Nicole Zimmermann Dynamics of Drivers of Organizational Change
GABLER RESEARCH
Nicole Zimmermann
Dynamics of Drivers of Organizational Change With a foreword by Prof. Dr. Dr. h. c. Peter Milling
RESEARCH
Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.d-nb.de.
Dissertation University of Mannheim, 2010
1st Edition 2011 All rights reserved © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011 Editorial Office: Stefanie Brich | Nicole Schweitzer Gabler Verlag is a brand of Springer Fachmedien. Springer Fachmedien is part of Springer Science+Business Media. www.gabler.de 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 copyright holder. Registered and/or industrial names, trade names, trade descriptions etc. cited in this publication are part of the law for trade-mark protection and may not be used free in any form or by any means even if this is not specifically marked. Cover design: KünkelLopka Medienentwicklung, Heidelberg Printed on acid-free paper Printed in the Netherlands ISBN 978-3-8349-3051-4
V
Foreword The design of organizational change has become a permanent managerial task in order for organizations to adapt to their environment or to exert influence on it. Accordingly, change in organizations can be reactive and proactive; either as a response to new demands which exert influence on the organization from outside, or as a deliberate measure to gain competitive advantage. Drivers of change and its management are the focus of Nicole Zimmermann’s dissertation. It explains change in organizational structures and processes by the interaction of internal and external stimuli. Here, it addresses the question of why some organizations are able to recognize new challenges early on and face them adequately, while others persist in their habituated structures and behavior patterns and are finally driven out of the market. Two major research questions are at the core of the investigation: (i) what are the drivers of organizational change, and (ii) is a history of successful change processes helpful or obstructive for further change? As such, problems are investigated as to why companies in a more specific sense and organizations in a more general sense have difficulties to change their strategies which have proven successful in the past whenever circumstances change. It is an important methodological characteristic of the stated problem and indeed a trivial fact that organizational change is a highly dynamic phenomenon. However, much research that addresses this topic employs methods that are adequate for static subjects of study. Often, this results in considerable differences in organizational theories and in their statements concerning change. The example of the New York Stock Exchange (NYSE) represents the basis of an extensive investigation of processes of inertia and change in an organization that has been successful for a very long period of time and then ran into difficulties that threatened its continued existence. The case study is used in order to collect data of a concrete example, to formulate hypotheses and to test them. Methodologically, the system dynamics approach is used because it is equally appropriate for theory formulation and the analysis of a case study. The author develops a formal model which can later be used for computer simulation. It is based on a multitude of also qualitative data that she derived predominantly from weblog entries and a number of interviews with employees of the New York Stock Exchange. She succeeds at developing hypotheses, at postulating causal relationships, at testing them by analyses, and at developing them further. The research question focuses on the transition from manual to electronic securities trading, which had been delayed for a long period of time, but which has since been implemented very quickly. Different periods, resistance, and impulses for change are analyzed, formed into hypotheses, and transferred to a model format. The comparison of simulated behavior over time with empirical data shows high consistency. The combination of exogenous pressure for the abandonment of manual
VI
Foreword
trading with endogenous actions by the management team creates a process that quickly transforms the habituated behaviors at the NYSE. The statements and insights which have been derived from the case study are generalized in order to derive the basic elements of a theory of organizational change. For this purpose, the original model that analyses a specific situation is transferred into a generic one that represents an entire class of applications. Elaborations on this generic model, its structure, and its behavioral patterns allow the author to gain insight into the processes of organizational change. Drivers of and resistance to change, external and internal factors, perception and actions of the management team are represented, analyzed, and consolidated towards the direction of a system of hypotheses—a theory—of organizational change. The analysis reveals that both developments in the environment and decisions by the managerial team determine the evolution of organizations. Additionally interests of stakeholders and the cognitive flexibility of the management team play a decisive role. In particular, the behavior of stakeholders can be used as a valuable source of information about the organizational environment.
Professor em. Dr. Dr. h. c. Peter Milling
VII
Acknowledgements Organizational phenomena have been a major topic of interest for me for a long time. In particular, the dynamic interplay of pressure for change and inertia has caught my attention. I have experienced its practical relevance and gained more and more interest in the theoretical side of this issue. Organizational change is situated at the interface between the system ‘organization’ and human beings. This dissertation concentrates on this interconnectedness of systemic drivers of and barriers to organizational change and on human interaction with this system. The dissertation grew out of my work during my time as an external PhD student and later as a research and teaching assistant at the Chair of Industrial Management (Industrieseminar) at the University of Mannheim. Additionally I profited from a research year at the University at Albany, State University of New York. The period as a doctoral student and assistant was interesting, challenging, and informative, which made it very valuable to me. Many people supported me during this phase and I would like to express my gratitude to them. I am greatly indebted to my advisor, Prof. Dr. Dr. h. c. Peter Milling, for giving me the possibility to work and develop under his guidance and for his continuing support and encouragement. I am also much obliged to researchers at the University at Albany. Here, I would like to express my gratitude to Professor George Richardson for many fruitful conversations and his support, as well as for co-refereeing this dissertation. Professor David McCaffrey aroused my enthusiasm for the case of the New York Stock Exchange, and Professor David Andersen helped to see the forest for the trees. Navid Ghaffarzadegan, Hyunjung Kim and Mohammad Mojtahedzadeh also made helpful suggestions. Additionally, I would like to thank Professor David Lane of the London School of Economics. Several years ago by his lecture he inspired me and aroused my enduring interest for system dynamics modeling and for systemic questions. My colleagues, Dr. Philipp Konecny, Dr. Christian Lehr, Switbert Miczka, Dr. Lena Oswald, Oliver Schmitzer, Prof. Dr. Jörn-Henrik Thun and Christian Weitert as well as our secretary Iris Scheuermann, have accompanied me during my years as a research and teaching assistant. They made it a pleasant time. In my personal environment, I have been able to rely on my close friends’ and family’s support, for which I am truly greateful. My sincere thanks go out to them all. I would also like to emphasize the assistance with final editing by Susanne Eschbach and, in particular, Dr. Lena Oswald. My parents have continuously encouraged me during my time as a doctoral student and on my entire path of life. I would like to thank them for their wonderful encouragement, for the motivation they gave me, and their loving support.
Nicole Zimmermann
IX
Contents Foreword.................................................................................................................... V Acknowledgements ................................................................................................ VII Contents ................................................................................................................... IX List of Figures ........................................................................................................ XIII List of Tables ........................................................................................................ XVII List of Abbreviations ............................................................................................ XIX A
The Challenge of Triggering Change in Organizations ................................... 1
B
Deterministic and Voluntaristic Theories of Organizational Change ............. 9 B.I
Incommensurable Drivers of Organizational Evolution ................................. 9 B.I.1
The Classical View of Adaptation to the Environment ..................... 11
B.I.1.a
Deterministic Adaptation and Impediments to the Adaptation Process ................................................................................... 11
B.I.1.b
Behavioral Elements of Adaptation ......................................... 15
B.I.2
Inertia and Routines as Determinants of Change ............................ 21
B.I.2.a
Inertia Leading to Environmental Selection ............................. 21
B.I.2.b
Routines Hindering and Driving Change ................................. 24
B.I.3
Transformation Triggered by Strategic Choice ................................ 28
B.II Reconciliation of Environmental Determinism and Managerial Choice ...... 33 B.II.1
Compatibility of Voluntaristic and Deterministic Views ..................... 33
B.II.2
Understanding Change by the Combination of Environmental and Managerial Forces ........................................................................... 37
B.II.3
A Reconciled Theory of Radical Change ......................................... 43
B.III Cognition and Attention as Drivers and Restraints of Alteration ................. 49 B.III.1 Perception of the Environment Through a Cognitive Managerial Lens ................................................................................................. 49 B.III.2 Selective Attention to Issues and Stakeholders ............................... 54 B.IV Need for a Dynamic Feedback View of Organizational Change ................. 61
X
Contents
C
The Phenomenon of Inertia and Change Exemplified in a Case Study of the New York Stock Exchange ........................................................................ 65 C.I
A Method Mix for Studying Change in Organizations and Especially at the New York Stock Exchange ................................................................... 67 C.I.1
Contribution of Studying Change with System Dynamics and Case Study Methodologies .............................................................. 69
C.I.2
The Process of Model Conceptualization Supported by Case Study Research Methods................................................................. 72
C.I.3
Confidence Through Model Analysis and Testing............................ 79
C.II Reacting to Automation in the U.S. Securities Market ................................ 80 C.II.1
Automation of Order Clearing, Routing, and Information Systems .. 80
C.II.2
Moving Towards an Electronic Market ............................................. 87
C.II.3
The New York Stock Exchange’s Recast of Trading Systems ......... 93
C.III Structure and Behavior of Forces for Retention and Change ................... 100 C.III.1 A Perspective of Adaptation to the External Environment ............. 101 C.III.2 An Endogenous Struggle of Culture and Resistance ..................... 111 C.III.3 Managerial Impact on Change ....................................................... 120 C.III.4 Full Model Behavior ....................................................................... 139 C.IV Analyses of Model Structure and Behavior .............................................. 145 C.IV.1 Confidence in Model Structure and Parameterization .................... 146 C.IV.2 Validation of Model Behavior and Sensitivity ................................. 149 C.V Implications of the New York Stock Exchange’s Recast of Trading Systems ................................................................................................... 160 D
Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change ................................................................................... 165 D.I
A Generic Model of Organizational Inertia and Change ........................... 165 D.I.1
Motivation for a Generic View ........................................................ 165
D.I.2
Generic Model Structure ................................................................ 168
D.II Structural-Behavioral Analysis and Causal Theory .................................. 182 D.II.1
Validation of the Generic Model ..................................................... 182
D.II.2
Effects of Reinforcing and Balancing Feedback on the Occurrence of Change ................................................................... 185
Contents
XI
D.III Possibilities of Managerial Intervention for Driving Change ..................... 192 D.III.1 Inertia and the Ambiguous Effects of the Responsiveness to Pressure ........................................................................................ 192 D.III.2 Effects of Increases in the Responsiveness of Attention ............... 198 D.IV Joint Management of Leverage Points ..................................................... 199 D.IV.1 Relationship Between the Responsiveness of Strategy to Pressure and Attention .................................................................. 200 D.IV.2 Policy Implications in Different Environments ................................ 205 D.V A Feedback Theory of Organizational Inertia, Change, and Attention...... 211 E
Realization of Change in Organizations ....................................................... 217
References ............................................................................................................. 227 Appendix................................................................................................................ 257
XIII
List of Figures Figure A-1: Innovation typology .................................................................................. 5 Figure B-1: Adaptive feedback .................................................................................. 12 Figure B-2: Impediments to adaptive feedback ......................................................... 15 Figure B-3: Limitations to adaptation......................................................................... 20 Figure B-4: A feedback view of strategic choice in a theory of organization ............. 31 Figure B-5: Sociological paradigms (Burrell and Morgan) ......................................... 34 Figure B-6: Convergence in the punctuated equilibrium model ................................. 44 Figure B-7: Determinants of stakeholder attention according to Mitchell, Agle, and Wood ............................................................................................... 59 Figure C-1: Trade participants and interactions in floor trade ................................... 66 Figure C-2: Percentage of leading stock exchanges enabling some or full e-trade (BOT) ..................................................................................................... 88 Figure C-3: NYSE market share in NYSE-listed securities (BOT) ............................. 90 Figure C-4: Adaptation process of stock exchanges ................................................. 93 Figure C-5: Spread.................................................................................................... 94 Figure C-6: Reference mode (BOT) .......................................................................... 98 Figure C-7: Diagramming conventions .................................................................... 100 Figure C-8: Sector diagram of the adaptation view ................................................. 101 Figure C-9: External influences in the remaining market (SFD) .............................. 103 Figure C-10: Diffusion of electronic trading in the securities market (BOT)............. 104 Figure C-11: External influences (CLD) .................................................................. 105 Figure C-12: Customer pressure for e-trade (SFD) ................................................. 106 Figure C-13: Limiting effect of e-trade on change ................................................... 106 Figure C-14: Customer pressure for e-trade (CLD) ................................................. 107 Figure C-15: Specialist participation from data (BOT) ............................................. 108 Figure C-16: Customer pressure for market quality (SFD) ...................................... 109 Figure C-17: Customer pressure for floor trade (CLD) ............................................ 110 Figure C-18: Adaptation view (BOT) ....................................................................... 111 Figure C-19: Sector diagram of the culture and resistance view ............................. 112 Figure C-20: Commissions and spread (SFD and BOT) ......................................... 113 Figure C-21: Resistance pressure from floor (SFD and effect) ............................... 114
XIV
List of Figures
Figure C-22: Resistance pressure from floor (SFD and CLD) ................................. 115 Figure C-23: Cultural pressure from floor (SFD and CLD) ...................................... 117 Figure C-24: Effect of institutional customers on power of floor firms ..................... 118 Figure C-25: Power of floor firms (SFD and CLD) ................................................... 119 Figure C-26: Culture and resistance (BOT) ............................................................. 120 Figure C-27: Sector diagram of the management view ........................................... 120 Figure C-28: Inertia and repetitive momentum (SFD) ............................................. 122 Figure C-29: Relationship between change and inertia........................................... 123 Figure C-30: Inertia and repetitive momentum (CLD) ............................................. 123 Figure C-31: Relationship between customer orientation and change (SFD) ......... 126 Figure C-32: Repetitive attention loop (SFD) .......................................................... 128 Figure C-33: Customer orientation (SFD) ............................................................... 129 Figure C-34: Perception bias from managerial attention (CLD) .............................. 129 Figure C-35: Spread and market share (SFD) ........................................................ 132 Figure C-36: Relationship between time at NBBO and market share ..................... 133 Figure C-37: Market share adjustment from speed and market quality (SFD) ........ 134 Figure C-38: Relationship between time to execution and market share ................ 134 Figure C-39: Relationship between market share and openness to change (SFD). 135 Figure C-40: Confidence effect of market share on openness to change ............... 136 Figure C-41: Market share (CLD) ............................................................................ 137 Figure C-42: Liquidity algorithms as a response to low market quality (SFD) ......... 138 Figure C-43: Adaptation, Culture and Resistance, and Mgmt (BOT) ...................... 139 Figure C-44: Underlying forces (BOT)..................................................................... 140 Figure C-45: Relationship between the time at the NBBO and market share ......... 141 Figure C-46: Comparison with reference mode (BOT) ............................................ 142 Figure C-47: Full NYSE model (CLD) ..................................................................... 143 Figure C-48: Importance of liquidity algorithms and market quality ......................... 144 Figure C-49: Sector diagram making explicit the model boundary .......................... 147 Figure C-50: Linear development of e-trade in market ............................................ 148 Figure C-51: Sensitivity for institutional customer pressure for e-trade ................... 151 Figure C-52: Sensitivity for resistance..................................................................... 152 Figure C-53: Sensitivity for resistance, power, and cohesiveness of floor firms ...... 154
List of Figures
XV
Figure C-54: Sensitivity for the fractional change in trading per pressure ............... 156 Figure C-55: Sensitivity for fractional change in customer orientation .................... 157 Figure C-56: Sensitivity for managerial parameters ................................................ 158 Figure C-57: Sensitivity for changes in stakeholder and management parameters 159 Figure D-1: Diffusion of B (SFD) ............................................................................. 169 Figure D-2: Adaptation pressure for strategy B (SFD) ............................................ 170 Figure D-3: Effect the relative quality B on the perceived inadequacy of strategy .. 171 Figure D-4: Adaptation pressure for strategy B (CLD) ............................................ 171 Figure D-5: Resistance pressure for strategy A (SFD) ............................................ 172 Figure D-6: Effect of adequacy of quality A on resistance pressure ........................ 173 Figure D-7: Resistance pressure for strategy A (CLD) ............................................ 173 Figure D-8: Inertia and repetitive momentum (SFD) ............................................... 174 Figure D-9: Effect of change on the decrease of inertia .......................................... 175 Figure D-10: Limitations to changes of strategy (SFD) ........................................... 176 Figure D-11: Attention to stakeholders (SFD) ......................................................... 177 Figure D-12: Repetitive momentum in the generic model (CLD) ............................. 178 Figure D-13: Limitations to changes in attention (SFD)........................................... 178 Figure D-14: Adaptation of Attention (CLD) ............................................................ 179 Figure D-15: Performance (SFD) ............................................................................ 180 Figure D-16: Relationship between performance and change (SFD) ...................... 181 Figure D-17: Performance (CLD) ............................................................................ 182 Figure D-18: Sensitivity for changes in stakeholder and management parameters 184 Figure D-19: Generic base run (BOT) ..................................................................... 186 Figure D-20: Phases of loop dominance (BOT and CLD) ....................................... 187 Figure D-21: Comparison of early and late radical adaptation (BOT) ..................... 189 Figure D-22: Relationship between difference in quality B and strategy inadequacy......................................................................................... 190 Figure D-23: Two environmental changes (BOT) .................................................... 191 Figure D-24: Effects of high inertia (BOT) ............................................................... 194 Figure D-25: Sensitivity to variations of inertia ........................................................ 195 Figure D-26: Sensitivity for change per perceived pressure.................................... 197 Figure D-27: Sensitivity for changes in adaptability of attention .............................. 199
XVI
List of Figures
Figure D-28: Delayed adaptation and early radical adaptation (BOT) ..................... 201 Figure D-29: Nonlinear relationship between fractional change and adaptation effectiveness ...................................................................................... 202 Figure D-30: Effect of attention to stakeholders on model behavior (BOT) ............. 204 Figure D-31: Adaptation of attention to stakeholder favoring B (CLD) .................... 205 Figure D-32: Reaction to quicker environmental change (BOT) .............................. 206 Figure D-33: Nonlinear relationship between the fractional change and adaptation effectiveness in the case of quick environmental change . 207 Figure D-34: Sensitivity to the variation in the permanent pressure for strategy A .. 208 Figure D-35: Inconsistent managerial setup (BOT and CLD) .................................. 210 Figure E-1: Full generic causal loop diagram (CLD) ............................................... 223
XVII
List of Tables Table B-1: Theoretical focus of the dissertation ........................................................ 10 Table B-2: Selected examples of multi-paradigm research in organizational change theory.......................................................................................... 40 Table C-1: Research design ..................................................................................... 69 Table C-2: Data sources for variable derivation ........................................................ 75 Table C-3: Exemplary derivation of variables, causal relationships, and behavior.... 78 Table C-4: Quantitative content analysis of the Exchanges Weblog ....................... 125 Table D-1: Strategies and their qualities ................................................................. 168
XIX
List of Abbreviations Amex
American Stock Exchange
BOT
behavior over time
CEO
chief executive officer
CLD
causal loop diagram
COO
chief operating officer
DEC
Digital Equipment Corporation
DMM
designated market maker
DOT
Designated Order Turnaround
i.e. (id est)
that is
NASDAQ
National Association of Securities Dealers Automated Quotations
NBBO
national best bid and offer
NGO
non-governmental organization
NYSE
New York Stock Exchange
PC
personal computer
SEC
U.S. Securities and Exchange Commission
SFD
stock and flow diagram
SIAC
Securities Industry Automation Cooperation
SOFFEX
Swiss Options and Financial Futures Exchange
vs.
versus
WHO
World Health Organization
1
A The Challenge of Triggering Change in Organizations Organizational environments are dynamic, and the speed of environmental change as well as its direction is often difficult to anticipate. In order to survive in the long term, organizations need to be able to successfully cope with a changing environment. They have to respond to emerging technologies or strategies even before they are fully established in the market.1 For this purpose, they need to be crafted in a way that makes them adaptable systems. Quick and appropriate organizational reactions are desired, and the ability to trigger organizational adaptations and transformations is important, but this poses great challenges to organizations. There is wide agreement that—in particular for established organizations—change is a difficult task. These establishments are often incapable of responding effectively to shifts in their external environment.2 “[… E]xisting organizations, especially the largest and most powerful, rarely change strategies and structures quickly enough to keep up with the demands of uncertain, changing environments.”3 This failure to understand the need for change or to initiate it threatens the performance and even the survival of organizations.4 Three examples will elaborate more clearly the difficulties organizations encounter when the demands from their environments change. The world's largest food corporation Nestlé, which originated from a breast-milk substitute producer, took more than a decade to adapt to its customers’ demands for ethical conduct. In the early 1970s, critics addressed Nestlé for its aggressive marketing practices for infant formula in the developing world, such as the failure to label products appropriately as well as the use of personnel dressed like nurses who advised mothers to use milk substitutes. Critics argued that, in combination with the prevailing conditions of water contamination, illiteracy, and poverty, infants often received a diluted and contaminated meal. They attributed the resulting malnutrition, diarrhea, and higher mortality rates to the use of breast-milk substitutes.5 Nestlé reacted by only minor modifications to its marketing practices.6 It even sued some of its 1
2
3
4
5
6
See Benner, Mary J.: Securities Analysts and Incumbent Response to Radical Technological Change: Evidence from Digital Photography and Internet Telephony, in: Organization Science, Vol. 21 (2010), No. 1, pp. 42 and 59. See ibid., p. 42; Tripsas, Mary and Giovanni Gavetti: Capabilities, Cognition, and Inertia: Evidence from Digital Imaging, in: Strategic Management Journal, Vol. 21 (2000), No. 10/11, p. 1147. See also Hannan, Michael T. and John Freeman: Structural Inertia and Organizational Change, in: American Sociological Review, Vol. 49 (1984), No. 2, p. 151; and Schaefer, Scott: Influence Costs, Structural Inertia, and Organizational Change, in: Journal of Economics and Management Strategy, Vol. 7 (1998), No. 2, pp. 237–238. Hannan, Michael T. and John Freeman: Organizational Ecology, Cambridge, MA [et al.] 1989, p. 12. See Hill, Charles W. L. and Frank T. Rothaermel: The Performance of Incumbent Firms in the Face of Radical Technological Innovation, in: The Academy of Management Review, Vol. 28 (2003), No. 2, p. 257. See Newton, Lisa H.: Truth is the Daughter of Time: The Real Story of the Nestle Case, in: Business and Society Review, Vol. 104 (1999), No. 4, p. 369. See Sethi, S. Prakash: Multinational Corporations and the Impact of Public Advocacy on Corporate Strategy: Nestlé and the Infant Formula Controversy, Boston 1994, p. 53.
N. Zimmermann, Dynamics of Drivers of Organizational Change, DOI 10.1007/978-3-8349-6811-1_1, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011
2
A The Challenge of Triggering Change in Organizations
critics for libel. The company continued to affirm that its marketing practices were in accordance with industry standards and law. Due to Nestlé’s reluctance to react to its critics, activists organized a boycott in 1977 that won support from the World Health Organization (WHO) and UNICEF. In 1980 the WHO drafted an industry-wide code of conduct. Discussions between the two parties began, and Nestlé slowly started to accept several of the boycotters’ demands.7 It had noticed that it needed to adapt its culture and management to changes in its socio-political environment.8 The boycott was called off when Nestlé accepted the code in 1984, seven years after the boycott started and more than a decade after its critics pointed to problems caused by its marketing practices. Although the Nestlé Company still has many critics, it has started to change the way it deals with its stakeholders. The next time when some European customers and Greenpeace aired their displeasure with a genetically modified candy bar produced by the organization, it reacted quickly by removing the bar from the European market. It also refrained from introducing further genetically engineered products. Although the organization was very reluctant to change its strategy in the infant formula case, with respect to candy bars it quickly aligned its strategic orientation with the public’s expectations. While Nestlé finally adapted to stakeholder demands, other organizations failed to undergo necessary change in their strategic orientation. Digital Equipment Corporation (DEC) used to be a leading computer manufacturer and a highly innovative company during the minicomputer generation in the 1970s and early 1980s. With the emergence of the much simpler personal computer (PC), DEC remained with its old strategy and continued to offer rather expensive and specialized all-in-one solutions including storage, processing, infrastructure, and applications. The emergence of the personal computer changed DEC’s market environment as well as the needs of a fast growing number of private customers. DEC’s management did not perceive this new group and thus failed to recognize the potential of the new product.9 The DEC culture kept the organizational focus on customers with a higher technological interest. Deeply embedded convictions about the prevalence of their technologically innovative products in the market led to a diminished perception of the significance of problems and of customer groups which developed in a radically altered environment.10 In the view of Kampas, the degree of environmental determinism, meaning the degree to which the market drives organizations, increased with the development of the PC as a dominant design in the computer market. DEC with its inwardly focused culture 7
8
9
10
See Post, James E.: Assessing the Nestlé Boycott: Corporate Accountability and Human Rights, in: California Management Review, Vol. 27 (1985), No. 2, p. 123; and Sethi: Multinational Corporations and the Impact of Public Advocacy on Corporate Strategy, 1994, pp. 226–227 and 273–282. See Sethi: Multinational Corporations and the Impact of Public Advocacy on Corporate Strategy, 1994, pp. 121 and 226–227. See Schein, Edgar H. with Peter DeLisi, Paul J. Kampas, and Michael M. Sonduck: DEC is Dead, Long Live DEC: The Lasting Legacy of Digital Equipment Corporation, San Francisco, CA 2003, p. 291. See ibid., pp. 251–252.
A The Challenge of Triggering Change in Organizations
3
could not keep up with external developments.11 After many of its businesses had been sold already, the company was bought up by Compaq in 1998, which was later acquired by Hewlett-Packard. DEC thus represents an example of a company that did not align its strategy with changing environmental conditions. The failure to perceive and orient to a new customer group despite existing environmental pressure was significant for the absence of change. A relatively young company represents the third example of an organization having problems to reorient. The digital photography company Linco, founded in 1996, soon became the technological market leader in digital photo memory, and faced problems of inertia as early as 2001. This means it was not sufficiently open to demands from its environment. Linco originated from an exploratory research project within another company, became a formal business unit in 1993 and independent in 1996. Shortly after its founding, the CEO and management established the firm’s identity as a digital photography company.12 Here, identity is defined as what insiders and outsiders perceive to be the core of an organization. It is closely associated with norms and shared beliefs about legitimate behavior and manifests itself in capabilities, routines, and procedures.13 In order to increase internal identification with digital photography, management even handed out digital cameras to employees. In 2000, by the time the organization went public, Linco was also regarded from the outside as a photography company. Resource allocation, capital investment, and human resource policies were guided by identity. Because the technological and industry context were viewed through the lens of digital photography, Linco’s perception of other opportunities was limited. Even when USB flash drives—which overlapped the functionality of Linco’s digital film—were launched by its competitors, Linco was reluctant to adapt because flash memory did not fit its identity.14 Customers adopted the competitors’ product. From 2001 on, a new CEO tried to change the organization’s identity, but both internally and externally the company did not converge on a new identity for several years. Linco faced significant difficulty adopting identity-challenging technologies that violate self-reinforcing core beliefs.15 It is surprising how quickly identity manifested once the organization embarked on the digital photography strategy. This then created a reinforcing cycle between cognitive elements of the organization and its actions, preventing it from deviating from the direction taken. It becomes evident from the examples that many established organizations have problems adapting to a new strategy demanded or pursued by other market partici11
12
13 14 15
See Kampas, Paul J.: The Impact of Changing Technology, in: Schein, Edgar H. (Ed.): DEC is Dead, Long Live DEC: The Lasting Legacy of Digital Equipment Corporation, San Francisco, CA 2003, pp. 138–139. Directly see Tripsas, Mary: Technology, Identity, and Inertia through the Lens of 'The Digital Photography Company', in: Organization Science, Vol. 20 (2009), No. 2, p. 444. See ibid., p. 450. See ibid., pp. 450–451. See ibid., p. 454.
4
A The Challenge of Triggering Change in Organizations
pants. Organization theorists argue that organizations are transformed all the time.16 For many of them, peripheral change or the fine-tuning of processes to further improve an old but also a relatively new strategy does not pose great challenges. DEC and Linco, for example, were very innovative in the minicomputer and digital photography business. However, they had great difficulties adopting a new strategy requiring a change in the organization’s core. In the view of Hannan and Freeman, the organizational core is comprised of stated goals, forms of authority, the core technology, as well as the marketing strategy, indicating which clients the organization orients to.17 Tushman and Romanelli hold a very similar notion of core strategy or strategic orientation. In contrast to a narrow understanding of strategic orientation as a product or business strategy, it not only tells what business the firm is in, but also how it competes in this area. Change of the strategic orientation can then be regarded as a shift in the business and product strategy as well as the structure, control systems, power relationships, or in the core values of an organization.18 The dissertation follows the broad definition of strategic orientation suggested by Hannan and Freeman as well as Tushman and Romanelli. This view of organizational change goes beyond the mere adoption of a technological innovation or disruptive technology with different technological features that customers value. Figure A-1 places the three examples discussed into a grid that measures the extent of technological innovation and the newness of the market served. The three market developments fall into very different categories, and it cannot be said that they all serve new markets or are radical technological innovations. Only digital photography can be subsumed under disruptive innovations.19 The concept described here additionally extends to non-technological changes in the market and includes all significant shifts in customer or stakeholder demands that impinge on an organization and require a change in the organization’s core. Many organizations seem to have difficulty with transformations that require such a form of rethinking. Organizations already fail to recognize the need and to make the decision to adopt a new strategy.20
16
17 18
19
20
See March, James G.: Footnotes to Organizational Change, in: Administrative Science Quarterly, Vol. 26 (1981), No. 4, p. 563. See also Kimberly, John R. and Robert H. Miles: The Organizational Life Cycle: Issues in the Creation, Transformation, and Decline of Organizations, in: The Jossey-Bass Social and Behavioral Science Series, San Francisco, CA 1980, p. 2. See Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 156. Directly see Tushman, Michael L. and Elaine Romanelli: Organizational Evolution: A Metamorphosis Model of Convergence and Reorientation, in: Cummings, Larry L. and Barry M. Staw (Ed.): Research in Organizational Behavior, Vol. 7, Greenwich, CT 1985, pp. 175–176. See Christensen, Clayton M: The Innovator's Dilemma: When New Technologies Cause Great Firms to Fail, Boston, MA 1997, p. xxv. Christensen defines a disruptive technology as an inferior technology which has different features that customers value. See Christensen: The Innovator's Dilemma, 1997, p. xv. See Mellahi, Kamel: The Dynamics of Boards of Directors in Failing Organizations, in: Long Range Planning, Vol. 38 (2005), No. 3, Special Issue: Organizational Failure, p. 270; Shattock, Michael: The academic profession in Britain: A study in the failure to adapt to change, in: Higher Education, Vol. 41 (2001), No. 1/2, pp. 45–46; and Sheppard, Jerry Paul and Shamsud
5
A The Challenge of Triggering Change in Organizations
new newness of market served
personal computers
ethical products
digital photography
existing incremental radical extent of technological innovation
Figure A-1: Innovation typology From the reluctance to change among organizations the problem arises of what enhances their adaptability. The recognition of drivers of change allows for a more appropriate strategic orientation of organizations that goes beyond the mere understanding of change obstacles. Theoretical positions concerting drivers of change are far from concordant with each other. Opinions on what triggers development in organizations are indeed diverse, as the following statement highlights. “Although development can be considered as the natural model of organizational behavior, there is no theoretical consensus as to which forces generate development or hold it back. There is a debate as to how far one should seek to interpret the development of organizations as the product of external forces rooted in the social and economic system, as opposed to interpreting it as the product of idiosyncratic purposive behavior on the part of those within organizations who decide on strategies. It is a debate over the significance of environmental forces as opposed to managerial action, over organizational dependence as opposed to autonomy. This has been a fundamental issue both in the economic theory of the firm and in organization theory […].”21 As the statement reveals, the question of what determines change is central in organization theory. Different theories propose contrasting drivers and also inhibitors of change. The disagreement among organization theories shows that there is need for a deeper exploration. This investigation in this dissertation will therefore be guided by the research question of what the drivers of change are. The query will be discussed also in relation to possible inhibitors of change that proved to be important in the short examples presented. Reasons for the difficulty to make the decision to change will be analyzed. As the DEC case showed, managerial perception of old and emerging stakeholder groups turned out to be essential. Also the Linco study revealed the significance of a cognitive lens in relation to self-reinforcing beliefs and routines. The
21
D. Chowdhury: Riding the Wrong Wave: Organizational Failure as a Failed Turnaround, in: Long Range Planning, Vol. 38 (2005), No. 3, Special Issue: Organizational Failure, p. 250. Child, John and Alfred Kieser: Development of organizations over time, in: Starbuck, William H. and Paul C. Nystorm (Ed.): Handbook of organizational design, Vol. 1. Adapting organizations to their environments, Oxford 1981, p. 28.
6
A The Challenge of Triggering Change in Organizations
Nestlé example demonstrated differences in the organization’s reaction to a first and second customer demand of a similar kind. These aspects deserve consideration in their relation to drivers of change. Overall, this dissertation addresses the questions of, first, what drives organizational change, and second, whether prior change serves as a driver for future transformations. In this way, it addresses the problem of why organizations have difficulties making the decision to alter their core strategy. A better understanding of what prevents and drives change in organizations is expected to lead to better organizational strategies and enhance organizations’ adaptability. Previous research on organizational change has often investigated drivers of change from a static perspective. Many approaches illuminate single aspects and disregard interdependencies and their evolution over time. This investigation employs a long-term focus of behavior and structure of organizations. Change is analyzed by system dynamics modeling and simulation as this approach accounts for the system’s complexity as well as the interaction of hindering and driving forces for transformation. It is also able to incorporate cognitive aspects and show how the system’s history is captured in accumulations, which may create path-dependent behavior.22 The relation of structure to behavior helps understand organizations’ evolution and elicit triggers of change. For the reason of educing determinants of organizational evolution, in chapter B, organizational theories are discussed in their relation to drivers of change. First, single-driver theories are addressed. Second, there may be multiple drivers and inhibitors. Therefore, it is clarified whether it is legitimate from a philosophical point of view to combine drivers and theories of change. After this discussion on legitimacy, several examples of combined or multiple-paradigm theories follow that include several of the elements of the single-paradigm theories. As there still remains a need for a more dynamic consideration of drivers of change, a case study helps elicit causal relationships of change and its absence. Chapter C deals with a case study of the New York Stock Exchange’s move to electronic trading and its methodical analysis. In the beginning, the system dynamics method is described together with its potential to reveal complex causal relationships, link structure and behavior, and provide understanding for what determines change. In particular the use of system dynamics for case study research in combination with qualitative methods is elaborated. Subsequently, the reader is introduced to important developments in the securities market which led to the emergence of electronic trading. The NYSE’s initial lack of response and later radical adaptation to electronic trading are analyzed in more detail by system dynamics methods. This part of the inquiry particularly focuses on different drivers of change. As the NYSE-specific analysis provides many examples of more generic relevance, a further, generic model is developed in chapter D. It represents a more ge22
For an overview of the principles system dynamics modeling see Forrester, Jay W.: Industrial Dynamics, Cambridge, MA 1961, pp. 67–72.
A The Challenge of Triggering Change in Organizations
7
neric system dynamics theory and investigates the interconnectedness of drivers of change and possibilities for managerial intervention for creating different patterns of change. Chapter E summarizes the findings and consolidates the implications.
9
B Deterministic and Voluntaristic Theories of Organizational Change There are many schools of thought concerned with organizations and their evolution. Innovation research, for example, focuses on process and product innovations in business.23 Institutional theories and industrial organization deal with market exchanges.24 Strategic management is concerned with the competitiveness of companies.25 Many theories pay attention to single aspects in organizations such as power, culture, or sensemaking.26 Additionally, there are practical approaches such as change management and organization development, aiming at successful change implementation. This dissertation, however, concentrates on organization theory and its sub-category of organizational change. The reason is their focus on behavioral elements of decision-making. Structure and process are both important in organization theory, and some theories even take into account cognitive elements of decisionmakers.27 The organization theory approach also incorporates many of the aspects of strategic management, the dynamic capabilities view, and innovation research. This, together with decision-behavioral elements, makes it particularly suited for the analysis of change in organizations.
B.I
Incommensurable Drivers of Organizational Evolution
The organization theory field combines many smaller organizational theories. These theories have various and often incommensurable opinions regarding the origin of change. Despite these differences, they hold similar views of the essence of organizations and consider them as socially constructed systems of human activity. Organi-
23
24
25
26
27
See Milling, Peter M.: Modeling innovation processes for decision support and management simulation, in: System Dynamics Review, Vol. 12 (1996), No. 3, pp. 215–218; Milling, Peter M.: Understanding and managing innovation processes, in: System Dynamics Review, Vol. 18 (2002), No. 1, pp. 75–80; and Rogers, Everett M.: Diffusion of Innovations, 5. Ed., New York, NY [et al.] 2003, pp. 136–157. See Nelson, Richard R. and Sidney G. Winter: An evolutionary theory of economic change, Cambridge, MA [et al.] 1982, pp. 3–4; and North, Douglass C.: Institutions, institutional change and economic performance, Cambridge, MA [et al.] 1990, pp. 93–96 and 108–109. See Chandler, Alfred D. Jr.: Strategy and Structure: Chapters in the History of the Industrial Enterprise, Cambridge, MA 1962, p. 11; and Porter, Michael E.: On Competition, updated and expanded Ed., Boston, MA 2008, pp. 3–4. See Pfeffer, Jeffrey: Managing With Power: politics and influence in organizations, Boston, Mass. 1992, pp. 8–13; Schein, Edgar H.: Organizational Culture and Leadership, 3. Ed., San Francisco, CA 2004, pp. 1–8; and Weick, Karl E.: Sensemaking in Organizations:, Thousand Oaks [et al.] 1995, pp. 4–16. See Scott, W. Richard: Reflections of a Half-Century of Organizational Sociology, in: Annual Review of Sociology, Vol. 30 (2004), No. 1, pp. 3–4 and 7–8.
N. Zimmermann, Dynamics of Drivers of Organizational Change, DOI 10.1007/978-3-8349-6811-1_2, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011
10
B Deterministic and Voluntaristic Theories of Organizational Change
zations are goal-directed and maintain boundaries that reflect their goals.28 Hence they are social systems that have persisting elements and a purpose. Following the organizational literature in general, the view of organizations employed in this piece of work mainly focuses on business organizations, but also extends to non-profit organizations, as they obey similar processes and share properties important in the organization theory field. Phenomena discussed in organization theory—such as decision-making in organizations, organizational behavior, and particularly organizational change—also apply to administrative, charitable, or health care organizations. The characteristics of organizational theories employed in this dissertation are summarized in Table B-1. Characteristics of organization theory employed Similar definitions of organizations, but different views on drivers of change
Focus on behavior and decision-making in combination with organizational elements
Focus on organizational change
Primary focus on organizational level of investigation
Table B-1: Theoretical focus of the dissertation Research has approached organizations and their evolution at different levels of investigation: the individual, group, organization, and even the industry. This dissertation concentrates on decisions and behavior at the group and organizational level. Accordingly, it relies on an aggregate view of individuals although they may be influenced by the psychological micro-perspective. Since the focus is on the organization, reasons for changes in the market will also be outside the boundary of the subject matter. They may be described, but will be included only in so far as they impinge on decision-makers in the organization as part of their decision environment. Much attention has been paid to the relationship between the environment or market and the organization. Environmental drivers of change have been found in changes in demand, technological innovations, and institutional conditions.29 The perspective according to which organizations adapt to the environment is one of the most prominent views of organizational theory and change. 28
29
See Aldrich, Howard E.: Organizations and Environments, Stanford, CA 2008, pp. 4–6; and Aldrich, Howard E. and Martin Ruef: Organizations Evolving, 2. Ed., London [et al.] 2006, p. 4. See also Barnard, Chester I.: The Functions of the Executive, Cambridge, MA 1938, p. 65. See Romanelli, Elaine and Michael L. Tushman: Organizational Transformation as Punctuated Equilibrium: An Empirical Test, in: Academy of Management Journal, Vol. 37 (1994), No. 5, p. 1145. See also Abernathy, William J. and James M. Utterback: Patterns of Industrial Innovation, in: Technology Review, Vol. 80 (1978), No. 7, pp. 41–46; Haveman, Heather A., Michael V. Russo and Alan D. Meyer: Organizational Environments in Flux: The Impact of Regulatory Punctuations on Organizational Domains, CEO Succession, and Performance, in: Organization Science, Vol. 12 (2001), No. 3, p. 253–254 and 269; and Meyer, Alan D., Geoffrey R. Brooks and James B. Goes: Environmental Jolts and Industry Revolutions: Organizational Responses to Discontinuous Change, in: Strategic Management Journal, Vol. 11 (1990), No. -, Special Issue: Corporate Entrepreneurship, pp. 94–97.
B.I Incommensurable Drivers of Organizational Evolution
B.I.1 B.I.1.a
11
The Classical View of Adaptation to the Environment Deterministic Adaptation and Impediments to the Adaptation Process
Well into the second half of the 20th century, organization theory was dominated by views of rational adaptation. Many theories fall into the rational adaptation category, ranging from scientific management to industrial organization economics and resource dependence theories.30 In their core these ‘modern’ theories are based on the assumption of human rationality and utility maximization, and they rely on humans as decision-makers according to the concept of the homo economicus.31 Based on this cognitive and behavioral assumption of humans, entire organizations are assumed to rationally adapt to the environment. Hannan and Freeman describe rational adaptation as “designed changes in strategy and structure of individual organizations in response to environmental changes, threats, and opportunities.”32 Since the environment is assumed to set the point of time and the direction of adaptation, many authors call these theories deterministic.33 Management choice plays a minor role in these theories that were originally rather focused on content; a management team is only assumed to direct the organization to match the environmental demands and the organization to be able to change in this adaptive direction that management sets. In this way, classical organization theory is thus also based on the premise of the malleability of the entire organization.
30
31
32 33
A more detailed list of theories includes scientific management, Fordism, Weber’s bureaucracy, industrial organization economics, contingency theories, and resource dependence theories. See Whittington, Richard: Environmental Structure and Theories of Strategic Choice, in: Journal of Management Studies, Vol. 25 (1988), No. 6, pp. 524–526. A concept of man is an assumption or abstraction about the nature of human beings. The deterministic concept of the homo economicus can be traced back to classical 19th century economists, particularly to Wilfried Pareto and John Stuart Mill. See Persky, Joseph: Retrospectives: The Ethology of Homo Economicus, in: The Journal of Economic Perspectives, Vol. 9 (1995), No. 2, p. 222. See Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 150. See Child, John: Organizational structure, environment and performance: The role of strategic choice, in: Sociology, Vol. 6 (1972), pp. 8 and 10; Mellahi, Kamel and Adrian Wilkinson: Organizational failure: a critique of recent research and a proposed integrative framework, in: International Journal of Management Reviews, Vol. 5/6 (2004), No. 1, pp. 23 and 27; Child, John: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment: Retrospect and Prospect, in: Organization Studies, Vol. 18 (1997), No. 1, p. 45; and de Rond, Mark and Raymond-Alain Thietart: Choice, chance, and inevitability in strategy, in: Strategic Management Journal, Vol. 28 (2007), No. 5, pp. 539 and 546.
12
B Deterministic and Voluntaristic Theories of Organizational Change
The adaptation process to environmental drivers of change can be described by a feedback loop. As Figure B-1 reveals, a gap between the target organization’s and the environment’s strategic orientation leads to adaptation pressure, consequently aligning the target’s strategic orientation with environmental demands and reducing the gap. This constitutes a balancing, also called negative, goal-seeking or adaptive feedback relationship—here the balancing loop Deterministic Adaptation. ENVIRONMENT'S STRATEGIC ORIENTATION
Strategic Orientation + - + (B) Deterministic gap Adaptation pressure to adapt + strategic orientation
Note on the Nomenclature of Feedback Figures Loop Polarity: The letters (B) or (R) in the center of a feedback loop indicate the polarity of the entire feedback cycle. A balancing loop (B) is a self-correcting loop that seeks to achieve an equilibrium stage and to remain at this stage. The antipode of a balancing feedback loop is a reinforcing feedback cycle (R). It enhances what happens in the system. Arrow Polarity: The signs next to the arrows specify the polarity of the respective causal relationship. If x changes, a plus indicates a change of y in the same direction, a minus indicates a change of y in the opposite direction. The mathematical representation is shown in the following: +
x՜y ֜ -
x ՜ y ֜
y x y x
t
>0ǡ and for accumulations Y= t=0ሺx+…ሻds+Yt0 ; t
<0ǡ and for accumulations Y= t=0൫-x+...൯ds+Yt0
Stocks/Levels: The box around a variable demonstrates that the respective variable is a stock (level), meaning it accumulates over time and has an observable value at each point of time. Figure B-1: Adaptive feedback Faced with rapid transformations in their environment, e.g. from technological developments, many organizations cannot meet the requirements their environments make because of organizational impediments to change.34 Practitioners and researchers soon realized that adaptation is difficult in organizational reality. They first 34
See Tripsas and Gavetti: Capabilities, Cognition, and Inertia, 2000, p. 1147.
B.I Incommensurable Drivers of Organizational Evolution
13
concentrated on problems in the implementation of adaptive change. Later they also recognized that organizations even fail to identify the need to adapt. With regards to implementation, since adaptations did not always proceed as smoothly as management teams desired, the management of human behavior became an important theme within the adaptation paradigm. Researchers who studied the diffusion of technological innovations noticed a central issue which could become a problem: the adoption of innovations.35 They noticed the importance of the consideration of the social element among human beings for change.36 Resistance as well as cultural socialization of organizational members can be effective impediments to the organization’s adaptation to its environment. Particularly the research into resistance as an impediment to the rational adaptation process began with the behavioral theories and with the human relations movement. Resistance behavior can form for different reasons: rational, self-serving, and emotional.37 Rational opposition forms e.g. because in the opinion of the employees, the costs for implementing change outweigh the expected benefits.38 Resistance may also be framed as intentional resistance based on the self-interest of the individual. This particularly includes opposition to change due to the loss of power or other benefits.39 Emotional resistance is triggered by the feeling of not being able to cope with new demands; in particular when they challenge identity, they create fear and defensive mechanisms.40 Without knowledge of the sources and consequences of resistance, neither an adequate reaction to it nor a viable planning of change can take 35
36
37
38
39
40
See Parker, Charles A.: The Literature on Planned Organizational Change: A Review and Analysis, in: Higher Education, Vol. 9 (1980), No. 4, pp. 431–432. See Lewin, Kurt: Field Theory in Social Science: Selected Theoretical Papers, in: Cartwright, Dorwin, New York 1951, Coch, Lester and John R. P. French, Jr.: Overcoming Resistance to Change, in: Human Relations, Vol. 1 (1948), No. 4; Lawrence, Paul R.: How to Deal with Resistance to Change, in: Harvard Business Review, Vol. 32 (1954), No. 3; and Zander, Alvin: Resistance to Change—Its Analysis and Prevention, in: Advanced Management, Vol. 4 (1950), No. 5. Research regularly focused on one or several of the three aspects of cognition, emotion, and behavior. See Piderit, Sandy Kristin: Rethinking Resistance and Recognizing Ambivalence: A Multidimensional View of Attitudes toward an Organizational Change, in: The Academy of Management Review, Vol. 25 (2000), No. 4, pp. 785–786; ; and Oreg, Shaul: Personality, context, and resistance to organizational change, in: European Journal of Work & Organizational Psychology, Vol. 15 (2006), No. 1, pp. 90–91. See Oreg: Personality, context, and resistance to organizational change, 2006, p. 79; and Beer, Michael: Organization Change and Development: A Systems View, Santa Monica, CA 1980, p. 103. See Judson, Arnold S.: Changing Behavior in Organizations: Minimizing Resistance to Change, rev. Ed., Cambridge, Mass. 1991, p. 34. See Beer: Organization Change and Development, 1980, p. 102; Conner, Daryl R.: Managing change: A business imperative, in: Business Quarterly, Vol. 58 (1993), No. 1, p. 91; Diamond, Michael A.: Resistance to Change: A Psychoanalytic Critique of Argyris and Schön's Contributions to Organization Theory and Intervention, in: Journal of Management Studies, Vol. 23 (1986), No. 5, p. 559; Doppler, Klaus and Christoph Lauterburg: Managing Corporate Change, Berlin [et al.] 2001, p. 48; and Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 156.
14
B Deterministic and Voluntaristic Theories of Organizational Change
place. Researchers noticed that resistance is a social rather than technical phenomenon and tried to manage and overcome resistance.41 Lewin’s concept of unfreezing—movement—refreezing is part of this approach. It incorporates a step of dismantling inertia and bypassing defensive mechanisms, a transitional stage, and a freeze step that ensures sustainability at a new equilibrium.42 The concept combines a mechanical approach with social, group, and psychological aspects and is thus part of the human relations movement that gave importance to social aspects. It aims at overcoming the lacking sustainability of change. What is important is the fact that at no point they question the idea of the organization’s adaptation to its environment. This perspective assumes that resistance will necessarily emerge.43 A second perspective has a rather contingent view. It regards resistance not as a general reaction to change, but a response to its side effects. It further eliminates the assumption that resistance is inherently bad and needs to be overcome. It can be regarded as a feedback reaction to change objectives. Recent research suggests that resistance be regarded as a multi-faceted phenomenon which includes positive utility of resistance. On the one hand resistance pushes an organization to greater stability; on the other hand, resistance has the ability to draw attention to hidden shortcomings.44 Disagreement seems to trigger further investigation and learning, discussion and improvisation. It may also help to break up old routines.45 Resistance can arise from dynamics within the organization.46 It is thus an aspect of the process of change rather than of its content.47 It develops in the process of changing and is an aspect of the dynamics of how change unfolds. Cultural elements constitute a further impediment to the organizational adaptation process. If employees and organizations are deeply embedded in a specific organizational culture, it is difficult to undergo fundamental changes which violate this culture because these changes then also violate people’s values and beliefs. Since they attack the organization’s core, changes to identity are particularly difficult.48 They are 41
42 43
44
45 46
47
48
See Coch and French: Overcoming Resistance to Change, 1948, pp. 517–520 and 529–531; Lawrence: How to Deal with Resistance to Change, 1954, pp. 49–52 and 56–57. See Lewin: Field Theory in Social Science, 1951, pp. 228–229. See Dent, Eric B. and Susan Galloway Goldberg: Challenging "Resistance to Change", in: Journal of Applied Behavioral Science, Vol. 35 (1999), No. 1, p. 29. See Ford, Jeffrey D., Laurie W. Ford and Angelo D'Amelio: Resistance to Change: The Rest of the Story, in: Academy of Management Review, Vol. 33 (2008), No. 2, p. 368; and Waddell, Dianne and Amrik S. Sohal: Resistance: a constructive tool for change management, in: Management Decision, Vol. 36 (1998), No. 7/8, pp. 543–545. See Piderit: Rethinking Resistance and Recognizing Ambivalence, 2000, p. 790. See Macrì, Diego Maria, Maria Rita Tagliaventi and Fabiola Bertolotti: A grounded theory for resistance to change in a small organization, in: Journal of Organizational Change Management, Vol. 15 (2002), No. 3, pp. 303–306. See Amburgey, Terry L., Dawn Kelly and William P. Barnett: Resetting the Clock: The Dynamics of Organizational Change and Failure, in: Administrative Science Quarterly, Vol. 38 (1993), No. 1, p. 51. See Doppler and Lauterburg: Managing Corporate Change, 2001, p. 48; and Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 156.
B.I Incommensurable Drivers of Organizational Evolution
15
likely to result in defensive mechanisms among organizational members. Organizational culture thus represents an impediment to change in the way that it is likely to be a trigger of resistance.
impaired by resistance
Strategic Orientation + - + (B) Deterministic gap Adaptation
ENVIRONMENT'S STRATEGIC ORIENTATION
pressure to adapt + strategic orientation
Figure B-2: Impediments to adaptive feedback Research focusing on impediments to change does not deviate from the general adaptive view of environment-induced change. Cultural identity and resistance only weaken the adaptation cycle shown in Figure B-2. They may represent impediments to the adaptation itself, meaning that the implementation of changes becomes thwarted and delayed.
B.I.1.b
Behavioral Elements of Adaptation
Impediments to the adaptation process discussed so far concerned implementation issues, but adaptation may also fail because the need for it remains unrecognized. The behavioral theory of organizations developed a decision-centered view of adaptation that helps explain why organizations have difficulties changing. It focuses on decision-makers in organizations and regards change as a theory of how an organization responds to its environment. While focusing on adaptation, it does not deny elements of resistance. “Organizations change. Although they often appear resistant to change, they are frequently transformed into forms remarkably different from the original.”49 According to March, organizations change continually, “routinely, easily, and responsively.”50 The strand of the behavioral theory that this dissertation focuses on is in the tradition of the Carnegie School of Organizational Research and of Cyert and March’s Behavioral Theory of the Firm.51 It will be called the behavioral theory or
49 50 51
March: Footnotes to Organizational Change, 1981, p. 563. ibid., p. 563. The research tradition developed as a reaction to the rational scientific and the motives and emotion-based human relations theory. Today it provides the foundation for transaction cost economics, evolutionary theories, and the dynamic capabilities view. See Argote, Linda and Henrich R. Greve: A Behavioral Theory of the Firm—40 Years and Counting: Introduction and Impact, in: Organization Science, Vol. 18 (2007), No. 3, p. 338; Augier, Mie and David J Teece: Dynamic Capabilities and the Role of Managers in Business Strategy and Economic Performance, in: Organization Science, Vol. 20 (2009), No. 2, pp. 412–413; and March, James G. and Herbert A. Simon: Organizations, New York [et al.] 1958, p. 210. For foundational work see Barnard: The Functions of the Executive, 1938; Cyert, Richard M. and James G. March: A Behavioral Theory of the Firm, Englewood Cliffs, NJ 1963; March and
16
B Deterministic and Voluntaristic Theories of Organizational Change
the behavioral theory of organizations. This stream of research investigates organizational goals, expectations, the choice of responses to a problem, and organizational control.52 Within these areas the theory particularly addresses decisions.53 Its central element is the process of organizational decision-making as it concerns the question of how organizations adapt to their environments in order to sustain an adequate performance. Bounded rationality is the theory’s most fundamental assumption underlying the adaptation to a complex and changing environment. The behavioral theory opposes the presumption of rationality which the authors reduce to profit maximization and perfect knowledge.54 Deviating from the rational homo economicus, Simon suggests a bounded rational concept of man as someone who perceives only a simplified model of the real world.55 Human decisions are influenced by limited information available as well as by biased information processing. Values, limited skills, and personal motives different from organizational goals may reduce decision efficiency.56 As a consequence, decision-makers have incomplete knowledge, difficulties in anticipating the future and in considering behavior possibilities. As an example of limited knowledge, humans can only base their decisions on a small, closed system of variables isolated from the rest of the world. They consider factors closely connected to the problem in cause and time; delays and far-reaching effects are neglected.57 Consequently, due to its limited cognitive capacity, a human mind is not able to grasp the consequences of behaviors in their entirety. The concept of bounded rationality describes the “limits in the decision maker's mental capacity compared with the complexity of the decision environment.”58 The importance of bounded rationality increases with environmental complexity. Simon as well as Dörner define complexity by the number of parts in a system and their interaction. These systems are complex because it is non-trivial to infer the properties of the
52 53
54 55
56
57 58
Simon: Organizations, 1958; and Simon, Herbert A.: Administrative Behavior: A Study of Decision-making Processes in Administrative Organizations, New York 1947. See Cyert and March: A Behavioral Theory of the Firm, 1963, p. 21. See Gavetti, Giovanni, Daniel A. Levinthal and William Ocasio: Neo-Carnegie: The Carnegie School's Past, Present, and Reconstructing for the Future, in: Organization Science, Vol. 18 (2007), No. 3, pp. 525–526. Other elements of the theory, such as individuals’ decision to participate in organizations will not be elaborated here. See Cyert and March: A Behavioral Theory of the Firm, 1963, p. 8. See Simon, Herbert A.: Administrative Behavior: A Study of Decision-making Processes in Administrative Organizations, 3. Ed., New York 1976, p. xxv. Simon even expresses that “[t]he social sciences suffer from a case of acute schizophrenia in their treatment of rationality.” Simon: Administrative Behavior, 1976, p. xxiii. See March and Simon: Organizations, 1958, p. 136; and Simon: Administrative Behavior, 1976, pp. 40 and 241. See Simon: Administrative Behavior, 1976, p. 82. Dequech, David: Bounded Rationality, Institutions, and Uncertainty, in: Journal of Economic Issues, Vol. 35 (2001), No. 4, p. 913.
B.I Incommensurable Drivers of Organizational Evolution
17
whole from the properties of the parts.59 Milling emphasizes that, apart from the number of parts and their connections in a system, their type of entanglement affects the system’s complexity. Non-linear relationships and delays often complicate the interdependence of parts in a system.60 Humans have serious difficulties inferring behavior from such a structure of a system. Difficulties arise in the estimation of exponential developments, side effects, and from isolated cause and effect thinking.61 The recognition of accumulations and feedback and their effect on the systems’ behavior is also subject to high error probability.62 As the behavioral theory focuses on organizational decision-making, the decision capabilities of the dominant coalition in a complex environment are of particular importance. The dominant coalition is the central deciding body in an organization, and it is composed of a coalition of members with different goals. Organizational decisionmaking is thus a group process. According to Argote and Greve, the behavioral theory opens the black box of processes of how organizations internally work.63 This process view of organizations is opposed to the many studies that either focus on the structure of organizations or that are content-oriented and neglect the importance of organizational processes and change. A central element of the behavioral decisionmaking is the analysis of policies, or decision strategies or rules, as Cyert and March call them.64 Yet, the behavioral theory does not only investigate organizations from within, the relationship between organizational processes and the environment also receives attention.65 The organizational environment and particularly the alignment with organizational aspiration levels (goals) serves as the driver of a firm’s evolution. If performance falls below an organization’s aspiration level, it is more likely to undergo change.66 This assertion is based on the assumption that performance below aspirations is problem59
60
61 62
63
64 65
66
See Dörner, Dietrich: Die Logik des Misslingens: strategisches Denken in komplexen Situationen, 5. extended Ed., Reinbek bei Hamburg 2003, pp. 60–61; and Simon, Herbert A.: The Sciences of the Artificial, 3. Ed., Cambridge, MA [et al.] 1999, pp. 183–184. See Milling, Peter: Systemtheoretische Grundlagen zur Planung der Unternehmenspolitik, Berlin 1981, p. 91; Milling, Peter: Kybernetische Überlegungen beim Entscheiden in komplexen Situationen, in: Milling, Peter (Ed.): Entscheiden in komplexen Situationen, Berlin 2002, p. 12. See also Sterman, John D.: Learning in and about complex systems, in: System Dynamics Review, Vol. 10 (1994), No. 2/3, pp. 297–299. See Dörner: Die Logik des Misslingens, 2003, pp. 32 and 57. See Sterman, John D.: Modeling Managerial Behavior: Misperceptions of Feedback in a Dynamic Decision Making Experiment, in: Management Science, Vol. 35 (1989), No. 3, pp. 334–337. See Argote and Greve: A Behavioral Theory of the Firm—40 Years and Counting, 2007, p. 344. See Cyert and March: A Behavioral Theory of the Firm, 1963, p. 20. See Argote and Greve: A Behavioral Theory of the Firm—40 Years and Counting, 2007, p. 344. See Cyert and March: A Behavioral Theory of the Firm, 1963, p. 121; and March and Simon: Organizations, 1958, pp. 173–174 and 184. Ideas of adaptive feedback are already apparent in Simon’s work who regards it as a process of adaptive learning. See Simon: Administrative Behavior, 1976, p. 85.
18
B Deterministic and Voluntaristic Theories of Organizational Change
atic. It triggers a myopic search process near the problem for finding a solution to the performance shortfall, called problemistic search. There also exists the conjecture of slack search, indicating that unused resources can also be an initiator of change and innovativeness.67 Yet, this type of search is of minor importance here since the focus is on stress-induced rather than slack-induced change. The aspiration level defines a satisfactory alternative and marks the point when problemistic search ends. The idea originates from Simon who called the process ‘satisficing’. This is a neologism deriving from the words ‘satisfy’ and ‘suffice’. It describes a strategy of problemistic search not with the aim of optimization, but with the aim of only finding a workable solution. Simon mentions the example of a seller of a house who—particularly when receiving sequential bids for her house—will not wait for the highest possible bid, but will sell the house as soon as somebody offers a price at or above her aspiration level.68 In day-to-day business, performance below the aspiration level will trigger problemistic search until a satisficing solution has been found. Satisficing is an important strategy that links aspiration levels with bounded rationality. There is empirical evidence for the fact that performance below aspiration levels has an effect on the overall strategy and strategic persistence of organizations. A study by Miller and Chen found poor performance to be a driver of adaptive tactical changes, but their data did not strongly support the idea that performance shortfalls initiate strategic moves.69 This was different in empirical research done by Wenneberg and Holmquist consisting of four case studies and a subsequent large quantitative analysis for aspiration levels in the context of internationalization. It tested the prevalence of the performance-action feedback as suggested by the behavioral theory and found that performance below aspiration levels increases firms’ search for international opportunities and internationalization attempts.70 Park conducted a causal statistical analysis of longitudinal data in the U.S. food processing industry and investigated how performance in relation to aspiration levels affects the choice of strategic positions. He found that low performance increases firms’ likelihood to align their strategy to that of their surrounding firms whereas high performance makes firms more likely to diverge from their competitors.71 In summary, empirical analyses in 67
68
69
70
71
See Cyert, Richard M. and James G. March: A Behavioral Theory of the Firm, 2. Ed., Englewood Cliffs, NJ 1992, pp. 188–190. See Simon, Herbert A.: Models of Man: Social and Rational: Mathematical Essays on Rational Human Behavior in a Social Setting, New York 1957, pp. 204–205 and 253. See also March and Simon: Organizations, 1958, pp. 140–141. See Miller, Danny and Ming-Jer Chen: Sources and Consequences of Competitive Inertia: A Study of the U.S. Airline Industry, in: Administrative Science Quarterly, Vol. 39 (1994), No. 1, pp. 14 and 18. See Wennberg, Karl and Carin Holmquist: Problemistic search and international entrepreneurship, in: European Management Journal, Vol. 26 (2008), No. 6, pp. 450 and 452. See Park, Kyung Min: Antecedents of Convergence and Divergence in Strategic Positioning: The Effects of Performance and Aspiration on the Direction of Strategic Change, in: Organization Science, Vol. 18 (2007), No. 3, pp. 389–390 and 399.
B.I Incommensurable Drivers of Organizational Evolution
19
general support the idea that performance below aspiration levels increases problemistic search and the attempts to adjust strategy. While the early behavioral school’s work regarded aspiration levels as fix and exogenous, Cyert and March explain how they are formed in the first place and how they then dynamically adapt.72 Concerning aspiration level formation, Cyert and March assume that organizational goals are the result of a negotiation process in which different participants bring in their interests and goals. Here, the decisionmaking of executive groups and the question of how these decisions are implemented are important. Additionally, dynamic adaptation and change of objectives is a central element of the adaptive Behavioral Theory of the Firm. It estimates aspiration levels to be a function of past aspiration levels, historical performance and past performance of comparable firms, i.e. of social aspiration levels.73 Lant conducted an early empirical analysis of Cyert and March’s aspiration sub-theory and analyzed the formation of aspiration levels in a simulation game in which the player teams assume the role of top management. She found that the adaptive and historically oriented model of adjustment to performance feedback, as it is suggested by the behavioral theory, fits empirical evidence better than perfectly rational or future-oriented models of expectation formation. It also became clear that aspirations are adapted depending on the discrepancies between the goal and actual performance. This adjustment can occur incrementally, but it can also induce substantial behavioral responses. Additionally, Lant’s experiments reveal an optimistic bias of aspirations being higher than average performance.74 A simulation analysis by Greve supports the importance of aspiration level adaptation. Slow adjustment of aspiration levels leads to higher organizational performance than quick or no adaptation when facing both slow and rapid environmental changes.75 Adaptation represents the core element of the classical to behavioral theories of the firm.76 The former is based on the complete and deterministic adaptation cycle as shown in Figure B-1. Resistance and bounded rationality pose limitations to this adaptation process, but they do not question the idea and the usefulness of an organization’s adaptation to its environment, confirming the environment as a trigger of change.
72
73 74
75
76
See Cyert and March: A Behavioral Theory of the Firm, 1963, p. 43; and Gavetti, Levinthal and Ocasio: Neo-Carnegie, 2007, p. 526. For a description of stable expectation levels see Simon: Administrative Behavior, 1976, p. 100. See Cyert and March: A Behavioral Theory of the Firm, 1963, pp. 21, 43, and 115. See Lant, Theresa K.: Aspiration Level Adaptation: An Empirical Exploration, in: Management Science, Vol. 38 (1992), No. 5, pp. 638–639 and 641–642. See Greve, Henrich R.: Sticky Aspirations: Organizational Time Perspective and Competitiveness, in: Organization Science, Vol. 13 (2002), No. 1, p. 12. For this adaptive model see Cyert and March: A Behavioral Theory of the Firm, 1963, p. 99; and March, James G. and Johan P. Olsen: Ambiguity and Choice in Organizations, Bergen [et al.] 1976, p. 56.
20
B Deterministic and Voluntaristic Theories of Organizational Change
The process of adapting to the environment and to aspiration levels is influenced by limited rationality. The concept of incomplete rationality got refined into a bounded rational cycle of learning. Departing from the model of organizational omniscience and rational calculation, it is assumed that learning takes place from experience. Here, limited human cognition can impede the different sub-steps of the adaptation process. Characteristic of the incomplete adaptive cycle are the difficulty to transform individual knowledge into individual action, of transforming individual into organizational knowledge and action, and the challenge of interpreting the environment.77 Experience requires interpretation, but success and failure are ambiguous and their causalities challenging to disentangle. In other words, interpretation is based on perception and beliefs, and mental models shape how somebody makes sense of information in the interpretation process. Since the adjustment to the environment represents the behavioral theory’s core, it can also be described by the adaptive feedback cycle presented earlier in Figure B-1. In contrast to the classical management theories, perfect adaptation is hindered by bounded rationality of humans and organizational decisions. The perception and interpretation of environmental feedback are incomplete, leading to delays and shortcomings in the adaptation process. Group decision processes have a great influence on an organization’s strategic orientation. They affect the formation of aspiration levels and the link from individual to organizational knowledge and action. Bounded rationality and organizational decision-making thus weaken the adaptive cycle. As shown in Figure B-3, they tackle it in two areas. Limited rationality has an influence on the perception of the gap between the environment’s and the focal organization’s strategic orientation. The ways in which the pressure to adapt develops into a new strategic orientation is additionally influenced by bounded rational and group influenced decision-making.
impaired by bounded rationality and group processes
Strategic Orientation +
- + (B) Deterministic gap Adaptation
pressure to adapt + strategic orientation
ENVIRONMENT'S STRATEGIC ORIENTATION
impaired by bounded rationality
Figure B-3: Limitations to adaptation While the behavioral theory and particularly some of its later derivative theories such as the garbage can model do not deny choice, the core theory focuses on ad-
77
See March, James G. and Johan P. Olsen: The Uncertainty of the Past: Organizational Learning Under Ambibuity, in: European Journal of Political Research, Vol. 3 (1975), No. 2, pp. 158–160; and March and Olsen: Ambiguity and Choice in Organizations, 1976, pp. 54–58. See also Kim, Daniel H.: The Link between Individual and Organizational Learning, in: Sloan Management Review, Vol. 35 (1993), No. 1, pp. 43–46.
B.I Incommensurable Drivers of Organizational Evolution
21
aptation to the environment, and choices are only possible within this adaptation cycle.78 The deterministic elements of this theory thus still prevail.
B.I.2
Inertia and Routines as Determinants of Change
With regards to the difficulties organizations have adapting, opinions were voiced criticizing the adaptive theories’ fundamental assumption of organizational mutability. A second stream of research does not share the belief that organizations are able to adapt if environmental conditions require them to do so. Based on the opposed assumption of organizational inertia, this research questions the adaptability of organizations and focuses on organizational survival and decline in the evolution of entire populations. It also highlights the importance of routines for organizations which can serve both as an impediment and as a driver of change. The focus on routines as well as what is called the organizational ecology approach thus developed as an alternative to the adaptive theories of organizational change.
B.I.2.a
Inertia Leading to Environmental Selection
The organizational ecology approach by Hannan and Freeman concentrates on a population perspective of organization-environment relations with a particular focus on the effects of environmental change on populations. It addresses the query of whether organizational features change by adaptation or by the selection and replacement of inflexible organizations.79 In contrast to the earlier theories, it questions the malleability of organizations, and it has little room for adaptive processes within organizations.80 Organizational inertia represents the theory’s core assumption, and it explains the lack of adaptation. A manifestation process of strong inertial forces within populations is supposed to prevent major changes.81 Inertia expresses the idea that organizations do not change as quickly or completely as some groups want them to change in order to be adequately adapted to the environment. In the view of Hannan and Freeman, inertia symbolizes a lack of adaptation ability which means that an organi78
79
80
81
Whittington calls the behavioral school’s theory deterministic in a second way, meaning that human reactions to external stimuli are predictable and governed by psychological traits. Programmed responses are part of a deterministic feedback-reaction system, this being called action determinism. See Whittington: Environmental Structure and Theories of Strategic Choice, 1988, pp. 525 and 531. See Hannan, Michael T. and John Freeman: The Population Ecology of Organizations, in: The American Journal of Sociology, Vol. 82 (1977), No. 5, particularly pp. 929–931, 934 and 957–958; Hannan and Freeman: Structural Inertia and Organizational Change, 1984, pp. 149–150; and Hannan and Freeman: Organizational Ecology, 1989, p. 69. See Adler, Paul S. and Bryan Borys: Materialism and Idealism in Organizational Research, in: Organization Studies, Vol. 14 (1993), No. 5, p. 663; Hannan and Freeman: The Population Ecology of Organizations, 1977; Hannan and Freeman: Structural Inertia and Organizational Change, 1984; and Hannan and Freeman: Organizational Ecology, 1989. See Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 149.
22
B Deterministic and Voluntaristic Theories of Organizational Change
zation is not able to sufficiently react to a change in circumstances, even if it would like to be adequately aligned with its environment.82 Opinions about the usefulness of inertia differ, but it is often referred to as a cause of the failure to adapt.83 Larsen and Lomi as well as Hannan and Freeman regard inertia as an important outcome of daily operations which gives the organization reliability and continuity. They admit that inertia prohibits organizational change, but they also emphasize that continuity is advantageous for experience and performance.84 Later Hannan and Freeman attenuate their earlier theory by acknowledging that minor organizational changes “[…] occur frequently and that organizations sometimes even manage to make radical changes in strategies and structures. Nevertheless, we argue that selection processes tend to favor organizations whose structures are difficult to change. That is, we claim that high levels of structural inertia in organizational populations can be explained as an outcome of an ecological-evolutionary process.”85 This evolutionary process abets organizations that are able to accumulate experience, reliability and continuity, making it unlikely for established organizations to be highly flexible. This bears similarity with the three examples discussed in the previous chapter. These organizations were also able to make small changes, even be highly innovative in their taken-for-granted strategy. Nevertheless they were too inert to deviate from that strategy. Organizational ecology assumes that organizational size and age increase inertia. Here, it is possible to distinguish internal constraints to adaptation and external pressure for stability. Internally, sunk costs—e.g. from the commitment to a certain type of technology—restrict the adaptive process as well as sub-optimal decision making due to information constraints or internal politics. As an example of politics, resistance can slow and block the change process. Past normative agreements may also enhance inertia. Entry and exit barriers, limits to the availability of information, and legitimacy constraints are examples for external pressure for stability.86 Organizational ecology is based on the assumption that inertia is a consequence of a selection process since more reliable and more accountable organizations have higher legitimacy and thus higher performance. These organizations’ accountability signalizes their rationality. They apply a repertoire of routines and a limited set of rules for switching between routines, leading to high reproducibility and little change. Particu-
82 83 84
85
86
See ibid., p. 151. E.g. Tushman and Romanelli: Organizational Evolution, 1985, pp. 178 and 201. See Amburgey, Kelly and Barnett: Resetting the Clock, 1993, p. 52; and Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 149. Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 149. An almost identical statement can be found in Hannan and Freeman: Organizational Ecology, 1989, p. 67. See Hannan and Freeman: Organizational Ecology, 1989, pp. 67–69.
B.I Incommensurable Drivers of Organizational Evolution
23
larly changes to the organization’s core and identity are difficult to enact.87 Inertia is thus a determinant of change and a strong factor leading to failure to change. The entire population becomes the unit of analysis in evolutionary organizational ecology, and an organization is regarded as a “cohesive organism”.88 In contrast to the approach of rational adaptation which was described earlier, the organizational ecology explains evolution mainly by the failure of maladjusted organizations and the creation of new organizations.89 Sociological and evolutionary aspects are combined.90 The socio-ecological theory links to evolutionary biology by its assumption that a population evolves by Darwinist selection. This also means that changes cannot easily be planned and enacted by organizational members. Organizational ecology is a point of view that is also part of what could be called the environmental imperative. Astley and Van de Ven argue that organizational ecologists conclude that the environment has primacy. Since the theory focuses its analysis on the population level, it attaches minor importance to internal dynamics, strategic choice, and an organization’s effect on its environment.91 With its focus on the creation of organizational reliability and inertia, the perspective thus denies the effectiveness of drivers of change. While the original theory suggests that organizations improve their performance by becoming reliable and inert, empirical evidence is ambiguous. Different processes operate in populations.92 Studies support Hannan and Freeman’s hypothesis for old organizations, but overall they detect a hump-shaped relationship between organizational age and the probability of organizational death. Newly founded organizations are protected by initial material and immaterial resources that work as a buffer. As it is used up, organizational mortality increases, before it starts to decrease again due to learning effects.93 Contrary to this, other studies even found evidence for increas-
87
88
89
90 91
92
93
See Hannan and Freeman: Structural Inertia and Organizational Change, 1984, pp. 153–156; and Hannan and Freeman: Organizational Ecology, 1989, pp. 72–77. See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, pp. 66–67. See Mellahi and Wilkinson: Organizational failure, 2004, pp. 21–41; Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 150; and Perrow, Charles: Complex Organizations: A Critical Essay, New York [u.a.] 1986, pp. 208–210 See Hannan and Freeman: The Population Ecology of Organizations, 1977, p. 962. See Astley, W. Graham and Andrew H. Van de Ven: Central Perspectives and Debates in Organization Theory, in: Administrative Science Quarterly, Vol. 28 (1983), No. 2, pp. 257–258. See Dobrev, Stanislav D., Tai-Young Kim and Glenn R. Carroll: The Evolution of Organizational Niches: U.S. Automobile Manufacturers, 1885–1981, in: Administrative Science Quarterly, Vol. 47 (2002), No. 2, p. 262. See Brüderl, Josef and Rudolf Schüssler: Organizational Mortality: The Liabilities of Newness and Adolescence, in: Administrative Science Quarterly, Vol. 35 (1990), No. 3, pp. 539 and 545–546; and Fichman, Mark and Daniel A. Levinthal: Honeymoons and the Liability of Adolescence: A New Perspective on Duration Dependence in Social and Organizational Relationships, in: The Academy of Management Review, Vol. 16 (1991), No. 2, pp. 454 and 462.
24
B Deterministic and Voluntaristic Theories of Organizational Change
ing mortality with age, when organizational size is controlled for.94 Dobrev, Kim, and Carroll observed that organizations experiencing high levels of inertia are less likely to undergo change. If they change nonetheless, they have a higher risk of failure as a result of this transformation as compared to less inert organizations. Their findings are interesting as they are in part counter-intuitive. The results suggest that organizations that fine-tune their strategy in periods when the environment is stable are more likely to remain locked in this strategy or misinterpret information from the environment when the environment undergoes a more radical transformation.95 Thus, while there is empirical evidence for the initial theory, studies can be found that support the hypothesis only in special circumstances or show inverse relationships. Larsen and Lomi formalized Hannan and Freeman’s approach, transformed it into a feedback model, and tested the theory with the help of system dynamics simulation. In this way they simulated and tested organization theory.96 They chose inertia and reliability to be a function of size as well as of experience. Pressure to change grows with reliability, but change is hindered by inertia. Larsen and Lomi are able to point to the importance of the internal dynamics between pressure to change and inertia. Findings also suggest an ideal level of resistance to change resulting from managerial attitudes. The model shows that prior change increases the likelihood of further change because it decreases inertia.97 As such it relates to drivers of change and addresses the question of whether prior change can serve as a trigger of later transformation. Larsen and Lomi’s model also relates to the concept of competence and routines that are regarded both as an impediment to change and as a source of flexibility.
B.I.2.b
Routines Hindering and Driving Change
In the organization theory tradition, much research has focused on inertia as an impediment to change. Opinions about the sources as well as about the effects of inertia—good or bad—differ. Some explanations for the reasons of inertia have already been discussed. They range from individual bounded rationality to management and organizational homogeneity coming from institutionalization processes.98 At the or94
95
96
97
98
See Barron, David N., Elizabeth West and Michael T. Hannan: A Time to Grow and a Time to Die: Growth and Mortality of Credit Unions in New York City, 1914–1990, in: The American Journal of Sociology, Vol. 100 (1994), No. 2, p. 414. See Dobrev, Stanislav D., Tai-Young Kim and Glenn R. Carroll: Shifting Gears, Shifting Niches: Organizational Inertia and Change in the Evolution of the U.S. Automobile Industry, 1885–1981, in: Organization Science, Vol. 14 (2003), No. 3, pp. 276–277. See Larsen, Erik R. and Alessandro Lomi: Representing change: a system model of organizational inertia and capabilities as dynamic accumulation process, in: Simulation Modelling Practice and Theory, Vol. 10 (2002), No. 5–7, pp. 383–384. See Larsen, Erik R. and Alessandro Lomi: Resetting the clock: a feedback approach to the dynamics of organisational inertia, survival and change, in: Journal of the Operational Research Society, Vol. 50 (1999), No. 4, pp. 412 and 417–418. For bounded rationality see March and Simon: Organizations, 1958, pp. 138–141; and Simon: Administrative Behavior, 1976, pp. 240–244. For management homogeneity see Murray,
B.I Incommensurable Drivers of Organizational Evolution
25
ganizational level, routines are of paramount importance. They have traditionally been associated with inert organizational behavior. Routines are perceived as repetitive patterns of behavior, learned capabilities that have evolved over time.99 Yet, this notion only captures the behavioral part of routines; they can also be understood as habituated and repetitive ways of thinking, cognitive patterns or filtering rules.100 Organizational routines are collective phenomena since multiple individuals are involved in their behavioral and cognitive outcomes. Many authors regard routines and the resulting path dependencies as a major impediment to change. For example, redundant processes may exist which have been implemented to solve a problem that no longer exists. Although these processes served the organization well in the past, they now represent deeply embedded, but inappropriate knowledge and impede any transition.101 Due to organizational history, there may by accumulations in the system that cannot change instantaneously and the system exhibits hysteretic, i.e. persisting behavior.102 The organization has difficulty deviating from routines because of institutionalization processes that have taken place earlier. According to Larsen and Lomi as well as Leonard-Barton, the extent of inertia depends on the feasibility and speed of the change of established routines.103 There is empirical evidence for the fact that good performance induces managers to strengthen their commitment to established routines and makes them reluctant to Alan I.: Top Management Group Heterogeneity and Firm Performance, in: Strategic Management Journal, Vol. 10 (1989), No. -, Special Issue, pp. 137–139; and Tushman and Romanelli: Organizational Evolution, 1985, p. 211. For institutionalization processes see Hannan and Freeman: Structural Inertia and Organizational Change, 1984, p. 154. 99 See Cohen, Michael D., et al.: Routines and Other Recurring Action Patterns of Organizations: Contemporary Research Issues, in: Industrial and Corporate Change, Vol. 5 (1996), No. 3, p. 663; Feldman, Martha S. and Brian T. Pentland: Reconceptualizing Organizational Routines as a Source of Flexibility and Change, in: Administrative Science Quarterly, Vol. 48 (2003), No. 1, p. 95; March and Simon: Organizations, 1958, p. 140 and 142; and Nelson and Winter: An evolutionary theory of economic change, 1982, p. 97. To some extent, the definition of routines used is consistent with Levitt and March as well as Cyert and March’s notion of standard operating procedures. See Cyert and March: A Behavioral Theory of the Firm, 1963, pp. 101–110; and Levitt, Barbara and James G. March: Organizational Learning, in: Annual Review of Sociology, Vol. 14 (1988), p. 320. 100 See Becker, Markus C.: Organizational routines: a review of the literature, in: Industrial and Corporate Change, Vol. 13 (2004), No. 4, pp. 645 and 662. 101 See Thun, Jörn-Henrik: Die zeitbasierte Fertigungsstrategie: Methoden zur Leistungssteigerung in Industriebetrieben, Wiesbaden 2002, p. 71. 102 Hysteresis (meaning “remaining”) was introduced by J. A. Ewing in relation to magnetism and established in economics by Schumpeter. See Ewing, J. A.: On the Production of Transient Electric Currents in Iron and Steel Conductors by Twisting them when Magnetised or by Magnetising them when Twisted, in: Proceedings of the Royal Society of London, Vol. 33 (1881), pp. 21–23; and Schumpeter, Joseph Alois: The Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest, and the Business Cycle: Harvard Economic Studies, Vol. 46. Cambridge, MA 1934, pp. 57–94. 103 See Larsen and Lomi: Representing change, 2002, particularly p. 291; Larsen and Lomi: Resetting the clock, 1999, p. 407; and Leonard-Barton, Dorothy: Core Capabilities and Core Rigidities: A Paradox in Managing New Product Development, in: Strategic Management Journal, Vol. 13 (1992), No. 5, p. 118.
