This paper examines a population of labor market intermediaries that has received scant attention from social scientists: JSOs (for an important exception see Korteweg, 2003). By comparing JSOs that serve two different class- and occupationally based populations, this paper sheds light on one organizational vehicle for the reproduction of occupational inequality in the U.S. We argue that this understudied organizational population shapes, blocks, and expands people’s ability to acquire job search skills that might enhance their occupational and career opportunities. Studying the pedagogical approach, resources (space, technology, and stafﬁng), and discourse about achievement in each organization allows us to pinpoint how this stratiﬁcation process occurs.

LABOR MARKET INTERMEDIARIES Labor market intermediaries typically provide information about skills, occupations, and career development. They purvey advice about how to ﬁnd jobs and thus shape the effectiveness of the job search process (Halaby, 1988). Most scholarly attention has been focused on labor market intermediaries that provide job training and placement services. Examples of the former include training agencies that teach new skills to displaced, underemployed, and unemployed workers (Deliganis, 2003; Grahame, 1998; Grubb, 1996; Indergaard, 1999; Lafer, 2002; Negrey, Golin, Lee, & Gault, 2003; Otis, 2001; Riemer, 2001; Solomon, 2003), while examples of the latter include temporary-help stafﬁng agencies that place low-wage workers (Benner, 2002; Neuwirth, 2004; Osterman,1999; Rogers, 2000). Headhunting and stafﬁng agencies that place well-paid professionals fall into this category as well (Barley & Kunda, 2004; Finlay & Coverdill, 2002). The type of labor market intermediary analyzed here provides information about and assistance with the job search itself. JSOs do not place people in jobs nor do they provide job training in a technical or narrow sense. Rather, they provide job search education and readiness by transmitting guidelines for ‘‘labor market behaviors’’ (Rosenbaum, 2001): learning where to go for information about positions (newspaper ads, the internet, networks, and cold calling) and learning how to present oneself in order to impress prospective employers (resume´ construction, interactive skills for successful interviews, and emotion management). Also integral to educating people about how to ﬁnd jobs is introducing them to the facts of occupational opportunities

378

VICKI SMITH ET AL.

in any given region. For example, teachers and facilitators in JSOs present information about local labor markets: who is the largest employer in the area, what industry is the leading source of job growth, and what qualiﬁcations or credentials are required for which jobs. JSOs take different forms. Public-sector organizations (the subject of this paper) organized by county, state, and federal governments usually require no membership fee, making the barriers to entry very low and, theoretically, open to anyone. On the other hand, private-sector career search ﬁrms have high barriers to entry; for a steep price, they offer customized, one-on-one counseling services to individuals who can afford to use their services. Examples of such ﬁrms are Bernard Haldane Associates (http://www.jobhunting.com) and Richard Bolles’s corporation, which offers books and career counseling seminars. (Bolles is the author of perhaps the most widely read job hunt and career change book in the U.S., if not the world, What Color is Your Parachute?).2 Analyzing how JSOs teach people to search for jobs is necessary to explain one part of the larger puzzle of how occupational inequality is created. JSOs are bridge organizations, stepping stones for many on a pathway between school and work, between welfare and work (Hayes, 2003), or between jobs. Public- and private-sector organizations can deﬁne occupational and mobility opportunities: they sponsor learning and education, and extend life-enhancing resources. Organizations’ ideologies about mobility and opportunity – delivered through day-to-day interaction and curriculum – channel participants’ aspirations toward JSO-sanctioned goals. They have special potency for shaping social stratiﬁcation. We predicted that JSOs might maintain and recreate differential opportunity but that this would not be an obvious mechanical process whereby facilitators instruct clients in what they should, will, or cannot do. Rather, we expected that administrative, instructional, and curricular processes combined would communicate asymmetrical information about occupational positions and behavior as well as handicap or enable people to maximize their job search. In making this prediction, we drew on three bodies of literature, which suggest that formal, employment-related organizations play a key role in creating and reproducing class inequality. One is the abundant literature on gendered and racialized organizations that analyzes how work organizations are premised on implicit assumptions about gender and race. These assumptions, when embodied in structure and policies, create inequality between women and men, and between people of different racial and ethnic groups at work (Acker, 1990; Britton, 2003; Collins, 1996; Williams, 1995). The second body of literature looks closely at

Getting Jobs and Building Careers

379

the way in which gatekeepers in job training and welfare ofﬁces – case managers, training directors, trainers, and so forth – treat clients differentially according to their class, race, and gender (Deliganis, 2003; Grahame, 1998; Negrey et al., 2003; Otis, 2001; Riemer, 2001; Solomon, 2003). Researchers have found that gatekeepers assign clients to segregated training opportunities and classes, and often advise different and unequal employment options based on the same criteria. Riemer (2001, p. 21) argues that such organizational actors ‘‘impose identities’’ on welfare clients, irrespective of clients’ actual experiences and competencies. The third is the extensive literature on schools and education, which highlights how tracking processes, unequal resource distribution, and stereotyping on the part of teachers and counselors prepares students differentially for the labor market (Bowles & Gintis, 1976; Eckert, 1989; Ferguson 2001; MacLeod, 1987; Weis, 1990; Willis, 1977). Extending these theoretical frameworks to JSOs, we wanted to investigate embedded assumptions that might lead these organizations to create dissimilar outcomes for people of different class and occupational positions. Organizational rules, policies, discourses, and resources – all pointed to in this literature – can contain such assumptions. Secondarily, we wanted to explore the utility of what Rosenbaum (2001) calls the linkage model, which looks at institutions that connect individuals and labor markets. Focusing on schools, he argues: ‘‘Because societal linkages tend to offer incentives to advantaged students but not to others, linkages often magnify preexisting differences in human capital’’ (p. 4). We wanted to apply this theory to JSOs to analyze whether, as a linkage between individuals and labor markets, JSOs might perpetuate or magnify the gap between people who bring different degrees of human and cultural capital to their job search process. Comparing organizations that serve different class- and occupationally based populations would enable us to explore this issue of inequality and how these organizations perpetuate or increase it.

METHODS AND CASES In order to focus on organizational interactions and programs, we used a qualitative, organizational case study approach, including participant observation, informal interviews, and analysis of organizational documents. Because the research agenda was concerned with whether and how JSOs contribute to occupational stratiﬁcation, we chose organizations that

380

VICKI SMITH ET AL.

appeared to link people to two different segments of the labor market. Choosing and comparing two disparate organizational cases (Ragin, 1987) would allow us to identify the speciﬁc mechanisms by which information was transmitted to participants. The organizations we studied are located in an urban area in Northern California. The ﬁrst type of organization, the One Stop Career Center (One Stop), has approximately 480 centers in California alone (http://www.edd.ca.gov/ONE-STOP). One Stops were established to increase the employment, retention, and earnings of participants, enhance their occupational skills, and, in addition, to reduce welfare dependency and improve the quality and productivity of the workforce. As suggested in its name, One Stop Centers have a cluster of ofﬁces and services that are supposed to enable users to conduct all their transactions related to employment with the state in one centralized place. Here, clients can typically ﬁnd assistance with unemployment insurance, job search and training services, childcare, vocational education, vocational rehabilitation, and youth programs. We studied the job search component of these organizations. Although the One Stop Centers are varied in minor ways, in general, they are organized quite similarly. Thus, we treat them as representing one coherent organizational type in this paper.3 The second, the Urban Professional Network (UPN), is one chapter of a statewide organization called Experience Unlimited.4 There are approximately 20 of these chapters sprinkled throughout California. UPN has served low- and mid-level managerial, professional, and technical workers for many decades. In the last two decades, UPN has focused speciﬁcally on how to search for jobs in an economy plagued by downsizing, mergers, and corporate restructuring. It is based on the philosophy that professional, managerial, and technical workers need to ‘‘work together and share jobseeking knowledge and experiences’’ (EDD Factsheet, n.d.). The One Stop Career Centers and Experience Unlimited (and thus UPN) are sponsored by the California Employment Development Department. Private-sector, for-proﬁt employment organizations such as career consulting ﬁrms were excluded from this study in order to control for the costs of entry to using the organization; public-sector organizations are available to all people. Moreover, in order to compare JSOs that serve people from different class and occupational positions we had to study public-sector organizations; low-income people are priced out of the expensive, privatesector organizations. Analyzing state-sponsored JSOs also would potentially allow us to theorize about the role of state agencies in organizing occupational and labor market opportunity. Many sociologists and political

Getting Jobs and Building Careers

381

scientists have identiﬁed the ways in which the state regulates labor, both its quality and its availability; comparing agencies serving different populations of workers may illuminate state practices in preserving social and economic inequality (Peck, 2001; Piven & Cloward, 1971; Ridzi & Banerjee, 2006, in this volume). The research design called for focusing on the speciﬁc organizational mechanisms, routines, and pedagogical practices by which people were taught about, directed to, and discouraged from different occupational opportunities. Participant observation was carried out in UPN and in ﬁve One Stop Career Centers, at general orientation meetings, and speciﬁc preparatory workshops on composing resume´s, how to interview, and how to negotiate compensation. In both settings, informal conversations were held with other job seekers. We observed relevant features of the organizational settings such as spatial and architectural arrangements, the content and display of organizational artifacts (brochures, handouts, signs, and posters), and the degree of user friendliness. These features can manifest subtle codes in expectations that organizations communicate to clients (Hasenfeld, 2000; Schwartzman, 1993). The analysis reported here is based on a total of 120 hours in the ﬁeld over a three-month period. Documents such as brochures about the formal goals or mission of the organizations, ﬂiers advertising job opportunities, curricular material used in workshops, self-help pamphlets, calendars of activities hosted by the organizations, and materials that advertise training opportunities were collected from the different organizations. We analyzed these organizational artifacts for discourses about jobs, mobility, and opportunity. Our study is limited in that our data do not allow us to determine the long-range effectiveness of either of the two types of programs we studied. While we were able to observe clients’ reactions to the programs during stints of participant observation, we did not formally interview them. The research reported here does not establish whether each organizational type successfully imparted its messages – a topic best addressed in an in-depth interview study that would explore the degree to which participants internalized or rejected such messages, or a longitudinal study that tracked actual labor market outcomes after participating in these organizations. Rather, we ask whether these state-sponsored organizations and the way they organize job search education directly and indirectly shape one’s access to occupational opportunity and if so, how they do it. Although ofﬁcially open to all job seekers, we found that the One Stops’ constituency was mainly unemployed and working poor women (African

382

VICKI SMITH ET AL.

American, Hispanic, white, and Asian/Asian American) and secondarily men of color of similar race and economic standing. Participants were either transitioning from welfare to work (primarily women), were moving around between the primary and secondary labor markets, or within the latter (women and men). Centers had small-paid staffs – which included welfare case managers, program managers, receptionists – but no single individual was exclusively dedicated to the job-training component of the organization. Job search seminars were conducted by individuals who were contracted by the organization to educate clients in labor market dynamics and in job search procedures. These trainers rotated in and out of the centers on an irregular basis. The UPN membership was comprised of a majority of men (approximately 60%, in descending order, White, African American, Hispanic, and recent immigrants) with white women and women of color comprising the remainder of the membership. The vast majority of UPN members was made up of unemployed or underemployed low- and mid-level professional and technical workers; only a few were higher-end professionals and managers. The majority, especially of the white men, tended to be over 40 years of age, with many closer to 50-years old. UPN operated with one paid Employment Development Department (EDD) staff member dedicated exclusively to overseeing this organization. Virtually all activities, meetings, and workshops, however, were conducted by volunteers, who were members themselves. Volunteers worked with a curriculum produced by former members, handed down to newer generations of job seekers, and regularly modiﬁed by those new cohorts. How is it that the two organizations’ constituencies are so differentiated, when One Stop Centers are open to anyone? As we will show, UPN members put up many boundaries to ensure their professional/managerial base, allowing this organization to stay homogeneous, with respect to class in particular. The explanation for why professional/managerial workers stay away from the One Stop Centers may lie in the gatekeepers of the EDD and of the One Stops themselves. In the EDD ofﬁces, staff are known to direct the unemployed to UPN speciﬁcally, when staff perceive the individual to be of the professional/managerial class. Staff relied on a variety of cues to make this judgment: information from resume´s, knowledge of job seekers’ credentials, and clothing, speciﬁcally professional attire. And in the One Stops, staff made comments about how UPN was the place to be for professional workers. These perceptions and comments play a sorting role and may explain who ends up where.

Getting Jobs and Building Careers

383

ORGANIZATIONAL LOGICS Publicly available organizational documents represent the ofﬁcial face of an organization. The content of many of these artifacts implied that the One Stop Centers and the UPN had a common approach to educating people about how to search for jobs. Their shared literature came from the EDD, which sponsored and funded the organizations. The content of a number of each organization’s documents, such as pamphlets about the importance of interviewing, networking, attitude adjustment, or skill assessment, struck a similar tone. Both organizations’ literature stressed the need for self-help, optimism, and individual initiative. Both urged participants to conduct background research on companies they were interested in. Inspirational posters proclaiming that ‘‘Attitude is a little thing that makes a big difference!’’ adorned the walls of UPN and two of the One Stops. Each organization offered workshops on topics such as how to interview, network, write a resume´, and ﬁnd ‘‘hidden jobs’’ in the economy. They provided contact information for other agencies that provided support services for the unemployed. Various documents emphasized how participants could improve their interpersonal skills through greater awareness of body language, eye contact, and projection of attitude. In other words, content analysis of these shared documents, their advice and their rhetoric, signiﬁed that UPN and the One Stops were neutral with regard to who their clients would be and their expectations for their clients.5 The technological infrastructure, also supplied by the state, similarly revealed common approaches to job search education. In each organizational setting there were anywhere between seven and fourteen personal computers (PCs) for clients to use, a toll-free telephone for placing calls about job prospects, and a copy machine. However, divergent logics were also revealed when we participated in organizational workshops and observed the recommendations and tools given for conducting the job search or for controlling job outcomes. We found, embedded in each organizations’ practices and structures, different assumptions about the capabilities of the clients and their occupational potential. Each organization embodied a unique logic that worked at two levels. First, the social-structural, programmatic, and discursive organization of each site equipped clients with disparate sets of job search skills, thus producing uneven capacities for exercising control over occupational outcomes. Second, each organization attempted to guide clients’ aspirations in ways that encouraged them to expect segregated occupational opportunities: low-wage work, in one case; middle-wage work but potentially downwardly mobile

384

VICKI SMITH ET AL.

middle-class work, in the other. Thus, each logic reproduced class at the level of organizational practice. We derived the logics of the two models by focusing on four organizational data points which we culled from the organizational literature discussed earlier: resources (space, time, and stafﬁng), the model of learning (pedagogy), behavioral norms modeled through scheduling practices and discourse, and the construction of aspirations. The job search logic of the One Stop Career Centers was a hierarchical approach that positioned clients as undifferentiated and passive consumers of information. The curriculum was organized top down, discouraged critical thinking, and cultivated a ‘‘quick ﬁx’’ solution to clients’ unemployment and underemployment. Through the One Stops’ social organization of resources and workshops, and because of tight linkages with low-wage employers, clients were pressed to quash any aspirations they had for work that was other than low wage and dead end. The routine operations of the Centers modeled dysfunctional work norms to clients. In sum, clients were not given the opportunity to develop job search skills or collaborative experiences that might allow them to plan for the long term, or might empower them in their efforts to move out of their current class location. The job search logic of UPN was a participative approach that enabled clients to become active creators of information. This approach cultivated a ‘‘long-term solution’’ to clients’ unemployment and underemployment. Clients collectively created and ran the organization, incorporating the norms of professional and managerial work into their codes for behavior, meetings, workshops and other events. The range of occupational possibilities presented in workshop discourse was comparatively open and the resources of the organization enabled clients to map out plans for career development and job acquisition. At the same time, through its collective practices, the organization signaled the need for participants to scale down occupational and mobility aspirations.