26
B Deterministic and Voluntaristic Theories of Organizational Change
change. Experiences of success them stick to established decision rules. Audia, Locke, and Smith’s archival study of the trucking and airline industries showed that past success in organizations leads to strategic persistence, planting “the seeds for their possible future decline.”104 When facing radical change, the success-persistence-success circle develops into a vicious circle. In a similar vein, Starbuck and Milliken’s analysis of the Challenger disaster revealed how past success leads to overconfidence, reduced reflection of actions, and a high commitment to the familiar course of action.105 The above paragraphs describe how institutionalization processes create organizational routines that manifest themselves in inertia and path-dependent behavior. The past strategic orientation perpetuates. Organizational momentum is the organization’s tendency to extrapolate past strategic orientations.106 Many scientists argue that these path dependencies also work in the direction of change. Once an organization changes, it loses inertia, develops change routines and will change more frequently in the future. Researchers have called this the repetitive momentum hypothesis. The idea relates to the research question concerning drivers of change and suggests that prior change serves as a driver for future transformations. A body of literature has investigated this question by statistical empirical research that attempts to elicit relations between change and the occurrence of subsequent transformations. It has been confirmed by a number of studies that change momentum is established. The accumulative number of prior strategic changes similar to previous ones enhances the probability of change of the same type.107 Collins et al. investigated the question in relation to merger activity and identified that prior acquisition activity in a specific area is positively associated with later acquisitions in the same area.108 Kelly and Amburgey confirmed this for the U.S. air carrier industry, and found that in some cases a transformation even enhances further ones in different areas.109 Additionally, Beck, Brüderl, and Woywode list 37 instances in 19 studies in which a positive corre-
104
Audia, Pino G., Edwin A. Locke and Ken G. Smith: The Paradox of Success: An Archival and a Laboratory Study of Strategic Persistence Following Radical Environmental Change, in: The Academy of Management Journal, Vol. 43 (2000), No. 5, p. 849. 105 See Starbuck, William H. and Frances J. Milliken: Challenger: Fine-Tuning the Odds until Something Breaks, in: Journal of Management Studies, Vol. 25 (1988), No. 4, pp. 323–324, 329 and 331. 106 See Jansen, Karen J.: From Persistence to Pursuit: A Longitudinal Examination of Momentum During the Early Stages of Strategic Change, in: Organization Science, Vol. 15 (2004), No. 3, p. 277. 107 See Kelly, Dawn and Terry L. Amburgey: Organizational Inertia and Momentum: A Dynamic Model of Strategic Change, in: Academy of Management Journal, Vol. 34 (1991), No. 3, p. 596. 108 See Collins, Jamie D., et al.: Learning by doing: Cross-border mergers and acquisitions, in: Journal of Business Research, Vol. 62 (2009), No. 12, p. 1333. 109 See Kelly and Amburgey: Organizational Inertia and Momentum, 1991, p. 606.
B.I Incommensurable Drivers of Organizational Evolution
27
lation between prior and subsequent change is supported.110 These studies do not give a causal explanation for the results; only two modeling studies are more successful at doing this. Both Sastry as well as Larsen and Lomi show by a system dynamics model that change reduces inertia, making organizations more ready for further alteration.111 Although many researchers come to the conclusion that prior change increases the occurrence of subsequent transformations, there exist conflicting results. E.g. a second assessment of merger activity could only partially confirm repetitive momentum.112 Further studies even find an inverse correlation. Baum and Singh identify a negative relationship between changes of the same type. Organizations rather backpedal and change into the opposite direction.113 Also Beck and Kieser find that the cumulative number of rule changes in German banks decreases the occurrence of further rule amendments. According to them, there exists a learning process. The previous amendment and experience with a rule increases the stability of the respective rules.114 Beck et al. thus challenge the repetitive momentum hypothesis by showing that change propensity can also decrease when changes accumulate. In their view, prior change does not serve as a driver of change, but hinders subsequent alterations. They justify their finding by indicating that the more the organization transformed in the past the less further changes are necessary. The authors find no evidence for a reinforcing process of organizational change.115 In a similar vein, Amburgey, Kelly, and Barnett contribute to the ambiguity of results. They empirically test the assumption of repetitive momentum and increased flexibility for content and frequency changes of newspapers and get mixed results. They are able to show that if organizational processes are understood as routines, change routines can also establish momentum for further changes. Change of content increases the probability of subsequent change of the same type. For changes in the frequency of publication, however, their hypothesis is not supported: early changes in frequency diminish the likelihood of further changes.116 These mixed results serve as an indicator for different effects of routines working in differing direc110
See Beck, Nikolaus, Josef Brüderl and Michael Woywode: Momentum or Deceleration? Theoretical and Methodological Reflections on the Analysis of Organizational Change, in: Academy of Management Journal, Vol. 51 (2008), No. 3, pp. 419–420. 111 See Larsen and Lomi: Resetting the clock, 1999, p. 418; and Sastry, M. Anjali: Problems and Paradoxes in a Model of Punctuated Organizational Change, in: Administrative Science Quarterly, Vol. 42 (1997), No. 2, p. 256. 112 See Amburgey, Terry L. and Anne S. Miner: Strategic Momentum: The Effects of Repetitive, Positional, and Contextual Momentum on Merger Activity, in: Strategic Management Journal, Vol. 13 (1992), No. 5, pp. 343–345. 113 See Baum, Joel A. C. and Jitendra V. Singh: Dynamics of Organizational Responses to Competition, in: Social Forces, Vol. 74 (1996), No. 4, p. 1277. 114 See Beck, Nikolaus and Alfred Kieser: The Complexity of Rule Systems, Experience and Organizational Learning, in: Organization Studies, Vol. 24 (2003), No. 5, p. 807. 115 See Beck, Brüderl and Woywode: Momentum or Deceleration, 2008, pp. 425–428. 116 See Amburgey, Kelly and Barnett: Resetting the Clock, 1993, pp. 66 and 69–70.
28
B Deterministic and Voluntaristic Theories of Organizational Change
tions. When faced with demands of their environments, e.g. due to rapid technological change, organizations thus show different reactions. Once change is initiated, it is uncertain whether it amplifies or whether accumulated changes decrease the propensity of further alteration. Beck, Brüderl, and Woywode address the ambiguous results and suggest that there must be different processes working in organizations that are responsible for the ambiguity. They assume the existence of a routine process that creates pathdependent behavior, but additionally suppose that there is an underlying process making further change unnecessary once an organization has changed. The more the organization transformed in the past, the less further changes are necessary.117 While Beck, Brüderl, and Woywode’s suggestion seems promising, the ambiguous results and the lacking causal explanations—prior research almost exclusively relies on statistical analyses—make it impossible to draw a conclusion of whether there exists repetitive momentum. This makes it worthwhile to further analyzing the question whether prior transformations serve as a driver of change. It needs to be done with measures that are able to explain the causalities and reasons for the ambiguous results. Both inert and changing behavior may be perpetuated so that path dependencies develop. The repetitive momentum idea still does not give real power to decisionmakers either, since it assumes that past behavior continues only. It also leaves the organization at the fate of the environment. A third stream of research objected to this environmentally-centered view and suggested the management team to be a powerful driver of change and of an organization’s evolution. This research stream’s arguments will be presented in the following sub-chapter.
B.I.3
Transformation Triggered by Strategic Choice
Authors challenge both the rational adaptation theory and organizational ecology for their deterministic explanation of change as a result of environmental constituencies alone, as these theories attach too much importance to constraints and too little to choice.118 Technological and organizational constituencies may be restrictive in the short-run, but decisions will finally be carried out by those in control of the organization. It is Child in particular who criticizes the assumption of contextual factors determining structural variables of organizations.119 He developed a theory of strategic choice in response to the approaches that regard organizational performance as contingent on the matching of capabilities and structure with the respective environment or niche. 117
See Beck, Brüderl and Woywode: Momentum or Deceleration, 2008, p. 428. See Child: Organizational structure, environment and performance, 1972, p. 19; Mellahi and Wilkinson: Organizational failure, 2004, pp. 23 and 27; and de Rond and Thietart: Choice, chance, and inevitability in strategy, 2007, pp. 539 and 546. 119 See Child: Organizational structure, environment and performance, 1972, p. 2. 118
B.I Incommensurable Drivers of Organizational Evolution
29
The strategic choice approach is different from a homonymous methodology developed by John Friend.120 The perspective discussed here builds on the importance of the consideration of choice, as was already proposed by Chandler.121 The role of agency and choice, the nature of the organizational environment, and the relationship between organizational agents and the environment are central.122 The discussion closely relates to the agency-structure debate in sociology which discusses whether actors’ independent choices or the structural-environmental circumstances determine the fate of society and of organizations. Overall, the strategic choice perspective reveals the importance of choice as a tool of the organization’s decision-makers. It concentrates on the actions of leading groups in organizations and their role in shaping the organization. According to Child, the contingency approach, the ecological perspective of Hannan and Freeman, and the focus on routines stress the selection by the organization’s environment.123 They fall short in explaining how actors make choices between alternatives.124 Child argues against the environment as the only source of variation of organizational structure and sheds light on the importance of the dominant coalition, i.e. the powerful group of decision-makers or management within an organization.125 “Strategic choice is contingent on causality, on the belief that strategies have causes as well as consequences.”126 In a world dominated by chance, strategic choices would have no meaning. Freedom is expressed in this gap between the context for choice and choice itself. In the view of de Rond and Thietart a theory of choice does not deny the influence of the environment, but the authors do not regard the environment as sufficiently influential to call it deterministic.127 The dominant coalition is important in strategic choice. Its position as organizational power-holders is adopted from Cyert and March’s behavioral theory of the firm.128 This group, which in composition may somewhat deviate from formally designated holders of authority, evaluates the organization’s position, exercises power, 120
Friend’s methodology is part of Soft Operations Research. Friend developed methods for decision-making in groups under time pressure and uncertainties. They aim at helping people handle decision situations in a strategic manner. See Friend, John K. and Allen Hickling: Planning under pressure: the strategic choice approach, 3. Ed., Oxford 2005, pp. 1–5. 121 See Chandler: Strategy and Structure, 1962, p. 8. 122 See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 43; and Child: Organizational structure, environment and performance, 1972, pp. 1–2. 123 See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 45. See also de Rond and Thietart: Choice, chance, and inevitability in strategy, 2007, p. 546. 124 See Whittington: Environmental Structure and Theories of Strategic Choice, 1988, p. 527. 125 See Child: Organizational structure, environment and performance, 1972, p. 16. 126 de Rond and Thietart: Choice, chance, and inevitability in strategy, 2007, p. 547. 127 See ibid., p. 547. 128 For the Carnegie behavioral theory’s concept of the dominant coalition see Cyert and March: A Behavioral Theory of the Firm, 1963, pp. 240–241.
30
B Deterministic and Voluntaristic Theories of Organizational Change
and has considerable freedom of decision of how the organization is maintained.129 Strategic choice is a (political) process. Those who have power make decisions on the strategy and structure of an organization. Structural and technological constituencies, on the other hand, have limited effects. Technology may constrain organizational choice in the short run, but it is also a result of previous decisions of those in control in the organization.130 In the long run these decisions may be revised and neither structure nor technology serves as a constraint on choice. Two further actions precede the choice of strategy: the evaluation of the situation and the choice of goals. Actions do not follow from environmental conditions per se. The exercise of choice requires a prior perception, interpretation, and evaluation of the environment.131 Even in the strategic choice perspective the organizationenvironment fit and performance represent an important input for decision-makers.132 Decision-makers interpret environmental constituencies, and the interests, values, and capability of those in control affect the outcomes of these decisions.133 Following the assessment of the situation, they exercise choice over performance standards, and in the last step they formulate their strategy. As Figure B-4 shows, the dominant coalition’s evaluation of the organizational performance and interpretation of the environment shapes the choice of goals or objectives. In the next step it translates it into the organizational strategy. Two adaptive processes influence the decision-making in organizations. Decision-makers choose an internal and external strategy, and they receive internal feedback on productivity as well as an external response on sales performance. The perspective regards the deliberate manipulation of the adaptive processes and the realm of deliberate decision-making based on states of the adaptive process as to what constitutes choice. Furthermore, the strategy affects the environment, giving organizational agents a further means of exercising choice by shaping the situation by their decisions. At the individual level this has been confirmed by Hoff and McCaffrey in a qualitative study of primary care physicians. When facing environmental change, individuals resist, but simultaneously accommodate and adapt to new demands. In this way they also shape their future work conditions.134 129
See Chandler: Strategy and Structure, 1962, pp. 8 and 383; and Child: Organizational structure, environment and performance, 1972, pp. 13–14. 130 See Child: Organizational structure, environment and performance, 1972, pp. 6 and 16. 131 See ibid., pp. 4–5. 132 See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 48. 133 See Child: Organizational structure, environment and performance, 1972, pp. 1–6, 15–16, and 19; Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 48; and Greenwood, Royston and C. R. Hinings: Understanding Radical Organizational Change: Bringing Together the Old and the New Institutionalism, in: Academy of Management Review, Vol. 21 (1996), No. 4, p. 1048. 134 See Hoff, Timothy J. and David P. McCaffrey: Adapting, Resisting, and Negotiating: How Physicians Cope With Organizational and Economic Change, in: Work and Occupation, Vol. 23 (1996), No. 2, p. 185.
31
B.I Incommensurable Drivers of Organizational Evolution
2. choice of goals 1. evaluation of situation
strategy (B) Internal Strategy
organizational performance
efficiency
environmental conditions
(B) External Strategy
Figure B-4: A feedback view of strategic choice in a theory of organization135 Internal determinism and bounded rationality are elements that Child added later to the strategic choice approach. Elster as well as Whittington point to the importance of drawing a distinction between environmental and internal forms of constraint. Human behavior results from two successive filtering processes. First, environmental determinism, by which structural-environmental constraints reduce the decisionmakers’ options, has been elaborated earlier. Second, people select from the field of feasible actions by a filtering mechanism.136 This action determinism draws attention to the predetermined mindset that defines how options are considered and choices made, particularly how basic beliefs toward stability inhibit adaptive choices to new conditions. Culture, values, and norms may become internalized and constrain choices through interpretative mechanisms rather than through constraints imposed from outside. Child gives the example of many organizations that ban smoking within their buildings although legislation does not require them to do so. Action determinism sheds light on the relevance of the interpretive process and the way it influences and limits the choice process.137 This individual view complements the political process-related view of strategic choice. Later, strategic choice theorists adopted a model of bounded rational strategic choice. It builds up on research on top managers, integrating the behavioral school’s approach of bounded rationality, conflicting goals, and varying aspiration levels with Child’s perspective of choice. Strategic—i.e. complex and highly significant—choices have a large behavioral content, reflecting decision-makers’ idiosyncrasies, that is their cognitive base and values.138 They are impaired by a filtering process of man135
On the basis of Child: Organizational structure, environment and performance, 1972, p. 18; and Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 71. 136 See Elster, Jon: Ulysses and the Sirens: Studies in Rationality and Irrationality, rev. Ed., Cambridge [et al.] 1984, p. 113; and Whittington: Environmental Structure and Theories of Strategic Choice, 1988, pp. 522–524. Child later adopted the concept. See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 49. 137 See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, pp. 49–50. 138 Directly see Hambrick, Donald C. and Phyllis A. Mason: Upper Echelons: The Organization as a Reflection of Its Top Managers, in: The Academy of Management Review, Vol. 9 (1984), No. 2, pp. 194–195.
32
B Deterministic and Voluntaristic Theories of Organizational Change
agement that involves limited vision, selective perception that concentrates on familiar stimuli, and attaching meaning by interpretation. The resulting managerial perception and values then result in a strategic choice that is based on a constructed reality.139 Child identifies three types of limitations to agency and choice: restraints from an intra-organizational political process, action determinism, and informational deficiencies. While action determinism is also a result of environmental circumstances, the internal processes are important as they shape and differentiate strategic choices of organizations acting within the same environment.140 Decision-makers’ preferences and rational considerations may prompt them to choose to ignore certain developments in the environment that would require a departure from the status quo.141 It is an element of managerial choice to either attend to or restrain from environmental developments, depending on decision-makers’ preferences, values and power. In his later work Child expresses the same idea by alluding to a double structuration loop which, first, through choice and feedback shapes the organizational design and, second, exerts influence on the environment. He calls this an evolutionary framework of pro-action and re-action. It builds on the mutuality of deliberate action and environmental constraint and overcomes the incommensurability of voluntarism and determinism. The strategic choice perspective assumes a view of recursive environment-action feedback. By information processing and evaluation, choice, action, and further processing of the information feedback, organizations evolve over time. This dynamic organizational process creates two cycles which Child recognizes as inner structuration and outer structuration. The inner structuration loop characterizes how actors are influenced by the organizational design, i.e. by structures and routines. Outer structuration symbolizes how actors influence or adapt to environmental conditions or the demands of environmental groups. As these two feedback loops mutually constraint actions, a mere polarization of voluntarism and determinism, of agency and structure is misleading—they dynamically interact.142 While emphasizing the importance of decision-making by the dominant coalition, the later work on strategic choice starts to soften the organizational imperative. Deliberate choice is embedded in the dynamics between the organization, its management, and the environment. Choice takes place in the openness towards and interpretation of environmental feedback and in the selection of actions as a response to environmental conditions. Whereas the strategic choice perspective does not deny 139
See Finkelstein, Sydney, Donald C. Hambrick and Albert A. Cannella Jr.: Strategic Leadership: Theory and Research on Executives, Top Management Teams, and Boards, New York [et al.] 2008, pp. 43–49; and Hambrick and Mason: Upper Echelons, 1984, p. 195. 140 See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 52. 141 See Child: Organizational structure, environment and performance, 1972, p. 9; and Elster, Jon: Ulysses Unbound: Studies in Rationality, Precommitment, and Constraints, New York [et al.] 2000, pp. 4–7. 142 See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, pp. 70–72.
B.II Reconciliation of Environmental Determinism and Managerial Choice
33
orientation towards efficiency criteria and performance, the focus does not lie on environmentally triggered adaptation, but on the idiosyncratic dealing with information from the environment and with the options that the situation provides. In the early development of the perspective, the interpretive voluntarist paradigm dominated, yet the approach later integrated and emphasized findings from the more deterministic perspectives concentrating on routines and bounded rationality in action determinism. Deliberate choice and the position of the management team are crucial factors in organizations, but their decision-making is embedded in the specific context.
B.II Reconciliation of Environmental Determinism and Managerial Choice Pressure for adaptation from the organizational environment, inertia, and managerial choice represent three very different determinants of change. In their original formulation, the theories advancing these drivers deny the impact of other determinants of change. Adaptive theories’ core assumption is that managerial choice plays an insignificant role, and vice versa. Each organization theory regards its presumptions as most compelling, but it is contentious which of these assumptions is most adequate or whether and how they might be compatible. As each view propones a different driver of change, an analysis into their compatibility is desirable.
B.II.1
Compatibility of Voluntaristic and Deterministic Views
As both voluntarist and deterministic elements are considered in the later strategic choice perspective, this is suggestive of the investigation of the integration and reconciliation of perspectives. The question arises whether their reconciliation is an epistemologically viable and legitimate endeavor. The topic of theory reconciliation is one of the central debates in organizational theory. Several positions relate to this debate: there are those researchers who argue for paradigm incommensurability, those who aim for an integrative meta-theory, and those who use a multi-paradigm perspective to see different angles of organizations. An investigation of the commensurability of angles of view lays ground into the analysis of drivers of change. Burrell and Morgan provide a conceptualization of organizational perspectives in the form of a philosophical classification of research paradigms. As shown in Figure B-5, they divide theories on the lines of their objective vs. subjective epistemology and they categorize among their assumptions about the degree of social conflict. Their classification positioned the approaches discussed in chapter B.I as well as systems theory, and even pluralism into what they called the ‘functionalist paradigm’.143 They distinguish these approaches from other theories that aim at radical
143
See Burrell, Gibson and Gareth Morgan: Sociological Paradigms and Organisational Analysis, London 1979, pp. 29–30 and 148–152; and Morgan, Gareth: Paradigms, Metaphors, and Puz-
34
B Deterministic and Voluntaristic Theories of Organizational Change
change, emphasize chaos, or are completely subjective. Burrell and Morgan’s view regards different research paradigms as opposed and incommensurable due to their assumptions and mutually exclusive views of the social world.144 In the view of Kuhn, different languages, i.e. different meanings attributed to words, used by the paradigms make them incompatible so that meaningful communication between them is not possible.145 Kuhn also argues that different theories and paradigms cannot judge each other’s validity. He even argues that people are so entangled in a paradigm they once adopted that they are likely to resist emerging paradigms and multiparadigm views. But a multi-lens perspective may emerge as a paradigm by itself.146 Conflictive element radical change Radical humanist
regulation and status quo
Interpretive subjective
Radical structuralist Functionalist objective
Epistemology
Figure B-5: Sociological paradigms (Burrell and Morgan)147 The integrative approach criticized the organizational science field as being too heterogeneous due to multiple existing research paradigms, and Pfeffer called for more consensus and agreement concerning relevant questions, variables, methods, and models of behavior.148 This view regards Burrell and Morgan’s framework as a “relatively static and myopic vision” and suggests a middle course between deterministic thinking and enactment of the environment.149 Hence the integrative position aims at the synthesis of paradigms into a new dominant paradigm in organizational science. But this position was not only criticized by the advocates of incommensurability who stand in the tradition of Burrell and Morgan, but also by those who advance multi-paradigm research. Multi-paradigm organizational research is a field of study that gains insight by using multiple views of organizations. It can be applied to the characterization of rezle Solving in Organization Theory, in: Administrative Science Quarterly, Vol. 25 (1980), No. 4, p. 608. 144 See Burrell and Morgan: Sociological Paradigms and Organisational Analysis, 1979, p. x. 145 See Kuhn, Thomas S.: The Structure of Scientific Revolutions, 2. Ed., Chicago 1970, p. 200. See also Jackson, Norman and Pippa Carter: In Defence of Paradigm Incommensurability, in: Organization Studies, Vol. 12 (1991), No. 1, pp. 117–118. 146 See Kuhn: The Structure of Scientific Revolutions, pp. 152 and 199–200. 147 On the basis of Burrell and Morgan: Sociological Paradigms and Organisational Analysis, 1979, p. 29. 148 See Pfeffer, Jeffrey: Barriers to the Advance of Organizational Science: Paradigm Development as a Dependent Variable, in: The Academy of Management Review, Vol. 18 (1993), No. 4, pp. 616 and 618. 149 Reed, Michael I.: Redirections in Organizational Analysis, London [et al.] 1985, pp. 204–205.
B.II Reconciliation of Environmental Determinism and Managerial Choice
35
search, to data collection and analysis as well as to theory building.150 The approach aims at the comprehension of paradigmatic differences, similarities, and interrelationships. It favors the simultaneous consideration of alternative views without aiming at paradigmatic synthesis as Pfeffer does, because assumptions, vocabularies, and goals are incompatible. Instead, the view suggests the perspective of an observer that allows for a more comprehensive consideration of phenomena of organizations than a single paradigmatic view.151 This means multi-paradigm research may arise from a traditional paradigm, but the perspective is enriched by the consideration of other views. Instead of focusing on incompatibilities between positions, it is important to clarify the relations between them.152 This is because different representations produce idiosyncratic insights. Mutually exclusive views of the social world do not represent immutable metaphysical principles that need to be obeyed. They have incommensurable as well as continuous elements.153 It is important to highlight tensions coming from incompatibilities between viewpoints.154 Then it is possible to learn from both their diversity as well as from their similarities.155 There are themes which are part of several approaches, although they talk about them by using different languages. For example, interpretive theories as well as functionalist perspectives talk about structure but have a different understanding of the concept. It may be objective reality or the result of a creation process. The functionalist perspective may enrich a more interpretive view by creating awareness that organizational members may regard their environment as an external and objective reality. An interpretive view may enrich functionalism by showing that an organization’s own choices helped the structuration of the environment.156 Multi-paradigm inquiry investigates phenomena from one theoretical perspective and enlarges this perspective by further perspectives and their relation to each other. It is thus often centered in a traditional paradigm, but enlarges this paradigm’s horizon by creating awareness of its assumptions. Then it enriches the central paradigm by views and insights of other perspectives. It holds the view that the sum of the insights generated by the consideration of many perspec150
See Lewis, Marianne W. and Andrew J. Grimes: Metatriangulation: Building Theory from Multiple Paradigms, in: The Academy of Management Review, Vol. 24 (1999), No. 4, p. 673. 151 See Gioia, Dennis A. and Evelyn Pitre: Multiparadigm Perspectives on Theory Building, in: The Academy of Management Review, Vol. 15 (1990), No. 4, pp. 596 and 599. 152 See Poole, Marshall Scott and Andrew H. Van de Ven: Using Paradox to Build Management and Organization Theories, in: The Academy of Management Review, Vol. 14 (1989), No. 4, p. 576. 153 See Willmott, Hugh: Breaking the Paradigm Mentality, in: Organization Studies, Vol. 14 (1993), No. 5, pp. 701–703 and 708. 154 See Lewis, Marianne W. and Mihaela L. Kelemen: Multiparadigm inquiry: Exploring organizational pluralism and paradox, in: Human Relations, Vol. 55 (2002), No. 2, p. 260; and Poole and Van de Ven: Using Paradox to Build Management and Organization Theories, 1989, p. 563. 155 See Gioia and Pitre: Multiparadigm Perspectives on Theory Building, 1990, p. 598. See also Morgan, Gareth: Images of Organization, updated Ed., Thousand Oaks, CA 2006, pp. 8 and 337–339. 156 See Gioia and Pitre: Multiparadigm Perspectives on Theory Building, 1990, p. 596.
36
B Deterministic and Voluntaristic Theories of Organizational Change
tives is greater than those won by the consideration of a single one. In more concrete terms this means that the joint consideration of environmental determinism and managerial choice as well as of environmental and managerial drivers of change is more insightful than information gained from the separate analysis of organizational phenomena with different approaches. While many scholars emphasize the usefulness of theory integration and multiparadigm research, others take the opposite stance and refer to potential risks. Jackson and Carter argue that those who advocate paradigm integration aim at a middle ground between structural determinism and cognitive realism; but then this middle ground may become the dominant paradigm in itself, providing no reason for theory integration.157 In the view of Deetz, different research traditions have different goals and assumptions that require different forms of evaluation. These orientations have been developed to answer their respective research questions, but are not necessarily appropriate to shed light on other questions.158 Jackson and Carter add, if scientific knowledge is derived from beliefs and assumptions about the nature of the world, it then can only be validated ideologically, never impartially. Pluralism and multiparadigm inquiry does not indicate how to judge between contradictory claims. Incommensurability serves the goal of protecting diversity in scientific inquiry. It uses different lenses provided by paradigms and guards them against a unified and totalitarian scientism.159 In this respect, however, the incommensurability view is not very different from multi-paradigm research because the latter also aims at illuminating different facets of phenomena by divergent lenses. Multiple paradigms provide divergent lenses which enable insights about different facets of interdependent and complex phenomena in organizations.160 From this view of multi-paradigm research one may understand change as a multi-faceted phenomenon ranging from adaptation to selection and choice. Some scientists argue that paradigms are incommensurable because meaningful communication between them is not possible. Still, others foster communication between paradigms for intra-paradigmatic development and in order not to develop in cognitive isolation, but they warn against paradigm assimilation.161 They see advantages in inter-paradigm communication. Lying out the claims for discussion on tensions and conflicts between areas helps approaches not to become narrowminded. Communication between paradigms is incomplete, but it is possible.162 157
See Jackson and Carter: In Defence of Paradigm Incommensurability, 1991, p. 121. They refer to Reed: Redirections in Organizational Analysis, 1985, p. 204. 158 Directly see Deetz, Stanley: Describing Differences in Approaches to Organization Science: Rethinking Burrell and Morgan and Their Legacy, in: Organization Science, Vol. 7 (1996), No. 2, p. 204. See also Willmott: Breaking the Paradigm Mentality, 1993, p. 702. 159 See Jackson and Carter: In Defence of Paradigm Incommensurability, 1991, pp. 110–112. 160 Directly see Lewis and Kelemen: Multiparadigm inquiry, 2002, pp. 258 and 272. 161 See Jackson and Carter: In Defence of Paradigm Incommensurability, 1991, p. 123. 162 See Deetz: Describing Differences in Approaches to Organization Science, 1996, pp. 193 and 204.
B.II Reconciliation of Environmental Determinism and Managerial Choice
37
Researchers hold different positions concerning the question whether it is legitimate to reconcile different theories. The risks of integrating fundamentally different points of view or epistemologies have been laid out, but it has also been said that the organization theories addressed in this dissertation are all part of what Burrell and Morgan call the functionalist paradigm. Thus, while the approaches do differ in some of their underlying philosophies and assumptions, they are not radically different. This gives ground for fruitful communication and also learning between these theories and application in multi-paradigm research. Child adopts the view that while different organization theories or paradigms are irreconcilable in their philosophical underpinnings, i.e. in their world view and assumptions, they are not incommensurable when they are applied to the study of organizations.163 What is important is the focus on process and on the dynamics between different views and between the different ideas put forward by them. These multiple views are then particularly helpful when they concern change processes. It is even necessary to consider multiple aspects of change for understanding its diverse drivers.
B.II.2
Understanding Change by the Combination of Environmental and Managerial Forces
The aforementioned positions reveal that many different theories can contribute to an understanding of organizations and how they change. The adaptive and the ecological view account for the relevance of the environment. Strategic choice highlighted the importance of individual decisions. It added understanding by showing that one theory cannot be applied universally and that the system is complex because there is feedback within the entire organization-environment system. Lately also theories of environmental selection and of firm adaptation have grown closer as organizational ecology has borrowed from the behavioral theory of the firm and has studied causes and consequences of change.164 Researchers have even jointly used both approaches.165 Adaptation and choice perspectives have also been applied together.166 Several calls for integration and respective applications in relation to organizational change will be elaborated in this sub-chapter.
163
See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 44. See Argote and Greve: A Behavioral Theory of the Firm—40 Years and Counting, 2007, p. 340. 165 E.g. Barnett, William P. and John Freeman: Too Much of a Good Thing? Product Proliferation and Organizational Failure, in: Organization Science, Vol. 12 (2001), No. 5, pp. 540–545 and 554; and Ingram, Paul and Joel A. C. Baum: Opportunity and Constraint: Organizations' Learning from the Operating and Competitive Experience of Industries, in: Strategic Management Journal, Vol. 18 (1997), No. -, Special Issue on Organizational and Competitive Interactions, particularly pp. 91–97. 166 E.g. Burgelman, Robert A.: Strategy as Vector and the Inertia of Coevolutionary Lock-in, in: Administrative Science Quarterly, Vol. 47 (2002), No. 2, pp. 352–354. 164
38
B Deterministic and Voluntaristic Theories of Organizational Change
Organization theories often specialize on one driver of organizational evolution and change only such as the environment or the management team. According to Adler and Borys, they then ignore the relationship between this aspect and other factors. The attribution of too much weight on one force may blur existing relationships and causalities. Similarly, the mere acknowledgement of the interwovenness of all forces seems intriguing, but it fails to uncover the general interdependence between different factors. Therefore, the inclusion of connections between different forces and drivers of change prooves fruitful, in particular the investigation of causality in organization theory and change.167 In the view of Astley and Van de Ven it is important to investigate how environment and choice-focused theories mutually interrelate. The appreciation of the causal relationships between deterministic and subjectivemanagerial views allows for a dynamic understanding of organizations. Organizations to some extent follow deterministic rules of the system which they are embedded in, and they are only partially modifiable by actors. Though, since they are maintained by individuals who react based on idiosyncratic perceptions, organizations are also subjective systems. Traditional theories are not able to discover relationships between the objective elements and individual actions, how these unfold over time, and how they are able to produce organizational change. Theorists often establish unidirectional causal models that only consider a limited number of isolated variables. The authors emphasize the importance of an investigation into the processes by which feedback cycles and relationships unfold over time.168 The perspective is thus dynamic. It is able to analyze the relationship between changes in the environment and in the organization as well as causal relations and feedback and the resulting process of how organizations evolve over time. Some researchers already regard organizational change theories as a mixture of four ideal types of change. They identify four drivers of change: an immanent lifecycle, environmental or competitive selection, purposeful enactment of adaptation, and dialectical conflict. For them, the organizational ecology approach is driven by selective forces, and they regard March and Simon’s behavioral theory of the firm as a theory of adaptive enactment rather than of deterministic adaptation. They advocate a process-related and pluralistic view because they consider the focus on one representation of organizational change as one-sided.169 The combination of ‘mutually exclusive’ but complementary approaches is able to generate a more holistic model of complex organizational dynamics.170
167
Adler and Borys: Materialism and Idealism in Organizational Research, 1993, pp. 663–666. See Astley and Van de Ven: Central Perspectives and Debates in Organization Theory, 1983, pp. 266–267. 169 See Van de Ven, Andrew H. and Marshall Poole: Explaining development and change in organizations, in: The Academy of Management Review, Vol. 20 (1995), No. 3, pp. 527–532. 170 See Van de Ven, Andrew H. and Marshall Scott Poole: Alternative Approaches for Studying Organizational Change, in: Organization Studies, Vol. 26 (2005), No. 9, pp. 1387 and 1395–1396. 168
B.II Reconciliation of Environmental Determinism and Managerial Choice
39
Organization researchers have increasingly employed a multi-paradigm view by the combination of different lenses. In this way they offer a more detailed picture of how organizations change. They have also started to focus on the importance of causal relationships and feedback between different drivers of change. Table B-2 gives an overview of these approaches. The integration of a content and process view of change by Rajagopalan and Spreitzer constitutes an example of a multi-lens perspective and integrative framework. They apply a rational adaptation lens, a learning lens, and a cognitive lens to understand change processes.171 The rational lens— somewhat comparable to the homo economicus type of thinking—is capable of understanding antecedents of change and its effects on performance. Yet, it treats managerial actions as a black box and neglects feedback of how managerial actions affect the environmental circumstances, and it thus assumes a deterministic and immutable context with little scope for decision-makers to intervene, experiment, and learn.172 The learning perspective’s advantages are its focus on managerial processes (i.e. change readiness and resistance) and the interdependencies among environmental, organizational, and strategic factors. The cognitive perspective elucidates the underlying logic for managerial actions and reveals an interactive process of dynamic learning. Yet, these single views have limitations since researchers rarely explicitly distinguish cognitions, actions, and strategies, and they fail to address economic outcomes, so that causal relationships between them remain opaque.173 As a solution, Rajagopalan and Spreitzer suggest an integrated multi-lens perspective incorporating rational adaptation, learning, and cognition. It explicitly distinguishes cognitions, actions, strategies, and outcomes. Strategy needs to reflect environmental requirements, offer insight into why firms react differently in similar contexts, and how managers learn from continuously reshaping their cognitions.174 It is important to apply multiple lenses which consider processes and feedback.175 With the help of this framework they are able to better address the question of the extent to which changes in strategies are explained by environmental variation, inertia, managerial cognition and action. The reconciliation of the organizational adaptation and selection theories of change is exemplified in the middle row of Table B-2. According to Péli it is able to give insight into situations in which adaptation and out-selection occur as well as into the effects of change on organizational mortality. Concerning the initiation of adaptive processes it is argued that incumbents do adapt their efficiency to quantitative envi171
See Rajagopalan, Nandini and Gretchen M. Spreitzer: Toward a Theory of Strategic Change: A Multi-lens Perspective and Integrative Framework, in: The Academy of Management Review, Vol. 22 (1997), No. 1, pp. 48–50. 172 See ibid., pp. 55–56. 173 See Buchanan, David and Patrick Dawson: Discourse and Audience: Organizational Change as Multi-Story Process, in: Journal of Management Studies, Vol. 44 (2007), No. 5, pp. 669 and 682; and Rajagopalan and Spreitzer: Toward a Theory of Strategic Change, 1997, p. 66. 174 See Rajagopalan and Spreitzer: Toward a Theory of Strategic Change, 1997, pp. 70–71. 175 See Buchanan and Dawson: Discourse and Audience, 2007, pp. 669 and 682.
40
B Deterministic and Voluntaristic Theories of Organizational Change
ronmental changes, e.g. to increases in demand. Nevertheless, when faced with substantial qualitative change, meaning customers require a different type of product, efficiency improvements are insufficient for adapting to the new demands. A substantive transformation is impossible because these organizations are highly adapted to matching the demand of the previous period. They are ‘selected out’.176 This explains when adaptation occurs and when inertia causes organizational decline. Levinthal argues for the combination of adaptive and selective theories. The effect of change on organizational mortality is not clear as there are different short and long-term effects. In the reorganization phase organizations face a short-run risk with innerorganizational turbulence and low experience. This enhances the likelihood that organizations are selected out by the environment and it increases their mortality rates. There also exists a long-term effect of change which in general results in improved adaptation to the environment. Hence, adaptation and selection are not mutually exclusive, but fundamentally interdependent processes of change.177 Paradigms employed by respective perspectives Rajagopalan and Spreitzer, 1997
Levinthal 1991 Adaptation Péli 2009
Astley, 1985
Mellahi and Wilkinson, 2004
Managerial Cognition
Adaptation
Understanding of interdependency of short-term and long-term effects of change on organizational mortality/ success.
Environmental Selection (Orgl Ecology)
Environmental Selection (Orgl Ecology)
Consideration of processes, causal relationships, and feedback among environmental variations, managerial cognition, and action to explain change.
Explanation of when adaptation occurs and when inertia causes out-selection.
Strategic Choice
Constraints by and enactment of the environment and niche. Dynamic interdependency between environmental and organizational determinants of adaptation.
Table B-2: Selected examples of multi-paradigm research in organizational change theory 176
See Péli, Gábor: Fit by Founding, Fit by Adaptation: Reconciling Conflicting Organization Theories with Logical Formalization, in: Academy of Management Review, Vol. 34 (2009), No. 2, p. 357. 177 See Levinthal, Daniel A.: Organizational Adaptation and Environmental Selection-Interrelated Processes of Change, in: Organization Science, Vol. 2 (1991), No. 1, pp. 142–144.
B.II Reconciliation of Environmental Determinism and Managerial Choice
41
The combination of environmental selection and strategic choice (bottom row of Table B-2) provides information on organizations’ mortality and success as well as about their ability to create new niches of opportunity. First, there is no single line of causation which determines organizational failure and success. Rather, both environmental and managerial causes have to be taken into consideration. Only an integrative framework is able to account for the “interplay between contextual factors and organizational dynamics”.178 It needs to consider the dynamic interdependence between choice and determinism in an organization’s adaptation to the environment. Second, the combination of organizational ecology and strategic choice demonstrates that organizations neither exclusively try to adapt to existing niches nor die if they fail to do so. Instead, they are also able to enact their own domains and niches. “Environmental constraint seen from one point of view is open opportunity from the other.”179 When conditions of environmental openness prevail and selective forces are weak, opportunistic choice drives organizational dynamics.180 Organizations are able to create their own paths. Multi-paradigm research has not only been discussed theoretically, but empirical research has also used multiple lenses. “[… S]ome of the most recent studies integrate elements from various previous perspectives, therefore leading to a certain convergence among different streams of research.”181 These studies combine adaptive and selective views, adaptation and choice, or even integrate more than two perspectives. For example, a historical analysis of the California wine industry confirmed that organizations do change frequently, but these changes do not necessarily increase adaptation to the environment. They also exhibit strong inertia, and they are particularly reluctant to change if it touches their core. Changes occur more frequently if they have transformed in the past. Organizations thus exhibit traits portrayed in the organizational ecology as well as the adaptation literature.182 An analysis of day care centers also found evidence that the evolution of organizations is shaped by adaptive and selective forces. The relationship between these two forces does not simply add up, but generates complex dynamics.183 A further case study of the Intel Corporation revealed how adaptive forces interfere with selective processes and inertia. Routine-based decision-making that evolved with the organization’s past success 178
Mellahi and Wilkinson: Organizational failure, 2004, p. 34. See also Hrebiniak, Lawrence G. and William F. Joyce: Organizational Adaptation: Strategic Choice and Environmental Determinism, in: Administrative Science Quarterly, Vol. 30 (1985), No. 3, p. 346. 179 Astley, W. Graham: The Two Ecologies: Population and Community Perspectives on Organizational Evolution, in: Administrative Science Quarterly, Vol. 30 (1985), No. 2, p. 234. 180 See ibid., p. 234. See also Milling: Systemtheoretische Grundlagen zur Planung der Unternehmenspolitik, 1981, p. 51. 181 Demers, Christiane: Organizational Change Theories: A Synthesis, Los Angeles, CA [et al.] 2007, p. xiv. 182 See Delacroix, Jacques and Anand Swaminathan: Cosmetic, Speculative, and Adaptive Organizational Change in the Wine Industry: A Longitudinal Study, in: Administrative Science Quarterly, Vol. 36 (1991), No. 4, pp. 651 and 656–657. 183 See Baum and Singh: Dynamics of Organizational Responses to Competition, 1996, p. 1286.
42
B Deterministic and Voluntaristic Theories of Organizational Change
together with strong socialization processes of employees created path-dependent behavior and commitment to the status quo.184 Thus there is strong empirical evidence for the mutual existence of adaptation, inertia, and selection, providing support for their analysis with a multi-paradigm lens. Further research has substantiated the link to managerial choice as well. An experimental study investigated the effect of the pace of environmental change on organizational decisions. It analyzed whether the respective environment moderates rationality and decision quality and thus tested the interconnectedness of environmental influences and rational choice. The authors came to the conclusion that the rationality requirements in dynamic and stable environments differ and that decision quality also depends on environmental dynamism.185 Moreover, a further qualitative and quantitative study of strategic decision-making suggests that choices may be contingent on decision, environmental, and firm characteristics. Additionally, the type of economic i.e. environmental system may have an influence on the decision outcome.186 A case study of technology-mediated organizational change identified multiple determinants of change. Environmental circumstances, organizational routines, roles, and choices add up to a mid-range process theory of organizational change. It understands change as an adaptive process, mediated by elements of choice.187 Existing examples show the relevance and viability of theoretical pluralism. A considerable number of researchers followed the calls for theory pluralism and integration. By the combination of adaptation and selection, or selection and choice, they consider multiple drivers of change. As Child expresses, irreconcilable paradigms are not incommensurable when applied to the study of organizations.188 In their application these approaches are not too different since they regard organizations as complex entities that need to be studied with different lenses, be it from multiple theories’ views or a single theory that applies multiple lenses. There is agreement about the significance of process, meaning a shift away from a static perspective towards behavior over time. Research also increasingly emphasizes the importance of feedback, of how causal relationships unfold over time, and how this contributes to a dynamic understanding of the evolution of organizations.
184
See Burgelman: Strategy as Vector and the Inertia of Coevolutionary Lock-in, 2002, pp. 351–354. 185 See Hough, Jill R. and Margaret A. White: Environmental dynamism and strategic decisionmaking rationality: an examination at the decision level, in: Strategic Management Journal, Vol. 24 (2003), No. 5, pp. 486–487. 186 See Elbanna, Said and John Child: The Influence of Decision, Environmental and Firm Characteristics on the Rationality of Strategic Decision-Making, in: Journal of Management Studies, Vol. 44 (2007), No. 4, pp. 579–580. 187 See Volkoff, Olga, Diane M. Strong and Michael B. Elmes: Technological Embeddedness and Organizational Change, in: Organization Science, Vol. 18 (2007), No. 5, pp. 842 and 845–846. 188 See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 44.
B.II Reconciliation of Environmental Determinism and Managerial Choice
B.II.3
43
A Reconciled Theory of Radical Change
The punctuated equilibrium approach, put forward by Tushman and Romanelli, represents an example of a multi-paradigm theory. As each new perspective adds a little different understanding,189 it demonstrates how multiple views can be combined. The punctuated equilibrium approach explains in more detail how the environment and management can be similarly important drivers of change. It proposes a holistic theory of organizational evolution that combines elements of the (bounded) rational adaptation theories, of organizational ecology and strategic choice. The research stream links structure and organizational processes, strategic choice, and the environment, and it reveals how strategic change is limited by inertia.190 This organizational evolution model is called punctuated equilibrium model because it assumes that organizations progress through convergent phases of incremental change and rather short reorientations of discontinuous change, called reorientations that represent a certain degree of threat.191 Tushman and Romanelli “define strategic reorientations as simultaneous and discontinuous changes in strategy, power, structure and controls. Recreations are reorientations which also involve discontinuous shifts in core values.”192 Re-creations are the most fundamental form of reorientations. In the view of Gersick, this concept alludes to several further approaches: apart from Schumpeterian thinking, she points out Kuhn’s theory of scientific revolution as well as evolutionary theories, all of which include elements of radical transformation.193 The following quotation summarizes the essence of the theory: “Patterns of organizational evolution are characterized by periods of convergence punctuated by reorientations leading to the next convergent period. These cycles are driven by the emergence of tension between organizational and institutional forces for inertia and competitive, technological and legal pressure on performance which are mediated by the perceptions and decisions of executive leadership.”194
189
See Morgan: Images of Organization, 2006, pp. 337–341. See Tushman and Romanelli: Organizational Evolution, 1985, p. 185. The punctuated equilibrium model also links to the industrial organization approach, a complement to the organizational ecology approach which also recognizes the role of management for organizational evolution. Industrial organization is a microeconomic theory focusing on market structure and firm behavior. 191 See Gersick, Connie J. G.: Revolutionary Change Theories: A Multilevel Exploration of the Punctuated Equilibrium Paradigm, in: The Academy of Management Review, Vol. 16 (1991), No. 1, p. 12; Nadler, David A. and Michael L. Tushman: Types of Organizational Change: From Incremental Improvement to Discontinuous Transformation, in: Nadler, David A., et al. (Ed.): Discontinuous Change: Leading Organizational Transformation, San Francisco 1995, p. 23; Tushman and Romanelli: Organizational Evolution, 1985, p. 181; and Wollin, Andrew: Punctuated equilibrium: reconciling theory of revolutionary and incremental change, in: Systems Research and Behavioral Science, Vol. 16 (1999), No. 4, pp. 365–366. 192 Tushman and Romanelli: Organizational Evolution, 1985, p. 203. 193 See Gersick: Revolutionary Change Theories, 1991, particularly pp. 13–16. 194 Tushman and Romanelli: Organizational Evolution, 1985, p. 181. 190
44
B Deterministic and Voluntaristic Theories of Organizational Change
The punctuated equilibrium theory of Tushman and Romanelli states that convergent periods become manifest by two different accumulations. Inner-organizational inertia as well as environmental pressure for change accumulate. Due to the buildup of inertia within the organization, the perception of and reaction to this pressure is biased. The executive team has great importance in this process; its homogeneity creates bias and hinders the perception of outside pressure. The elements of inertia, pressure from the environment, as well as the role of the management team and its perception of pressure will be discussed below. Concerning organizational inertia, the punctuated equilibrium model strongly follows the organizational ecology approach. In periods of convergence, which are characterized by stability and incremental change, the organization consolidates on its current strategic orientation. During these periods, values, beliefs, and ideologies develop and strengthen among individuals, groups, and the entire organization, thus creating inertia. This is shown in Figure B-6. The level of commitment to values or the level of inertia affects leadership behavior. Senior management and its values strongly shape an institutionalization process by which employees become socialized and ‘learn’ the cultural view. Consistency also builds because organizations attract and select employees whose values are consistent with those of the organization. Over time, commitment to the current work strategies and processes deepens; the more the organization becomes institutionalized and consistent, the more it becomes inert.195 It then only changes incrementally. The institutionalization processes further deepen values and inertia, creating reinforcing mechanisms of convergence. Values and Beliefs and Ideologies = Inertia
socialization and institutionalization (R)
(R) training
recruiting
leadership behavior
Figure B-6: Convergence in the punctuated equilibrium model In the view of Tushman and Romanelli, long and quiet convergent periods enhance organizational effectiveness.196 This means organizational consistency and institutionalization render an organization experienced and effective in what it does. At the same time, the convergence process also consolidates a firm’s strategic orientation; individuals and groups become increasingly committed to the current strategy. This reduces the likelihood to perceive a need to change the strategic orientation. As a consequence, the effects of convergence are twofold: increasing effectiveness en-
195 196
See ibid., pp. 192–193. See also Wollin: Punctuated equilibrium, 1999, pp. 363–364. See Tushman and Romanelli: Organizational Evolution, 1985, p. 195.
B.II Reconciliation of Environmental Determinism and Managerial Choice
45
hances success, but the increased commitment to the current alignment can become a risk when environmental conditions shift.197 In internally consistent organizations, reorientations face political challenges. “Coalitions of interests in large organizations are made up of stable, self-perpetuating groups who have a vested interest in the status quo, and who make consequential decisions slowly and with frequently biased and distorted information.”198 These interest groups may render attempts at change difficult, and information processing may be biased. The processes by which organizations develop inertia biases executives in a likewise manner. Homogeneous executive teams enhance inertial processes at the individual level since group tenure and homogeneity increase a management team’s convergence on decision-making procedures, norms and values. A long and particularly a successful period of convergence creates greater commitment to the current processes and strategy. Homogeneous structures filter information and they lower the openness to adverse information, supporting the status quo.199 Due to these biased perceptions and inertia, reorientations will most likely occur after a sustained period of unsatisfactory performance. They may be initiated by current management, but the likelihood for change increases with heterogeneity as well as recent changes in the executive team.200 Tushman and Romanelli see a reciprocal relationship between strategy and structure; in convergent periods, they say, the homogeneity or consistency of organizational structure drives the organizational members’ commitment to the current strategic orientation. During times of reorganization, inertia decreases and strategy drives structure.201 Hence, the punctuated equilibrium model favors a combination of environmental determinism and choice, as the approach assumes that executive leadership indeed has a crucial function. “Environments do not cause reorientations. Rather, direct responsive activity which intervenes on prior activity patterns and establishes new patterns is required for reorientations to occur. Direct executive leadership is required because internal inertial forces operate to maintain the status quo.”202 Middle man197
See Tushman, Michael L. and Charles A. O'Reilly III: Ambidextrous organizations: Managing evolutionary and revolutionary change, in: California Management Review, Vol. 38 (1996), No. 4, pp. 17–19; and Tushman and Romanelli: Organizational Evolution, 1985, pp. 190 and 197. 198 Tushman and Romanelli: Organizational Evolution, 1985, p. 192. 199 See Staw, Barry M., Lance E. Sandelands and Jane E. Dutton: Threat Rigidity Effects in Organizational Behavior: A Multilevel Analysis, in: Administrative Science Quarterly, Vol. 26 (1981), No. 4, pp. 502 and 519; and Tushman and Romanelli: Organizational Evolution, 1985, pp. 210–211. 200 See Naranjo-Gil, David and Frank Hartmann: Management accounting systems, top management team heterogeneity and strategic change, in: Accounting, Organizations and Society, Vol. 32 (2007), No. 7/8, p. 747; Naranjo-Gil, David, Frank Hartmann and Victor S. Maas: Top Management Team Heterogeneity, Strategic Change and Operational Performance, in: British Journal of Management, Vol. 19 (2008), No. 3, pp. 229–231; and Tushman and Romanelli: Organizational Evolution, 1985, pp. 180 and 212–213. 201 See Tushman and Romanelli: Organizational Evolution, 1985, p. 215. 202 ibid., p. 210.
46
B Deterministic and Voluntaristic Theories of Organizational Change
agement often rather interpolates structures and systems, and upper management is responsible for the initiation, shaping, and directing of change.203 This makes managerial choice an important driver of change. In the view of Tushman and Romanelli, population ecology, strategic management, and industrial organization theory literatures all regard organizational competence as related to the fit between the organization’s strategic orientation and its internal and external environmental conditions. These views advocate different degrees of determinism, but they agree that organizations need to be both adapted to their environment and internally consistent in order to achieve adequate performance.204 According to Nadler and Tushman, there are alternating periods of minor and radical change in the entire industry. During periods of disequilibrium, companies change strategy, structure, processes, and even people. They find that early movers in this process have a tendency to be more successful.205 Concerning drivers of change, transformations in many sectors are initiated by the coupling of technological innovations and consequent change in legal and political conditions. Similar to performance, pressure from the competitive and technological developments, pressure may also arise internally from a redefinition of performance indicators and a shifting perception of strategic contingencies. Likewise, shifts in the balance of power may create internal pressure for fundamental change.206 Stakeholder groups are able to put pressure on an organization to change, from within the organization as well as from outside. Perception of the environment and subsequent management choices are crucial for achieving a fit with the environment, but distorted or one-sided perception filters the recognition of new information. Perception and choices of decisionmakers are shaped by the length of the organization’s prior convergent period and its level of success during this time. High success is likely to distort perception and increase resistance, so that change rarely takes place in the absence of a crisis of sustained low performance, or a major shift in the distribution of power.207 In summary, the punctuated equilibrium approach links environmental and managerial drivers of change. It receives its name from the assumption that organizations progress through convergent periods of incremental change and short periods of radical transformations. During convergent periods, inertia consolidates and also influences reorientations. The management team has a crucial role in triggering change, but it is also biased in its information processing and by inertia. Its perception of the environment is important for an organization’s evolution. 203
See Nadler and Tushman: Types of Organizational Change, 1995, pp. 30 and 33; and Tushman and Romanelli: Organizational Evolution, 1985 p. 173. 204 Directly see Tushman and Romanelli: Organizational Evolution, 1985, pp. 183–189. Mellahi and Wilkinson regard industrial organization economics as deterministic whereas Tushman and Romanelli regard it as a compound of determinism and choice. See also Mellahi and Wilkinson: Organizational failure, 2004, pp. 22–23. 205 See Nadler and Tushman: Types of Organizational Change, 1995, pp. 20–21. 206 See Tushman and Romanelli: Organizational Evolution, 1985, pp. 201–202 and 205. 207 See ibid., pp. 180 and 204–207.
B.II Reconciliation of Environmental Determinism and Managerial Choice
47
Empirical research supports many of the approach’s theoretical propositions. Based on a 3-year study of 25 minicomputer companies, Romanelli and Tushman affirmed that organizations change by the pattern of the punctuated equilibrium model; they rarely change incrementally. The authors revealed a significant relationship between major environmental changes as well as CEO succession and organizational revolutionary change; nevertheless, the impact of performance crises was insignificant.208 Overall they show that pressure from outside the organizational system has the potential to initiate organizational change. Wischnevsky and Damanpour replicated Romanelli and Tushman’s test in a less turbulent environment, the banking industry. Based on 20-year data of 50 bank holding companies, they found that both revolutionary as well as incremental transformations are common. These results are congruent with two other studies indicating that organizations also change incrementally.209 New top executives as well as changes in the banks’ regulatory environment often triggered reorientations. Contrary to their expectations, severe performance declines decrease the likelihood of transformations.210 This result might be explained by the fact that decision-makers revert to superficial and automatic information processing when facing a threat.211 Wischnevsky and Damanpour later expanded their analysis and distinguished between seldom strategic and more frequent structural changes. In this later analysis, low performance caused radical strategic change as it is proposed by the punctuated equilibrium theory.212 Sabherwal, Hirschheim, and Goles could also support the punctuated equilibrium model, but their evidence suggests some limitations for the three cases they researched. Occasionally, organizations undergo revolutionary changes, but they may be inhibited by structural and cultural inertia. In that case, the transformation may not extend to all necessary dimensions. As a consequence, long convergent periods may sometimes be characterized by low alignment between the strategy and the environment. Additionally, post-revolutionary adjustments which either reinforce the move or take a step back may follow. Apart 208
See Romanelli and Tushman: Organizational Transformation as Punctuated Equilibrium, 1994, pp. 1156–1158. 209 See Wischnevsky, J. Daniel and Fariborz Damanpour: Punctuated Equilibrium Model of Organizational Transformation: Sources and Consequences in the Banking Industry, in: Research in Organizational Change and Development, Vol. 15 (2004), p. 219. Brown and Eisenhardt as well as Greenwood and Hinings also support that revolutionary change may not be the only way organizations evolve. See Brown, Shona L. and Kathleen M. Eisenhardt: The Art of Continuous Change: Linking Complexity Theory and Time-Paced Evolution in Relentlessly Shifting Organizations, in: Administrative Science Quarterly, Vol. 42 (1997), No. 1, p. 25; and Greenwood, Royston and C. R. Hinings: Understanding Strategic Change: The Contribution of Archetypes, in: Academy of Management Journal, Vol. 36 (1993), No. 5, pp. 1073–1075. 210 See Wischnevsky and Damanpour: Punctuated Equilibrium Model of Organizational Transformation, 2004, pp. 222 and 204. 211 See Grégoire, Denis A., Pamela S. Barr and Dean A. Shepherd: Cognitive Processes of Opportunity Recognition: The Role of Structural Alignment, in: Organization Science, Vol. 21 (2010), No. 2, pp. 424 and 426. 212 See Wischnevsky, J. Daniel and Fariborz Damanpour: Radical strategic and structural change: occurrence, antecedents and consequences, in: International Journal of Technology Management, Vol. 44 (2008), No. 1/2, p. 65.
48
B Deterministic and Voluntaristic Theories of Organizational Change
from environmental shifts, sustained low performance, and influential outside stakeholders, they depicted new leadership and a transformation of perception to be drivers of reorientations, the latter of which had received little attention earlier.213 The study also confirms findings of an earlier study indicating the existence of a combination of change triggers. It already depicted a threat from environmental shifts, low performance, and the entry of a new client group to be able to initiate radical organizational change. She also showed that change benefits organizational performance, a finding opposed to the assumptions of the organizational ecology literature.214 Apart from empirical studies, Sastry formalized the punctuated equilibrium theory and tested it by simulation. As in her system dynamics model organizations did not terminate periods of reorganization, she proposed two extensions; she added the monitoring of the current fit with the environment as well as a trial period. These measures avoid a further reaction if performance has not yet recovered by the time the organization has adapted due to low experience shortly after a reorientation.215 The trial period is consistent with empirical evidence of post-revolutionary adjustments some time after the original transformation as found by Sabherwal, Hirschheim, and Goles.216 The original theory of the punctuated equilibrium model is therefore not exhaustive, and empirical evidence is not unambiguously in favor of the punctuated equilibrium model. Revolutionary change may not be the only way organizations evolve. For instance, Brown and Eisenhardt rather detect continuous than punctuated changes among six firms in the context of multiple-product innovation, and Feldman showed that routines can also be a source of continuous change in organizations.217 Nevertheless, the punctuated equilibrium approach serves as a valuable example of a theory of multiple drivers of change. Many factors responsible for the evolution of organizations are included in this model. It combines elements of adaptation to the environment with hindrances to this process such as inertia, resistance, and routines, and it also comprises elements of managerial choice. It provides insight into how organizational decision-making takes place by opening the black box of inertia and management choice indicating how perception of managers can contribute to an organization’s evolution. 213
See Sabherwal, Rajiv, Rudy Hirschheim and Tim Goles: The Dynamics of Alignment: Insights from a Punctuated Equilibrium Model, in: Organization Science, Vol. 12 (2001), No. 2, pp. 193–194. 214 See Haveman, Heather A.: Between a Rock and a Hard Place: Organizational Change and Performance Under Conditions of Fundamental Environmental Transformation, in: Administrative Science Quarterly, Vol. 37 (1992), No. 1, pp. 72–73. 215 See Sastry, M. Anjali: Time and tide in organizations: Simulating change processes in adaptive, punctuated, and ecological theories of organizational evolution, Ph.D. Thesis, Alfred P. Sloan School of Management, Massachusetts Institute of Technology, Boston, Massachusetts 1995, pp. 141–145; and Sastry: Problems and Paradoxes in a Model of Punctuated Organizational Change, 1997, pp. 257–265. 216 See Sabherwal, Hirschheim and Goles: The Dynamics of Alignment, 2001, p. 195. 217 See Brown and Eisenhardt: The Art of Continuous Change, 1997, p. 32; and Feldman, Martha S.: Organizational Routines as a Source of Continuous Change, in: Organization Science, Vol. 11 (2000), No. 6, pp. 620–621.
B.III Cognition and Attention as Drivers and Restraints of Alteration
49
B.III Cognition and Attention as Drivers and Restraints of Alteration B.III.1 Perception of the Environment Through a Cognitive Managerial Lens Managerial cognition, perception, and attention are repeating concepts not only in the punctuated equilibrium approach, but also in some of the other organization theories described. They provide further insight into how multiple views of drivers of change interrelate. Cognition describes the process of perceiving and interpreting the environment and translating this information into strategic choice and action.218 Recently a number of researchers have developed cognitive approaches to strategy and organizational change which particularly unite the behavioral theory, the influence of the environment, inertia and routines. They mold theories of organizations and their environment.219 Thus, while situated in the adaptive behavioral theory, the cognitive view draws from organizational ecology, strategic choice, and the punctuated equilibrium approach. It is able to enrich knowledge on organizational change because it bridges the gap between the different perspectives.220 It covers divergent aspects, as it was suggested by proponents of the multi-paradigm focus. The cognitive approach seeks to understand the processes in organizations leading both to organizational prosperity and decline, and to failure to change. This research tradition assumes that information about managerial cognition results in an improved understanding of managers’ mental models, of how they are influenced by environmental factors, and how these models lead to organizational decisions and actions.221 It is often analyzed by in-depth case studies. The concept of cognition links to psychological and behavioral research of individuals.222 The psychological stream of literature focuses on the influence of personali218
See Kaplan, Sarah, Fiona Murray and Rebecca Henderson: Discontinuities and senior management: assessing the role of recognition in pharmaceutical firm response to biotechnology, in: Industrial and Corporate Change, Vol. 12 (2003), No. 2, p. 203. 219 E.g. Gavetti, Giovanni and Jan W. Rivkin: On the Origin of Strategy: Action and Cognition over Time, in: Organization Science, Vol. 18 (2007), No. 3, p. 420.; Rajagopalan and Spreitzer: Toward a Theory of Strategic Change, 1997, pp. 67–74; and Walsh, James P.: Managerial and Organizational Cognition: Notes from a Trip Down Memory Lane, in: Organization Science, Vol. 6 (1995), No. 3, p. 280. 220 See Huff, Anne S., Joseph W. Huff and Pamela S. Barr: When Firms Change Direction, New York, NY 2000, pp. 29 and 38; Mintzberg, Henry, Joseph Lampel and Bruce Ahlstrand: Strategy Safari: A Guided Tour Through The Wilds of Strategic Management, New York, NY 1998, p. 151. 221 See Schwenk, Charles R.: The Cognitive Perspective on Strategic Decision Making, in: Journal of Management Studies, Vol. 25 (1988), No. 1, p. 53. 222 E.g. Gigerenzer, Gerd, Peter M. Todd and the ABC Research Group: Simple Heuristics That Make Us Smart, New York 1999, pp. 20–21; Kahneman, Daniel, Jack L. Knetsch and Richard H. Thaler: Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias, in: Journal of Economic Perspectives, Vol. 5 (1991), No. 1, p. 199; and Kahneman, Daniel and Amos
50
B Deterministic and Voluntaristic Theories of Organizational Change
ties and individual characteristics on cognition. Since this dissertation focuses on the group and organizational level, these characteristics will only be of importance as they shape aggregate decision-making and behavior. Two major cognitive approaches can be distinguished that bridge the gap between the individual and organizational level. The first once regards cognition as a meaning creation process. An example is Weick’s theory of organizational perception that focuses on the interpretation process by which humans make sense of their environment.223 The second approach investigates how decision-makers interpret information by their mental models and interact with the environment.224 The focus of this chapter will be on the latter one as it is closely related to the literatures discussed before as well as to organizational decision-making. Building on the behavioral theory as well as its more recent integrative developments, Kim, Payne, and Tan argue for the integration of decision-makers’ perception of environmental stimuli, cognition and affect, strategic decision-making, and organizational adaptation. In particular they focus on the individual level of cognitiveaffective environmental interpretation.225 Parting from deterministic views of environmental influence, the authors adopt an interpretive view of strategic choice. They suggest that whether environmental determinism prevails or whether there is room for choice and organizational diversity depends on the perception of internal and environmental conditions. Taking the cognitive aspects into consideration gives a more comprehensive model of organizational change processes.226 Decision-makers often revert to intuitive and heuristic decision-making when they are not able to draw from normative decision rules or experience, for example during times of quick environmental change.227 A combined cognitive lens also helps explain a lack of response to environmental threats. A threat restricts information processing, evoking a welllearned behavioral response in individuals, groups, and organizations. This inhibits a correct interpretation of and reaction to environmental stimuli.228 Affective states often underlie these cognitive outcomes. They are emotional experiences such as success, failure, and fear, but also previous experiences and decisions that are associated Tversky: Prospect Theory: An Analysis of Decision under Risk, in: Econometrica, Vol. 47 (1979), No. 2, pp. 274–280. 223 See Weick: Sensemaking in Organizations, 1995, p. 14; and Weick, Karl E.: Making Sense of the Organization, Malden, MA [et al.] 2001, pp. 196 and 243. 224 See Demers: Organizational Change Theories, 2007, p. 41. 225 See Kim, Kong-Hee, G. Tyge Payne and James A. Tan: An examination of cognition and effect in strategic decision making, in: International Journal of Organizational Analysis, Vol. 14 (2006), No. 4, p. 278. 226 See ibid., pp. 280 and 283–286. 227 See Eisenhardt, Kathleen M.: Making Fast Strategic Decisions in High-Velocity Environments, in: The Academy of Management Journal, Vol. 32 (1989), No. 3, pp. 544–545. See also Tversky, Amos and Daniel Kahneman: Judgment under Uncertainty: Heuristics and Biases, in: Science, Vol. 185 (1974), No. 4157, p. 1124. 228 See Staw, Sandelands and Dutton: Threat Rigidity Effects in Organizational Behavior, 1981, pp. 502–503 and 517–519.
B.III Cognition and Attention as Drivers and Restraints of Alteration
51
with strong emotions. They can help or hinder a rational decision-making process.229 These attachments may support decision-making, but they may also cause good ‘intuitive’ feelings for the wrong decision. While these individual psychological and cognitive concepts provide a basis, the focus will now shift back to determinants of the organizational cognition which are inter-subjectively comparable and thus affect entire groups. In managerial and organizational cognition, research on cognitive inertia has become important. Cognition and cognitive inertia also relates to mental models, the focus of attention, and identity. Mental models are important for the understanding of the dynamic processes of organizational adaptation. Here, a shared mental model represents an individually as well as collectively held simplified representation of the decision-makers’ world.230 Information processing is biased by the availability of information, selective perception, conservatism in forecasting, wishful thinking, and more. This biased information processing lays ground for the interpretive lenses of mental models.231 Chosen strategies, organizational performance and efficiency are a reflection of values and mental models of the dominant coalition in the organization.232 The problem is that mental models may not only filter and bias information, but are difficult to revise once they are firmly established, even in the presence of adverse information.233 Individuals’ cognitive limitations in many cases become obvious through neglecting feedback in decision-making. Sterman finds evidence that people regularly misinterpret or ignore the impact of their decisions on their wider environment. Based on his study he proposes an anchoring and adjustment heuristic which more realistically describes human behavior.234 The failure to perceive feedback relationships and the inability to revise mental models based on performance feedback might also explain 229
See Finkelstein, Sydney, Jo Whitehead and Andrew Campbell: Think Again: Why Good Leaders Make Bad Decisions and How to Keep it From Happening to You, Boston 2009, pp. 73–106. 230 See Hodgkinson, Gerard P.: Cognitive Inertia in a Turbulent Market: the Case of UK Residential Estate Agents, in: Journal of Management Studies, Vol. 34 (1997), No. 6, p. 922. Kim makes aware that the conceptions of mental model as well as of managerial cognition are somewhat problematic. There exist many similar notions like collective cognition, shared mental model, schema, belief structure, and else. Different terms that are used to describe ‘shared mental model’ hold different assumptions about location and form of the model. Kim points to the difficulty arising e.g. from different interpretations of ‘shared mental model’ as collectively or individually held. See Kim, Hyunjung: In search of a mental model-like concept for grouplevel modeling, in: System Dynamics Review, Vol. 25 (2009), No. 3, pp. 212–215. 231 See Schwenk uses the terms cognitive maps and schemata instead of mental model, but the concepts are similar. See Schwenk: The Cognitive Perspective on Strategic Decision Making, 1988, pp. 43–45. 232 See Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 51; and Hambrick and Mason: Upper Echelons, 1984, pp. 193–196. 233 See Schwenk, Charles R.: Cognitive Simplification Processes in Strategic Decision-Making, in: Strategic Management Journal, Vol. 5 (1984), No. 2, p. 116. 234 See Sterman: Modeling Managerial Behavior, 1989, p. 334. See also Tversky and Kahneman: Judgment under Uncertainty, 1974, pp. 1128–1130.
52
B Deterministic and Voluntaristic Theories of Organizational Change
research done by Hodgkinson. He conducted an empirical analysis of the effect of cognitive inertia on companies’ ability to re-conceptualize their business. In a longitudinal study of the UK real estate industry, which underwent radical environmental change, he found that despite these significant changes in market conditions mental models remained stable over time. As a result of stable mental models, organizations continue with strategies that have become inappropriate in a changed environment.235 The lag of change in the mental model behind the changes in the market demonstrates high levels of cognitive inertia. Mental models also explained a lack of reaction at the Polaroid Company. Based on a case study, Tripsas and Gavetti showed how managerial cognition shapes adaptation processes and inertia in the face of radical technological change. In so doing they aimed at explaining why radical transformations in the organizational environment often lead to established firm failure.236 Prior research was concerned with the evolution and adaptation difficulties of managers’ mental models and with how capabilities and routines limit adaptation.237 Tripsas and Gavetti’s inductive case study of the Polaroid Corporation’s reaction to possibilities of digital imaging portrays a strong belief in the ‘accustomed’ business model that prevented a smooth transition from instant to digital photography. Polaroid was a clear technology-driven company that spent large amounts of money on research and development. It believed in two things, first, that the market will accept a new product if it is technologically superior to the competitors’ ones, and second, that money is made on the film as the dependent replacement good instead of the camera hardware. The latter represented the building block and underlying belief of the business model. It was also the reason why—although the company had invested in digital imaging and had the leading technology by about 1990—the management team did not believe in the digital camera’s success and did not bring the product to the market. Polaroid did not sell digital cameras until 1996, when 40 companies had already entered the market.238 Tripsas and Gavetti attribute the failure to develop new capabilities to the strong managerial beliefs that remained unaltered. The business model proved to be the main source of inertia. The managers’ mental model and cognitive representation of their environment fundamentally contributed to the observed path dependency. Top management team turnover then helped initiate change because it also changed the belief system
235
See Hodgkinson: Cognitive Inertia in a Turbulent Market, 1997, pp. 936 and 938. See Tripsas and Gavetti: Capabilities, Cognition, and Inertia, 2000, p. 1147. 237 E.g. Barr, Pamela S., J. L. Stimpert and Anne S. Huff: Cognitive Change, Strategic Action, and Organizational Renewal, in: Strategic Management Journal, Vol. 13 (1992), No. -, Summer Special Issue, pp. 32–33; and Cohen, et al.: Routines and Other Recurring Action Patterns of Organizations, 1996, p. 663. 238 See Gavetti, Giovanni: Cognition and Hierarchy: Rethinking the Microfoundations of Capabilities' Development, in: Organization Science, Vol. 16 (2005), No. 6, pp. 601–602; and Tripsas and Gavetti: Capabilities, Cognition, and Inertia, 2000, pp. 1150–1157. 236
B.III Cognition and Attention as Drivers and Restraints of Alteration
53
which was necessary to develop new capabilities and to be successful in a different competitive landscape.239 Gavetti also developed an agent-based axiomatic model that was informed by the qualitative case study of digital imaging at Polaroid. It combines routine-based and path-dependent behavior with cognition. Differences in information processing, e.g. by individuals in varying hierarchical levels, lead to unequal cognitive representations and capabilities. It confirms cognition as an important aspect in the development of capabilities in organizations.240 Myopic self-focus in decision-making is a further concept related to cognitive inertia. Building on the behavioral school’s ideas of limited search for information and limited informational attention, Moore, Oesch, and Zietsma researched the concept with the help of a qualitative study. This study was based on interview data of 54 entrepreneurs who had recently decided for or against founding a new business. The authors discovered that the founding decision depends on internal factors, i.e. on perception of personal ability as well as factors internal to the organization. Conditions in the external environment such as industry and market characteristics do not appear to have high influence on the decision.241 They supported their findings in a second, experimental study of market entry. Here as well, participants based their choice on their perceived subjective ability, representing an internal focus; they ignored their competitors’ entry rates, i.e. their environment.242 Tripsas’ longitudinal and inductive analysis of the digital photography company Linco in the first decade after founding also provides insight into cognitive inertia. She links the punctuated equilibrium approach with cognitive research and puts special emphasis on organizational identity. Tripsas comprehends identity as a filter through which organizational members perceive and interpret external stimuli. Internal identity guides strategic decisions and represents a shared understanding about the core business.243 In her view, the dynamics between identity and technological change remain neglected in the literature. Therefore she studied the filtering effect of identity and the process of occasional subsequent technological change. She examined feedback processes of change dynamics over time. Her sources reveal a threephase transition of the organization. In the first phase, self-reinforcing dynamics maintain the original identity. Following a performance decrease and CEO renewal, the corporate strategy changed, indicating the second period of identity ambiguity and organizational transformation. In the last phase, the company converged on a new identity and on a respective self-reinforcing dynamic. In the view of Tripsas, iner239
See Tripsas and Gavetti: Capabilities, Cognition, and Inertia, 2000, pp. 1158–1159. See Gavetti: Cognition and Hierarchy, 2005, p. 612. 241 See Moore, Don A., John M. Oesch and Charlene Zietsma: What Competition? Myopic SelfFocus in Market-Entry Decisions, in: Organization Science, Vol. 18 (2007), No. 3, p. 444. 242 See ibid., p. 449. 243 See Tripsas: Technology, Identity, and Inertia through the Lens of 'The Digital Photography Company', 2009, p. 441. 240
54
B Deterministic and Voluntaristic Theories of Organizational Change
tia arises, first, from a cognitive filter placed on environmental stimuli. Identity serves as a filter that has an influence on perception, interpretation, technical choices, and actions taken. Identity-challenging environmental developments may remain unnoticed. Second, it also arises from a positive feedback loop which reinforces consistent elements of the system.244 The reinforcing dynamics concerning identity and action represent a significant impediment to change. The study is thus consistent with the literature on organizational inertia as well as with the punctuated equilibrium model and evidences that young organizations can exhibit cognitive inertia as well. Furthermore the study is a valuable illustration of the importance of feedback and of the consideration of dynamics over time. Approaches that take into consideration cognitive aspects often portray multiparadigm views. They bridge the gap between different theories and drivers of change. The aforementioned examples disclose that existing resource commitments, behavioral routines, capabilities, and cognitive frames all represent inertia. They thus limit an organization’s response to environmental change in the same way as they enhanced success under stable conditions before.245 Durable organizational change does not only derive from an adaptation process, but involves a second learning cycle by which cognitions and attitudes change as well.246 This learning results in more appropriate reactions to environmental drivers of change and in better results arising from intentional initiatives by management.
B.III.2 Selective Attention to Issues and Stakeholders „[… T]heories of decision making are often better described as theories of attention or search than as theories of choice. They are concerned with the way in which scarce attention is allocated.”247 As the above quote reveals, for theories of decision-making in general not only cognition and choice, but also attention is an important notion. Attention can be a means of measuring cognition.248 It is the degree of consciousness of and orientation to something. One of the first approaches in which attention was discussed is the 244
See ibid., pp. 447 and 454. See Kaplan, Sarah: Cognition, Capabilities, and Incentives: Assessing Firm Response to the Fiber-Optic Revolution, in: Academy of Management Journal, Vol. 51 (2008), No. 4, p. 690; Tripsas: Technology, Identity, and Inertia through the Lens of 'The Digital Photography Company', 2009, p. 441; and Tripsas and Gavetti: Capabilities, Cognition, and Inertia, 2000, p. 1158. 246 See Schimmel, Remco and Dennis R. Muntslag: Learning Barriers: A Framework for the Examination of Structural Impediments to Organizational Change, in: Human Resource Management, Vol. 48 (2009), No. 3, p. 413. 247 March, James G., with the assistance of Chip Heath: A Primer on Decision Making: How Decisions Happen, New York, NY [et al.] 1994, p. 10. 248 See Eggers, J. P. and Sarah Kaplan: Cognition and Renewal: Comparing CEO and Organizational Effects on Incumbent Adaptation to Technical Change, in: Organization Science, Vol. 20 (2009), No. 2, p. 462. 245
B.III Cognition and Attention as Drivers and Restraints of Alteration
55
Carnegie behavioral school and in particular the writings by Herbert Simon. In his seminal book on organizational behavior Simon laid out how organizations and their people behave and how they change and adapt over time. He concentrates on managerial decision-making and on the analysis of factors influencing these decisions. In the view of Simon, organizations are complex networks of attention processes.249 One important aspect of his theory is the bounded rationality of individuals which he considers to be a result of limited human perception and attention. Stimuli direct human attention to selected aspects of a situation.250 This also means competing aspects of the situation that the stimulus does not refer to may be excluded and not be perceived. Choices would be different if these competing aspects were taken into consideration. Vice versa, Simon also focuses on how structure and cognitive capabilities influence perception. Attention then also has an influence on the intake of subsequent stimuli, so that the role of attention is the channelizing of stimuli. The allocation of attention affects the selection of information and consequently has an effect on the resulting decision and behavior.251 “Attention and behavior, once initiated in a particular direction, tend to persist in that direction for a considerable amount of time.”252 The ways in which individuals distribute attention become in large parts habitual, and the span of attention as well as habitual skills and behaviors limit docility. Consequently, to a great extent the limits to rationality result from limits in the area of attention.253 Thus Simon describes causal relationships related to attention which work in two different directions. While he demonstrates how stimuli shape attention, he also indicates how habituated attention shapes the intake of new stimuli. Taken together the causal relations represent a feedback loop of attention and stimuli which develops into a habit and has an influence on bounded rational decisions in organizations. Likewise, later research by the behavioral theory school has touched attention, particularly its sources routines and bounded rationality. Related research concentrating on how attention develops by coincidence will not be considered here as it lacks causal relationships and thus controllability.254 In this dissertation, it is not understood as a short-term attention span, e.g. of visual perception as it is analyzed in 249
See Simon: Administrative Behavior, 1976, pp. 220–222. See ibid., p. 90. 251 See March: A Primer on Decision Making, 1994, pp. 23–24; and Simon: Administrative Behavior, 1976, pp. 90–93. 252 Simon: Administrative Behavior, 1976, p. 95. 253 Directly see ibid., pp. 90–91 254 See Ocasio, William: Towards an Attention-based View of the Firm, in: Strategic Management Journal, Vol. 18 (1997), No. -, Summer Special Issue, p. 188. Work by Cyert and March as well as by March and Simon focuses on routines and bounded rationality, whereas their later work centers on random aspects, described by Cohen, March, and Olsen as well as Weick. See Cyert and March: A Behavioral Theory of the Firm, 1963, p. 35; and March and Simon: Organizations, 1958, pp. 152 and 154. Vs. Cohen, Michael D., James G. March and Johan P. Olsen: A Garbage Can Model of Organizational Choice, in: Administrative Science Quarterly, Vol. 17 (1972), No. 1, p. 8; and Weick, Karl E.: The Social Psychology of Organizing, Reading, MA [et al.] 1979, p. 32. 250
56
B Deterministic and Voluntaristic Theories of Organizational Change
cognitive psychology.255 Rather, it represents medium to long-term orientation of decision-makers in an organization. According to Ocasio—who later took up and enhanced Simon’s and the behavioral schools attention-based view—attention is able to explain how organizations adapt to a changing environment. He sees reason for the failure to adjust in the failure to attend to relevant issues.256 His view distinguishes three aspects. First, there is the focus of attention, meaning which issues and answers decision-makers attend to. Selective distribution of attention limits perception to a section of reality. Second, he mentions the impact of the context or situation, called situated attention. The environment has an influence on what a decision-maker focuses on. Third, the structural and social environment affects the decision, this being called the structural distribution of attention. Organizational structures have an influence on the assessment and legitimacy of issues and answers taken into consideration. Here, past decisions become important influencing factors on the current decision context.257 Ocasio considers his point of view as an alternative to both deterministic theories such as the organizational ecology approach as well as to Child’s theory of strategic choice. He explicitly states that adequate change as well as insufficient change and inertia can be a result of situated attention. Referring to the debate in organization theory of the determination and reconciliation of inertia and adaptation as discussed in chapter B.II, Ocasio enhances the reconciliatory view by the incorporation of attention because this can conciliate between the different views. The attention-based view has the advantage of explaining the different reactions to a changing environment— reactions such as both inertia and change—which theories of rational choice or deterministic theories of environmental selection are not capable of doing.258 Here the view bears similarity with Hambrick and Mason’s multi-paradigm lens including individual cognitive and attentive elements that lead to inertia, adaptation, and strategic choice.259 Lieberman and Montgomery state as well that established firms’ inertia
255
See Lamme, Victor A. F.: Why visual attention and awareness are different, in: Trends in Cognitive Sciences, Vol. 7 (2003), No. 1, p. 12; Tomasello, Michael: Joint Attention as Social Cognition, in: Moore, Chris and Philop J. Dunham (Ed.): Joint Attention, Hillsdale, NJ 1995, p. 124. 256 See Ocasio: Towards an Attention-based View of the Firm, 1997, p. 204; and Ocasio, William and John Joseph: An Attention-Based Theory of Strategy Formulation: Linking Micro- and Macroperspectives in Strategy Processes, in: Gabriel Szulanski, Joe Porac and Yves Doz (Ed.): Strategy Process, Advances in Strategic Management, Vol. 22, 2005, pp. 56–57. 257 See Ocasio: Towards an Attention-based View of the Firm, 1997, pp. 190–193 and 199. 258 See ibid., pp. 188, 202 and 204. Concerning the debate of inertia vs. adaptation and the reconcilability of theories see Adler and Borys: Materialism and Idealism in Organizational Research, 1993; pp. 666–675; Astley and Van de Ven: Central Perspectives and Debates in Organization Theory, 1983, pp. 253–255; and Barnett, William P. and Glenn R. Carroll: Modeling Internal Organizational Change, in: Annual Review of Sociology, Vol. 21 (1995), No. 1; pp. 217–219. 259 See Hambrick and Mason: Upper Echelons, 1984, pp. 194–195.