JOB SEARCH EDUCATION: A HIERARCHICAL, PASSIVE APPROACH Resources One Stop Career Centers were neighborhood based, and thus were decentralized geographically. Situated on the premises of other community organizations (such as Urban League ofﬁces) and bundled together with other social service units, One Stops lacked sufﬁcient dedicated organizational

Getting Jobs and Building Careers

385

space and time to allow clients to search for jobs or exercise skills learned in workshops. Because the facilities were modestly sized, opportunities for sustained collective interactions among clients were rare. Although technologically well equipped, each One Stop restricted clients to 45 minutes use of a PC in one sitting and computer labs were monitored by ‘‘lab wardens’’ (paid staff). Restriction of time had two implications. First, members were limited in their ability to use computers for their job search (such as exploring the internet, or editing resume´s or cover letters). Second, the restriction on time spent at the computers diminished the amount of time spent in the facility; we observed little ‘‘hanging out’’ during which clients could share information with others. Restricting computer time meant restricting Center time. Pedagogy The Career Centers’ pedagogical model of learning was hierarchical, based on the assumption that the leaders were the experts and the students were passive vessels (consumers) into which scripted information was poured. Each Center offered a three-day set of workshops on core issues about the job search: resume´ writing, learning how to tap the hidden job market, and interviewing. In workshops, leaders principally lectured to participants. They asked rhetorical questions that elicited one-dimensional answers (for example, ‘‘How much money would you like to make?’’) but participants were not asked for feedback about whether the workshops addressed their speciﬁc needs, constraints, or experiences. It was a ‘‘one-size-ﬁts-all’’ program. None of the workshops was centered around reﬂection, either on personal experience or on the larger conditions of the economy. By limiting information and feedback and by not involving clients in their own learning process, One Stops foreclosed one opportunity for clients to learn how to purposefully take more control over their job search process. Building time for group discussion into workshops could have allowed participants to share success stories as well as stories of failure from which others could learn: how to approach employers, how people could emphasize their personal strengths, and so forth. Instead, the workshops created a passive learning environment. The hierarchical pedagogy we observed resembled the ‘‘skills and drills’’ approach (Grubb & Kalman, 1994) taken in remedial education classes. The key elements of the passive skills and drills approach are that students are given few opportunities to express or elaborate on their own proﬁciencies and interests, they learn only from the standardized curriculum and not from one another, and complex ideas and concepts are broken down into simplistic units. In addition, students ‘‘are deﬁned in terms of what they are

386

VICKI SMITH ET AL.

unable to do – their deﬁciencies within a rigid hierarchy of skills’’ (Grubb & Kalman, 1994, p. 73). Teachers assume students are not capable of learning in a less-routinized way and appear to devalue their capacity for digesting more complicated material. In the absence of more collectively structured discussions, clients were unable to get feedback from leaders about how they might possibly modify some of the advice in order to adapt it to their speciﬁc circumstances. One important consequence of the hierarchical approach, in which it was assumed that workshop facilitators had expert knowledge that clients should follow, was that the advice and prescriptions were often irrelevant, given clients’ resources and their capabilities. Clients were often visibly discouraged when they realized that they could not apply certain snippets of expert advice to their own lives because of material constraints and limitations in their social capital and budgets. For example, a common piece of advice to people searching for jobs was that everyone should constantly network. Networking, these workshop leaders argued, would connect job seekers to information and opportunities. A One Stop workshop leader repeatedly mentioned that everyone should have business cards and they should circulate their cards to absolutely everyone in their network, ‘‘to hundreds of people.’’ On the face of it this may be a reasonable suggestion. However, clients indicated that they did not possess the ﬁnancial means, the technology, or technological savvy to make business cards. Several participants noted that the network of people to whom they could circulate business cards was extremely limited and tended to consist of friends and family who were in similar positions – they were either unemployed or had secondary labor market jobs, not professional acquaintances who might be able to help them ﬁnd better jobs – a point the trainer shrugged off. It was difﬁcult to meaningfully critique this piece of advice because it too was handed to participants as a simplistic unit of information (Grubb & Kalman, 1994). Modeling Behavioral Norms Workshops were often run in irregular ways that conveyed norms of unreliability and lack of professionalism to Center clients. In so doing, One Stops devalued participants’ time and failed to model marketable employment norms. Virtually every workshop we attended started late (anywhere from 15 to 45 min) with no reasonable explanation given to those who showed up on time. One workshop had been poorly advertised, had only two participants, and was subsequently canceled, but only after the two had

Getting Jobs and Building Careers

387

arrived. At a Career Center that served primarily Hispanic women, staff had promised that several representatives from local ﬁrms would be there to give information to participants and talk to them about job opportunities; none showed up and the staff person was not sure whether anyone had called to conﬁrm his/her participation. Workshop leaders’ presentations were only occasionally well prepared, often meandered, and did not spend adequate time on the central topic that clients had hoped to learn about. For example, in a 2 hour workshop on preparing resume´s, the leader only started talking speciﬁcally about how to create an effective resume´ 10 minutes before the end of the scheduled time. When two of the workshop members expressed their distress with this, the leader instructed them to return the next day for more information and was not sympathetic to their assertion that they could not return then. In general, the job search workshops lacked professionalism, consistency, and continuity, signaling that clients’ time was unimportant. While the Centers had very small paid staff, including receptionists and program managers, the job preparation seminars were conducted by different individuals who were contracted by the organization to educate clients in labor market dynamics and in job search procedures. The fact that these ‘‘labor market entrepreneurs’’ were independent contractors meant that the curricular material – advice, curriculum, and interpretation of opportunity and the job search process – was irregular from seminar to seminar. It also meant that these entrepreneurs had little personal or professional investment in actual job outcomes for their clients. Organizational staff transmitted unprofessional behavioral norms that, if enacted, would undermine clients’ ability to hold down a job on a long-term basis, thus compromising their ability to break out of the low-wage sector of the job market. Speciﬁcally, leaders gave self-defeating advice about work behavior. In one workshop, the leader was discussing jobs in call centers (all the One Stop Centers emphasized call center employment, near-minimumwage telephone work) and encouraged participants to seek them out, saying, ‘‘All you need to do is to be alive and breathing,’’ contradicting the organizational literature that stressed the need for an attentive attitude and lively demeanor in interviews. (Another layer of meaning in her comment was that if someone exhibiting slug-like behavior could get a job in a call center, certainly the workshop attendees could too.) In another workshop, a male participant worried aloud that, because he relied on public transportation, he would have to leave his home quite early in the morning to get to a fork-lift training class he was interested in taking. A staff member who was helping him with paperwork dismissed his concern and told him to just take a later bus and get to the class a few minutes after it

388

VICKI SMITH ET AL.

started each morning – without acknowledging the possibility that this lessthan-professional pattern might make it difﬁcult to get a reference from the course trainer and eventually to be taken seriously as a candidate for a job in this ﬁeld. Finally, the normalization of weak commitment to organizations was conveyed routinely through the ‘‘drop in’’ approach to participation. Staff at One Stops encouraged but did not require participation in ongoing or multiple activities. Although individuals could sign up for workshops in advance, they were not required to. In the lexicon of the One Stops’, the participants were called ‘‘clients,’’ which suggested that participants should just ‘‘shop around’’ for something useful to ‘‘consume.’’ Nor did clients have to devote any particular number of hours to their search. The open-ended norms about participation communicated that there were no beneﬁts to developing a stable commitment to the program and that there was no problem, from the staff’s point of view, in using the Centers’ services in a casual, disorganized way. An indicator of weak commitment was reﬂected in turnover and attendance patterns. In various three-day series of workshops we attended, which on any given day would have seven to nine participants, it was typical that only two participants other than the researcher attended all three classes. Tempering Aspirations The discourse and the practice of the workshops discouraged One Stop clients from developing aspirations aimed above low-wage and otherwise disadvantaged jobs. They did so in a variety of ways. First, workers were actively discouraged from setting their sights on jobs that might pay living wages. For example, one leader asked participants to state the wages they hoped to earn. When a participant said that she hoped to earn about six dollars higher than the minimum wage, the leader warned her to scale back her expectations. He emphasized to the class that she would never be able to command that particular wage level. Leaders discussed low-wage service and health care jobs (for example, cashier jobs at big box stores, food preparation jobs at a large, local sports arena, and temporary light assembly and clerical positions). No one discussed alternative strategies for improving one’s long-range standing, such as ﬁnding out how to attend community college or other credentialing programs outside the purview of EDD. Leaders similarly gave advice that would disadvantage participants when it came to accepting jobs. Moe, for example, advised participants in his workshop to end their interview by saying ‘‘I’d like to say that I would be very happy to accept this position if it were offered to me,’’ an acquiescent

Getting Jobs and Building Careers

389

contrast to what we would ﬁnd later at UPN. One workshop trainer tossed out a question: ‘‘If offered a position, do you accept the wage or do you try and jack it up?’’ Two workshop participants said, ‘‘Accept the offer,’’ and the class leader concurred: ‘‘Chances are you’re not going to get a higher wage; you should accept,’’ adding that, ‘‘Life’s a bitch, baby; live with it.’’ In these ways clients were discouraged from negotiating for better wages, work conditions, or schedules. Indicative of the low expectations that the One Stop Centers had for clients, recruiters from Wal-Mart were on site several times a month accepting applications from center clients. This ‘‘partnership’’ constituted a direct linkage between the Centers and low-wage, low-skill jobs. Indeed, a leader in one workshop told a client to leave the workshop – which was on resume´ writing, a skill that conceivably could enable the client to conduct his or her own independent job search – to go meet with an on-site recruiter from Wal-Mart. In addition, representatives from temporary help agencies encouraging participants to sign up for assembly and clerical jobs were consistently observed at the Centers. In this way, the One Stops directly channeled low-income workers into the low-wage, low-skill sector of employment. The programmatic organization of the Centers placed limits on clients’ occupational outlooks and ambitions. Clients were pressured to accept jobs rather than visualize careers or decipher the game of upward mobility. The limited resources, hierarchical and passive model of learning, inculcation of weak professional norms, and quashing of aspirations collectively signaled to One Stop participants that they should use the organization on a shortterm basis, that their time was not valued, that they should settle for quick, unempowering solutions (take a job vs. develop a plan, the quick ﬁx), and that their occupational placement would be limited.

JOB SEARCH EDUCATION: A PARTICIPATIVE, ACTIVE APPROACH The degree to which learning processes and organizational structure limited job seekers in the Career Centers becomes especially salient when compared to processes and structure at UPN, another state-run JSO. Resources UPN was a comparatively time- and space-rich organizational environment that allowed people to adopt a long-term approach to their job search. It

390

VICKI SMITH ET AL.

was located in a large site adjacent to the EDD itself. UPN members who needed the services of EDD had convenient access to both. The space itself was substantial compared to the One Stop Centers, with a central area that could seat 80–100 people for regular membership meetings. At the periphery of this area were individual cubicles in which people read through job binders or newspapers, or otherwise worked on individual projects related to the job search. In a separate area of UPN, members used a phone bank to make toll-free phone calls to employers in the state. There was also a line of computers, which members used to conduct their job search (e.g., check the internet, prepare resume´s, download forms, make inquiries, and so forth) without explicit limits on time. Many members worked together, learning how to mine the Internet expertly for job information for the ﬁrst time. Finally, an additional room was used for smaller workshops. Long tables and a white board projected the feel of a professional meeting room. On any given day that we attended, we found 10–20 individuals at the facility, attending workshops and committee meetings, using computers, browsing through job binders, and quietly socializing. Often they were brainstorming about appropriate tactics to use to achieve speciﬁc professional outcomes (such as where one could network with lawyers or how one could gain access to ‘‘good’’ high-tech employers in the area). It was common to see the same clients repeatedly over the course of the three months in the ﬁeld. A large desk was positioned at the entrance to the facility and a UPN member served as the receptionist, policing its boundaries. One had to be a registered member to enter and use UPN. If not a member, one would be given general information and an application to ﬁll out in order to join. The receptionist answered the front phone and was available to take messages (typically from inquiring employers) for members, giving them a near-professional message service. The ample space and support technology for UPN members invited sustained and long-term work on individuals’ job search processes. Pedagogy UPN’s model of learning was based upon the philosophy that networking between members was essential, both to reproduce the organization over time and to help everyone who came to use its services. Ostensibly, networking enabled people to make contacts and ﬁnd jobs, but at a deeper pedagogical level it required members to view their own insights as crucial. Rather than being positioned as passive recipients, UPN members were

Getting Jobs and Building Careers

391

positioned as active agents, experts about work and employment, and able to beneﬁt from others’ insights about ﬁnding good jobs and making connections. UPN’s approach made it possible for people to build and maximize social capital and to customize their searches to maximize their professional goals. Members began their relationship with UPN as members of a cohort. Each cohort went through a four-day training session together. Subsequently, the cohort became a source of information, critique, and moral support. In fact, each cohort stood as a ready-made social network. In a required orientation, cohort members learned about the philosophy of UPN, its history, and its pedagogical structure. The organizational creed was that they were each other’s teachers and that the organization was little but the collective experiences and insights of the members. Former members had developed the curriculum for orientations and for a variety of workshops, ranging from learning about the labor market to handling emotional distress, preparing resume´s, using the internet, and videotaping and critiquing mock job interviews. This information had been assembled into a binder of course training materials. Members who ran workshops customized and modiﬁed the workshop by using their own experiences as illustration. Reciprocity and interdependence were enforced in the way that staff and leaders talked about relations between UPN members. The staff person stressed in an orientation meeting, ‘‘Treat this place like work. This is your new ofﬁce; act like you are in the workplace. Sexual harassment is not tolerated here. Someone here might be your connection for your dream job and that would not serve you well here.’’ Comments such as this provided an underpinning of social control over members’ behavior. In workshops, participants commented on one another’s attitudes, behavior, and experiences. In a workshop on interviewing, for example, cohort members gave Kevin, a geologist in their group, a block of cheese to go with his ‘‘whine,’’ good-naturedly criticizing him for complaining continually about his circumstances. Another day a cohort member, Fred, saw a job listing for a geologist in a locale where Kevin had expressed desire to work. He reported this to Kevin. Kevin claimed there were no jobs in that area, whereupon Fred went and printed out the listing. Kevin applied, interviewed, and shortly thereafter got the job. This story was heralded as living proof of the power of collaboration and networking. The participatory learning model both opened up avenues of information but also exerted a modicum of peer pressure on each member of UPN. Interdependence increased the degree to which all members monitored fellow members’ job

392

VICKI SMITH ET AL.