B.III Cognition and Attention as Drivers and Restraints of Alteration
57
may be a result of cognitive elements or “inattention to shifts in technology or customer needs.”260 Sullivan enriched the concept which Ocasio calls situated attention. His study of airline industry regulations showed that attention is directed to the area in which problems arise, other areas are disregarded. The likelihood of attending to and solving problems is higher when performance is low. In general, urgency created by new problems increases problem solving attention to them. Then, under the influence of problem urgency, attention is channeled more towards simple solutions. The nature of environmental stimuli and the processing capacity of the organization together affect attention to problems and how fast they are resolved.261 Sullivan’s study makes clear that attention is closely related to cognitive limitations and that it is part of what constitutes organizational bias. Concepts of attention often assume attention and interpretation to be different concepts, but so intertwined that it does not make sense to separate them.262 An analysis of airline deregulation follows this assumption. It shows that attention can change rapidly. Changes in top management team characteristics are associated with attentional shifts, and differences in attention translate to strategy.263 Further studies confirmed this in the area of innovation and market entry. An empirical test of the effect of CEO attention on innovation outcomes discovered that CEOs’ focus on the future and on the environment increases the speed of detecting new technological developments. The future and environmental focus enhances the speed of reaction to environmental developments as well as of new product development.264 Further research analyzed the effects of CEO attention to new technologies and to industries in which technologies emerge. The authors investigated their impact on the speed of adoption and entry into a new market. Results support that attention to existing technologies is correlated with a later entry and introduction of a new product. Attention to the new industry is associated with more rapid change. For instance, the authors of the study state that for one standard deviation increase in words used in documents that relate to the new technology, a firm is 138 percent more likely to in-
260
Lieberman, Marvin B. and David B. Montgomery: First-Mover Advantages, in: Strategic Management Journal, Vol. 9 (1988), p. 54. 261 See Sullivan, Bilian Ni: Competition and Beyond: Problems and Attention Allocation in the Organizational Rulemaking Process, in: Organization Science, Vol. 21 (2010), No. 2, pp. 442, 444, and 446. 262 See Cho, Theresa S. and Donald C. Hambrick: Attention as the Mediator Between Top Management Team Characteristics and Strategic Change: The Case of Airline Deregulation, in: Organization Science, Vol. 17 (2006), No. 4, p. 454; LaBerge, David: Attentional Processing: The Brain's Art of Mindfulness: Perspectives in Cognitive Neuroscience, 1995, p. 64; and Ocasio: Towards an Attention-based View of the Firm, 1997, p. 189. 263 See Cho and Hambrick: Attention as the Mediator Between Top Management Team Characteristics and Strategic Change, 2006, pp. 461 and 464–466. 264 See Yadav, Manjit S., Jaideep C. Prabhu and Rajesh K. Chandy: Managing the Future: CEO Attention and Innovation Outcomes, in: Journal of Marketing, Vol. 71 (2007), No. 4, pp. 93–97.
58
B Deterministic and Voluntaristic Theories of Organizational Change
troduce the new product.265 Hence, while the study revealed the importance of attention for strategy, it concludes with statements that do not take into consideration possible changes over time in the relationships. The authors of the attention-based perspective emphasize decision-makers’ inaptness to sufficiently adapt to the environment because they cannot adequately distribute their attention to different issues due to their bounded rationality. They do not necessarily attend to those problems in the environment which indicate that transformations within the organization are necessary. Similar to Simon, they understand attention as the focus of a decision-maker on stimuli. Importantly, attention is understood as the issues and answers organizational decision-makers focus on. Issues might be problems, opportunities, and threats while answers are proposals, routines, and procedures.266 Yet, managers may not only attend to issues but also to important groups, i.e. to stakeholders. However, groups have been given little attention. Only stakeholder management theory accomplishes this inclusion of groups. It concentrates on the mutual interference of groups and the organization. It commonly defines a stakeholder as “any group or individual who can affect of is affected by the achievement of the organization’s objectives.”267 Instrumental, normative, and empirical stakeholder research can be distinguished. Based on the above definition of stakeholders, research in the instrumental realm of stakeholder theory is based on the assumption that the influence of stakeholders on the organization calls for good management of these groups in order to maximize profits.268 Corporate prosperity then depends on some fit between the stakeholders’ and the organization’s interests and values.269 In the intrinsic (normative) stakeholder model, mere profit maximization makes way for a normative commitment to values. Here, the interests of stakeholders are accepted to have intrinsic value, so that they become a part of corporate strategy and are considered prior to competing strategic considerations.270 There is empirical research, particularly studies that try to substantiate a relationship between stakeholder orientation and performance, but they are highly influenced by the normative theory. Little research has been done in the descriptive and empirical realm of how managers deal with stakeholders.271 Sparse exemptions deal with 265
See Eggers and Kaplan: Cognition and Renewal, 2009, pp. 471–472. See Ocasio: Towards an Attention-based View of the Firm, 1997, p. 189. See also Cho and Hambrick: Attention as the Mediator Between Top Management Team Characteristics and Strategic Change, 2006, p. 454. 267 Freeman, R. Edward: Strategic Management: A Stakeholder Approach, Boston, London 1984, p. 46. See also Donaldson, Thomas and Lee E. Preston: The Stakeholder Theory of the Corporation: Concepts, Evidence, and Implications, in: The Academy of Management Review, Vol. 20 (1995), No. 1, pp. 72 and 76. 268 See Berman, Shawn L., et al.: Does Stakeholder Orientation Matter? The Relationship between Stakeholder Management Models and Firm Financial Performance, in: Academy of Management Journal, Vol. 42 (1999), No. 5, p. 491 269 See Freeman: Strategic Management, 1984, p. 107. 270 See Berman, et al.: Does Stakeholder Orientation Matter, 1999, p. 494. 271 See ibid., p. 488. 266
B.III Cognition and Attention as Drivers and Restraints of Alteration
59
determinants as well as effects of organizational attention. Mitchell, Agle, and Wood’s work represents an exception as it adds to stakeholder theory by providing a descriptive theory of stakeholder attention. The concept is not within the tradition of organization theory discussed so far, but rather a concept of stakeholder theory. It explains the conditions under which managers consider certain groups (with their claims) as important. According to the authors of that view, attention derives from stakeholder salience which they understand as the priority that they attach to a stakeholder claim.272 It is a function of the perceived group’s power, the perceived urgency expressed by the stakeholders, and their perceived legitimacy, as can also be seen in Figure B-7. Importantly, stakeholder attention is not fixed. When determinants of salience change, this concept allows for alterations in the management team’s attention to stakeholders over time. pcvd stakeholder power pcvd urgency (expressed by stakeholders)
moderating effect of management + salience + +
+
attention
pcvd legitimacy (within society)
Figure B-7: Determinants of stakeholder attention according to Mitchell, Agle, and Wood273 Agle, Mitchell, and Sonnenfeld empirically tested the theory developed by Mitchell, Agle, and Wood. They regarded the cumulative perceived stakeholder attributes— power, legitimacy, and urgency—as important for stakeholder salience.274 They found support for their hypotheses, and this finding suggests that stakeholder attributes have an effect on the degree to which managers prioritize competing stakeholders. In their sample, urgency is the best predictor for stakeholder salience, and the moderating effect of normative management and values prove to be non-significant.275 A number of studies is based on the same framework and provided empirical support for it.276 A study of manufacturing firms also found that urgency has greatest in272
See Mitchell, Ronald K., Bradley R. Agle and Donna J. Wood: Toward a Theory of Stakeholder Identification and Salience: Defining the Principle of Who and What Really Counts, in: The Academy of Management Review, Vol. 22 (1997), No. 4, pp. 853–854. 273 See ibid., pp. 865–868. 274 See Agle, Bradley R., Ronald K. Mitchell and Jeffrey A. Sonnenfeld: Who Matters to CEOs? An Investigation of Stakeholder Attributes and Salience, Corporate Performance, and CEO Values, in: Academy of Management Journal, Vol. 42 (1999), No. 5, p. 509. 275 See ibid., pp. 518–520. Here Agle, Mitchell, and Sonnenfeld tested the moderating effect of normative values. 276 E.g. Agle, Bradley R., Ronald K. Mitchell and Jeffrey A. Sonnenfeld: A Report on Stakeholder Attibutes and Salience, Corporate Performance, and CEO Values, in: Logsdon, Jeanne M., Donna J. Wood and Lee E. Benson (Ed.): Research in Stakeholder Theory, 1997–1998: The Sloan Foundation Minigrant Project, Toronto, Canada 2000, p. 47; Knox, Simon and Colin
60
B Deterministic and Voluntaristic Theories of Organizational Change
fluence on stakeholder salience among managers, followed by legitimacy and power.277 Three further studies showed that power is the strongest indicator for salience and attention.278 In another research project, a positive effect of power could be confirmed, but there were different results for group and claim legitimacy and no significant effect of urgency on salience.279 According to this, so far no unambiguous scheme of interference can be derived. The different and particularly opposed results are suggestive of dynamics within power, urgency, and legitimacy. Consequently, the strength of the determinants of attention may not be constant over time. The theoretical concepts and empirical studies of stakeholder attention were concerned so far with determinants of attention. But similar to attention to issues, researchers have started to analyze the effect of attention on change as well. A study of failing firms and their matched pairs revealed differences in organizations’ reactions to demand-decline crises. While top managers of surviving firms pay more attention to their external environment, those of failing firms become inward-oriented and have a tendency to neglect their outer environment. The authors emphasize the importance of causal relationships, but also argue that further research is needed in order to establish causal directions.280 A further empirical test of managerial cognition and attention constructed causal maps based on data of 22 firms in four industries. The study supported that top managers’ attention focus influences the speed of the strategic response to major changes in the organization’s environment. In a quickly changing industry, organizations give more attention to competitors, suppliers, and customers than in a slowly changing industry.281 An additional study analyzed the relationship between the group affiliation of an organization, its focus of attention, and its responsiveness to performance feedback. It helped align the behavioral theoGruar: The Application of Stakeholder Theory to Relationship Marketing Strategy Development in a Non-profit Organization, in: Journal of Business Ethics, Vol. 75 (2007), No. 2, p. 127; Magness, Vanessa: Who are the Stakeholders Now? An Empirical Examination of the Mitchell, Agle, and Wood Theory of Stakeholder Salience, in: Journal of Business Ethics, Vol. 83 (2008), No. 2, pp. 187–190; and Winn, Monika I.: Building Stakeholder Theory with a Decision Modeling Methodology, in: Business & Society, Vol. 40 (2001), No. 2, pp. 159–160. 277 See Fernández Gago, Roberto and Mariano Nieto Antolín: Stakeholder salience in corporate environmental strategy, in: Corporate Governance, Vol. 4 (2004), No. 3, p. 71. 278 See Harvey, Brian and Anja Schaefer: Managing Relationships with Environmental Stakeholders: A Study of U.K. Water and Electricity Utilities, in: Journal of Business Ethics, Vol. 30 (2001), No. 3, p. 253; Parent, Milena M. and David L. Deephouse: A Case Study of Stakeholder Identification and Prioritization by Managers, in: Journal of Business Ethics, Vol. 75 (2007), No. 1, pp. 9, 13 and 17; and Ryan, Lori Verstegen and Marguerite Schneider: Institutional Investor Power and Heterogeneity: Implications for Agency and Stakeholder Theories, in: Business Society, Vol. 42 (2003), No. 4, p. 416. 279 See Eesley, Charles and Michael J. Lenox: Firm responses to secondary stakeholder action, in: Strategic Management Journal, Vol. 27 (2006), No. 8, p. 777. 280 See D'Aveni, Richard A. and Ian C. MacMillan: Crisis and the Content of Managerial Communications: A Study of the Focus of Attention of Top Managers in Surviving and Failing Firms, in: Administrative Science Quarterly, Vol. 35 (1990), No. 4, pp. 641, 645 and 650. 281 See Nadkarni, Sucheta and Pamela S. Barr: Environmental Context, Managerial Cognition, and Strategic Action: An Integrated View, in: Strategic Management Journal, Vol. 29 (2008), No. 13, pp. 1414–1416.
B.IV Need for a Dynamic Feedback View of Organizational Change
61
ry of organizations with organizational ecology research on inertia. Business group affiliation and attention alter the balance between the prevalence of inert and adaptive forces.282 While not part of the attention literature, stakeholders or the influence of these groups have come up in several studies before. In particular the punctuated equilibrium model has been connected to stakeholder influence. Theoretical work considered the influence of outside individuals to support a transitional period. Empirical studies observed that influential stakeholders—such as lending banks, consulting firms, firms entering the market—or new client groups can trigger radical changes and reorientations in organizations.283 There are different theoretical frameworks that relate to managerial and organizational attention and cognition. The stakeholder approach examines the determinants of organizational and managerial attention and its effects on performance. It considers the influence of stakeholder attributes on the attention of decision-makers. The cognitive perspective on attention is rather concerned with attention to stimuli. It tries to generate answers to the question of the relevance of attention for adaptation and inertia. While theoretical frameworks are different, they both add complementary insights. Research that bridges between several theoretical approaches by cognition and attention has strongly increased in importance in the beginning of the 21st century. Yet the dynamics between influencing factors and processes have not yet been exhaustively studied and deserve particular ‘attention’.
B.IV Need for a Dynamic Feedback View of Organizational Change Since change in organizations may be influenced by many factors, different schools of thought and their drivers of organizational change have been presented. Early theories in particular emphasize the prevalence of the environment. They expect organizations to adapt to outside demands in order to ensure their survival and high performance. Whereas rational scientific theories suppose that adaptation takes place perfectly and rationally, in particular the behavioral theory presumes a bounded rational adaptation process, but it still considers the environment as the driving force of an organization’s evolution. The prevalence of the environment as driver of change renders the adaptive theories deterministic. Concerning environmental determinism the organization ecology approach is even more extreme as it assumes almost complete organizational inertia. Organizations which do no longer fit a niche are selected 282
See Vissa, Balagopal, Henrich R. Greve and Wei-Ru Chen: Business Group Affiliation and Firm Search Behavior in India: Responsiveness and Focus of Attention, in: Organization Science, Vol. 21 (2010), No. 3, pp. 708–709. 283 See Gersick: Revolutionary Change Theories, 1991, pp. 27–28; Haveman: Between a Rock and a Hard Place, 1992, pp. 57–58 and 72–73; and Sabherwal, Hirschheim and Goles: The Dynamics of Alignment, 2001, p. 194.
62
B Deterministic and Voluntaristic Theories of Organizational Change
out. Routines contributing to stability and experience create path dependent dynamics inhibiting a deviation from a taken direction. But empirical research proved ambiguous showing no clear answer to the question whether—by the impact of routines—change induces further change or reduces the likelihood of subsequent change. In opposition to these deterministic points of view, the strategic choice approach regards decision-makers as capable of making deliberate strategic choices. This view in particular concentrates on the management team as the driving force of change in organizations. At the same time, it does not completely deny environmental impact. Hence it does not limit its focus on just one driver of change as previous approaches have done, but to some extent allows for multiple drivers. A number of theories are limited to one or few aspects, such as only one determinant of change. Since organizational change is a multi-faceted phenomenon, many researchers argued that one theoretical lens does not do justice to its complexity. They call for a multi-paradigm view in order to generate a more complete picture that captures several aspects or drivers of change. They argue that reality is neither completely deterministic nor does it allow for complete choice, but that both deterministic and deliberate elements are present in organizational decision-making and evolution. This statement needs to be qualified by noting that there are voices against the legitimacy of theory integration due to their different assumptions. Although theories may not be completely compatible, they can inform and complement each other. One preeminent example of a multi-paradigm theory that considers several drivers of change is the punctuated equilibrium model which represents a radical change theory. Simultaneously it assumes strategic choice in an adaptation process, inhibited by routines and inertia. Similarly, cognition and attention theories often are integrative approaches, and one could say that the consideration of cognitive factors to some extent is present already in the punctuated equilibrium literature. Managerial cognition strongly relates to the notion of organizational inertia and it considers influences from both the environment and the management team. It also closely relates to the adaptive and routine-based theories. Human cognitive models and decision-making may be incomplete due to limited attention. The attention-based perspective shows that attention serves as a habituated filter on issues and answers, and the stakeholder perspective reveals that this filter even extends to the perception of stimuli from groups. There may be different drivers of change. Yet they do not indicate the relation between causes and consequences, and what kind of behavior results from this over time. Richardson and Weick point out that much research simply considers unidirectional causation and distinguishes between independent and dependent variables. This only reveals half-truths. Factors are related interdependently, and researchers therefore need to think in circles.284 For this reason it is important to gain information about causal relationships between different forces and about how they dynamically 284
See Richardson, George P.: Feedback Thought in Social Science and Systems Theory, 2 Ed., Waltham, MA 1999, p. 2; and Weick: The Social Psychology of Organizing, 1979, p. 86.
B.IV Need for a Dynamic Feedback View of Organizational Change
63
interact. This helps understand the resulting behavior over time. Even in view of the question whether change induces further change, a consideration of causal relationships, the incorporation of balancing and reinforcing feedback processes is necessary. While much research neglects causality and feedback, many authors have called for a causal view, for the integration of feedback, or have already applied such a lens in their research.285 The world is complex and non-linear, and especially the field of organizational change has been shown to be multi-faceted. Accordingly, research methods need to capture this complexity. Additionally, previous statistical analyses have shown ambiguous results. A case study is expected to explain such ambiguous results and also shed light into the complexity of organizational change processes. As it will be analyzed by the system dynamics methodology that includes causal modeling and simulation, it will give deep insight into causal relationships, feedback, and dynamics. It is a long-term qualitative study of an organization facing technological change in its environment and of the resulting managerial decisions. More concretely, the New York Stock Exchange’s reaction to an ongoing trend in the securities market towards automation of trading will be studied.
285
E.g. Adler and Borys: Materialism and Idealism in Organizational Research, 1993, pp. 665–669; Astley and Van de Ven: Central Perspectives and Debates in Organization Theory, 1983, p. 266; Buchanan and Dawson: Discourse and Audience, 2007, pp. 669 and 682; Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, pp. 70–72; Larsen and Lomi: Resetting the clock, 1999, pp. 408–410; Rajagopalan and Spreitzer: Toward a Theory of Strategic Change, 1997, p. 66; de Rond and Thietart: Choice, chance, and inevitability in strategy, 2007, p. 547; Sastry: Problems and Paradoxes in a Model of Punctuated Organizational Change, 1997, p. 240; and Tripsas: Technology, Identity, and Inertia through the Lens of 'The Digital Photography Company', 2009, pp. 447–448.
65
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the New York Stock Exchange The question of what are the drivers and inhibitors of change as well as of the repetitive momentum proposition will be analyzed in a case study of the New York Stock Exchange (NYSE). Since the organization’s institutionalization, it conducted business in the matching of incoming buy and sell orders for securities. Its trading system will be analyzed because it has always been a core capability. Today the company also offers derivatives trading, is in the listing business, and sells market data, but these divisions will not be discussed since they were added just recently or developed out of the trading business. The NYSE’s trading mechanism has its roots in the beginnings of the organization in the 18th century when traders held auctions on Wall Street. In 1792, 24 of them signed an agreement from which the NYSE originates, stipulating to give each other preference in the trading of specific securities and to charge a fixed commission. The NYSE’s and many other stock exchanges’ trading systems were auction markets in which a market maker conducts an auction for specific securities and allows for negotiation over the price. At the NYSE, the market maker used to be called the specialist and is now referred to by designated market maker.286 The specialist conducts an auction and ‘makes the market’ for the securities of about a handful of corporations. This means he brings together supply and demand and always gives traders a price at which brokers can buy and sell shares of the securities he is responsible for. Specialists are required to provide a fair and orderly auction process, match orders, and to step in with their own capital to minimize imbalances and stabilize prices. They help provide liquidity and minimize volatility.287 Specialists as well as brokers maintain close relationships to the NYSE, but are separate entities and not part of its workforce. Customer orders are executed either against the specialist quote, or prices are negotiated with the specialist, or among floor brokers on the NYSE trading floor. Since the specialist matches orders manually on the trading floor, the trading system is called manual trading or floor trade. Figure C-1 shows a graphical representation of the participants and interactions in the manual floor trading mechanism. A buyer of e.g. 1000 shares of the Walt Disney Corporation contacts his broker A, e.g. his bank. This firm is in contact with a floor broker D, a person who is physically situated on the NYSE trading floor. This person then walks to the specialist who ‘makes the market’ in the Walt Disney Corporation securities and who matches the buy order against his own quote or against a sell interest from another floor broker E if that gives the customer a better price. The general process of floor trading used to 286 287
Appendix A provides a glossary of the most important financial terms used in this dissertation. See NYSE Euronext Inc.: Types of Members, no date-n, electronic source (website no longer active); and NYSE Group Inc., Annual Report pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 for the fiscal year ended December 31, 2006, No. 001-32829, New York 2007, pp. 4 and 33. Today, as specialists are called designated market makers their responsibilities have slightly changed.
N. Zimmermann, Dynamics of Drivers of Organizational Change, DOI 10.1007/978-3-8349-6811-1_3, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011
66
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
be similar at many stock exchanges, but at the NYSE specialists had a more central role and their obligations used to be stricter than at other trading venues. The frequent bidding of many floor brokers around the specialist booth created the energetic trading process symbolic for the New York Stock Exchange. Buyer
Seller
Broker A
Broker B
Floor Broker D
Specialist
Floor Broker E
The Floor
Figure C-1: Trade participants and interactions in floor trade The last decades have witnessed a tendency towards more and more electronic, meaning automated trading in both U.S. and worldwide securities markets.288 In electronic trading (e-trade) a computer takes over the formerly manual matching of buy and sell orders. This renders the entire trading floor—i.e. the grey part in Figure C-1—redundant. Many researchers and industry experts alike expected stock exchanges which had not yet done so to further automate and adapt to the transformation in their environment.289 They thus predicted an adaptive behavior of exchanges, driven by deterministic forces as outlined by adaptive organization theory. Most noticeable, while many believed in the automation trend, few expected the New York Stock Exchange to go along. Even during the 1990s, many people believed that the exchange was threatened by extinction.290 Thus while the industry seemed to follow the pattern of adaptation, the NYSE was expected to be selected out by environmental evolution, as suggested by the organizational ecology literature.
288
See McAndrews, James and Chris Stefanadis: The Emergence of Electronic Communication Networks in the U.S. Equity Markets, in: Federal Reserve Bank of New York: Current Issues in Economics and Finance, Vol. 6 (2000), No. 12, p. 3. Electronic trading is understood here as the full automation of the trading process which eliminates the trading floor as well as the need for people on the floor. 289 See Clemons, Eric K. and Bruce W. Weber: Information Technology and Screen-Based Securities Trading: Pricing the Stock and Pricing the Trade, in: Management Science, Vol. 43 (1997), No. 12, pp. 1695 and 1706; Feldman, Stuart: Electronic Marketplaces, in: IEEE Internet Computing, Vol. 4 (2000), No. 4, p. 95; Handa, Puneet, Robert A. Schwartz and Ashish Tiwari: The Economic Value of a Trading Floor: Evidence from the American Stock Exchange, in: The Journal of Business, Vol. 77 (2004), No. 2, p. 354; and Picot, Arnold, Christine Bortenlänger and Heiner Röhrl: The Automation of Capital Markets, in: Journal of ComputerMediated Communication, Vol. 1 (1995), No. 3, 1995, electronic source. 290 See Abolafia, Mitchel Y.: Making Markets: Opportunism and Restraint on Wall Street, Cambridge, MA and London 1996, pp. 130–151; Naidu, G. N. and Michael S. Rozeff: Volume, volatility, liquidity and efficiency of the Singapore Stock Exchange before and after automation, in: Pacific-Basin Finance Journal, Vol. 2 (1994), No. 1, p. 24; and Welles, Chris: Is it time to make the Big Board a black box?, in: Business Week (1990), No. 3144, issued 5 Feb. 1990, p. 74.
C.I A Method Mix for Studying Change in Organizations and Especially at the NYSE
67
The case is particularly suitable for gaining insight into the research questions of the determinants of change because the NYSE represents an organization in the struggle between adaptation and inertia. It finally underwent a reorientation that resulted in adaptation to the environment. This move was overdue, but at the same time somewhat surprising for those who expected the stock exchange’s demise. The organization’s situation exhibited many ambiguities, and there were forces pulling the organization in opposing directions of stability and change. The case is able to show how these different kinds of pressure develop and interact, and the analysis of the underlying dynamics helps to understand the stock exchange’s reaction and decision-making. The example of the NYSE demonstrates the difficulties which are associated with a change process. It will be informative for determinants of organizational change because, on the one hand, it represents a peculiar example as the organization was an island of a manual trading system in an environment that increasingly automated its trading system. On the other hand, the NYSE represents a typical example of a ‘dinosaur’ that has been successful not only for decades, but for centuries. It is representative of a class of organizations confronted with changes in the market. The dynamics of its evolution will thus help gain insight into dynamics of organizational change and momentum. It may also shed light on similar processes in related cases and industries, such as the U.S. automobile manufacturers which were threatened by Japanese competitors, mini computer manufacturers which had to cope with the development of personal computers, or film camera producers facing digital photography. A more detailed description of the developments in the securities market and of the NYSE’s reaction will be given in this chapter. It will apply organizational rather than financial perspectives on the exchange’s trajectory. The information will then be used to build, simulate, and analyze a causal model of the New York Stock Exchange’s behavior. But before, the following sub-chapter will concentrate on the research design employed and elaborate the case study and modeling methodology.
C.I
A Method Mix for Studying Change in Organizations and Especially at the New York Stock Exchange
In order to capture the complexity and dynamic behavior of how electronic trading at the NYSE was upheld and then unfolded, a long-term investigation of the kinds of pressure pulling at the organization is required. This investigation needs to incorporate causes and consequences distant in time and in space, thus requiring a broad view of the important elements in the entire system. As Table C-1 summarizes, this research will thus follow a research paradigm that particularly focuses on causal relationships. The analysis will include the simultaneous consideration of individual and collective elements as they are interdependent.291 The system dynamics methodolo291
See Reihlen, Markus, Thorsten Klaas-Wissing and Torsten Ringberg: Metatheories in Management Studies: Reflections Upon Individualism, Holism, and Systemism, in: M@n@gement,
68
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
gy is well suited for the analysis of complex systems. It is a theory of decision-making and change through time, and it is also a feedback theory of the structure and behavior of social systems.292 The nature of system dynamics places it in the middle of the epistemological continuum between an objective view of the world and subjective construction.293 It is thus able to take several points of view and investigate organizations and their behavior from different angles. This makes it particularly suitable for the analysis of deterministic vs. deliberate change. The system dynamics methodology will be used to craft a model of the interaction of driving and inhibiting processes of change. It will help elucidate objective elements of reality as well as how humans make sense of it through perception and interpretation. It has proven useful for the analysis of bounded rational decision-making and misperceptions of feedback.294 The methodology of qualitative case study research will complement the system dynamics investigation. The combination of methodologies in this dissertation follows Crotty’s suggestion of a case-related development of methodology and a combination of qualitative and quantitative methods.295 System dynamics, qualitative and case study research can fruitfully be integrated as they share similar assumptions as well as consistent goals of theory building and explanation.296 System dynamics and case study methodologies will be elaborated further as well as their respective methods which include data collection and analysis, text coding, system dynamics modeling and validation.
Vol. 10 (2007), No. 3, p. 65. Research can be described, first, by concrete research methods used, second, by their underlying methodology, third, by the theoretical perspective revealing the philosophy and assumptions behind the methodology, and fourth, by the implied epistemology, i.e. the theory of knowledge and knowledge generation. See Crotty, Michael: The foundations of social research: Meaning and perspective in the research process, London [et al.] 1998, pp. 6–8. 292 See Milling, Peter: Der technische Fortschritt beim Produktionsprozeß: Ein dynamisches Modell für innovative Industrieunternehmen, Wiesbaden 1974, pp. 55–56; and Richardson: Feedback Thought in Social Science and Systems Theory, 1999, p. 296. 293 Directly see Lane, David C.: Rerum cognoscere causas: Part I—How do the ideas of system dynamics relate to traditional social theories and the voluntarism/determinism debate?, in: System Dynamics Review, Vol. 17 (2001), No. 2, pp. 113–114. See also Bowen, Michael G.: System dynamics, determinism, and choice: Toward a reconsideration of the image of "systems man", in: System Dynamics Review, Vol. 10 (1994), No. 1, pp. 88–89. 294 E.g. Salge, Markus: Struktur und Dynamik ganzheitlicher Verbesserungsprogramme in der industriellen Fertigung: Ein systemdynamisches Modell zur nachhaltigen Gestaltung des Wandels in Industrieunternehmen, Hamburg 2009, pp. 37–43; and Sterman: Modeling Managerial Behavior, 1989, pp. 334–336. 295 See Crotty: The foundations of social research, 1998, pp. 13–14. 296 See Kopainsky, Birgit and Luis Felipe Luna-Reyes: Closing the Loop: Promoting Synergies with Other Theory Building Approaches to Improve System Dynamics Practice, in: Systems Research and Behavioral Science, Vol. 25 (2008), No. 4, pp. 474–476; Luna-Reyes, Luis Felipe and Deborah Lines Andersen: Collecting and analyzing qualitative data for system dynamics: methods and models, in: System Dynamics Review, Vol. 19 (2003), No. 4, pp. 284 and 288–291.
C.I A Method Mix for Studying Change in Organizations and Especially at the NYSE
Epistemology
Reconciliation of objective and subjective epistemology
Theoretical Perspective
Systemic research paradigm, in particular: feedback theory of the structure and behavior of social systems
Methodology
System dynamics and qualitative case study research
Methods
System dynamics modeling and validation, case study data collection and analysis, text coding
69
Table C-1: Research design
C.I.1
Contribution of Studying Change with System Dynamics and Case Study Methodologies
System dynamics is an aggregate theory that focuses on decision-making processes and aims at improving human judgment. It concentrates on problematic behavior or interesting phenomena occurring in reality.297 According to Milling, the theory’s main building blocks are the focus on, first, the structure of social systems, second, their inherent feedback character, and third, the construction of formal models and their simulation as the process of learning about the feedback structure of social systems.298 The structure of systems as one of the main features of system dynamics tells how systemic parts are related to one another.299 It causes their resulting behavior. Yet, the aggregate view of system dynamics leaves room for individual interpretation and intervention into the system, even within a structure that determines behavior.300 The assumption of accumulations, delays, and nonlinearities in social systems are further central structural elements of social systems.301 Structure reveals the reg297
See Forrester: Industrial Dynamics, 1961, pp. 13–18; and Forrester, Jay W.: Principles of Systems, Cambridge, MA 1968c, ch. 1.2. For a view of SD as a way to explain interesting phenomena see e.g. Größler, Andreas: System Dynamics Modelling as an Inductive and Deductive Endeavour: Comment on the Paper by Schwaninger and Grösser, in: Systems Research and Behavioral Science, Vol. 25 (2008b), No. 4, p. 468. Examples of the modeling of phenomena can be found in Repenning, Nelson P. and John D. Sterman: Nobody Ever Gets Credit for Fixing Problems that Never Happened, in: California Management Review, Vol. 43 (2001), No. 4, p. 66; Rudolph, Jenny W. and Nelson P. Repenning: Disaster Dynamics: Understanding the Role of Quantity in Organizational Collapse, in: Administrative Science Quarterly, Vol. 47 (2002), No. 1, p. 3; and Sastry: Problems and Paradoxes in a Model of Punctuated Organizational Change, 1997, p. 257. 298 See Milling, Peter: Leitmotive des System-Dynamics-Ansatzes, in: Wirtschaftswissenschaftliches Studium (WiSt), Vol. 13 (1984), No. 10, p. 507. For an analysis of system dynamics as a feedback theory see Richardson: Feedback Thought in Social Science and Systems Theory, 1999, pp. 296–317. 299 Directly see Forrester: Industrial Dynamics, 1961, p. 15. 300 For a discussion of this point see Lane: Rerum cognoscere causas: Part I, 2001, p. 112. 301 See Forrester, Jay W.: Industrial Dynamics—After the First Decade, in: Management Science, Vol. 14 (1968b), No. 7, pp. 404–409; and Lane, David C.: Should System Dynamics be Described as a 'Hard' or 'Deterministic' Systems Approach?, in: Systems Research and Behavioral Science, Vol. 17 (2000), No. 1, p. 4; and Sterman, John D.: All models are wrong: reflec-
70
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
ularity and stability of human behavior in a system.302 Yet, it will also be used to investigate and explain fundamental change. The assumption of feedback in the structure of systems constitutes the second central element of system dynamics theory. In closed loop feedback thinking, endogenous variables are assumed to cause respective behavior; the environment is not left out, but is of secondary importance. The feedback view denies unidirectional thinking. It often describes circles in which a decision results in an action which then forms the basis of decision-making in the future.303 The focus is on the endogenous creation of behavior patterns. Simulation analysis, constituting the third building block besides feedback and structure, is a convenient tool for analyzing highly complex models in social science.304 Davis, Eisenhardt, and Bingham define simulation as the use of computer software to model the underlying logic of real world processes. They consider it as particularly useful when the research question addresses intertwined and longitudinal processes, timing effects (delays) and nonlinearities, as well as existing tensions. Here, they take the tension of inertia vs. change as an example, which is also the focus of this dissertation.305 The use of simulation as a method for theory generation is not undisputed. Simulation methods may be regarded as lacking importance, based on unrealistic assumptions, being inaccurate, or overly complex.306 In order to avoid these possible shortcomings, modeling assumptions will be made explicit, and the system’s inherent complexity will be kept in a homomorphous way,307 and they will be comprehensible through the elicitation of feedback loops and scenario analyses. Simulation will be helpful in specifying boundary conditions, enhancing the model theory’s internal validity and adding value by creative experimentation in a testing laboratory.308 It provides a process perspective. Further, it enables an imtions on becoming a systems scientist, in: System Dynamics Review, Vol. 18 (2002), No. 4, p. 506. 302 See Rogers: Diffusion of Innovations, 2003, p. 24. 303 See Forrester: Industrial Dynamics, 1961, pp. 14 and 51; Forrester, Jay W.: Industrial Dynamics—A Response to Ansoff and Slevin, in: Management Science, Vol. 14 (1968a), No. 9, p. 610; and Morgan, Gareth and Linda Smircich: The Case for Qualitative Research, in: The Academy of Management Review, Vol. 5 (1980), No. 4, pp. 493–497. See also Wilden, Anthony: System and Structure: Essays in Communication and Exchange, 2. Ed., London 2003, p. 39. 304 See Cohen, Kalman J. and Richard M. Cyert: Computer Models in Dynamic Economics, in: The Quarterly Journal of Economics, Vol. 75 (1961), No. 1, p. 118. 305 See Davis, Jason P., Kathleen M. Eisenhardt and Christopher B. Bingham: Developing Theory though Simulation Methods, in: Academy of Management Review, Vol. 32 (2007), No. 2, pp. 481, 485, and 495. 306 See ibid., pp. 480 and 496. 307 See Milling: Systemtheoretische Grundlagen zur Planung der Unternehmenspolitik, 1981, p. 97. 308 See Davis, Eisenhardt and Bingham: Developing Theory though Simulation Methods, 2007, pp. 495 and 497. For the use and usefulness of simulation see also Forrester: Industrial Dynamics, 1961, pp. 18 and 23–24.
C.I A Method Mix for Studying Change in Organizations and Especially at the NYSE
71
proved understanding and model of dynamic behavior, structure, and feedback in the real system. Models are used to capture structure and feedback and to conduct simulation tests. A model is a simplified representation of the real system, and in particular it captures accumulations, how these accumulations are influenced by rates, and important feedback relationships that cause the behavior observed in reality. It is a homomorphous, meaning a structure-preserving mapping. It represents a dynamic theory that is able to explain behavior that is observed in the real world.309 A system dynamics model is thus empirical and descriptive.310 In the view of Forrester, a model may not only be a representation of one system, meaning a single case or organization, but it can be a theory of a particular type or class of systems—also called a midrange theory. It offers insight into the generic structure and behavior patterns of this type of system.311 Mid-range theory is in the middle of the continuum between a grand theory and the minor theory of a single case because it explains phenomena occurring in the entire class of systems. The system dynamics modeling process is a dynamic process of theory building. Using the system dynamics methodology and theory building, researchers are able to contribute to a rigorous research process.312 The process of theory building and modeling in system dynamics is highly iterative and involves repeated steps of the construction of a quantified model of the real system as well as its validation.313 The 309
See Cohen and Cyert: Computer Models in Dynamic Economics, 1961, p. 113; Forrester, Jay W.: System dynamics, systems thinking, and soft OR, in: System Dynamics Review, Vol. 10 (1994), No. 2/3, p. 253; and Kopainsky and Luna-Reyes: Closing the Loop, 2008, pp. 474 and 483. 310 See Bertrand, J. Will M. and Jan C. Fransoo: Operations management research methodologies using quantitative modeling, in: International Journal of Operations & Production Management, Vol. 22 (2002), No. 2, pp. 249–251; and Größler, Andreas, Jörn-Henrik Thun and Peter M. Milling: System Dynamics as a Structural Theory in Operations Management, in: Production and Operations Management, Vol. 17 (2008), No. 3, p. 378. 311 Forrester: Industrial Dynamics—A Response to Ansoff and Slevin, 1968, p. 607. For an assessment of system dynamics as a useful methodology to develop mid-range theory see Kopainsky and Luna-Reyes: Closing the Loop, 2008, pp. 474–476; and Schwaninger, Markus and Stefan Grösser: System Dynamics as Model-Based Theory Building, in: Systems Research and Behavioral Science, Vol. 25 (2008), No. 4, p. 461. Schwaninger and Grösser also exemplify how to develop mid-range theory based on SD modeling and a case study. 312 Milling, Peter M.: Afterword, in: Systems Research and Behavioral Science, Vol. 25 (2008), No. 4, pp. 581–582. 313 For this two-step account of the modelling process see Schwaninger and Grösser: System Dynamics as Model-Based Theory Building, 2008, p. 449. Luna-Reyes and Andersen provide a summary of different conceptualizations of the modelling process often including problem articulation, conceptualization, model formulation, analysis and validation. See Luna-Reyes and Andersen: Collecting and analyzing qualitative data for system dynamics, 2003, pp. 275–279 as well as Randers, Jørgen: Guidelines for Model Conceptualization, in: Randers, Jørgen (Ed.): Elements of the System Dynamics Method, Cambridge, MA [et al.] 1980, particularly p. 135; Richardson, George P. and Alexander L. Pugh III: Introduction to System Dynamics Modeling with DYNAMO, Cambridge, MA [et al.] 1981, p. 62; and Sterman, John D.: Business Dynamics: Systems Thinking and Modeling for a Complex World, Boston [et al.] 2000, p. 86.
72
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
model construction and formulation process involves several sub-steps which will be described in the following chapter.
C.I.2
The Process of Model Conceptualization Supported by Case Study Research Methods
As a first step, model conceptualization and construction involves problem articulation in which the reference mode, i.e. the problematic or interesting behavior of the real world, is outlined. In this case the interesting phenomenon is the long phase in which the NYSE did not change followed by a radical transformation. The problem description also includes the identification of key variables. The second step is the formation of a dynamic hypothesis which is a “working theory of how the problem arose” that explains the dynamic behavior observed in the real world (reference mode).314 Although the outcome is a quantitative model, this is a highly qualitative process as it relates to the identification of relationships and structure.315 Third, based on this information the model is formalized. This system dynamics modeling process is in line with theory building processes in general. Particularly in the early steps, the model building process can be supported by methods typical for qualitative research and case studies. There are obvious similarities to methods used particularly in grounded theory, case study research, and content analysis.316 These methodologies focus on processes and are concerned with how patterns of behavior unfold over time.317 They also share iterative methods of data collection (theoretical sampling), analysis and theory generation. Analysis is done by the rigorous coding of data by which the researcher looks for repeating codes, concepts, and categories in the data which are then investigated in their relationship to generate theory.318 The theory follows a hypothesis-building, integrative process involving the search for frequently recurring evidence in the data for the
314
Sterman: Business Dynamics, 2000, pp. 86 and 95. See Forrester: Industrial Dynamics, 1961, pp. 57–58. Luna-Reyes and Andersen: Collecting and analyzing qualitative data for system dynamics, 2003, pp. 276–277; and Richardson and Pugh III: Introduction to System Dynamics Modeling with DYNAMO, 1981, p. 19. 316 See Kopainsky and Luna-Reyes: Closing the Loop, 2008, pp. 474–476; and Luna-Reyes and Andersen: Collecting and analyzing qualitative data for system dynamics, 2003, pp. 284–286. Grounded theory is a qualitative methodology that uses rigorous data analysis such as coding to develop theory. Content analysis is a mostly quantitative methodology that measures word frequencies and/or deductive coding according to defined rules. 317 See Bryman, Alan and Emma Bell: Business Research Methods, 2. Ed., Oxford, UK 2007, p. 418. 318 See Glaser, Barney G. and Anselm L. Strauss: The Discovery of Grounded Theory: Strategies for Qualitative Research, Chicago 1967, pp. 105–113; as well as Bryant, Anthony and Kathy Charmaz (Ed.): The SAGE Handbook of Grounded Theory, Los Angeles [et al.] 2007, pp. 12–13; and particularly Strauss, Anselm L. and Juliet M. Corbin: Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory, 2. Ed., Thousand Oaks, CA [et al.] 1998, pp. 12–14. 315
C.I A Method Mix for Studying Change in Organizations and Especially at the NYSE
73
emerging theory.319 By a dynamic hypothesis and model formulation process, emerging categories and themes provide information on subjective meaning and on the richness of relationships and structure. Categories and themes can be variables as well as dynamic behaviors and even policies. Through the provision of a record of structure formulation, categories and themes may help interpreting simulation results also in the policy analysis phase.320 Both case study research and the system dynamics process of model and theory building draw on qualitative data. According to Sterman, interviews, participant observation and archival data are a common source of information for developing system dynamics models. Particularly a mix of qualitative and quantitative data proves useful so that the modeler is able to triangulate from many sources.321 The data for the case study of the New York Stock Exchange’s transformation were thus collected from multiple origins. First, information was taken from scientific journal articles. Second further written data sources were used such as magazine and newspaper articles, press releases, books, annual reports, SEC filings, published transcripts of conference calls with managers and industry experts, transcripts of analyst meetings, published interviews and management speeches. Third, the NYSE Facts and Figures, made available on the NYSE Euronext website, provided useful time series data of variables relating to NYSE trading, ownership, customers, and the U.S. securities market. In order to capture cultural information, fourth, the Exchanges Blog, a weblog published by the New York Stock Exchange, provided information through entries of NYSE staff, reprint of management speeches, magazine and newspaper articles, and through comments to the entries.322 Since NYSE employees moderate the weblog and since customers make comments and/or involve in a discussion with NYSE staff, the weblog illustrates both the New York Stock Exchange’s and some its customer groups’ point of view. Fifth, two preliminary and four main interviews with individuals of the NYSE, its related parties, and customers supported the general understanding and helped the analysis of events and data. 319
See Bryant and Charmaz (Ed.): The SAGE Handbook of Grounded Theory, 2007, p. 16; Strauss and Corbin: Basics of Qualitative Research, 1998, pp. 22 and 136–137. Many researchers also regard qualitative research or grounded theory as a purely inductive methodologies, but this research follows the above-mentioned authors who consider it to have deductive and abductive elements, meaning that it builds structural hypotheses which are tested for every bit of data and that causes are inferred from consequences. 320 See Luna-Reyes and Andersen: Collecting and analyzing qualitative data for system dynamics, 2003, pp. 284–286 and 288–291. They actually consider the model formulation process as being inductive. See also Schwaninger and Grösser: System Dynamics as Model-Based Theory Building, 2008, pp. 448–450. 321 See Sterman: Business Dynamics, 2000, p. 157. See also Glaser and Strauss: The Discovery of Grounded Theory, 1967, pp. 61–62 and 111–112; and Lewis and Grimes: Metatriangulation, pp. 272 and 685–686. 322 See http://exchanges.nyse.com/. For a discussion of advantages and disadvantages of the use of data from weblogs for conducting research see Zimmermann, Nicole S. and Peter M. Milling: Data Generation from Weblogs in Qualitative Research and System Dynamics, in: Proceedings of the 17th International Annual EurOMA Conference 2010, pp. 1–3 and 9.
74
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
The NYSE weblog was used to gain insight about issues and culture in a time of change. Its entries from late 2006 until March 2008 were analyzed with a follow up analysis until March 2010. The weblog was used in order to get a more thorough understanding of the ‘softer’ elements. A qualitative analysis and respective coding of entries (topics) of the NYSE weblog plus their following developing discussions was helpful particularly for the elucidation of management and customer points of view. They proved important particularly for the construction of the managerial elements of the model. The emergence of the internet has led to new types of data and has enlarged the possible sources of material. By analyzing the Exchanges Weblog, this dissertation makes use of such a new internet data format. These electronic sources are indicated as such in the footnotes. Their web addresses and access dates can be found in the bibliography. In the cases in which the weblog reprinted content, the original source will be cited. It has become increasingly common to use internet sources for academic research.323 Consumer comments on the internet provide rich and meaningful information for researchers.324 Concerning the quality of the sources for research, it needs to be said that respondents may self-select and the internet sample may not be random. However, several comparative studies of online and paper-and-pencil surveys found very similar results, providing support for the usefulness of the analysis of internet sources.325 On the one hand, the internet can be used as a technical tool aiding the acquisition of knowledge. On the other hand, it can itself be understood as a social system that serves as a mirror of society. This means researchers can abstract from the data on websites or weblogs and infer to other categories such as psychological, cultural, or organizational aspects. It has even been suggested to use weblog research for detecting attention and emotions.326
323
See Dholakia, Nikhilesh and Dong Zhang: Online Qualitative Research in the Age of ECommerce: Data Sources and Approaches, in: Forum: Qualitative Social Research, Vol. 5 (2004), No. 2, pp. 2–3; and Dresing, Thorsten and Udo Kuckartz: Neue Datenquellen für die Sozialforschung: Analyse von Internetdaten, in: Kuckartz, Udo, Heiko Grunenberg and Thorsten Dresing (Ed.): Qualitative Datenanalyse: computergestützt, 2. Ed., Wiesbaden 2007, p. 144. 324 See Allen, Gove N., Dan L. Burk and Gordon B. Davis: Academic Data Collection in Electronic Environments: Defining Acceptable Use of Internet Resources, in: MIS Quarterly, Vol. 30 (2006), No. 3, p. 600; and Romano Jr., Nicholas C., et al.: A Methodology for Analyzing WebBased Qualitative Data, in: Journal of Management Information Systems, Vol. 19 (2003), No. 4, pp. 242–243. 325 See Eaton, Judy and C. Ward Struthers: Using the Internet for Organizational Research: A Study of Cynicism in the Workplace, in: CyberPsychology & Behavior, Vol. 5 (2002), No. 4, pp. 310–312; Gosling, Samuel D., et al.: Should We Trust Web-Based Studies? A Comparative Analysis of Six Preconceptions About Internet Questionnaires, in: American Psychologist, Vol. 59 (2004), No. 2, p. 102; and Stanton, Jeffrey M.: An Empirical Assessment Of Data Collection Using The Internet, in: Personnel Psychology, Vol. 51 (1998), No. 3, pp. 720–721. 326 See Ammer, Daniela: Die Umwelt des World Wide Web: Bildung für nachhaltige Entwicklung im Medium World Wide Web aus pädagogischer und systemtheoretischer Perspektive, Ph.D. Thesis, Department of Social and Behavioral Sciences, University of Tübingen, Tübingen 2010, p. 215; Dresing and Kuckartz: Neue Datenquellen für die Sozialforschung, 2007, p. 146; and Glance, Natalie S., Matthew Hurst and Takashi Tomokiyo: BlogPulse: Automated Trend Discovery for Weblogs, in: Proceedings of the WWW 2004 Workshop on the Weblogging Eco-
75
C.I A Method Mix for Studying Change in Organizations and Especially at the NYSE
Ideas and concepts that showed up repeatedly in the data and particularly in multiple data sources were used as model variables. Table C-2 provides an exemplary overview of several variables or constructs that were included in the model. In many cases important variables represent a stock in the model, indicating that the respective variable accumulates over time. They are indicated by a box around their name. Variable/concept name
Journal articles
Newspaper articles, reports, …
NYSE time series data
Exchanges Weblog
Interviews
(Extent/fraction of) electronic trading
yes
yes
yes
yes
yes
(No./fraction of) institutional customers
yes
yes
yes
yes
Speed, time to execution
yes
yes
Market share
yes
yes
yes
yes
Market quality
yes
yes
outdated
yes
Specialist participation
yes
yes
yes
yes
yes
yes
yes
Power/pressure of floor firms/specialists
yes
some
yes
Culture, cultural aspects
yes
yes
yes
Stakeholder/customer orientation
yes
yes
yes
…
Table C-2: Data sources for variable derivation The data sources were used to infer concepts and more concrete variables, but they also offered information about causal relationships and dynamic behavior as Table C-3 exemplifies. Text passages helped the elicitation of structure and its connection to dynamics. Model conceptualization followed a process in which many fragments of data and information were used to generate the most plausible causal explanation for themes and the observed behavior. The data was used to establish causal relationships and build the stock-and-flow structure of the system. From this structure, feedback loops were derived and will be described in relation to their polarity, i.e. in relation to whether they exhibit reinforcing or balancing (equilibrating) behavior. Qualitative case study and system dynamics methods were employed jointly in the present analysis. The triangulation from quantitative and qualitative data and the use of the NYSE weblog supported the piecemeal and iterative construction of the system dynamics model. system: Aggregation, Analysis and Dynamics, New York, NY 2004, pp. 6–8. Further applications include Wang, Shan, et al.: A Literature Review of Electronic Marketplaces: Themes, Theories and an Integrative Framework, in: Information Systems Frontiers, Vol. 10 (2008), No. 5, p. 566.
76 Variable NYSE market share
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
Definition “percent[age] of the daily trades in shares issued by companies listed on the NYSE … [and] executed at NYSE.”327
Causal relationships
Dynamic behavior
“After two centuries of ା ା (market share ՜ … ՜ market share) dominance, the exchange was steadily losing “The New York Stock Exchange is ground, surrendering cruplanning to unveil a series of proposed cial market share to elecnew rules Friday in an effort to turn tronic rivals; a decade around its falling market share. The ago, … 80 percent of the changes by the NYSE Euronext unit daily trades in shares iswould further overhaul the role that sued by companies listed "specialist" traders play.”328 on the NYSE were exeି (market share ՜ pressure to change) cuted at NYSE. In October that figure was 42.8 percent. … NYSE Euronext gained share in September, reversing a year-long slide. […] That’s still well below the 59.8 percent it logged at the end of 2007, but at least the decline has been reversed.”327 “Liquidity begets liquidity.”
“Last September, the NYSE was trading 1.4 billion shares per day on average, giving it a market share of 44 percent. This month, the NYSE is trading 1.2 billion shares per day, giving it a market share of 29 percent.”329
Time to execution
“The decline (in the NYSE's share of trading in NYSE-listed issues) underscores the urgency of adopting the Hybrid Market, a project that will cut the “time to com- time to complete a trade to less than a 330 plete a trade” second, just like all-electronic rivals. NYSE floor traders take an average of nine seconds, Thain has said.”330
Monthly time series data from the NASDAQ website.
“At the NYSE itself, the average speed for a market order fell from 12 seconds in April 2005 to twoି tenths of a second in April (market share ՜ pressure to change 2007.”331 ା ՜ e-trade (associated with the Hybrid Market)) ି
(e trade ՜ time to execution) 327
McGee, Suzanne: Up Off the Floor, in: Institutional Investor, Vol. 42 (2008), No. 10, p. 40. These numbers refer to the NYSE Group market share which includes NYSE and the electronic platform Arca (Archipelago). 328 Lucchetti, Aaron: NYSE Plans to Revise Specialist-Trader Rules, in: Wall Street Journal Eastern Edition (2008b), issued 13 Jun. 2008, p. C.4. 329 Chapman, Peter: Rule Changes: NYSE Euronext Bets on Specialists to Revive Broken Floor, Traders Magazine Online, 2008b, electronic source. 330 Ortega, Edgar: NYSE's Thain Fails to Stem Losses in Market Share (Update 2) (26 Jul. 2006), Bloomberg, 2006, electronic source. 331 Chapman, Peter: Reg NMS Is a Winner, SEC Says, in: Traders Magazine, Vol. 21 (2008a), No. 278, p. 22.
C.I A Method Mix for Studying Change in Organizations and Especially at the NYSE
Variable Specialist participation
Definition “… add liquidity, and stabilize the price's [sic!].”332
Causal relationships
77
Dynamic behavior
“Specialist participation in the market Monthly time series data has indeed diminished, and actually we from the NYSE Facts and Figures website. aim to reverse that trend […].”333 specialist participation ି
ା
(՜
market “Specialist participation in ା indeed ՜) the market has diminished […].”333
quality) ՜ (liquidity algorithms specialist participation
“… the five DMMs [formerly specialist firms] …, accounting for an aggregate 9.1% of NYSE volume in September 2009, up from 3.6% in September 2008, prior to the algorithmic trading (associated with rollout of NYSE’s new ି e-trade) ՜ specialist participation market model.”335 Since 2005, various forces have caused specialists’ participation in total NYSE trading to fall. […] The growing role of algorithmic trading is credited with dropping specialists’ participation ….”334
Market quality
“market quality: promoting more quoting, matching and price improvement; … over all, creating a deeper, more liquid market for all traders.”336 “It will include level of price improvement, level of size improvement, management of volatility, tightness of spread, liquidity offered at the NBBO, percent of time you're on the NBBO.”337
momentum has “Your customers … want the specialists “The to create a better quality market to swung to[o] far towards 338 trade on ….” speed and in the process “I would feel alot better if i saw some has really destroyed marreal increase in specialist participation ket quality which in the in the form of price improvement & past was always a342tradematching. Your customers dont want mark of the NYSE.” another NASDAQ, we want a TRUE Hybrid with liquidity, depth of market, and stability which only seems to come with MORE specialist involvement. … with a few minor changes the NYSE would really see a huge improvement in thier overall market quality and once again become the best place to trade [sic!].”339
“…volatility has picked up slightly.”337 “And since the New York expanded automatic executions, the quality of its quoting has deteriorated.”341
”Results suggest that market quality as a whole has “We believe the end result will be in- improved. […] Quoted creased participation …, which in turn spreads, effective spreads, further contributes to market quality.”340 and realized spreads uni“ … improve the quality of quotes on formly decreased ….”343 the exchange to attract more orders.”341
(specialist participation ା
ା
՜ market quality ՜ market share)
332
Dey, Tony: Comment on: Men At Work (at re-making the markets) (27 Nov. 2007, commented on 29 Nov. 2007), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2007a, electronic source. 333 Pellecchia, Ray: Comment on: We, Robots: UAL Hit By Perfect Storm Where Nobody Is at the Switch (10 Sep. 2008), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2008b, electronic source. 334 Lucas Jr., Henry C., Wonseok Oh and Bruce W. Weber: The defensive use of IT in a newly vulnerable market: The New York Stock Exchange, 1980–2007, in: The Journal of Strategic Information Systems, Vol. 18 (2009), No. 1, p. 5. 335 Clark, Colin: Growth in the NYSE's Liquidity Provider Programs (8 Oct. 2009), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2009, electronic source.
78
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
Variable Customer orientation
Definition
Causal relationships
"Part of being very successful for a very long time and having a large market share [is that] the New York Stock Ex“customerchange did become complacent. We focused”345 have to be receptive to change. We “serving have to give our customers what they're the needs of its looking for."347 ି customers”341 (market share ՜ pressure to change) “customer focus”344
Dynamic behavior ”… lack of responsiveness of the NYSE in the past to its customers”349 “… our customer base, which we hadn’t previously been listening to well enough.” 348
“It has had to reconnect ା Ȃ (market share (՜ … ) ՜ customer with the sources of its or345 orientation) der flow.” “That push from customers was really how this all started. … we moved in this direction because of the reaction from our customer base, ….”348 ା
(customer orientation ՜ e-trade)
Table C-3: Exemplary derivation of variables, causal relationships, and behavior 336
Pellecchia, Ray: Comment on: Ten Things I Like About the Coming Changes at NYSE (13 Jun. 2008), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2008a, electronic source. 337 President & Co-COO & Head of U.S. Cash Markets, NYSE Euronext, in: Niederauer, Duncan: Statement in speech on U.S. Cash Equities, in: Thomson Financial: NYX - NYSE Euronext Analyst Day, Corporate Analyst Meeting (6 Jun. 2007) [Conference Call Transcript], 2007b, electronic source. 338 Dey, Tony: Comment on: 'The floor is not going away. OK?' (27 Jun. 2007, commented on 28 Jun. 2008), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2007d, electronic source. 339 Dey, Tony: Comment on: Readers discuss automating NYSE openings (23 Oct. 2007, commented on 28 Oct. 2007), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2007c, electronic source. 340 Mark Schaedel, Vice President NYSE Data Products, in: NYSE Euronext Inc.: New NYSE Order Imbalances Product Goes Live Today, in: News Releases, 1 Jul. 2008, 2008b, electronic source. 341 Chapman, Peter, Nina Mehta and Michael Scotti: Men At Work, in: Traders Magazine, Vol. 20 (2007), No. 274, issued Nov. 2007, p. 52. 342 Dey, Tony: Comment on: NYSE to introduce Do Not Ship orders (13 Mar. 2007, commented on 14 Mar. 2007), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2007b, electronic source. 343 Storkenmaier, Andreas and Ryan Riordan: The Effect of Automated Trading on Market Quality: Evidence from the New York Stock Exchange, in: Kundisch, Dennis, et al. (Ed.): Enterprise Applications and Services in the Finance Industry, Berlin, Heidelberg 2009, p. 28. 344 Pellecchia, Ray: Shaping the blog to a new mark (21 May 2007), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2007, electronic source. 345 Chapman, Mehta and Scotti: Men At Work, 2007, p. 48. 346 NYSE Euronext Inc.: NYSE Euronext Appoints Todd B. Abrahall and Michael J. Rutigliano as Liaisons to NYSE Specialists and Brokers, in: News Releases, 10 Jan. 2008, 2008d, electronic source. 347 Ewing, Jack: When CEOs Talk to Each Other, Business Week Online, issued 31 Jan. 2005, 2005, electronic source.
C.I A Method Mix for Studying Change in Organizations and Especially at the NYSE
C.I.3
79
Confidence Through Model Analysis and Testing
Once some model structure is derived, it needs to be analyzed and ‘validated’ in order to gain confidence in its usefulness for generating insight. It is assumed that no model or theory can be verified and that all theories and models can even always be falsified since models always represent a simplified mapping of reality.350 In order to gain confidence in the model’s usefulness, it needs to be checked in the areas of model structure and parameterization, resulting behavior, and policies.351 The quality of the model depends on its adequacy for a certain purpose, and simulation and testing are used to check and enhance the model’s suitability to purpose by revealing possible researcher biases and assumptions.352 In the view of Milling, since validity is judged against purpose, models are not either valid or invalid; even if test results are not equally sound in all instances, the model can nevertheless be very helpful in fulfilling its purpose, generate knowledge, and help decision-making.353 Model analysis and testing is an iterative process that involves many rounds of testing and model revision. It should also draw upon all available sources of information, qualitative as well as quantitative.354 The results of these tests lead to further and further model advancement. Once the system dynamics model is sufficiently tested, it will be elaborated how a model may also serve as a testing laboratory for different scenarios. It can be used for testing alternative behavior and decision policies. Overall, the research design for analyzing the evolution of the New York Stock Exchange can be described as a model conceptualization process alternating with model testing. Different scenarios may also give insight into diverse drivers of change and provide information on when inhibitors come into play, such as inertia and limited cognition—as is discussed in the organization theory literature. There has been no 348
Pellecchia, Ray: I'd like an Auction Limit order with soy milk, please (2 Mar. 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, 2006a, electronic source. 349 Dillon, Tim: NYSE leader looks to its reputation (8 Mar. 2004), in: USA Today, 2004, electronic source. 350 See Sterman: Business Dynamics, 2000, pp. 846–851. Popper believes falsifiability to be an important criterion to judge about a theory, but according to Kuhn the decision to accept or abandon a theory usually depends on nonfalsifiable assumptions. See Kuhn: The Structure of Scientific Revolutions, pp. 152–156; and Popper, Karl R.: The Logic of Scientific Discovery, reprint Ed., London, New York 2007, pp. 61–66 and 274–276. 351 See Forrester, Jay W. and Peter M. Senge: Tests for Building Confidence in System Dynamics Models, in: TIMS Studies in the Management Sciences, Vol. 14 (1980), pp. 224–226; and Milling: Der technische Fortschritt beim Produktionsprozeß, 1974, p. 209. 352 See Forrester and Senge: Tests for Building Confidence in System Dynamics Models, 1980, p. 211; and Sterman: Business Dynamics, 2000, p. 851. See also Davis, Eisenhardt and Bingham: Developing Theory though Simulation Methods, 2007, p. 497. For a view of validation as a deductive process see Schwaninger and Grösser: System Dynamics as Model-Based Theory Building, 2008, p. 448; and moreover Cohen and Cyert: Computer Models in Dynamic Economics, 1961, p. 115. 353 See Milling: Der technische Fortschritt beim Produktionsprozeß, 1974, p. 208. See also Barlas, Yaman: Formal aspects of model validity and validation in system dynamics, in: System Dynamics Review, Vol. 12 (1996), No. 3, p. 187. 354 See Randers: Guidelines for Model Conceptualization, 1980, p. 129.
80
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
study of the transformation of the New York Stock Exchange to electronic trading neither from an organizational nor a system dynamics perspective. Existing research either focuses on financial aspects, on the organization’s IT strategy, or on personalities and events.355 Yet, this dissertation analyzes the underlying structure of behavior and processes. It centers on how the behavior emerges from the drivers of change, and how the development of dynamic drivers influences the occurrence the observed events. Since Forrester argued that in the best of cases a system dynamics model is applicable to an entire class of systems, the model of the New York Stock Exchange will then be used to abstract from the specific case and generalize the findings. Here, a second, generic model will be developed and investigated. It examines what the analysis of the case study can tell about organizations in general that face changes in their environment. The aim is to generate knowledge that is applicable to an entire class of systems. It is supposed not only to represent further stock exchanges, but explain behavior of organizational ‘dinosaurs’ or organizations in general struggling between inert, adaptive, and deliberate behavior. Before a simulation model will be developed and analyzed, the following sub-chapter will concentrate on the observed events and developments. This helps to get an understanding of how the U.S. securities market moved towards automation within the last decades.
C.II Reacting to Automation in the U.S. Securities Market In the second half of the 20th century, the securities industry faced changes in several areas. Trading professionalized, trading technology developed quickly, and the market moved towards automation. Since the New York Stock Exchange was part of the market, the developments greatly affected the organization and its trading system, and it reacted to them.
C.II.1
Automation of Order Clearing, Routing, and Information Systems
Based on research in academia around the 1950s, trading in the securities industry started to professionalize. Part of this change was a movement towards diversification resulting from the development of the Modern Portfolio Theory by Harry M. Mar355
For a financial perspective see Battalio, Robert, Andrew Ellul and Robert Jennings: Reputation Effects in Trading on the New York Stock Exchange, in: Journal of Finance, Vol. 62 (2007), No. 3; Bennett, Paul and Li Wei: Market structure, fragmentation, and market quality, in: Journal of Financial Markets, Vol. 9 (2006), No. 1; Hendershott, Terrence and Pamela C. Moulton: Speed and Stock Market Quality: The NYSE's Hybrid, Working Paper, 2009; and Storkenmaier and Riordan: The Effect of Automated Trading on Market Quality, 2009. For the development of the NYSE’s IT strategy see Lucas Jr., Oh and Weber: The defensive use of IT in a newly vulnerable market, 2009. Personalities and events are described in Gasparino, Charles: King of the Club: Richard Grasso and the Survival of the New York Stock Exchange, New York 2007.
C.II Reacting to Automation in the U.S. Securities Market
81
kowitz and others. It shifted the focus away from single stocks to the consideration of risk and volatility of an entire portfolio.356 Additionally, in the 1970s and 1980s, academics started to gain interest in patterns of price behavior and they rejected the idea of the “random walk” of securities, i.e. the idea of randomness in price movements.357 Particularly after the publication of the Black-Scholes formula in 1973, which allowed for the calculation of the value of an option, the belief in efficient markets diminished and the interest in derivatives and arbitrage increased.358 Studies found that it was possible to beat the market by recognizing its seasonal and demand patterns.359 The two developments—portfolio management and information patterns-based trading— had tremendous effects on the market. It “facilitated the rise of a new force in the world of investments—a breed of customer that would not [be] intimidated by the New York Stock Exchange—and that would change forever the Exchange’s way of doing business.”360 The combination of investment managing and research led to a professionalization of investors. The amount and fraction of capital held by institutional investors instead of households increased significantly. Institutional investors are no homogeneous group, but differ in their power and legitimacy.361 The focus of dissertation exempts institutions such as non-governmental organizations that are not important for trading, and it rather concentrates on financial institutions. These are banks, insurance companies, investment funds, pension funds, proprietary trader organizations and else. While there are differences among financial institutions, they are viewed as an aggregated group because their interests differ greatly from those of retail (i.e. private or individual) traders who, in general, do not use professional trading strategies.362 Their growth had a significant bearing on the U.S. securities market. From 1950 until 356
See Blume, Marshall E., Jeremy J. Siegel and Dan Rottenberg: Revolution on Wall Street: The Rise and Decline of the New York Stock Exchange, New York and London 1993, pp. 95–101. 357 See Lo, Andrew W. and A. Craig MacKinlay: A Non-Random Wald Down Wall Street, Princeton, NJ 2002, pp. 3–6. 358 See Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, pp. 150–151. See also Geisst, Charles, R.: Wall Street: A History: From Its Beginnings to the Fall of Enron, rev. and expand. Ed., Oxford [et al.] 2004, pp. 308–309. Before, particularly academics believed in efficient markets and in the “random walk” of securities. If a market is efficient, all relevant information about a specific security is already represented in its price so that price movements would just occur randomly through unanticipated changes. This idea dominated the thinking in academia during the 1960s, but it had just started to reach the investment community. 359 See Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, pp. 92–93. Price pressure and shifts in the demand curve determine the price. See Ryan and Schneider: Institutional Investor Power and Heterogeneity, 2003, p. 409. 360 Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 102. 361 See Ryan and Schneider: Institutional Investor Power and Heterogeneity, 2003, p. 421. 362 Retail traders or retail investors are also called private, individual or small investors. They buy or sell securities for their personal account instead of on behalf of a company or an organization. Usually retail investors trade in much smaller quantities than institutional investors. This work uses the term ‘financial institution’ as an equivalent to ‘financial organization’. In the sociology literature the term ‘institution’ is often understood differently as ‘social mechanism’ or ‘pattern of social order’.
82
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
present, capital held by U.S. institutions instead of individuals has increased from 7 to well above 50 percent, and this has been a global phenomenon.363 From 1995 to 2005 alone, the annual average growth rate of financial assets held by institutional investors was almost 7 percent.364 This represented a significant change in the New York Stock Exchange’s environment for the following reasons: First, many of these institutions very actively manage the money they hold and thus conduct many trades and replace retail traders to quite a large extent. Over two-thirds of individual or retail investors do not effect more than two transactions per year.365 Second, the preferences of institutional and individual investors often differ. Third, institutions are larger and they have more resources than individual investors, so that they are often better at developing professional trading strategies. Last, since one institutional investor on average holds much greater amounts of capital than the average retail trader, each of them accounts for a significantly larger share of trading. As such, institutional traders have more bargaining power and are thus more powerful at exerting pressure for their interests on the NYSE. For example, the Exchange’s largest two customers— the leading investment and wealth management banks Goldman Sachs Group and UBS AG—each account for about 10% of the trading.366 Their significance for the NYSE’s income to some extent leaves the NYSE to the discretion of these large investors. Additionally, there are examples of when institutional investors even influenced the Securities and Exchange Commission’s (SEC) regulations.367 Blume, Siegel, and Rottenberg note that, “[b]y the 1970s, institutional investors had replaced individual investors as the dominant force in the market.” They also report that during the period from 1950 to the late 1970 alone, the percentage of trades which were done by institutions rather than individual investors rose from 20 to 75 percent.368 Although further sources report somewhat smaller numbers, institutions’ share of 363
See NYSE Euronext Inc.: Holdings of corporate equities in the U.S. by type of institution, no date-f, electronic source. For the London market see Michie, Ranald C.: The London Stock Exchange: A History, Oxford, New York 1999, p. 631. 364 See Gonnard, Eric, Eun Jung Kim and Isabelle Ynesta: Recent Trends in Institutional Investors Statistics, in: OECD: Financial Market Trends (2008), No. 95, p. 6. 365 See United States Securities and Exchange Commission (SEC), Division of Market Regulation: Market 2000: An Examination of Current Equity Market Developments, New York, NY 1994, p. II-1. 366 See NYSE Group Inc., Annual Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 for the fiscal year ended December 31, 2005, No. 001-32829, New York 2006, p. 139. The numbers refer to the years 2004 and 2005. 367 See Ryan and Schneider: Institutional Investor Power and Heterogeneity, 2003, p. 413. They refer to Hawthorne, Fran: What the SEC rules do for activism, in: Institutional Investor, Vol. 27 (1993), No. 4, p. 51. 368 See Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 108 and see p. 105. See also Geisst: Wall Street: A History, 2004, p. 316. In the mid-1970s, the London Stock Exchange also regarded institutional investors as the powerful force in the market. See Michie: The London Stock Exchange, 1999, pp. 536 and 631. Blume, Siegel, and Rottenberg talk about a rise to 75 percent in the entire market in 1975, the NYSE Facts and Figures pages indicate that in 1980 institutions and NYSE member (owner) firms together account for 75 percent of trading and 80 percent of dollar value. See NYSE Euronext Inc.: Major sources of NYSE volume, no date-h, electronic source.
C.II Reacting to Automation in the U.S. Securities Market
83
trading has kept increasing ever since. Parallel with the rise in institutional customers, computer technology as the prerequisite for electronic trading has developed, technically allowing for the elimination of people in the order matching process.369 As part of their professional trading, institutional investors often manage large portfolios, trading in large blocks of shares—10,000 shares or an equivalent of USD 200,000 and more—increased highly both at the New York Stock Exchange and in the remaining market.370 Initially, this development was beneficial for the NYSE as block trading makes use of specialists. Since large orders have great price effects that investors like to avoid, specialists are valuable at splitting up the order and slowly matching it with orders on the opposite side of the market. With the growth of institutional investors and of block trading the number of trades increased, and U.S. stock exchanges underestimated this trend. The growth in trading required higher capacities which led to automation in the securities industry. In 1966, the New York Stock Exchange fully automated its transmission of quote and trade data.371 Back offices and in particular the delivery of securities between the NYSE and brokerage houses as well as between market participants in general followed. The back offices of Wall Street firms had been managed poorly, but a high profitability of floor brokers and specialists in general had not put pressure on the firms to deal with their poor back office management. The failure to act on the increasing trading volume resulted in what is called the back office or paper crisis. At the New York Stock Exchange and in the entire industry back offices could not keep up the post-trade paperwork and physical delivery with clearing deadlines. In the short run, other functions were synchronized with this constraint, and the industry reacted with cutbacks on the trading hours and brokers. But then the industry and the Security and Exchange Commission (SEC) looked for a solution in the automation of back offices, clearing, and settlement in order to eliminate the constraint.372 By that time, information technology was sufficiently developed to perform this task. The NYSE formed the Central Certificate Service which allowed transferring stock ownership by a single book entry. By the mid-1970s, this service was transformed into a market-wide electronic settlement and clearing company.373 The moves toward automation continued. In the beginning of the 1970s the process was only in its infancy, and over the course of time the NYSE continuously further automated its back office and order handling. In cooperation with the American Stock Exchange (Amex) it founded the Securities Industry Automation Cooperation (SIAC) that pro-
369
See McAndrews and Stefanadis: The Emergence of Electronic Communication Networks in the U.S. Equity Markets, 2000, p. 4. 370 See NYSE Euronext Inc.: NYSE Glossary: Block, no date-j, electronic source. 371 See NYSE Euronext Inc.: Chronology of the New York Stock Exchange (1930–1979), no date-c, electronic source. 372 See Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 120. 373 See ibid., pp. 123–125.
84
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
vides automation and data processing services.374 A further crisis in October 1987 resulted in further automation of systems. Instead of dealing with 5 million shares on peak days in 1960 and 600 million in 1989, today the NYSE floor is handling up to 5 billion.375 Technology has to be in place for this. At least in the area of information systems as well as clearing and settlement, the increase in trading volume caused by a rise in institutional investors led to a turn towards automation of Wall Street. Further automation concerned order routing, and this threatened the position of floor brokers. The NYSE’s Designated Order Turnaround (DOT) System eliminated floor brokers from the trading process by allowing brokerage firms to electronically transmit orders directly to the specialist. Since it allowed large institutions to go around floor brokers, it was initially met with fierce resistance from floor brokers. The DOT System was therefore restricted to orders of 199 shares or less, but this limit was more and more extended. It soon cut the order execution speed from five minutes to a minute and a half.376 Automation thus also reached floor brokers, but still did not touch the specialist position. The rise in institutional investors not only led to back office automation, but institutions were also sufficiently powerful to substantially lower commissions paid for trading. By the New York Stock Exchange’s creation through the Buttonwood Agreement in 1972, the contracting brokers established a cartel for fixed commissions. While the Exchange mainly dealt with individual customers, it could maintain the cartel for over 200 years. But whilst these commissions represent revenue for the brokers and specialists, they are costs on the side of customers, and the increasingly powerful institutions were less and less willing to pay commissions for the tasks performed by the floor. It was evident that investors could get better deals if they traded away from exchanges and circumvent commissions at the over the counter market, i.e. in the market apart from stock exchanges.377 A specialist described the developments as follows: “It was always a great business because we earned a commission on whatever we acted as agent on. There is tremendous pressure on us to give up that business now.”378 On May 1, 1975, the SEC introduced new regulations ending the NYSE cartel by ending fixed commissions. In the following period, commissions diminished.379 Since shrinking commissions meant a loss in the specialists’ and floor brokers’ income; and these floor firms saw themselves confronted with much pressure to be374
See NYSE Euronext Inc.: Timeline: Technology, no date-m, electronic source. See NYSE Euronext Inc.: Daily reported share volume: average, high, and low days (thous. of shares), no date-e, electronic source; and NYSE Euronext Inc.: Daily NYSE Group Volume in NYSE Listed, no date-d, electronic source. 376 See Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, pp. 203–206. 377 See Sobel, Robert: N.Y.S.E.: A History of the New York Stock Exchange 1935–1975, New York 1975, p. 213. 378 Quoted in Abolafia: Making Markets, 1996, p. 131. Similar statements by specialists can be found on pp. 135, 141–142 and 145. 379 See ibid., p. 109; Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 25; and Gasparino: King of the Club, 2007, p. 48 375
C.II Reacting to Automation in the U.S. Securities Market
85
come more professional, to automate, and to consolidate. They lost their power to the firms in their environment.380 This development meant a move away from the floor firms’ domination of the market towards increasing importance of institutional investors. While the NYSE was not sufficiently powerful to uphold fixed commissions, its power was still adequate to prevent a centralized market system. Based on recommendations from an Institutional Investor Study, the SEC as well as U.S. Congress favored a centralized market system, but the NYSE’s opposition was too powerful. Upon the advice of academics and the pressure of the New York Stock Exchange which would have lost much of its power to the centralized system, a restricted version, the Intermarket Trading System, finally was implemented. It gives market professionals the opportunity to send orders to other markets if these markets display a better price.381 Its most important component is what is called the “trade-through rule” which helps price protection by demanding that a trade is not traded through inferior markets, but should instead be directed to the market which offers the best price. The Intermarket Trading System applied to exchange-listed stocks that were traded at exchanges and in the over-the-counter market.382 It was rather a recommendation for self-regulation, and it maintained a decentralized system. Apart from the abandonment of fixed commissions and the implementation of the Intermarket Trading System, the securities markets of the 1970s saw a further development that was important for the NYSE. Many new forms of stock derivatives were created, and particularly options and index futures gained in importance.383 This paved the way for index arbitrage and for program trading. In index arbitrage traders use disequilibria e.g. between futures and options and the underlying stock, and program trading in general involves the use of an arbitrage strategy for a large portfolio
380
See Abolafia: Making Markets, 1996, p. 149. In order to remain profitable and to fulfill the tasks of providing liquidity, many specialist firms consolidated, and mane even went out of business. The number of specialist at the NYSE decreased from 150 in 1960 to 40 in 1990 and 5 firms today. 381 See NYSE Euronext Inc.: NYSE Glossary: Intermarket Trading System (ITS), no date-k, electronic source. For information on the struggle between the SEC and opposition by the NYSE see also Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, pp. 163–180. The ITS Plan was approved in 1978. It requires trading venues to provide electronic access to their best bid and offer. It also provides an electronic mechanism for routing orders to other market places if they display a better price. While these rules are not mandatory, the participants agreed to adopt the rules. See footnote no. 30 in: Securities and Exchange Commission (SEC): Regulations NMS (Final Rule) (Release No. 34–51808; File No. S7–10–04), in: Federal Register, Vol. 70 (2005), No. 124, issued 29. June 2005, p. 37501. 382 See Securities and Exchange Commission (SEC): Regulations NMS, 2005, p. 37501; and Securities and Exchange Commission (SEC): Adoption of Amendments to the Intermarket Trading System Plan To Expand the ITS/Computer Assisted Execution System Linkage to all Listed Securities (FR Doc. 99–32684, SEC Release No. 34–42212, File No. 4–208), in: Federal Register, Vol. 64 (1999), No. 241, issued 16 Dec. 1999, chapter I.A. 383 Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, pp. 147 and 152.