search practices and professional behavior. Unlike the One Stop Centers, workshop leaders and other UPN members were personally and professionally invested in their comrades’ outcomes. Every occasion on which a member obtained a ‘‘good job’’ was taken as evidence of an outcome that others might experience as well. Modeling Behavioral Norms From the ﬁrst moment of entry through the doors of UPN, members and the staff person patrolled professional codes of appearance, behavior, and activity. UPN’s policies, found in the printed orientation material, required predictability, commitment, and respect for others’ time. Participants were called ‘‘members,’’ underscoring an expectation that they were part of a meaningful community rather than passive consumers. Members were required to attend a week of orientation classes, volunteer a set number of hours per week at the organization either as a member of a committee (computer, resume´, and so forth), a workshop facilitator, or to assist new members with their own searches. They were also required to attend weekly membership meetings, where new members introduced themselves, the staff person made general announcements, and current members shared information about jobs, interviews they had had, or job offers they had received. In contrast to the irregularity of the Career Centers, UPN did not convey norms of unreliability or chaos. UPN seminars started when advertised and if a workshop was undersubscribed, members were called by the receptionist (an UPN member) and informed of the cancellation. Combined, these practices signaled that UPN members’ time was valued by the organization and that the organization’s professional norms must be upheld. UPN’s organizational and curricular model, therefore, required members to enact sustained involvement and participation, and professional codes of behavior and interaction. Much of the workshop material stressed the dangers of falling between the cracks, of not being part of a network, or of not staying on one’s toes 100% of the time. Participating in this organization and staying embedded in the UPN network allowed participants to engage in conditions of action that mirrored what they hoped would be the conditions of managerial and professional work: teamwork, continual learning, and networking opportunities. Finally, adherence to professional norms was encouraged through clothing and behavioral rules. UPN had a written dress code specifying that members should project a ‘‘conservative business appearance.’’ If an individual showed up inappropriately dressed, there was a rack of clothes from

Getting Jobs and Building Careers

393

which they could borrow while they were at UPN. When members violated these rules, there were processes in place to bring the individual in question in line. For example, inappropriately dressed individuals could be negatively sanctioned. According to the ‘‘Professional Attire Guidelines,’’ ‘‘If an individual’s attire distresses three people, and it is brought to the attention of the President, then the President may approach that person in a tactful and respectful manner to discuss closer compliance to the dress code.6 There is no reason for a ‘dress police’’’ (authors’ emphasis).7 Second, the print material provided a lengthy list of unacceptable behavior (people were required to be civil, punctual, to control their tempers, and not use telephones or computers for uses other than job searching). What happened if members failed to live up to these requirements? Violators of UPN rules about participation were not ‘‘punished’’ visibly. One way that participation requirements were enforced was to bar access to the resources at the facility if individuals did not complete the four-day orientation session. Because many of the same people were observed over the three-month period of research, we concluded that members by and large adhered to UPN’s requirements for participation. Everyone we observed adhered to the dress code as well.

Tempering Aspirations UPN did not quash or limit aspirations as bluntly as the One Stop Centers. Rather, this organization gently signaled to participants the need to scale down occupational and mobility aspirations. Participants were cautioned to adjust their goals downward in recognition of the turbulent professional labor market and the destabilization of professional/managerial employment. A recurrent theme throughout the staff person’s proclamations, which was also integral to curricular material, was that to survive in the current economy members must leave opportunity doors open, and therefore must be willing to accept a ‘‘bread and butter job.’’ The bread-and-butter job fell well below the threshold of the ‘‘dream job,’’ the well-paid professional position with opportunities for advancement. When people spoke of bread-and-butter jobs, they typically referred to jobs that were entry level (thus, not ‘‘professional’’); temporary (rather than career oriented); and would barely pay a living wage (in contrast to a professional salary). At general meetings, the staff person who read out updates about past and current members always emphasized when someone had to accept a bread-and-butter job instead of his/her dream job, normalizing the bread and butter job as something that

394

VICKI SMITH ET AL.

was acceptable even if not ideal. In one workshop, the leader told the group that it had to ‘‘discover who you are and what your needs are,’’ that it was ‘‘important to recognize the characteristics of your dream job but to recognize that reality entails survival,’’ and often a bread-and-butter job. Throughout the week long orientation, participants reminded each other of the deleterious conditions of the labor market for professionals and managers in the new economy. While leaders often cautioned people against dwelling on structural forces that were out of anyone’s control (indeed, one workshop was devoted to acknowledging one’s emotions about this but learning how to let them go), few ever denied that such forces were a reality or that they powerfully conditioned anyone’s ability to land their dream job. A process of adjusting members’ expectations for the possibility of downwardly mobile middle-class status also emerged when repeated references were made to a labor market of scarcity, in which ‘‘dream jobs’’ were few and the competition to obtain them stiff. A trainer presented a statistical schema which outlined how only about 15% of people had jobs they love, 55% of people are ‘‘punching in and punching out’’ and 30% of people ‘‘actively sabotage the system,’’ cementing the sense that few dream jobs existed and there was much competition for them. At the same time, trainers insisted that UPN members should maintain an upper hand by refusing to state their desired salary to employers and always leave room for negotiating the best salary possible. The effect of telling people there was a scarcity of good jobs (employers have the upper hand) but stressing the importance of negotiating savvy (employees have the upper hand) conveyed contradictory messages about the professional/managerial labor market.

COMPARATIVE FINDINGS The JSOs analyzed here reveal two different logics about how to ﬁnd jobs and what to expect from the opportunity structure, as summarized in Table 1. Taking a hierarchical and short-term approach to low-income and working-class people, and a participative and long-term approach to middle-class individuals, One Stop Career Centers and UPN create differently abling environments in which people can learn about how to ﬁnd jobs and what they can achieve. Thus, state-sponsored JSOs seem to operate much in the way Rosenbaum (2001) suggests about the unequal ways that schools link people to the labor market; institutions connecting individuals

Getting Jobs and Building Careers

Table 1.

Target population Resources

Pedagogy

Modeling of behavioral norms Tempering aspirations Linkages to employers

Solution to ﬁnding a job

395

Comparison of Two Job Search Organizations. Career Centers: Hierarchical

UPN: Participative

Low income and working class Restricted: resources discourage taking time to ﬁnd a job and ‘‘hanging out’’ Top-down, hierarchical: onesize-ﬁts-all, treats clients as passive recipients

Middle class Expansive: resources encourage collaboration and spending time at the facility Participative, collaborative: treats members as active agents, customized to individual needs Regular scheduling, heavy emphasis on professional work norms and conventions Aim for the top but brace yourself for something less: possible downward mobility Loose: information is available about different companies but actual recruitment does not happen The long-term approach

Erratic scheduling, counterproductive advice, weak work ethic A ceiling is placed on aspirations aimed higher than working class jobs Tight: presence of low-wage employers recruiting and accepting applications on site The quick ﬁx

and labor markets can have the effect of magnifying and at the very least maintaining preexisting differences in human capital. The comparative analysis of this paper highlights this linkage and supports numerous other research studies, which have found that formal organizations play a crucial role in tracking and sorting people on the basis of class. Organizations serving low-income clients rendered these clients as passive consumers. Clients were encouraged to accept a limited range of jobs and not given tools to interpret and navigate the job market on their own. Norms of unreliability and underdeveloped work ethics were embedded in organizational processes: in the ways that workshops were structured, in the inconsistent and sometimes irrelevant advice offered, and in the discouraging feedback given to clients about their potential to demand more from their employers. The One Stop Centers’ setup did not encourage people to investigate, shop around, or hold out for a job that would confer greater opportunity or dignity. The social organization of Career Centers reﬂected the assumption that clients did not have the time, inclination, or resources to engage in a more protracted job search. But in organizing its activities on such an

396

VICKI SMITH ET AL.