86
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
of stocks.384 Information technology had greatly advanced to allow for this computerized trading. It makes sense to use computer programs for calculating the strategy and placing orders at the right point in time. Therefore, large clients like to use computer algorithms which work best when the exchange also allows for trading in subsecond speed. If possible, they make use of electronic trading possibilities, and the time to execute a trade becomes highly important because institutions desire immediate execution.385 Algorithmic trading programs are also able to split large orders and feed them into the system in small portions over time and even to route them to different exchanges and trading platforms. Often, it is important for institutional investors to hide their buy or sell interest when they like to trade large quantities of shares, which would have a significant impact on the market price. A fully electronic execution system allows them to ‘test’ the price with a small order and remain anonymous. They can also send many small orders instead of a large one which then have a much smaller price impact because other investors are not aware that these orders all come from one source. Since institutions often trade professionally, sometimes even with a short-time horizon, the trading speed is of much greater importance to them than price quality, i.e. receiving the best price. Traditionally, stock exchanges competed on price quality which private customers still prefer. Institutional traders seize the opportunity to trade electronically because it gives them more rapid access to quotes and trade information, and it allows them to react more quickly to changes in the market. Ellul et al. find that orders which are sent to an automated system are less price sensitive than those sent to the floor.386 In the time period which the specialist needs to match an order, the market may move to another price. Slow execution is thus interpreted as a cost by many investors.387 As a consequence, institutional investors often prefer the price security that comes with fast executions. It allows them to react almost infinitesimally quickly to new information. The importance of the trading speed has also been confirmed by an interviewee working for the New York subsidiary of an international bank. Electronic trading in which the order execution process is automated allows for just this sub-second order execution speed, whereas manual handling of orders has the advantage of higher price quality due to the obligations of the specialist and the bargaining of floor brokers.
384
See ibid., p. 153; Kim, Kendall: Electronic and Algorithmic Trading Technology: The Complete Guide, in: Kaljuvee, Ayesha and Jürgen Kaljuvee: Complete Technology Guides for Financial Services Series, Amsterdam [et al.] 2007, pp. 8–9; and Stoll, Hans R.: Electronic Trading in Stock Markets, in: Journal of Economic Perspectives, Vol. 20 (2006), No. 1, p. 168. 385 See Hendershott and Moulton: Speed and Stock Market Quality, 2009, pp. 1 and 27; and Lucchetti, Aaron: Nasdaq Passes NYSE in Sales Milestone, in: Wall Street Journal - Eastern Edition (2008a), issued 14 Jul. 2008, p. C.3; Lucchetti: NYSE Plans to Revise Specialist-Trader Rules, p. C.4; and Tabb, Larry: The NYSE Floor: A Question of Control, in: Wall Street & Technology (2005), issued Mar. 2005, p. 54. 386 See Ellul, Andrew, et al.: Order dynamics: Recent evidence from the NYSE, in: Journal of Empirical Finance, Vol. 14 (2007), No. 5, p. 659. 387 See Boehmer, Ekkehart: Dimensions of execution quality: Recent evidence for US equity markets, in: Journal of Financial Economics, Vol. 78 (2005), No. 3, pp. 554 and 581.
C.II Reacting to Automation in the U.S. Securities Market
C.II.2
87
Moving Towards an Electronic Market
Despite the professionalization in trading strategies, the NYSE “failed to keep pace with the new explosion of market needs” and it lost some of its importance.388 It had automated the distribution of trade information, of clearing and settlement in back offices and also allowed for automated routing of orders. However, the NYSE did not make any move towards full electronic trading. According to Clemons and Weber, the organization remained committed to its specialist system and used automation only in the area of clerical details where it enforced the existing manual trading system. The NYSE did not attempt to change and the authors regarded it as impossible that the NYSE would make a move towards e-trade.389 Other exchanges and trading venues had automated their order matching system already, meaning they had implemented electronic trading (e-trade), or they were further along. NASDAQ was founded in 1971 as an electronic trading platform and later purely electronic stock exchange and is the NYSE’s largest competitor today. Already in the late 1970s, its technology included a computer-assisted execution system. Participating firms could use it to route their orders that would be executed against each other as well as against bids and offers of third market makers.390 The Cincinnati Stock Exchange implemented an electronic system in the late 1970s as well. It closed its trading floor and became all electronic in 1980 and was the first fully automated exchange.391 In the year 1981, the Philadelphia Stock Exchange introduced an automated order matching system for orders smaller than 400 shares which it continuously expanded in the following years.392 The Boston Stock Exchange became electronic in 1989. One year later, the Chicago Mercantile Exchange introduced a fully automated futures and options trading system.393 The move towards electronic trading was an international phenomenon. In 1986, the London Stock Exchange changed from open outcry to screenbased electronic trading. But the move was also accompanied by a loss of interaction and feel for the market.394 Shortly afterwards, in 1988, the Swiss Options and Financial Futures Exchange (SOFFEX) represented the first fully automated options ex-
388
Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 147. See Clemons, Eric K. and Bruce W. Weber: London's big bang: a case study of information technology, competitive impact, and organizational change, in: Journal of Management Information Systems, Vol. 6 (1990), No. 4, pp. 48 and 52. 390 Directly see Geisst: Wall Street: A History, 2004, p. 506. NASDAQ stands for National Association of Securities Dealers Automated Quotations. 391 See National Stock Exchange (NSX): History, no date, electronic source; and Seligman, Joel: The Transformation of Wall Street: A History of the Securities and Exchange Commission and Modern Corporate Finance, 3rd Ed., New York, NY 2003, p. 521. The Cincinnati Stock Exchange is today called the National Stock Exchange (NSX). 392 See Hamilton, Martha M.: Philadelphia Exchange Booming: Philadelphia Expands Option, in: The Washington Post (1981), issued 4 Aug. 1981, p. D6. 393 See Welles: Is it time to make the Big Board a black box, 1990, p. 74. 394 See Clemons and Weber: London's big bang, 1990, pp. 42 and 52; and Michie: The London Stock Exchange, 1999, pp. 586–587. 389
88
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
change and it also included futures two years later.395 Stockholm and Paris abolished the trading floor in 1991.396 These examples show that there was significant movement towards e-trade in the market; and by the early 1990s most of the major stock exchanges had automated the matching of orders. Many experts expected the development towards automation of exchanges to proceed. Schmidt expected financial markets to be almost completely electronic in 2003. He explains that, in the beginning of the 1990s, stock exchanges still tried to maintain their traditional organizational form, but that it is a question of time only until change will be forced by the clients or until other parties will step in with new systems.397 The behavior over time (BOT) graph in Figure C-2 shows the fraction of the largest foreign competitors of the NYSE enabling some e-trade.398 It refers to the leading 22 non-U.S. exchanges that have implemented some portion of electronic trading. The largest proportion of them implemented electronic trading around 1990, and by the mid-1990s, 95 percent of them allowed for e-trade.
Fraction of largest foreign competitors enabling some e-trade 1
Dmnl
0.75 0.5 0.25 0 1970
1980
1990
2000 Date
2010
2020
2030
Figure C-2: Percentage of leading stock exchanges enabling some or full e-trade (BOT)398 According to Gasparino, the pressure to automate its trading system grew also at the New York Stock Exchange. He expresses that institutional investors regarded trading at the NYSE’s floor as too costly, questioned the specialist system as well as the NYSE’s monopoly position.399 Their voices gained weight with the growth of NASDAQ as an electronic stock market, and with other exchanges to implement 395
See Schmid, Beat F.: Elektronische Märkte, in: Wirtschaftsinformatik, Vol. 35 (1993), p. 469. See Lucas Jr., Oh and Weber: The defensive use of IT in a newly vulnerable market, 2009, p. 9. 397 See Schmid: Elektronische Märkte, 1993, p. 470. 398 Data is taken from Jain, Pankaj K.: Financial Market Design and the Equity Premium: Electronic versus Floor Trading, in: Journal of Finance, Vol. 60 (2005), No. 6, p. 2960. Further data can be found in Gorham, Michael and Nidhi Singh: Electronic Exchanges: The Global Transformation from Pits to Bits, Amsterdam [et al.] 2009, pp. 66–71. 399 See Gasparino: King of the Club, 2007, p. 56. See also Gorham and Singh: Electronic Exchanges, 2009, p. 74. 396
C.II Reacting to Automation in the U.S. Securities Market
89
electronic order execution. In December 2000, the NYSE did implement e-trade for small orders. However, it greatly restricted it so that the fraction of trades with electronic execution remained small. The NYSE greatly lagged behind its competitors. The move towards electronic trading is in line with what is either called the electronic market hypothesis, electronic brokerage effect, or disintermediation. It suggests that electronic communication between agents will replace intermediaries (brokers) in a market and that a central database or electronic market will fulfill their intermediary function.400 Picot, Bortenlänger, and Röhrl predicted this to happen worldwide for stock exchanges. To the extent that brokers do not execute more tasks than order handling, they become obsolete as customers enter into direct electronic contact with each other.401 The electronic brokerage effect is limited by two factors only. In securities trading, there may be a critical mass problem related to liquidity when shifting to new markets. Additionally, the replacement of brokers may make institutional changes necessary. Threatened intermediaries may form cooperative groups and react with resistance.402 For example, in the Australian beef industry the resistance of intermediaries caused the adoption of an electronic market system to fail. Strong influence of brokers in the industry made cattle producers refrain from adopting the electronic system and it prevented the direct link of producers and meat processors. The attempt succeeded where intermediaries were less common.403 The most prominent example of the electronic brokerage effect are Electronic Communication Networks (ECNs). They are trading platforms which electronically match orders or route them to other platforms or exchanges if supply and demand cannot be matched.404 ECNs allow for the electronic matching of orders outside a stock exchange, usually with extended trading hours. The first one was established in 1969; since the 1990s, their number grew. Order execution times for ECNs were 3 seconds in 1999, 1 second in 2004, and often ranged in the sub-millisecond speed 400
See Malone, Thomas W., Joanne Yates and Robert I. Benjamin: Electronic Markets and Electronic Hierarchies, in: Allen, Thomas J. and Michael S. Scott Morton (Ed.): Information Technology and the Corporation of the 1990s: Research Studies, New York and Oxford 1994, pp. 67–68. 401 See Picot, Arnold, Christine Bortenlänger and Heiner Röhrl: Organization of Electronic Markets: Contributions from the New Institutional Economics, in: The Information Society, Vol. 13 (1997), No. 1, pp. 113–115. 402 See Clemons and Weber: Information Technology and Screen-Based Securities Trading, 1997, p. 1706; and Picot, Bortenlänger and Röhrl: Organization of Electronic Markets, 1997, p. 115. 403 See Driedonks, Caroline, et al.: Economic and Social Analysis of the Adoption of B2B Electronic Marketplaces: A Case Study in the Australian Beef Industry, in: International Journal of Electronic Commerce, Vol. 9 (2005), No. 3, pp. 63–65. 404 Apart from stock exchanges, trading venues include alternative trading systems such as electronic communication networks, dark pools, and crossing networks. Electronic communication networks allow for the electronic matching of orders outside a stock exchange, usually with extended trading hours. Dark pools and crossing networks do not publicly display orders, but rather serve as a black box, particularly for large orders. It is their purpose to hide information about quotes in order to have a minimal impact on the price.
90
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
in 2010.405 ECNs gained importance particularly for institutional investors since they could go around brokers and could directly trade with each other. For these institutions it is also of great significance that orders could be matched much more quickly. NASDAQ in particular started to gain importance in trading NYSE-listed stocks. From 1980 to 2004, it handled trading volume in NYSE-listed stocks rose from about 2 to 13 percent.406 Until 2007, NASDAQ’s volume in NYSE-listed stocks even rose to 37 percent.407 Figure C-3 provides information about the NYSE’s market share in its stocks. It reveals that it could maintain its prominent position as a market leader for a rather long time. Its market share used to range between 75 and 85 percent of trading in NYSE-listed securities. Starting around 2005, its share of trading declined rapidly by nearly 15 percentage points per year. Trading volume and market share fell; in September 2007 the NYSE had an average trading volume of 1.4 billion shares per day with a market share of trading in NYSE-listed securities of 44 percent. In June 2008 the NYSE on average traded 1.2 billion shares per day, and its market share had fallen to 29 percent. 408 It leveled out at about 25 percent in 2009.
NYSE market share in NYSE-listed issues 1
Dmnl
0.75 0.5 0.25 0 2000
2002
2004
2006 Date
2008
2010
Figure C-3: NYSE market share in NYSE-listed securities (BOT)409 405
See Dwyer, Paula, Amy Borrus and Gary Weiss: Big Bang at The Big Board, in: BusinessWeek (2004), issued 16 Feb. 2004, p. 67; NYSE Euronext Inc.: NYSE Arca Submillisecond Execution Speed with Universal Trading Platform, in: U.S. Equities News (2010b), issued Mar. 2010, p. 1; Smith, Randall, Greg Ip and Charles Gasparino: Bitter Rivals Jointly Seek Major Changes in the Markets, in: Wall Street Journal (1999), issued 1 Oct. 1999, p. C.2; and Tabb, Larry: A Tale of Two Exchanges, in: Wall Street & Technology, Vol. 4 (2008b), No. 5, p 49. 406 See NYSE Euronext Inc.: Market Share of consolidated tape volume by year (1976–2003), no date-i, electronic source. Numbers from other sources differ slightly. Lucas Jr, Oh, and Weber speak of 12.6 % in 2004. See Lucas Jr., Oh and Weber: The defensive use of IT in a newly vulnerable market, 2009, p. 5. 407 See Lucas Jr., Oh and Weber: The defensive use of IT in a newly vulnerable market, 2009, p. 5. They refer to a study by Barrons Market Lab. 408 See Chapman: Rule Changes, 2008, electronic source. 409 For data up to the year 2003 see NYSE Euronext Inc.: Market Share of consolidated tape volume by year (1976–2003), no date, electronic source; and for data from 2003 to May 2009 see NASDAQ OMX Group Inc.: NYSE Market Share in NYSE-Listed Securities [Excel Chart], no
C.II Reacting to Automation in the U.S. Securities Market
91
Until about 2005, although NASDAQ and ECNs were in place and gained some importance, in many cases institutional investors could not execute their trades there because the NYSE’s “monopoly was maintained by arcane rules that favored the floor.”410 Traders particularly criticized the SEC’s trade-through rule that is based on the Securities Exchange Act of 1934 and on the Intermarket Trading System of 1978. The rule mandated that an order be executed at the exchange that currently displays the best price.411 Since the New York Stock Exchange displayed the best price most of the time, it attracted most of the order flow. But institutional investors preferred a fast execution even at a slightly inferior price because this decreased their risk that the price of a security would move.412 The manual execution at the NYSE takes about 10 to 30 seconds, allowing for changes in the security’s price. The pressure from institutional investors to abolish this trade-through rule was thus fierce. The institutions with a preference for speed turned to electronic communication platforms as far as regulation allowed. They received invaluable support when the SEC in 2005 announced the Regulation NMS, whose Order Protection Rule meant a great regulatory shift in favor of electronic and automated trading. This rule amended the trade through rule of the Intermarket Trading System. According to Regulation NMS, orders now only needed to be sent to another trading venue offering a better price if that was a fast market, meaning if it allowed for execution in sub-second speed.413 Additionally, the new regulation did not mandate the full order to be sent, but required order fragments to be sent, too. Regulation NMS represents a great advantage for institutional investors since they had often complained about the risk of price change between the point of order entry and execution. Additionally they said that before Regulation NMS if the NYSE offered the best price for a small amount of shares, it would still have received the full order, while it now only receives a respective fragment, and the remaining fragments are executed against the new best prices in the market. Institutional investors could exert
date, electronic source; and for data of the last full trading week of a month after May 2009 see BATS: Market Volume Summary, no date, electronic source. Data reported by Chapman and by NASDAQ OMX differ by no more than one percentage point. Data illustrate the fraction of shares matched in NYSE-listed securities as part of the Total Consolidated Tape. The Total Consolidated Tape aggregates shares matched at U.S. exchanges and volume of transactions effected otherwise than on an exchange which are reported to the Financial Industry Regulatory Authority. Some of these exchanges have ECN-like properties. 410 Gasparino: King of the Club, 2007, p. 57. 411 See Securities and Exchange Commission (SEC): Adoption of Amendments to the Intermarket Trading System Plan, 1999, p. 70298. 412 See Borrus, Amy: No More Breaks for the Big Board, in: Business Week (2004), No. 3911, issued 6 Dec. 2004, p. 46. 413 See Securities and Exchange Commission (SEC): Regulations NMS, 2005, pp. 37500–37501. Promulgated by the US Securities and Exchange Commission (SEC), the Regulation NMS (Regulation National Market Share) is supposed to strengthen the national market system. It requires orders and order fragments to be sent to the trading venue that offers the best price as well as sub-second execution. Orders sent to manual trading floors are exempt from this order protection rule.
92
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
coercive power by influencing the rules of the game through their influence on the inauguration of Regulation NMS. Developing technology, professional financial instruments, the growth of ECNs, and regulatory requirements represent important changes in the U.S. financial industry during the 1960s to 2010.414 Technological developments and the rise in institutional investors caused a professionalization of trading and led to automation. Overall, the automation of exchanges has been distinguished in four areas: information systems, clearing and settlement, order routing, and order execution systems.415 The NYSE and the industry have already arrived at the first three phases. In 1978 the exchange had implemented electronic communication tubes to allow electronic communication with the specialist booths. In order to remain competitive and increase the speed of its transactions, it automated its information systems, back offices, and order routing. Information systems providing information about quotes and trades were automated e. g by the introduction of the fully automated quotation service in 1965 and its later extensions e.g. to off-floor customers.416 Clearing and settlement systems attracted attention after the back office crisis in the late 1960s and the 1970s. Also due to regulatory requirements significant automation of clearing and settlement followed. Order routing systems were automated by the introduction of the DOT System by the New York Stock Exchange because it allowed for the automatic routing of orders to the specialist.417 Thus, in the three aforementioned areas, the NYSE has undergone change. The last area that can be automated is the order execution system that would eliminate the matching of orders by specialists and replace them by an electronic system. This area remained untouched. The automation of the fourth area represents electronic trading because it eliminates the manual matching of orders and replaces trading floors by an electronic system. Since large investors have been in favor of automation even before 1980, researchers and practitioners alike expected this trend towards e-trade to continue, i.e. they expected those exchanges which had not yet automated their order execution systems to fully implement electronic trading.418 Intermediation by the trading floor proves beneficial particularly for
414
See Madhavan, Ananth: Market microstructure: A survey, in: Journal of Financial Markets, Vol. 3 (2000), No. 3, p. 206. 415 See Picot, Bortenlänger and Röhrl: The Automation of Capital Markets, 1995, electronic source. Lucas, Oh, and Weber distinguish four technology categories: order processing, broker support, specialist support, and transparency and disclosure. See Lucas Jr., Oh and Weber: The defensive use of IT in a newly vulnerable market, 2009, p. 7. 416 See NYSE Euronext Inc.: Timeline: Technology, no date, electronic source. Information systems include the Consolidated Tape which since 1975 disseminated real-time trade and quotation information, and the NYSE OpenBook in 2002, enabling off-floor customers to view buy and sell interest at every price point. 417 See Picot, Bortenlänger and Röhrl: The Automation of Capital Markets, 1995, electronic source. 418 See Clemons and Weber: London's big bang, 1990, p. 56; Clemons and Weber: Information Technology and Screen-Based Securities Trading, 1997, p. 1695; Feldman: Electronic Mar-
93
C.II Reacting to Automation in the U.S. Securities Market
low-volume securities, but trading volume continues to rise, so that the need for a trading floor with human intervention will diminish.419 But academics, economists, and ‘non-Wall Streeters’ did not believe that the NYSE would follow; instead they describe it as a relic, ready to be replaced by electronic marketplaces.420 Focal Stock Exchange's Fraction of E-Trade +
fraction of e-trade in remaining market
+ gap in fraction of (B) e-trade Adaptation
pressure to adapt +
Figure C-4: Adaptation process of stock exchanges This view of the adaptation of stock exchanges to the demands that their institutional and technological environment imposes can be represented by a balancing feedback mechanism that regulates the adaptation to exogenous input. The causal mechanism of adaptation to the market as an example of negative feedback in a social system can be seen in Figure C-4. This view assumes development in the market environment to control stock exchanges’ trajectories. It regards the environment as the driver of change, forcing the other stock exchanges’ move towards electronic trading. Only for the NYSE, many did not expect this adaptation process to work.
C.II.3
The New York Stock Exchange’s Recast of Trading Systems
The NYSE management strongly believed in the superiority of the manual floor trading system in comparison with electronic trading. By the specialists’ obligation to provide liquidity and to provide for an auction process in a fair and orderly market, the NYSE expected to improve market quality. Market quality is a composition of characteristics of a market place and particularly includes price efficiency (high), liquidity (high), and volatility (low).421 Research on the economic value of the trading floor often supported the NYSE decision-makers’ view, as will be described. Price improvement is the principal value the floor can provide. A market order is normally filled at ketplaces, 2000, p. 95; and Picot, Bortenlänger and Röhrl: The Automation of Capital Markets, 1995, electronic source; Stoll: Electronic Trading in Stock Markets, 2006, pp. 154 and 173. 419 See Barclay, Michael J., Terrence Hendershott and Kenneth Kotz: Automation versus Intermediation: Evidence from Treasuries Going Off the Run, in: Journal of Finance, Vol. 61 (2006), No. 5, p. 2413. 420 See Naidu and Rozeff: Volume, volatility, liquidity and efficiency of the Singapore Stock Exchange before and after automation, 1994, p. 24; and Welles: Is it time to make the Big Board a black box, 1990, p. 74. 421 The definition of market quality differs. Apart from price efficiency (effective and realized spreads) liquidity, and volatility authors sometimes also include order consolidation as a cause of liquidity and market quality as well as speed. See Bennett and Wei: Market structure, fragmentation, and market quality, 2006, p. 51; and Boehmer: Dimensions of execution quality, 2005, p. 554.
94
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
the respective market price. The quote may be that of the specialist or be a public order with a price limit.422 The order is price improved if the specialist executes it against a further market order in his order book or against his own money at a price better than the best market-wide quotation, i.e. better than the national best bid and offer (NBBO). The NBBO is the highest revealed price at which there is interest to buy and the lowest revealed price at which someone in the entire market is willing to sell. The price difference between the revealed buy and sell price is called the spread. As is exemplified by Figure C-5, if an order is executed within the quoted spread, this order is price improved. If a market buy order is filled at USD 20.07 while the NBBO offer is at USD 20.10, the buyer receives price improvement of USD 0.03 in comparison to what she could otherwise have gotten in the market. USD 20.10
lowest offer to sell
USD 20.00
best bid to buy
spread, area for price improvement Figure C-5: Spread Finance researchers often distinguish between the quoted, effective, and realized half-spread of which the first one assesses expected execution costs. The effective spread is a better measure for price improvement because it accounts for price changes between the point of order entry and execution. The realized spread even includes price movements after execution, and it provides information on the sustained price improvement.423 Handa, Schwartz, and Tiwari are able to show that in 1996 at the American Stock Exchange (AMEX) about 54 percent of orders, also representing about 56 percent of the shares traded, were price improved with an average of 6.81 cents per share. Numbers refer to orders arriving via floor brokers. Orders that were routed electronically to the AMEX specialist received price improvement only about 20 percent of the time and 14 percent of shares traded by an average of 1.43 cents per share.424 In a follow up study based on data from 2001 the same authors also found that while an 422
See Conroy, Robert M. and Robert L. Winkler: Market Structure: The Specialist as Dealer and Broker, in: Journal of Banking & Finance, Vol. 10 (1986), No. 1, p. 22. 423 Handa, Schwartz and Tiwari: The Economic Value of a Trading Floor, 2004, p. 336; and Huang, Roger D. and Hans R. Stoll: Dealer versus auction markets: A paired comparison of execution costs on NASDAQ and the NYSE, in: Journal of Financial Economics, Vol. 41 (1996), No. 3, pp. 322–327. The quoted half-spread is a measure at the time of order entry and indicates one-half of the ratio of the bid-ask spread and the mid-quote. The effective half-spread measures the ratio of the execution price and the prevailing mid-quote. The realized half-spread includes the price impact of the respective trade, notifies about informed trading, and measures the average revenue of specialists. It is the negative of the logarithmic return of the transaction in comparison with the mid-quote five minutes or 15 trades after the transaction. It thus excludes short-term price impacts. 424 See Handa, Puneet, Robert A Schwartz and Ashish Tiwari: Price Improvement and Price Discovery on a Primary Market: Evidence from the American Stock Exchange, in: Journal of Portfolio Management, Vol. 25 (1999), No. 3, p. 58.
C.II Reacting to Automation in the U.S. Securities Market
95
increasing number of institutional investors routes their orders electronically, trades handled by floor brokers receive a significantly better price, expressed by the realized half-spread, equivalent to savings of about 4 cents per share. The trading floor provides value by reducing execution costs through superior order handling.425 A study of the Singapore Exchange’s move towards electronic trading in 1989 showed that substantial increases in volume and in volatility, the latter particularly for low-volume stocks, followed the automation. Furthermore spreads increased slightly.426 A comparison of the manual NYSE with the electronic Paris Bourse also shows significantly lower effective spreads at the NYSE.427 A sample of stocks that switched from being listed on NASDAQ to the NYSE in 2002 showed substantially better market quality because of higher order consolidation and liquidity at the NYSE. 428 This shows that in areas where order routing is automated price improvement and thus market quality decrease, supporting the NYSE decision-makers’ view that the floor does have value. Specialists participate particularly in low-volume stocks where their service of liquidity provision is most highly valued. They step in when bid-ask spreads are wide and when there is significant price movement in the respective security.429 The fact that intermediaries are especially valued in low volume securities is supported by an analysis of Barclay, Hendershott, and Kotz who find evidence that for treasury securities which are on the run, meaning they are relatively new and actively traded, electronic brokers’ market share is about 80 percent. But when the securities go off the run, meaning they are traded less frequently, electronic brokers’ market share falls to 12 percent. They also support the notion that intermediaries participate more when there are market imbalances. Yet, if trading volume increases, they also expect intermediaries to continue to diminish.430 A growing number of institutional customers use smart order routing systems which allow them to work their orders strategically and to bypass human intermediaries such as floor brokers. For example, these systems react to market circumstances and split up large orders so as to avoid a price impact. However, while researchers
425
See Handa, Schwartz and Tiwari: The Economic Value of a Trading Floor, 2004, pp. 333–334 and 353–354. 426 See Naidu and Rozeff (1994) Volume, volatility, liquidity and efficiency of the Singapore Stock Exchange before and after automation, pp. 37 and 40–41. 427 See Venkataraman, Kumar: Automated Versus Floor Trading: An Analysis of Execution Costs on the Paris and New York Exchanges, in: Journal of Finance, Vol. 56 (2001), No. 4, p. 1479. 428 See Bennett and Wei: Market structure, fragmentation, and market quality, 2006, pp. 69 and 71. 429 See Venkataraman: Automated Versus Floor Trading, 2001, p. 1480. 429 See Ellul, et al.: Order dynamics, 2007, pp. 637 and 658; and Madhavan, Ananth and George Sofianos: An empirical analysis of NYSE specialist trading, in: Journal of Financial Economics, Vol. 48 (1998), No. 2, pp. 208–209. 430 See Barclay, Hendershott and Kotz: Automation versus Intermediation, 2006, pp. 2412–2413.
96
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
expect all markets to adapt e-trade, they also state that these need to include some of the intelligence that human intermediaries provide such as smart order routing.431 Already in the 1990s, the NYSE was aware of the growth of institutional investors and their trading preferences. The organization’s research department made aware of the fact that institutional traders are attracted to ECNs and prefer the abandonment of intermediaries, such as specialists and floor brokers. Nevertheless they still believed that customers would continue to come to the floor for the advantages it provides, e.g. liquidity and the flexibility of floor brokers.432 The New York Stock Exchange relied heavily on the superiority of its floor trading process. The organization was known for its often loud and hectic trading process with a crowd clustered around the specialists’ posts and vivid negotiation for price improvement. It was symbolic for the institution. The specialists used to be a very strong force at the NYSE that helped shape its outside image and reputation. Its trading process may have been boisterous, but as an organization it portrayed both stability and strength. With the image of conservatism and strength, the NYSE was for a long time a symbol for the US economy as the largest and most prestigious stock exchange of the world. Yet, following the rise of institutional investors and electronic trading, dissatisfaction with the NYSE’s slow execution speed continued to grow. In fact, whereas the NYSE implemented many changes to accelerate its current trading system, the core of it remained untouched. The NYSE did not want to follow the adaptation process to electronic trading that most other stock exchanges had already undergone. Quite the contrary, still in 2003 the NYSE declared its commitment to the manual trading system. The then CEO Richard Grasso confirmed that “… change at the NYSE is likely to be incremental at best. … Elimination of the exchange's floor-trading system, as urged by some exchange critics, … is not about to happen.“433 The NYSE under Grasso already enabled purely electronic trading for small orders by its Direct+ system in December 2000. Yet, the way of trading changed incrementally only, since the NYSE placed restrictions on order type, frequency, and size. For instance, there used to be a restriction against entering orders for the account of the same owner in less than 30-second intervals.434 Consequently, in 2003, only 7 per431
See Handa, Schwartz and Tiwari: The Economic Value of a Trading Floor, 2004, p. 354; and Venkataraman: Automated Versus Floor Trading, 2001, p. 1480. 432 See Shapiro, James E.: U.S. Equity Markets: A View of Recent Competitive Developments, NYSE Working Paper 93-02, New York 1993, pp. 9 and 11. 433 Weiss, Gary: The $140,000,000 Man, in: Business Week (2003), No. 3849, issued 15 Sep. 2003, pp. 90–92. 434 See Jiang, Christine, Thomas McInish and James Upson: The information content of trading halts, in: Journal of Financial Markets, Vol. 12 (2009), No. 4, p. 712; Pellecchia, Ray: WSJ: Lucent first stock to get unrestricted auto-ex in Hybrid Market (3 May 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2006e, electronic source; and Pellecchia, Ray: On crisis and opportunity: Thomas Friedman and the Hybrid Market (1 Mar. 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, 2006c, electronic source.
C.II Reacting to Automation in the U.S. Securities Market
97
cent of volume was handled electronically, and by 2005 this number had risen to 11.4 percent (see Figure C-6). Since around 2007, the formerly lively trading floor has been empty and quiet.435 Two years after the implementation of electronic trading for small orders, the NYSE’s CEO Richard Grasso was replaced by a new management team led by John Thain. Grasso became involved in a scandal when he tried to cash in his pay and retirement package which was considered inappropriate and outrageous for a CEO of a nonprofit company. Then in 2004, the trading system somewhat changed when the limit on the size of orders that could be handled electronically was removed. But the lively picture of manual floor trading dominated the exchange until about mid-2006. Then, electronic trading rapidly replaced manual floor trading. From Oct. 6, 2006 to April 2, 2007 the New York Stock Exchange completed the rollout of the Hybrid Market—a mostly electronic market with elements of both automation and floor trading. It combines specialist obligations and floor broker expertise with the speed of electronic markets in order to be liquid and provide better market quality than purely electronic exchanges.436 Customers can choose whether they send their orders to the computer or to the specialist. In the view of Boehmer, Jennings, and Wei, the NYSE’s slow execution speed triggered the organization’s decision to further automate its Direct+ system.437 According to Lucchetti, the transformation of the trading system happened in response to the acknowledgement of customer demands and falling market share.438 Figure C-6 indicates how rapidly electronic trading developed once the decision to automate had been made. The graph reveals the reference behavior of the NYSE’s fraction of e-trade over the full time horizon. Whereas in 2005 only 11.4 percent of volume was executed electronically, in 2007, 86 percent of volume and 96 percent of trades were handled by computers. The move to e-trade heavily reduced the need for people, and the number of people on the floor fell from about 3500 to 1500.439 Three of the five trading floors closed down permanently, and the remaining two convey an almost idle atmosphere. In October 2008, the New York Stock Exchange introduced further changes. It even turned away from the ‘old’ specialist system which it had protected for so long, not completely discarding it, but abandoning some of the specialist responsibilities and advantages and made them
435
See Steverman, Ben: NYSE: Hooray for Market Volatility, in: Business Week Online (2007), issued 5 Nov. 2007, p. 25. 436 See NYSE Regulation Inc., Market Surveillance: Hybrid Market Implementation - Phase IV, in: Information Memo, No. 07-12, New York 2007. 437 See Boehmer, Ekkehart, Robert Jennings and Li Wei: Public Disclosure and Private Decisions: Equity Market Execution Quality and Order Routing, in: Review of Financial Studies, Vol. 20 (2007), No. 2, p. 317. 438 See Lucchetti: NYSE Plans to Revise Specialist-Trader Rules, p. C.4. 439 See NYSE Euronext Inc., Annual Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 for the fiscal year ended December 31, 2007, No. 001-368007, New York 2008a, p. 13.
98
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
designated market makers (DMMs).440 The management team added special liquidity provider firms and introduced algorithms for designated market makers and floor brokers to allow them to improve market quality in an electronic environment. Most surprisingly, although the NYSE made great efforts to keep the floor and the respective floor firms, by automating it inflicted damage on its floor brokers and specialists who until quite recently had a great influence in the Exchange’s board and were the symbol for the institution. The NYSE underwent a sincere transformation.
Reference Mode 1
Dmnl
0.75 0.5 0.25 0 1970
1980
1990
2000 Date
2010
2020
2030
"fraction of e-trade from data (volume)" :
Figure C-6: Reference mode (BOT)441 An anonymous electronic survey conducted by Traders Magazine confirmed that most customers were generally in favor of the NYSE’s move towards automation. It revealed that among those who used floor brokers in the manual trading environment, 66 percent will not use them anymore or will use them to a smaller extent. 14 percent have never used them. This also becomes obvious as the floor brokers’ share of NYSE volume traded on the exchange sank by half in the period of the first
440
Acording to NYSE Euronext, the Designated Market Maker has similar responsibilities to the specialist and will: - be obliged to maintain an orderly market in assigned stocks; - be required to quote at the national best bid and offer a specified percentage of the time; - facilitate price discovery at the open, close and in periods of significant imbalances; - provide liquidity, price improvement and match incoming orders based on a Capital Commitment Schedule; - no longer have an advance ’look’ at incoming orders, but trade on parity with others; - have transparent economic incentives and will be reviewed regularly; - be supported by Supplemental Liquidity Provider Firms. See NYSE Euronext Inc.: NYSE Aims to Maximize Market Quality and Competitiveness With Newly Approved Enhancements of Trading Model, in: News Releases, 24 Oct. 2008, 2008c, electronic source; and NYSE Euronext Inc.: Types of Members, no date, electronic source. 441 See NYSE Group Inc., Annual Report for the fiscal year 2005, 2006, p. 6; NYSE Group Inc., Annual Report for the fiscal year 2006, 2007, p. 12; and NYSE Euronext Inc., Annual Report for the fiscal year 2007, 2008, pp. 12–13.
C.II Reacting to Automation in the U.S. Securities Market
99
quarter of 2006 to the first quarter of 2007.442 In an interview in May 2008 a floor broker commented that of the former fifty brokers of his company on the floor only six remained. Some customers express their dissatisfaction with this fact. They miss the human involvement and the respective market quality that humans provide. On the other hand, some are of the opinion that the move towards the Hybrid Market was insufficient and was only an advancement of which the “intentions have been countered by the efficiency growth on other exchanges.443 Although the move towards electronic trading looks like a one-time event, it is rather an ongoing process. While automating, the NYSE initially aimed at a sub-second speed and reached times to execution of about 300 milliseconds. By 2009, they diminished to 5 milliseconds and by 2010 to about one.444 It as well as its competitors continue to increase speed. System response time, called latency in financial terms, has become a benchmark. It takes data about 7 milliseconds to travel from Chicago to New York. For this reason, in e-trade it has even become common to co-locate the trading computer of a large institutional investor in close proximity to Wall Street exchanges. An advantage of one millisecond is estimated to save a brokerage firm USD 100 million per year.445 Since 2006, by implementing the Hybrid Market and by further enhancing its market model the NYSE has joined this competition for speed. Apart from the NYSE’s alteration of its trading system, the organization underwent further changes. In 2006 it merged with the ECN Archipelago in order to additionally provide a purely electronic trading platform. This move also required the former nonprofit organization to demutualize and become a corporation. In line with a recent worldwide trend of stock exchange mergers, in 2007 it fused with the European Euronext exchanges, and it acquired the American Stock Exchange in 2008. This is part of a worldwide consolidation wave that represents a move towards more efficient structures.446 The NYSE receives revenues from listing fees, from the distribution of market data, from regulatory and facilities fees, and of course from trading fees.447 442
See Mehta, Nina: Traders Sound Off on the NYSE's Hybrid Market, in: Traders Magazine, Vol. 20 (2007), No. 270, issued 15 Jul. 2007, p. 21. The survey is based on replies by 87 respondents, representing a response rate of three percent. 443 ibid., p. 24. 444 See Bunge, Jacob: NYSE Cuts Trade Times to 5 Milliseconds, in: Wall Street Journal (2009), issued 2 Jul. 2009, p. C.4. 445 See Martin, Richard: Business At Light Speed, in: InformationWeek (2007), No. 1135, issued 23 Apr. 2007, p. 42. See also The Economist: Ahead of the tape; Algorithmic trading, in: The Economist, Vol. 383 (2007), No. 8534, issued 23 Jun. 2007, p. 99; and Strobl, Thomas: Das Herzkammerflimmern der Börse, in: Frankfurter Allgemeine Zeitung (2010), issued 18 May 2010, p. 29. 446 See Hasan, Iftekhar, Markku Malkamaki and Heiko Schmiedel: Technology, automation, and productivity of stock exchanges: International evidence, in: Journal of Banking & Finance, Vol. 27 (2003), No. 9, p. 1770. 447 See NYSE Euronext Inc., Annual Report pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 for the fiscal year ended December 31, 2009, No. 1-33392, New York 2010a, p. 42; and United States Securities and Exchange Commission (SEC), Market 2000, 1994, p. 12.
100
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
Furthermore, NYSE Euronext, as it is called now, expanded its range of products and services and now also trades funds, options, and futures. Yet while these business areas and changes are important for the NYSE, the focus of this work and of the subsequent system dynamics model will be limited to the exchange’s trading and order matching system as this represents its historical and actual core capability.
C.III Structure and Behavior of Forces for Retention and Change Both stability and adaptation characterize the behavior of the NYSE. In the securities industry, the automation of the order matching process often took place in the 1980s and 1990s, but it took the New York Stock Exchange about 20 years to follow suit. An investigation of the underlying forces will be able to shed light into this delayed and radical change. It will be interesting to analyze the process of how the forces causally emerge, interact and create the behavior seen at this organization. The system dynamics modeling technique used for mapping causal relationships of the respective problem and for detecting interacting forces uses a special pictogram language to distinguish different types of variables. Figure C-7 gives an example. A stock (or level) is surrounded by a box and describes the condition of a system. It accumulates over time and has an observable value at each point of time (Stockt ) as well as an initial value (INI STOCK or Stockto ). Equation C-1 illustrates the general computation of stocks. Rates—marked by a valve—determine the changes per time unit of the condition of a system, i.e. of the stocks. They represent the policies in a system.448 Rates are further broken up into the substructure of auxiliaries and constants as well as table functions that map the relationship of one variable onto another. Further labels of delays, polarities and feedback loops explained on page 12 in chapter B.I.1.a remain valid. rate
Stock
INI STOCK
Feedback Loop TIME CONSTANT
auxiliary
CONSTANT
Figure C-7: Diagramming conventions t
Stockt = Stockto + න ሺrates ሻds to
where rates indicates the rate flow at time t.
448
See Forrester: Principles of Systems, 1968, pp. 4-1–4-17.
C-1
C.III Structure and Behavior of Forces for Retention and Change
101
The pictogram language will be used to provide a graphical explanation or causal relations that support verbal statements. Additionally, equations of selected variables that cannot be derived intuitively will be presented. Equations for the entire model can be found in Appendix D. With the help of this modeling language, the following sub-chapters will present causal structures of the simulation model representing the principal factors determining the behavior of the NYSE.
C.III.1 A Perspective of Adaptation to the External Environment The rise in institutional investors as well as technological advancements that facilitate communication and allow for automation represent important developments in the U.S. stock exchanges’ environment. The fundamental changes in these areas point towards the environment as an essential driver of change; the working of the adaptation mechanism seems important. In the view of Stoll, competition from ECNs and exchanges forced the NYSE to adapt.449 Therefore, a system dynamics model will be developed that focuses on the effects of these processes on the NYSE. In order to fully capture these developments, time frames shown will cover the period from 1970 to 2030. This allows for the replication of developments before the NYSE adapted, and it also gives sufficient time to develop their ceteris paribus effects. PRESSURE FOR E-TRADE (-) Pressure by institutional customers
MGMT DECISION ON EXTENT OF E-TRADE
MARKET CHARACTERISTICS
(-) Pressure by noninstitutional customers
PRESSURE FOR FLOOR TRADE
Figure C-8: Sector diagram of the adaptation view The view of the NYSE as adapting to the market compares the focal organization’s characteristics with the market’s characteristics. It includes the pressure by customers who were summarized into two important groups: institutional customers who generally like speed and who exert pressure for e-trade as well as non-institutional customers representing those who have a desire for price quality and exert pressure for the floor. As Figure C-8 indicates, these two types of pressure determine the management team’s decision on e-trade. The view is simplified in the way that it focuses on customers who trade on the NYSE. Additionally, the NYSE may listen to the organizations that list their stocks at the NYSE. Since listed companies are particular449
See Stoll: Electronic Trading in Stock Markets, 2006, p. 153.
102
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
ly interested in price quality, they are included indirectly by aggregating them with customers who exert pressure for manual floor trading. The growth of large financial institutions had a significant impact on the NYSE and other exchanges. The institutions hold large amounts of equities and become customers of stock exchanges. Figure C-9 a) goes more into detail and shows this phenomenon in a full stock and flow diagram (SFD) of the system dynamics model. The fraction of equities held by institutions450, for which numbers were taken from data (see Figure C-10), has an effect on the indicated fraction of institutional customers that trade in the market. The effect of the capital distribution on customers is a concave function that is situated above the 45° line, indicating that institutions participate disproportionally frequently in trading. They have higher turnover rates than individual investors, and there is evidence that the percentage of retail participation relative to institutional investors has declined.451 In order to account for trading frequency, the fraction of institutional customers represents their participation in trading rather than their absolute number. It adapts with a time delay to the indicated fraction so as to neglect short-time changes and to account for the delay between the creation and the amendment of a portfolio of securities.452 When this fraction is multiplied with the total number of customers that is normalized to 100 customers, the resulting number of institutional customers represents the group impact and strength of large institutions trading at the NYSE. As the SFD in Figure C-9 b) indicates, the fraction of institutional customers serves as an input variable (shadow variable) in a further part of the model. A large fraction causes a high degree of trading professionalization, representing the first factor leading to the implementation of e-trade in the market.453 Even in the absence of institutional traders, professionalization starts at 10 percent because retail customers also require some professional strategies. It increases linearly in the beginning, and slowly approaches a fraction of 1 as soon as about 60 percent of customers are institutions and fully dominate trading.
450
Italic letters indicate the name of model variables as well as of feedback loops. See United States Securities and Exchange Commission (SEC), Market 2000, 1994, p. II-1. 452 The fraction of institutional customers is a ‘smooth’ or information delay of the indicated fraction of institutional customers. A ‘smooth’ always represents an exponential moving average of an input variable. See Forrester: Industrial Dynamics, 1961, p. 408. 453 See Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, pp. 146–153. 451
103
C.III Structure and Behavior of Forces for Retention and Change
a) FRACTION OF EQUITIES HELD BY INSTITUTIONS
effect of capital + distribution on customers
indicated fraction of institutional + customers
Fraction of Institutional change in fraction Customers of institutional + customers - (B)
TIME TO BECOME CUSTOMER
+
number of institutional customers +
TOTAL NO. OF CUSTOMERS
b) ACCESS TO INFORMATION TECHNOLOGY
degree of trading professionalization
TIME TO DEVELOP E-TRADE POSSIBILITIES
+ fraction of e-trade in remaining market -
+
TIME TO EXECUTION E-TRADE
-
time to execution in market
TIME TO EXECUTION FLOOR TRADE
Figure C-9: External influences in the remaining market (SFD) The access to information technology represents the second impact factor because it enhances possibilities for fast trading.454 Figure C-9 b) depicts its influence. It has been drawn as an s-shaped diffusion process that develops during the 1970s to 1990s.455 Together, professionalization and technology trigger the development of e-trade in the market. The implementation of the fraction of e-trade in the remaining market reacts with a 5-year delay that corresponds to the time it takes to develop e-trade possibilities.456 The grey box around the fraction of e-trade is an indicator that the variable is a smoothed average of its input. The computation of e-trade in the market is given below. The equation shows that both a professionalization of institutional customers as well as technological advancements are necessary for e-trade to develop.
454
See Gorham and Singh: Electronic Exchanges, 2009, p. 314; and Madhavan: Market microstructure: A survey, 2000, p. 206. 455 The diffusion of computing technology mainly took place during 1970s to 1990s. See Attewell, Paul: Technology Diffusion and Organizational Learning: The Case of Business Computing, in: Organization Science, Vol. 3 (1992), No. 1, p. 9; and Caselli, Francesco and Wilbur John Coleman II: Cross-Country Technology Diffusion: The Case of Computers, in: The American Economic Review, Vol. 91 (2001), No. 2, p. 329. 456 For example, at the Chicago Mercantile Exchange, there was a 5-year delay between the decision to implement an electronic platform in 1987 and its launch in 1992. At the NYSE, there was a 5-year delay between electronic trading for small orders and the full rollout of the Hybrid Market. At the Bourse de Paris, e-trade started to be implemented in 1986 and the exchange went fully electronic in 1991.
104
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
fraction of eTrade in remaining market [dmnl] = SMOOTH3(degree of trading professionalization [dmnl] • ACCESS TO INFORMATION TECHNOLOGY [dmnl], TIME TO DEVELOP eTRADE POSSIBILITIES [year])
C-2
The resulting model behavior of the fraction of e-trade in the remaining market is plotted in line 3 in Figure C-10. This model output matches real-world data. Line 1 of the same figure shows the fraction of the 22 largest foreign competitors of the NYSE allowing for at least some electronic trading, as it could be observed in reality.457 Assuming a time delay of five years to fully implement e-trade, line 2 reveals the resulting fraction of electronic trading among these foreign competitors. The model output (line 3) highly overlaps with this delayed function of the fraction of e-trade among foreign competitors (line 2). The comparison with data on the international market is legitimate because, first, trading has become an international phenomenon, and second, both U.S. and international exchanges went electronic at about the same time (see chapter C.II.2). The implementation of electronic trading in the market follows an s-shaped adaptation curve. This behavior is in line with a diffusion concept of electronic trading in the rest of the market. It is consistent with knowledge about technology diffusion in general.458
E-Trade in the Market from Data 1
Dmnl
0.75
123 0 123123 1970 1980
2
1
12312312312312 23
3
3
0.5 0.25
1
1
2 1
2
3
1990
2000 Date
2010
2020
2030
"Fraction of largest foreign competitors allowing some e-trade" : adaptation 1 "fract. of e-trade among foreign competitors" : adaptation 2 2 2 "fraction of e-trade in remaining market" : adaptation 3 3 3 3
Figure C-10: Diffusion of electronic trading in the securities market (BOT) Figure C-11 summarizes the aggregate causal structure of the main outside influences on the system in a simplified causal loop diagram (CLD). New technological 457
Data has been taken from Jain: Financial Market Design and the Equity Premium, 2005, p. 2960. 458 For information on the diffusion of innovations see Milling: Modeling innovation processes for decision support and management simulation, 1996, pp. 215–218; Milling: Understanding and managing innovation processes, 2002, pp. 75–80; and Rogers: Diffusion of Innovations, 2003, pp. 136–157.
C.III Structure and Behavior of Forces for Retention and Change
105
possibilities and the change in customer preferences went hand in hand and contributed to the growing importance of electronic trading. Storkenmaier and Riordan also emphasize the changes in the securities trading industry, new information technology as well as the resulting electronic systems and use of algorithmic trading as important for the decision to introduce the Hybrid Market.459 ACCESS TO INFORMATION TECHNOLOGY
FRACTION OF EQUITIES HELD BY INSTITUTIONS
+ fraction of e-trade in remaining market +
+
Fraction of Institutional Customers
Figure C-11: External influences (CLD) Large institutional investors’ preferences and trading strategies often differ from those of individual clients. Many institutions value a good execution speed (time to complete a trade, time to execution) higher than receiving the best possible price.460 According to an NYSE specialist as well as a floor broker who were interviewed, there is a large proportion of customers that care about milliseconds. The use of arbitrage, algorithmic trading, and the preference for anonymity went hand in hand with a preference for speed and electronic trading in the whole market.461 Apart from price quality, the time to complete a trade became highly important. The NYSE’s competitors which had already implemented electronic trading could offer a much higher speed. As the SFD in Figure C-12 demonstrates, this also meant that before the NYSE’s transformation its relative time to execution in comparison with other market players was very high. Even with all possible technical support, manual floor trading involves a specialist decision and cannot be done in less than about nine seconds. Electronic trading is fast; and a market classifies as a fast market if the time to execution is one second or less.462 Hence, in the system dynamics model the fraction of electronic trading is the major factor determining the time to execution, both at the NYSE and in the remaining market. The NYSE’s slow order handling in combination with the institutional customers’ preference for speed led to dissatisfaction among these powerful clients.463 In proportion with the relative time to execution institutional customers develop dissatisfaction with time per institutional customer and then exert 459
See Storkenmaier and Riordan: The Effect of Automated Trading on Market Quality, 2009, p. 11. 460 See Hendershott and Moulton: Speed and Stock Market Quality, 2009, p. 1. They also refer to Boehmer, Jennings and Wei: Public Disclosure and Private Decisions, 2007, pp. 252–253. 461 See Hendershott, Terrence and Pamela C. Moulton: The Shrinking New York Stock Exchange Floor and the Hybrid Market, Working Paper, 2007, p. 24; and Lucchetti: Nasdaq Passes NYSE in Sales Milestone, 2008, p. C.3; and Lucchetti: NYSE Plans to Revise SpecialistTrader Rules, p. C.4. 462 See NYSE Euronext Inc.: Hybrid Market: Key Attributes, 2007, electronic source. 463 See Tabb: The NYSE Floor, 2005, p. 54; and Tully, Shawn: Bringing Down the Temple, in: Fortune, Vol. 148 (2003), No. 10, issued 10 Nov. 2003, p. 120.
106
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
pressure for more e-trade.464 The pressure may also be interpreted as a customer desire that develops into pressure for the NYSE since its competitors have already implemented electronic trading. The total pressure for more e-trade from customers results from the pressure per customer and its multiplication with the size, i.e. the strength and importance of the institutional customer group. REF. PRESSURE PER DISSATISF. UNIT
dissatisf. with time + per inst. customer
-
time to execution in market
-
relative time to execution (B) Customer Pressure for Speed
+
TIME TO EXECUTION E-TRADE
NYSE time to execution +
TIME TO EXECUTION FLOOR TRADE
NYSE Fraction of Floor Trade
+ pressure for more e-trade per inst. customer
+ total pressure for more e-trade from customers +
NYSE Fraction of E-Trade
pcvd pressure for more e-trade
change in fraction + of e-trade + (B) (B) + effect of floor effect of e-trade trade on change on change
Figure C-12: Customer pressure for e-trade (SFD) The pressure triggers a decision about automation among the NYSE management. Since the NYSE adapts to pressure from its environment, management is assumed to perceive this pressure in an unbiased way so that (for now) the perceived pressure for more e-trade equals the total pressure for more e-trade from customers. Perceived pressure serves as a measure for the management team’s awareness that its strategy meets resistance and is disaligned with desired from stakeholders. Effect of E-Trade on Change 1
effect
0.75 0.5 0.25 0 0
0.10
0.20
0.30 0.40 0.50 0.60 0.70 "NYSE Fraction of E-Trade"
0.80
0.90
1
Figure C-13: Limiting effect of e-trade on change The difference in perceived pressure for e-trade and for floor trade multiplied with the fractional change per perceived pressure per year generates the change in the 464
See Blake, Rich: NYSE Technology Threatens Survival of Floor Traders, in: Institutional Investor, Vol. 43 (2009), No. 6, issued Jul./Aug. 2009, p. 29.
107
C.III Structure and Behavior of Forces for Retention and Change
fraction of e-trade. This amount of change is limited by the current NYSE fraction of e-trade through the effects of e-trade and floor trade on change. In the rare case the fraction of e-trade (floor trade) has almost reached one, but the management team still perceives pressure for more e-trade (floor trade), it would not be willing any more to fully react to this pressure. This is shown in detail in Figure C-13 which reveals the nonlinearity between the fraction of e-trade on the abscissa and its effect on change on the ordinate. The effect of floor trade works in an analogous manner. Equation C-3 illustrates the computation of the change in the fraction of e-trade. change in fraction of eTrade [dmnl / year] = (pcvd pressure for more eTrade [pressure unit] • effect of eTrade on change [dmnl] C-3 – pcvd pressure for more floor trade [pressure unit] • effect of floor trade on change [dmnl]) • fractional change per perceived pressure [dmnl / pressure unit / year] Since change determines the NYSE fraction of e-trade, the balancing feedback loop Customer Pressure for Speed is closed. Its simplified causal structure is shown in Figure C-14 in order to illustrate how this feedback loop fits in with the model elements explained before. Depending on the fraction of e-trade, dissatisfaction with the time to execution may develop and institutional customers exert pressure for the implementation of electronic trading. When e-trade is implemented the relative speed of the NYSE vs. that of its competitors improves and dissatisfaction decreases. ACCESS TO INFORMATION TECHNOLOGY
+ fraction of e-trade in remaining market + Fraction of Institutional + Customers
FRACTION OF EQUITIES HELD BY INSTITUTIONS
+ rel. time to execution -
+ (B) Customer Pressure for Speed
dissatisfaction with time per inst. customer + pressure for more e-trade
NYSE Fraction + of E-Trade
Figure C-14: Customer pressure for e-trade (CLD) The move towards electronic trading has consequences among those customers that prefer price quality to speed, i.e. among retail customers in particular. Noninstitutional customers such as individuals who directly invest in securities have no interest in fast executions since they do not engage in arbitrage and algorithmic trading, but they have a high preference for receiving the best price.465 The analysis of the NYSE weblog reveals that after the introduction of e-trade a customer group showed dissatisfaction with the missing price improvement, and it demanded more specialist participation. Two traders commented: 465
See Schwartz, Robert A. and Robert A. Wood: Best Execution, in: Journal of Portfolio Management, Vol. 29 (2003), No. 4, p. 38; Securities and Exchange Commission (SEC): Regulations NMS, 2005, p. 37498; and Shapiro: U.S. Equity Markets, 1993, p. 5.
108
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
“[W]e need the specialists to add liquidity, and stabilize the price's. The value the specialists add is what makes the NYSE the best place to trade and its what makes the exchange different from the other exchanges [sic!].”466 “Your customers want the ability to execute [orders in large] blocks, be price improved, and not be charged an arm and a leg to do it. We also want the specialists to create a better quality market […].”467 “[…H]opefully the NYSE management is … finding out just how dissatisfied traders are with the new market.”468 The impact of the rise of electronic trading on specialist participation can be seen in Figure C-15. Before the implementation of the Hybrid Market between October 2006 and March 2007, the NYSE was able to offer high price quality to its customers because its specialist participation used to range between 10 and 15 percent. Since a rise above 10 percent during the years 2001 to 2003 is attributed to decimalization,469 i.e. to quoting in pennies, a maximum or reference specialist participation of 10 percent is used. When the trading mechanism was automated, specialist participation decreased to below 3 percent.470 Market quality deteriorated and the fraction of time the NYSE quoted at the national best bid and offer (NBBO) fell sharply from more than 60 to 10 percent.471 Specialist participation stands for the involvement of floor brokers as well since the entire floor is connected to the manual trading model.
Specialist Participation from Data 0.12
implementation of Hybrid Market
Dmnl
0.08
0.04
0 2004
2005
2006 Time (Year)
2007
2008
Figure C-15: Specialist participation from data (BOT) 466
Dey: Comment on: Men At Work (at re-making the markets), 2007, electronic source. Dey: Comment on: 'The floor is not going away. OK?', 2007, electronic source. 468 Brandon: Comment on: When a quote goes 'slow,' speed vs. price is for customer to decide (11 Jan. 2007, commented on 16 Jan. 2007), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2007, electronic source. 469 See Lucas Jr., Oh and Weber: The defensive use of IT in a newly vulnerable market, 2009, p. 5. 470 See NYSE Euronext Inc.: Specialist Activity, no date-l, electronic source. 471 See NYSE Euronext Inc.: Significant Progress in New NYSE Market Model, in: U.S. Equities News (2009d), issued Apr. 2009, p. 1; and Clark: Growth in the NYSE's Liquidity Provider Programs, 2009, electronic source. 467
109
C.III Structure and Behavior of Forces for Retention and Change
As Figure C-16 illustrates by an SFD, non-institutional customers criticize the missing specialist participation and market quality. With a rise in electronic execution, the involvement of specialists in trading (specialist participation) proportionally decreases, e.g. since computer programs take over the management of (block) trades. According to a study by Storkenmaier and Riordan automation of the NYSE causes a decrease in floor broker and specialist participation and a reduction of price improvement.472 Thus, in line with the reduction of their activity, the market quality from specialist participation diminishes. This variable aggregates the extent of price improvement and stabilization, volatility reduction, and liquidity improvement. There is a diminishing marginal utility of the floor participation so that the effect of specialist participation on market quality is drawn as a slightly concave function. Market quality develops into a floating goal: it is compared with desired market quality that adapts with a time delay to the actual quality. This smoothed goal is marked by a grey box around the variable and represents a smoothed average of the actual market quality. Expressed formally and in general terms, a desired variable, i.e. a goal gt is a function of its past value gt–W, the actual value at, and an adjustment time W. at – gt–W
gt = gt–W +
C-4
W
If market quality falls and customers perceive an inadequacy, dissatisfaction rises. As a consequence, non-institutional customers start pressuring for more floor trading. The total pressure for more floor trade from customers also depends on the strength or size of the non-institutional customer group as well as their reference pressure. It is set to 0.5, half the value for institutions, in order to capture the concentration of power which is lower for fragmented private clients than for financial institutions.473 +
(B) Customer Pressure for Market Quality
effect of floor trade on sp. participation specialist participation + REF. SP PARTICIPATION
effect of sp. participation on market quality
TIME TO CHANGE DESD MARKET QUALITY desired market quality from sp. part. by customers + + (R) pcvd adequacy of market quality from market quality by sp. participation customer
<no of non-institutional customers> + total pressure for more floor trade from customers + pressure for more floor trade per customer + + dissatisfaction effect of REF. PRESSURE market quality on PER NON.INST. pressure CUSTOMER +
REF. MARKET QUALITY
Figure C-16: Customer pressure for market quality (SFD) 472
See Storkenmaier and Riordan: The Effect of Automated Trading on Market Quality, 2009, p. 22. 473 See Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 108; and Ryan and Schneider: Institutional Investor Power and Heterogeneity, 2003, pp. 399 and 416–417.
110
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
As is shown in an aggregate CLD in Figure C-17, the mechanisms explained above create a balancing feedback loop called Customer Pressure for Market Quality. A perceived inadequacy of market quality from a falling fraction of floor trade creates customer pressure for more floor trade. The implementation of electronic trading thus triggers pressure for the retention of the specialist system. ACCESS TO INFORMATION TECHNOLOGY
+ fraction of e-trade in remaining market +
+ rel. time to execution -
Fraction of Institutional + Customers FRACTION OF EQUITIES HELD BY INSTITUTIONS
+ (B) Customer Pressure for Speed
dissatisfaction with time per inst. customer + pressure for more e-trade
NYSE Fraction + of E-Trade specialist participation + market quality from specialist
pressure for more floor (B) Customer Pressure for Market + Quality pressure for floor - from customers
Figure C-17: Customer pressure for floor trade (CLD) Competition for order flow takes place on price and speed.474 Since the two types of customer pressure represent environmental drivers of organizational change, the NYSE can be regarded as adapting to the external forces of the market. Figure C-18 reveals the output produced when the model presented so far is simulated. The behavior over time (BOT) graph shows a quick adaptation of the NYSE (line 2) to the trend in the market (line 1). Since the customers and the NYSE react to what they perceive in the remaining market, there is an implementation delay of about two years. Its length depends on how strongly the NYSE reacts to pressure, i.e. on the fractional change per perceived pressure p.a.475 When change starts to become implemented, non-institutional customers perceive the disadvantages in market quality and start pressuring for the retention of the old system (line 4), creating a deceleration of the implementation of e-trade at the NYSE. The non-institutional customers’ pressure vanishes after some years since they become used to the new situation and since their number decreases.
474 475
See Boehmer, Jennings and Wei: Public Disclosure and Private Decisions, 2007, p. 318. Here, the fractional change per perceived pressure per year was set to 0.1 which can be interpreted as a fractional change of 0.5 per year (i.e. an implementation delay of two years) and a fractional change per perceived pressure of 0.2. The more willing the organization is to react to pressure (meaning the higher the reference fractional change in trading per pressure per year), the quicker the implementation of e-trade.
111
C.III Structure and Behavior of Forces for Retention and Change
NYSE Fraction of E-Trade 1 Dmnl 6 pressure unit
1
0.5 Dmnl 3 pressure unit
1 2
34
12
12 2
12
3
4
3 3
0 Dmnl 12 3 4 0 pressure unit 123 4123 4 1970 1980 1990
4 2000 Date
2010
1 "fraction of e-trade in remaining market" : adaptation "NYSE Fraction of E-Trade" : adaptation 2 2 "total pressure for more e-trade from customers" : adaptation total pressure for more floor trade from customers : adaptation
1
4 2020
3 4 2030
1
2
Dmnl Dmnl 2 3 pressure unit 4 pressure unit
Figure C-18: Adaptation view (BOT) This adaptive behavior does not yet give an adequate picture of reality. Even a further run with an assumed increase in the implementation time does not produce realistic behavior; it only results in a time-delayed, but smooth adaptation. Change happened later and more radically. Remembering the difference between the simulation results and the observed radical implementation of e-trade mainly between 2006 and 2007, the environment may be an important driver for change, but impediments to change deserve closer investigation. For this reason, the pressure for the floor system will now be elaborated.
C.III.2 An Endogenous Struggle of Culture and Resistance The pressure for the NYSE trading floor that prevented the adaptation to environmental drivers of change also originated from the floor firms (i.e. from specialists and floor brokers and their respective firms). They, first, met automation attempts with resistance, second, dominated the stock exchange’s culture, and third, they were powerful. These influences by stakeholders are added in bold face in the sector diagram of Figure C-19.
112
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
PRESSURE FOR E-TRADE (-) Pressure by institutional customers
MGMT DECISION ON EXTENT OF E-TRADE
MARKET CHARACTERISTICS
(-) Pressure by non-inst. cust. (-) Resistance by floor (+) Cultural pressure by floor (+) Power of floor
PRESSURE FOR FLOOR TRADE
Figure C-19: Sector diagram of the culture and resistance view First, the floor’s opposition to electronic trading is able to create a balancing mechanism Floor Resistance that diminishes the implementation of e-trade once it is launched. It has been said that the electronic market hypothesis or disintermediation effect will be limited by the extent at which there is resistance and by the extent at which intermediaries form cooperative groups.476 In 1971, an attempt to launch an Automated Trading System at the NYSE that would execute orders of 100 or less shares without human intervention was literally torn into pieces. On the weekend before its commissioning someone destroyed its cover with a saw. The system was perceived as a threat to the floor, and nobody really wanted it to work.477 In the years following, the Securities Industry Automation Cooperation (SIAC) which provides automation and data processing services to the NYSE made several attempts to automate the floor. These attempts usually were not very successful.478 A floor broker who was interviewed reported a rather hostile reaction of the floor to electronic trading. Formally, the development of resistance is expressed by the linkage of employability and income. Resistance arises when floor brokers and specialists have fewer possibilities to participate in trading and at the same time feel restrictions in their income. Floor firms earn money from the commissions they receive for trading, specialists also from the spread between their bid and offer (ask) price, and from trading for their own account.479 They could maintain their cartel for fixed commissions until the first 476
See Clemons and Weber: Information Technology and Screen-Based Securities Trading, 1997, p. 1706; and Picot, Bortenlänger and Röhrl: Organization of Electronic Markets, 1997, p. 115. Gasparino also expects floor firms to react with resistance. See Gasparino: King of the Club, 2007, p. 318. 477 See Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, pp. 194–198. 478 See Keith, Christopher and Allan Grody: Electronic Automation at the New York Stock Exchange, in: Guile, Bruce R. and James Brian Quinn (Ed.): Managing Innovation: Cases from the Services Industries, Washington, DC 1988, p. 92. 479 Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 40; and Conroy and Winkler: Market Structure, 1986, p. 22.
113
C.III Structure and Behavior of Forces for Retention and Change
half of the 1970s, but afterwards institutional investors gained in influence and commissions declined.480 As Figure C-20 a) illustrates, due to the power and number of institutional investors, the commission per share diminished. This is also illustrated in Figure C-20 b), comparing model output with historical data. While brokers still receive commissions, designated market makers are not any more allowed to charge them. Shrinking commissions negatively affect the floor’s earnings per share. a)
-
REF. COMMISSION PER SHARE + + commission per share
effect of inst. customers on commission
effect of inst. - customers on spread
+ NYSE spread + REF. SPREAD
+ +
floor earnings per share handled HALF SPREADS
b)
NYSE Commissions 1
$/share
0.75
12
data
12
0.5
12 12
0.25 0 1970
1980
model output
12
1990
1
2
12 2000 Time (Year)
commission per share from data : adaptation 2 commission per share : adaptation 2
1
2 2 2010 1
2
1 2
2
2 2020
2
1
1 2
2 2030 1 2
Figure C-20: Commissions and spread (SFD and BOT)481 Additionally, the effect of institutional investors on spread diminishes earnings. In reality this often took place piecemeal in events, such as the transitions from quoting
480 481
See Abolafia: Making Markets, 1996, p. 109; and Gasparino: King of the Club, 2007, p. 48. Data for commission per share was derived by dividing the floor firms’ entire income from commissions by the NYSE share volume. The numbers match specifications about the size of commissions by Tinic and West as well as Tully. See NYSE Euronext Inc.: Annual reported volume, turnover rate, reported trades (mils. of shares), no date-a, electronic source; NYSE Euronext Inc.: Income statement for NYSE member firms ($ in mils.), no date-g, electronic source; Tinic, Seha M. and Richard R. West: The Securities Industry under Negotiated Brokerage Commissions: Changes in the Structure and Performance of New York Stock Exchange Member Firms, in: The Bell Journal of Economics, Vol. 11 (1980), No. 1, p. 36; and Tully: Bringing Down the Temple, 2003, p. 126.
114
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
in eights of a dollar to sixteenth to pennies, but in the model this is a continuous effect. The average NYSE spread declined from about USD 0.20 to USD 0.03.482 a) (B) Floor Resistance from Profitability
<specialist participation> + proportional floor earnings per share traded + +
+ pcvd adequacy of profitability (R)
-
effect of profitability on resistance
total pressure for more floor trade from floor + +
resistance pressure for floor system per floor firm + +
desired earnings per share
TIME TO ADJUST DESIRED EARNINGS
<effect of employability on resistance>
REF. RESISTANCE PRESSURE PER FLOOR FIRM
b) Effect of Profitability on Resistance 1
effect
0.75 0.5 0.25 0 0
0.10
0.20
0.30 0.40 0.50 0.60 0.70 pcvd adequacy of profitability
0.80
0.90
1
Figure C-21: Resistance pressure from floor (SFD and effect) A decline in spread and commissions reduces the floor earnings per share. The spread captures the liquidity providers’ revenues.483 It is divided by two, i.e. by the number of half spreads since a floor participant does not necessary deal with both the buyer and seller. As can be seen in Figure C-21 a), when multiplied by the percentage at which the specialist participates in trades, the proportional specialist earnings result, which provide information on revenues per share traded. A perceived inadequacy of profitability leads to resistance pressure for the floor system. Resistance pressure for the floor system rises when adequacy falls below its normal level of 1, as Figure C-21 b) exhibits in more detail. Since there may always be some variation, it increases somewhat slower in the beginning. Since floor firms need a certain extent
482
See NYSE Euronext Inc.: Average NYSE Spreads, (in dollars, rounded to penny) (1994–2003), no date-b, electronic source. See also Huang and Stoll: Dealer versus auction markets, 1995, p. 323; and Bennett and Wei: Market structure, fragmentation, and market quality, 2006, p. 62. 483 See Storkenmaier and Riordan: The Effect of Automated Trading on Market Quality, 2009, p. 23.
115
C.III Structure and Behavior of Forces for Retention and Change
of participation in order to remain in the business, resistance is already fully present when employability and thus its perceived adequacy fall to 50 percent. Yet, by the above formulation floor firms may also show opposition when their income decreases for other reasons than automation. They would thus only resist electronic trading if they could attribute the loss in their ability to earn money to lower possibilities of employment. Specialists and floor brokers are “less and less required to arrange trades” in an automated environment.484 It is not surprising that they showed resistance against the implementation of e-trade which would reduce the specialist participation and ability of specialists and floor brokers to carry on their profession. Figure C-22 a) illustrates the structural relationships in an SFD. Structurally this floor resistance from employability is highly similar to resistance from profitability. The floor’s perceived adequacy of employability adapts to a floating goal of desired specialist participation. In the same manner as it works with profitability (see Figure C-21 b), resistance pressure for the floor system may rise when adequacy falls below its normal level of 1. a) <specialist participation>
total pressure for more floor trade from floor +
(B) Floor Resistance from Employability
+ (R)
+
desired specialist participation
- pcvd adequacy of employability
resistance pressure for floor system per floor firm + +
effect of + employability on resistance
TIME TO ADJUST DESIRED PARTICIPATION
REF. RESISTANCE PRESSURE PER FLOOR FIRM
b) ACCESS TO INFORMATION TECHNOLOGY
+ fraction of e-trade in remaining market +
+ rel. time to execution -
Fraction of Institutional + Customers FRACTION OF EQUITIES HELD BY INSTITUTIONS
+ (B) Customer Pressure for Speed
dissatisfaction with time per inst. customer + pressure for more e-trade
NYSE Fraction + of E-Trade specialist participation
pressure for more floor +
(B) Resistance -
+
profitability of floor
-
resistance pressure for floor
Figure C-22: Resistance pressure from floor (SFD and CLD) The effects of employability and profitability are multiplied so that pressure for more floor trade only develops if low profitability results from the floor’s low participa484
ibid., p. 28.
116
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
tion in trading. Both effects together create the balancing feedback loop Floor Resistance which is shown in Figure C-22 b). As soon as e-trade is implemented, resistance builds up from the floor to reduce the amount of automated trading and to increase the floor’s involvement again. Second, a strong floor culture developed during the 20th century at the NYSE and helped exert pressure for the retention of the floor. The stock and flow structure in Figure C-23 a) shows how floor firms value the floor culture depending on their ability to make profits and on their merits in providing market quality. Concerning market quality, people on the floor strongly believed in the superiority of their manual system. “We’re a national asset.”485 They were convinced that only the auction process which involves the floor brokers and specialists is able to generate high levels of price improvement. Additionally, due to their high position and commissions, the specialist (and broker) profession used to be a license to make money.486 This further contributed to the floor’s valuation of its trading system. This culture has also emerged, has been deeply embedded for many decades, but may also decline. To the extent either the perceived adequacy of market quality or of profitability is below the value one, the floor reduces its valuation of the floor culture. The valuation is modeled as an accumulation because the strength of the floor culture only faded slowly, although its profitability steadily declined with the rising power of institutional investors, decimalization, and others. The valuation of the floor culture divided by its reference value serves as an indicator for the importance of cultural aspects, i.e. the relative valuation of the floor culture. Since individual valuation does not produce concerted action, the relative value needs to be adjusted by the cohesiveness of floor firms to determine the cultural multiplier of pressure from the floor. The latter variable moderates the strength of the resistance pressure discussed on the previous pages. The cohesiveness of floor firms is important and floor firms had established long-term relationships based on trust, but concerted action was not as high as, for example, among members of an orchestra who depend on each other for a good individual outcome. Despite individual valuation of the floor, high valuation does therefore not fully translate into concerted resistance, and resistance is reduced to about 70 percent of its potential value. As is depicted in Figure C-23 b), this closes a reinforcing loop from which many decades ago high pressure for the floor developed which then diminished only slowly due to the imprinting of culture.
485
Cited in Abolafia: Making Markets, 1996, p. 104, see also p. 133; and Tabb: The NYSE Floor, 2005, p. 54. 486 See Abolafia: Making Markets, 1996, p. 131; and Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 25.
117
C.III Structure and Behavior of Forces for Retention and Change
a) + effect of market quality on culture + + effect of profitability on culture
(R) Pressure from Floor Culture
+
fractional change in valuation of floor culture +
+
total pressure for more floor trade from floor +
cultural multiplier of pressure from floor + + Valuation of rel. valuation of Floor Culture + floor culture change in + by Floor + valuation - (R)
REF. FRACTIONAL CHANGE OF VALUATION PER YEAR
REF. VALUATION OF FLOOR CULTURE
DEGREE OF COHESIVENESS OF FLOOR FIRMS
b) ACCESS TO INFORMATION TECHNOLOGY
+ fraction of e-trade in remaining market +
+ rel. time to execution -
Fraction of Institutional + Customers FRACTION OF EQUITIES HELD BY INSTITUTIONS
+ (B) Customer Pressure for Speed
dissatisfaction with time per inst. customer + pressure for more e-trade
NYSE Fraction + of E-Trade specialist participation + market quality from specialist + - profitability of floor
pressure for more floor +
(R) Pressure from Floor Culture
+ +
Valuation of Floor Culture by Floor
Figure C-23: Cultural pressure from floor (SFD and CLD) Third, floor firms’ power constitutes a remarkable factor that determines the floor’s influence as well as the exerted pressure for the floor. Specialists in particular used to be the force at the exchange. In the beginning of the 1990s, they NYSE was still described as “wedded to the concept of the specialist market”487, but the floor’s power had begun to decline.488 According to an interviewee, during the months following the implementation of the Hybrid Market, one of the specialists attached around 20 stickers to his suit showing his opposition to electronic trading and to the way it was implemented. Although he used to be a member of the board of executives, his opposition was powerless at that point of time because the floor firms’ influence had already deteriorated. Picot et al. state that these firms’ power can prevent electronic trading. They define the floor’s power particularly by the volume the firms attract to the exchange.489 The floor’s ability to attract volume can be operationalized by the market quality floor brokers and in particular specialists provide, i.e. the market quality from 487
Clemons and Weber: London's big bang, 1990, p. 52. See Abolafia: Making Markets, 1996, p. 129. 489 See Picot, Bortenlänger and Röhrl: The Automation of Capital Markets, 1995, electronic source, no page. 488
118
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
specialist participation in Figure C-25 a). Due to the absence of more detailed information, it is assumed that market quality raises the indicated power of floor firms in a proportional way. As a matter of completeness, the indicated power is also shaped by the effect of institutional customers on power. Over the last decades of the 20th century, the power of floor firms diminished in the manner institutional investors’ importance grew—particularly because they represent a large amount of order flow. Some floor firms were even taken over by institutional customers. The power of floor firms relative to that of institutions thus declines. This put specialists under pressure, and their role in providing liquidity was crowded out by computer-based trading.490 Figure C-24 shows that the existence of some institutional trading has no significant effect on the indicated power—as could be observed by about the end of the 1970s. The effect becomes more severe as the fraction of institutions rises. The increasingly negative slope of the relationship reflects what could be observed in reality. A possible effect that changing securities market regulations had on the power of the floor is outside the boundary of this investigation.491 Effect of Institutional Customers on Power of Floor Firms 1
effect
0.75 0.5 0.25 0 0
0.10
0.20
0.30 0.40 0.50 0.60 0.70 Fraction of Institutional Customers
0.80
0.90
1
Figure C-24: Effect of institutional customers on power of floor firms Floor firms have the same potential power (reference power of floor firms) as customers. As described in the beginning of chapter 0, the customers’ power is expressed by their group size or share of trading which is normalized to 100. Therefore, the reference power of floor firms also takes the value of 100. Although both groups have the same potential power, the real power distribution between the two groups is determined by the specific circumstances represented e.g. by the effects which shape the indicated power in the model (Figure C-25 a on page 119). The power of floor firms adapts with a time delay of two years to this indicated power of floor firms. Since the power represents the firms’ usefulness and legitimacy for the trading business, it shapes the total pressure for more floor trade from the floor. 490
See Abolafia: Making Markets, 1996, p. 131; and Lucas Jr., Oh and Weber: The defensive use of IT in a newly vulnerable market, 2009, p. 5. 491 For an analysis of the effects of regulation on the power of specialists see Abolafia: Making Markets, 1996, pp. 111–128.
119
C.III Structure and Behavior of Forces for Retention and Change
total pressure for more floor trade from floor [pressure unit] = resistance pressure for floor system per floor firm [pressure unit / entity] • cultural multiplier of pressure from floor [dmnl] • Power of Floor Firms [entity]
C-5
A decrease in market quality resulting from the introduction of electronic trading thus diminishes power and then the pressure for the floor. This creates the reinforcing mechanism Power shown in Figure C-25 b). It reveals that once electronic trading is initiated, the decreasing power of floor firms makes it easier to implement more. a) <market quality from sp. participation>
(R) Floor Power + effect of market quality on power
<cultural multiplier of pressure from floor> Power of
+
total pressure + for more floor trade from floor + +
effect of institutional customers on power -
indicated power of floor firms + + REF. POWER OF FLOOR FIRMS
Floor Firms + change in power of floor firms (B)
TIME TO CHANGE POWER OF FLOOR FIRMS
b) ACCESS TO INFORMATION TECHNOLOGY
+ fraction of e-trade in remaining market +
+ rel. time to execution -
Fraction of Institutional + Customers FRACTION OF EQUITIES HELD BY INSTITUTIONS
dissatisfaction with time per inst. customer +
+ (B) Customer Pressure for Speed
pressure for more e-trade
NYSE Fraction + of E-Trade specialist participation + market quality from specialist
pressure for more floor
(R) Floor Power
+
Power of Floor Firms
+ -
Figure C-25: Power of floor firms (SFD and CLD) The model now incorporates drivers of change from the environment as well as impediments to these drivers from exchange-related stakeholders. These impediments are the product of the floor firms’ resistance, their culture, and power. When the model is simulated, the implementation of electronic trading is shaped by the pressure for and against electronic trading from the floor and from customers. Line 3 of Figure C-26 reveals that the pressure coming from the valuation of the floor culture and from resistance is influential and that it can defer and diminish the implementa-
120
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
tion of electronic trading. Nevertheless, the system dynamics model is still not able to reproduce a behavior similar to the real course of events.
NYSE Fraction of E-Trade 1
1
Dmnl
0.75
3
12 1231231 3
3
12
0.5
12 2
3 12
0.25 0 123123123 1970 1980
3
1 23 1990
2000 Time (Year)
"fraction of e-trade in remaining market" : adaptation "NYSE Fraction of E-Trade" : adaptation 2 2 3 3 "NYSE Fraction of E-Trade" : pressures
2010
2020 1
2
1 2
3
2030 1
2 3
2 3
3
Figure C-26: Culture and resistance (BOT) So far it has been assumed that management implemented whatever the joint pressure of customers and stakeholders called for. Since it could be ruled out that the observed radical shift is based on sudden changes in the forces for floor and e-trade only, the role of management as a driver of electronic trading is worth consideration.
C.III.3 Managerial Impact on Change How the management team reacts to pressure from its environment is also shaped by the recognition and interpretation of this pressure which—as Figure C-27 displays—depends on the organization’s market share and inertia.