assumption, One Stops perpetuated a cycle of disadvantage by denying clients knowledge of how to engage in a long-term search. Even when lowincome people are constrained by poverty, lack of education, or marginalization in social networks, it is reasonable to argue that these conditions will never change if people are not given the tools and the insights to escape them, a dilemma noted by Hayes (2003) in her examination of welfare workers who were assisting women moving off welfare into the workforce. Furthermore, like Grubb and Kalman’s (1994) remedial education programs, One Stop Career Centers’ pedagogy failed to impart the skills and competencies that are crucial for the 21st century occupational structure, a criticism that similarly has been leveled at job training organizations (Indergaard, 1999; Lafer, 2002). The authors note how abilities to problem solve, listen, negotiate, and ‘‘know how to learn’’ are crucial for the contemporary economy (Grubb & Kalman, 1994, p. 75). The top-down, passive approach – the skills and drills approach – taken by the Career Centers diminished the potential for gaining such competencies. The Centers’ pedagogy assumes that the clients were not capable of learning from one another or working collaboratively, that there was nothing of value they could learn from each other, and that they had no need to develop longer-term strategies for employment and mobility. Thus, the One Stop Career Centers’ approach to working with its clients potentially was a self-fulﬁlling prophecy that handicapped them from moving up in the occupational structure and failed to deliver the tools for long-term self-sufﬁciency. UPN resources, in contrast, allowed members to engage in the job search in a comparatively generous manner, under conditions that approximated those of professional work. With their tools, technologies, and resources, members had the ability to pursue job leads in an autonomous fashion. Unlike at One Stops, UPN pedagogy assumed that middle-class people were active agents, capable of learning from one another and working collaboratively, and that there was value in what they could teach and learn from one another. Embedded in the social organization of UPN was an assumption that members were in it for the long haul, that the job search was an unfolding process, not a one-shot deal. This is not to say that all members could support themselves through a prolonged job search but it is to say that if they were able to, this organization would buffer them from the worst of the market by helping them maximize the ‘‘opportunity’’ to be choosy and wait for the best offer possible. At the same time, the combined practices of UPN worked on members to qualify their aspirations for ‘‘good’’ professional and managerial positions. While encouraging them to keep their sight on such

Getting Jobs and Building Careers

397

positions – dream jobs, – the curriculum and discourse continually reminded them that they might be unattainable. A subtext of much of the material presented constituted a rhetoric of ‘‘brace yourself,’’ thus normalizing unpredictability of employment and possible downward mobility. Some observers might say that the emphasis on networking in UPN was ‘‘natural’’ because these professional and managerial workers had more extensive and more complex experiences, and were more advantageously located in social networks that could beneﬁt their fellow UPN comrades. Correspondingly, networking and collaborative learning in the One Stop Centers would not yield much because Career Center clients lacked these insights, experiences, and connections. Yet this overlooks the self-empowerment that can occur from collective reﬂection, when groups of people discuss and comment on shared experience, no matter how that experience has been limited and conditioned by the inequalities of class, race, or gender. Mutually discussing failures, experiments, and successes could motivate and model alternative perspectives: developing long-term strategies for mobility, acquiring educational credentials, and learning about expectations for professional work.

CONCLUSION JSOs encourage and discipline aspirations for occupational outcomes. As they relay information about labor markets and how to ﬁnd jobs, JSO trainers and facilitators shape clients’ insights about their occupational strengths and weaknesses, and about their chances of ﬁnding certain kinds of jobs. They equip people differently to control their labor market outcomes. Studies cited at the beginning of this article have documented how employers’ biases about gender and race inform their beliefs about who can perform particular jobs and their behavioral expectations for different groups of individuals: in general, such biases determine who they are willing to incorporate into the workplace and where. Our study strongly suggests that assumptions and biases about the class positions of different groups similarly inform the views of trainers in an understudied population – JSOs – as well as in the social organization of the agencies themselves. In other words, selectively rendered advice and prescription are shaped by JSO leaders’ as well as the state’s class-speciﬁc perceptions of the employment and mobility prospects of clients. This directly affects racial inequality as well given that each organization serves racially distinct populations (primarily people of color at the One Stops and primarily whites at UPN). This

398

VICKI SMITH ET AL.

information affects clients’ options, opening up opportunities or constricting them. One important ﬁnding from this comparative study is that the state, via JSOs, actively regulates the ﬂow of labor and conditions its quality. For middle-class workers, the state’s role in shaping information and advice is embodied in its support for an environment of abundant resources that allows middle-class job seekers to take a long-term approach to ﬁnding work and its support for a collaborative environment which is run by members. At UPN, the state gives people an opportunity to survey changes in the economy and occupational structure and to adjust to those changes. In the organizations for less-privileged people, the state creates a culture of passivity and discouragement and increases the degree to which they are likely to take any job. This study suggests that the state plays a critical and routine role in directing a disempowered population into the low-wage labor market. The individuals who are pointed toward low-wage, disadvantaging jobs have not been able to develop the capacity to negotiate their way out of this niche or to determine what alternative routes there might be to employment. Our study indicates that state-sponsored organizations serving diverse populations – JSOs targeted at people located at different levels of our class system – actively construct workers with divergent aptitudes, behaviors, and expectations. This is of particular concern given current political rhetoric about ‘‘empowering’’ poor people by forcing them off welfare and expecting them to become self-sufﬁcient with respect to earning a living wage. JSOs open a sociologically valuable window onto the creation and denial of social status and opportunity in our society. Hasenfeld (2000) has argued that human service organizations, especially those sponsored by the state, confer ‘‘moral judgment’’ about the worth of their clients through organizational discourses, labels, and practices. Drawing on new institutionalism, Hasenfeld (2000) notes that organizations engaged in moral work ‘‘uphold dominant values about desired behavior; they enforce these values through laws, rules, and regulations; and they provide typiﬁcation schemes to categorize and classify people’’ (p. 33). In this view the state, acting through its various human service organizations and relying on a variety of organizational mechanisms, afﬁrms the social worth of high-status people while denying or minimizing the social worth of those at the bottom of the social hierarchy. State-sponsored JSOs are an example of such human service organizations: they process people and serve them, but in strikingly unequal ways. They are organizational vehicles that sort, categorize, and conﬁrm or deny power to people trying to make their way through the occupational structure in the United States.

Getting Jobs and Building Careers

399

NOTES 1. See Granovetter (1995) for the importance of information ﬂows. 2. It would be extremely difﬁcult to specify how many people use job search organizations. The U.S. Census Bureau does not track this phenomenon although it does track job search activity on the internet (Statistical Abstracts of the United States, 2004–2005, Table 604). There are countless mass-market books written for job searchers (as is evident from a trip to any major American chain or college bookstore). Sales ﬁgures for What Color is Your Parachute? indirectly give a sense of the magnitude of the topic. By 2004 Bolles’s book had sold eight million copies, and was on the New York Times best selling-paperback list for 288 weeks. In 1995 the Library of Congress named it one of ‘‘25 books that have shaped readers’ lives’’ (www.jobhuntersbible.com) and it has been translated into 12 different languages. A new edition of this book is published every year. Job search web sites such as careerbuilder.com and monster.com have been used by millions of Americans in recent years. While not all readers of Parachute or all visitors to web sites actually use job search organizations, we feel conﬁdent in asserting that a considerable number of Americans look to them for information and guidance. 3. These ofﬁces are not synonymous with the public welfare ofﬁces identiﬁed by Handler and Hasenfeld (1997), Hayes (2003), Peck, (2001), and Riemer (2001), whose explicit emphasis is on personal responsibility, work-ﬁrst and workfare, and in general, getting welfare recipients off welfare. Such agencies mandate quick entry into the low-wage labor market. Although women previously on welfare are one constituency of the One Stops, and while the One Stops are clearly a potential source of information and assistance for women on welfare (or TANF, Temporary Aid for Needy Families), One Stops’ mission in the state’s panoply of employment services is to serve a more diverse clientele. 4. UPN is a pseudonym for this particular chapter of Experience Unlimited. 5. One important indication that the two organizations might serve different clientele is that the One Stop Centers provided workshops on employment issues for exoffenders, while UPN did not. 6. The President was a member who was elected by the general membership. 7. There were no written rules about clothing at the One Stops. Participants were nearly always dressed casually, ranging from shorts to jeans to Dockers slacks. With one exception, workshop leaders were usually dressed in casual business attire, with pressed slacks but casual shirts like Hawaiian-print shirts.