MARKET SHARE
(+) Market share from market quality
(-) Market share from speed
(-) Pressure by institutional customers
MGMT DECISION ON EXTENT OF E-TRADE
(+) Repetitive momentum (+) Repetitive attention (customer orientation)
INERTIA
PRESSURE FOR E-TRADE
MARKET CHARACTERISTICS
(-) Pressure by non-inst. cust. (-) Resistance by floor (+) Cultural pressure by floor (+) Power of floor
PRESSURE FOR FLOOR TRADE
Figure C-27: Sector diagram of the management view
C.III Structure and Behavior of Forces for Retention and Change
121
“A few years ago, the NYSE, owned by NYSE Euronext, was considered a dinosaur in the increasingly electronic exchange universe, hobbled by its slowmoving, human-run system and outdated infrastructure.”492 "Part of being very successful for a very long time and having a large market share [is that] the New York Stock Exchange did become complacent. We have to be receptive to change. We have to give our customers what they're looking for."493 The statement by the NYSE CEO reveals that since about the year 2005 the New York Stock Exchange management regards its past success and the resulting complacency as reasons for the missing reaction to changing demands in its environment. The past success and high levels of inertia made the NYSE inattentive so that the organization focused on what it had always done. The NYSE was not known to be an innovative organization.494 Inertia could grow over a long period of several decades because the NYSE had not changed the basic principles of its way of trading for about 100 years. During this convergence period, there was much opportunity for the institutionalization of routines and the growth of consistency. This idea is represented by the institutionalization rate in the SFD of Figure C-28 a). Inertia increases in a reinforcing way by a reference fraction of about 0.3 per year, but a further process limits inertia to a maximum value of 100 percent or 1. Additionally, fluctuation and an ongoing process of rethinking make an organization lose its inertness. An NYSE interviewee reported that new employees in middle and high positions were “grown from within the organization” and seldom came from outside bringing fresh ideas. While fluctuation in general reduces inertia, inner-organizational replacements hardly do. The NYSE’s fluctuation and inertia decrease rate is therefore assumed to be only 15 percent per year, which equals half the recent fluctuation rate in the financial sector of about 30 percent.495 The resulting inertia led to the observed complacency.
492
Pellecchia, Ray: Faster, Faster: High Tech Hits the NYSE Floor (29 Mar. 2010), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2010, electronic source. 493 Statement by John Thain, CEO, quoted in Ewing: When CEOs Talk to Each Other, 2005, electronic source. 494 See Thain, John: Statement on A CEO's Views on IT and Innovation (28 Jun. 2007), in: McGee, Ken: Gartner Fellows Interviews, 2007, electronic source. 495 Data is based on the rate of total US hires in the finance and insurance industry for the three years before the CEO of the NYSE changed. See Bureau of Labor Statistics, U.S. Department of Labor: Job Openings and Labor Turnover Survey (JOLTS), 2009, electronic source.
122
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
a)
b) REF. OPENNESS PER INERTIA +
REF. FRACT. institutionalization INSTITUTIONALIZATION + + + (R) (B) limiting effect on institutionalization
Inertia (B)
REF. FRACT. INERTIA DECREASE
+
+ inertia decrease
institutionalization + (R) Inertia (B)
+ inertia decrease + effect of change on inertia NYSE Fraction of Floor Trade
openness to change
+
effect of openness on change
(R) Repetitive Momentum REF. FRACT. CHANGE IN TRADING PER PRESSURE P.A. + + fract. change per pcvd pressure p.a.
change in fraction + of e-trade
NYSE Fraction of E-Trade
pcvd pressures
Figure C-28: Inertia and repetitive momentum (SFD) An extension of the inertia mechanism is shown in Figure C-28 b) which focuses on the stock and flow structure of the Repetitive Momentum Loop. The organization’s level of inertia is inversely related to its openness to change. This openness has an influence on how willing the organization is to react to perceived pressure from its environment, expressed by the fractional change per perceived pressure. The effect of openness on change is an s-shaped curve indicating that the organization quickly reacts to perceived pressure if it has a rather high openness. It becomes less responsive as openness decreases until its reactivity reaches a lower bound. Since the fractional change determines the change in the fraction of e-trade, and since change reduces inertia, the reinforcing feedback loop Repetitive Momentum is closed. The effect of change on inertia amplifies the reference fractional inertia decrease that was explained in relation to Figure C-28 a) on page 122. The effect was modeled in such a way that incremental—meaning slow—changes reduce inertia only slightly. Net inertia still develops during times of incremental change.496 As soon as changes become more profound, they quickly reduce inertia, but with an upper limit of 6.5 times the reference fractional decrease of 15 percent so that a very rapid and extensive transformation has the potential to fully destroy inertia. These assumptions lead to an s-shaped effect of change on inertia, shown by Figure C-29. This is consistent with statements by Nadler and Tushman as well as Wollin according to which discontinuous change is disruptive and involves the unlearning of routines, ways of thinking, and assumptions.497 The structure implies that inertia grows when no or very little changes take place, leading to a low openness to change and making changes unlikely. But once the pressure on the NYSE is sufficiently strong to force the imple496
See Kaplan, Sarah and Mary Tripsas: Thinking about technology: Applying a cognitive lens to technical change, in: Research Policy, Vol. 37 (2008), No. 5, pp. 795 and 800. 497 See Nadler and Tushman: Types of Organizational Change, 1995, p. 23; and Wollin: Punctuated equilibrium, 1999, p. 362.
123
C.III Structure and Behavior of Forces for Retention and Change
mentation of some e-trade, inertia diminishes and a reinforcing change mechanism is initiated. Effect of Change on Inertia Decrease 8
effect
6 4 2 0 0
0.100
0.200 0.300 "change in fraction of e-trade"
0.400
0.500
Figure C-29: Relationship between change and inertia How this mechanism fits in with other parts of the model can be seen in the CLD of Figure C-30. Due to its reinforcing character, the Repetitive Momentum Loop creates path-dependent behavior. Repetitive behavior may occur in both directions of stability and change, but does not explain how the NYSE switched from one to the other, and it does thus not explain the sudden NYSE transformation. The path dependent behavior is challenged, first, by customer pressure and its influence on customer orientation, and second, by inadequate performance. These two concepts will be explained on the following pages. ACCESS TO INFORMATION TECHNOLOGY Fraction of Institutional Customers + FRACTION OF EQUITIES HELD BY INSTITUTIONS
+
+ fraction of e-trade in + remaining market
rel. time to execution -
+
dissatisfaction with time per inst. customer +
pressure for (B) Customer Pressure for Speed more e-trade openness to change Inertia
(R) Repetitive Momentum
NYSE Fraction of E-Trade specialist participation
+ (B) Customer Pressure for Market Quality
+ market quality from specialist
(R) Floor Power
pressure for more floor
(B) Resistance -
+ profitability of floor
(R) Pressure from Floor Culture
Figure C-30: Inertia and repetitive momentum (CLD) The orientation of the NYSE’s market model on floor firms and its culture strongly contributed to the observed inertia. The NYSE had always been dedicated to providing the best market for individual investors.498 Its focus on the floor originated from 498
See Blume, Siegel and Rottenberg: Revolution on Wall Street, 1993, p. 108.
124
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
the belief that the floor provided the best trading mechanism. As a result, the organization centered around and benefited floor brokers and specialists, disregarding the sell side’s, i.e. the institutional investment banks’ demands. The NYSE COO who used to work for several large financial institutions expressed this fact. "In my 20 years on the sellside, I never felt like a client of the New York Stock Exchange, … That's despite the fact that at Morgan Stanley, Credit Suisse, and UBS, I represented a pretty good chunk of their order flow.”499 He went on to say that the exchange was run for its owners—often floor broker and specialist firms—but not for investment houses. The NYSE admitted that it had not listened to large customers. It had restricted quick and anonymous trading with intent.500 But this view changed radically, and the focus shifted from floor firms to customers that use the NYSE for trading. Now, NYSE executives “… are trying to transform what was traditionally an inward-looking, exchange floor-dominated culture into an outward-looking, customer-focused culture. Instead of running the organization for the benefit of specialists, for example, Niederauer and Leibowitz say they are trying to satisfy the people that deliver the order flow—[i.e.] the broker-dealers.” “With its market share slipping away, the New York [Stock Exchange] can no longer afford such arrogance. It has had to reconnect with the sources of its order flow.” 501 Yet, an interviewee also reported that the rising customer focus was rather the result of the reinvention of the company than of falling market share. “…because what we were doing was, we were responding to a set of customer needs. There's a set of customers who said, "We want to trade instantaneously, electronically and anonymously. We do not want our orders being routed to the floor." So, we said okay — great. We have to respond to that group of customers, because if we don't they're going to go trade somewhere else.”502 “That push from customers was really how this all started. Making ourselves fit into Regulation NMS is also very important to us. But we moved in this direction because of the reaction from our customer base, which we hadn’t previously been listening to well enough.”503 “If you have a group of big and important customers who want to trade in a certain way and you don’t give them the capability and somebody else does, that’s where they’re going to go. So the first objective of the Hybrid Market is to allow those institutions that want to trade electronically, instantaneously and anonymously to do so.” 504 499
Statement by the NYSE COO and Executive Vice President Lawrence Leibowitz. See Chapman, Mehta and Scotti: Men At Work, 2007, p. 48. 500 See Pellecchia: On crisis and opportunity, 2006, electronic source. 501 Chapman, Mehta and Scotti: Men At Work, 2007, p. 48. 502 Thain: Statement on A CEO's Views on IT and Innovation, 2007, electronic source. 503 Pellecchia: I'd like an Auction Limit order with soy milk, please, 2006, electronic source. 504 ibid., electronic source.
125
C.III Structure and Behavior of Forces for Retention and Change
As a consequence, the NYSE shifted to “serving the needs of its customers,”505 and customer orientation has become central at the New York Stock Exchange. This also manifests itself in a large number of entries in its weblog and in utterances by its management. The importance of customer orientation for the change becomes obvious through a quantitative content analysis of the Exchanges Weblog of which Table C-4 lists the results. It was used and able to provide a quantitative confirmation of qualitative findings. According to Duriau, Reger, and Pfarrer content analysis has received increasing attention particularly in the investigation of constructs related to strategic management and managerial cognition as well as in longitudinal research.506 In the analysis made of the NYSE weblog, searches in March 2008 and December 2009 for the word customer(s) retrieve more entries than searches for floor, specialist(s), speed, fast (market), electronic, automated/automation, value, price improvement, and volatility. The rising awareness of complacency and particularly its impact on the incoming order flow served as a wake-up call. The organization finally noticed that its environment had changed.507 The management team became aware that its strategic orientation was in conflict with what its environment demanded, and it readjusted its attention towards those groups which were most dissatisfied: its large institutional customers who wanted to trade electronically. As a consequence of the growing awareness that the NYSE was not responding to their stakeholders’ demands, it made a strong move away from its inward floor orientation towards an adaptation to its customers. Key Words Entered on 11 Dec. 2009
Number of Hits
Customer(s) / client(s)
447
Floor / manual
411
Specialist(s) / DMM(s) / designated market maker(s)
379
Electronic / automated / automation
354
Speed / fast (market)
243
Value
222
Price improvement
151
Volatility
140
Market share / performance
125
Table C-4: Quantitative content analysis of the Exchanges Weblog 505
Statement by Louis G. Pastina, Executive Vice President, NYSE Operations, quoted in: NYSE Euronext Inc.: NYSE Euronext Appoints Todd B. Abrahall and Michael J. Rutigliano as Liaisons to NYSE Specialists and Brokers, 2008, electronic source. 506 See Duriau, Vincent J., Rhonda K. Reger and Michael D. Pfarrer: A Content Analysis of the Content Analysis Literature in Organization Studies: Research Themes, Data Sources, and Methodological Refinements, in: Organizational Research Methods, Vol. 10 (2007), No. 1, pp. 14–16 and 23. 507 See Pellecchia: Shaping the blog to a new mark, 2007, electronic source.
126
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
The NYSE management team faces pressure from institutional as well as noninstitutional customers and floor firms. The majority of institutional customers expresses a desire or exerts total pressure for more e-trade. Many non-institutional ones apply total pressure for more floor trade from customers. Additionally, the floor firms impose total pressure for more floor trade from the floor which is a product of resistance, culture, and power of floor firms. These three types of pressure are shown in the bottom right corner of Figure C-31, and they were already discussed in the previous chapters. Customer orientation comes in when it concerns the perception of pressure. The management team’s perception of pressure is a result of customer and floor pressure and particularly its orientation to its stakeholders, i.e. its customer orientation as opposed to the orientation to the floor. As the figure and the equations C-6 and C-7 elucidate, customer orientation serves as a weighting factor for the perception of pressure that stakeholders exert on the NYSE. It tells how much weight the NYSE management attributes to the respective stakeholder pressure. In this way, a shift in customer orientation is able to explain the increased perceived pressure to implement e-trade and the resulting shift of the organization. REF. OPENNESS PER INERTIA
institutionalization + (R) Inertia (B) + inertia decrease + effect of change on inertia NYSE Fraction of Floor Trade effect of floor trade on change
+ openness to change
+
effect of openness on change
(R) Repetitive Momentum REF. FRACT. CHANGE IN TRADING PER PRESSURE P.A. + +
Customer Orientation
fractional change per pcvd pressure p.a.
change in fraction + of e-trade + + -
NYSE Fraction of E-Trade effect of e-trade on change
+ pcvd pressure for + more e-trade pcvd pressure for + more floor trade +
Figure C-31: Relationship between customer orientation and change (SFD)
pcvd pressure for more eTrade [pressure unit] = total pressure for more eTrade from customers [pressure unit] • Customer Orientation [dmnl]
C-6
C.III Structure and Behavior of Forces for Retention and Change
pcvd pressure for more floor trade [pressure unit] = total pressure for more floor trade from customers [pressure unit] • Customer Orientation [dmnl] + total pressure for more floor trade from floor [pressure unit] • (1 – Customer Orientation [dmnl])
127
C-7
An imbalance in the perceived pressure triggers change. When the perceived pressure for e-trade is higher than the perceived pressure for floor trade, the NYSE moves from floor to e-trade and vice versa. How the NYSE translates perceived pressure into change action confirms with earlier representations. The limiting effects of e-trade and floor trade on change still represent the idea that in the theoretical case when the customers still exert a large amount of pressure for e-trade (floor trade) although the fraction of e-trade (floor trade) is already high, the management team is less willing to fully react to the pressure. Depending on the fractional change per perceived pressure—i.e. the NYSE’s general openness to change—the organization reacts more or less quickly to the pressure it perceives in its environment. Overall this means, managerial attention provides the lens through which the NYSE interprets the pressure which affects the organization. A high or low value of customer orientation always expresses a bias towards one of the stakeholder groups. Figure C-32 indicates how the stock customer orientation itself is influenced and adapts. The management team can distribute its attention between floor firms and customers. First, the rate of change in the customer orientation shifts attention the quicker the larger the yearly fractional change in customer orientation becomes. Since the latter directly depends on the effect of openness on change that also affects changes in e-trade, the feedback loop Repetitive Attention is closed. This reinforcing feedback loop allows the organization to adapt to a level of customer importance at a specific point in time and to freeze here as inertia grows, biasing the perception of outside pressure. This relationship captures how a reinforcing character of Repetitive Attention perpetuates and how customer orientation changes when the organization becomes more open.
128
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
REF. OPENNESS PER INERTIA
institutionalization + (R)
+ openness to change
+
effect of openness on change
-
REF. FRACT. CHANGE IN CUST. ORIENT. P.A. + fract. change in + cust. orient. per pressure p.a.
Inertia (B) + inertia decrease + effect of change on inertia NYSE Fraction of Floor Trade - (R) effect of floor trade on change
Floor Orientation
change in cust. orient.
Customer Orientation
change in fraction + of e-trade + + -
NYSE Fraction of E-Trade (B) effect of e-trade on change
(R) Repetitive Attention + pcvd pressure for + more e-trade pcvd pressure for + more floor trade +
Figure C-32: Repetitive attention loop (SFD) Second, as illustrated in Figure C-33, customer orientation adapts to the pressure which comes from stakeholder groups in a comparable way as the fraction of e-trade adjusts to pressure. This means customer orientation rises when the perceived pressure from customers is higher than the perceived pressure from the floor. Similar to the perception of pressure for change, the perception of pressure from stakeholders is biased by customer (or floor) orientation itself, leading to the following computation of change in the customer orientation: change in customer orientation [dmnl / year] = (pcvd pressure for more cust. orient. [pressure unit] • effect of cust. orient. on change [dmnl] – pcvd pressure for more floor orient. [pressure unit] • effect of floor orient. on change [dmnl]) • fract. change in cust. orient. per pressure p.a. [dmnl / pressure unit / year]
C-8
Importantly, the perceived pressure from customers and from the floor is different from that for e-trade and floor trade since one set of variables aggregates stakeholders and the second one aggregates their desires. Additionally, the balancing effect of customer (floor) orientation on change limits a further orientation to customers (the floor) when customer (floor) orientation is already high, but customers (the floor) continue to pressure for their aims. The management team would no longer be willing to fully react to this pressure.
129
C.III Structure and Behavior of Forces for Retention and Change
Floor Orientation
Customer Orientation change in cust. orient. + + (B) (B) - + effect of floor effect of cust. orient. on change orient. on change +
(R)
(R)
pcvd pressure from the floor
+
pcvd pressure from customers
+
+
Figure C-33: Customer orientation (SFD) As the relationship between pressure and customer orientation in Figure C-33 and Figure C-34 reveals, customer orientation adapts if there is a pressure imbalance. It does so even faster if the organization is open to change. The reinforcing feedback loop that shapes the Repetitive Attention is shown in a simplified causal form in Figure C-34. It shapes and biases the pressure for e-trade and for the floor in a similar way the Repetitive Momentum Loop does. ACCESS TO INFORMATION TECHNOLOGY
+
+ fraction of e-trade in + remaining market +
Fraction of Institutional Customers + FRACTION OF EQUITIES HELD BY INSTITUTIONS
Customer Orientation openness to change Inertia
(B) rel. time to execution -
+
dissatisfaction with time per inst. customer +
pressure for (B) Customer Pressure for Speed more e-trade
(R) Repetitive Attention
(R) Repetitive Momentum
NYSE Fraction of E-Trade specialist participation
+ (B) Customer Pressure for Market Quality
+ market quality from specialist
(R) Floor Power
pressure for more floor
(B) Resistance -
+ profitability of floor
(R) Pressure from Floor Culture
Figure C-34: Perception bias from managerial attention (CLD) In the case of the New York Stock Exchange, not only the acknowledgement of customer demands, but also performance represents a driver of change.508 Performance of the NYSE can be operationalized by several concepts—market share, market quality, liquidity, and speed being the most important ones. Two of these are of particular importance: Much of the discussion centers around the market quality that the 508
See Lucchetti: NYSE Plans to Revise Specialist-Trader Rules, p. C.4.
130
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
participation of specialists provides. Additionally, market share is central because all concepts listed above have an influence on the NYSE’s market share of volume in NYSE-listed securities. Therefore, the determinants of market share and market quality as well as how these performance measures translate into managerial decisionmaking will be described in more detail. Two reinforcing liquidity mechanisms of the NYSE and of the remaining U.S. market are the main determinants of market share. Figure C-35 visualizes these loops. The mechanism is commonly referred to by the words “liquidity begets liquidity”.509 The same reinforcing mechanism is also reported for the London Stock Exchange.510 The feedback loop NYSE Spread from Liquidity on the right side sketches the mechanism for the NYSE, and the loop Spread in the Remaining Market mirrors it for the remaining trading venues. The further division of market share among other exchanges is not of interest so that the remaining market can be represented in an aggregate manner. The description will focus on the NYSE Spread from Liquidity loop. The NYSE market share is the central stock at the top of Figure C-35. Multiplied by the total U.S. share volume in NYSE-listed issues, it results in the NYSE trading volume expressed in shares per year. The relative trading volume in comparison to the total shares traded (which is another expression for market share) has an effect on the NYSE spread, i.e. the quoted price. Liquidity is often regarded as the most important factor characterizing the attractiveness of an exchange, which—since difficult to measure—is often described with relation to daily turnover and the size of the bid ask spread.511 This means, liquidity from market share will attract volume to the exchange—here orders with a price limit—and will increase market share again. The more limit orders there are and the more quoted depth (i.e. orders) there is at each price point, the smaller the spread. An analysis of market quality at the NYSE and NASDAQ also illustrates that a high consolidation of order flow to one exchange reduces an exchange’s spread.512 The effect of the relative trading volume on the NYSE spread is drawn as a linear function because, first, there is no further information on the shape of this relationship, and second, the idea that liquidity begets liquidity develops by the reinforcing character of the entire loop, not by this relationship between two variables. When market share (or trading volume) increase by one percent, the spread falls by 0.2 percent. The NYSE spread is additionally influenced by the reference spread and the effect of institutional customers on the spread. The latter decreased the spread and is supposed to symbolize events such as the transi509
Duncan Niederauer, Deputy CEO, NYSE Euronext, in: Niederauer, Duncan: Statement 17 in Questions and Answers, in: Thomson Financial: NYX - Q2 2007 NYSE Euronext Earnings Conference Call (2 Aug. 2007) [Conference Call Transcript], 2007a, electronic source; Tabb, Larry: Liquidity Begets Liquidity, in: Advanced Trading (2008a), issued Feb. 2008, p. 47; and Weber, Bruce W.: Adoption of electronic trading at the International Securities Exchange, in: Decision Support Systems, Vol. 41 (2006), No. 4, p. 741. Additionally, in an interview a specialist referred to the mechanism as “volume begets volume”. 510 See Clemons and Weber: London's big bang, 1990, p. 51. 511 See ibid., pp. 54–55. 512 See Bennett and Wei: Market structure, fragmentation, and market quality, 2006, p. 69.
C.III Structure and Behavior of Forces for Retention and Change
131
tion to quoting in sixteenth and to pennies (commonly referred to as decimalization) that is supposed to have favored institutional customers. Data shows that the quoted spread used to average at USD 0.25, declined to USD 0.13 in 1991, and fell further to USD 0.06 for a sample of NYSE stocks from January 2002 to March 2003. During the latter period a NASDAQ sample averaged at USD 0.09.513 The system dynamics model explicitly refers to these sources of the quoted average spread, not to data on the median spread or on a spread measure that is not further defined.514 The model exhibits a spread of USD 0.23 in the beginning which falls to USD 0.13 in 1991 and to USD 0.07 in early 2003. It is thus able to closely replicate behavioral patterns observed in the real world. The simulated spread for the remaining market diminishes from USD 0.25 to USD 0.08 as compared to USD 0.09 at NASDAQ. Hendershott and Moulton observe a 10 percent increase in the NYSE spread directly after the introduction of the Hybrid Market.515 Others report a falling quoted spread or ambiguous results.516 In the simulation runs, the spread continuously decreases, too, but the relative spread of the NYSE increases by about 10 percent after the introduction of the Hybrid Market. Further influences on the spread such as inventory and informational issues are discussed particularly in the market microstructure literature.517 They are left out since they often concern short-term or external effects.
513
For data during the 1960s see Demsetz, Harold: The Cost of Transacting, in: The Quarterly Journal of Economics, Vol. 82 (1968), No. 1, p. 40; for data on the year 1991 see Huang and Stoll: Dealer versus auction markets, 1995, p. 323; and for data on the period from 2002 to 2003 see Bennett and Wei: Market structure, fragmentation, and market quality, 2006, p. 62. 514 Data on some measure of spread and on the median spread can be found in NYSE Euronext Inc.: Average NYSE Spreads, no date, electronic source; and United States Securities and Exchange Commission (SEC), Market 2000, 1994, exhibit 30. 515 See Hendershott and Moulton: Speed and Stock Market Quality, 2009, p. 26. 516 See Storkenmaier and Riordan: The Effect of Automated Trading on Market Quality, 2009, pp. 21 and 28. Gutierrez and Tse find a falling quoted spread for several order types and sizes, but they observe contrary evidence for large market orders and insignificant results for other order types and sizes. See Gutierrez, Jose A. and Yiuman Tse: NYSE execution quality subsequent to migration to hybrid, in: Review of Quantitative Finance and Accounting, Vol. 33 (2009), No. 1, p. 75. 517 Market microstructure, trading costs, and the spread are analyzed in Biais, Bruno, Larry Glosten and Chester Spatt: Market microstructure: A survey of microfoundations, empirical results, and policy implications, in: Journal of Financial Markets, Vol. 8 (2005), No. 2, pp. 221–235; Conroy and Winkler: Market Structure, 1986, pp. 22–30; and Madhavan: Market microstructure: A survey, 2000, pp. 208–211. For an analysis of transaction costs see also Demsetz: The Cost of Transacting, 1968, pp. 46–47.
132
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE NYSE Market Share
TOTAL U.S. SHARE VOLUME IN NYSELISTED ISSUES
<Time>
+
TIME FOR CHANGING MARKET SHARE
+ trading volume of the remaining market
+ + NYSE trading volume
indicated NYSE market share +
effect of relative trading volume on market's spread
effect of relative trading volume on NYSE spread
NYSE market share from NBBO (cons.) + (R) Market Liquidity and Spread
fraction of time at NBBO -
(R) NYSE Liquidity and Spread
+
+
spread in market -
+
<effect of inst. customers on spread>
relative spread + of NYSE
NYSE spread - +
REF. SPREAD
Figure C-35: Spread and market share (SFD) The relative price expressed by the relative spread of the NYSE in comparison with the market is important. The better the relative spread, the higher the fraction of time at the NBBO, i.e. the fraction of time the NYSE displays the national best bid and offer. The specialist quote or a public limit order may set the best quote at the NYSE.518 “[The] NBBO is a key driver of market share.”519 How the time at the NBBO translates into market share is depicted in Figure C-36. In the consolidated market, before the implementation of Regulation NMS, the SEC’s trade-through rule required an order to be sent to the exchange displaying the best price.520 The line is situated slightly above the 45°line in order to capture that the NYSE is one market where liquidity consolidates whereas the remaining market consists of several trading venues where liquidity is scattered. The line rises below proportionally in the very beginning so as to account for the fact that an exchange that displays bad prices most of the time has difficulties to attract a sufficient depth of liquidity at the best price. A critical mass of liquidity is necessary for an exchange to attract volume.521 The market share from the NBBO of a single exchange reaches a maximum of 80 percent as not all trades are executed at exchanges and only reported later. Some are executed internally at large institutions and off the exchanges after the regular trading hours.522 Just from quoting at the NBBO a single exchange can thus not reach a 100 percent mar518
See Conroy and Winkler: Market Structure, 1986, pp. 22 and 34. A limit order is an order with a maximum (or minimum) price. 519 Statement by Duncan Niederauer, as Deputy CEO of the NYSE, quoted in: Chapman, Mehta and Scotti: Men At Work, 2007, p. 54. 520 A detailed description of the mechanism can be found in Hasbrouck, Joel, George Sofianos and Deborah Sosebee: New York Stock Exchange Systems and Trading Procedures, NYSE Working Paper, 1993, pp. 25–29. See also Securities and Exchange Commission (SEC): Adoption of Amendments to the Intermarket Trading System Plan, 1999, p. 70298. 521 See Picot, Bortenlänger and Röhrl: Organization of Electronic Markets, 1997, p. 115. 522 See United States Securities and Exchange Commission (SEC), Market 2000, 1994, p. II-7.
133
C.III Structure and Behavior of Forces for Retention and Change
ket share. Since the focus is on the consolidated market before the implementation of Regulation NMS and since further influences are left out at the moment, the NYSE market share from the NBBO directly translates into an indicated NYSE market share. The actual NYSE market share is a stock that adapts to the indicated variable with a time delay of one year. The feedback loop NYSE liquidity and spread is now closed. While benefiting exchanges with high market share, the reinforcing character of this feedback mechanism also aggravates situations of low liquidity.
market share from NBBO
NYSE Market Share from NBBO 0.8 0.6
maximum due to internal and off exchange execution
critical mass problem
0.4 0.2 0 0
0.10
0.20
0.30
0.40 0.50 0.60 0.70 fraction of time at NBBO
0.80
0.90
1
Figure C-36: Relationship between time at NBBO and market share While the time at the NBBO is a key driver of market share, the “other half is based on the specialist’s trading performance.”523 This may be an exaggeration, but the extra market quality the specialists and floor brokers provide as well as the trading speed are responsible for a market share adjustment. Execution quality has an influence on order routing decisions. Markets with low execution costs and fast executions receive more orders.524 The NYSE assumes that improving the quality of quotes will attract more orders, i.e. market quality from the specialist will improve market share.525 “My general view is if I get market quality right, market share will follow.”526 Therefore, as depicted in the bottom left of Figure C-37, market quality from specialist participation has an influence on the indicated NYSE market share via the market share adjustment. Since market quality ranges between the values 1 when there is no specialist participation and 1.1 when the specialist participates in 10 percent of the trades, it serves as a multiplier for the NYSE market share from the NBBO. High market quality from high floor involvement may thus adjust market share upwards by 10 percent. 523
Statement by Duncan Niederauer, Deputy CEO of the NYSE, quoted in: Chapman, Mehta and Scotti: Men At Work, 2007, p. 54. 524 See Boehmer, Jennings and Wei: Public Disclosure and Private Decisions, 2007, pp. 318 and 353. 525 See Chapman, Mehta and Scotti: Men At Work, 2007, p. 52. 526 Statement by Duncan Niederauer, President and Co-COO and Head of U.S. Cash Markets. Niederauer: Statement in speech on U.S. Cash Equities, 2007, electronic source. See also McAndrews and Stefanadis: The Emergence of Electronic Communication Networks in the U.S. Equity Markets, 2000, p. 3.
134
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE <wt. on time vs. spread among all customers> <effect of time to execution on market share>
<market quality from sp. participation>
indicated NYSE market share + + - market share - adjustment +
NYSE market share from NBBO (cons.) + fraction of time at NBBO
Figure C-37: Market share adjustment from speed and market quality (SFD) Additionally, customers do not only value market quality, but speed has gained significant importance. Customers decide to trade at electronic exchanges in order to have their orders executed more quickly.527 A slow time to execution has a negative effect on market share.528 Figure C-37 expresses this by the effect of the time to execution on market share. The nonlinear relationship is also drawn in Figure C-38. In the area around the point (1,1) where the speed at the NYSE and in the market are similar, customers are rather indifferent and market share is determined by the frequency an exchange sets the NBBO. Yet, the further the times to execution diverge, the more important speed becomes, and it turns into an important order winning criterion.529 Effect of Time to Execution on Market Share 1.1
effect
0.9
0.7
0.5 0
1
2
3 4 5 6 relative time to execution
7
8
9
Figure C-38: Relationship between time to execution and market share While competition originally focused on price, with the growth of institutional customers and technology, it increasingly shifted towards speed. Based on an analysis of customer order choices, Ellul et al. find that orders which are sent to an automated 527
See Ortega: NYSE's Thain Fails to Stem Losses in Market Share (Update 2), 2006, electronic source. 528 See Hendershott and Moulton: Speed and Stock Market Quality, 2009, p. 1. They also refer to several further authors, such as Boehmer, Jennings and Wei: Public Disclosure and Private Decisions, 2007, pp. 252–253. 529 For information on the shape of the relationship between a performance criterion and the customer reaction see Salge: Struktur und Dynamik ganzheitlicher Verbesserungsprogramme in der industriellen Fertigung, 2009, pp. 55–56; and Slack, Nigel, Stuart Chambers and Robert Johnston: Operations Management, 5. Ed., Harlow [et. al] 2007, pp. 69–70.
135
C.III Structure and Behavior of Forces for Retention and Change
system are less price sensitive than those sent to the floor.530 This suggests that customers attach a certain weighting to the time to execution as well as to the spread or price. As equation C-9 indicates, the weight on time vs. spread among all customers thus weights among the importance of the effect of time to execution on market share and the impact of the market quality from specialist participation. The resulting market share adjustment works as a multiplier of the NYSE market share from the NBBO to derive the indicated NYSE market share. market share adjustment [dmnl] = wt. on time vs. spread among all customers [dmnl] • effect of time to execution on market share [dmnl] + (1 – wt. on time vs. spread among all customers [dmnl]) • market quality from sp. participation [dmnl]
+
pcvd adequacy of market share -
+
confidence effect of market share
REF. OPENNESS PER INERTIA + openness to change -
+ desired market share TIME TO ADJUST DESD MARKET SHARE
C-9
Figure C-39: Relationship between market share and openness to change (SFD) The CEO explained that the NYSE had lost market share due to a general increase of electronic trading with which the NYSE could not keep up.531 JP Morgan Securities estimates that with every lost percentage point of market share the NYSE looses about USD 2.3 million in net income.532 Respectively, the organization’s projected earnings per share are expected to diminish by 2 cents per lost percentage point of market share.533 In the view of Storkenmaier and Riordan, the NYSE’s reluctance to change led to decreases in its market share, and the NYSE management reacted by the introduction of the Hybrid Market. Further authors support this view.534 “The decline [in the NYSE's share of trading NYSE-listed issues] underscores the urgency of adopting the Hybrid Market, a project that will cut the time to complete a
530
See Ellul, et al.: Order dynamics, 2007, p. 659. See CEO, NYSE Euronext in: Thain, John: Statement in an Interview with CSBN (6 Oct. 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2006, electronic source. 532 Cited in: Ortega: NYSE's Thain Fails to Stem Losses in Market Share (Update 2), 2006, electronic source. 533 See Lucchetti, Aaron and Alistair MacDonald: Stock-Exchanges Grudge Match: Old Rivals NYSE and Nasdaq Use Bids for European Markets As Latest Way to Pin Their Foe, in: Wall Street Journal - Eastern Edition (2006), issued 8 Sep. 2006, p. C.1. 534 See Storkenmaier and Riordan: The Effect of Automated Trading on Market Quality, 2009, p. 11. See also Ortega: NYSE's Thain Fails to Stem Losses in Market Share (Update 2), 2006, electronic source. 531
136
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
trade to less than a second, just like all-electronic rivals.”535 They thus see a causal relationship between an inadequate market share and subsequent change, which is understandable with respect to the negative consequences of market share. The causal link from market share to the NYSE decision making can be seen in Figure C-39. When the management team perceives the NYSE market share to be inadequate as compared to the floating goal of desired market share, it loses its confidence that it is pursuing the right strategy. As Storkenmaier and Riordan reported, this confidence effect of market share increases the organization’s openness to change.536 Figure C-40 indicates how this happens. As soon as the adequacy level falls below the value of one, confidence decreases. This happens slowly for minor inadequacies close to the value one since there always may be some variation in market share, but the effect quickly aggravates. Since the product of inertia and the confidence effect are subtracted from the value one to derive the openness to change (equation C-10), a low confidence effect from low market share increases the openness to change. Confidence effect of market share on opennes to change 1
effect
0.75 0.5 0.25 0 0
0.20
0.40 0.60 0.80 pcvd adequacy of market share
1
1.20
Figure C-40: Confidence effect of market share on openness to change openness to change [dmnl] = 1 [dmnl] – Inertia [consistency unit] • confidence effect of market share [dmnl] • REF. OPENNESS PER INERTIA [dmnl / consistency unit]
C-10
Figure C-41 illustrates that the relationships market share Æ NYSE decision on e-trade Æ market share create two endogenous feedback mechanisms. Low market share creates more openness for electronic trading, and e-trade improves the relative time to execution and raises market share again. This closes the balancing feedback loop Market Share from Speed. A higher fraction of electronic trading also lowers market quality and thus market share, creating the reinforcing mechanism Market 535
CEO John Thain in: Ortega: NYSE's Thain Fails to Stem Losses in Market Share (Update 2), 2006, electronic source. 536 See Storkenmaier and Riordan: The Effect of Automated Trading on Market Quality, 2009, p. 11.
137
C.III Structure and Behavior of Forces for Retention and Change
Share from Market Quality. This indicates that a move towards electronic trading calls for even more electronic trading. Therefore, the implementation of e-trade has ambiguous effects on market share, but due to the rising importance of institutional customers and their weighting of speed, the effect of speed finally prevails. ACCESS TO INFORMATION TECHNOLOGY Fraction of Institutional Customers + FRACTION OF EQUITIES HELD BY INSTITUTIONS
+ fraction of e-trade in + remaining market
-
spread
Customer Orientation
(R) Liquidity Market openness to Share + change +
Inertia
+ (B) Market Share from Speed
rel. time to execution
+
-
dissatisfaction with time per inst. customer +
pressure for (B) Customer Pressure for Speed more e-trade
(R) Repetitive Attention
NYSE Fraction + of E-Trade (B) Customer specialist Pressure for participation Market Quality + (R) Floor market quality (R) Market Share Power from Market Quality from specialist (B) Resistance (R) Repetitive Momentum
-
+ profitability of floor
pressure for more floor
(R) Pressure from Floor Culture
Figure C-41: Market share (CLD) In order to reduce the reinforcing effects created by the Market Share from Market Quality Loop, the NYSE introduced liquidity algorithms which provide specialists and floor brokers with the ability to participate in electronic trading and improve market quality at the same time. When specialist participation and market quality dropped to a level much below that of the pre-Hybrid stage, the NYSE management made a significant revision of its new market model. It slightly amended the role of the specialists and changed their name to designated market makers (DMMs). While the management team did not depart from electronic trading, it implemented liquidity algorithms in order to make sure that floor brokers and DMMs participate in the electronic environment. It also introduced Special Liquidity Providers: six firms which complement and compete with DMMs and provide bids and offers for assigned securities.537 With the help of liquidity algorithms, DMMs and Special Liquidity Providers are able to constantly provide automated bids and offers also in electronic trading. The NYSE management made these changes because the electronic environment did not provide sufficient incentives for specialists to participate in trading. The transformation of specialists to DMMs and the introduction of liquidity algorithms are based on the belief that the DMMs’ provision of liquidity and quality improves the NYSE’s market share.538 537 538
See NYSE Euronext Inc.: Significant Progress in New NYSE Market Model, 2009, p. 2. See Niederauer: Statement in speech on U.S. Cash Equities, 2007, electronic source; Pellecchia: Comment on: Ten Things I Like About the Coming Changes at NYSE, 2008, electronic source.; Pellecchia: Comment on: We, Robots, 2008, electronic source; and Ross, Jim (Vice President of NYSE MatchPoint and ATS (alternative trading system) Strategy): Together, MatchPoint and Algorithms Take Performance to a New Level (29 Apr. 2009), in: NYSE Eu-
138
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE TIME TO DEVELOP ALGORITHMS
REF. LIQUIDITY ALGORITHMS effect of liquidity algorithms on + participation
<effect of floor trade on sp. participation>
Liquidity Algorithms
+
+ specialist participation + REF. SP PARTICIPATION
+
development of + liquidity algorithms +
market quality adequacy gap + -
(B) Liquidity Algorithms
effect of sp. participation on market quality + +
market quality from sp. participation
ALGORITHMS PER GAP PER YEAR
+
pcvd adequacy of market quality by customer
DESD ADEQUACY OF MARKET QUALITY
REF. MARKET QUALITY
Figure C-42: Liquidity algorithms as a response to low market quality (SFD) The causal mechanism between floor trade, specialist participation, and the perceived adequacy of market quality has been described and is depicted again from the top left to the bottom right of Figure C-42. Algorithms are introduced to increase participation and market quality when there is a gap between the perceived adequacy of market quality and its desired value one. Algorithms take about one and a half years to be initiated since the gap needs to be perceived as being problematic; this is why there is a third order smooth in the development decision instead of a development delay. Then, algorithms accumulate and positively affect specialist participation as expressed by equations C-11 and C-12. Together, this creates the balancing loop Liquidity Algorithms. effect of liquidity algorithms on participation [dmnl] = Liquidity Algorithms [algorithms] + REF. LIQUIDITY ALGORITHMS [algorithms] / REF. LIQUIDITY ALGORITHMS [algorithms]
C-11
specialist participation [dmnl] = effect of floor trade on sp. participation [dmnl] • effect of liquidity algorithms on participation [dmnl] • REF. SP PARTICIPATION [dmnl]
C-12
Floor algorithms serve the goal of offering greater choices to customers and serving their needs. In the view of the managing director of the company providing the algorithms, markets characterized by fragmentation and millisecond timescales demand algorithms which additionally bring human judgment.539 Even with the revision ronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2009, electronic source. See also Chapman, Mehta and Scotti: Men At Work, 2007, p. 56; and Lucchetti: NYSE Plans to Revise Specialist-Trader Rules, p. C.4. 539 See NYSE Euronext Inc.: NYSE Euronext Appoints Todd B. Abrahall and Michael J. Rutigliano as Liaisons to NYSE Specialists and Brokers, 2008, electronic source.
139
C.III Structure and Behavior of Forces for Retention and Change
of the Hybrid Market model, in the pursuit of customer orientation the NYSE remained committed to the advantages of the floor and combined this with high-speed automated trading.
C.III.4 Full Model Behavior The system dynamics model now includes exogenous pressure, stakeholder reactions, and the NYSE management’s response to these kinds of pressure, as it has been shaped by the past. Endogenous processes of the management’s decisionmaking create the rather radical shift from manual to electronic trading, as can be seen in line number 3 in the BOT graph of Figure C-43. The consideration of how inertia and market share impact the openness to change as well as the consideration of attention to customers explain the long period during which electronic trading is not implemented. They also account for the observed radical shift.
NYSE Fraction of E-Trade 1
1 1
Dmnl
0.75
123 1231231
2 1
0.5 0.25
23
2 1
1 2 0 3 123123123 1970 1980 1990
2 3
3 2000 Date
3 2010
2020
2030
"NYSE Fraction of E-Trade" : adaptation 1 1 1 1 1 "NYSE Fraction of E-Trade" : pressures 2 2 2 2 2 2 3 3 3 3 3 "NYSE Fraction of E-Trade" : management 3
Figure C-43: Adaptation, Culture and Resistance, and Mgmt (BOT) Figure C-44 a) reveals the different forces leading to this shift. When the pressure for more e-trade (line 1 a) rises, a small portion of e-trade is implemented. At the same time when e-trade is implemented, the floor exerts pressure against it, trying to avert the change (line 2 a). Its pressure is less strong than that by institutional customers because the floor has lost much of its power and cannot truly slow down change. As Figure C-44 b) shows, a decline in market share (line 2 b) from the rising dissatisfaction with the trading mechanism also helps the implementation of electronic trading by increasing the openness to change. But the radical behavior comes about due to the decline of inertia from change (line 1 b). This drop is accommodated by a rise in the openness to change, allowing for more e-trade. The base run of the model now to a great extent matches the reference mode. It fits the idea that after the
140
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
transformation the NYSE is much more aware of competition it faces from ECNs, alternative trading systems, and stock exchanges.540 a)
Pressure and E-Trade 1
50 pressure unit 1 Dmnl
3
3
3
1
25 pressure unit 0.5 Dmnl
2 0 pressure unit 1 3 0 Dmnl 12 12 12 312 31 231 2 3 23 2 1970 1980 1990 2000 2010 2020 Date
12 2030
"total pressure for more e-trade from customers" : management 1 pressure unit total pressure for more floor trade : management pressure unit 2 2 3 3 3 3 "NYSE Fraction of E-Trade" : management 3 Dmnl
b)
Market Share and Inertia 1 consistency unit 1 2 1 Dmnl
12
12
1
2
1
2
2
2
2
0.5 consistency unit 0.5 Dmnl 0 consistency unit 0 Dmnl 3 3 1970 1980
3
3
1
1
3 12
1
1
3 3 3 1990
3 2000 Date
2010
2020
2030
Inertia : management consistency unit 1 1 1 1 1 1 2 2 2 2 2 NYSE Market Share : management Dmnl 3 3 3 3 "NYSE Fraction of E-Trade" : management 3 Dmnl
Figure C-44: Underlying forces (BOT) As the reader may have noticed, the reviewed forces are able to explain the general behavior pattern of the radical shift from floor to e-trade. Environmental as well as endogenous drivers of change collectively create the behavioral pattern that can be observed in the real world. The inclusion of two external effects will additionally help to be more exact concerning the timing of the radical transformation. These external effects are a scandal involving the former CEO Richard Grasso and the implementation of Regulation NMS with the resulting fragmentation of the formerly consolidated market. The former NYSE CEO tried to cash in his USD 140 million retirement 540
See NYSE Euronext Inc., Annual Report pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 for the fiscal year ended December 31, 2008, No. 1-33392, New York 2009a, p. 15.
141
C.III Structure and Behavior of Forces for Retention and Change
package. Due to the high amount, this was considered inappropriate and outrageous for a CEO of a non-profit company. As a result of the scandal, the CEO Grasso was asked to resign.541 His resignation and replacement diminished organizational inertia suddenly and exogenously, which is added to the model as a diminishing effect on organizational inertia in the year 2004. About two years after the scandal, the Securities and Exchange Commission implemented Regulation NMS affecting the reinforcing “liquidity begets liquidity” loop of market share. Regulation NMS reduces market share for two reasons. First, it requires markets to trade fast if they want to participate in the national market system in which orders are directly sent to the trading venue with the best price. Second, even if an exchange quotes the best price, after the regulatory change it may often only execute the first portion of the order if it does not have sufficient liquidity at the best price. As a consequence the remainder gets fragmented, which is represented by the comparatively lower thick line in Figure C-45. Technically, starting in mid-2005, a different function is used for translating the fraction of time at the NBBO into the NYSE market share from the NBBO. Studies provide evidence that market fragmentation results in reduced price efficiency, liquidity, and higher volatility.542 This weakens the reinforcing liquidity loop and diminishes the NYSE’s market share. The scandal and the regulatory change are thus responsible for the radical shift towards electronic trading to happen about two years earlier.
market share from NBBO
NYSE Market Share from NBBO 0.8
consolidated market
0.6 0.4
fragmented market
0.2 0 0
0.10
0.20
0.30
0.40 0.50 0.60 0.70 fraction of time at NBBO
0.80
0.90
1
Figure C-45: Relationship between the time at the NBBO and market share As can be seen in Figure C-46, the base run of the model now to a great extent matches the reference mode, and the radical automation of NYSE trading starts around late 2006. It also becomes obvious that, due to the regulatory change, the market share will not recover to the former high standard any more. This base run
541 542
See Gasparino: King of the Club, 2007, pp. 275í281. See Bennett and Wei: Market structure, fragmentation, and market quality, 2006, pp. 69 and 71; and Madhavan, Ananth: Consolidation, Fragmentation, and the Disclosure of Trading Information, in: The Review of Financial Studies, Vol. 8 (1995), No. 3, p. 594.
142
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
reproduces the real events and behavior as closely as possible and marks the reference point for further model analysis.
Market Share and E-Trade 3
1
2
4
4
2
4
2 4
2
Dmnl
0.75
3
3
3 4
0.5 2 4
0.25 0 1 3 1970
1
3 1 1980
3 1 1990
3
3 1 2000 Date
4
4
1 2010
2020
2030
"fraction of e-trade from data (volume)" : base run 1 1 1 market share from data : base run 2 2 2 2 2 "NYSE Fraction of E-Trade" : base run 3 3 3 3 3 4 4 4 4 4 NYSE Market Share : base run
Figure C-46: Comparison with reference mode (BOT) In the beginning, while there is no pressure from customers, high institutionalization and inertia dominate the system dynamics model’s behavior. A significant level of market share as well as the reinforcing Liquidity Loop keep the management team inert and support the concentration on floor firms (see Figure C-47). Initially, the balancing Customer Pressure for Speed Loop is not able to affect greater changes in the way of trading. But once the organization makes small changes to its trading mechanism, the Repetitive Momentum Loop and the Repetitive Attention Loop, which explain path-dependent behavior, become less rigid. The Liquidity Loop around market share also looses much of its strength due to Regulation NMS and allows market share to decline. When the feedback from the relative time to execution to market share additionally gets stronger, i.e. the Market Share from Speed Loop, these mechanisms trigger a rethinking in the organization’s strategic orientation. Once some change is initiated, decreasing inertia also triggers more and more change so that the pendulum of the reinforcing Repetitive Momentum Loop swings to the opposite direction towards more electronic trading. Now the Repetitive Attention Loop also helps bias the perception of resistance from the floor. The adaptation process then slows down the more the adaptive Customer Pressure for Speed Loop approaches its goal.
143
C.III Structure and Behavior of Forces for Retention and Change
ACCESS TO INFORMATION TECHNOLOGY Fraction of Institutional Customers + FRACTION OF EQUITIES HELD BY INSTITUTIONS
+ fraction of e-trade in + remaining market
-
spread
Customer Orientation
(R) Liquidity openness to Market change Share + + -
Inertia
+ (B) Market Share from Speed
rel. time to execution
+
-
dissatisfaction with time per inst. customer +
pressure for (B) Customer Pressure for Speed more e-trade
(R) Repetitive Attention
NYSE Fraction + of E-Trade (B) Customer specialist Pressure for participation Market Quality + (R) Floor market quality (R) Market Share Power from Market Quality from specialist (B) Resistance (R) Repetitive Momentum
+ -
profitability of floor
pressure for more floor
(R) Pressure from Floor Culture
Figure C-47: Full NYSE model (CLD) While responding to pressure from stakeholders, the NYSE management team also shaped the situation. This is the case in particular concerning floor firms and their participation by algorithms. The importance of the implementation of liquidity algorithms and the modification of specialists to designated market makers becomes obvious when comparing simulation runs of Figure C-48. When no algorithms are developed, market quality declines (line 2) due to the implementation of electronic trading, and floor firms as well as customers who prefer floor trading have a much greater potential to hold up electronic trading (line 4). A strong Liquidity Algorithms Loop weakens the balancing loops Customer Pressure for Market Quality and Floor Resistance. As a consequence, more electronic trading can be implemented as is shown in the base run behavior (line 3). This happens without significant shortcomings in market quality (line 1) so that several stakeholder groups can be satisfied at the same time. Market quality only reveals a short-term drop before sufficient algorithms are in place and then rises again. This is also what could be observed in reality. A recent study in the area of finance has investigated the impact of the NYSE’s migration to the Hybrid Market on market quality and speed. The latter has improved significantly as compared to pre-Hybrid. But the move to electronic trading has also reduced the proportion of shares which are price-improved by specialists and floor brokers.543 But since late 2008 participation has increased and so has market quality and the time the floor quotes at the NBBO. Less volume is routed away to other stock exchanges.544
543
See Gutierrez and Tse: NYSE execution quality subsequent to migration to hybrid, 2009, pp. 72 and 79. See also Chapman, Mehta and Scotti: Men At Work, 2007, p. 54. 544 See Clark: Growth in the NYSE's Liquidity Provider Programs, 2009, electronic source; NYSE Euronext Inc.: Cash Equities: Designated Market Makers, 2009b, electronic source, p. 1; and NYSE Euronext Inc.: NYSE Equities: NYSE Market Model Update, 2009c, electronic source, pp. 1–2.
144
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
Liquidity Algorithms 3
1.2 Dmnl 1 Dmnl 1.1 Dmnl 12 0.5 Dmnl
12
1 Dmnl 0 Dmnl 3 4 34 1970 1980
1 2
12
34 1990
1 2
34 2000 Date
1
1 3
4
3 4
4
2 2010
34
1 2
1 2
2020
2030
"market quality from sp. participation" : base run 1 1 1 Dmnl "market quality from sp. participation" : no algorithms Dmnl 2 2 3 3 3 3 "NYSE Fraction of E-Trade" : base run Dmnl "NYSE Fraction of E-Trade" : no algorithms Dmnl 4 4 4
Figure C-48: Importance of liquidity algorithms and market quality By keeping the floor, the organization was able to reduce some of the negative consequences the shift has for some of its customers. The different behavior that liquidity algorithms cause explains the importance the floor still has for the NYSE. The NYSE’s hybrid strategy also serves those customers well who prefer market quality—something which could not have been achieved if the organization had turned into a purely electronic exchange. While institutional customers gained a lot of importance for the New York Stock Exchange, the organization did not entirely reorient towards electronic trading. It still kept many of its old values, particularly concerning market quality and specialist participation. Apart from allowing for e-trade, it kept the floor at a reduced level to add value by the provision of liquidity, price improvement, and reduction of volatility. After the number of specialist or DMM firms has declined for decades, in February 2010 a pivot firm of electronic trading joined the floor to become a DMM. Traders interpreted this as a sign of the continuing importance of a physical trading floor despite the dominance of electronic trading.545 Nevertheless, for a long time specialists and floor brokers dominated the exchange, and the management’s decision-making was highly biased to their favor. This has fundamentally changed, and they have lost the privileges they formerly had. The valuation of the floor is now assessed by the value it contributes to a fair and orderly as well as high quality market. In 2010, stock exchanges were competing on the basis of speed. One may imagine that once all stock exchanges have implemented electronic platforms and technology reaches physical boundaries, the importance of speed will deteriorate since it
545
See Peterson, Kristina and Jacob Bunge: Getco Becomes NYSE Market Maker, in: Wall Street Journal (2010), issued 12 Feb. 2010, p. C.5.
C.IV Analyses of Model Structure and Behavior
145
then becomes not an order winner, but an order qualifier.546 Then, the NYSE’s involvement of specialists/DMMs may gain in importance again, giving the stock exchange an advantage over its competitors. The human element is thus likely to prove useful also in a highly electronic environment.
C.IV Analyses of Model Structure and Behavior Now, after the structure of the model as well as its behavior have been revealed, it will be important to gain more confidence in the system dynamics model. This is done with the help of several tests which are applied to its structure and behavior. The model needs to be able to describe the aspects that explain the respective problem, here the radical adaptation of the New York Stock Exchange to electronic trading. Cyert and March suggest the comparison of model output with actual data as an indicator of fit.547 This test alone is not yet sufficient because the comparison cannot validate the appropriateness of the model, but only serves as one indicator of fit. One of the benefits of the system dynamics method becomes apparent here: it is not only possible to test model behavior, but model structure as well. Model structure, parameterization, resulting behavior, as well as policies need to match the problem case of the NYSE.548 The model thus needs to be checked in all of these areas. It is important that in a homomorphous way the model elucidates the essential structural elements of the real system that generate important behavioral patterns. Parameter values also have to be adjusted to the case. Particularly if data is qualitative and ordinal measures are used, the influence of inaccuracy and estimation errors on the behavior must be checked in order to judge the quality of the model.549 Additionally, a thorough examination of behavior follows, although a fit of model behavior to behavior observed in the real world does not yet establish validity. Therefore, first, plausibility of the model itself is tested as well as, second, the consistency of model with real world behavior. The behavioral consistency analysis helps to make sure that the behavior of the model matches schematic patterns of behavior observed in the real world.550 Third, scenario tests are included in order to check the plausibility of behavior in possible different worlds. The model then constitutes a testing laboratory which can be used to check different policies. It can be employed to analyze the New York 546
The shift from speed being an order winner to an order qualifier will also change the shape of the function that transforms the relative speed into an adjustment of market share (effect of time to execution on market share, see Figure C-38 on page 134). 547 See Cyert and March: A Behavioral Theory of the Firm, 1963, pp. 93–96. Behavioral tests of models that involve transient behavior such as s-shaped growth or cycles are difficult to do with standard statistical measures. In the view of Barlas it is best to compare output visually to behavior patterns observed in reality. See Barlas: Formal aspects of model validity and validation in system dynamics, 1996, p. 194. 548 See Forrester and Senge: Tests for Building Confidence in System Dynamics Models, 1980, p. 213; and Milling: Der technische Fortschritt beim Produktionsprozeß, 1974, p. 209. 549 See Milling: Der technische Fortschritt beim Produktionsprozeß, 1974, pp. 212–213. 550 See ibid., pp. 214–216; and Sterman: Business Dynamics, 2000, p. 860.
146
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
Stock Exchange’s possible decision-making rules and reactions to changes in their environment. The simulator may also add and subtract model structure in order to test the sensitivity of behavior policies to changes in the model boundary.551 Nevertheless, it is difficult to separate and label model tests since one test often provides information on several aspects.
C.IV.1 Confidence in Model Structure and Parameterization Concerning the structural elements, first, the structure of the model itself, second, its boundary, and third, parameters are important. To begin with, structural elements and relationships need to be validated. It is even more important to show this for soft variables, such as customer orientation, power of floor firms, and market quality. The derivation of model structure from aspects of data strengthens the confidence that all elements of model structure conceptually correspond to the real world.552 This structural analysis has been done in parallel with the model building process by the close linkage of variables and causal relationships to data from the case study. Model analysis is an iterative process in which the boundaries between construction and testing become blurred so that some tests could already be done and were described in the chapter on model construction. Second, the piecemeal building process of the model was important for the validation of structure since it led to the adequate model boundary. An adequate boundary aggregates the model in such a way that important feedback loops are still included.553 This was, for example, done in the representation of market share which is highly simplified in relation to reality, but still includes the important Liquidity and Spread Loops that bundle liquidity. A too narrow system boundary, i.e. when only environmental developments are taken into consideration, leads to inadequate model behavior.554 It also became obvious that the reactions of important stakeholders to the NYSE decisions and strategy need to be included in an endogenous manner. As Figure C-49 shows, these stakeholder responses are the customers’ development of 551
See Forrester and Senge: Tests for Building Confidence in System Dynamics Models, 1980, p. 225. 552 See ibid., p. 213. 553 See Größler, Andreas: Modelltests, in: Strohhecker, Jürgen (Ed.): System Dynamics für die Finanzindustrie, Frankfurt a. M. 2008a, pp. 261–262. The way commissions and prices were included into the model represent an example of aggregation. The impact of commissions on market share has not explicitly been modeled although there is evidence that execution costs also have an impact on order flow and market share. They were incorporated indirectly by including the price improvement of the specialist. This can also be regarded as a part of lowering execution costs. For evidence of the impact of execution costs see Boehmer, Jennings and Wei: Public Disclosure and Private Decisions, 2007, p. 317. 554 The work by Kim exemplifies the importance of setting the right system boundary for analyzing a problem. See Kim, Hyunjung: Broadening Boundary Perception in a Multi-organizational Context: Study of a Community Mental Health Program in New York State, in: Dangerfield, Brian C. (Ed.): Proceedings of the 26th International Conference of the System Dynamics Society, Athens 2008, pp. 9 and 14–15.
147
C.IV Analyses of Model Structure and Behavior
dissatisfaction, reactions by floor firms, and their importance for market share. Additionally, the management decision process and its reaction to stakeholders are central as well. Since they interact with each other, including them in an endogenous manner proved necessary. While developments in the securities market were important for management and stakeholder dynamics, the market itself is not the focus of this dissertation and is thus not included in an endogenous manner. The causal mechanisms which led to these developments in the market were outside the model boundary. Several important market characteristics, such as the number of securities held by institutional investors as well as technological advancements were included exogenously. Some of them, e.g. the reasons for the growth of institutional investors or for the inauguration of Regulation NMS, were described verbally in order to gain a thorough understanding of the situation. MARKET SHARE - Spread (Liquidity) - Speed
PRESSURE FOR E-TRADE (-) Market share from speed
- Market quality
(+) Market share from market quality
- From institutional customers deriving from dissatisfaction with speed (-) Pressure by inst. customers
MGMT DECISION ON EXTENT OF E-TRADE
MARKET CHARACTERISTICS
- Fractional change
- Institutions
- Customer orientation
- Technology
- Pcvd relative pressure
- Market speed
(+) Repetitive momentum (+) Repetitive attention (customer ortientation)
(-) Pressure by non-inst. cust. (-) Resistance by floor (+) Cultural pressure by floor
INERTIA
PRESSURE FOR FLOOR TRADE
- Institutionalization
- From non-institutional customers
- Market share
deriving from market quality - From floor firms as resistance by stakeholders as cultural pressure by stakeholders
Figure C-49: Sector diagram making explicit the model boundary Several further aspects were outside the boundary. The system dynamics model makes simplifications concerning the representation of market share. To a large extent market share is determined by a simplified Liquidity Loop, but influences from e.g. order types and trade anonymity are left out. The latter has been integrated in an indirect manner as part of electronic trading. Additionally, already in the 1970s to 1990s, automation was put into practice in order to accelerate the manual trading process and to support the floor. The NYSE automated information systems, order routing, broker and specialist support, and else. These developments were excluded because, first, they do not allow for the leap from a trade that takes multiple seconds to sub-second speed, and second, they represent a continuation of the old market model, not a deviation from it. The same is true for enhancements in trading speed
148
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
that occurred after the implementation of the Hybrid Market. While the Hybrid initially allowed for order execution within 300 milliseconds, it took only about 5 milliseconds to execute a trade in the year 2009. These factors are included indirectly by a higher fraction of electronic trading that is possible in and after 2009, but they are what can be called incremental modifications of the current strategic orientation as opposed to more radical transformations or discontinuities.555 The case study of the NYSE reveals how managerial and inertial forces as well as endogenous pressure from culture and resistance impinge on environmental drivers of change. Figure C-50 supports the importance of endogenous forces. Concerning the trend of the remaining market, two different assumptions are modeled: a linear growth of electronic trading as well as the more realistic limited exponential growth. These two different market behaviors help analyze how closely the adoption pattern of e-trade in the remaining marked shaped electronic trading within the NYSE. The resulting behavior shows no difference. Further runs revealed that the timing of e-trade in the market may slightly change the point of time of the implementation of e-trade at the NYSE, but the pattern of radical change is always prevalent. Thus the development of electronic trading in the market may be necessary for the NYSE to automate, but how the automation is executed and how the implementation process unfolds within the NYSE is subject to endogenous management and stakeholder dynamics. In the following sub-chapters, these endogenous aspects will be analyzed more closely, particularly the effects of each of these determinants on the implementation of e-trade, pressure for and against e-trade, and market share.
NYSE and Market Fraction of E-Trade 12
1
41234123412
12
0.75 Dmnl
3
0.5
2
0.25
2
1 0 123412 34 1970 1980
1 34 1990
34 2000 Time (Year)
34
2010
2020
2030
"fraction of e-trade in remaining market" : base run 1 1 1 "fraction of e-trade in remaining market" : linear techn 2 2 2 "NYSE Fraction of E-Trade" : base run 3 3 3 3 3 "NYSE Fraction of E-Trade" : linear techn 4 4 4 4
Figure C-50: Linear development of e-trade in market 555
For a grundlegend typology of incremental and discontinuous change see Nadler and Tushman: Types of Organizational Change, 1995, p. 22. Tushman and O’Reilly III emphasize the necessity of organizations to be ambidextrous and capable of mastering evolutionary and revolutionary change. See Tushman and O'Reilly III: Ambidextrous organizations, 1996, p. 24.
C.IV Analyses of Model Structure and Behavior
149
Third, in addition to structural elements, parameters have been adapted to mirror the conditions of the case study for external validity. For instance, cohesiveness of floor firms was found to range on a fair average level. Additionally, the reference fractional inertia decrease at the NYSE was adapted to real-world data and the respective circumstances of the exchange. Many other parameters, such as the reference fractional change in trading and in customer orientation have been adjusted as well. Areas in which the model reacts sensitively to changes in parameters will be pointed out. When the initial market share parameter is changed, it is not only possible to simulate trajectories of the NYSE, but also that of its smaller competitors in the securities market. For a stock exchange that starts out with a market share of only a few percent and that otherwise has the same parameterization as the NYSE, the resulting behavior almost overlaps with that of the NYSE base run.556 The smaller market participant might need a different organizational setup in the area of inertia, for example. However, its simulation revealed a broader applicability of the system dynamics model. A validation test for extreme parameter assumptions serves as an internal parameter test and helps gain confidence in model structure because it illustrates the model’s reasonable behavior under extreme conditions. For example, the model shows sensible results when it is assumed that no technology develops or no institutional customers hold securities. In both cases e-trade is not implemented. Furthermore, tests in which e.g. the NYSE fraction of e-trade, customer orientation or the power of floor firms had extreme initial values produced sensible results. Tests with simultaneous changes in multiple parameters will be subject matter of the following chapter. Despite its ability to explain the interaction of drivers of change, the system dynamics model makes several simplifying assumptions. Floor brokerage and specialist firms are increasingly owned by large financial institutions. This has implicitly been incorporated by the power of floor firms which decreases with the rise of institutional customers. The effect of speed becoming an order qualifier instead on an order winner could more closely be investigated, but was not considered important for the phenomenon of inertia and change that rather concerned the NYSE’s and the securities market’s past.
C.IV.2 Validation of Model Behavior and Sensitivity Sensitivity analyses are conducted to analyze model robustness and the relationship between structure and behavior.557 They investigate the strength of the model’s reac556 557
Appendix B compares the simulation runs. See Sterman: Business Dynamics, 2000, p. 830. In the sensitivity analyses, the model is run 500 to 10000 times while a certain parameter or a group of parameters is changed. In all sensitivity analyses it is assumed that Regulation NMS
150
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
tion to parameter changes, i.e. to changes concerning the model’s assumptions. Minor reactions to changing assumptions point to the robustness of the policies and results. If the model reacts sensitively, this can also point to important feedback mechanisms, to levers of human intervention and the likelihood of different scenarios. The sensitivity of the model behavior to changes in the pressure of institutional customers, for the pressure of floor firms, and for management attitudes will be presented. A sensitivity analysis for the pressure that develops out of the dissatisfaction of institutional customers was run, and its boundaries compare to the extreme conditions test in which no e-trade develops in the market. The reference pressure per dissatisfaction unit was varied from 0.01 to 5 with a normal value of 1. Figure C-51 indicates that the customer reaction is an important driver of change, meaning that extremely strong institutional customers can force an earlier and more gradual implementation of electronic trading. If institutional customers do not develop strong pressure, then in rare cases e-trade is not implemented to the full extent. If an organization thus operates in a niche market and enjoys stable relationships to customers that do not go along with the new trend, it may be relatively unaffected by a major shift of the main market. While the variation in the reference pressure from institutional customers has effects on model behavior, variation in the reference pressure per non-institutional customer neither has an effect on the timing nor on the extent of the implementation of e-trade.558
558
comes into effect, but the effect of the Grasso scandal is left out since Regulation NMS is assumed not be influenced by the NYSE’s behavior, but the scandal is. The output graphs of the sensitivity test are shown in APPENDIX B. Here the reference pressure per non-institutional customer was changed in the range of 0.01 to 5 with a base-run value of 0.5.
151
C.IV Analyses of Model Structure and Behavior sens institutional pressure 50% 75% 95% 100% "total pressure for more e-trade from customers" 100
75
50
25
0 1970
1980
sens institutional pressure 50% 75% 95%
1990
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
100%
"NYSE Fraction of E-Trade" 1
0.75
0.5
0.25
0 1970
1980
1990
changed parameter ref. pressure per dissatisfaction unit
base run value 1
parameter range 0.01 – 5
Figure C-51: Sensitivity for institutional customer pressure for e-trade The investigation of pressure from floor firms for the retention of the floor sheds light into the influence floor firms have on the implementation of electronic trading. A first step analyzes the effect of their resistance on the NYSE decision-making. The upper graph in Figure C-52 shows that the extent to which e-trade is implemented is not sensitive to variations in resistance from inadequacies in employability and profitability. Here, the reference resistance pressure per floor firm, the time to adjust desired participation and the time to adjust desired earnings were changed simultaneously to express differences in the strength of resistance and in their durability. Different parameter values create pressure to a greater or lesser extent for the specialist system (bottom graph). But they do not affect the decision for e-trade since resistance occurs only after the New York Stock Exchange started implementing at least some e-trade and had already shifted its attention towards institutional customers. In this case the management team’s reaction is more important than the re-
152
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
sistance reaction by stakeholders on the floor since managerial attention and openness determine to what extent the management team actually considers stakeholders to be important. sens resistance 50% 75% 95% "NYSE Fraction of E-Trade" 1
100%
0.75 0.5 0.25 0 1970
1980
sens resistance 50% 75%
95%
1990
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
100%
total pressure for more floor trade from floor 80 60 40 20 0 1970
1980
1990
changed parameters ref. resistance pressure per floor firm time to adjust desired participation time to adjust desired earnings
base run value 1 5 5
parameter range 0.01 – 5 1 – 20 1 – 20
Figure C-52: Sensitivity for resistance The power of floor firms may be a more sensitive determinant of the extent of e-trade. Floor firms gain power from the participation of the specialist in trading and the positive effect of participation on market quality. They lose power due to the rising number of institutional traders who like to bypass floor intermediaries. When the sensitivity of the model is checked for changes of the reference power of floor firms, the fraction of e-trade shows a very low sensitivity, too. In general, sensitivity analyses for resistance, the floor firms’ power, and even for their cohesiveness show that variances in each pressure for itself do not have much effect on the fraction of e-trade at
153
C.IV Analyses of Model Structure and Behavior
the NYSE.559 It seems that a group of forces determines the implementation of e-trade, as simultaneous changes of the cultural variables have complementary effects. Therefore, the sensitivity of the model is checked for several floor constituencies at the same time—for the resistance variables as above, for the reference power of floor firms, and the degree of cohesiveness of floor firms. Here, here the output graphs show more variation. Management was parameterized as in the base scenario. Figure C-53 a) suggests that when all forces work together the effects on the implementation of e-trade are much greater. The effect of feedback becomes obvious in the figure on this and the following page. Different extents of the institutional pressure for e-trade (b) affect the fraction of e-trade (a) which then leads to more or less pressure for the floor (c) and differences in market share (d).560 Concerning the NYSE this reveals that even with a different value of stakeholder pressure change could not have been implemented much earlier. If culture, resistance and related aspects had been stronger and institutional pressure weaker, e-trade could have been held up and been implemented to a lesser extent. This shows that affected stakeholders have some influence on the extent to which change is implemented if they are sufficiently strong and cohesive at the same time. There are several feedback loops which include the floor’s reaction to the management team’s decisions. When the strength of only one of these loops is changed, the system exhibits ‘policy’ resistance and the behavior remains very similar to the base run. The management loops’ dominance is responsible for the immutable behavior. The simultaneous alteration of several floor parameters is much more effective because the influences are multiplicative and this strength of pressure is needed to affect the persistent management behavior. a)
sens floor 50% 75%
95%
100%
"NYSE Fraction of E-Trade" 1 0.75 0.5 0.25 0 1960
559
560
1971
1983
1995 Date
2006
2018
2030
Appendix B illustrates the respective sensitivity analyses for the reference power of floor firms and the degree of cohesiveness of floor firms. The total pressure for more floor trade from customers has been excluded from the figure because it shows the same behavioral pattern as the total pressure for more floor trade from the floor that is shown in graph c) of Figure C-53.
154 b)
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE sens floor 50% 75%
95%
100%
"total pressure for more e-trade from customers" 60 45 30 15 0 1970
c)
sens floor 50% 75%
1980
95%
1990
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
100%
total pressure for more floor trade from floor 80 60 40 20 0 1970
d)
1980
sens floor 50% 75%
95%
1990
100%
NYSE Market Share 1 0.75 0.5 0.25 0 1970
1980
1990
changed parameters ref. resistance pressure per floor firm time to adjust desired participation time to adjust desired earnings ref. power of floor firms cohesiveness of floor firms
base run value 1 5 5 100 0.7
parameter range 0.1 – 5 1 – 20 1 – 20 20 – 200 0.1 – 2
Figure C-53: Sensitivity for resistance, power, and cohesiveness of floor firms The management team’s importance has already become obvious in the descriptions above. The influence of the reinforcing loops which represent managerial decisions will be analyzed now. It will be tested how sensitively the system reacts when stakeholder parameters are kept at their base run values but management attributes are changed. These changes comprise the adaptability of the trading mechanism, i.e. the organization’s general responsiveness to pressure, as well as the management’s adaptability of customer orientation.
155
C.IV Analyses of Model Structure and Behavior
Figure C-54 shows a sensitivity analysis of the organization’s general openness to change in which the management’s reference fractional change in trading per pressure p.a. was altered. Graph a) of Figure C-54 reveals that the management team’s general openness to pressure decides between an earlier and smoother adaptation and the observed radical change. The NYSE adapts earlier for high values of the reference fractional change. Radical change develops when the reference fractional change is low. Yet, the earlier adaptation is also delayed by about ten years in comparison to the market since the organization still starts with a high level of inertia. Missing adaptation leads to strong institutional pressure for electronic trading, as depicted in graph c) on the following page. Together with falling market share this pressure triggers some change which quickly amplifies by the reinforcing loops Repetitive Momentum and Repetitive Attention. Graphs a) and b) illustrate this phenomenon: Low initial change is followed by a radical shift and a resulting sudden depletion of inertia. Only when the NYSE management is not at all reactive to pressure, is e-trade not implemented to the full extent. The floor also reacts with pressure when automation threatens its survival (graph d), but this reaction only lasts several years. First of all, the floor firms become accustomed to the situation, and second, liquidity algorithms improve their situation by improving participation and market quality, depicted in graph e). a)
sens change per pressure 50% 75% 95%
100%
"NYSE Fraction of E-Trade" 1 0.75 0.5 0.25 0 1970
b)
1980
sens change per pressure 50% 75% 95%
1990
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
100%
Inertia 1 0.75 0.5 0.25 0 1970
1980
1990
156
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
c)
sens change per pressure 50% 75% 95%
100%
"total pressure for more e-trade from customers" 60 45 30 15 0 1970
d)
1980
sens change per pressure 50% 75% 95%
1990
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
100%
total pressure for more floor trade from floor 20 15 10 5 0 1970
e)
1980
sens change per pressure 50% 75% 95%
1990
100%
"market quality from sp. participation" 1.100 1.090 1.080 1.070 1.060 1970
1980
1990
changed parameter base run value ref. fract. change in trading per pressure p.a. 0.02
parameter range 0.005 – 0.2
Figure C-54: Sensitivity for the fractional change in trading per pressure A similar reaction to variations in the reference fractional change in customer orientation p.a and in the initial customer orientation can be observed. Customer orientation responds sensitively, shown in the lower part of Figure C-55. The little dip around the year 2008 in some of the simulation runs indicates pressure from the floor reducing customer orientation. The fraction of e-trade is somewhat affected from the changes in attention. A high initial as well as a responsive orientation to stakeholders is able to expedite the shift to e-trade, low adaptability of customer orientation slows down the process a little bit, but the radical behavioral pattern generally remains. A sensitivity analysis for different initial specifications of inertia has similar results and shifts the timing of the radical change (see Appendix B).
157
C.IV Analyses of Model Structure and Behavior sens change customer orientation 50% 75% 95% 100% "NYSE Fraction of E-Trade" 1 0.75 0.5 0.25 0 1970
1980
1990
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
sens change customer orientation 50% 75% 95% 100% Customer Orientation 1 0.75 0.5 0.25 0 1970
1980
1990
changed parameters ref. fract. change in customer orientation per pressure p.a. ini customer orientation
base run value 0.03 0.1
parameter range 0.005 – 0.2 0.05 – 0.95
Figure C-55: Sensitivity for fractional change in customer orientation The analyses of the management team’s general openness to change and of its concentration on institutional customers support the impact of the management and the organization on the implementation of electronic trading. Yet, so far only the reactivity to pressure has been analyzed. Figure C-56 below shows the model output for a sensitivity analysis in which additionally the levels of inertia and of customer orientation were altered. As before, it is assumed that the organization is at least somewhat responsive to performance and stakeholder pressure. The graph reveals that the shift towards e-trade could have happened somewhat faster had the organization been more reactive, and the implementation could also have gone slower if the organization had been more rigid in its reaction to outside pressure. Thus, management turns out to be an important factor concerning the implementation of electronic trading. By
158
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
different levels of inertia, openness to change and customer orientation it particularly sets the point of time when change is initiated. Strong stakeholder pressure turned out to have an effect rather on the extent to which electronic trading is implemented. Overall, the sensitive behavior towards variation in managerial disposition reveals the scope of managerial choice. Given the historic environmental conditions, the management team is an important driver of change. It can trigger adaptive behavior to external forces quite early or remain inert for a long time and then rather change radically to the strong pressure that has built up. sens management 50% 75% 95% "NYSE Fraction of E-Trade" 1
100%
0.75 0.5 0.25 0 1960
1971
1983
1995 Date
2006
changed parameters base run value ref. fract. change in trading per pressure p.a. 0.02 ref. fract. change in customer orientation per pressure p.a. 0.03 ini customer orientation 0.1 ini inertia 0.9 ref. fract. institutionalization 0.3
2018
2030
parameter range 0.005 – 0.2 0.005 – 0.2 0.05 – 0.95 0.05 – 0.95 0.15 – 0.35
Figure C-56: Sensitivity for managerial parameters So far examples concerned the effect of changes in a single or a few variables on the implementation of electronic trading. Figure C-57 shows a sensitivity analysis for a simultaneous random change of environmental, stakeholder, and managerial conditions. Subjected to a careful sensitivity analysis, the results are robust to a wide range of parameters, and they are important in two ways. They portray the robustness of the model and additionally reveal the full range of theoretical possibilities of how an NYSE-like stock exchange may react to different environments and interact with pressure from multiple sides.
159
C.IV Analyses of Model Structure and Behavior sens all 50% 75% 95% "NYSE Fraction of E-Trade" 1
100%
0.75 0.5 0.25 0 1970
1980
sens all 50% 75% 95% NYSE Market Share 1
1990
2000 Date
2010
2020
2030
100%
0.75 0.5 0.25 0 1970
1980
1990
2000 Date
2010
changed parameters base run value ref. pressure per dissatisfaction unit 1 ref. resistance pressure per floor firm 1 time to adjust desired participation 5 time to adjust desired earnings 5 ref. power of floor firms 100 cohesiveness of floor firms 0.7 ref. fract. change in trading per pressure p.a. 0.02 ref. fract. change in customer orientation per pressure p.a. 0.03 ini customer orientation 0.1 ini inertia 0.9 ref. fract. institutionalization 0.3 algorithms per gap per year 50
2020
2030
parameter range 0.01 – 5 0.1 – 5 1 – 20 1 – 20 20 – 200 0.1 – 5 0.005 – 0.2 0.005 – 0.2 0.05 – 0.95 0.05 – 0.95 0.2 – 0.35 1 – 100
Figure C-57: Sensitivity for changes in stakeholder and management parameters
160
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
The sensitive graphs of Figure C-57 make it obvious that the history and the future of the New York Stock Exchange could well have been different. Only in rare cases the shift to electronic trading would not have happened, but the extent to which etrade was implemented could have varied. The differences arise from variations in the pressure for and against automation, how vigorously the stakeholder groups pressure for their aims and how powerful their pressure is. Differing setups in the management can also lead to different trajectories. By their pressure and resistance the surrounding stakeholders exert influence on the extent to which the decision is carried out. In their combination the different stakeholder and management dispositions are able to generate different modes of behavior, ranging from smooth adaptation to radical change and even the decline of the organization. As the market share graph of Figure C-57 indicates, these theoretical possibilities lead to different positions of a stock exchange in the market so that multiple futures would theoretically have been possible.
C.V Implications of the New York Stock Exchange’s Recast of Trading Systems The present study provides an example of a radical organizational transformation. It presents an investigation into structure and behavior of organizations—here of the New York Stock Exchange—with particular focus on feedback relationship and the emergence of behavior. The different scenarios simulated by a system dynamics model have shown that, on the one hand, the change towards electronic trading is an adaptation process of the NYSE to its competitive environment. It proved those right who predicted stock exchanges to adopt electronic trading, and proved those wrong who expected the NYSE to be too inert to change and fail.561 Since technology provided for the possibility of e-trade and since there was a growth in institutional customers, electronic trading finally was implemented. Without these drivers for change from the environment, no change in the NYSE’s way of trading would have emerged. On the other hand, the mere adaptation point of view does not provide the full picture and cannot explain how and why the shift to electronic trading happened at the NYSE —i.e. the timing, pacing, and the significance of decision-making. The modeling process is indeed able to illustrate why the changes appeared to be this radical. It gave a structural account for path-dependent processes that have also been discussed in some of the literature on the punctuated equilibrium approach and 561
For an adaptive point of view for stock exchanges see Clemons and Weber: London's big bang, 1990, p. 56; Clemons and Weber: Information Technology and Screen-Based Securities Trading, 1997, p. 1695; Feldman: Electronic Marketplaces, 2000, p. 95; and Picot, Bortenlänger and Röhrl: The Automation of Capital Markets, 1995, electronic source; Picot, Bortenlänger and Röhrl: Organization of Electronic Markets, 1997, pp. 113–115; and Stoll: Electronic Trading in Stock Markets, 2006, pp. 154 and 173. For the NYSE’s expected decline see Naidu and Rozeff: Volume, volatility, liquidity and efficiency of the Singapore Stock Exchange before and after automation, 1994, p. 24; and Welles: Is it time to make the Big Board a black box, 1990, p. 74.