ACKNOWLEDGMENTS We are grateful to Tom Beamish, Fred Block, Julie Collins, Joan Meyers, Sarah Ovink, Ofer Sharone, several anonymous reviewers, and the participants at the Workers and Class Relations Roundtable, 2004 American Sociological Association meetings for their very helpful feedback on this research. The research was supported by a generous grant from the Institute for Labor and Employment, University of California.

400

VICKI SMITH ET AL.

REFERENCES Acker, J. (1990). Hierarchies, bodies and jobs: A theory of gendered organizations. Gender & Society, 4, 139–158. Barley, S., & Kunda, G. (2004). Gurus, hired hands, and warm bodies. Princeton and Oxford: Princeton University Press. Benner, C. (2002). Work in the new economy: Flexible labor markets in Silicon Valley. London: Blackwell. Bielby, W., & Baron, J. (1986). Men and women at work: Sex segregation and statistical segregation. American Journal of Sociology, 91, 759–799. Blau, F., Ferber, M., & Winkler, A. (1998). The economics of women, men, and work. Upper Saddle River, NJ: Prentice-Hall. Bowles, S., & Gintis, H. (1976). Schooling in capitalist America: Educational reform and the contradictions of economic life. New York: Basic Books. Braddock, J., & McPartland, J. (1987). How minorities continue to be excluded from equal employment opportunities: Research on labor market and institutional barriers. Journal of Social Issues, 43, 5–39. Bradshaw, T. (1973). Jobseeking methods used by unemployed workers. Monthly Labor Review, February, 35–40. Britton, D. (2003). At work in the iron cage: The prison as gendered organization. New York: New York University Press. Cohn, S. (2000). Race, gender, and discrimination at work. Boulder, CO: Westview Press. Collins, S. (1996). Black corporate executives: The making and breaking of a Black middle class. Philadelphia: Temple University Press. Deliganis, S. (2003). Gender and racialized assumptions within a job training organization. Unpublished manuscript. Department of Sociology, University of California, Davis. Eckert, P. (1989). Jocks and burnouts: Social categories and identity in the high school. New York: Teacher’s College/Columbia University. EDD Factsheet: ‘‘Experience Unlimited’’ Job Search Associations. n.d. DE 8714 V Rev. 9. Elliott, J., & Sims, M. (2001). Ghettos and barrios: The impact of neighborhood poverty and race on job matching among Blacks and Latinos. Social Problems, 48, 341–361. Ferguson, A. A. (2001). Bad boys: Public schools in the making of Black masculinity. Ann Arbor: University of Michigan Press. Finlay, W., & Coverdill, J. (2002). Headhunters: Matchmaking in the labor market. Ithaca: Cornell University/ILR Press. Grahame, K. M. (1998). Asian women, job training and the social organization of immigrant labor markets. Qualitative Sociology, 21, 75–90. Granovetter, M. (1995). Getting a job: A study of contacts and careers. Chicago: University of Chicago Press. Grubb, W. N. (1996). Learning to work: The case for reintegrating job training and education. New York: Russell Sage Foundation Publications. Grubb, W. N., & Kalman, J. (1994). Relearning to earn: The role of remediation in vocational education and job training. American Journal of Education, 103, 54–93. Halaby, C. (1988). Action and information in the job mobility process: The search decision. American Sociological Review, 53, 9–25. Handler, J., & Hasenfeld, Y. (1997). We the poor people: Work, poverty, and welfare. New Haven: Yale University Press.

Getting Jobs and Building Careers

401

Hanson, S., & Pratt, G. (1995). Gender, work and space. New York: Routledge Press. Hasenfeld, Y. (2000). Organizational forms as moral practices: The case of welfare departments. Social Service Review, 74, 329–351. Hayes, S. (2003). Flat broke with children: Women in the age of welfare reform. Oxford: Oxford University Press. Holzer, H. (1988). Search method used by unemployed youth. Journal of Labor Economics, 6, 1–20. http://www.jobhunting.com. http://www.edd.ca.gov/ONE-STOP. Indergaard, M. (1999). Retrainers as labor market brokers: Constructing networks and narratives in the Detroit area. Social Problems, 46, 67–87. Jacobs, J. (Ed.) (1995). Gender inequality at work. Thousand Oaks, CA: Sage. Kennelly, I. (1999). ‘That single-mother element’: How white employers typify Black women. Gender and Society, 13, 168–192. Knoke, D., & Ishio, Y. (1998). The gender gap in company job training. Work and Occupations, 25, 141–167. Korteweg, A. (2003). Welfare reform and the subject of the working mother: ‘Get a job, then a Career’. Theory and Society, 32, 445–480. Lafer, G. (2002). The job training charade. Ithaca: Cornell University/ILR Press. MacLeod, J. (1987). Ain’t no makin’ it: Leveled aspirations in a low-income neighborhood. Boulder, CO: Westview Press. McGuire, G. (2002). Gender, race, and the shadow structure: A study of informal networks and inequality in a work organization. Gender and Society, 16, 303–322. Montgomery, J. (1992). Job search and network composition: Implications of the strength of weak ties hypothesis. American Sociological Review, 57, 586–596. Moss, P., & Tilly, C. (2001). Stories employers tell: Race, skill, and hiring in America. New York: Russell Sage Foundation Publications. Neckerman, K., & Kirschenman, J. (1991). Hiring strategies, racial bias, and inner-city workers. Social Problems, 38, 433–447. Negrey, C., Golin, S., Lee, S., & Gault, B. (2003). Job training for women leaving welfare: Assessing interest in nontraditional employment. The Sociology of Job Training, 12, 231–257. Neuwirth, E. (2004). Blurring corporate boundaries: Staffing agencies, human resource practices, and unions in the new employment relationship. Unpublished doctoral dissertation. University of California, Davis. Osterman, P. (1999). Securing prosperity: The American labor market, how it has changed and what to do about it. Princeton: Princeton University Press. Otis, E. (2001). The reach and limits of Asian panethnic identity: The dynamics of gender, race, and class in a community-based organization. Qualitative Sociology, 24, 349–379. Peck, J. (2001). Workfare states. New York: The Guilford Press. Piven, F. F., & Cloward, R. (1971). Regulating the poor: The functions of public welfare. New York: Vintage Books. Ports, M. (1993). Trends in job search methods, 1970–92. Monthly Labor Review, 116, 63–67. Ragin, C. (1987). The comparative method: Moving beyond qualitative and quantitative strategies. Berkeley: University of California Press. Riemer, F. J. (2001). Working at the margins: Moving off welfare in America. New York: State University New York Press.

402

VICKI SMITH ET AL.