C.V Implications of the New York Stock Exchange’s Recast of Trading Systems
161
on cognitive determinants of change.562 The management team is highly important for the way the organization reacts to its environment. A reinforcing process of repetitive momentum made it develop path-dependent behavior and concentrate on what it had always done. The strength of this reinforcing process prevented a smoother adaptation and was responsible for the observed radical behavior. Together with the management team’s generally weak conversion of pressure to action, it caused the missing adaptation. Missed adjustment then resulted in even higher pressure and a market share decrease that eventually overpowered the reluctance to change. For a substantial period of time there had been dissatisfaction pressure that was initially held up by management, but managerial inertia started to be overwhelmed by the pressure from institutional stakeholders. The slowly rising awareness of pressure together with the shift in management and with the effects of the falling market share functioned like a valve, so that the pressure was released. Since inertia is a part of reinforcing feedback loops, a small step towards change decreased inertia, meaning it created further replacement and rethinking among the management team. The New York Stock Exchange seized the opportunity for electronic trading quite late, but it did make the decision to change. It also decided against full automation and kept parts of its old structure in a customized way in alignment with the new trading system. This was a decision of the NYSE management that revealed high elements of choice. After the shift in its strategic orientation the NYSE expressed about itself: “We adapt and evolve.”563 This statement summarizes the environment and the management as dual drivers of change since the NYSE adapts to e-trade but finds its own way of evolution in combination with a trading floor and liquidity algorithms. Liquidity algorithms as well as the conversion of specialists into DMMs take a special position since they happened later than the initial move to e-trade. They provide empirical evidence of a post-revolutionary adjustment some time after the original transformation.564 The later modification led to more full alignment with the NYSE’s environment. This later adjustment can be regarded as part of the full transformation process towards e-trade because it fine-tunes the new strategy and implements learnings from the change process. By this post-revolutionary adjustment the NYSE also provided an example of how an organization is able to simultaneously address different stakeholder demands and to cope with conflicting demands of its environment.565
562
See Jansen: From Persistence to Pursuit, 2004, p. 277; Tripsas: Technology, Identity, and Inertia through the Lens of 'The Digital Photography Company', 2009, p. 447; Tushman and Romanelli: Organizational Evolution, 1985, p. 192; and Wollin: Punctuated equilibrium, 1999, p. 363. 563 Pellecchia, Ray: It's go time for go fast (6 Oct. 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2006b, electronic source. 564 See Sabherwal, Hirschheim and Goles: The Dynamics of Alignment, 2001, pp. 193–195. 565 For the need to cope with conflicting demands see Pache, Anne-Claire and Filipe Santos: When Worlds Collide: The Internal Dynamics of Organizational Responses to Conflicting Institutional Demands, in: Academy of Management Review, Vol. 35 (2010), No. 3, p. 471.
162
C The Phenomenon of Inertia and Change Exemplified in a Case Study of the NYSE
In relation to managerial decision-making, the degree of attention to customers proved to be of specific importance. The existence of strong and then decreasing inertia is not able to fully explain why the shift towards electronic trading took place so late and was radical. An attentional bias towards the floor and a rather sudden acknowledgement of the customers’ importance supported the radical nature of the shift. Attention is part of a reinforcing and path-dependent process. The distribution of attention often biases the disposition and orientation of the management team towards a specific stakeholder group and its demands. If an organization accumulates inertia, it may reinforce its current attention to powerful groups such as floor firms and thus perpetuate tradition. While attention also follows the pressure that comes from the two stakeholder groups, as long as the management team is inert and committed to its current strategic orientation, adaptation takes place only slowly. Some existing literature on decision-making, change and adaptation take up the topic of attention to issues.566 The present case study, however, provides an example of organizational attention to legitimate stakeholders. It shows how the combination of a stakeholder group’s urgency from dissatisfaction and power determines the pressure for their respective aim. It takes into account how attention to customers developed and how it affected decisions of the management team. The case study of the NYSE gives first insights into the research question of what the drivers of change are.567 It shows that and clarifies how the evolution of the NYSE has been affected by multiple drivers. While including decision-making by the management team into the analysis proved to be necessary, narrowing the focus only to the management would not provide the desired results. Hereby, the analysis also reveals that one view alone also does not explain the NYSE’s trajectory. Both the environmental developments as well as the management team’s decisions create endogenous forces among stakeholder groups such as customer and floor firm pressure that returns to the management’s subsequent decision-making by means of feedback. The modular build-up of the system dynamics model reveals that while the environment represents an important driver of change, the endogenous dynamics of pressure by stakeholders and the management team also drive organizational evolution. Different determinants of change have to be seen in their interaction in order to make sense of the radical shift to electronic trading by the New York Stock Exchange. Jointly these drivers explain why “this old stone is shaking off the moss and rolling again.”568 As Morgan points out, each new perspective adds a little different 566
See Cho and Hambrick: Attention as the Mediator Between Top Management Team Characteristics and Strategic Change, 2006, p. 454, referring to Ocasio: Towards an Attention-based View of the Firm, 1997, p. 189; Ocasio and Joseph: An Attention-Based Theory of Strategy Formulation, 2005, pp. 56–57; and Simon: Administrative Behavior, 1976, p. 90. 567 For an answer to this question see also Milling, Peter M. and Nicole S. Zimmermann: Modeling Drivers of Organizational Change, in: Kybernetes, Vol. 39 (2010), No. 9/10, pp. 1452–1490. 568 Pellecchia, Ray: The right time (6 Oct. 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2006d, electronic source.
C.V Implications of the New York Stock Exchange’s Recast of Trading Systems
163
understanding, thus leading to a more complete picture.569 Only from a multi-paradigm view is it possible to capture how elements of deliberate choice and deterministic adaptation drive the NYSE’s evolution. The combination of perspectives shed light into the dynamics of how changes unfold. The investigation of the case study of the New York Stock Exchange and the three-step development of the system dynamics model point out that parameters have an influence on loop dominance and can put loops as well as entire model sectors out of action. This suggests that different boundaries: the environment, stakeholders, and the management—i.e. the assumption of what needs to be included in the analysis—create different assumptions about the prevalence of drivers of organizational change. Due to the focus on feedback relationship and the emergence of behavior, this investigation helps open the black box of the process of organizational change. The analysis reveals that the implementation of electronic trading at the New York Stock Exchange could have followed a different trajectory. Organizational decline due to the high inertia would have well been another possible path that also many researchers and industry experts assumed the NYSE would take. As shown by the last sensitivity graph, an early adaptation to electronic trading was also within the realm of possibilities. The importance of the pressure for the floor also became obvious in the runs in which no liquidity algorithms were included and the pressure against e-trade was higher. The failure to introduce algorithms in combination with an even stronger floor could have somewhat impeded the full implementation of electronic trading. The system dynamics analysis also demonstrates that the management team’s initial orientation to specialists, floor brokers and non-institutional customers was successful at delaying the adaptation process by masking the pressure for the automated system. The historical importance of the floor and the low initial customer orientation averted the shift towards electronic trading in the 1990s. A management team which does not direct its attention to customers even in times of performance crisis could have caused a further delay in e-trade. Since it is clear now why the NYSE followed its specific trajectory, it will be interesting to further investigate how other modes of behavior can be achieved by a management team when it faces rapid environmental change.
569
See Morgan: Images of Organization, 2006, pp. 337–341.
165
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change D.I D.I.1
A Generic Model of Organizational Inertia and Change Motivation for a Generic View
Although the New York Stock Exchange takes a special position among U.S. and even world-wide stock exchanges, its behavior is typical for a class of organizations. Organizational ‘dinosaurs’ which frequently are long-established corporations but also younger organizations often have difficulties adapting to a changing environment. They exhibit inertia and often adapt late or not at all. The short recall of several known examples will demonstrate this similar pattern of behavior. Just like the NYSE did for a long period of time, the computer manufacturer DEC also failed to undergo necessary change in its strategy and culture. The innovation of the personal computer altered the organization’s market environment. A new group of private customers emerged who had different preferences and were served by new competitors. Since DEC’s management did not perceive the needs of this customer group, it missed major market opportunities.570 The organizational culture remained focused on clients with a strong technological interest. The grown belief that technologically sophisticated computing products will prevail in the market was deeply embedded in DEC’s culture and resulted in a reduced perception of the radically altered environment with its new solutions and stakeholder groups.571 The DEC example shows that apparently similar reasons led to the failure to change at DEC and at the NYSE. In both cases a missing orientation to an important stakeholder group has had significant impact on the biased perception of pressure to change. Attention to customers was also a main factor leading to the demise of the camera and film manufacturer Polaroid. Although the company did invest in a new technology, it misjudged the rising importance of a new group of customers and their preference for digital photography. The failure to depart from its established business model was grounded in the management team’s cognitive representations. Adequate adaptation to the new environment would have required change in strategic beliefs.572 The case of Nestlé was somewhat different from DEC and Polaroid. Customers accused Nestlé of and boycotted it for its unethical marketing practices in the developing world. The question did thus not center on the adoption of a new technology, but rather on a new strategy of ethical conduct and respective marketing. At the same time it is also an example of a long-term neglect of stakeholder preferences 570
See Schein: DEC is Dead, Long Live DEC, 2003, p. 291. See ibid., pp. 74. 572 See Tripsas and Gavetti: Capabilities, Cognition, and Inertia, 2000, p. 1158. 571
N. Zimmermann, Dynamics of Drivers of Organizational Change, DOI 10.1007/978-3-8349-6811-1_4, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011
166
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
and of a strong orientation towards its taken for granted strategy. The company regarded the boycotters’ complaints as an ephemeral phenomenon and paid little attention to them. Even more than a decade after Nestlé’s marketing practices had been denunciated, the organization was still regarded as not having undergone a true ideological change.573 Finally Nestlé learned, and also started to promote its ethical behavior on its company website. In the three cases described above, organizational inertia played a critical role. In relation to managerial cognition and bounded rationality, it often resulted in the failure to attend to an important group of stakeholders. In this respect, the reaction of selected organizations to the personal computer, digital photography, and to a rising demand of consumers for corporate ethical conduct bears resemblance to the NYSE’s reaction to the emergence of e-trade and the growing emphasis on speed vs. price among institutional customers. The similarities between the different examples give reason into a further investigation of the causalities that generate behaviors at the NYSE and that were also apparent at Nestlé, DEC and Polaroid. These causal relationships and their informational value for the research questions guiding this investigation will be analyzed in the following sub-chapters. They will concentrate on causal relationships and dynamics among drivers of change and the influence of previous changes on further transformations. Causal relations and resulting behavior will be examined with a generic system dynamics model. It will portray the characteristics of a ‘canonical situation model’. According to Lane, such a model is general and applies to a specific domain (or class) of systems and is often derived from a more specific application case. Depending on the parameter and policy choices employed, it is able to generate significantly different modes of behavior. These types of generic models serve “as general theories of structure and behaviours of a domain.”574 The system dynamics model is supposed to explain behavior observed in the three cased described. Although the emphasis in system dynamics modeling is on modeling a problem instead of a system, many models which are developed for specific purposes are able to make generic contributions, e.g. by revealing causal relationships that generate problems also in related cases. Forrester points out that in the best of cases a generalized model is created. It is a theory for a particular class of systems that can be adapted to specific circumstances by parameterization. The generalized structure explains phenomena and modes of behavior encountered in similar situations.575 System dynamics models are structural theories of social and in 573
See Post: Assessing the Nestlé Boycott, 1985, p. 123; Richter, Judith: Holding Corporations Accountable: Corporate Conduct, International Codes, and Citizen Action, New York 2001, pp. 77–78; and Sethi: Multinational Corporations and the Impact of Public Advocacy on Corporate Strategy, 1994, p. 70. 574 Lane, David C. and Chris Smart: Reinterpreting 'generic structure': Evolution, application and limitations of a concept, in: System Dynamics Review, Vol. 12 (1996), No. 2, p. 102, see also p. 91. 575 See Forrester: Industrial Dynamics—A Response to Ansoff and Slevin, 1968, p. 607.
D.I A Generic Model of Organizational Inertia and Change
167
particular of socio-economic systems.576 Hence, a system dynamics model as the one described above reaches beyond the explanatory value of the single case of the New York Stock Exchange. It does not only shed light onto behavior and structural causalities of a specific example, but is able to explain generic characteristics of specific social phenomena. Since the generic system dynamics model accounts for a specific class of organizations—i.e. established organizations that face environmental change—on the continuum between small and middle range theories it is situated somewhere towards the latter. It represents a dynamic theory that is able to explain adaptive, inertial, and radical patterns of behavior observed in the real world. The system dynamics approach can not only be used for supporting concrete decisions in organizations. It can also be used to understand the structure of a decision context and behavioral patterns in general. There are many examples when system dynamics models were employed to test, enlarge, or develop theories. E.g. Größler developed a concept model that links key strategic competitive factors in production: i.e. time, cost, quality, and flexibility. Rudolph and Repenning showed how interruptions in the organizational routines can lead to organizational collapse. Sastry explicitly tested and enlarged Tushman and Romanelli’s punctuated equilibrium theory.577 Based on a concrete example, Saysel and Barlas developed a generalized simplification process that has been applied, for example, by Kopinsky et al. In the view of Saysel and Barlas, generic structures, simplified from one or several case-specific models, are able to transfer insights within or across application domains.578 Schwaninger and Grösser also demonstrated the use of a case study as a means and locus of theory building in the area of product launch strategies. Based on SD modeling and a case study, they exemplified how to develop theory that is applicable to an entire class of systems and is thus closer to a middle range than a small range theory.579 This places the generic system dynamics model or theory which is developed here in the tradition of previous system dynamics research. 576
See Barlas: Formal aspects of model validity and validation in system dynamics, 1996, p. 187; and Größler, Andreas: Struktur und Dynamik strategischer Fähigkeiten in der Produktion, Mannheim 2007, p. 150. For information on system dynamics models or simulation models as a theory see Cohen and Cyert: Computer Models in Dynamic Economics, 1961, p. 113; Forrester: System dynamics, systems thinking, and soft OR, 1994, p. 253; and Kopainsky and Luna-Reyes: Closing the Loop, 2008, pp. 474 and 483. 577 See Größler: Struktur und Dynamik strategischer Fähigkeiten in der Produktion, 2007, in particular p. 158; Rudolph and Repenning: Disaster Dynamics, 2002, p. 24; Sastry: Problems and Paradoxes in a Model of Punctuated Organizational Change, 1997, p. 266; and Sastry, M. Anjali: Understanding Dynamic Complexity in Organizational Evolution: A System Dynamics Approach, in: Lomi, Alessandro and Erik R. Larsen (Ed.): Dynamics of Organizations: Computational Modeling and Organization Theories, Menlo Park, CA [et al.] 2001, p. 400. 578 See Saysel, Ali Kerem and Yaman Barlas: Model simplification and validation with indirect structure validity tests, in: System Dynamics Review, Vol. 22 (2006), No. 3, p. 260. For a further application see Kopainsky, Birgit, et al.: A Blend of Planning and Learning: Simplifying a Simulation Model of National Development, in: Simulation & Gaming (2010), pp. 6–7. 579 See Schwaninger and Grösser: System Dynamics as Model-Based Theory Building, 2008, pp. 448, 457 and 461.
168
D.I.2
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
Generic Model Structure
Causally, the generic model of the organization-environment relationship is highly similar to the NYSE-specific model. The managerial decision-making structure was kept in order to adequately capture the reaction of organizational decision-makers to their environment. This part has been condensed already. Pressure from stakeholders for the new strategy—called strategy B—is represented by one feedback loop as it was also in the NYSE case. The pressure from the old stakeholders for the retention of the old system—here for strategy A—was simplified since in the case of the NYSE the floor firms displayed idiosyncratic features of their culture and power in relation to the NYSE management that cannot be expected to hold in the majority of cases. The causal loops and the respective stock and flow structure will be described next. The order of description will be the same as for the NYSE model: starting with the new developments in the market and stakeholder pressure for a new strategy— here for strategy B—resistance from stakeholders favoring the old strategy A will be added. Finally, the managerial mechanisms will be presented. In an analogous manner to floor and e-trade, the two strategies A and B will be discussed. Table D-1 gives an overview. Strategies may stand for a focus on mini vs. personal computers, for analog vs. digital photography, or for a pure cost strategy vs. a strategy that incorporates customer demands for ethical conduct. While the focal organization and the remaining market pursue the traditional strategy A, a new strategy B is developed, e.g. by a competitor in the market. This new strategy offers a different set of qualities that are preferred by a group of stakeholders. Strategy A (old strategy)
Strategy B (new strategy)
floor trade, analog photography, minicomputer, …
e-trade, digital photography, PC, …
price quality in trading, Quality A resolution quality in photography, computing power
high
often low
speed in trading, Quality B digital editing ability in photography, applications of home computing
low
high
Table D-1: Strategies and their qualities In the base scenario, the development is assumed to be no single invention, but to be a process such as in the case of electronic trading that takes about 15 years to develop and on average 5 more years to be implemented in the market. The long development seems reasonable: The interest in ethical conduct started to increase in the 1980s and is still current in the year 2010. Digital photography started to develop in the 1980s, and became the dominant design around the year 2000. Yet, the struc-
169
D.I A Generic Model of Organizational Inertia and Change
ture also allows the testing of alternative developments. The development of strategy B, pictured on the left of Figure D-1, diffuses in the remaining market with the respective time delay of 5 years, as in the NYSE case.580 It nd increases the fraction of stakeholders favoring B. The latter fraction, when multiplied by the total number of stakeholders, results in the number of stakeholders favoring B, shown in the upper part of the figure. Compared to the NYSE case, these external influences on the model have been simplified.581 TOTAL NO. OF STAKEHOLDERS number of stakeholders favoring B + +
DEVELOPMENT OF STRATEGY B
+
diffusion of B in remaining market -
fraction of stakeholders favoring B +
+ TIME TO DIFFUSE B IN REMAINING MARKET
REF.QUALITY B OF STRATEGY B
desired quality B +
REF. QUALITY B OF STRATEGY A
Figure D-1: Diffusion of B (SFD) Many stakeholders prefer B because it succeeds at discovering the potential that is unused by the old strategy A. Strategy B offers a special quality B such as speed in the case of stock trading or usage in households in the case of personal computers. On the bottom right of Figure D-1, the desired quality B forms from the diffusion of B in the remaining market which works as a weight for the reference quality B of strategy A and B respectively. The quality B of strategy B is set to 1 whereas that of strategy A is a rather small number (here 0.1). The computation of the desired quality B is broken down in the following equation D-1: desired quality B [quality unit] = diffusion of B in remaining market [dmnl] • REF. QUALITY B OF STRATEGY B [quality unit] + (1 – diffusion of B in remaining market [dmnl]) • REF. QUALITY B OF STRATEGY A [quality unit]
D-1
The concept of the diffusion of a new strategy that entails an idiosyncratic quality does not only comply with the emergence of e-trade and the growing emphasis on speed vs. price among institutional customers. It also extends to the increasing preference for digital storage in photography, and the applications of home computing as well as the availability of mobile telephony also serve as examples of the rising quality B of a new product or strategy. Apart from the invention of new technological appli580 581
Appendix E lists the model equations of the generic model. For simplification of parameters and their causes see Saysel and Barlas: Model simplification and validation with indirect structure validity tests, 2006, pp. 256–257.
170
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
cations, the mechanism also extends to developments that require a new strategy that involves a rethinking in the management or a change of the corporate culture. In the case of ethical conduct, the development of B may be an increasing interest in ethics, resulting in a group of stakeholders demanding ethical behavior from organizations.582 Quality B could in this case be interpreted as a product’s ethical quality. These market changes are similar to what Christensen called disruptive innovation. The latter is a technologically inferior product with features different from the old product but which customers value. Christensen subsumes digital photography and the emergence of ECNs under disruptive innovations.583 The concept described in the generic model of this dissertation extends to non-technological changes in the market to include significant shifts in customer or stakeholder demands that impinge on an organization. While disruptive technologies rather describe changes in the market, the focus is on transformations that require radical change and re-thinking in organizations. + <desired quality B>
-
pcvd inadequacy of strategy per stakeholder B
rel. quality B +
REF.QUALITY B OF STRATEGY B
(B) Adaptation Pressure for B
+
+ total stakeholder pressure for more B
quality B + + -
REF. QUALITY B OF STRATEGY A
+ + stakeholder pressure for more B
REF. PRESSURE PER STAKEHOLDER FAVORING B
+
Orientation to Strategy A
change in strategy +
Orientation to Strategy B
+ pcvd pressure from stakeholders favoring B
Figure D-2: Adaptation pressure for strategy B (SFD) Those stakeholders favoring strategy B compare the developments of this new strategy and its quality dimension to the focal organization’s orientation, as described in the upper left of Figure D-2. In this respect, the focal organization’s quality B, that derives from its orientation to strategy B rather than to the old strategy A, gets compared with the desired quality B in the market. The resulting relative quality B, shown in Figure D-3, is a measure for the adequacy of strategy that those stakeholders favoring B perceive. This perceived inadequacy of strategy per stakeholder B is the 582
For more information on the rising demand of ethical conduct see Miczka, Switbert, et al.: Walk the Talk: Implementing Ethical Conduct in Industrial Production Systems, in: Strohhecker, Jürgen and Andreas Größler (Ed.): Strategisches und operatives Produktionsmanagement: Empirie und Simulation, Wiesbaden 2009, pp. 91–92. For an opposite view see Carrigan, Marylyn and Ahmad Attalla: The myth of the ethical consumer - do ethics matter in purchase behaviour?, in: The Journal of Consumer Marketing, Vol. 18 (2001), No. 7, pp. 569–573. 583 See Christensen: The Innovator's Dilemma, 1997, p. xxv. For a definition of disruptive technology see Christensen: The Innovator's Dilemma, 1997, p. xv.
171
D.I A Generic Model of Organizational Inertia and Change
inverse of the relative quality B, at the time when the focal organization lags behind the market. Inadequacy remains at zero for when the focal organization outperforms it. Perceived Inadequacy of Strategy per Stakeholder Favoring B perceived inadequacy
1 0.75 0.5 0.25 0 -1
-0.80 -0.60 -0.40 -0.20 0 0.20 "rel. quality B"
0.40
0.60
0.80
1
Figure D-3: Effect the relative quality B on the perceived inadequacy of strategy Just as in the NYSE-specific case, each stakeholder perceiving an inadequacy exerts stakeholder pressure for more B that adds up to total stakeholder pressure for more B. Compared to the NYSE case in which the individual reference pressure per stakeholder favoring B was set to one, it is here set to 0.6 in order to refrain from the special power that institutional clients had at the NYSE. In more general terms the pressure may also be interpreted as a stakeholder desire that the focal organization perceives or does not notice. The weighting of this total pressure with the organization’s attention to stakeholders favoring B results in the perceived pressure for more B which then leads to the focal organization’s change in strategy towards more orientation to strategy B. In the area of stakeholder theory, it is established knowledge that stakeholder pressure motivates organizations to implement new practices.584 This closes the balancing feedback loop Adaptation Pressure for B that is also shown in a simplified CLD in Figure D-4. DEVELOPMENT OF STRATEGY B
+ desired quality B
rel. quality B + (B) Adaptation Pressure for B Orientation to Strategy B
pressure for more B
+
Figure D-4: Adaptation pressure for strategy B (CLD)
584
See Eesley and Lenox: Firm responses to secondary stakeholder action, 2006, pp. 775–777; and Sarkis, Joseph, Pilar Gonzalez-Torre and Belarmino Adenso-Diaz: Stakeholder pressure and the adoption of environmental practices: The mediating effect of training, in: Journal of Operations Management, Vol. 28 (2010), No. 2, p. 164.
172
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
It can be expected that resistance from the stakeholder favoring the old strategy follows the implementation of the new strategy B.585 The resulting resistance pressure is composed similar to the non-institutional customers’ resistance pressure rather than the more complicated structure described for the NYSE floor firms. The mediation of resistance by cultural aspects is left out since it also represents a phenomenon that was special at the NYSE. The power of the stakeholders favoring A is expressed by their group size, e.g. by the number of customers desiring the old product. Additionally permanently powerful stakeholders may exist whose power is independent of their group size and thus constant over time. They may be floor firms, employees, and else. In the base scenario their number is kept at zero, but it can be varied. The respective stock and flow structure is depicted in Figure D-5.
<no of stakeholders favoring A>
(B) Resistance Pressure for A
(R)
desired quality A by stakeholders favoring A
TIME TO ADJUST DESIRED QUALTIY
+
stakeholder resistance pressure for more A + +
quality A +
REF. QUALITY A OF STRATEGY A
total stakeholder pressure for more A +
+ +
+
+ -
pcvd adequacy of quality A
PERMANENTLY POWERFUL STAKEHOLDERS FAVORING A
+ effect of quality A on resistance REF. RESISTANCE PRESSURE PER STAKEHOLDER FAVORING A
Figure D-5: Resistance pressure for strategy A (SFD) The orientation to strategy A involves a specific quality, such as market quality of stock exchanges, resolution quality in analog photography, or the computation capacity of mini-computers. This results in an absolute quality A that the focal organization offers. Quality A is presented as an absolute quality since the difference to the past quality, not to that of competitors is essential for the rise of resistance pressure. The perceived adequacy of quality A results from this comparison of the actual quality value with the floating goal of desired quality A by stakeholders favoring A. Stakeholder resistance pressure for more A rises when adequacy falls below its normal level of one, as Figure D-6 exhibits in more detail. The effect of quality on resistance is slightly inversely s-shaped to account for a convergence on a maximum value of one and a lower rise of resistance when adequacy is close to one.
585
See Oreg: Personality, context, and resistance to organizational change, 2006, p. 79; and Beer: Organization Change and Development, 1980, p. 103.
173
D.I A Generic Model of Organizational Inertia and Change
Effect of Quality on Resistance 1
effect
0.75 0.5 0.25 0 0
0.10
0.20
0.30 0.40 0.50 0.60 0.70 pcvd adequacy of quality A
0.80
0.90
1
Figure D-6: Effect of adequacy of quality A on resistance pressure Individual stakeholder resistance pressure multiplied with the sum of the number of stakeholders favoring A and the permanently powerful stakeholders favoring A results in the total stakeholder pressure for more A. The total stakeholder pressure for more A is thus computed as follows: total stakeholder pressure for more A [pressure unit] = stakeholder resistance pressure for more A [pressure unit / entity] D-2 • (no of stakeholders favoring A [entity] + PERMANENTLY POWERFUL STAKEHOLDERS FAVORING A [entity]) Since the total stakeholder pressure for more A feeds back to the management’s decision-making, the balancing feedback mechanism Resistance Pressure for A is closed. How this loop fits in with the structure described before is shown in Figure D-7. It represents the direct antipode to the adaptation loop, but only gets triggered once adaptation has started. DEVELOPMENT OF STRATEGY B
+ desired quality B
rel. quality B + (B) Adaptation Pressure for B
pressure for more B
(B) Resistance Pressure for A
pressure for more A -
Orientation to + Strategy B
quality A
Figure D-7: Resistance pressure for strategy A (CLD)
174
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
Concerning managerial decision-making, management is represented in an analogous way to the NYSE case with a repetitive momentum and a repetitive attention mechanism. The Repetitive Momentum Loop, shown on the right part of Figure D-8, influences the change rate by which the focal organization shifts its orientation between the accustomed strategy A and the new strategy B. The rate of change in strategy reduces inertia which—together with the confidence effect of performance— determines the organization’s actual openness to change. This openness then affects the yearly fractional change per perceived pressure and feeds back to the rate of change in strategy.586 The reference fractional change in strategy determines the general rigidity of this feedback loop and organization’s normal responsiveness to pressure that is independent of the current situation. REF. OPENNESS PER INERTIA
REF. FRACT. institutionalization INSTITUTIONALIZATION + + + (R) (B) limiting effect on institutionalization
Inertia (B)
REF. FRACT. INERTIA DECREASE
+
+ inertia decrease
institutionalization + (R) Inertia
+ openness to change (R) Repetitive Momentum
(B)
REF. FRACT. CHANGE IN STRATEGY PER PRESSURE P.A. + +
+ inertia decrease +
fract. change per pcvd pressure p.a.
effect of change on inertia Orientation to Strategy A
+ effect of openness on change
-
change in + strategy
Orientation to Strategy B
pcvd pressures
Figure D-8: Inertia and repetitive momentum (SFD) As depicted in the left part of Figure D-8, inertia itself grows by institutionalization, in congruence with writings of the organizational evolution and organization ecology literature. Researchers stated and also found evidence that, as time passes, institutionalization processes contribute to inertia. During times in which changes are not required, inertia and persistence consolidate.587 Mollona pointed out the distinction of resource-like inertia and cognitive inertia by two different accumulations.588 In the 586
A technically somewhat different, but conceptually similar mechanism can be found in Larsen and Lomi: Resetting the clock, 1999, pp. 412 and 419–420; and Sastry: Problems and Paradoxes in a Model of Punctuated Organizational Change, 1997, pp. 244 and 270–272. Sterman and Wittenberg also show that reinforcing feedback loops produce path-dependent behavior and lock-in. See Sterman, John D. and Jason Wittenberg: Path Dependence, Competition, and Succession in the Dynamics of Scientific Revolution, in: Organization Science, Vol. 10 (1999), No. 3, pp. 332–333. 587 See Hannan and Freeman: Structural Inertia and Organizational Change, 1984, pp. 152–155; Péli: Fit by Founding, Fit by Adaptation, 2009, pp. 344–345. For empirical evidence see Audia, Locke and Smith: The Paradox of Success, 2000, p. 849; and Starbuck and Milliken: Challenger, 1988, pp. 323–324, 329 and 331. 588 See Mollona, Edoardo: A Competence View of Firms as Resource Accumulation Systems: A Synthesis of Resource-Based and Evolutionary Models of Strategy-Making, in: Morecroft,
175
D.I A Generic Model of Organizational Inertia and Change
present stock and flow diagram, the accumulation of the orientation to strategy A or B shows similarity to the concept of resource-like inertia since the stocks are inert and inflexible in the way that they comprise the accumulation of their history. The variable inertia itself, which particularly symbolizes the institutionalized routines and inflexibility in the thinking of the focal organization’s management, has similarity to the concept of cognitive inertia. Inertia decreases by the replacement of old with new employees who bring new ideas to the organization as well as by the learning of new and unlearning of old patterns of thinking and behavior. This unlearning increases when change takes place.589 The s-shaped relationship between a change in strategy and its effect on the decrease of inertia is shown in Figure D-9. Its shape symbolizes a less than proportional disruption of routines and thinking when changes are incremental. Effect of Change on Inertia 8
effect
6 4 2 0 0
0.100
0.200 0.300 change in strategy
0.400
0.500
Figure D-9: Effect of change on the decrease of inertia The rate of change in strategy not only affects inertia, but is itself affected mainly by perceived pressure from stakeholders for more A (or B), illustrated in Figure D-10. Depending on which pressure is greater, the organization shifts to strategy A (or B). Additionally, and only in the case that the orientation to A or to B is already very high, a limiting effect influences the rate of change so that it looks as follows: change in strategy [dmnl / year] = (pcvd pressure from stakeholders favoring B [pressure unit] • effect of B on change [dmnl] - pcvd pressure from stakeholders favoring A B [pressure unit • effect of A on change [dmnl]) • fract. change per pcvd pressure p.a. [dmnl / pressure unit / year]
D-3
John, Ron Sanchez and Aimé Heene (Ed.): Systems Perspectives on Resources, Capabilities, and Management Processes, Amsterdam [et al.] 2002, p. 111. For a distinction of resource and routine rigidity see Gilbert, Clark G.: Unbundling the Structure of Inertia: Resource Versus Routine Rigidity, in: Academy of Management Journal, Vol. 48 (2005), No. 5, p. 742. 589 See Nadler and Tushman: Types of Organizational Change, 1995, p. 23; and Wollin: Punctuated equilibrium, 1999, p. 362.
176
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
The limiting effects work when stakeholders still exert pressure for a strategy that is almost fully implemented. The management team becomes hesitant in reacting to the full pressure. Orientation to Strategy A effect of A on change
change in + strategy+ + -
Orientation to Strategy B
INI ORIENTATION TO STRATEGY B
effect of B on change
Figure D-10: Limitations to changes of strategy (SFD) Phenomena such as the concentration on floor firms do not represent an unparalleled example. The Polaroid management misjudged the rising importance of a new group of customers and their preference for digital photography. DEC, for instance, was a highly client-oriented organization. The company even maintained the Digital Equipment Corporation Users Society in order to provide for the possibility of mutual exchange, feedback, and learning. At the same time, exactly this strong relationship to loyal customers and its culture made the DEC management inattentive to the growth of a new group of customers that favored the PC. DEC appeared to be customer oriented, but concentrated only on one customer group.590 The organization’s cultural inertia was responsible for the lacking attention to an altered environment with different stakeholders. Magness supports in an empirical analysis that “stakeholder status is impermanent, and determined through the eyes of the decisionmaker.“591 Therefore, it is also necessary to include the concept of attention to stakeholders into the generic model of organizational inertia and change.
590 591
See Schein: DEC is Dead, Long Live DEC, 2003, pp. 74 and 252. Magness: Who are the Stakeholders Now, 2008, p. 177, abstract.
177
D.I A Generic Model of Organizational Inertia and Change
REF. OPENNESS PER INERTIA + institutionalization + (R)
openness to change
+ effect of openness on change
-
REF. FRACT. CHANGE IN ATTENTION P.A. + fract. change in attention per + pressure p.a.
Inertia (B) + inertia decrease +
Attention to Stakeholders Favoring A
effect of change on inertia Orientation to Strategy A
change in attention
Attention to Stakeholders Favoring B
change in strategy +
Orientation to Strategy B
(R) Repetitive Attention + + pcvd pressure from stakeholders favoring A
+ + pcvd pressure from stakeholders favoring B
Figure D-11: Attention to stakeholders (SFD) Via the openness to change, inertia also affects the attention to stakeholders. Figure D-11 demonstrates that openness increases the annual fractional change in attention per pressure and thus allows for a faster reaction of attention to pressure from stakeholders. The organization also has a general responsiveness of attention, the yearly reference fractional change in attention. It represents its normal openness to new trends and stakeholder groups, independent of the situation. How the adaptation to pressure takes place is further detailed in Figure D-13 and will be described in the next paragraph. Back to Figure D-11, a modified accumulated attention to stakeholders also weights the types of pressure and results in an altered perceived pressure from stakeholder favoring A (or B) for more A (or B). Since the perceived pressure returns to the change in strategy and to inertia by means of feedback, the reinforcing Repetitive Attention Loop is closed. The weighting relationship of attention has generic value because it not only could be found at the NYSE, but González-Benito and González-Benito found a similar relationship not for stakeholder attention, but managerial environmental awareness. The latter increases the perceived pressure for environmental issues and the implementation of environmental practices.592 The full Repetitive Attention Mechanism as well as the Repetitive Momentum Loop shown in the CLD of Figure D-12 exhibit reinforcing behavior.
592
See González-Benito, Javier and Óscar González-Benito: The role of stakeholder pressure and managerial values in the implementation of environmental logistics practices, in: International Journal of Production Research, Vol. 44 (2006), No. 7, p. 1368.
178
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change + desired quality B
DEVELOPMENT OF STRATEGY B
Attention to Stakeholder B openness to change -
(R) Repetitive Attention
(R) Repetitive Momentum
Inertia
rel. quality B + (B) Adaptation Pressure for B
pressure for more B
(B) Resistance Pressure for A
pressure for more A -
+ Orientation to + Strategy B
quality A
Figure D-12: Repetitive momentum in the generic model (CLD) While in particular the freeze of attention is a result of high inertia, attention also shows adaptive behavior. The adaptive mechanism is displayed in Figure D-13. The change in attention is positive and attention to stakeholders favoring B rises when the perceived pressure from stakeholders favoring B is higher than that from stakeholders favoring A. The balancing effect of attention to B (A) limits a further orientation to stakeholders favoring B (A) when attention to these stakeholders is already very high. While certain stakeholders may continue to exert pressure, the management team would not be willing any more to fully react to these forces and to further change its attention. The rate of change in attention is thus computed as follows: change in attention [dmnl / year] = (pcvd pressure from stakeholders favoring A [pressure unit] • effect of attention to A on change [dmnl] D-4 – pcvd pressure from stakeholders favoring B [pressure unit] • effect of attention to B on change [dmnl]) • fract. change in attention per pressure p.a. [dmnl / pressure unit / year]
Attention to Stakeholders Favoring A
Attention to Stakeholders change in Favoring B + attention + (B) (B) - + effect of attention effect of attention to A on change to B on change +
(R)
pcvd pressure from stakeholders favoring A +
(R)
+
pcvd pressure from stakeholders favoring B +
Figure D-13: Limitations to changes in attention (SFD)
179
D.I A Generic Model of Organizational Inertia and Change
As Figure D-13 explains structurally, the perception of pressure from stakeholders itself is biased by the current distribution of attention, leading to the following computation of the perceived pressure, here exemplified for the perceived pressure from stakeholders favoring B: pcvd pressure from stakeholders favoring B [pressure unit] = total stakeholder pressure for more B [pressure unit] • Attention to Stakeholders Favoring B [dmnl]
D-5
Attention serves as a weighting factor for the incoming forces from stakeholder pressure. This structure creates two reinforcing mechanisms which bias the perception of forces towards those stakeholders the management team listens to. Nevertheless, while attention may be biased, it also adapts to existing total stakeholder pressure for more A or B. Once those who demand change receive more attention, this triggers change. Then the interests of the stakeholders demanding change are better met so that they do not need to exert as much pressure any more. These feedback mechanisms of the adaptation of attention are detailed in the causal loops of Figure D-14. pressure for more B
rel. quality B + (B) Adaptation of Attention to B
+ Attention to Stakeholder B -
+
Orientation to Strategy B
(B) Adaptation of Attention to A -
pressure for more A -
quality A
Figure D-14: Adaptation of Attention (CLD) A further important feedback relationship can be found between managerial decision-making and performance. In different organizations, performance may represent varying concepts such as market share, the sales level, or the size of the customer base. Therefore, its representation is kept general and plain to match all of these interpretations. It is assumed that—unlike in the NYSE case—past performance does not reinforce the current value.593 The idea that different effects adjust performance upwards and downwards are also known from works by Salge and Sterman, for example, in which cases market share is affected by effects of quality and attractive-
593
For information on the simplification of decision rules and the aggregation of stock and flow structure see Saysel and Barlas: Model simplification and validation with indirect structure validity tests, 2006, p. 257.
180
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
ness.594 A similar computation of performance, by which a normal or reference performance is adjusted by a multiplier effect and real performance then adapts to this indicated value, has also been proposed by Milling.595 Figure D-15 indicates that quality A and the relative quality B that come with the pursuit of strategy A or B both adjust performance. Which of these effects prevails depends on the weight that customers attribute to the qualities, e.g. the weight on speed vs. price or on digital editing vs. resolution quality in photography. The weight on quality B vs. quality A directly emanates from the distribution of stakeholder preferences (i.e. the fraction of stakeholders favoring B). The performance adjustment amends the reference performance, set here to 0.5 performance units. Since information about an organization’s offerings needs to diffuse in the market and customers may show some loyalty, performance adapts to its indicated value with a time delay of one year.
effect of quality A on performance
+
+
+
(B) Performance Adaptation
+ performance adjustment
wt. on quality B vs. quality A
+
effect of rel. quality B on performance
(R) Performance Decline
indicated + performance + REF. PERFORMANCE
change in -(B) + performance
Performance
TIME FOR CHANGING PERFORMANCE
Figure D-15: Performance (SFD) In the NYSE case, the descent of market share helped the introduction of a new strategy, the Hybrid Market.596 In a more general sense, this idea conforms to the concept of aspiration levels and failure-induced change of the behavioral theory of organizations. If performance falls below the aspiration level, the organization is more likely to search for a solution and undergo change.597 Figure D-16 clarifies this process by a detailed SFD. If performance is below the aspiration level of desired performance, it is perceived as inadequate and decision-makers lose confidence in the current strategy. This confidence effect of performance is weak for minor inadequacies as they may reflect normal variations of performance not related to the organization’s strategy. For greater perceived shortcomings the effect quickly aggravates and increases the organization’s openness to change. The relationships and their 594
See Salge: Struktur und Dynamik ganzheitlicher Verbesserungsprogramme in der industriellen Fertigung, 2009, p. 51; and Sterman: Business Dynamics, 2000, p. 393. 595 See Milling: Der technische Fortschritt beim Produktionsprozeß, 1974, pp. 195–197. 596 See Storkenmaier and Riordan: The Effect of Automated Trading on Market Quality, 2009, p. 11. 597 See Cyert and March: A Behavioral Theory of the Firm, 1963, p. 121; and March and Simon: Organizations, 1958, pp. 173–174 and 184. For empirical support see e.g. Greve, Henrich R.: Performance, Aspirations, and Risky Organizational Change, in: Administrative Science Quarterly, Vol. 43 (1998), No. 1, pp. 74–75.
181
D.I A Generic Model of Organizational Inertia and Change
strength were already specified for the NYSE and described on page 136. These causal relationships also comply with the view of Lant and Mezias who maintain that the impetus for change and for the adaptation to the environment is triggered by a gap between current and desired performance.598 They also found empirical evidence that historical performance provides the most robust description of aspiration levels.599 +
+
pcvd adequacy of performance -
+
confidence effect of performance
+ openness to change -
desired performance
TIME TO ADJUST DESD PERFORMANCE
REF. OPENNESS PER INERTIA
Figure D-16: Relationship between performance and change (SFD) In the view of Forrester and Senge, a model of the loss and gain of market share should include the effect of different companies’ contrasting policies on market share.600 This has been achieved by the linking of qualities A and B to performance and further linking the latter to the openness to change. The full resulting feedback cycles involving performance and the orientation to a strategy are shown by bold arrows in Figure D-17. Low performance increases the openness to change, and—in the case of a pressure imbalance in favor of strategy B—the organization orients towards strategy B, increases its relative quality B, and performance increases in an adaptive manner. This balancing mechanism is called Performance Adaptation. It has a reinforcing side effect since the reorientation further reduces quality A and diminishes performance. As in the NYSE case, this Performance Decline Loop makes the organization reorient to the alternative direction even more quickly.
598
See Lant, Theresa K. and Stephen J. Mezias: An Organizational Learning Model of Convergence and Reorientation, in: Organization Science, Vol. 3 (1992), No. 1, p. 48. 599 See Lant: Aspiration Level Adaptation, 1992, pp. 641–642. 600 See Forrester and Senge: Tests for Building Confidence in System Dynamics Models, 1980, p. 221.
182
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
DEVELOPMENT OF STRATEGY B
+ desired quality B (B) Performance Adaptation Attention to Stakeholder B
+ Performance +
-
openness to change Inertia
(R) Repetitive Attention
(R) Repetitive Momentum
rel. quality B + (B) Adaptation Pressure for B
pressure for more B
(B) Resistance Pressure for A
pressure for more A -
+ Orientation to + Strategy B
(R) Performance Decline
quality A
Figure D-17: Performance (CLD) The structure of the generic system dynamics model has now fully been specified. It includes the environment as an external driver of change. Endogenously it incorporates stakeholders in the close environment of the focal organization and managerial decision-making as it relates to cognitive elements. The model’s explanatory power will be analyzed in the next chapters.
D.II Structural-Behavioral Analysis and Causal Theory D.II.1
Validation of the Generic Model
Validation of a generic structure is more difficult than gaining confidence in a model that maps a specific example. Nevertheless it is possible.601 In particular as regards to generic system dynamics models it is useful to distinguish two different kinds of validation: internal and external, both of which will be addressed. Internal validity exists if the model is consistent and sound.602 It can be tested just as in a case-specific model. This has initially been done by tests of model structure. Very fundamentally, the generic model is dimensionally consistent.603 Additionally, it is possible to have confidence in the structure because the behavior it produces is insensitive to changes in the choice of integration frequency (time step) and method.604 Many aspects of structural validity were already described together with the model structure. It has 601
See Größler: Struktur und Dynamik strategischer Fähigkeiten in der Produktion, 2007, p. 152; Lane, David C.: Can We Have Confidence in Generic Structures?, in: The Journal of the Operational Research Society, Vol. 49 (1998), No. 9, p. 942; and Lane and Smart: Reinterpreting 'generic structure', 1996, p. 113. 602 See Größler: Struktur und Dynamik strategischer Fähigkeiten in der Produktion, 2007, p. 146. 603 See Forrester and Senge: Tests for Building Confidence in System Dynamics Models, 1980, pp. 215–216; and Größler: Modelltests, 2008, p. 262. 604 See Sterman: Business Dynamics, 2000, p. 872.
D.II Structural-Behavioral Analysis and Causal Theory
183
been laid out how model variables and in particular parameters correspond to reality. For instance, the importance of attention has been established in relation to several examples, and there is theoretical and empirical evidence for an effect of performance on decisions of the management team. Parameters shaping inertia were compared to institutionalization and employee turnover, and the reference change in attention might represent the intensiveness of the search for new trends and stakeholder groups, e.g. by means of consultation of market search institutions. Parameters in this case cannot be compared to exact data. Since the model includes many soft variables and parameters, an exact quantification is not possible and it is not necessary. In the view of Richardson, the degree of accuracy is always judged against model purpose.605 Forrester mentions that for many purposes it is sufficient to estimate parameters within the plausible range because it will not affect results significantly.606 For this reason, different numerical parameters values are rather understood as qualitative values such as low, rather low, medium, or high levels of e.g. reference fractional changes in inertia or attention. Extreme conditions tests were useful to validate the causal structure because they uncover whether the model produces results that are inconsistent with e.g. physical laws. In this way, they are a means to uncover inconsistencies in assumptions made. A sensitivity analysis including a broad number of extreme and simultaneous parameter changes also produces sensible outcomes, as shown in Figure D-18. It assumes the environment to proceed as before and investigates how organizations with different managerial setups that may also face varied degrees of resistance pressure may react to this environmental development. It results in orientations to strategy B ranging between 20 and 100 percent, rather high sensitivity in performance, as well as high variation in inertia and attention—all of them within reasonable bounds. The analysis serves as an extreme test of the system dynamics model, but it also demonstrates that many different behaviors ranging from the failure to change to smooth adaptation are within the realms of possibility. According to Lane as well as Milling, the aim is to have a homomorphous mapping of encountered phenomena.607 The quality of this mapping can and will be analyzed in particular with the family member test suggested by Forrester and Senge. This test is part of the behavioral validation of a generic model. It checks the applicability of the generic model to a class of phenomena, and it tests whether a modification of parameter values is able to generate behavior appropriate for different organizations within a class.608 The validation procedure helps to clarify whether different modes of 605
See Richardson and Pugh III: Introduction to System Dynamics Modeling with DYNAMO, 1981, p. 230. 606 See Forrester: Industrial Dynamics, 1961, p. 171. 607 See Lane: Can We Have Confidence in Generic Structures, 1998, p. 942; and Milling: Der technische Fortschritt beim Produktionsprozeß, 1974, p. 212. 608 See Forrester and Senge: Tests for Building Confidence in System Dynamics Models, 1980, p. 220; Lane: Can We Have Confidence in Generic Structures, 1998, p. 942; and Lane and Smart: Reinterpreting 'generic structure', 1996, p. 110.
184
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
behavior known to occur in the class of systems that the generic model stands for can be reproduced. In this way, it also establishes external validity of the model. generic sens all 50% 75%
95%
100%
Orientation to Strategy B 1 0.75 0.5 0.25 0
0
10
generic sens all 50% 75% Performance 0.6
20
30
40
50
30
40
50
Time (Year)
95%
100%
0.45 0.3 0.15 0
0
10
20 Time (Year)
changed parameters base run value ini inertia 0.9 ref. fract. institutionalization 0.3 ref. fract. change in strategy per pressure p.a. 0.02 ref. fract. change in attention p.a. 0.05 ini attention to stakeholders favoring B 0.1 permanently powerful stakeholders favoring A 0 ref. resistance pressure per stakeholder favoring A 1 time to adjust desired quality 5
parameter range 0.2 – 0.9 0.15 – 0.45 0.002 – 0.2 0.002 – 0.2 0.2 – 0.3 0 – 100 0.2 – 5 1 – 20
Figure D-18: Sensitivity for changes in stakeholder and management parameters
D.II Structural-Behavioral Analysis and Causal Theory
185
In close relation to the family member test, behavioral correspondence can be investigated by the behavioral anomalies and the surprise behavior test.609 Behavioral anomalies and surprise behavior can be used on a continuous basis in the modeling process to detect flaws in the model’s assumptions.610 In the present study, tracing back the reasons for the behavior has helped to decide whether the anomalies required a modification of assumptions or conveyed surprise behavior that advances the understanding of the system.611 System understanding is also enhanced by the testing of different policies and the sensitivity of recommendations to parameter changes. This way it becomes evident which policies lead to a system improvement. Additionally, the further analysis of policy sensitivity provides information on the robustness of strategy and policy recommendations.612 The generic model structure includes elements of the environment, inertia, cognition as well as stakeholder reactions. With the structure laid out, the next step will be the investigation into the generic model’s behavior. The tests mentioned above that have not yet been presented will be elaborated. For this behavioral analysis, a simulation period of 50 years was chosen. While this time horizon may seem long at first glance, it roughly equals two to three times the implementation time of important innovations and of customer demand changes. The rising demand for corporate ethical conduct and for digital cameras constitutes an example.
D.II.2
Effects of Reinforcing and Balancing Feedback on the Occurrence of Change
In the base run scenario, the generic model exhibits a radically changing behavior similar to the NYSE case. While a smooth s-shaped adaptation takes place in the remaining market, expressed by the grey line 1 in Figure D-19, the target organization (line 2) initially does not react, but then shows a much steeper s-shaped growth than the remaining market.
609
See Lane: Can We Have Confidence in Generic Structures, 1998, p. 942. See Forrester and Senge: Tests for Building Confidence in System Dynamics Models, 1980, p. 220. 611 See ibid., pp. 220–221. 612 See Richardson and Pugh III: Introduction to System Dynamics Modeling with DYNAMO, 1981, pp. 349–352. 610
186
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
Orientation to Strategy B in Focal Organization and Market 1 0.75 Dmnl
12
1
1 2
1 2
12
35
40
45
1 2
1 2
0.5 1 0.25 1 0
1 2 0
1 2 5
12 10
2 15
2 20
2 25 30 Time (Year)
diffusion of B in remaining market : generic base run Orientation to Strategy B : generic base run 2
1 2
1 2
50
1 2
2
Figure D-19: Generic base run (BOT) The observed behavior can be explained by relating the behavior of selected variables shown in Figure D-20 to the full causal structure. The upper part shows an excerpt of model behavior in period 15 to 35, broken down into different phases during which loop activity differs. The four variables presented in the behavior over time (BOT) graph are also highlighted in a CLD further below in the figure. In phase I, hardly any change in the organization’s strategy takes place (line 3). The Repetitive Momentum and Repetitive Attention Loop reinforce the current orientation to strategy A. As a consequence, pressure for change is not fully perceived and what is perceived is just marginally implemented. When performance declines (line 1) the organization becomes somewhat more open to change and enters into phase II, initiated by the balancing Performance Adaptation mechanism. Attention adapts more quickly than the organizational strategy, and the small performance effect additionally helps the organization to become more attentive to the stakeholders who exert pressure. The Repetitive Attention Loop begins to soften. The stakeholders favoring B attract more of the management team’s attention, and attention adapts to their pressure. The shift in attention is more important for triggering change than performance itself because it makes the management team aware of what it has previously neglected. This change in stakeholder attention is sufficient for the change towards strategy B to take off (line 3) in phase III, triggered by the balancing effect Adaptation Pressure for B. Change reduces inertia (line 4), and it enables the Repetitive Momentum Loop (and Repetitive Attention Loop) to turn its repetitive character towards the direction of more change. In phase IV, the repetitive momentum still allows for alteration, but the balancing forces of the loop Adaptation Pressure for B become weaker as the organization increasingly orients towards strategy B. In phase V, adaptation has basically been completed and consolidation becomes important again. Institutionalization processes increase inertia, and the Repetitive Momentum and Attention Loop shift again towards stability and rigidity.
187
D.II Structural-Behavioral Analysis and Causal Theory
Model Behavior I II 1 performance unit 1 Dmnl 1 consistency unit
4
4
III
4
2
4 1
0 performance unit 0 Dmnl 0 consistency unit
1
3
1
1
15
2
2 3
3 20
3
2
V
2 3
3
1
1
1
2 2
IV
1 4
4
4
3 25 Time (Year)
30
35
Performance : generic base run performance unit 1 1 1 Attention to Stakeholders Favoring B : generic base run 2 Dmnl 2 3 3 3 Orientation to Strategy B : generic base run Dmnl Inertia : generic base run 4 4 4 4 4 consistency unit
Phase Loop Activity I Repetitive Momentum and Repetitive Attention reinforce current strategy II Performance Adaptation active, softening of Repetitive Attention III Adaptation Pressure for B Loop active, reversal of repetitive character of Repetitive Momentum and Attention Loop IV Adaptation Pressure for B Loop starts to weaken, Repetitive Momentum still oriented to change V Consodidation and growth of inertia, Repetitive Momentum and Attention Loop move towards stability DEVELOPMENT OF STRATEGY B
+ desired quality B (B) Performance Adaptation Attention to Stakeholder B
+ Performance +
-
openness to change Inertia
(R) Repetitive Attention
(R) Repetitive Momentum
(R) Performance Decline
rel. quality B + (B) Adaptation Pressure for B
pressure for more B
(B) Resistance Pressure for A
pressure for more A -
+ Orientation + to Strategy B -
quality A
Figure D-20: Phases of loop dominance (BOT and CLD) Since—at least in the scenario presented here—in the end all stakeholders favor strategy B, the balancing Resistance Pressure and the reinforcing Performance De-
188
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
cline mechanisms are of minor importance. When the organization finally reacts, stakeholder preferences have almost completely shifted to strategy B so that there is only minor resistance from the few favoring A. The structural-behavioral relationships discussed above have a direct connection to the research question of whether prior change increases or reduces the occurrence of subsequent change. This question has mainly been analyzed by statistical means, whereas this dissertation will give a causal explanation. While a significant number of studies have found that prior change increases the occurrence of subsequent transformations, the results were not unambiguous and a proportion of studies found contrary or inconclusive evidence.613 Relating the two different behavioral patterns exhibited in Figure D-21 below to the causal structure of Figure D-20 above, will help explain in what cases change increases the occurrence of subsequent transformations. Line 2 in the upper part of Figure D-21 reveals the behavior of an organization which—in comparison with the base run organization—is more aware of the existence and desires of a new customer group, less inert and more reactive to perceived pressure. After only a short period during which the Repetitive Momentum and Attention Loops reinforce the old strategy, around period 20 the management team radically changes its strategy and adapts to the situation in the market. This ‘earlier radical adopter’ might represent companies such as Canon or Nikon that introduced digital cameras already around the year 1990, or it might be one of the stock exchanges that implemented a significant portion of electronic trading in the late 1980s. For example, the London Stock Exchange was inert during the 1970s, but radically adapted in 1986 despite the opposition of the disadvantaged.614 As in the base run, the radical adaptation reduces inertia (line 2 in the lower part of Figure D-21) and the Repetitive Momentum Loop switches to the direction of further change. The repetitive loops allow the organization to become more malleable; its openness to change increases. Lower inertia entails a greater willingness to react to pressure and to adapt its attention towards those who demand a transformation. This means the organization’s ability to react to pressure and thus its ability to change increases. Change propagates due to the reduction of inertia. This is the same for both organizations, the inert one and the earlier radical adopter.
613
Among others, the following studies support the hypothesis that prior change increases the occurrence of subsequent changes: Amburgey, Kelly and Barnett: Resetting the Clock, 1993, for change of the same type, pp. 66 and 69; Collins, et al.: Learning by doing, 2009, p. 1333; Dobrev, Kim and Carroll: Shifting Gears, Shifting Niches, 2003, p. 274; Kelly and Amburgey: Organizational Inertia and Momentum, 1991, for change of the same type, negative relationship for change of different type, p. 606. Inverse correlations were found in Beck, Brüderl and Woywode: Momentum or Deceleration, 2008, p. 426; Beck and Kieser: The Complexity of Rule Systems, Experience and Organizational Learning, 2003, p. 807; and Wischnevsky and Damanpour: Radical strategic and structural change, 2008, p. 65. 614 See Michie: The London Stock Exchange, 1999, pp. 593–595.
189
D.II Structural-Behavioral Analysis and Causal Theory
When the earlier adopter (line 2 of the upper part of Figure D-21) closes the gap to its environment around period 22, the environment still changes (line 1) and requires further adaptation from the focal organization. The repetitive mechanisms are flexible, and the adaptation mechanisms pull the organization towards a greater implementation of strategy B. In this scenario, the initial change leads to subsequent changes.
Orientation to Strategy B in Focal Organization and Market 1
3
0.75 Dmnl
1 2 3 12 3 1 2 31 2 3 12
1 2 12
0.5 1 0.25 1 0
1 23 1 2 3 12 3 0 5 10
2
23 15
3 20
3 25 30 Time (Year)
35
40
45
50
diffusion of B in remaining market : generic early radical adopter 1 1 2 2 2 Orientation to Strategy B : generic early radical adopter Orientation to Strategy B : generic base run 3 3 3 3 3
Inertia
consistency unit
1
1
1 2
0.75
1 2
1 2
1 2
1
1
1
2
1 2
1
0.5
2 1
0.25
2
0 0
5
10
15
20
Inertia : generic base run 1 1 Inertia : generic early radical adopter
1
2
2
2
25 30 Time (Year)
1
1 2
35
1 2
2
40
1 2
45
1 2
changed parameters early radical adopter base run value ref. fract. change in strategy per pressure p.a. 0.02 ref. fract. change in attention p.a. 0.05 ref. fract. inertia decrease 0.15 ini inertia 0.9 ini attention to stakeholders favoring B 0.1
50
1 2
2
parameter value 0.05 0.1 0.21 0.78 0.3
Figure D-21: Comparison of early and late radical adaptation (BOT)
190
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
However, the situation is different in the base run scenario (line 3). Here, the adaptation takes place later, and once the focal organization reaches its environment, the latter does not change any further. There may be other unrelated changes in the focal organization’s environment, but the shift from strategy A to strategy B has fully taken place. In this case the initial change does not lead to any subsequent transformations. The reason is that the balancing Adaptation Pressure for B Loop by which stakeholders exert pressure for more B is inactive. Thus, in the base run case, the transformation process comprises one adaptive shift that decreases the occurrence of further changes. While both transformation reduce inertia and increase the malleability of the focal organization, the occurrence of subsequent alterations also depends on the further development of the organizational environment. The importance of the relation between the focal organization’s ability to change and the pressure to adapt also becomes obvious when two environmental changes are simulated. Figure D-23 displays a modified version of the base run scenario. It overlaps the base run until period 30, but then a second shift takes place, triggered by stakeholders favoring strategy B, to which the remaining market adapts with an average time delay of five years (line 1). For reasons of feasibility, this has been modeled as a move back to strategy A instead of a shift towards strategy C. The stakeholders originally favoring B now pressure for the diminution of strategy B (meaning for less B). Technically, this is achieved by making the perceived inadequacy of the strategy the inverse of the relative quality B. Figure D-22 displays the relationship which is established hereby, by which not only an underachievement in the relative quality B (below 0) leads to a perceived inadequacy, but also an overachievement gets sanctioned. Stakeholders favoring B now not only exert pressure for more B, but also for less B when the focal organization’ relative quality B reaches values above market average. Since the market transformation takes both directions, it needs to be assumed that both over- and underachievement in quality B leads to some kind of pressure for change. Perceived Inadequacy of Strategy per Stakeholder Favoring B perceived inadequacy
1 0.5 0 -0.5 -1 -1
-0.80 -0.60 -0.40
-0.20 0 0.20 "rel. quality B"
0.40
0.60
0.80
1
Figure D-22: Relationship between difference in quality B and strategy inadequacy
191
D.II Structural-Behavioral Analysis and Causal Theory
Line 2 of Figure D-23 exhibits the known delayed but radical adaptation process of the focal organization that is upheld by the working of the reinforcing character of the Repetitive Momentum and Attention Loops which then also enforce how radical the behavior is. The reaction to the second market move is very different, as the organization now quite quickly adapts to the new direction of the market and implements what is implemented in the market with a time delay of only little more than one year. The behavior in the scenario including two environmental changes is different from the base run in which the organization remained with strategy B. The reason is that, while the repetitive loops are flexible, the Adaptation Pressure for B loop, by which the organization adapts to the market, gains in importance again.
Orientation to Strategy B in Focal Organization and Market 1 0.75 Dmnl
12
1 1 2
0.5
1
1
2
0.25 1 0
1 2 0
1 2 5
12 10
2 15
2 20
12
2 25 30 Time (Year)
35
diffusion of B in remaining market : generic two envl changes Orientation to Strategy B : generic two envl changes 2 2
1 2 45
40 1
1 2 50
1 2
2
Figure D-23: Two environmental changes (BOT)615 A second reason why the adaptation to the second market shift takes place more quickly is that the management team has already focused its attention on those stakeholders who matter and who demand the second change. The stakeholders originally favoring B now pressure for the diminution of strategy B. As a consequence, attention remains with stakeholders who originally favored B. The quick adaptation is thus possible because, first, inertia is lower than in the outset and the Repetitive Momentum Loop more flexible, and second, because the management team already focuses its attention on those stakeholders who demand the subsequent change. The hypothetical twofold change scenario is a clear example for the effects of previous change on the occurrence of further alterations. Since initial change enhances the organization’s ability to change, the existence of a continuously strong adaptive mechanism increases the occurrence of subsequent change. But the base
615
The development of strategy B is shaped differently after period 30 and includes a hypothetical second environmental change back to strategy A. The number of stakeholders favoring B is kept at 100 after period 30.
192
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
run scenario reveals in comparison that without any further activity of adaptive mechanisms, future change fails to appear. Hence, there is no single answer to the question whether prior change increases or reduces the occurrence of subsequent change. While it is clear that it increases the malleability of the organization and its ability to change, the strength of the adaptive mechanism, meaning the standing of the organization in relation to its environment, is also important. A one-time environmental shift leads to a new equilibrium, but a dynamic and competitive environment that has not yet finished changing or that frequently transforms may trigger subsequent change in the focal organization. Behavior is shaped by the interrelationship of the management setup, endogenous forces, and environmental requirements. The consideration of how balancing and reinforcing feedback mechanisms are intertwined is necessary and important.
D.III Possibilities of Managerial Intervention for Driving Change In the latter scenario, reduced inertia and the focus of attention proved to be important for the quicker adaptation to the second environmental change. Attention and inertia will therefore be analyzed as possible leverage points for intervention. It will be investigated how the management team’s influence on inertia and attention can act as a driver of change and shape the evolution of an organization and its alignment with the environment. This view concurs with Bowen’s position indicating that while affected by the pressure that arises from the system’s structure, decision-makers still have the ability to either follow these types of pressure or make an autonomous decision.616
D.III.1 Inertia and the Ambiguous Effects of the Responsiveness to Pressure The effect that a diminished level of inertia has on the behavior of the organization will be considered. At the NYSE, one interviewee reported the reason for high inertia within the organization to be rooted in the inward-orientation of recruiting. People were grown from within, and there was very little turnover with people from outside the NYSE. This was different with Polaroid; 90 percent of the employees initially involved in the development of digital photography were new to the company. They developed a sound product, but its commercialization was thwarted by the management team’s grown convictions and beliefs in an old business model that did not fit digital photography. In the management area, no turnover had taken place. Once the management started to change, new people from outside were less entangled and 616
See Bowen: System dynamics, determinism, and choice, 1994, pp. 87–88 and 90. See also Lane: Should System Dynamics be Described as a 'Hard' or 'Deterministic' Systems Approach, 2000, p. 10.
D.III Possibilities of Managerial Intervention for Driving Change
193
locked in the old business model and embraced market developments with greater openness.617 The replacement of culturally and ideologically aligned employees and managers with more open ones could have been a viable solution for the NYSE and Polaroid. Therefore the effect of different degrees of the decrease in inertia on the model behavior will be simulated to test flexible turnover and unlearning. For example, a consequence of the London Stock Exchange’ move to e-trade was a permanent transformation of its membership. The stock exchange re-emerged as a more dynamic institution.618 Additionally, the strength of institutionalization processes in organizations deserves attention. Even at the relatively young digital photography company Linco, the opportunity for entering the USB flash drives business bypassed unnoticed. The company’s quickly arising insistence on its identity as a photo memory or digital film producer refrained it from exploiting all options that opened up, e.g. in the area of MP3 players or flash memory.619 Similar effects of a quick socialization of new employees have been reported for the Intel Corporation.620 Quick institutionalization can thus also cause high inertia and lock-in. It may result in a missing adaptation to the market and to the opportunities it offers. In a similar vein, DEC’s inertia did not result from missing employee turnover. While the rate at which employees left the company may have been small, the organization grew rapidly and had a strong inflow of new employees from outside. But the culture of technological overconfidence and missing market orientation was enforced by the leadership style and development programs which decreased the organization’s ability to react to changed environmental circumstances.621 At DEC the institutionalization process worked particularly well. The effects of an especially high institutionalization as well as of a low turnover rate on an organization’s strategy are shown in Figure D-24. They lead to a very slow adaptation to the market and have long-term side effects on performance that are detrimental to any company having fixed costs.
617
See Tripsas and Gavetti: Capabilities, Cognition, and Inertia, 2000, pp. 1152–1157. See Michie: The London Stock Exchange, 1999, pp. 633 and 441. 619 See Tripsas: Technology, Identity, and Inertia through the Lens of 'The Digital Photography Company', 2009, pp. 450–451. 620 Directly see Burgelman: Strategy as Vector and the Inertia of Coevolutionary Lock-in, 2002, p. 354. 621 See Gibbons, Tracy C.: DEC's "Other" Legacy: The Development of Leaders, in: Schein, Edgar H. (Ed.): DEC is Dead, Long Live DEC: The Lasting Legacy of Digital Equipment Corporation, San Francisco, CA 2003, pp. 97–102; and Schein: DEC is Dead, Long Live DEC, 2003, pp. 80–89. 618
194
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
Orientation to Strategy B 1
Dmnl
0.75 0.5
1
0.25
2
1
2
1
1 2
12
2
2
1 0
1 2 0
1 2 5
12 10
1 2 1 2 12 15 20 25 30 Time (Year)
35
40
Orientation to Strategy B : generic less inertia decrease 1 Orientation to Strategy B : generic more institutionalization 2
changed parameters 1: ref. fract. inertia decrease 2: ref. fract. institutionalization
base run value 0.15 0.3
45
1
50
1 2
2
parameter value 0.1 0.45
Figure D-24: Effects of high inertia (BOT) Since the processes described above resulted in high inertia and missed opportunities for change, an analysis of the effects of low inertia is promising for testing the management team’s different possibilities of intervention. For this reason, the sensitivity for parameter changes of the initial level of inertia (ini inertia), the reference fractional institutionalization and the reference fractional inertia decrease is tested. The scenarios show the management team’s ability to intervene and to create an organization that is more flexible and adaptive to change. The upper part of Figure D-25 illustrates the area of parameter changes by a grey box and the variable which it may affect by a black box. The lower part of the figure displays the results of the sensitivity runs in comparison with the diffusion of B in the remaining market. The simulation runs reveal that reduced institutionalization as well as a higher rate of inertia decrease are able to trigger an earlier and often also radical orientation to strategy B. It proves beneficial to bring in new people with fresh ideas and a greater openness. The reduction of strong institutionalization processes, i.e. in the form of special trainings or the creation of an open-minded culture different from, for example, the engineering culture present at DEC also turns out to be valuable. They represent ways in which the management team can make the organization more malleable and drive change. Sensing opportunities and threats in shifting markets is a necessary requirement for an organization’s ability to adapt to a changed market environment.622 An open-minded management that exhibits low inertia is a necessary requirement for 622
See O'Reilly III, Charles A. and Michael L. Tushman: Ambidexterity as a dynamic capability: Resolving the innovator's dilemma, in: Research in Organizational Behavior, Vol. 28 (2008), p. 191.
195
D.III Possibilities of Managerial Intervention for Driving Change
an organization to be able to sense and seize the opportunities developing in the market. + desired quality B
DEVELOPMENT OF STRATEGY B
(B) Performance Adaptation Attention to Stakeholder B + Performance +
openness to change -
(R) Repetitive Attention
(R) Repetitive Momentum
Inertia
(R) Performance Decline
generic sens inertia 50% 75% 95% Orientation to Strategy B 1
rel. quality B + (B) Adaptation Pressure for B
pressure for more B
(B) Resistance Pressure for A
pressure for more A -
+ Orientation + to Strategy B -
quality A
100%
0.75
0.5
diffusion of B in remaining market
0.25
0
0
5
10
changed parameters ini inertia ref. fract. institutionalization ref. fract. inertia decrease
15
20
25 30 Time (Year)
35
base run value 0.9 0.3 0.15
40
45
50
parameter range 0.2 – 0.9 0.2 – 0.3 0.15 – 0.2
Figure D-25: Sensitivity to variations of inertia It turns out that the lower inertia the better for the adaptive ability of the organization. Nevertheless, the comparison of the focal organization’s to the market’s orientation to strategy B in Figure D-25 indicates that even very low levels of inertia are not able to trigger an immediate adaptation. Even in the extreme situation in which no additional amount inertia develops (i.e. when fractional institutionalization and inertia decrease are equal at 0.2), several years pass during which the focal organization does not react. The level of inertia has an influence on how quickly the organization adapts attention and reacts to perceived pressure. The reference reaction time is still
196
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
rather long—or expressed differently, the reference fractional change in strategy per pressure is low. The latter is a measure for the speed and the intensiveness of the reaction to perceived pressure. It decides when the system achieves a critical load that it reacts to this pressure by the amendment of its strategy.623 Therefore, the potential of a more intense reaction to perceived pressure will be tested. It distinguishes the rather change aversive base run organization from, e.g., a more decentralized organization. In the latter one, employees are free to initiate their own changes if they perceive pressure to do this or see room for improvement in their respective area. Größler, Grübner, and Milling also illustrate that it might represent an organization that has previously reacted to a complex environment by a more complex internal structure and built-in flexibility that is autonomous of the management function.624 In this way, changes may take place more easily despite initial managerial inertia. At the same time, the repetitive momentum mechanism is not put out of action and alterations still transform the organization and reduce inertia. The upper part of Figure D-26 demonstrates that ceteris paribus an increase of the reference change in strategy per pressure does not have a great impact on the system’s behavior. The increase in the change per pressure somewhat reduces the strength of the Repetitive Momentum Loop which keeps the organization locked at its initial strategy. The radical shift is triggered by the concurrence of the strong Adaptation Pressure for B Loop and the Performance Adaptation Loop with the sudden decrease of the formerly dominant Repetitive Momentum Loop.625 As the bottom graph of Figure D-26 shows, even extreme values of the reference change in strategy per pressure do not lead to the desired result of an early and smooth adaptation. The transformation of perceived pressure into change action is still hampered by inertia and a biased perception of pressure, and adaptation takes place somewhat quicker in the beginning, but not in line with the rest of the industry. A result which may not be intuitive is that a somewhat quicker adaptation in the beginning even results in a much slower implementation of strategy B in the end. The reason for the different behavior is rooted in the difference of the dominance of the Adaptation Pressure for B Loop, not in the nonlinearity of the effect of change on inertia. In their overall behavior simulation runs with a linear effect of change on inertia are hardly different from those shown. While the restrictive character of the Repetitive Momentum Loop is reduced, adaptation initially is quicker and pressure by stakeholders favoring B 623
See Merten, Peter P.: Loop-based Strategic Decision Support Systems, in: Strategic Management Journal, Vol. 12 (1991), No. 5, p. 375. However, the view employed here to some extent deviates from that of Merten since it does not consider binary but continuous effects. 624 See Größler, Andreas, André Grübner and Peter M. Milling: Organisational adaptation processes to external complexity, in: International Journal of Operations & Production Management, Vol. 26 (2006), No. 3/4, pp. 256 and 272–273. 625 It was also tested whether the limiting effects that come into play when the orientation to a strategy is already very high affect the behavior. These limiting effects were explained and displayed in Figure D-10 on page 178, and they limit an overreaction to pressure. The simulation tests did not find any differences in the pattern of behavior when it was assumed that the management fully implements all perceived pressure.
197
D.III Possibilities of Managerial Intervention for Driving Change
never builds up as strongly as in the base run scenario. This means the Adaptation Pressure for B Loop never becomes as dominant. Minor inadequacies that arise when the organization is, in general, reactive only create reactions with less than proportional strength. generic sens change 50% 75% 95% Orientation to Strategy B 1
100%
0.75
0.5
diffusion of B in remaining market
0.25
0
0
5
10
15
20
25 30 Time (Year)
35
40
changed parameter base run value ref. fract. change in strategy per pressure p.a. 0.02 generic sens change 50% 75% 95% Orientation to Strategy B 1
45
50
parameter range 0.02 – 0.2
100%
0.75
0.5
diffusion of B in remaining market
0.25
0
0
5
10
15
20
25 30 Time (Year)
35
changed parameter base run value ref. fract. change in strategy per pressure p.a. 0.02
40
45
50
parameter range 0.02 – 0.5
Figure D-26: Sensitivity for change per perceived pressure Overall, the mere increase of the reference change in strategy per pressure—that may symbolize a more decentralized organization that hands responsibility for change to its employees—does not arrive at the desired result of a quick adaptation to the environment. Additionally, the quick reference change in strategy per pressure
198
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
can be combined with the previously discussed measures to decrease the general level of inertia.626 The combination of a higher reference change and lower inertia has some effect, but it is minor. Even in the best of cases, this only leads to an adaptation delayed by four to five years. Additionally, the effect of attention on the dominance of the loops discussed is worth consideration. Therefore it will now be analyzed how the management team’s influence on attention can leverage the evolution of an organization.
D.III.2 Effects of Increases in the Responsiveness of Attention Attention as a possible lever for management intervention can work in two different ways. An organization can either try to equally distribute its attention to stakeholder groups, or it can change its attention more flexibly than in the base case when stakeholders start to exert pressure for the implementation of another strategy. An equal distribution of attention may be desirable, but it would be unrealistic to assume that a management team is completely unaffected by past developments and by the intensity with which stakeholder groups exert pressure on the organization. The change in importance of new stakeholder groups and the process of allocating attention to them would remain unclear. Hence, there may be a small variability in the initial level of attention, but the main focus will be on the examination of attention allocation over time and on the effects of an enhancement of the responsiveness of attention. An organization that changes attention more easily is simulated by increasing the yearly reference fractional change in attention. This represents a management team that trains employees to sense rising stakeholder groups more quickly, commissions market surveys, or buys information from market research institutes. In an empirical study of small and medium-sized firms, Durand identified differences in the effectiveness of these measures. He found that investment in market information is a factor reducing forecast errors, but that investment in employee capability increases them because it does not sufficiently shift attention outside the organization.627 While there is need for further research into the effects of different measures, it is assumed that an effective measure is chosen by the management team. These measures all aim at being informed about customers’ and other stakeholders’ desires as well as about the organizational strategy’s reputation among these groups in order to then direct adequate attention to them. The simulation reveals a sensitive reaction to an enhancement of the flexibility of attention. The sensitivity to these measures can be seen in Figure D-27. Compared to the base run (black line) it brings forward the change by two to three years. The 626
A sensitivity analysis was run in which the reference fractional change in strategy was changed as described in the upper part of Figure D-26 and inertia was varied as explained in Figure D-25. 627 See Durand, Rodolphe: Predicting a Firm's Forecasting Ability: The Roles of Organizational Illusion of Control and Organizational Attention, in: Strategic Management Journal, Vol. 24 (2003), No. 9, pp. 829 and 833.
199
D.IV Joint Management of Leverage Points
significance of attention also becomes obvious from the sensitivity runs portraying parameter choices for the responsiveness of attention lower than in the base run. Even a minor reduction of the reference change in attention compared to the base run significantly reduces the adaptability of the organization. generic sens attention 50% 75% 95% Orientation to Strategy B 1
100%
0.75
0.5
0.25
0
0
5
10
15
changed parameter ref. fract. change in attention p.a.
20
25 30 Time (Year)
35
base run value 0.05
40
45
50
parameter range 0.03 – 0.5
Figure D-27: Sensitivity for changes in adaptability of attention The model reacts sensitively to changes in the responsiveness of attention, but even a high responsiveness does not trigger a quick adaptation to changes in the organizational environment. Merely reducing the restrictive character of the Repetitive Attention Loop, e.g. by the active fostering of search activities for new stakeholders and their desires is thus no solution. Nevertheless, it can be said that ceteris paribus higher attention to the new stakeholders favoring B is favorable for a quicker adaptation process, and a greater flexibility in adaptation is desirable. It increases the speed of adaptation by a couple of months or years.
D.IV Joint Management of Leverage Points So far the sensitivity to managerial intervention has been analyzed while one lever was changed and the other parameters were kept at the base run values. Now, an investigation into the joint influence on points of leverage will give an idea of the freedom of action of the management team. Changing one lever only revealed that the decrease of inertia and enhancement in the attentiveness has positive outcomes, whereas an increase in the reaction to pressure has mixed effects on organizational adaptation. The simultaneous amendment of several leavers can show whether these findings are stable in different situations as well.
200
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
D.IV.1 Relationship Between the Responsiveness of Strategy to Pressure and Attention The behavior of an early radical adopter was described on page 189, and it is again depicted below by line 2 of Figure D-28. The early radical adopter may represent the London Stock Exchange or one of the early adopters of digital photography. In comparison to this early radical adopter, a second organization is described that seems highly flexible at first glance and should adapt to environmental changes in an instantaneous manner. The organization starts out with a low level of inertia and rather high turnover, high flexibility in attention, a somewhat higher level of attention at the outset, and a strong reaction to perceived pressure. While an equal distribution of attention would be difficult to explain, a somewhat higher level of attention than in the base run can still be legitimized. An organization may try to focus on all stakeholders to be aware if dissatisfaction arises, but it is likely to put more emphasis on those groups who proved important in the past. Contrary to expectations, while the focal organization adapts more quickly in the beginning, it never manages to catch up with its competitors (line 1), so that its behavior resembles a delayed adaptation with a time lag of about three years. It is also remarkable that in the end it even lags behind the early radical adopter. The behavior of this ‘delayed adopter’ is shown by line 3 of Figure D-28. In those runs in which only one lever was improved and the other parameters were kept at base run values, the increase of the reference change in strategy per pressure had ambiguous effects. Since the change in strategy takes a special position among the possible managerial levers for change, it will be further investigated. It will be interesting to see the effects of the variation of the responsiveness to perceived pressure when the management team simultaneously influences inertia and attention to make the organization more adaptive. Therefore, in the following simulation runs, inertia is at a desirable low level, attention changes quickly, and the model’s sensitivity for changes in the reference fractional change in strategy is tested.
201
D.IV Joint Management of Leverage Points
Orientation to Strategy B in Focal Organization and Market 1
3 1 2
Dmnl
0.75
12
0.5
1 2 3 12 3 1 2 31 2 3 12
3
3
1 0.25 1 0
1 23 1 2 3 12 3 0 5 10
3 2 15
2 20
25 30 Time (Year)
35
40
45
50
diffusion of B in remaining market : generic early radical adopter 1 1 Orientation to Strategy B : generic early radical adopter 2 2 2 3 3 3 3 Orientation to Strategy B : generic delayed adopter
changed parameters (delayed adopter) base run value ref. fract. change in strategy per pressure p.a. 0.02 ref. fract. change in attention p.a. 0.05 ref. fract. inertia decrease 0.15 ini inertia 0.9 ini attention to stakeholders favoring B 0.1
parameter value 0.5 0.3 0.28 0.2 0.3
Figure D-28: Delayed adaptation and early radical adaptation (BOT) Since the market is assumed to implement customer requirements, quick adaptation is desired, and it results in the highest performance. In order to measure the adaptation and implementation effectiveness, a new and cumulative measure needs to be introduced. Performance has been modeled in a rather aggregate way. Additionally, it is formulated in such a way that minor inadequacies in its determining factors have a less than proportional influence on it. Therefore, the cumulative strategic orientation gives a finer grained picture of the implementation effectiveness than cumulative performance. Technically, the computation of the adaptation effectiveness was calculated by accumulating the orientation to strategy B over the modeling period. The higher cumulative orientation, the earlier or more fully the organization adapted, and the better. The level of the cumulative orientation to Strategy B is derived as follows: Cumulative Orientation to Strategy B = 50
න ሺOrientation to Strategy Bሻtdt + (Cumulative Orientation to Strategy B)t ǡ D-6 0
t=0
where(Cumulative Orientation to Strategy B)t stands for the initial valueǤ 0
The relationship between the reference fractional change in strategy and the cumulative orientation is shown in Figure D-29. The graph illustrates that a higher frac-
202
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
tional change in strategy does not necessarily lead to better results. There is a maximum point somewhere in the middle for the reference change in strategy. Initially, increasing the organization’s responsiveness to pressure turns out to be useful, but usefulness peaks at a value for the reference fractional change of about 0.06. It plateaus and then decreases for values above 0.1 representing an organization that very quickly responds to perceived pressure. While an organization should not neglect pressure for change it senses in its environment, the results indicate that it should not focus its entire energy on reacting to all different kinds of pressure it perceives. If an organization wants to follow the market as closely as possible, it can be beneficial to wait and see whether an innovation or customer demand gets substantial and receives some establishment in the market. In order to explain the reason for the split results, three strategies that represent a low (1), a desirable (2), and a high (3) fractional change per pressure will be analyzed and compared.