Reskin, B., & Padavic, I. (2002). Women and men at work (2nd ed.). Thousand Oaks: Pine Forge Press. Ridgeway, C. (1997). Interaction and the conservation of gender inequality: Considering employment. American Sociological Review, 62, 218–235. Ridzi, F., & Banerjee, P. (2006). The spirit of outsourcing: Corporate and state regulations of labor under the H-1B visa and TANF policies in the U.S. Research in the Sociology of Work. Special Issue on Workplace Participation, 16. Oxford: Elsevier/JAI Press. Rogers, J. (2000). Temps: The many faces of the changing workplace. Ithaca: Cornell University/ ILR Press. Rosenbaum, J. (2001). Beyond college for all: Career paths for the forgotten half. New York: Russell Sage Foundation. Schwartzman, H. (1993). Ethnography in Organizations. Newbury Park: Sage Publications. Shih, J. (2002). Yeah, I could hire this one but I know it’s gonna be a problem: How race, nativity, and gender affect employers’ perceptions of the manageability of job seekers. Ethnic and Racial Studies, 25, 99–119. Smith, R., & Elliott, J. (2002). Does ethnic concentration inﬂuence employees’ access to authority? An examination of contemporary urban labor markets. Social Forces, 81, 255–279. Solomon, B. (2003). A ‘know it all’ with a ‘pet peeve’ meets ‘underdogs’ who ‘let her have it’: Producing low-waged women workers in a welfare-to-work training program. Journal of Contemporary Ethnography, 32, 693–727. Statistical abstracts of the United States. (2004–2005). U.S. Census Bureau. http://www.census.gov/statab/ Thomas, W. (2003). The meaning of race to employers: A dynamic qualitative perspective. The Sociological Quarterly, 44, 227–242. Waldinger, R., & Lichter, M. (2003). How the other half works: Immigration and the social order of labor. Berkeley: University of California Press. Weis, L. (1990). Working class without work: High school students in a de-industrializing economy. New York: Routledge. Williams, C. (1995). Still a man’s world: Men who do women’s work. Berkeley: University of California Press. Willis, P. (1977). Learning to labor: How working class kids get working class jobs. New York: Columbia University Press.

ABOUT THE AUTHORS Payal Banerjee is a doctoral candidate in the Department of Sociology at Syracuse University and is completing a dissertation on Indian immigrant IT workers in the U.S., which foregrounds the intersecting contexts of gender and race/ethnicity, U.S. immigration and visa policies, global economic restructuring, and transnational mobility of labor. Joshua L. Carreiro is a doctoral student in the Department of Sociology at the University of Massachusetts, Amherst. He is currently researching the unionization efforts of retail workers. His past research examined news media coverage of the United States labor movement. George Cheney (Ph.D., Purdue University, 1985) is professor of Communication at the University of Utah where he is Director of Peace and Conﬂict Studies. Also, he is adjunct professor of Management Communication at The University of Waikato, Hamilton, New Zealand. George is the author or co-author of ﬁve books and more than 75 articles and chapters. His teaching and research interests include power and identity at work, quality of worklife, workplace democracy, business and professional ethics, the marketization of society, and the rhetoric of war and peace. Heather Kohler Flynn is a Ph.D. candidate in the Department of Sociology at the University of California, Davis. She has conducted empirical research on close relationships, friendship, and self-esteem. Her dissertation examines friendship quality and evaluates the developmental and structural characteristics of friendship across the life course. Shlomo Getz (Ph.D. from the Hebrew University of Jerusalem) has been a researcher at the Institute for Research of the Kibbutz and the Cooperative Idea at the University of Haifa since 1986, and currently serves as its head. He has been conducting surveys of all kibbutzim annually since 1990. Reports on this research have appeared in Kibbutz Trends, the Journal of Rural Cooperation, and other publications in English and in Hebrew. 403

404

ABOUT THE AUTHORS

Michael J. Handel is associate professor of Sociology at Northeastern University. He studies the growth of earnings inequality in the United States and its relationship to the changing nature of work and organizations, particularly the diffusion of information technology and high-performance work practices. He is currently conducting a national survey of job skill requirements, workplace technology use, and involvement in participatory work practices. Robert Hanneman is professor of Sociology at the University of California, Riverside. His primary research interests are in methodology and mathematical modeling. In collaboration with Raymond Russell, he has done work on employment dynamics in other Israeli cooperatives, and in Russia during privatization. Jonathan Isler is a lecturer at the University of California, Davis, where he received his Ph.D. in Sociology in the summer of 2005. His dissertation research, drawing on the career analysis of web designers, highlights their unstable, insecure career trajectories and strategies they use to ﬁnd meaning in their work. Joan S.M. Meyers is a doctoral candidate at the University of California, Davis. Her current research compares two large worker-owned companies with similar histories but different organizational structures and workplace cultures. Her next project explores the construction of the diversity training industry. Esther B. Neuwirth is an applied researcher at Kaiser Permanente’s Care Management Institute where she develops methods for studying, implementing, and spreading innovative practices aimed at improving the quality of health care. She earned her Ph.D. in sociology from the University of California, Davis and her B.A. from Brandeis University. Nancy Plankey Videla is assistant professor in the Department of Sociology at Texas A&M University. She earned a Ph.D. from the University of Wisconsin, Madison. Through ethnographic work, her research examines how changes in the global economy affect ﬁrm strategies, and in turn, the experience of work on the shop ﬂoor. Jeremy Reynolds is assistant professor in the Department of Sociology at the University of Georgia. His research interests include work hours, work– family issues, nonstandard employment relationships, employee involvement, and how these phenomena are related to inequality.

About the Authors

405

Frank Ridzi is assistant professor of Sociology at Le Moyne College where he is Director of Urban and Regional Studies and Director of the Center for Regional and Urban Applied Research. He earned a Ph.D. and Master of Public Administration from Syracuse University’s Maxwell School of Citizenship and Public Affairs. Joyce Rothschild is professor of Sociology in the School of Public and International Affairs at Virginia Tech. Her previous research on grassroots collectives and cooperatives focused on the conditions that support or defeat collectivist-democratic forms of organizational governance; her recent research on whistleblowers addresses issues of organizational corruption and accountability. Her ongoing research returns to the question of how substantive values may shape innovative organizational and institutional forms. She can be contacted at [email protected]. Jeffrey S. Rothstein is a research fellow at the Douglas A. Fraser Center for Workplace Issues at Wayne State University. He received his Ph.D. in Sociology from the University of Wisconsin-Madison in 2005. His research explores the impact of globalization on autoworkers and compares the ways local actors shape global processes on the shop ﬂoor, based on their own experiences of economic change. Raymond Russell is professor of Sociology at the University of California, Riverside. His past research has included studies of diverse forms of employee participation in ownership and decision-making in the United States, Israel, and Russia. His previous research on the history of worker cooperatives in Israel is reported in the monograph, Utopia in Zion: The Israeli Experience with Worker Cooperatives, published in 1995 by State University of New York Press. Joel Schoening is a Ph.D. candidate at the University of Oregon. His other areas of interest include urban sociology, mass media, political economy, and education. He is an avid proponent of anything that may result in greater worker autonomy, more democratic decision making, and less reliance on fossil fuels. Vicki Smith is professor of Sociology at the University of California, Davis, and the author of Managing in the Corporate Interest: Control and Resistance in an American Bank (University of California Press, 1990) and Crossing the Great Divide: Worker Risk and Opportunity in the New Economy (Cornell/ILR Press, 2001).

406

ABOUT THE AUTHORS

Amy Tomchin recently received her Master’s Degree in Public and International Affairs at Virginia Tech, completing her thesis on gender equality and the division of labor at Twin Oaks. She is now in Tel Aviv working with a nonprofit organization on issues of social justice in Israel. She may be reached at [email protected]. Steven Peter Vallas chairs the Department of Sociology and Anthropology at George Mason University in Fairfax, Virginia. His research has focused on the transformation of work and managerial authority in several distinct industries, ranging from rural manufacturing plants in the American south to biotechnology laboratories in Silicon Valley. His articles have appeared in various sociological journals. He is currently completing an ethnography of industrial workers confronting social and organizational change.