Cumulative Orientation to Strategy B cumulative orientation
30
smooth adopter (ref. change pt06)
29.5
delayed adopter (ref. change pt2)
29 28.5 ref. change pt02
28 0
0.040 0.080 0.120 0.160 (Year) PER PRESSURE P.A. REF. FRACT. CHANGE INTime STRATEGY
changed parameters base run value ref. fract. change in stratetgy per pressure p.a. 0.02 ref. fract. change in attention p.a. 0.05 ref. fract. inertia decrease 0.15 ini inertia 0.9 ini attentio to stakeholders favoring B 0.1
0.200
parameter range 0.02 – 0.2 0.3 0.28 0.2 0.3
Figure D-29: Nonlinear relationship between fractional change and adaptation effectiveness The effects on performance of the three different strategies differ only slightly because, first, in all three cases the organization follows the market fairly quickly, and second, performance is fairly aggregated. In the transition phase, the delayed adopter rather satisfies those favoring strategy A, whereas the smooth adopter pleases stakeholders favoring B. Nevertheless, the analysis of the different patterns of behavior is important because smooth or delayed adaptation might have positive
D.IV Joint Management of Leverage Points
203
and negative side effects on reputation and customer loyalty.628 Knowing about them gives decision-makers higher planning reliability. The upper part of Figure D-30 presents the different behavioral patterns of the three strategies during the period 10 to 30. The strategy leading to the greatest implementation effectiveness is shown by line 2. The organization adapts after an initial period during which the organization is somewhat restrictive in implementing change. In comparison, an organization with a low fractional change (line 1) lags behind due to its lacking transformation of pressure into change. The organization with a high fractional change, i. e the delayed adopter (line 3), takes a special position because it initially adapts even more quickly than the smooth adopter depicted by line 2. But after the initial period, it yet stays behind the organization exhibiting a low fractional change, before in the end it catches up again.629 The reason for the different patterns of behavior becomes apparent from differences in the degree to which the organizations focus their attention on stakeholders favoring B (lower part of Figure D-30). In an organization that is less reactive to pressure, attention shifts rather quickly to those stakeholders favoring B while it only changes slowly in an organization that reacts very flexibly.
628
It is often argued that being a first mover as well as adapting early is the most effective and desirable strategy in terms of customer loyalty and reputation. See Lieberman, Marvin B. and David B. Montgomery: First-Mover (Dis)Advantages: Retrospective and Link with the Resource-Based View, in: Strategic Management Journal, Vol. 19 (1998), No. 12, pp. 1113–1114 and 1122. 629 It has also been tested whether the limiting effect that comes into play if the orientation to a strategy is already very high affects the behavior. Here as well, the behavioral pattern was not affected.
204
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
NYSE Fraction of E-Trade 1
Dmnl
0.75
2 2
diffusion of B in remaining market
0.5
2 231
0.25 0
123 1 10
2 23 1
3
1
3
1
3
1
2 31
2 31
2 312
231
3
1 15
20 Time (Year)
25
Orientation to Strategy B : generic ref. change pt02 Orientation to Strategy B : generic smooth adopter Orientation to Strategy B : generic delayed adopter
1
30
1 2
1 2
3
1 2
2
3
3
3
12
12
Attention to Stakeholders Favoring B 1
1
1 2
1
2
1
2
1
2
1
2
Dmnl
3
12
3
0.75 0.5 0.25
2
1 2 123 1 3
3
2
3
3
3
3
3 3
0 10
15
20 Time (Year)
25
Attention to Stakeholders Favoring B : generic ref. change pt02 Attention to Stakeholders Favoring B : generic smooth adopter Attention to Stakeholders Favoring B : generic delayed adopter
changed parameters base run value ref. fract. change in strategy per pressure p.a. 0.02 ref. fract. change in attention p.a. 0.05 ref. fract. inertia decrease 0.15 ini inertia 0.9 ini attentio to stakeholders favoring B 0.1
30 1
1 2
2 3
3
parameter value 0.02 / 0.06 / 0.2 0.3 0.28 0.2 0.3
Figure D-30: Effect of attention to stakeholders on model behavior (BOT) The interrelationship between the organization’s strategy and the reaction of stakeholders to changes in strategy causes the different trajectories of attention. There is a stakeholder group which expresses a strong desire for a strategy B. In the case of the smooth adopter which exhibits medium responsiveness, in the beginning
205
D.IV Joint Management of Leverage Points
the focal organization remains locked to strategy A. This initial lack of response triggers a strong growth of stakeholder pressure for more B. While the Repetitive Momentum Loop is still strong and does not allow for much change, the growing stakeholder pressure affects attention. A real world example constitutes Nestlé’s customers who expressed their wish for ethical marketing practices, but initially did not see their demands fulfilled at all by Nestlé’s marketing strategy. This dissatisfaction pressure triggers the shift towards higher attention to this stakeholder group. The causal structure of this balancing Adaptation of Attention Mechanism has already been described in Figure D-14 on page 179, but is again illustrated in relation to the full causal structure in Figure D-31. Once the organization becomes more aware of the new stakeholder group, it starts to react to pressure for more B, and this change reduces the rigidity of the Repetitive Momentum and Attention Loop, leading to further change until the organization has adapted. (B) Adaptation of Attention to B
+ Attention to Stakeholder B + Performance +
openness to change Inertia
(B) Performance Adaptation (R) Repetitive Attention
(R) Repetitive Momentum
(R) Performance Decline
rel. quality B + (B) Adaptation Pressure for B
pressure for more B
+ Orientation to + Strategy B (B) Resistance Pressure for A quality A
pressure for more A -
Figure D-31: Adaptation of attention to stakeholder favoring B (CLD) The explanations above clarified the importance of attention for the behavior of the organization. While attention by itself does not determine the fate of an organization, it has an effect on how radical as well as on how fully the organization adapts. A strategic reorientation is supported by a shift of attention. The quick shift of attention is further enhanced by the fact that attention itself also biases the perception of pressure. As a consequence, perceived pressure for more B further diverges from perceived pressure for more A. The behavior is thus influenced by the high interaction between variables and by the mutual interference of balancing and reinforcing loops that include elements of the management, stakeholders, and the environment.
D.IV.2 Policy Implications in Different Environments The investigation into the joint influence on points of leverage has given an idea of the freedom of action of management. In order to test policy sensitivity and the ro-
206
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
bustness of the policy implications of levers of change, further analyses were conducted. They also provide information on the interaction of managerial decisionmaking and environment as drivers of change. So far, low inertia, high flexibility in the distribution of attention, and a medium responsiveness to pressure proved to be useful. However, it remains to see whether this is also the case in different environments. Therefore, reactions to a quicker environmental change will be simulated as well as the impact of strong resisting groups. First, different organizational dispositions were combined with a different development in the environment. Figure D-32 depicts a quicker environmental transformation (line 1) in comparison with the known environmental change (line 2) and the respective response to these two different developments by the smooth adopter. The quick transformation may represent two situations: firstly, a faster development of a new strategy in the market in general. Secondly, the focal organization may have a different reference group in the market and may orient towards those who are first at implementing a new strategy. Multiple simulations were conducted with both environmental developments, further detailed in Appendix C.630 They evidenced the advantage of low inertia and responsive attention over the broad range of parameters tested and are thus an indicator for policy robustness.
Orientation to Strategy B in Focal Organization and Market 1
1
3
1
3 2
4
1 2 3 41 2 3 4 12 3 4 1 23
Dmnl
0.75 4 0.5
3
0.25 0
2
1 1 23 4 1 2 3 4 0 5 10
4
2 15
20
25 30 Time (Year)
35
40
45
50
diffusion of B in remaining market : generic smooth adopter quick 1 2 2 diffusion of B in remaining market : generic smooth adopter Orientation to Strategy B : generic smooth adopter quick 3 3 3 Orientation to Strategy B : generic smooth adopter 4 4 4
Figure D-32: Reaction to quicker environmental change (BOT) Apart from inertia and attention, effects of variation in the reference change in strategy were tested in relation to the quicker environmental transformation. Irrespective of inertia and attention, a medium responsiveness of the strategy to pressure for 630
Sensitivity simulations were run that kept most variables at values as shown in Figure D-28, while one parameter (set) was varied in the range between the base run value and the value shown in Figure D-28.
207
D.IV Joint Management of Leverage Points
change proves to be best. If the environmental transformation takes place more quickly, the peak of the adaptation effectiveness moves von 0.06 to 0.08 for the reference change, but the general relationship and conclusion remains the same, as can be derived from Figure D-33.631 A medium responsiveness of the reference fractional change in strategy per pressure gives the best results. The counterintuitive effect of the responsiveness to pressure thus still holds if the organization is embedded in a different environment. The simultaneous amendment of several leavers is able to show that the recommendations to keep inertia low, attention flexible, and the responsiveness to pressure at a medium level are stable also in different situations.
Cumulative Orientation to Strategy B cumulative orientation
35
ref. change pt08
34.5 34 33.5 33 0
0.100 0.200 0.300 0.400 (Year) PER PRESSURE P.A. REF. FRACT. CHANGE INTime STRATEGY
0.500
Figure D-33: Nonlinear relationship between the fractional change and adaptation effectiveness in the case of quick environmental change Second, concerning different environmental developments, not only the speed of innovation in the market can be considered. Differences may also arise in the close stakeholder environment. In the previous simulation runs it was assumed that all stakeholders finally shift from favoring the old to favoring the new strategy. The case of the New York Stock Exchange exemplified by floor firms that there may exist a stakeholder group that continues to be powerful and to exert pressure for the retention of the old system. This might also represent a situation in which an advocacy group exerts pressure or a permanently loyal customer group exists that favors the old strategy. Figure D-34 reveals how such a group may affect the base run organization. The permanently powerful stakeholders favoring A were varied between the base run value 0 and the value 100. In the latter case the simulation represents circumstances in which a customer group orients to strategy B, but an equally powerful group continues to exert pressure for strategy A. Within certain bounds the decisionmakers react rather sensitively. Initially, they implement strategy B, but the resulting resistance causes a resurgence of the importance of the stakeholders favoring A. If 631
If the reference fractional change in strategy is in the interval between 0.1 and 0.2, a short period of oscillatory behavior occurs. The orientation to stakeholders favoring B weakly oscillates because
208
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
resistance is strong, the management reorients towards those who exert strong pressure. This is exemplified by the dashed line in Figure D-34. It displays the attention to stakeholders favoring B in a situation when the permanently powerful stakeholders favoring A are set to 100. Attention initially shifts to stakeholders favoring B, but when resistance develops, it partially shifts back to those favoring A. With fading resistance attention slowly moves towards stakeholders favoring B again. In summary, the resulting variance in the level of orientation to strategy B shows that powerful stakeholders who develop resistance against the management’s decisions and actions can be powerful at shaping the extent to which a new strategy is implemented. generic sens permanent A 50% 75% 95%
100%
Orientation to Strategy B 1
0.75
Attention to Stakeholders Favoring B (permanently powerful stakeholders favoring A = 100)
0.5
0.25
0
0
10
20
30
40
50
Time (Year)
changed parameter base run value permanently powerful stakeholders favoring A 0
parameter range 0 – 100
Figure D-34: Sensitivity to the variation in the permanent pressure for strategy A At first glance, it does not seem bad if the organization partially shifts back to its old strategy. Indeed, if it serves loyal customers who favor strategy A, this does not pose a problem. However, the permanently powerful stakeholders might represent a group that can significantly impede an organization’s operations such as changeaverse employees. Then they are able to deliver low quality work or damage an organization’s reputation. When at the same time customers already desire strategy B, this poses a problem for performance. In this case the organization needs to take appropriate action to address both stakeholder groups. This is exactly why the NYSE implemented liquidity algorithms. They allowed for an almost complete shift to electronic trading (strategy B), without giving up market quality (quality A). The high specialist involvement and resulting market quality pleases those favoring the floor while customers that want to trade electronically do not have to accept restrictions to fast trading. Concerning the other examples that
D.IV Joint Management of Leverage Points
209
have been discussed this may mean that a photography company shifts towards the production of digital cameras but keeps offering a limited range of traditional models. Since this may tie up too many resources, a strategy that moves towards resolution quality within digital photography could be an alternative solution. It may mean an industry-wide implementation of a minimum ethical code of conduct while further pursuing a profit maximizing strategy. The simulation results and the examples described above reveal that the forces developing from the decisions of the focal organization also provide it with feedback and may in some cases suggest a further revision of the strategy. In the latter example which involved permanent stakeholders, the balancing Resistance Pressure for A Loop became more active. Figure D-35632 portrays a further example in which there is high resistance from permanent stakeholders. In this organizational-environmental setup, permanent stakeholders encounter a very flexible organization with a low level of inertia and a high reference change in attention. Yet, despite its general flexibility, the organization is at the same time reluctant to react to perceived pressure. The simulation run reveals that once the environment (line 1) starts to change, the organization quickly adapts its attention (line 2) to those who become dissatisfied and starts to change its strategy (line 3). The strategic change triggers resistance so that in period 18 the organization quickly shifts attention back to those resisting. Attention then slightly oscillates. Extreme combinations of strong stakeholders and a high responsiveness of attention have the tendency to create oscillatory behavior for which the balancing feedback loops Adaptation of Attention to A (and B) are responsible, shown in the lower part of Figure D-35. The two accumulation delays—attention and strategy—cause oscillations if both Adaptation of Attention Loops are rather active. It hardly happens when management parameters are within reasonable ranges, but extreme pressure by stakeholders in combination with an extremely flexible attention may create oscillatory behavior. This also makes sense in reality. It is intuitive that if an organization changes its attention quickly and if stakeholders react to changes in strategy by varying their pressure, then organizational attention shifts quickly between the stakeholder groups. Hence, having a highly flexible attention (i.e. a very large reference fractional change in attention) is not recommendable in all situations.
632
Figure D-35 portrays a CLD of the full generic model.
210
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
Model Behavior 1 2
Dmnl
0.75
1
2
2
0.5
3
1 2
0.25
2
2
1 0
31 5
2
1
2
2
1 3
3
3
3
3
1
2
2
2
40
45
50
3 1
0
1
1
1
31 3 10
3 15
20
25 30 Time (Year)
35
diffusion of B in remaining market : generic permanent A 1 1 1 Attention to Stakeholders Favoring B : generic permanent A 2 2 2 3 3 3 3 3 Orientation to Strategy B : generic permanent A
changed parameters base run value permanently powerful stakeholders favoring A 0 ref. fract. change in attention p.a. 0.05 ref. fract. inertia decrease 0.15 ini inertia 0.9 ini attention to stakeholders favoring B 0.1 DEVELOPMENT OF STRATEGY B
+
+ Performance +
(B) Adaptation of Attention to B
+ desired quality B
Attention to Stakeholder B openness to change -
-
Inertia
parameter value 100 0.3 0.28 0.2 0.3
(B) Performance Adaptation (R) Repetitive Attention
(R) Repetitive Momentum
(R) Performance Decline
rel. quality B + (B) Adaptation Pressure for B
pressure for more B
+ Orientation to + Strategy B (B) Resistance Pressure for A -
pressure for more A -
quality A (B) Adaptation of Attention to A
Figure D-35: Inconsistent managerial setup (BOT and CLD) The robustness of the policy implications of levers of change has now been analyzed over diverse organizational setups and in different environments. It has been shown that the management team can be a significant driver of change. Concerning
D.V A Feedback Theory of Organizational Inertia, Change, and Attention
211
the managerial policy recommendations for intervening into the system the following can be said:
Independent of the flexibility of attention and responsiveness to pressure, it turns out that the lower inertia the better.
Flexibility in attention is beneficial, but when resistance pressure is strong, the usefulness of flexibility in attention has limits.
Irrespective of inertia and attention, a medium responsiveness of the strategy to pressure for change proves to be best.
Overall, for the organization to be highly adaptive, a low level of inertia needs to be accompanied by a relatively reactive attention and by a medium level of responsiveness to pressure.
D.V A Feedback Theory of Organizational Inertia, Change, and Attention The generic system dynamics model, which builds on the NYSE-specific one, represents a causal theory of organization-environment relations. It is able to give more general answers to the research questions of, first, what the drivers of change in organizations are, and second, whether previous change enhances the occurrence of subsequent transformations. It also explains why and how organizations may have difficulty to adequately adapt to their environment. Concerning the validity of the model and its general applicability for answering the research questions, it can be pointed out that a single causal structure was able to generate different modes of organizational behavior. Simulation runs demonstrated the possibility of failure to change due to inertia, of radical and of smooth adaptation. This bears comparison with the family member test that analyzes whether one system dynamics structure is suitable not only for a single case but for a class of phenomena and situations and whether it is able to generate behavior appropriate for the class.633 In the concrete case here, it explained how different managerial policies create idiosyncratic behavioral patterns and how these policies interact with environmental settings. Concerning the question of drivers of change, the analysis revealed that both the environmental developments and managerial choice are determinants of an organization’s evolution. They are further affected by stakeholders as well as aspects of the management team’s cognitive inertia and attention. The close investigation of the generic base run related the model’s behavior to the underlying causal structure. It shows that organizational change is a multi-faceted process and that the evolution of the focal organization is driven by the complex interaction of reinforcing and balancing feedback relationships. Mechanisms of environmental adaptation that represent 633
See Forrester and Senge: Tests for Building Confidence in System Dynamics Models, 1980, p. 220; and Lane and Smart: Reinterpreting 'generic structure', 1996, p. 110.
212
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
the environment as a driver of change interfere with managerial choice and bounded rational decision-making that thwarts and supports adaptation mainly by reinforcing mechanisms. An adaptive mechanism of stakeholder dissatisfaction aligns the organization with its environment. The modeling process made clear that environmental developments are important to generate change. Differences in the speed of change in the market or different reference points create variations in the behavior of the focal organization. An organization needs to adapt to the demands, be it preferences of a customer group or claims of other stakeholder groups that can affect the organization. Otherwise, the misalignment causes stakeholders to be dissatisfied and organizational performance to deteriorate. Thus, a balancing adaptation mechanism serves as one driver of change. The strength and dominance of this mechanism, however, can differ to a great extent, as the simulation runs have shown. Actions do not follow from environmental conditions per se. The managerial scope of decision proved substantial and makes the management team and its deliberate choices a further driver of change. The generic base run showed that the reinforcing mechanisms of the management team have a great effect on how the organization reacts to alterations in the environment. These mechanisms enforce an organization’s lock-in on its accustomed strategy A. But due to their reinforcing character, these feedback loops also amplify changes once it is initiated due to the reduction of inertia and due to a reorientation of the management team’s attention. The influence of bounded rationality and cognitive elements as part of the choices of decision-makers became obvious in the examples discussed. The case of the NYSE and further instances of Polaroid and DEC revealed that behavior is influenced by bounded rationality as well as routine behavior, expressed by reinforcing structures in both models. E.g., the grown belief in the traditional business model at the NYSE and Polaroid as well as DEC’s cultural blindness prevented a reorientation to new stakeholders. Inertia and the attention to stakeholders shaped the decisionmakers’ exercise of choice. These are central elements in the system dynamics model. The simulation of different organizational and environmental setups supported the importance of these elements for the evolution of organizations. This endorsed prior statements that the exercise of choice requires a prior perception and interpretation of the environment.634 Attention to stakeholders is an important aspect of managerial cognition and serves as a filter for information from the environment. The causal structure of the system dynamics model together with the simulation runs demonstrate that the direct influence of a performance inadequacy on change does not necessarily trigger transformations. Rather attention to stakeholders reacts more flexibly, so that a performance decline together with strong pressure from stakeholders loosens the reinforc634
See Child: Organizational structure, environment and performance, 1972, pp. 4–5; and Child: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment, 1997, p. 48.
D.V A Feedback Theory of Organizational Inertia, Change, and Attention
213
ing Repetitive Attention Loop. The resulting shift in attention is more important for triggering change than performance itself. The shift in attention makes the management team aware of what it has previously neglected and gives the transformation a direction and momentum. Overall, there is strong evidence for the coexistence of the environment and the management as drivers of change. An organization’s evolution depends of the interconnection between these two mechanisms as well as influences from stakeholders, inertia, and attention. It is hence not possible to speak of ‘one’ driver of change. Behavioral outcomes are diverse. Organizational trajectories are driven by the combination of environmental and managerial, i.e. of deterministic and deliberate elements. Milling observes that behavior of firms is controlled by exogenous factors that exert high pressure. But it is not only determined by these forces. Organizational policies have considerable effects and often even cause negative organizational outcomes. There is feedback and mutual influence between the organization and the environment. Social economic systems are able to react to their environment, but also to anticipate its evolution and actively exert influence on it. ”Action and reaction, stimulus and response are tied together in a complex, causal relationship.”635 Several authors suggest a recursive view between structuring and structure.636 On the one hand, decisions of agents in social systems are shaped by the structure of the system. On the other hand, there is a feedback process between decisions and how they create the future decision environment. The second research question of whether prior change serves as a driver of subsequent transformations has also been answered in relation to the interconnectedness of change determinants. The focal organization’s reaction to a two-fold transformation in its environment provided useful insights into the organization’s adaptive ability. As a result of the mutual existence and even interdependence of drivers of change, it can neither be said that past change strictly increases nor decreases the occurrence of future transformations. Several authors found similar evidence. By testing the effect of performance on change, Greve could not fully explain change by per635
Milling, Peter: Business Systems as Control Systems, in: D'Amato, Vittorio and Carlo Maccheroni (Ed.): Dynamic Analysis of Complex Systems, Milano 1989, p. 44. See also Milling: Systemtheoretische Grundlagen zur Planung der Unternehmenspolitik, 1981, p. 106. 636 See Black, Laura J., Paul R. Carlile and Nelson P. Repenning: A Dynamic Theory of Expertise and Occupational Boundaries in New Technology Implementation: Building on Barley's Study of CT Scanning, in: Administrative Science Quarterly, Vol. 49 (2004), No. 4, p. 603; Bloomfield, B: Cosmology, Knowledge and Social Structure: The case of Forrester and system dynamics, in: Journal of Applied Systems Analysis, Vol. 9 (1982), pp. 3–15 (cited in Lane); and Lane: Should System Dynamics be Described as a 'Hard' or 'Deterministic' Systems Approach, 2000, p. 18. Giddens, Orlikowski and Barley hold similar views on structuration. See Barley, Stephen R.: Technology as an Occasion for Structuring: Evidence from Observations of CT Scanners and the Social Order of Radiology Departments, in: Administrative Science Quarterly, Vol. 31 (1986), No. 1; Giddens, Anthony: The Construction of Society: Outline of the Theory of Structuration, Berkeley, CA [et al.] 1984; and Orlikowski, Wanda J.: The Duality of Technology: Rethinking the Concept of Technology in Organizations, in: Organization Science, Vol. 3 (1992), No. 3.
214
D Generic Interpretation of Organizational-Environmental Forces, Feedback, and Change
formance shortfall. Apart from performance effects, recent experience with change lowers inertia and enhances the organizational capability to change.637 Similar to Greve’s analysis, the simulation experiments evidenced that the adaptability of organizations increases by prior change; but this does not necessarily mean the organization will continue to change in the future. Depending on the strength of the adaptive mechanism at that point of time, the organization will find a new equilibrium or continue to evolve. Here again, the interrelationship of the managerial orientation, the organizational past, stakeholder and environmental forces contribute to organizational behavior. Managerial levers of change have received special attention because the setup of the organization and its management team decides about the adaptive or radical pattern of change. Inertia, the flexibility of attention, and the responsiveness of the strategy to pressure serve as policy levers. By their manipulation, the management team can determine the strength of repetitive feedback loops and manage the response to adaptive and resistant feedback pressure. Concerning inertia, an overall low level turned out to increase an organization’s adaptive ability. Therefore, given a certain type of environmental change, the decisions that guide the setup of the management team and its inertia are influential in deciding how the organization reacts to a given exogenous change. A restrictive organization with a homogeneous management team that drives all decisions is likely to overlook important developments in its environment and neglect the rise of new stakeholder groups or of shifting stakeholder preferences. A management team can enhance its organization’s adaptive ability by the active recruitment of employees from outside the organization who bring fresh ideas and are less inward oriented to the accustomed routines as people grown from within. Apart from the active management of the composition of decision-makers in an organization, their institutionalization plays a decisive role. If they are too quickly engrained into the organizational culture and routines, the adaptive ability of the organization is thwarted.638 The management team’s level of inertia is thus an important determinant of organizational outcomes and an important driver as well as inhibitor of change. A low level of inertia is best combined with a high, but not too high responsiveness of attention. This finding derives from a test for policy robustness that included different environments. The combination of a highly flexible reaction of attention with strong and permanent stakeholder pressure for the old strategy A is able to create oscillatory behavior. This is because attention is influenced by several feedback loops. They link the disposition of the management team and the respective strength 637
See Greve: Performance, Aspirations, and Risky Organizational Change, 1998, pp. 78–79 and 81. See also Amburgey, Kelly and Barnett: Resetting the Clock, 1993, pp. 66 and 69–70. 638 Quick institutionalization has also been exemplified for the Intel Corporation and for Linco. See Burgelman: Strategy as Vector and the Inertia of Coevolutionary Lock-in, 2002, p. 354; and Tripsas: Technology, Identity, and Inertia through the Lens of 'The Digital Photography Company', 2009, pp. 450–451.
D.V A Feedback Theory of Organizational Inertia, Change, and Attention
215
and urgency of stakeholder pressure and once again provide evidence for the interconnectedness of determinants of organizational evolution. Additionally, the split behavior resulting from modifications of the reference fractional change in strategy per pressure was insightful for demonstrating the interrelation of a managerial lever for change with environmental aspects. It pointed out that a quick transformation of pressure to strategy has counterintuitive effects on the further development of pressure so that the implementation is somewhat delayed in the end. In this way the interrelationship between the management and the environment as drivers becomes obvious again. While the environment requires the organization to change, first, managerial decisions have an influence on the organization’s reaction to external forces, and second, initial decisions of the management feed back to the environment. They affect the pressure from the environment that the decision-makers will encounter in the future. The generic system dynamics model thus relates behavior and structure, and it points out levers of managerial intervention. It helps explain when the environment and when the management team may be more influential in triggering or upholding change. As such, it unravels the interaction of the environment, the management, and of stakeholders in the closer environment of the organization, and it offers a structural theory of the interrelation of drivers of change in organizations.
217
E Realization of Change in Organizations Many different streams of literature emphasize the importance of organizational change. It is considered necessary that organizations are flexible and adapt to a changing environment.639 However there are many examples of organizations that have failed to change adequately. This dissertation therefore investigated the underlying forces of organizational change and its absence. It concentrated on the reasons for change and its absence. In particular it pursued the question of what are the drivers of change and what forces may inhibit organizational alteration. Since it is often argued that there is repetitive momentum once a shift is initiated, the dissertation also analyzed whether previous change increases or reduces the occurrence of subsequent transformations. In order to answer these questions, the literatures focusing on different drivers of change have been outlined. Their assumptions about organizations and the resulting triggers of change differ. Organizational ecology does not allow organizations the capability of adapting to the environment. The traditional behavioral theory on the other hand assumes them to be malleable, its decision-makers bounded rational, but generally willing and able to adapt to the environment. It grants the environment the role of a trigger of change. The strategic choice approach questions the prominence of the environment. According to this theory, management and decision-makers in an organization have much greater freedom of decision and action than the rather deterministic theories assume. It regards management as the most important driver of change while not denying the influence of organizational and environmental constituencies. A possible combination of epistemological points of view and of drivers of change has been heavily discussed in the sociological and organizational literature. While based on different assumptions, recently there has been an increased use of the combination of theories and a more joint consideration of drivers of change. The analysis of the case study of the New York Stock Exchange’s move towards electronic trading enriches this discussion from a multi-paradigm view. The NYSE is a peculiar example since it adapted later than most of its competitors, but it also stands for a class of organizational dinosaurs. It serves as an example of inertia and change, and the causal and dynamic analysis proved helpful for clarifying the interaction of drivers of change. The system dynamics methodology was chosen because it is able to provide a long-term and process analysis of important drivers of change and their interdependence. The piecemeal development of the system dynamics model of the NYSE’s move to e-trade revealed the importance of a simultaneous consideration of multiple drivers of change. In the NYSE-specific case it proved necessary to consider the environment and management both as drivers of change and to additionally include stake639
See Benner: Securities Analysts and Incumbent Response to Radical Technological Change, 2010, pp. 42 and 59; and Hill and Rothaermel: The Performance of Incumbent Firms in the Face of Radical Technological Innovation, 2003, p. 257.
N. Zimmermann, Dynamics of Drivers of Organizational Change, DOI 10.1007/978-3-8349-6811-1_5, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011
218
E Realization of Change in Organizations
holder dynamics such as cultural and resistance pressure that develops as a response to managerial decisions. Sensitivity analyses with different parameter constellations opened the managerial decision space. At the same time they elucidated the impact by which stakeholders and the market environment constrain the organization. The NYSE provided an example of how an organization is able to simultaneously address different stakeholder demands and to cope with conflicting demands of its environment. The case study explained how and why the NYSE shifted towards electronic trading. It pointed to the importance of reinforcing feedback loops of repetitive momentum and attention that create path-dependent behavior and constrain the organization’s vision and its willingness to change. The strength of these reinforcing processes prevented a smoother adaptation and created the observed radical behavior. This in combination with the further feedback relationships of the model added a causal explanation to the process view of change in organizations. The structures which were important in the NYSE case turned out to have generic relevance since, first, sub-elements of the structure are discussed in separate literatures, and second, since there are similarities to other cases such as the delayed adoption of digital photography at Polaroid or the missing reorientation from mini to personal computers at DEC. The generic system dynamics model and simulation process contributed even more to the idea of the interconnectedness of drivers of change. It provided a structural theory of how an organizational management team, its cognitive side, stakeholders, and the environment are connected and how they are able to create different outcomes of smooth adaptation, radical transformation, or organizational trajectories guided by inertia as a response to an environmental shift. Concerning the research question of what triggers change, the structuralbehavioral analysis of the generic model made clear that there is no single or most important determinant of change. Behavior is driven by the combination of environmental and managerial, i.e. of deterministic and deliberate factors. In this way the present study enriched the stream of literature focusing on organizational choice and on the importance of managerial freedom of action.640 It shows that organizational decision makers do have an influence on how the organization is maintained.641 In particular if an organization wants to be adaptive, the management team has a role to play in asset selection and orchestration.642 In addition to the previous theoretical and empirical studies on managerial choice, the present analysis provides a structural-causal explanation of the relationship between the environment, management, and 640
E.g. Child: Organizational structure, environment and performance, 1972, p. 19; Mellahi and Wilkinson: Organizational failure, 2004, pp. 23 and 27; and de Rond and Thietart: Choice, chance, and inevitability in strategy, 2007, pp. 539 and 546. 641 For similar views also see for example Chandler: Strategy and Structure, 1962, pp. 8 and 383; Child: Organizational structure, environment and performance, 1972, pp. 13–14; and Cyert and March: A Behavioral Theory of the Firm, 1963, pp. 240–241. 642 See Augier and Teece: Dynamic Capabilities and the Role of Managers in Business Strategy and Economic Performance, 2009, p. 417.
E Realization of Change in Organizations
219
organizational outcomes. Decision-makers exercise choice, but their choices are also influenced by conditions provided by their respective environment. Much research has also focused on the environment as a driver of change. Changes in demand, technological innovations, and institutional change have been recognized in the literature as triggers of organizational transformations.643 The simulation analyses make clear that a shift in the environment is needed to trigger any major change. However, even in a generally adaptive organization, specifications in the intensity to which the organization implements perceived pressure for change can lead to different patterns of behavior. In particular, the hump-shaped relationship between the reference fractional change and implementation effectiveness has pointed to the nonlinear emergence of patterns. Behavioral outcomes depend on the continuous interaction and feedback between managerial decisions and stakeholder reactions in the closer organizational environment. Drivers of change are thus intertwined. Highly biased repetitive loops may restrict information intake and conversion to change so that the behavior exhibited may be very different than that of other organizations and well influenced by the managerial setup. The existence of strong reinforcing loops is able to postpone change, but once they lose their strength, they often trigger a radical transformation. A more flexible adaptation on the other hand may also be triggered by the flexible setup of management. Additionally, the relation of the focal organization to its environment has to be taken into account and the existence and strength of stakeholders pressuring for the adoption of a new strategy. Thus, the management, the environment and the setup of stakeholders in the organization’s close environment determine the evolution of organizations. This also means that it cannot be said that prior change increases the occurrence of change in the future. A universally valid answer to the second research question therefore does not exist. Prior change enhances the organization’s adaptability to the environment, but it then also depends on the evolution of this environment and on the compatibility of managerial attention to stakeholders with environmental demands whether one transformation leads to further change. Managerial attention to stakeholders was important in this process although it is not the only determinant of change. In most organizational theories it is outside the boundary of consideration. Previous work on attention exists in the area of the behavioral theory. Its concept of problemistic search represents a search attention to the environment.644 This rather represents attention to issues and solutions; stakeholder attention is rarely discussed. Seldom examples include the business ethics literature in which a normative concept of stakeholder salience has received recognition. But here as well, the explanatory and descriptive side remains a minor matter. 643
See Abernathy and Utterback: Patterns of Industrial Innovation, 1978, pp. 41–46; Haveman, Russo and Meyer: Organizational Environments in Flux, 2001, pp. 253–254 and 269; Meyer, Brooks and Goes: Environmental Jolts and Industry Revolutions, 1990, pp. 94–97; and Romanelli and Tushman: Organizational Transformation as Punctuated Equilibrium, 1994, p. 1145. 644 See Cyert and March: A Behavioral Theory of the Firm, 1992,. pp. 188–190.
220
E Realization of Change in Organizations
What has remained unsolved, however, is the development of managerial attention to stakeholders.645 This has been addressed in this dissertation. This work pointed to the importance of managerial or also an organization’s attention to stakeholders and integrated it with managerial cognition and inertia. The present system dynamics model adds to the existing literature by providing a structural account and theory of how managerial attention to stakeholders evolves. It includes the determinants of stakeholder attention as well as the behavioral effects of attention for the strategy of the entire organization. The present causal model has shown that many factors have an influence on the salience of a stakeholder claim within an organization. It is important to distinguish the real stakeholder pressure from the perceived one. An essential aspect of the postulated system dynamics model and theory is that attention to stakeholders shapes the extent to which different types of real pressure are perceived. Additionally, the degree to which perceived pressure then is implemented also depends on the organization’s inertia and its general disposition or willingness to react to pressure. Hence the relationship between stakeholder attributes and the resulting pressure on the one hand and the organizational outcome on the other hand is complex. Knowing about this complex relationship is important because in many cases an attentional shift precedes a larger strategic change. Thus adaptation, choice, and cognitive-attentional elements have an influence on the evolution of organizations. Theories focusing on one driver of change are informative, but only illuminate one part. The need for the combination of drivers of change also points to the requirement to combine elements of several organizational theories. It has been elaborated in chapter B.II.1 that difficulties with the assumptions and philosophical foundations of theories arise when radically different theories are combined. Nevertheless, the differences among the organization theories discussed here are not as great since they all reside within Burrell and Morgan’s functionalist paradigm.646 The analysis of the New York Stock Exchange as well as the more generic investigation provided support for the usefulness of the combination of organizational theories. It allows for the analysis of organizational phenomena not only with one, but with two or multiple lenses. As a consequence, the observer receives a richer and more multi-faceted picture of an organization. In the present example it contains elements of environmental determinism and managerial decision-making at the same time. Stakeholder influences and cognitive aspects also enrich the picture. As proponents of the multi-paradigm perspective affirm, it allows for a more comprehensive consideration of phenomena than a single paradigmatic perspective. Mutually exclusive views of the social world have incommensurable as well as continuous el-
645
Mitchell, Agle, and Wood’s theory of stakeholder salience represents a first step into this direction. See Mitchell, Agle and Wood: Toward a Theory of Stakeholder Identification and Salience, 1997, pp. 868 and 879–880. 646 See Burrell and Morgan: Sociological Paradigms and Organisational Analysis, 1979, p. 25–30.
E Realization of Change in Organizations
221
ements, and this diversity is a possibility for learning.647 In the present study, the integration of different elements emerged naturally. Poole and Van de Ven argue that a multi-paradigm perspective is able to clarify the relations between different points of view.648 The elicitation of the causal structure in combination with the structuralbehavioral analysis in the NYSE and the generic case clearly support this statement. It explicitly shows how phenomena that are discussed in different perspectives are entangled. However, one has to be aware that the diverse picture is still a limited view which is more informative than the investigation of organizations by one lens only, but will never be exhaustive. The multi-paradigm analysis also allows for an investigation of different levels of abstraction. This dissertation combines a group and organizational level of abstraction with phenomena such as bounded rationality and cognition which are usually placed at the micro level. It succeeds in doing so by providing a causal explanation of how inertia and attention to stakeholders develop and lead to bounded rational decisions in organizations. The aggregated view of system dynamics makes it possible to integrate the bounded rational reasoning of the decision-making group with organizational outcomes.649 Latour emphasizes the usefulness of the combination of different levels of abstraction because the levels do not stand by themselves, but are linked.650 In the system dynamics model presented above, this linkage forms by shared team properties including shared mental models.651 These team properties of managerial attention and cognition derive from individual (bounded) rationality. Although individual understandings and representations differ, they are understood as what is shared among individuals in a group.652 The focus is on their behavior and change over time. The two team concepts of managerial attention and cognition bridge the gap between the underlying individual and micro-level to the organizational level. The present analysis is able to reveal how a shared cognition evolves over time.
647
See Willmott: Breaking the Paradigm Mentality, 1993, pp. 701–703 and 708. See also Gioia and Pitre: Multiparadigm Perspectives on Theory Building, 1990, pp. 598–599; and Morgan: Images of Organization, 2006, pp. 8 and 337–339. 648 See Poole and Van de Ven: Using Paradox to Build Management and Organization Theories, 1989, p. 576. 649 On the linkage of bounded rationality and system dynamics see Größler, Andreas, Peter Milling and Graham Winch: Perspectives on rationality in system dynamics—a workshop report and open research questions, in: System Dynamics Review, Vol. 20 (2004), No. 1, pp. 77–78. 650 See Latour, Bruno: We Have Never Been Modern, Cambridge, MA 1993, p. 121. 651 Klein and Kozlowski subsume shared mental models, norms, and team cohesion under the concept of shared team properties. See Klein, Katherine J. and Steve W. J. Kozlowski: From Micro to Meso: Critical Steps in Conceptualizing and Conducting Multilevel Research, in: Organizational Research Methods, Vol. 3 (2000), No. 3, p. 215. 652 For a typology of group mental model-like concepts and their modes of analysis see Kim: In search of a mental model-like concept for group-level modeling, 2009, pp. 213 and 216–219.
222
E Realization of Change in Organizations
Cognitive change, as explained by the structural-behavioral investigation of the system dynamics model, can also be compared to a double-loop learning process.653 The Adaptation Pressure and Performance Adaptation Loops, depicted in Figure E-1, serve as a first-order learning structure by which the organization reduces the gap between its strategic orientation and that of the environment. Information feedback from the organizational environment does not only trigger new decisions. It also involves double-loop learning by altering the mental model decision-makers have of the real world. It thus exerts influence on managerial inertia, cognition and in particular on managers’ attention to stakeholders, as described by the Adaptation of Attention Loops. This influence not always leads to better mental models; more important, they may be biased in just a different way. As part of this process, the decision-makers’ mental models alter the information collected or perceived from the environment. In this way, attention to stakeholders serves as a weight on incoming pressure, and perception alters attention even further. This confirms prior findings that durable organizational change not only derives from adaptation processes, but require a further learning cycle by which cognition changes as well.654 The system dynamics model adds a causal structure that explains how cognitive changes develop and interact with adaptive processes. This view accounts for the claim that the mind and the environment need to come to terms which each other by mutual adaptation.655 In particular perceived pressure is shaped by both, the environment and managerial perception. In the case that management is biased, it may distort the perception of the need to change. In future research, the analysis of perception and attention in relation to stakeholders in the organization’s environment may be enlarged: it may incorporate managerial attention to more than two stakeholder groups. In the present generic model, the investigation has been limited to two possible organizational strategies and two stakeholder groups because this is the standard situation when an organization needs to decide between an old and a new strategy. Concerning stakeholder attention, the analysis of the New York Stock Exchange’s move to electronic trading included three stakeholder groups: institutional customers, non-institutional customers, and floor firms. It is thus also possible to integrate and analyze more than two stakeholder groups and the effect of their demands on the evolution of organizations if the situation analyzed is characterized by diversified stakeholders.
653
For a causal account of double loop learning see Radzicki, Michael J.: Mr. Hamilton, Mr. Forrester, and a Foundation for Evolutionary Economics, in: Journal of Economic Issues, Vol. 37 (2003), No. 1, p. 154; and Sterman: Business Dynamics, 2000, pp. 18–19. 654 See Schimmel and Muntslag: Learning Barriers, 2009, p. 413. 655 See Gigerenzer, Todd and the ABC Research Group: Simple Heuristics That Make Us Smart, p. 22. They refer to a concept developed by Brunswik. See Brunswick, Egon: Scope and Aspects of the Cognitive Problem, in: University of Colorado at Boulder, Department of Psychology (Ed.): Contemporary Approaches to Cognition: A Symposium Held at the University of Colorado, Cambridge, MA 1957, p. 5.
223
E Realization of Change in Organizations
DEVELOPMENT OF STRATEGY B
+ Attention to Stakeholder B + Performance +
(B) Adaptation of Attention to B
+ desired quality B
openness to change Inertia
(B) Performance Adaptation (R) Repetitive Attention
(R) Repetitive Momentum
(R) Performance Decline
rel. quality B + (B) Adaptation Pressure for B
pressure for more B
(B) Resistance Pressure for A
pressure for more A -
+ Orientation to + Strategy B
quality A (B) Adaptation of Attention to A
Figure E-1: Full generic causal loop diagram (CLD) It would additionally be possible to tackle the levers of change in more detail. The process of inertia decrease could be fine-tuned and elaborated. In particular the effects of employee turnover could be compared to other measures such as job rotation systems, more diversified recruiting, unlearning and else. The institutionalization process is equally important. It would be interesting to investigate what measures have limiting effects on the growth of inertia without hampering the growth of organizational experience and competence. Experience has been associated with better performance because team members learn to work with each other. They learn about each other’s mental models.656 In this respect the inclusion of experience into the analysis would prove helpful. In further research it could for example be incorporated in the way Sastry included it in her system dynamics model of the punctuated equilibrium approach.657 The reason why experience was excluded from the present analysis is that it neither represents a central element in the NYSE’s transition from manual to electronic trading nor were the other examples related to missing experience. The omission of unethical marketing practices is not linked to competence, and Polaroid, for instance, had already built its experience in digital photography, but failed to change its business model. While floor trading requires a lot of human intervention and experience, this is different with e-trade. Participants are anonymous and a computer matches orders without any human intervention. Of course, traders also need to become accustomed to the elec656
657
See Huber, George P. and Kyle Lewis: Cross-understanding: Implications for Group Cognition and Performance, in: Academy of Management Review, Vol. 35 (2010), No. 1, p. 2010. For a formal representation of experience in the punctuated equilibrium approach see Sastry: Problems and Paradoxes in a Model of Punctuated Organizational Change, 1997, pp. 243–246 and 249.
224
E Realization of Change in Organizations
tronic environment, but the growth of experience is much less important than in settings that involve human interaction. Therefore, the effects of experience with a certain strategy on performance were excluded in the present study. Since competence with a new strategy can be important in cases of organizational change and since it might conflict with low inertia, the present study could be enlarged in this direction. In a similar manner, as the introduction of new trading technology did not require the active support of those working with it, difficulties that can arise in the implementation phase of a new strategy or technology are outside the boundary of this dissertation. It was not essential in the examples analyzed, but inner-organizational difficulties such as employee resistance could be incorporated in the further investigation of cases in which implementation effectiveness depends on human support. Concerning the implementation of change initiatives, future research might include the effectiveness and direction of measures management takes.658 If the change concerns the transition from an old to a new technology the direction of change is straightforward. Nevertheless, it is within the realms of possibility that the management team launches change initiatives that are ineffective or have unintended consequences which go beyond resistance from those favoring the old strategy. The analysis of these effects, however, represents an implementation problem that may build up on the present study. The latter is rather concerned with the cognitive side and with the circumstances that determine the decision to initiate change. The present study contributed to organizational change theory, but the knowledge derived from the investigation of New York Stock Exchange and of the generic system dynamics model can be used in practice. Several concrete recommendations can be given. The analyses have shown the scope for managerial action is great, but in order to maintain adequate performance, it is also necessary that an organization is in accordance with the demands its environment poses. Outcomes can be ameliorated in different ways; by increasing adaptability or the overall benefits of adaptation. In order for this to happen, the management team can reduce managerial and organizational inertia by lowering institutionalization and by enhancing the loss of inertia. It may reduce institutionalization by training and by the active maintenance of diversity. The decrease of inertia can be enhanced by turnover, job rotation systems, more diversified recruiting, learning, and else. In addition to inertia, the organization’s general handling of perceived pressure turned out to be a lever for change. A medium responsiveness of its strategy to pressure proved best for an organization’s adaptability. While a low transformation of pressure into action will hamper and delay change, it is beneficial to wait and see whether a stakeholder claim gets substantial and then quickly react to it.
658
E.g. Baum and Singh: Dynamics of Organizational Responses to Competition, 1996, p. 1287 and Delacroix and Swaminathan: Cosmetic, Speculative, and Adaptive Organizational Change in the Wine Industry, 1991, p. 657.
E Realization of Change in Organizations
225
Concerning managerial attention to stakeholders, rather high flexibility proved best. Decision-makers may commission market surveys or mandate market research institutes so as to early perceive emergent stakeholder groups and their claims. At the same time they need to make sure that their attention does not swing back as a result of a too high reaction to pressure in combination with strong stakeholder forces. Examples of the NYSE and DEC revealed that organizations often have strong relationships to existing stakeholders—here to floor firms and to business clients, respectively. But organizational managers often neglect emerging groups and their desires. The mere awareness of this problem and in particular knowledge of how it causally develops can help the organization. The inclusion of this knowledge into day-to-day business would increase decision-makers’ consciousness of the risk of growing inertia and failure to attend to important stakeholder groups. This can reduce their inward orientation and make the organization and its management more attentive to changes in its environment. The outcome of change initiatives can be increased by a reduction of negative consequences that come along with all organizational changes. As the case of the NYSE exemplified, it is not necessary to imitate the environment. For instance, by keeping the floor, the NYSE was able to reduce the negative consequences electronic trading had for some of its stakeholder groups. This shows that—while adopting a new strategy—it can be useful not to forget about the advantages of the old one. The NYSE did this by simultaneously striving for speed and market quality. Digital photography increasingly strives for resolution quality. PCs now also achieve high computation power. Customers who demand ethical behavior by the entire supply chain do not forget about price aspects either. It is important that organizations adapt and fulfill the demands of their environment, but they do not need to mechanically imitate other organizations. They are free to develop intelligent strategies to meet demands of several stakeholder groups at the same time. Managerial action thus bears great importance. Decision-makers shape the way an organization reacts to its environment. The overall evolution on an organization depends on the combination of the environment, its stakeholders, and organizational decision-makers. They constructively irritate and inspire each other. Nevertheless, the management team has sufficient freedom of action to shape its response to external factors and to enact its future decision environment.
227
References Abernathy, William J. and James M. Utterback: Patterns of Industrial Innovation, in: Technology Review, Vol. 80 (1978), No. 7, pp. 40–47. Abolafia, Mitchel Y.: Making Markets: Opportunism and Restraint on Wall Street, Cambridge, MA and London 1996. Adler, Paul S. and Bryan Borys: Materialism and Idealism in Organizational Research, in: Organization Studies, Vol. 14 (1993), No. 5, pp. 657–679. Agle, Bradley R., Ronald K. Mitchell and Jeffrey A. Sonnenfeld: Who Matters to CEOs? An Investigation of Stakeholder Attributes and Salience, Corporate Performance, and CEO Values, in: Academy of Management Journal, Vol. 42 (1999), No. 5, pp. 507–525. Agle, Bradley R., Ronald K. Mitchell and Jeffrey A. Sonnenfeld: A Report on Stakeholder Attibutes and Salience, Corporate Performance, and CEO Values, in: Logsdon, Jeanne M., Donna J. Wood and Lee E. Benson (Ed.): Research in Stakeholder Theory, 1997–1998: The Sloan Foundation Minigrant Project, Toronto, Canada 2000, pp. 39–53. Aldrich, Howard E.: Organizations and Environments, Stanford, CA 2008. Aldrich, Howard E. and Martin Ruef: Organizations Evolving, 2. Ed., London [et al.] 2006. Allen, Gove N., Dan L. Burk and Gordon B. Davis: Academic Data Collection in Electronic Environments: Defining Acceptable Use of Internet Resources, in: MIS Quarterly, Vol. 30 (2006), No. 3, pp. 599–610. Amburgey, Terry L., Dawn Kelly and William P. Barnett: Resetting the Clock: The Dynamics of Organizational Change and Failure, in: Administrative Science Quarterly, Vol. 38 (1993), No. 1, pp. 51–73. Amburgey, Terry L. and Anne S. Miner: Strategic Momentum: The Effects of Repetitive, Positional, and Contextual Momentum on Merger Activity, in: Strategic Management Journal, Vol. 13 (1992), No. 5, pp. 335–348. Ammer, Daniela: Die Umwelt des World Wide Web: Bildung für nachhaltige Entwicklung im Medium World Wide Web aus pädagogischer und systemtheoretischer Perspektive, Ph.D. Thesis, Department of Social and Behavioral Sciences, University of Tübingen, Tübingen 2010. Argote, Linda and Henrich R. Greve: A Behavioral Theory of the Firm—40 Years and Counting: Introduction and Impact, in: Organization Science, Vol. 18 (2007), No. 3, pp. 337–349. Astley, W. Graham: The Two Ecologies: Population and Community Perspectives on Organizational Evolution, in: Administrative Science Quarterly, Vol. 30 (1985), No. 2, pp. 224–241. Astley, W. Graham and Andrew H. Van de Ven: Central Perspectives and Debates in Organization Theory, in: Administrative Science Quarterly, Vol. 28 (1983), No. 2, pp. 245–273. Attewell, Paul: Technology Diffusion and Organizational Learning: The Case of Business Computing, in: Organization Science, Vol. 3 (1992), No. 1, pp. 1–19.
N. Zimmermann, Dynamics of Drivers of Organizational Change, DOI 10.1007/978-3-8349-6811-1, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011
228
References
Audia, Pino G., Edwin A. Locke and Ken G. Smith: The Paradox of Success: An Archival and a Laboratory Study of Strategic Persistence Following Radical Environmental Change, in: The Academy of Management Journal, Vol. 43 (2000), No. 5, pp. 837–853. Augier, Mie and David J Teece: Dynamic Capabilities and the Role of Managers in Business Strategy and Economic Performance, in: Organization Science, Vol. 20 (2009), No. 2, pp. 410–421. Barclay, Michael J., Terrence Hendershott and Kenneth Kotz: Automation versus Intermediation: Evidence from Treasuries Going Off the Run, in: Journal of Finance, Vol. 61 (2006), No. 5, pp. 2395–2414. Barlas, Yaman: Formal aspects of model validity and validation in system dynamics, in: System Dynamics Review, Vol. 12 (1996), No. 3, pp. 183–210. Barley, Stephen R.: Technology as an Occasion for Structuring: Evidence from Observations of CT Scanners and the Social Order of Radiology Departments, in: Administrative Science Quarterly, Vol. 31 (1986), No. 1, pp. 78–108. Barnard, Chester I.: The Functions of the Executive, Cambridge, MA 1938. Barnett, William P. and Glenn R. Carroll: Modeling Internal Organizational Change, in: Annual Review of Sociology, Vol. 21 (1995), No. 1, pp. 217–236. Barnett, William P. and John Freeman: Too Much of a Good Thing? Product Proliferation and Organizational Failure, in: Organization Science, Vol. 12 (2001), No. 5, pp. 539–558. Barr, Pamela S., J. L. Stimpert and Anne S. Huff: Cognitive Change, Strategic Action, and Organizational Renewal, in: Strategic Management Journal, Vol. 13 (1992), No. -, Summer Special Issue, pp. 15–36. Barron, David N., Elizabeth West and Michael T. Hannan: A Time to Grow and a Time to Die: Growth and Mortality of Credit Unions in New York City, 1914–1990, in: The American Journal of Sociology, Vol. 100 (1994), No. 2, pp. 381–421. BATS:
Market Volume Summary, no http://batstrading.com/market_summary/.
date,
accessed
12 Jun. 2010:
Battalio, Robert, Andrew Ellul and Robert Jennings: Reputation Effects in Trading on the New York Stock Exchange, in: Journal of Finance, Vol. 62 (2007), No. 3, pp. 1243–1271. Baum, Joel A. C. and Jitendra V. Singh: Dynamics of Organizational Responses to Competition, in: Social Forces, Vol. 74 (1996), No. 4, pp. 1261–1297. Beck, Nikolaus, Josef Brüderl and Michael Woywode: Momentum or Deceleration? Theoretical and Methodological Reflections on the Analysis of Organizational Change, in: Academy of Management Journal, Vol. 51 (2008), No. 3, pp. 413–435. Beck, Nikolaus and Alfred Kieser: The Complexity of Rule Systems, Experience and Organizational Learning, in: Organization Studies, Vol. 24 (2003), No. 5, pp. 793–814. Becker, Markus C.: Organizational routines: a review of the literature, in: Industrial and Corporate Change, Vol. 13 (2004), No. 4, pp. 643–678.
References
229
Beer, Michael: Organization Change and Development: A Systems View, Santa Monica, CA 1980. Benner, Mary J.: Securities Analysts and Incumbent Response to Radical Technological Change: Evidence from Digital Photography and Internet Telephony, in: Organization Science, Vol. 21 (2010), No. 1, pp. 42-62. Bennett, Paul and Li Wei: Market structure, fragmentation, and market quality, in: Journal of Financial Markets, Vol. 9 (2006), No. 1, pp. 49–78. Berman, Shawn L., Andrew C. Wicks, Suresh Kotha and Thomas M. Jones: Does Stakeholder Orientation Matter? The Relationship between Stakeholder Management Models and Firm Financial Performance, in: Academy of Management Journal, Vol. 42 (1999), No. 5, pp. 488–506. Bertrand, J. Will M. and Jan C. Fransoo: Operations management research methodologies using quantitative modeling, in: International Journal of Operations & Production Management, Vol. 22 (2002), No. 2, pp. 241–264. Biais, Bruno, Larry Glosten and Chester Spatt: Market microstructure: A survey of microfoundations, empirical results, and policy implications, in: Journal of Financial Markets, Vol. 8 (2005), No. 2, pp. 217–264. Black, Laura J., Paul R. Carlile and Nelson P. Repenning: A Dynamic Theory of Expertise and Occupational Boundaries in New Technology Implementation: Building on Barley's Study of CT Scanning, in: Administrative Science Quarterly, Vol. 49 (2004), No. 4, pp. 572–607. Blake, Rich: NYSE Technology Threatens Survival of Floor Traders, in: Institutional Investor, Vol. 43 (2009), No. 6, issued Jul./Aug. 2009, pp. 28–32. Bloomfield, B: Cosmology, Knowledge and Social Structure: The case of Forrester and system dynamics, in: Journal of Applied Systems Analysis, Vol. 9 (1982), pp. 3–15. Blume, Marshall E., Jeremy J. Siegel and Dan Rottenberg: Revolution on Wall Street: The Rise and Decline of the New York Stock Exchange, New York and London 1993. Boehmer, Ekkehart: Dimensions of execution quality: Recent evidence for US equity markets, in: Journal of Financial Economics, Vol. 78 (2005), No. 3, pp. 553–582. Boehmer, Ekkehart, Robert Jennings and Li Wei: Public Disclosure and Private Decisions: Equity Market Execution Quality and Order Routing, in: Review of Financial Studies, Vol. 20 (2007), No. 2, pp. 315–358. Borrus, Amy: No More Breaks for the Big Board, in: Business Week, (2004), No. 3911, issued 6 Dec. 2004, p. 46. Bowen, Michael G.: System dynamics, determinism, and choice: Toward a reconsideration of the image of "systems man", in: System Dynamics Review, Vol. 10 (1994), No. 1, pp. 87–90. Brandon: Comment on: When a quote goes 'slow,' speed vs. price is for customer to decide (11 Jan. 2007, commented on 16 Jan. 2007), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2007, accessed 9 Dec. 2009: http://exchanges.nyse.com/archives/ 2007/01/speed_vs_price.php.
230
References
Brown, Shona L. and Kathleen M. Eisenhardt: The Art of Continuous Change: Linking Complexity Theory and Time-Paced Evolution in Relentlessly Shifting Organizations, in: Administrative Science Quarterly, Vol. 42 (1997), No. 1, pp. 1–34. Brüderl, Josef and Rudolf Schüssler: Organizational Mortality: The Liabilities of Newness and Adolescence, in: Administrative Science Quarterly, Vol. 35 (1990), No. 3, pp. 530–547. Brunswick, Egon: Scope and Aspects of the Cognitive Problem, in: University of Colorado at Boulder, Department of Psychology (Ed.): Contemporary Approaches to Cognition: A Symposium Held at the University of Colorado, Cambridge, MA 1957, pp. 5–40. Bryant, Anthony and Kathy Charmaz (Ed.): The SAGE Handbook of Grounded Theory, Los Angeles [et al.] 2007. Bryman, Alan and Emma Bell: Business Research Methods, 2. Ed., Oxford, UK 2007. Buchanan, David and Patrick Dawson: Discourse and Audience: Organizational Change as Multi-Story Process, in: Journal of Management Studies, Vol. 44 (2007), No. 5, pp. 669–686. Bunge, Jacob: NYSE Cuts Trade Times to 5 Milliseconds, in: Wall Street Journal (2009), issued 2 Jul. 2009, p. C.5. Bureau of Labor Statistics, U.S. Department of Labor: Job Openings and Labor Turnover Survey (JOLTS), 2009, accessed 22 Nov. 2009: http://data.bls.gov/PDQ/ outside.jsp?survey=jt. Burgelman, Robert A.: Strategy as Vector and the Inertia of Coevolutionary Lock-in, in: Administrative Science Quarterly, Vol. 47 (2002), No. 2, pp. 325–357. Burrell, Gibson and Gareth Morgan: Sociological Paradigms and Organisational Analysis, London 1979. Carrigan, Marylyn and Ahmad Attalla: The myth of the ethical consumer - do ethics matter in purchase behaviour?, in: The Journal of Consumer Marketing, Vol. 18 (2001), No. 7, pp. 560–577. Caselli, Francesco and Wilbur John Coleman II: Cross-Country Technology Diffusion: The Case of Computers, in: The American Economic Review, Vol. 91 (2001), No. 2, pp. 328–335. Chandler, Alfred D. Jr.: Strategy and Structure: Chapters in the History of the Industrial Enterprise, Cambridge, MA 1962. Chapman, Peter: Reg NMS Is a Winner, SEC Says, in: Traders Magazine, Vol. 21 (2008a), No. 278, p. 22. Chapman, Peter: Rule Changes: NYSE Euronext Bets on Specialists to Revive Broken Floor, Traders Magazine Online, 2008b, accessed 20. Aug. 2008: http://www.tradersmagazine.com/news/101217-1.html. Chapman, Peter, Nina Mehta and Michael Scotti: Men At Work, in: Traders Magazine, Vol. 20 (2007), No. 274, issued Nov. 2007, pp. 44–75. Child, John: Organizational structure, environment and performance: The role of strategic choice, in: Sociology, Vol. 6 (1972), pp. 1–22.
References
231
Child, John: Strategic Choice in the Analysis of Action, Structure, Organizations and Environment: Retrospect and Prospect, in: Organization Studies, Vol. 18 (1997), No. 1, pp. 43–76. Child, John and Alfred Kieser: Development of organizations over time, in: Starbuck, William H. and Paul C. Nystorm (Ed.): Handbook of organizational design, Vol. 1. Adapting organizations to their environments, Oxford 1981, pp. 28–64. Cho, Theresa S. and Donald C. Hambrick: Attention as the Mediator Between Top Management Team Characteristics and Strategic Change: The Case of Airline Deregulation, in: Organization Science, Vol. 17 (2006), No. 4, pp. 453–469. Christensen, Clayton M: The Innovator's Dilemma: When New Technologies Cause Great Firms to Fail, Boston, MA 1997. Clark, Colin: Growth in the NYSE's Liquidity Provider Programs (8 Oct. 2009), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2009, accessed 10 Oct. 2009: http://exchanges.nyse.com/archives/ 2009/10/growth.php. Clemons, Eric K. and Bruce W. Weber: London's big bang: a case study of information technology, competitive impact, and organizational change, in: Journal of Management Information Systems, Vol. 6 (1990), No. 4, pp. 41–60. Clemons, Eric K. and Bruce W. Weber: Information Technology and Screen-Based Securities Trading: Pricing the Stock and Pricing the Trade, in: Management Science, Vol. 43 (1997), No. 12, pp. 1693–1708. Coch, Lester and John R. P. French, Jr.: Overcoming Resistance to Change, in: Human Relations, Vol. 1 (1948), No. 4, pp. 512–532. Cohen, Kalman J. and Richard M. Cyert: Computer Models in Dynamic Economics, in: The Quarterly Journal of Economics, Vol. 75 (1961), No. 1, pp. 112–127. Cohen, Michael D., Roger Burkhart, Giovanni Dosi, Massimo Egidi, Luigi Marengo, Massimo Warglien and Sidney G. Winter: Routines and Other Recurring Action Patterns of Organizations: Contemporary Research Issues, in: Industrial and Corporate Change, Vol. 5 (1996), No. 3, pp. 653–698. Cohen, Michael D., James G. March and Johan P. Olsen: A Garbage Can Model of Organizational Choice, in: Administrative Science Quarterly, Vol. 17 (1972), No. 1, pp. 1–25. Collins, Jamie D., Tim R. Holcomb, S. Trevis Certo, Michael A. Hitt and Richard H. Lester: Learning by doing: Cross-border mergers and acquisitions, in: Journal of Business Research, Vol. 62 (2009), No. 12, pp. 1329–1334. Conner, Daryl R.: Managing change: A business imperative, in: Business Quarterly, Vol. 58 (1993), No. 1, pp. 88–92. Conroy, Robert M. and Robert L. Winkler: Market Structure: The Specialist as Dealer and Broker, in: Journal of Banking & Finance, Vol. 10 (1986), No. 1, pp. 21–36. Crotty, Michael: The foundations of social research: Meaning and perspective in the research process, London [et al.] 1998. Cyert, Richard M. and James G. March: A Behavioral Theory of the Firm, Englewood Cliffs, NJ 1963.
232
References
Cyert, Richard M. and James G. March: A Behavioral Theory of the Firm, 2. Ed., Englewood Cliffs, NJ 1992. D'Aveni, Richard A. and Ian C. MacMillan: Crisis and the Content of Managerial Communications: A Study of the Focus of Attention of Top Managers in Surviving and Failing Firms, in: Administrative Science Quarterly, Vol. 35 (1990), No. 4, pp. 634–657. Davis, Jason P., Kathleen M. Eisenhardt and Christopher B. Bingham: Developing Theory though Simulation Methods, in: Academy of Management Review, Vol. 32 (2007), No. 2, pp. 480–499. Deetz, Stanley: Describing Differences in Approaches to Organization Science: Rethinking Burrell and Morgan and Their Legacy, in: Organization Science, Vol. 7 (1996), No. 2, pp. 191–207. Delacroix, Jacques and Anand Swaminathan: Cosmetic, Speculative, and Adaptive Organizational Change in the Wine Industry: A Longitudinal Study, in: Administrative Science Quarterly, Vol. 36 (1991), No. 4, pp. 631–661. Demers, Christiane: Organizational Change Theories: A Synthesis, Los Angeles, CA [et al.] 2007. Demsetz, Harold: The Cost of Transacting, in: The Quarterly Journal of Economics, Vol. 82 (1968), No. 1, pp. 33–53. Dent, Eric B. and Susan Galloway Goldberg: Challenging "Resistance to Change", in: Journal of Applied Behavioral Science, Vol. 35 (1999), No. 1, pp. 25–41. Dequech, David: Bounded Rationality, Institutions, and Uncertainty, in: Journal of Economic Issues, Vol. 35 (2001), No. 4, pp. 911–929. Dey, Tony: Comment on: Men At Work (at re-making the markets) (27 Nov. 2007, commented on 29 Nov. 2007), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2007a, accessed 25 Mar. 2008: http://exchanges.nyse.com/archives/2007/11/ men_at_work.php. Dey, Tony: Comment on: NYSE to introduce Do Not Ship orders (13 Mar. 2007, commented on 14 Mar. 2007), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2007b, accessed 12 Oct. 2009: http://exchanges.nyse.com/archives/2007/03/ do_not_ship.php. Dey,
Tony: Comment on: Readers discuss automating NYSE openings (23 Oct. 2007, commented on 28 Oct. 2007), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2007c, accessed 10 Oct. 2009: http://exchanges.nyse.com/archives/2007/10/ discuss_automated.php.
Dey, Tony: Comment on: 'The floor is not going away. OK?' (27 Jun. 2007, commented on 28 Jun. 2008), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2007d, accessed 25 Mar. 2008: http://exchanges.nyse.com/archives/2007/06/duncan.php. Dholakia, Nikhilesh and Dong Zhang: Online Qualitative Research in the Age of ECommerce: Data Sources and Approaches, in: Forum: Qualitative Social Research, Vol. 5 (2004), No. 2, pp. 1–9.
References
233
Diamond, Michael A.: Resistance to Change: A Psychoanalytic Critique of Argyris and Schön's Contributions to Organization Theory and Intervention, in: Journal of Management Studies, Vol. 23 (1986), No. 5, pp. 543–562. Dillon, Tim: NYSE leader looks to its reputation (8 Mar. 2004), in: USA Today, 2004, accessed 10 Dec. 2009: http://www.usatoday.com/money/markets/us/200403-08-thain_x.htm. Dobrev, Stanislav D., Tai-Young Kim and Glenn R. Carroll: The Evolution of Organizational Niches: U.S. Automobile Manufacturers, 1885–1981, in: Administrative Science Quarterly, Vol. 47 (2002), No. 2, pp. 233–264. Dobrev, Stanislav D., Tai-Young Kim and Glenn R. Carroll: Shifting Gears, Shifting Niches: Organizational Inertia and Change in the Evolution of the U.S. Automobile Industry, 1885–1981, in: Organization Science, Vol. 14 (2003), No. 3, pp. 264–282. Donaldson, Thomas and Lee E. Preston: The Stakeholder Theory of the Corporation: Concepts, Evidence, and Implications, in: The Academy of Management Review, Vol. 20 (1995), No. 1, pp. 65–91. Doppler, Klaus and Christoph Lauterburg: Managing Corporate Change, Berlin [et al.] 2001. Dörner, Dietrich: Die Logik des Misslingens: strategisches Denken in komplexen Situationen, 5. extended Ed., Reinbek bei Hamburg 2003. Dresing, Thorsten and Udo Kuckartz: Neue Datenquellen für die Sozialforschung: Analyse von Internetdaten, in: Kuckartz, Udo, Heiko Grunenberg and Thorsten Dresing (Ed.): Qualitative Datenanalyse: computergestützt, 2. Ed., Wiesbaden 2007, pp. 143–162. Driedonks, Caroline, Shirley Gregor, Arjen Wassenaar and Eric Van Heck: Economic and Social Analysis of the Adoption of B2B Electronic Marketplaces: A Case Study in the Australian Beef Industry, in: International Journal of Electronic Commerce, Vol. 9 (2005), No. 3, pp. 49–72. Durand, Rodolphe: Predicting a Firm's Forecasting Ability: The Roles of Organizational Illusion of Control and Organizational Attention, in: Strategic Management Journal, Vol. 24 (2003), No. 9, pp. 821–838. Duriau, Vincent J., Rhonda K. Reger and Michael D. Pfarrer: A Content Analysis of the Content Analysis Literature in Organization Studies: Research Themes, Data Sources, and Methodological Refinements, in: Organizational Research Methods, Vol. 10 (2007), No. 1, pp. 5–34. Dwyer, Paula, Amy Borrus and Gary Weiss: Big Bang at The Big Board, in: BusinessWeek, (2004), issued 16 Feb. 2004, pp. 66–67. Eaton, Judy and C. Ward Struthers: Using the Internet for Organizational Research: A Study of Cynicism in the Workplace, in: CyberPsychology & Behavior, Vol. 5 (2002), No. 4, pp. 305–313. The Economist: Ahead of the tape; Algorithmic trading, in: The Economist, Vol. 383 (2007), No. 8534, issued 23 Jun. 2007, p. 99. Eesley, Charles and Michael J. Lenox: Firm responses to secondary stakeholder action, in: Strategic Management Journal, Vol. 27 (2006), No. 8, pp. 765–781.
234
References
Eggers, J. P. and Sarah Kaplan: Cognition and Renewal: Comparing CEO and Organizational Effects on Incumbent Adaptation to Technical Change, in: Organization Science, Vol. 20 (2009), No. 2, pp. 461–477. Eisenhardt, Kathleen M.: Making Fast Strategic Decisions in High-Velocity Environments, in: The Academy of Management Journal, Vol. 32 (1989), No. 3, pp. 543–576. Elbanna, Said and John Child: The Influence of Decision, Environmental and Firm Characteristics on the Rationality of Strategic Decision-Making, in: Journal of Management Studies, Vol. 44 (2007), No. 4, pp. 561–591. Ellul, Andrew, Craig W. Holden, Pankaj Jain and Robert Jennings: Order dynamics: Recent evidence from the NYSE, in: Journal of Empirical Finance, Vol. 14 (2007), No. 5, pp. 636–661. Elster, Jon: Ulysses and the Sirens: Studies in Rationality and Irrationality, rev. Ed., Cambridge [et al.] 1984. Elster, Jon: Ulysses Unbound: Studies in Rationality, Precommitment, and Constraints, New York [et al.] 2000. Ewing, J. A.: On the Production of Transient Electric Currents in Iron and Steel Conductors by Twisting them when Magnetised or by Magnetising them when Twisted, in: Proceedings of the Royal Society of London, Vol. 33 (1881), pp. 21–23. Ewing, Jack: When CEOs Talk to Each Other, Business Week Online, issued 31 Jan. 2005, 2005, accessed 2 Apr. 2008: http://www.businessweek.com/bwdaily/ dnflash/jan2005/nf20050131_5572.htm. Feldman, Martha S.: Organizational Routines as a Source of Continuous Change, in: Organization Science, Vol. 11 (2000), No. 6, pp. 611–629. Feldman, Martha S. and Brian T. Pentland: Reconceptualizing Organizational Routines as a Source of Flexibility and Change, in: Administrative Science Quarterly, Vol. 48 (2003), No. 1, pp. 94–118. Feldman, Stuart: Electronic Marketplaces, in: IEEE Internet Computing, Vol. 4 (2000), No. 4, pp. 93–95. Fernández Gago, Roberto and Mariano Nieto Antolín: Stakeholder salience in corporate environmental strategy, in: Corporate Governance, Vol. 4 (2004), No. 3, pp. 65–76. Fichman, Mark and Daniel A. Levinthal: Honeymoons and the Liability of Adolescence: A New Perspective on Duration Dependence in Social and Organizational Relationships, in: The Academy of Management Review, Vol. 16 (1991), No. 2, pp. 442–468. Finkelstein, Sydney, Donald C. Hambrick and Albert A. Cannella Jr.: Strategic Leadership: Theory and Research on Executives, Top Management Teams, and Boards, New York [et al.] 2008. Finkelstein, Sydney, Jo Whitehead and Andrew Campbell: Think Again: Why Good Leaders Make Bad Decisions and How to Keep it From Happening to You, Boston 2009.
References
235
Ford, Jeffrey D., Laurie W. Ford and Angelo D'Amelio: Resistance to Change: The Rest of the Story, in: Academy of Management Review, Vol. 33 (2008), No. 2, pp. 362–377. Forrester, Jay W.: Industrial Dynamics, Cambridge, MA 1961. Forrester, Jay W.: Industrial Dynamics—A Response to Ansoff and Slevin, in: Management Science, Vol. 14 (1968a), No. 9, pp. 601–618. Forrester, Jay W.: Industrial Dynamics—After the First Decade, in: Management Science, Vol. 14 (1968b), No. 7, pp. 398–415. Forrester, Jay W.: Principles of Systems, Cambridge, MA 1968c. Forrester, Jay W.: System dynamics, systems thinking, and soft OR, in: System Dynamics Review, Vol. 10 (1994), No. 2/3, pp. 245–256. Forrester, Jay W. and Peter M. Senge: Tests for Building Confidence in System Dynamics Models, in: TIMS Studies in the Management Sciences, Vol. 14 (1980), pp. 209–228. Freeman, R. Edward: Strategic Management: A Stakeholder Approach, Boston, London 1984. Friend, John K. and Allen Hickling: Planning under pressure: the strategic choice approach, 3. Ed., Oxford 2005. Gasparino, Charles: King of the Club: Richard Grasso and the Survival of the New York Stock Exchange, New York 2007. Gavetti, Giovanni: Cognition and Hierarchy: Rethinking the Microfoundations of Capabilities' Development, in: Organization Science, Vol. 16 (2005), No. 6, pp. 599–617. Gavetti, Giovanni, Daniel A. Levinthal and William Ocasio: Neo-Carnegie: The Carnegie School's Past, Present, and Reconstructing for the Future, in: Organization Science, Vol. 18 (2007), No. 3, pp. 523–536. Gavetti, Giovanni and Jan W. Rivkin: On the Origin of Strategy: Action and Cognition over Time, in: Organization Science, Vol. 18 (2007), No. 3, pp. 420–439. Geisst, Charles, R.: Wall Street: A History: From Its Beginnings to the Fall of Enron, rev. and expand. Ed., Oxford [et al.] 2004. Gersick, Connie J. G.: Revolutionary Change Theories: A Multilevel Exploration of the Punctuated Equilibrium Paradigm, in: The Academy of Management Review, Vol. 16 (1991), No. 1, pp. 10–36. Gibbons, Tracy C.: DEC's "Other" Legacy: The Development of Leaders, in: Schein, Edgar H. (Ed.): DEC is Dead, Long Live DEC: The Lasting Legacy of Digital Equipment Corporation, San Francisco, CA 2003, pp. 90–112. Giddens, Anthony: The Construction of Society: Outline of the Theory of Structuration, Berkeley, CA [et al.] 1984. Gigerenzer, Gerd, Peter M. Todd and the ABC Research Group: Simple Heuristics That Make Us Smart, New York 1999. Gilbert, Clark G.: Unbundling the Structure of Inertia: Resource Versus Routine Rigidity, in: Academy of Management Journal, Vol. 48 (2005), No. 5, pp. 741–763.
236
References
Gioia, Dennis A. and Evelyn Pitre: Multiparadigm Perspectives on Theory Building, in: The Academy of Management Review, Vol. 15 (1990), No. 4, pp. 584–602. Glance, Natalie S., Matthew Hurst and Takashi Tomokiyo: BlogPulse: Automated Trend Discovery for Weblogs, in Proceedings of the WWW 2004 Workshop on the Weblogging Ecosystem: Aggregation, Analysis and Dynamics, New York, NY 2004, pp. 1–8. Glaser, Barney G. and Anselm L. Strauss: The Discovery of Grounded Theory: Strategies for Qualitative Research, Chicago 1967. Gonnard, Eric, Eun Jung Kim and Isabelle Ynesta: Recent Trends in Institutional Investors Statistics, in: OECD: Financial Market Trends (2008), No. 95. González-Benito, Javier and Óscar González-Benito: The role of stakeholder pressure and managerial values in the implementation of environmental logistics practices, in: International Journal of Production Research, Vol. 44 (2006), No. 7, pp. 1353–1373. Gorham, Michael and Nidhi Singh: Electronic Exchanges: The Global Transformation from Pits to Bits, Amsterdam [et al.] 2009. Gosling, Samuel D., Simine Vazire, Sanjay Srivastava and Oliver P. John: Should We Trust Web-Based Studies? A Comparative Analysis of Six Preconceptions About Internet Questionnaires, in: American Psychologist, Vol. 59 (2004), No. 2, pp. 93–104. Greenwood, Royston and C. R. Hinings: Understanding Strategic Change: The Contribution of Archetypes, in: Academy of Management Journal, Vol. 36 (1993), No. 5, pp. 1052–1081. Greenwood, Royston and C. R. Hinings: Understanding Radical Organizational Change: Bringing Together the Old and the New Institutionalism, in: Academy of Management Review, Vol. 21 (1996), No. 4, pp. 1022–1054. Grégoire, Denis A., Pamela S. Barr and Dean A. Shepherd: Cognitive Processes of Opportunity Recognition: The Role of Structural Alignment, in: Organization Science, Vol. 21 (2010), No. 2, pp. 413–431. Greve, Henrich R.: Performance, Aspirations, and Risky Organizational Change, in: Administrative Science Quarterly, Vol. 43 (1998), No. 1, pp. 58–86. Greve, Henrich R.: Sticky Aspirations: Organizational Time Perspective and Competitiveness, in: Organization Science, Vol. 13 (2002), No. 1, pp. 1–17. Größler, Andreas: Struktur und Dynamik strategischer Fähigkeiten in der Produktion, Mannheim 2007. Größler, Andreas: Modelltests, in: Strohhecker, Jürgen (Ed.): System Dynamics für die Finanzindustrie, Frankfurt a. M. 2008a, pp. 253–269. Größler, Andreas: System Dynamics Modelling as an Inductive and Deductive Endeavour: Comment on the Paper by Schwaninger and Grösser, in: Systems Research and Behavioral Science, Vol. 25 (2008b), No. 4, pp. 467–470. Größler, Andreas, André Grübner and Peter M. Milling: Organisational adaptation processes to external complexity, in: International Journal of Operations & Production Management, Vol. 26 (2006), No. 3/4, pp. 254–281.
References
237
Größler, Andreas, Peter Milling and Graham Winch: Perspectives on rationality in system dynamics—a workshop report and open research questions, in: System Dynamics Review, Vol. 20 (2004), No. 1, pp. 75–87. Größler, Andreas, Jörn-Henrik Thun and Peter M. Milling: System Dynamics as a Structural Theory in Operations Management, in: Production and Operations Management, Vol. 17 (2008), No. 3, pp. 373–384. Gutierrez, Jose A. and Yiuman Tse: NYSE execution quality subsequent to migration to hybrid, in: Review of Quantitative Finance and Accounting, Vol. 33 (2009), No. 1, pp. 59–81. Hambrick, Donald C. and Phyllis A. Mason: Upper Echelons: The Organization as a Reflection of Its Top Managers, in: The Academy of Management Review, Vol. 9 (1984), No. 2, pp. 193–206. Hamilton, Martha M.: Philadelphia Exchange Booming: Philadelphia Expands Option, in: The Washington Post (1981), issued 4 Aug. 1981, p. D6. Handa, Puneet, Robert A Schwartz and Ashish Tiwari: Price Improvement and Price Discovery on a Primary Market: Evidence from the American Stock Exchange, in: Journal of Portfolio Management, Vol. 25 (1999), No. 3, p. 55. Handa, Puneet, Robert A. Schwartz and Ashish Tiwari: The Economic Value of a Trading Floor: Evidence from the American Stock Exchange, in: The Journal of Business, Vol. 77 (2004), No. 2, pp. 331–355. Hannan, Michael T. and John Freeman: The Population Ecology of Organizations, in: The American Journal of Sociology, Vol. 82 (1977), No. 5, pp. 929–964. Hannan, Michael T. and John Freeman: Structural Inertia and Organizational Change, in: American Sociological Review, Vol. 49 (1984), No. 2, pp. 149–164. Hannan, Michael T. and John Freeman: Organizational Ecology, Cambridge, MA [et al.] 1989. Harvey, Brian and Anja Schaefer: Managing Relationships with Environmental Stakeholders: A Study of U.K. Water and Electricity Utilities, in: Journal of Business Ethics, Vol. 30 (2001), No. 3, pp. 243–260. Hasan, Iftekhar, Markku Malkamaki and Heiko Schmiedel: Technology, automation, and productivity of stock exchanges: International evidence, in: Journal of Banking & Finance, Vol. 27 (2003), No. 9, pp. 1743–1773. Hasbrouck, Joel, George Sofianos and Deborah Sosebee: New York Stock Exchange Systems and Trading Procedures, NYSE Working Paper 1993. Haveman, Heather A.: Between a Rock and a Hard Place: Organizational Change and Performance Under Conditions of Fundamental Environmental Transformation, in: Administrative Science Quarterly, Vol. 37 (1992), No. 1, pp. 48–75. Haveman, Heather A., Michael V. Russo and Alan D. Meyer: Organizational Environments in Flux: The Impact of Regulatory Punctuations on Organizational Domains, CEO Succession, and Performance, in: Organization Science, Vol. 12 (2001), No. 3, pp. 253–273. Hawthorne, Fran: What the SEC rules do for activism, in: Institutional Investor, Vol. 27 (1993), No. 4, pp. 47–54.
238
References
Hendershott, Terrence and Pamela C. Moulton: The Shrinking New York Stock Exchange Floor and the Hybrid Market, Working Paper, 2007. Hendershott, Terrence and Pamela C. Moulton: Speed and Stock Market Quality: The NYSE's Hybrid, Working Paper, 2009. Hill, Charles W. L. and Frank T. Rothaermel: The Performance of Incumbent Firms in the Face of Radical Technological Innovation, in: The Academy of Management Review, Vol. 28 (2003), No. 2, pp. 257–274. Hodgkinson, Gerard P.: Cognitive Inertia in a Turbulent Market: the Case of UK Residential Estate Agents, in: Journal of Management Studies, Vol. 34 (1997), No. 6, pp. 921–945. Hoff, Timothy J. and David P. McCaffrey: Adapting, Resisting, and Negotiating: How Physicians Cope With Organizational and Economic Change, in: Work and Occupation, Vol. 23 (1996), No. 2, pp. 165–189. Hough, Jill R. and Margaret A. White: Environmental dynamism and strategic decision-making rationality: an examination at the decision level, in: Strategic Management Journal, Vol. 24 (2003), No. 5, pp. 481–489. Hrebiniak, Lawrence G. and William F. Joyce: Organizational Adaptation: Strategic Choice and Environmental Determinism, in: Administrative Science Quarterly, Vol. 30 (1985), No. 3, pp. 336–349. Huang, Roger D. and Hans R. Stoll: Dealer versus auction markets: A paired comparison of execution costs on NASDAQ and the NYSE, in: Journal of Financial Economics, Vol. 41 (1996), No. 3, pp. 313–357. Huber, George P. and Kyle Lewis: Cross-understanding: Implications for Group Cognition and Performance, in: Academy of Management Review, Vol. 35 (2010), No. 1, pp. 6–26. Huff, Anne S., Joseph W. Huff and Pamela S. Barr: When Firms Change Direction, New York, NY 2000. Ingram, Paul and Joel A. C. Baum: Opportunity and Constraint: Organizations' Learning from the Operating and Competitive Experience of Industries, in: Strategic Management Journal, Vol. 18 (1997), No. -, Special Issue on Organizational and Competitive Interactions, pp. 75–98. Jackson, Norman and Pippa Carter: In Defence of Paradigm Incommensurability, in: Organization Studies, Vol. 12 (1991), No. 1, pp. 109–127. Jain, Pankaj K.: Financial Market Design and the Equity Premium: Electronic versus Floor Trading, in: Journal of Finance, Vol. 60 (2005), No. 6, pp. 2955–2985. Jansen, Karen J.: From Persistence to Pursuit: A Longitudinal Examination of Momentum During the Early Stages of Strategic Change, in: Organization Science, Vol. 15 (2004), No. 3, pp. 276–294. Jiang, Christine, Thomas McInish and James Upson: The information content of trading halts, in: Journal of Financial Markets, Vol. 12 (2009), No. 4, pp. 703–726. Judson, Arnold S.: Changing Behavior in Organizations: Minimizing Resistance to Change, rev. Ed., Cambridge, Mass. 1991.
References
239
Kahneman, Daniel, Jack L. Knetsch and Richard H. Thaler: Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias, in: Journal of Economic Perspectives, Vol. 5 (1991), No. 1, pp. 193–206. Kahneman, Daniel and Amos Tversky: Prospect Theory: An Analysis of Decision under Risk, in: Econometrica, Vol. 47 (1979), No. 2, pp. 263–292. Kampas, Paul J.: The Impact of Changing Technology, in: Schein, Edgar H. (Ed.): DEC is Dead, Long Live DEC: The Lasting Legacy of Digital Equipment Corporation, San Francisco, CA 2003, pp. 131–145. Kaplan, Sarah: Cognition, Capabilities, and Incentives: Assessing Firm Response to the Fiber-Optic Revolution, in: Academy of Management Journal, Vol. 51 (2008), No. 4, pp. 672–695. Kaplan, Sarah, Fiona Murray and Rebecca Henderson: Discontinuities and senior management: assessing the role of recognition in pharmaceutical firm response to biotechnology, in: Industrial and Corporate Change, Vol. 12 (2003), No. 2, pp. 203–233. Kaplan, Sarah and Mary Tripsas: Thinking about technology: Applying a cognitive lens to technical change, in: Research Policy, Vol. 37 (2008), No. 5, pp. 790–805. Keith, Christopher and Allan Grody: Electronic Automation at the New York Stock Exchange, in: Guile, Bruce R. and James Brian Quinn (Ed.): Managing Innovation: Cases from the Services Industries, Washington, DC 1988. Kelly, Dawn and Terry L. Amburgey: Organizational Inertia and Momentum: A Dynamic Model of Strategic Change, in: Academy of Management Journal, Vol. 34 (1991), No. 3, pp. 591–612. Kim, Daniel H.: The Link between Individual and Organizational Learning, in: Sloan Management Review, Vol. 35 (1993), No. 1, pp. 37–50. Kim, Hyunjung: Broadening Boundary Perception in a Multi-organizational Context: Study of a Community Mental Health Program in New York State, in Dangerfield, Brian C. (Ed.): Proceedings of the 26th International Conference of the System Dynamics Society, Athens 2008, pp. 1–16. Kim, Hyunjung: In search of a mental model-like concept for group-level modeling, in: System Dynamics Review, Vol. 25 (2009), No. 3, pp. 207–223. Kim, Kendall: Electronic and Algorithmic Trading Technology: The Complete Guide, in: Kaljuvee, Ayesha and Jürgen Kaljuvee: Complete Technology Guides for Financial Services Series, Amsterdam [et al.] 2007. Kim, Kong-Hee, G. Tyge Payne and James A. Tan: An examination of cognition and effect in strategic decision making, in: International Journal of Organizational Analysis, Vol. 14 (2006), No. 4, pp. 277–294. Kimberly, John R. and Robert H. Miles: The Organizational Life Cycle: Issues in the Creation, Transformation, and Decline of Organizations, in: The Jossey-Bass Social and Behavioral Science Series, San Francisco, CA 1980. Klein, Katherine J. and Steve W. J. Kozlowski: From Micro to Meso: Critical Steps in Conceptualizing and Conducting Multilevel Research, in: Organizational Research Methods, Vol. 3 (2000), No. 3, pp. 211–236.
240
References
Knox, Simon and Colin Gruar: The Application of Stakeholder Theory to Relationship Marketing Strategy Development in a Non-profit Organization, in: Journal of Business Ethics, Vol. 75 (2007), No. 2, pp. 115–135. Kopainsky, Birgit and Luis Felipe Luna-Reyes: Closing the Loop: Promoting Synergies with Other Theory Building Approaches to Improve System Dynamics Practice, in: Systems Research and Behavioral Science, Vol. 25 (2008), No. 4, pp. 471–486. Kopainsky, Birgit, Matteo Pedercini, Pål I. Davidsen and Stephen M. Alessi: A Blend of Planning and Learning: Simplifying a Simulation Model of National Development, in: Simulation & Gaming, Vol. 4 (2010), No. 5, pp. 641–662. Kuhn, Thomas S.: The Structure of Scientific Revolutions, 2. Ed., Chicago 1970. LaBerge, David: Attentional Processing: The Brain's Art of Mindfulness, Perspectives in Cognitive Neuroscience, 1995. Lamme, Victor A. F.: Why visual attention and awareness are different, in: Trends in Cognitive Sciences, Vol. 7 (2003), No. 1, pp. 12–18. Lane, David C.: Can We Have Confidence in Generic Structures?, in: The Journal of the Operational Research Society, Vol. 49 (1998), No. 9, pp. 936–947. Lane, David C.: Should System Dynamics be Described as a 'Hard' or 'Deterministic' Systems Approach?, in: Systems Research and Behavioral Science, Vol. 17 (2000), No. 1, pp. 3–22. Lane, David C.: Rerum cognoscere causas: Part I—How do the ideas of system dynamics relate to traditional social theories and the voluntarism/determinism debate?, in: System Dynamics Review, Vol. 17 (2001), No. 2, pp. 97–118. Lane, David C. and Chris Smart: Reinterpreting 'generic structure': Evolution, application and limitations of a concept, in: System Dynamics Review, Vol. 12 (1996), No. 2, pp. 87–120. Lant, Theresa K.: Aspiration Level Adaptation: An Empirical Exploration, in: Management Science, Vol. 38 (1992), No. 5, pp. 623–644. Lant, Theresa K. and Stephen J. Mezias: An Organizational Learning Model of Convergence and Reorientation, in: Organization Science, Vol. 3 (1992), No. 1, pp. 47–71. Larsen, Erik R. and Alessandro Lomi: Resetting the clock: a feedback approach to the dynamics of organisational inertia, survival and change, in: Journal of the Operational Research Society, Vol. 50 (1999), No. 4, pp. 406–421. Larsen, Erik R. and Alessandro Lomi: Representing change: a system model of organizational inertia and capabilities as dynamic accumulation process, in: Simulation Modelling Practice and Theory, Vol. 10 (2002), No. 5–7, pp. 271–296. Latour, Bruno: We Have Never Been Modern, Cambridge, MA 1993. Lawrence, Paul R.: How to Deal with Resistance to Change, in: Harvard Business Review, Vol. 32 (1954), No. 3, pp. 49–57. Leonard-Barton, Dorothy: Core Capabilities and Core Rigidities: A Paradox in Managing New Product Development, in: Strategic Management Journal, Vol. 13 (1992), No. 5, pp. 111–125.
References
241
Levinthal, Daniel A.: Organizational Adaptation and Environmental SelectionInterrelated Processes of Change, in: Organization Science, Vol. 2 (1991), No. 1, pp. 140–145. Levitt, Barbara and James G. March: Organizational Learning, in: Annual Review of Sociology, Vol. 14 (1988), pp. 319–340. Lewin, Kurt: Field Theory in Social Science: Selected Theoretical Papers, in: Cartwright, Dorwin, New York 1951. Lewis, Marianne W. and Andrew J. Grimes: Metatriangulation: Building Theory from Multiple Paradigms, in: The Academy of Management Review, Vol. 24 (1999), No. 4, pp. 672–690. Lewis, Marianne W. and Mihaela L. Kelemen: Multiparadigm inquiry: Exploring organizational pluralism and paradox, in: Human Relations, Vol. 55 (2002), No. 2, pp. 251–275. Lieberman, Marvin B. and David B. Montgomery: First-Mover Advantages, in: Strategic Management Journal, Vol. 9 (1988), pp. 41. Lieberman, Marvin B. and David B. Montgomery: First-Mover (Dis)Advantages: Retrospective and Link with the Resource-Based View, in: Strategic Management Journal, Vol. 19 (1998), No. 12, pp. 1111–1125. Lo, Andrew W. and A. Craig MacKinlay: A Non-Random Wald Down Wall Street, Princeton, NJ 2002. Lucas Jr., Henry C., Wonseok Oh and Bruce W. Weber: The defensive use of IT in a newly vulnerable market: The New York Stock Exchange, 1980–2007, in: The Journal of Strategic Information Systems, Vol. 18 (2009), No. 1, pp. 3–15. Lucchetti, Aaron: Nasdaq Passes NYSE in Sales Milestone, in: Wall Street Journal Eastern Edition (2008a), issued 14 Jul. 2008, p. C.3. Lucchetti, Aaron: NYSE Plans to Revise Specialist-Trader Rules, in: Wall Street Journal - Eastern Edition (2008b), issued 13 Jun. 2008, p. C.4. Lucchetti, Aaron and Alistair MacDonald: Stock-Exchanges Grudge Match: Old Rivals NYSE and Nasdaq Use Bids for European Markets As Latest Way to Pin Their Foe, in: Wall Street Journal - Eastern Edition (2006), issued 8 Sep. 2006, p. C.1. Luna-Reyes, Luis Felipe and Deborah Lines Andersen: Collecting and analyzing qualitative data for system dynamics: methods and models, in: System Dynamics Review, Vol. 19 (2003), No. 4, pp. 271–296. Macrì, Diego Maria, Maria Rita Tagliaventi and Fabiola Bertolotti: A grounded theory for resistance to change in a small organization, in: Journal of Organizational Change Management, Vol. 15 (2002), No. 3, pp. 292–310. Madhavan, Ananth: Consolidation, Fragmentation, and the Disclosure of Trading Information, in: The Review of Financial Studies, Vol. 8 (1995), No. 3, pp. 579–603. Madhavan, Ananth: Market microstructure: A survey, in: Journal of Financial Markets, Vol. 3 (2000), No. 3, pp. 205–258.
242
References
Madhavan, Ananth and George Sofianos: An empirical analysis of NYSE specialist trading, in: Journal of Financial Economics, Vol. 48 (1998), No. 2, pp. 189–210. Magness, Vanessa: Who are the Stakeholders Now? An Empirical Examination of the Mitchell, Agle, and Wood Theory of Stakeholder Salience, in: Journal of Business Ethics, Vol. 83 (2008), No. 2, pp. 177–192. Malone, Thomas W., Joanne Yates and Robert I. Benjamin: Electronic Markets and Electronic Hierarchies, in: Allen, Thomas J. and Michael S. Scott Morton (Ed.): Information Technology and the Corporation of the 1990s: Research Studies, New York and Oxford 1994, pp. 61–83, originally published in: Communications of the ACM, Vol. 30 (1987), No. 6, pp. 484-497. March, James G.: Footnotes to Organizational Change, in: Administrative Science Quarterly, Vol. 26 (1981), No. 4, pp. 563–577. March, James G. and Johan P. Olsen: The Uncertainty of the Past: Organizational Learning Under Ambibuity, in: European Journal of Political Research, Vol. 3 (1975), No. 2, pp. 147–171. March, James G. and Johan P. Olsen: Ambiguity and Choice in Organizations, Bergen [et al.] 1976. March, James G. and Herbert A. Simon: Organizations, New York [et al.] 1958. March, James G., with the assistance of Chip Heath: A Primer on Decision Making: How Decisions Happen, New York, NY [et al.] 1994. Martin, Richard: Business At Light Speed, in: InformationWeek, (2007), No. 1135, issued 23 Apr. 2007, pp. 42–47. McAndrews, James and Chris Stefanadis: The Emergence of Electronic Communication Networks in the U.S. Equity Markets, in: Federal Reserve Bank of New York: Current Issues in Economics and Finance, Vol. 6 (2000), No. 12, pp. 1–6. McGee, Suzanne: Up Off the Floor, in: Institutional Investor, Vol. 42 (2008), No. 10, pp. 38–42. Mehta, Nina: Traders Sound Off on the NYSE's Hybrid Market, in: Traders Magazine, Vol. 20 (2007), No. 270, issued 15 Jul. 2007, pp. 20–28. Mellahi, Kamel: The Dynamics of Boards of Directors in Failing Organizations, in: Long Range Planning, Vol. 38 (2005), No. 3, Special Issue: Organizational Failure, pp. 261–279. Mellahi, Kamel and Adrian Wilkinson: Organizational failure: a critique of recent research and a proposed integrative framework, in: International Journal of Management Reviews, Vol. 5/6 (2004), No. 1, pp. 21–41. Merten, Peter P.: Loop-based Strategic Decision Support Systems, in: Strategic Management Journal, Vol. 12 (1991), No. 5, pp. 371–386. Meyer, Alan D., Geoffrey R. Brooks and James B. Goes: Environmental Jolts and Industry Revolutions: Organizational Responses to Discontinuous Change, in: Strategic Management Journal, Vol. 11 (1990), No. -, Special Issue: Corporate Entrepreneurship, pp. 93–110. Michie, Ranald C.: The London Stock Exchange: A History, Oxford, New York 1999.
References
243
Miczka, Switbert, Nicole Zimmermann, David Andersen and George Richardson: Walk the Talk: Implementing Ethical Conduct in Industrial Production Systems, in: Strohhecker, Jürgen and Andreas Größler (Ed.): Strategisches und operatives Produktionsmanagement: Empirie und Simulation, Wiesbaden 2009, pp. 91–114. Miller, Danny and Ming-Jer Chen: Sources and Consequences of Competitive Inertia: A Study of the U.S. Airline Industry, in: Administrative Science Quarterly, Vol. 39 (1994), No. 1, pp. 1–23. Milling, Peter: Der technische Fortschritt beim Produktionsprozeß: Ein dynamisches Modell für innovative Industrieunternehmen, Wiesbaden 1974. Milling, Peter: Systemtheoretische Grundlagen zur Planung der Unternehmenspolitik, Berlin 1981. Milling, Peter: Leitmotive des System-Dynamics-Ansatzes, in: Wirtschaftswissenschaftliches Studium (WiSt), Vol. 13 (1984), No. 10, pp. 507–513. Milling, Peter: Business Systems as Control Systems, in: D'Amato, Vittorio and Carlo Maccheroni (Ed.): Dynamic Analysis of Complex Systems, Milano 1989, pp. 27–34. Milling, Peter: Kybernetische Überlegungen beim Entscheiden in komplexen Situationen, in: Milling, Peter (Ed.): Entscheiden in komplexen Situationen, Berlin 2002, pp. 11–26. Milling, Peter M.: Modeling innovation processes for decision support and management simulation, in: System Dynamics Review, Vol. 12 (1996), No. 3, pp. 211–234. Milling, Peter M.: Understanding and managing innovation processes, in: System Dynamics Review, Vol. 18 (2002), No. 1, pp. 73–86. Milling, Peter M.: Afterword, in: Systems Research and Behavioral Science, Vol. 25 (2008), No. 4, pp. 581–582. Milling, Peter M. and Nicole S. Zimmermann: Modeling Drivers of Organizational Change, in: Kybernetes, Vol. 39 (2010), No. 9/10, pp. 1452–1490. Mintzberg, Henry, Joseph Lampel and Bruce Ahlstrand: Strategy Safari: A Guided Tour Through The Wilds of Strategic Management, New York, NY 1998. Mitchell, Ronald K., Bradley R. Agle and Donna J. Wood: Toward a Theory of Stakeholder Identification and Salience: Defining the Principle of Who and What Really Counts, in: The Academy of Management Review, Vol. 22 (1997), No. 4, pp. 853–886. Mollona, Edoardo: A Competence View of Firms as Resource Accumulation Systems: A Synthesis of Resource-Based and Evolutionary Models of StrategyMaking, in: Morecroft, John, Ron Sanchez and Aimé Heene (Ed.): Systems Perspectives on Resources, Capabilities, and Management Processes, Amsterdam [et al.] 2002, pp. 93–125. Moore, Don A., John M. Oesch and Charlene Zietsma: What Competition? Myopic Self-Focus in Market-Entry Decisions, in: Organization Science, Vol. 18 (2007), No. 3, pp. 440–454. Morgan, Gareth: Paradigms, Metaphors, and Puzzle Solving in Organization Theory, in: Administrative Science Quarterly, Vol. 25 (1980), No. 4, pp. 605–622.
244
References
Morgan, Gareth: Images of Organization, updated Ed., Thousand Oaks, CA 2006. Morgan, Gareth and Linda Smircich: The Case for Qualitative Research, in: The Academy of Management Review, Vol. 5 (1980), No. 4, pp. 491–500. Murray, Alan I.: Top Management Group Heterogeneity and Firm Performance, in: Strategic Management Journal, Vol. 10 (1989), No. -, Special Issue, pp. 125–141. Nadkarni, Sucheta and Pamela S. Barr: Environmental Context, Managerial Cognition, and Strategic Action: An Integrated View, in: Strategic Management Journal, Vol. 29 (2008), No. 13, pp. 1395–1427. Nadler, David A. and Michael L. Tushman: Types of Organizational Change: From Incremental Improvement to Discontinuous Transformation, in: Nadler, David A., et al. (Ed.): Discontinuous Change: Leading Organizational Transformation, San Francisco 1995, pp. 15–34. Naidu, G. N. and Michael S. Rozeff: Volume, volatility, liquidity and efficiency of the Singapore Stock Exchange before and after automation, in: Pacific-Basin Finance Journal, Vol. 2 (1994), No. 1, pp. 23–42. Naranjo-Gil, David and Frank Hartmann: Management accounting systems, top management team heterogeneity and strategic change, in: Accounting, Organizations and Society, Vol. 32 (2007), No. 7/8, pp. 735–756. Naranjo-Gil, David, Frank Hartmann and Victor S. Maas: Top Management Team Heterogeneity, Strategic Change and Operational Performance, in: British Journal of Management, Vol. 19 (2008), No. 3, pp. 222–234. NASDAQ OMX Group Inc.: NYSE Market Share in NYSE-Listed Securities, [Excel Chart], no date, accessed 29 Sep. 2009: http://www.nasdaqtrader.com/ content/marketstatistics/marketshare/floor.xls, accessible via http://www.nasdaqtrader.com/trader.aspx?id=TradingNYSE. National Stock Exchange (NSX): History, no date, accessed 25 Aug. 2009: http://www.nsx.com/content/history. Nelson, Richard R. and Sidney G. Winter: An evolutionary theory of economic change, Cambridge, MA [et al.] 1982. Newton, Lisa H.: Truth is the Daughter of Time: The Real Story of the Nestle Case, in: Business and Society Review, Vol. 104 (1999), No. 4, pp. 367–395. Niederauer, Duncan: Statement 17 in Questions and Answers, in: Thomson Financial: NYX - Q2 2007 NYSE Euronext Earnings Conference Call (2 Aug. 2007), [Conference Call Transcript], 2007a, accessed 24 Feb. 2009: http://library.corporate-ir.net/library/12/129/129145/items/256935/NYXTranscript-2007-08-02T12-00.pdf. Niederauer, Duncan: Statement in speech on U.S. Cash Equities, in: Thomson Financial: NYX - NYSE Euronext Analyst Day, Corporate Analyst Meeting (6 Jun. 2007), [Conference Call Transcript], 2007b, accessed 24 Feb. 2009: http://library.corporate-ir.net/library/12/129/129145/items/250134/ 2007_Analyst_Day_Transcript.pdf. North, Douglass C.: Institutions, institutional change and economic performance, Cambridge, MA [et al.] 1990.
References
245
NYSE Euronext Inc.: Hybrid Market: Key Attributes, 2007, accessed 30 Sep. 2007: http://www.nyse.com/productservices/nyseequities/1126821290345.html. NYSE Euronext Inc.: Annual Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 for the fiscal year ended December 31, 2007, No. 001-368007, New York 2008a. NYSE Euronext Inc.: New NYSE Order Imbalances Product Goes Live Today, in: News Releases, 1 Jul. 2008, 2008b, accessed 4 Feb. 2010: http://www.nyse.com/press/1214908476746.html. NYSE Euronext Inc.: NYSE Aims to Maximize Market Quality and Competitiveness With Newly Approved Enhancements of Trading Model, in: News Releases, 24 Oct. 2008, 2008c, accessed 11 Oct. 2009: http://www.nyse.com/press/ 1224844297395.html. NYSE Euronext Inc.: NYSE Euronext Appoints Todd B. Abrahall and Michael J. Rutigliano as Liaisons to NYSE Specialists and Brokers, in: News Releases, 10 Jan. 2008, 2008d, accessed 14 Dec. 2009: http://www.nyse.com/press/ 1199963130941.html. NYSE Euronext Inc.: Annual Report pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 for the fiscal year ended December 31, 2008, No. 1-33392, New York 2009a. NYSE Euronext Inc.: Cash Equities: Designated Market Makers, 2009b, accessed 27 May 2010: http://www.nyse.com/pdfs/fact_sheet_dmm.pdf. NYSE Euronext Inc.: NYSE Equities: NYSE Market Model Update, 2009c, accessed 27 May 2010: http://www.nyse.com/pdfs/6892_NYSE_MKT_QLTY.pdf. NYSE Euronext Inc.: Significant Progress in New NYSE Market Model, in: U.S. Equities News (2009d), issued Apr. 2009, pp. 1–2. NYSE Euronext Inc.: Annual Report pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 for the fiscal year ended December 31, 2009, No. 1-33392, New York 2010a. NYSE Euronext Inc.: NYSE Arca Sub-millisecond Execution Speed with Universal Trading Platform, in: U.S. Equities News (2010b), issued Mar. 2010, p. 1. NYSE Euronext Inc.: Annual reported volume, turnover rate, reported trades (mils. of shares), no date-a, accessed 4 Dec. 2009: http://www.nyxdata.com/factbook/ (Facts and Figures > Historical > Annual reported volume, turnover rate, reported trades (mils. of shares)). NYSE Euronext Inc.: Average NYSE Spreads, (in dollars, rounded to penny) (1994–2003), no date-b, accessed 4 Dec. 2009: http://www.nyxdata.com/ factbook/ (Facts and Figures > Historical > Average NYSE spreads, (in dollars, rounded to penny) (1994-2003)). NYSE Euronext Inc.: Chronology of the New York Stock Exchange (1930–1979), no date-c, accessed 27 Aug. 2009: http://www.nyxdata.com/factbook (Facts and Figures > Historical > Chronology of New York Stock Exchange (1930–1979)). NYSE Euronext Inc.: Daily NYSE Group Volume in NYSE Listed, no date-d, accessed 25 May 2010: http://www.nyxdata.com/factbook/ (Facts and Figures > Market Activity > Daily NYSE Group Volume in NYSE Listed > View All Years).
246
References
NYSE Euronext Inc.: Daily reported share volume: average, high, and low days (thous. of shares), no date-e, accessed 25 May 2010: http://www.nyxdata.com/factbook/ (Facts and Figures > Historical > Daily reported share volume: average, high, and low days (thous. of shares)). NYSE Euronext Inc.: Holdings of corporate equities in the U.S. by type of institution, no date-f, accessed 15 Feb. 2009: http://www.nyxdata.com/factbook (Facts and Figures > Institutional Investors > Holdings of corporate equities in the U.S. by type of institution). NYSE Euronext Inc.: Income statement for NYSE member firms ($ in mils.), no dateg, accessed 5 Dec. 2009: http://www.nyxdata.com/factbook/ (Facts and Figures > NYSE and Membership > Income statement for NYSE member firms ($ in mils.)). NYSE Euronext Inc.: Major sources of NYSE volume, no date-h, accessed 28 Aug. 2009: http://www.nyxdata.com/factbook (Facts and Figures> The Investing Public > Major sources of NYSE volume). NYSE Euronext Inc.: Market Share of consolidated tape volume by year (1976–2003), no date-i, accessed 13. Jan. 2009: http://www.nyxdata.com/ factbook (Facts and Figures > Historical > Market Share of consolidated tape volume by year (1976–2003)). NYSE Euronext Inc.: NYSE Glossary: Block, no date-j, accessed 27 Aug. 2009: http://www.nyse.com/glossary/Glossary.html (Glossary > Block). NYSE Euronext Inc.: NYSE Glossary: Intermarket Trading System (ITS), no date-k, accessed 11 Feb. 2008: http://www.nyse.com/glossary/Glossary.html (Glossary > Intermarket Trading System). NYSE Euronext Inc.: Specialist Activity, no date-l, accessed 2 Dec. 2009: http://www.nyxdata.com/factbook (Facts and Figures > Market Activity > Specialist Activity). NYSE Euronext Inc.: Timeline: Technology, no date-m, accessed 27 Aug. 2009: http://www.nyse.com/about/history/timeline_technology.html. NYSE Euronext Inc.: Types of Members, no date-n, accessed 24 Oct. 2007: http://www.nyse.com/productservices/nyseequities/1167954368183.html (now available with adapted content at: http://www.nyse.com/equities/nyseequities/ 1167954368183.html). NYSE Euronext Inc.: Types of Members, no date-o, accessed 11 Oct. 2009: http://www.nyse.com/equities/nyseequities/1167954368183.html. NYSE Group Inc.: Annual Report Pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 for the fiscal year ended December 31, 2005, No. 001-32829, New York 2006. NYSE Group Inc.: Annual Report pursuant to Section 13 or 15(d) of the Securities Exchange Act of 1934 for the fiscal year ended December 31, 2006, No. 001-32829, New York 2007. NYSE Regulation Inc., Market Surveillance: Hybrid Market Implementation Phase IV, in: Information Memo, No. 07-12, New York 2007. Ocasio, William: Towards an Attention-based View of the Firm, in: Strategic Management Journal, Vol. 18 (1997), No. -, Summer Special Issue, pp. 187–206.
References
247
Ocasio, William and John Joseph: An Attention-Based Theory of Strategy Formulation: Linking Micro- and Macroperspectives in Strategy Processes, in: Gabriel Szulanski, Joe Porac and Yves Doz (Ed.): Strategy Process, Advances in Strategic Management, Vol. 22, 2005, pp. 39–61. Oreg, Shaul: Personality, context, and resistance to organizational change, in: European Journal of Work & Organizational Psychology, Vol. 15 (2006), No. 1, pp. 73–101. O'Reilly III, Charles A. and Michael L. Tushman: Ambidexterity as a dynamic capability: Resolving the innovator's dilemma, in: Research in Organizational Behavior, Vol. 28 (2008), pp. 185–206. Orlikowski, Wanda J.: The Duality of Technology: Rethinking the Concept of Technology in Organizations, in: Organization Science, Vol. 3 (1992), No. 3, pp. 398–427. Ortega, Edgar: NYSE's Thain Fails to Stem Losses in Market Share (Update 2) (26 Jul. 2006), Bloomberg, 2006, accessed 10 Nov. 2009: http://www.bloomberg.com/apps/news?pid=20601103&sid=aUyW4.K5ljZA&ref er=us#. Pache, Anne-Claire and Filipe Santos: When Worlds Collide: The Internal Dynamics of Organizational Responses to Conflicting Institutional Demands, in: Academy of Management Review, Vol. 35 (2010), No. 3, pp. 455–476. Parent, Milena M. and David L. Deephouse: A Case Study of Stakeholder Identification and Prioritization by Managers, in: Journal of Business Ethics, Vol. 75 (2007), No. 1, pp. 1–23. Park, Kyung Min: Antecedents of Convergence and Divergence in Strategic Positioning: The Effects of Performance and Aspiration on the Direction of Strategic Change, in: Organization Science, Vol. 18 (2007), No. 3, pp. 386–402. Parker, Charles A.: The Literature on Planned Organizational Change: A Review and Analysis, in: Higher Education, Vol. 9 (1980), No. 4, pp. 429–442. Péli, Gábor: Fit by Founding, Fit by Adaptation: Reconciling Conflicting Organization Theories with Logical Formalization, in: Academy of Management Review, Vol. 34 (2009), No. 2, pp. 343–360. Pellecchia, Ray: I'd like an Auction Limit order with soy milk, please (2 Mar. 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2006a, accessed 15 Jan. 2009: http://exchanges.nyse.com/ archives/2006/03/ auction_limit_soy.php. Pellecchia, Ray: It's go time for go fast (6 Oct. 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2006b, accessed 15 Jan. 2009: http://exchanges.nyse.com/archives/2006/10/ go_time_for_go_fast.php. Pellecchia, Ray: On crisis and opportunity: Thomas Friedman and the Hybrid Market (1 Mar. 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets [Weblog], 2006c, accessed 10 Dec. 2009: http://exchanges.nyse.com/archives/2006/03/crisis_and_opportunity.php. Pellecchia, Ray: The right time (6 Oct. 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2006d, accessed 10 Nov. 2009: http://exchanges.nyse.com/archives/2006/05/right_time.php.
248
References
Pellecchia, Ray: WSJ: Lucent first stock to get unrestricted auto-ex in Hybrid Market (3 May 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2006e, accessed 11 Oct. 2009: http://exchanges.nyse.com/archives/2006/05/nyse_picks_lucent.php. Pellecchia, Ray: Shaping the blog to a new mark (21 May 2007), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2007, accessed 18 Oct. 2009: http://exchanges.nyse.com/archives/2007/05/ new_mark.php. Pellecchia, Ray: Comment on: Ten Things I Like About the Coming Changes at NYSE (13 Jun. 2008), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2008a, accessed 10 Oct. 2009: http://exchanges.nyse.com/archives/2008/06/ten_things.php. Pellecchia, Ray: Comment on: We, Robots: UAL Hit By Perfect Storm Where Nobody Is at the Switch (10 Sep. 2008), Pellecchia, Ray, in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2008b, accessed 10 Oct. 2009: http://exchanges.nyse.com/archives/2008/09/ we_robots.php. Pellecchia, Ray: Faster, Faster: High Tech Hits the NYSE Floor (29 Mar. 2010), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2010, accessed 19 May 2010: http://exchanges.nyse.com/archives/ 2010/03/high_tech.php. Perrow, Charles: Complex Organizations: A Critical Essay, New York [u.a.] 1986. Persky, Joseph: Retrospectives: The Ethology of Homo Economicus, in: The Journal of Economic Perspectives, Vol. 9 (1995), No. 2, pp. 221–231. Peterson, Kristina and Jacob Bunge: Getco Becomes NYSE Market Maker, in: Wall Street Journal (2010), issued 12 Feb. 2010, p. C.5. Pfeffer, Jeffrey: Managing With Power: politics and influence in organizations, Boston, Mass. 1992. Pfeffer, Jeffrey: Barriers to the Advance of Organizational Science: Paradigm Development as a Dependent Variable, in: The Academy of Management Review, Vol. 18 (1993), No. 4, pp. 599–620. Picot, Arnold, Christine Bortenlänger and Heiner Röhrl: The Automation of Capital Markets, in: Journal of Computer-Mediated Communication, Vol. 1 (1995), No. 3, 1995: http://jcmc.indiana.edu/vol1/issue3/picot.html. Picot, Arnold, Christine Bortenlänger and Heiner Röhrl: Organization of Electronic Markets: Contributions from the New Institutional Economics, in: The Information Society, Vol. 13 (1997), No. 1, pp. 107–123. Piderit, Sandy Kristin: Rethinking Resistance and Recognizing Ambivalence: A Multidimensional View of Attitudes toward an Organizational Change, in: The Academy of Management Review, Vol. 25 (2000), No. 4, pp. 783–794. Poole, Marshall Scott and Andrew H. Van de Ven: Using Paradox to Build Management and Organization Theories, in: The Academy of Management Review, Vol. 14 (1989), No. 4, pp. 562–578. Popper, Karl R.: The Logic of Scientific Discovery, reprint Ed., London, New York 2007.
References
249
Porter, Michael E.: On Competition, updated and expanded Ed., Boston, MA 2008. Post, James E.: Assessing the Nestlé Boycott: Corporate Accountability and Human Rights, in: California Management Review, Vol. 27 (1985), No. 2, pp. 113–131. Radzicki, Michael J.: Mr. Hamilton, Mr. Forrester, and a Foundation for Evolutionary Economics, in: Journal of Economic Issues, Vol. 37 (2003), No. 1, pp. 133–173. Rajagopalan, Nandini and Gretchen M. Spreitzer: Toward a Theory of Strategic Change: A Multi-lens Perspective and Integrative Framework, in: The Academy of Management Review, Vol. 22 (1997), No. 1, pp. 48–79. Randers, Jørgen: Guidelines for Model Conceptualization, in: Randers, Jørgen (Ed.): Elements of the System Dynamics Method, Cambridge, MA [et al.] 1980, pp. 117–139. Reed, Michael I.: Redirections in Organizational Analysis, London [et al.] 1985. Reihlen, Markus, Thorsten Klaas-Wissing and Torsten Ringberg: Metatheories in Management Studies: Reflections Upon Individualism, Holism, and Systemism, in: M@n@gement, Vol. 10 (2007), No. 3, pp. 49–69. Repenning, Nelson P. and John D. Sterman: Nobody Ever Gets Credit for Fixing Problems that Never Happened, in: California Management Review, Vol. 43 (2001), No. 4, pp. 64–88. Richardson, George P.: Feedback Thought in Social Science and Systems Theory, 2 Ed., Waltham, MA 1999. Richardson, George P. and Alexander L. Pugh III: Introduction to System Dynamics Modeling with DYNAMO, Cambridge, MA [et al.] 1981. Richter, Judith: Holding Corporations Accountable: Corporate Conduct, International Codes, and Citizen Action, New York 2001. Rogers, Everett M.: Diffusion of Innovations, 5. Ed., New York, NY [et al.] 2003. Romanelli, Elaine and Michael L. Tushman: Organizational Transformation as Punctuated Equilibrium: An Empirical Test, in: Academy of Management Journal, Vol. 37 (1994), No. 5, pp. 1141–1166. Romano Jr., Nicholas C., Christina Donovan, Hsinchun Chen and Jay F. Nunamaker Jr.: A Methodology for Analyzing Web-Based Qualitative Data, in: Journal of Management Information Systems, Vol. 19 (2003), No. 4, pp. 213–246. de Rond, Mark and Raymond-Alain Thietart: Choice, chance, and inevitability in strategy, in: Strategic Management Journal, Vol. 28 (2007), No. 5, pp. 535–551. Ross, Jim (Vice President of NYSE MatchPoint and ATS (alternative trading system) Strategy): Together, MatchPoint and Algorithms Take Performance to a New Level (29 Apr. 2009), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2009, accessed 18 Oct. 2009: http://exchanges.nyse.com/archives/2009/04/together.php. Rudolph, Jenny W. and Nelson P. Repenning: Disaster Dynamics: Understanding the Role of Quantity in Organizational Collapse, in: Administrative Science Quarterly, Vol. 47 (2002), No. 1, pp. 1–30.
250
References
Ryan, Lori Verstegen and Marguerite Schneider: Institutional Investor Power and Heterogeneity: Implications for Agency and Stakeholder Theories, in: Business Society, Vol. 42 (2003), No. 4, pp. 398–429. Sabherwal, Rajiv, Rudy Hirschheim and Tim Goles: The Dynamics of Alignment: Insights from a Punctuated Equilibrium Model, in: Organization Science, Vol. 12 (2001), No. 2, pp. 179–197. Salge, Markus: Struktur und Dynamik ganzheitlicher Verbesserungsprogramme in der industriellen Fertigung: Ein systemdynamisches Modell zur nachhaltigen Gestaltung des Wandels in Industrieunternehmen, Hamburg 2009. Sarkis, Joseph, Pilar Gonzalez-Torre and Belarmino Adenso-Diaz: Stakeholder pressure and the adoption of environmental practices: The mediating effect of training, in: Journal of Operations Management, Vol. 28 (2010), No. 2, pp. 163–176. Sastry, M. Anjali: Time and tide in organizations: Simulating change processes in adaptive, punctuated, and ecological theories of organizational evolution, Ph.D. Thesis, Alfred P. Sloan School of Management, Massachusetts Institute of Technology, Boston, Massachusetts 1995. Sastry, M. Anjali: Problems and Paradoxes in a Model of Punctuated Organizational Change, in: Administrative Science Quarterly, Vol. 42 (1997), No. 2, pp. 237–275. Sastry, M. Anjali: Understanding Dynamic Complexity in Organizational Evolution: A System Dynamics Approach, in: Lomi, Alessandro and Erik R. Larsen (Ed.): Dynamics of Organizations: Computational Modeling and Organization Theories, Menlo Park, CA [et al.] 2001, pp. 377–404. Saysel, Ali Kerem and Yaman Barlas: Model simplification and validation with indirect structure validity tests, in: System Dynamics Review, Vol. 22 (2006), No. 3, pp. 241–262. Schaefer, Scott: Influence Costs, Structural Inertia, and Organizational Change, in: Journal of Economics and Management Strategy, Vol. 7 (1998), No. 2, pp. 237–263. Schein, Edgar H.: Organizational Culture and Leadership, 3. Ed., San Francisco, CA 2004. Schein, Edgar H. with Peter DeLisi, Paul J. Kampas, and Michael M. Sonduck: DEC is Dead, Long Live DEC: The Lasting Legacy of Digital Equipment Corporation, San Francisco, CA 2003. Schimmel, Remco and Dennis R. Muntslag: Learning Barriers: A Framework for the Examination of Structural Impediments to Organizational Change, in: Human Resource Management, Vol. 48 (2009), No. 3, pp. 399–416. Schmid, Beat F.: Elektronische Märkte, in: Wirtschaftsinformatik, Vol. 35 (1993), pp. 465–480. Schumpeter, Joseph Alois: The Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest, and the Business Cycle, Harvard Economic Studies, Vol. 46, Cambridge, MA 1934.
References
251
Schwaninger, Markus and Stefan Grösser: System Dynamics as Model-Based Theory Building, in: Systems Research and Behavioral Science, Vol. 25 (2008), No. 4, pp. 447–465. Schwartz, Robert A. and Robert A. Wood: Best Execution, in: Journal of Portfolio Management, Vol. 29 (2003), No. 4, pp. 37–48. Schwenk, Charles R.: Cognitive Simplification Processes in Strategic DecisionMaking, in: Strategic Management Journal, Vol. 5 (1984), No. 2, pp. 111–128. Schwenk, Charles R.: The Cognitive Perspective on Strategic Decision Making, in: Journal of Management Studies, Vol. 25 (1988), No. 1, pp. 41–55. Scott, W. Richard: Reflections of a Half-Century of Organizational Sociology, in: Annual Review of Sociology, Vol. 30 (2004), No. 1, pp. 1–21. Securities and Exchange Commission (SEC): Adoption of Amendments to the Intermarket Trading System Plan To Expand the ITS/Computer Assisted Execution System Linkage to all Listed Securities (FR Doc. 99–32684, SEC Release No. 34–42212, File No. 4–208), in: Federal Register, Vol. 64 (1999), No. 241, issued 16 Dec. 1999, pp. 70297–70307. Securities and Exchange Commission (SEC): Regulations NMS (Final Rule) (Release No. 34–51808; File No. S7–10–04), in: Federal Register, Vol. 70 (2005), No. 124, issued 29. June 2005, pp. 37495–37644. Seligman, Joel: The Transformation of Wall Street: A History of the Securities and Exchange Commission and Modern Corporate Finance, 3rd Ed., New York, NY 2003. Sethi, S. Prakash: Multinational Corporations and the Impact of Public Advocacy on Corporate Strategy: Nestlé and the Infant Formula Controversy, Boston 1994. Shapiro, James E.: U.S. Equity Markets: A View of Recent Competitive Developments, NYSE Working Paper 93-02, New York 1993. Shattock, Michael: The academic profession in Britain: A study in the failure to adapt to change, in: Higher Education, Vol. 41 (2001), No. 1/2, pp. 27–47. Sheppard, Jerry Paul and Shamsud D. Chowdhury: Riding the Wrong Wave: Organizational Failure as a Failed Turnaround, in: Long Range Planning, Vol. 38 (2005), No. 3, Special Issue: Organizational Failure, pp. 239–260. Simon, Herbert A.: Administrative Behavior: A Study of Decision-making Processes in Administrative Organizations, New York 1947. Simon, Herbert A.: Models of Man: Social and Rational: Mathematical Essays on Rational Human Behavior in a Social Setting, New York 1957. Simon, Herbert A.: Administrative Behavior: A Study of Decision-making Processes in Administrative Organizations, 3. Ed., New York 1976. Simon, Herbert A.: The Sciences of the Artificial, 3. Ed., Cambridge, MA [et al.] 1999. Slack, Nigel, Stuart Chambers and Robert Johnston: Operations Management, 5. Ed., Harlow [et. al] 2007. Smith, Randall, Greg Ip and Charles Gasparino: Bitter Rivals Jointly Seek Major Changes in the Markets, in: Wall Street Journal (1999), issued 1 Oct. 1999, p. C.2.
252
References
Sobel, Robert: N.Y.S.E.: A History of the New York Stock Exchange 1935–1975, New York 1975. Stanton, Jeffrey M.: An Empirical Assessment Of Data Collection Using The Internet, in: Personnel Psychology, Vol. 51 (1998), No. 3, pp. 709–725. Starbuck, William H. and Frances J. Milliken: Challenger: Fine-Tuning the Odds until Something Breaks, in: Journal of Management Studies, Vol. 25 (1988), No. 4, pp. 319–340. Staw, Barry M., Lance E. Sandelands and Jane E. Dutton: Threat Rigidity Effects in Organizational Behavior: A Multilevel Analysis, in: Administrative Science Quarterly, Vol. 26 (1981), No. 4, pp. 501–524. Sterman, John D.: Modeling Managerial Behavior: Misperceptions of Feedback in a Dynamic Decision Making Experiment, in: Management Science, Vol. 35 (1989), No. 3, pp. 321–339. Sterman, John D.: Learning in and about complex systems, in: System Dynamics Review, Vol. 10 (1994), No. 2/3, pp. 291–330. Sterman, John D.: Business Dynamics: Systems Thinking and Modeling for a Complex World, Boston [et al.] 2000. Sterman, John D.: All models are wrong: reflections on becoming a systems scientist, in: System Dynamics Review, Vol. 18 (2002), No. 4, pp. 501–531. Sterman, John D. and Jason Wittenberg: Path Dependence, Competition, and Succession in the Dynamics of Scientific Revolution, in: Organization Science, Vol. 10 (1999), No. 3, pp. 322–341. Steverman, Ben: NYSE: Hooray for Market Volatility, in: Business Week Online (2007), issued 5 Nov. 2007, p. 25. Stoll, Hans R.: Electronic Trading in Stock Markets, in: Journal of Economic Perspectives, Vol. 20 (2006), No. 1, pp. 153–174. Storkenmaier, Andreas and Ryan Riordan: The Effect of Automated Trading on Market Quality: Evidence from the New York Stock Exchange, in: Kundisch, Dennis, et al. (Ed.): Enterprise Applications and Services in the Finance Industry, Berlin, Heidelberg 2009, pp. 11–30. Strauss, Anselm L. and Juliet M. Corbin: Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory, 2. Ed., Thousand Oaks, CA [et al.] 1998. Strobl, Thomas: Das Herzkammerflimmern der Börse, in: Frankfurter Allgemeine Zeitung (2010), issued 18 May 2010, pp. 29. Sullivan, Bilian Ni: Competition and Beyond: Problems and Attention Allocation in the Organizational Rulemaking Process, in: Organization Science, Vol. 21 (2010), No. 2, pp. 432–450. Tabb, Larry: The NYSE Floor: A Question of Control, in: Wall Street & Technology, (2005), issued Mar. 2005, p. 54. Tabb, Larry: Liquidity Begets Liquidity, in: Advanced Trading, (2008a), issued Feb. 2008, p. 47. Tabb, Larry: A Tale of Two Exchanges, in: Wall Street & Technology, Vol. 4 (2008b), No. 5, p. 49.
References
253
Thain, John: Statement in an Interview with CSBN (6 Oct. 2006), in: NYSE Euronext Inc.: Exchanges: Blogging About NYSE Euronext Markets, [Weblog], 2006, accessed 19 Oct. 2009: http://exchanges.nyse.com/archives/2006/10/ smooth_start.php. Thain, John: Statement on A CEO's Views on IT and Innovation (28 Jun. 2007), in: McGee, Ken: Gartner Fellows Interviews, 2007, accessed 11 Jan. 2010: http://www.gartner.com/research/fellows/asset_181078_1176.jsp. Thun, Jörn-Henrik: Die zeitbasierte Fertigungsstrategie: Methoden zur Leistungssteigerung in Industriebetrieben, Wiesbaden 2002. Tinic, Seha M. and Richard R. West: The Securities Industry under Negotiated Brokerage Commissions: Changes in the Structure and Performance of New York Stock Exchange Member Firms, in: The Bell Journal of Economics, Vol. 11 (1980), No. 1, pp. 29–41. Tomasello, Michael: Joint Attention as Social Cognition, in: Moore, Chris and Philop J. Dunham (Ed.): Joint Attention, Hillsdale, NJ 1995, pp. 103–130. Tripsas, Mary: Technology, Identity, and Inertia through the Lens of 'The Digital Photography Company', in: Organization Science, Vol. 20 (2009), No. 2, pp. 441–460. Tripsas, Mary and Giovanni Gavetti: Capabilities, Cognition, and Inertia: Evidence from Digital Imaging, in: Strategic Management Journal, Vol. 21 (2000), No. 10/11, pp. 1147–1161. Tully, Shawn: Bringing Down the Temple, in: Fortune, Vol. 148 (2003), No. 10, issued 10 Nov. 2003, pp. 119–126. Tushman, Michael L. and Charles A. O'Reilly III: Ambidextrous organizations: Managing evolutionary and revolutionary change, in: California Management Review, Vol. 38 (1996), No. 4, pp. 8–30. Tushman, Michael L. and Elaine Romanelli: Organizational Evolution: A Metamorphosis Model of Convergence and Reorientation, in: Cummings, Larry L. and Barry M. Staw (Ed.): Research in Organizational Behavior, Vol. 7, Greenwich, CT 1985, pp. 171–222. Tversky, Amos and Daniel Kahneman: Judgment under Uncertainty: Heuristics and Biases, in: Science, Vol. 185 (1974), No. 4157, pp. 1124–1131. United States Securities and Exchange Commission (SEC), Division of Market Regulation: Market 2000: An Examination of Current Equity Market Developments, New York, NY 1994. Van de Ven, Andrew H. and Marshall Poole: Explaining development and change in organizations, in: The Academy of Management Review, Vol. 20 (1995), No. 3, pp. 510–540. Van de Ven, Andrew H. and Marshall Scott Poole: Alternative Approaches for Studying Organizational Change, in: Organization Studies, Vol. 26 (2005), No. 9, pp. 1377–1404. Venkataraman, Kumar: Automated Versus Floor Trading: An Analysis of Execution Costs on the Paris and New York Exchanges, in: Journal of Finance, Vol. 56 (2001), No. 4, pp. 1445–1485.
254
References
Vissa, Balagopal, Henrich R. Greve and Wei-Ru Chen: Business Group Affiliation and Firm Search Behavior in India: Responsiveness and Focus of Attention, in: Organization Science, Vol. 21 (2010), No. 3, pp. 696–712. Volkoff, Olga, Diane M. Strong and Michael B. Elmes: Technological Embeddedness and Organizational Change, in: Organization Science, Vol. 18 (2007), No. 5, pp. 832–848. Waddell, Dianne and Amrik S. Sohal: Resistance: a constructive tool for change management, in: Management Decision, Vol. 36 (1998), No. 7/8, pp. 543–548. Walsh, James P.: Managerial and Organizational Cognition: Notes from a Trip Down Memory Lane, in: Organization Science, Vol. 6 (1995), No. 3, pp. 280–321. Wang, Shan, Shi Zhang, Lida Xu, Dezheng Li and Huan Meng: A Literature Review of Electronic Marketplaces: Themes, Theories and an Integrative Framework, in: Information Systems Frontiers, Vol. 10 (2008), No. 5, pp. 555–571. Weber, Bruce W.: Adoption of electronic trading at the International Securities Exchange, in: Decision Support Systems, Vol. 41 (2006), No. 4, pp. 728–746. Weick, Karl E.: The Social Psychology of Organizing, Reading, MA [et al.] 1979. Weick, Karl E.: Sensemaking in Organizations, Thousand Oaks, CA [et al.] 1995. Weick, Karl E.: Making Sense of the Organization, Malden, MA [et al.] 2001. Weiss, Gary: The $140,000,000 Man, in: Business Week, (2003), No. 3849, issued 15 Sep. 2003, pp. 84–92. Welles, Chris: Is it time to make the Big Board a black box?, in: Business Week, (1990), No. 3144, issued 5 Feb. 1990, p. 74. Wennberg, Karl and Carin Holmquist: Problemistic search and international entrepreneurship, in: European Management Journal, Vol. 26 (2008), No. 6, pp. 441–454. Whittington, Richard: Environmental Structure and Theories of Strategic Choice, in: Journal of Management Studies, Vol. 25 (1988), No. 6, pp. 521–536. Wilden, Anthony: System and Structure: Essays in Communication and Exchange, 2. Ed., London 2003. Willmott, Hugh: Breaking the Paradigm Mentality, in: Organization Studies, Vol. 14 (1993), No. 5, pp. 681–719. Winn, Monika I.: Building Stakeholder Theory with a Decision Modeling Methodology, in: Business & Society, Vol. 40 (2001), No. 2, pp. 133–166. Wischnevsky, J. Daniel and Fariborz Damanpour: Punctuated Equilibrium Model of Organizational Transformation: Sources and Consequences in the Banking Industry, in: Research in Organizational Change and Development, Vol. 15 (2004), pp. 207–239. Wischnevsky, J. Daniel and Fariborz Damanpour: Radical strategic and structural change: occurrence, antecedents and consequences, in: International Journal of Technology Management, Vol. 44 (2008), No. 1/2, pp. 53–80. Wollin, Andrew: Punctuated equilibrium: reconciling theory of revolutionary and incremental change, in: Systems Research and Behavioral Science, Vol. 16 (1999), No. 4, pp. 359–367.
References
255
Yadav, Manjit S., Jaideep C. Prabhu and Rajesh K. Chandy: Managing the Future: CEO Attention and Innovation Outcomes, in: Journal of Marketing, Vol. 71 (2007), No. 4, pp. 84–101. Zander, Alvin: Resistance to Change—Its Analysis and Prevention, in: Advanced Management, Vol. 4 (1950), No. 5, pp. 9–11. Zimmermann, Nicole S. and Peter M. Milling: Data Generation from Weblogs in Qualitative Research and System Dynamics, in Proceedings of the 17th International Annual EurOMA Conference, 2010, pp. 1–10.
257
Appendix Appendix A: Financial Glossary
auction market
A market in which a market maker conducts an auction for specific securities and allows for negotiation over the price.
designated market maker A market professional who has the responsibility to provide a fair and orderly market for the securities he has been assigned. He combines a physical and an automated auction that includes algorithmic quotes. His responsibilities are to bring together demand and supply and to quote at the NBBO a specified percentage of time. designated order turnaround (DOT)
NYSE system which allows brokerage firms to electronically transmit orders directly to the specialist.
Direct+
A high-speed electronic system for immediate automatic execution of limit orders that was implemented at the NYSE in 2001.
DMM
See designated market maker.
DOT
Designated Order Turnaround. It is an electronic system that allows brokers to route orders directly to the specialist instead of a floor broker.
Financial Industry Regulatory Authority
A private regulatory body governing the business between brokers, dealers, and the investing public. It was formerly known as Securities Industry Regulatory Authority (SIRA).
floor broker
Broker physically located on the NYSE trading floor who competes with other brokers to receive the best price for his customer.
institutional investor
Financial institutions such as banks, insurance companies, investment funds, pension funds, proprietary trader organizations and else that frequently engage in securities trading.
Intermarket Trading System
System that gives market professionals the opportunity to send orders to other markets if these markets display a better price.
liquidity algorithm
With the help of liquidity algorithms, the NYSE’s DMMs and Special Liquidity Providers are able to constantly provide automated bids and offers also in electronic trading.
N. Zimmermann, Dynamics of Drivers of Organizational Change, DOI 10.1007/978-3-8349-6811-1, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011
258
Appendix
market maker
By providing a bid and an offer price, a market maker makes sure that a specific group of securities can always be traded.
NBBO
National best bid and offer, the highest bid and lowest offer in the U.S. securities market.
order flow
Incoming orders.
proprietary trader
A firm trading with its own instead of the customers’ money.
SEC
The Securities and Exchange Commission is the regulating body of the U.S. securities industry.
specialist
A market professional who manages the auction market trading in the specific securities he (or she) has been assigned. He has the responsibility to provide a fair and orderly market, brings together demand and supply, and steps in with his own money in order to match imbalances in the market.
spread
Difference between the asking price at which shares of a certain security are offered and the bid price which someone is willing to pay for shares of this security.
Supplementary Liquidity Providers
Electronic, high-volume members who are incented to add liquidity on the NYSE.
Total Consolidated Tape The Total Consolidated Tape aggregates shares matched at U.S. exchanges and volume of transactions effected otherwise than on an exchange which are reported to the Financial Industry Regulatory Authority. trade-through rule
The trade-through rule helps price protection by demanding that an order is not traded through inferior markets, but should instead be directed to the market which offers the best price.
259
Appendix
Appendix B: Simulation Runs and Sensitivity Analyses of the NYSE Model
Insensitive reaction in response to changes in stakeholder power. sens power 50% 75% 95% "NYSE Fraction of E-Trade" 1
100%
0.75
0.5
0.25
0 1970
1980
1990
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
sens power 50% 75% 95% 100% total pressure for more floor trade from floor 60
45
30
15
0 1970
1980
changed parameter ref. power of floor firms
1990
base run value 100
parameter range 10 – 500
260
Appendix
Insensitive reaction to changes in the cohesiveness of floor firms sens cohesiveness 50% 75% 95% "NYSE Fraction of E-Trade" 1
100%
0.75
0.5
0.25
0 1970
1980
sens cohesiveness 50% 75% 95%
1990
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
100%
total pressure for more floor trade from floor 80
60
40
20
0 1970
1980
1990
changed parameter degree of cohesiveness of floor firms
base run value 0.7
parameter range 0.1 – 5
261
Appendix
Simulation run representing an exchange with a small market share
Fraction of E-Trade 12
12
12
1
1
2
Dmnl
0.75 1
0.5 0.25 0 12 1970
12
12 1980
1 2 12 1990
12 1 2000 Date
2
"NYSE Fraction of E-Trade" : small exchange 1 "NYSE Fraction of E-Trade" : base run 2 2
changed parameters ini market share grasso effect strength
.
2010 1
2020 1
2
1 2
base run value 0.8747 1
2030 1
2
2
parameter value 0.03 0
262
Appendix
Pressure by non-institutional customers sens non-inst pressure 50% 75% 95% "NYSE Fraction of E-Trade" 1
100%
0.75 0.5 0.25 0 1970
1980
1990
2000 Date
2010
2020
2030
2010
2020
2030
sens non-inst pressure 50% 75% 95% 100% total pressure for more floor trade from customers 20 15 10 5 0 1970
1980
1990
changed parameter ref. pressure per non-inst customer
2000 Date
base run value 1
parameter range 0.01 – 5
263
Appendix
Sensitivity for differences in inertia sens inertia 50% 75% 95% "NYSE Fraction of E-Trade" 1
100%
0.75 0.5 0.25 0 1970
sens inertia 50% 75%
1980
95%
1990
2000 Date
2010
2020
2030
2000 Date
2010
2020
2030
100%
Inertia 1 0.75 0.5 0.25 0 1970
1980
changed parameters ref. fract. institutionalization ini inertia
1990
.
base run value 0.3 0.95
parameter range 0.2 – 0.35 0.05 – 0.95
264
Appendix
Appendix C: Simulation Runs and Sensitivity Analyses of the Generic Model
Comparative sensitivity analyses with quick environmental change: Inertia generic sens inertia flex 50% 75% 95%
100%
Orientation to Strategy B 1
0.75
0.5
0.25
0
0
5
10
15
20
25 30 Time (Year)
35
40
45
50
20
25 30 Time (Year)
35
40
45
50
generic sens inertia flex quick 50% 75% 95% 100% Orientation to Strategy B 1
0.75
0.5
0.25
0
0
5
10
15
changed parameters base run value ref. fract. change in strategy per pressure p.a. 0.02 ref. fract. change in attention p.a. 0.05 ref. fract. inertia decrease 0.15 ref. fract. institutionalization 0.3 ini inertia 0.9 ini attention to stakeholders favoring B 0.1
parameter range 0.06 0.3 0.15 – 0.2 0.2 – 0.3 0.5 0.3
265
Appendix
Comparative sensitivity analyses with quick environmental change: Attention generic sens attention flex 50% 75% 95% Orientation to Strategy B 1
100%
0.75
0.5
0.25
0
0
5
10
15
20
25 30 Time (Year)
35
40
45
50
20
25 30 Time (Year)
35
40
45
50
generic sens attention flex quick 50% 75% 95% 100% Orientation to Strategy B 1
0.75
0.5
0.25
0
0
5
10
15
changed parameters base run value ref. fract. change in strategy per pressure p.a. 0.02 ref. fract. change in attention p.a. 0.05 ref. fract. inertia decrease 0.15 ini inertia 0.9 ini attention to stakeholders favoring B 0.1
parameter range 0.06 0.05 – 0.5 0.28 0.2 0.3
266
Appendix
Appendix D: NYSE Model Equations
access to information technology
ACCESS TO INFORMATION TECHNOLOGY = WITH LOOKUP( Time , ([(1970,0)-(2030,1)],(1970,0),(1975,0), (1980,0.09),(1985,0.18),(1990,0.57),(1995,0.9),(1997,0.97),(1999,1), (2030,1) ) ) Units: Dmnl State of technology that is necessary for electronic trading. Access to Information Technology 1
0.5
0 1970
1980
1990
2000 Time (Year)
2010
2020
2030
ALGORITHMS PER GAP PER YEAR = 50 Units: algorithms/(Year*Dmnl) "change in cust. orient." = ( pcvd pressure from customers * "effect of cust. orient. on change" - pcvd pressure from the floor * "effect of floor orient. on change" ) * "fract. change in cust. orient. per pressure p.a." Units: Dmnl/Year "change in fraction of e-trade" = ( "pcvd pressure for more e-trade" * "effect of e-trade on change" - pcvd pressure for more floor trade * effect of floor trade on change ) * "fract. change per pcvd pressure p.a." Units: Dmnl/Year change in fraction of institutional customers = ( indicated fraction of institutional customers - Fraction of Institutional Customers ) / TIME TO BECOME CUSTOMER Units: Dmnl/Year
267
Appendix
change in power of floor firms = ( indicated power of floor firms - Power of Floor Firms ) / TIME TO CHANGE POWER OF FLOOR FIRMS Units: entity/Year change in valuation = Valuation of Floor Culture by Floor * fractional change in valuation of floor culture Units: valuation unit/Year commission per share = "REF. COMMISSION PER SHARE" * "effect of inst. customers on commission" Units: $/share Part of the floor’s earnings. confidence effect of market share = WITH LOOKUP( pcvd adequacy of market share , ([(0,0)-(1.2,1)],(0,0),(0.2,0.04),(0.4,0.14),(0.5,0.22),(0.6,0.33), (0.7,0.5), (0.8,0.75),(0.9,0.95),(0.95,0.985),(1,1),(1.2,1) ) ) Units: Dmnl Effect by which performance inadequacies increase the management team's openness to change. Minor inadequacies have less than proportional effect, but the effect on openness quickly rises before it slowly approaches the limit of a fully open organization in the case of organizational collapse. Confidence Effect of Market Share confidence effect
1 0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 pcvd adequacy of market share
1
cultural multiplier of pressure from floor = "rel. valuation of floor culture" * DEGREE OF COHESIVENESS OF FLOOR FIRMS Units: Dmnl Effect of culture and cohesiveness.
268
Appendix
Customer Orientation = INTEG( "change in cust. orient." , INI CUSTOMER ORIENTATION ) Units: Dmnl Attention to and orientation towards institutional and non-institutional customers. DEGREE OF COHESIVENESS OF FLOOR FIRMS = 0.7 Units: Dmnl Degree to which floor firms need to rely on each other. Cooperative groups may react with resistance. degree of trading professionalization = WITH LOOKUP( Fraction of Institutional Customers , ([(0,0)-(1,1)],(0,0.1),(0.1,0.27),(0.2,0.45),(0.3,0.63),(0.4,0.8), (0.5,0.92),(0.6,0.98), (0.7,0.995),(0.8,1),(0.9,1),(1,1) ) ) Units: Dmnl Portfolio management, information patterns-based trading, hedging, etc. Degree of Trading Professionalization professionalization
1 0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 Fraction of Institutional Customers
1
DESD ADEQUACY OF MARKET QUALITY = 1 Units: Dmnl desired earnings per share = SMOOTH ( proportional floor earnings per share traded , TIME TO ADJUST DESIRED EARNINGS ) Units: $/share The floor's floating goal of earnings. "desired market quality from sp. part. by customers" = SMOOTH3 ( "market quality from sp. participation" , TIME TO CHANGE DESD MARKET QUALITY ) Units: Dmnl Floating goal of desired market quality.
269
Appendix
desired market share = SMOOTH ( NYSE Market Share , TIME TO ADJUST DESD MARKET SHARE ) Units: Dmnl Floating goal of desired market share. desired specialist participation = SMOOTH ( specialist participation , TIME TO ADJUST DESIRED PARTICIPATION ) Units: Dmnl The floor's floating goal of desired specialist participation. development of liquidity algorithms = SMOOTH3 ( market quality adequacy gap * ALGORITHMS PER GAP PER YEAR , TIME TO DEVELOP ALGORITHMS ) Units: algorithms/Year Algorithms take about one and a half years to be initiated since a quality gap needs to be perceived as being problematic; this is why there is a third order smooth in the development decision instead of a development delay. "dissatisf. with time per inst. customer" = WITH LOOKUP( relative time to execution , ([(0,0)-(10,1)],(0.9,0),(1,0), (9,1) ) ) Units: dissatisfaction unit/entity Institutional customers' extent of dissatisfaction with or dislike of the NYSE's relative speed of execution. Dissatisfaction with Time per Institutional Customer
dissatisfaction
1 0.75 0.5 0.25 0 1
2
3
4 5 6 relative time to execution
7
8
9
270
Appendix
dissatisfaction effect of market quality on pressure = WITH LOOKUP( pcvd adequacy of market quality by customer , ([(0.9,0)-(1.05,1)],(0.9,1),(0.91,0.98),(0.92,0.95),(0.93,0.9),(0.94,0.75), (0.95,0.5),(0.96,0.25),(0.97,0.1),(0.98,0.05),(0.99,0.02),(1,0),(1.05,0) ) ) Units: Dmnl Inversely s-shaped. Slowly approaches maximum, and slowly starts in the beginning because minor inadequacies cause less than proportional reactions. Dissatisfaction Effect of Market Quality on Pressure dissatisfaction effect
1 0.75 0.5 0.25 0 0.900
0.920 0.940 0.960 0.980 pcvd adequacy of market quality by customer
1
effect of captial distribution on customers = WITH LOOKUP( FRACTION OF EQUITIES HELD BY INSTITUTIONS , ([(0,0)-(1,1)],(0,0),(0.1,0.19),(0.2,0.36),(0.3,0.52),(0.4,0.62),(0.5,0.71), (0.6,0.79),(0.7,0.86),(0.8,0.91),(0.9,0.96),(1,1) ) ) Units: Dmnl Institutions participate in trading in a more than proportional way Æ concave function. Effect of Capital Distribution on Customers 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 FRACTION OF EQUITIES HELD BY INSTITUTIONS
1
271
Appendix
effect of change on inertia = WITH LOOKUP( ABS ( "change in fraction of e-trade" ) , ([(0,0)-(0.5,7)],(0,1),(0.05,1.4),(0.1,2.4),(0.15,4.2),(0.2,5.4),(0.3,6.2), (0.5,6.5) ) ) Units: Dmnl Small changes have an underproportional effect on consistency loss. This allows an organization to change very slowly without disruption in its internal consistency. The consistency decrease from change represents turnover rates which became higher, but it also captures changes in the people's thinking even if they remain in the organization Effect of Change on Inertia 8
effect
6 4 2 0 0
0.100
0.200 0.300 "change in fraction of e-trade"
0.400
0.500
"effect of cust. orient. on change" = WITH LOOKUP( Customer Orientation , ([(0,0)-(1,1)],(0,1),(0.25,1),(0.5,0.95),(0.8,0.75),(0.9,0.5),(0.96,0.04), (0.98,0.005),(1,0) ) ) Units: Dmnl Limit to the willingness to further react to pressure. Effect of Customer Orientation on Change 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 Customer Orientation
1
272
Appendix
"effect of e-trade on change" = WITH LOOKUP( "NYSE Fraction of E-Trade" , ([(0,0)-(1,1)],(0,1),(0.5,1),(0.75,0.95),(0.9,0.75),(0.95,0.5),(0.99,0.01), (1,0) ) ) Units: Dmnl Limit to the willingness to further react to pressure. Effect of E-Trade on Change 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 "NYSE Fraction of E-Trade"
1
effect of employability on resistance = WITH LOOKUP( pcvd adequacy of employability , ([(0,0)-(1.2,1)],(0,1),(0.5,1),(0.55,0.97),(0.6,0.88),(0.75,0.5),(0.9,0.13), (0.95,0.05),(1,0),(1.2,0) ) ) Units: Dmnl Inversely s-shaped. Slowly approaches maximum, when adequacy is only 0.5 and slowly starts in the beginning because minor inadequacies cause less than proportional reactions. Effect of Employability on Resistance 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 pcvd adequacy of employability
1
"effect of floor orient. on change" = WITH LOOKUP( Floor Orientation , ([(0,0)-(1,1)],(0,1),(0.25,1),(0.5,0.95),(0.8,0.75),(0.9,0.5),(0.96,0.04), (0.98,0.005),(1,0) ) ) Units: Dmnl Limit to the willingness to further react to pressure. The graph has the same shape as the one indicating the effect of customer orientation on change.
273
Appendix
effect of floor trade on change = WITH LOOKUP( NYSE Fraction of Floor Trade , ([(0,0)-(1,1)],(0,1),(0.5,1),(0.75,0.95),(0.9,0.75),(0.95,0.5),(0.99,0.01), (1,0) ) ) Units: Dmnl Limit to the willingness to further react to pressure. The graph has the same shape as the one indicating the effect of e-trade on change. "effect of floor trade on sp. participation" = WITH LOOKUP( NYSE Fraction of Floor Trade , ([(0,0)-(1,1)],(0,0.1),(1,1) ) ) Units: Dmnl It is the special feature of the Hybrid Market that even when all trades are electronic, there is some specialist participation. In general, the effect of floor trade on specialist participation is assumed to grow linearly. Effect of Floor Trade on Specialist Participation 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 NYSE Fraction of Floor Trade
1
EFFECT OF GRASSO SCANDAL = 1 + PULSE ( 2004, GRASSO SCANDAL DURATION ) * GRASSO EFFECT STRENGTH Units: Dmnl In January 2004 a scandal around the former CEO Richard Grasso triggered a change of the CEO that decreased inertia.
274
Appendix
"effect of inst. customers on commission" = WITH LOOKUP( number of institutional customers , ([(10,0)-(100,1)],(10,1),(20,0.95),(65,0.06),(75,0.02),(100,0) ) ) Units: Dmnl Institutional customers became so powerful that they were able to strongly reduce the amount of money they need to pay for the floor's services. Effect of Institutional Customers on Commission 1
effect
0.75 0.5 0.25 0 0
10
20
30 40 50 60 70 80 number of institutional customers
90
100
"effect of inst. customers on spread" = WITH LOOKUP( number of institutional customers , ([(10,0)-(100,1)],(10,1),(25,0.95),(44,0.82),(55,0.6),(65,0.33),(80,0.15), (100,0.05) ) ) Units: Dmnl Regulatory effects that came with the rise of institutional customers such as the transition of quoting in eights of a dollar to sixteenth to pennies. Effect of Institutional Customers on Spread 1
effect
0.75 0.5 0.25 0 0
10
20
30 40 50 60 70 80 number of institutional customers
90
100
275
Appendix
effect of institutional customers on power = WITH LOOKUP( Fraction of Institutional Customers , ([(0,0)-(1,1)],(0,1),(0.2,0.98),(0.4,0.93),(0.6,0.75),(0.8,0.5),(1,0.1) ) ) Units: Dmnl Institutional customers diminish the floor's power since they are powerful as well. Some fraction of institutions has a low effect, but the strength of the effect rises Effect of Institutional Customers on Power 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 Fraction of Institutional Customers
1
effect of liquidity algorithms on participation = ( Liquidity Algorithms + "REF. LIQUIDITY ALGORITHMS" ) / "REF. LIQUIDITY ALGORITHMS" Units: Dmnl Liquidity Algorithms allow the floor to participate also in electronic trades and thus increase specialist participation. Due to the balancing nature of the Liquidity Algorithms Loop, specialist participation remains in reasonable bounds, although it may oscillate slightly around the goal value. effect of market quality on culture = WITH LOOKUP( pcvd adequacy of market quality by customer , ([(0.9,-1)-(1.1,1)],(0.93,-0.7),(1,0),(1.07,0.7) ) ) Units: Dmnl Floor participants reduce their valuation of their own culture and way of doing things if they do not provide adequate market quality. Effect of Market Quality on Culture 0.8
effect
0.4 0 -0.4 -0.8 0.930
0.950 0.970 0.990 1.010 1.030 1.050 pcvd adequacy of market quality by customer
1.070
276
Appendix
effect of market quality on power = WITH LOOKUP( "market quality from sp. participation" , ([(1,0)-(1.1,1)],(1,0),(1.1,1) ) ) Units: Dmnl The higher market quality is, i.e. the higher the floor's contribution is, the more powerful is the floor. Effect of market Quality on Power 1
effect
0.75 0.5 0.25 0 1
1.020 1.040 1.060 1.080 "market quality from sp. participation"
1.100
effect of openness on change = WITH LOOKUP( openness to change , ([(0,0)-(1,1)],(0,0.1),(0.1,0.11),(0.2,0.14),(0.3,0.21),(0.4,0.3),(0.5,0.435), (0.6,0.63),(0.7,0.81),(0.8,0.92),(0.9,0.97),(1,1) ) ) Units: Dmnl Low openness to change may reduce fractional change to 10 percent of its reference value. The effect of openness on change is an s-shaped curve indicating that the NYSE quickly reacts to perceived pressure if it has a rather high openness. It becomes less responsive as openness decreases until its reactivity reaches a lower bound. Effect of Openness on Change 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 openness to change
1
277
Appendix
effect of profitability on culture = WITH LOOKUP( pcvd adequacy of profitability , ([(0,-0.5)-(2,0.5)],(0,-0.5),(1,0),(2,0.5) ) ) Units: Dmnl Floor participants reduce their valuation of their own culture and way of doing things if they are not profitable. Effect of Profitabiltiy on Culture 0
effect
-0.125 -0.25 -0.375 -0.5 0
0.20
0.40 0.60 pcvd adequacy of profitability
0.80
1
effect of profitability on resistance = WITH LOOKUP( pcvd adequacy of profitability , ([(0,0)-(1.1,1)],(0,1),(0.5,1),(0.55,0.97),(0.6,0.88),(0.75,0.5),(0.9,0.13), (0.95,0.05),(1,0),(1.1,0) ) ) Units: Dmnl Inversely s-shaped. Slowly approaches maximum, when adequacy is only 0.5 and slowly starts in the beginning because minor inadequacies cause less than proportional reactions. Effect of Profitability on Resistance 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 pcvd adequacy of profitability
1
EFFECT OF REGULATION = STEP ( 1, 2005.5) Units: Dmnl Regulation NMS changed the situation in the market. It came into effect in the year 2005.
278
Appendix
effect of relative trading volume on market's spread = WITH LOOKUP( trading volume of the remaining market / TOTAL U.S. SHARE VOLUME IN NYSE- LISTED ISSUES , ([(0,0.8)-(1,1.2)],(0,1.1),(0.5,1),(1,0.9) ) ) Units: Dmnl The market with the higher trading volume usually has more quoted depth which reduces the spread. Effect of Relative Trading Volume on Market's Spread 1.1
effect
1.05 1 0.95 0.9 0 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1 "relative trading volume of the remaining market ( = trading volume / TOTAL )
effect of relative trading volume on NYSE spread = WITH LOOKUP( NYSE trading volume / TOTAL U.S. SHARE VOLUME IN NYSE- LISTED ISSUES, ([(0,0.8)-(1,1.2)],(0,1.1),(0.5,1),(1,0.9) ) ) Units: Dmnl The graph has the same shape as the one indicating the effect of relative trading volume on market’s spread.
279
Appendix
"effect of sp. participation on market quality" = WITH LOOKUP( specialist participation , ([(0,1)-(0.1,1.1)],(0,1),(0.01,1.015),(0.02,1.03),(0.04,1.055),(0.06,1.075), (0.08,1.09),(0.1,1.1) ) ) Units: Dmnl The curve is concave due to the diminishing marginal utility of specialist participation. This non-linear relationship bases on the fact that there is an absolute limit to the effect that specialist participation can have on market quality. This value always depends on the specific security, but on average it can be assumed that specialist involvement is needed in no more than 10 percent of trades. Then, specialists are able to increase market quality by 10 percent. Effect of Specialist Participation on Market Quality 1.1
effect
1.075 1.05 1.025 1 0
0.020
0.040 0.060 specialist participation
0.080
0.100
effect of time to execution on market share = WITH LOOKUP( relative time to execution , ([(0,0.5)-(9,1.136)],(0,1.1),(0.3,1.04),(0.5,1.02),(0.75,1.005),(1,1), (1.5,0.995), (2,0.99),(3,0.985),(4,0.965),(5,0.91),(6,0.82),(7,0.73), (9,0.5) ) ) Units: Dmnl Upward or downward adjustment of market share based on the NYSE's relative speed. Effect of Time to Execution on Market Share 1.2
effect
1 0.8 0.6 0.4 0
FINAL TIME = 2030 Units: Year
1
2
3 4 5 6 relative time to execution
7
8
9
280
Appendix
floor earnings per share handled = commission per share + NYSE spread / HALF SPREADS Units: $/share The floor makes money from commissions and the half-spread, i.e. the difference between the price and the mid-point between the bid and ask quote. Floor Orientation = INTEG( - "change in cust. orient." , 1- INI CUSTOMER ORIENTATION ) Units: Dmnl The NYSE’s attention to and orientation towards floor firms. "fract. change in cust. orient. per pressure p.a." = "REF. FRACT. CHANGE IN CUST. ORIENT. P.A." * effect of openness on change Units: Dmnl/(Year*pressure unit) Flexibility of attention. Mix of the NYSE management team's general flexibility of attention and situational factors. "fract. change per pcvd pressure p.a." = "REF. FRACT. CHANGE IN TRADING PER PRESSURE P.A." * effect of openness on change Units: Dmnl/(Year*pressure unit) Responsiveness of the NYSE’s strategy. Mix of the management team's general responsiveness to pressure and situational factors. "fract. of e-trade among foreign competitors" = SMOOTH3 ( "Fraction of largest foreign competitors allowing some e-trade" , "TIME TO FULLY IMPLEMENT E-TRADE" ) Units: Dmnl FRACTION OF CAPITAL HELD BY INSTITUTIONS FROM NYSE DATA = WITH LOOKUP (Time, ([(1950,0)-(2010,1)],(1950,0.072),(1970,0.282),(1990,0.414), (1992,0.417),(1993,0.411),(1995,0.437),(1997,0.477),(1999,0.432), (2000,0.47),(2001,0.483) ) ) Units: Dmnl Data has been taken from NYSE Facts and Figures. There is not data for 1960, so the table goes back to the year 1950. It serves as a comparison to data from ICI.org which is used as model input.
281
Appendix
"fraction of e-trade in remaining market" = SMOOTH3 ( degree of trading professionalization * ACCESS TO INFORMATION TECHNOLOGY , "TIME TO DEVELOP E-TRADE POSSIBILITIES" ) Units: Dmnl Adoption of e-trade in the market. FRACTION OF EQUITIES HELD BY INSTITUTIONS = WITH LOOKUP( Time , ([(1970,0)-(2030,1)], (1970,0.185154),(1971,0.20556),(1972,0.201695), (1973,0.224875),(1974,0.258656),(1975,0.263879),(1976,0.247148), (1977,0.267071),(1978,0.290532),(1979,0.282315),(1980,0.274132), (1981,0.2911),(1982,0.324992),(1983,0.354418),(1984,0.376369), (1985,0.398256),(1986,0.374692),(1987,0.390658),(1988,0.359247), (1989,0.364516),(1990,0.375896),(1991,0.370787),(1992,0.372372), (1993,0.400006),(1994,0.422731),(1995,0.420065),(1996,0.433062), (1997,0.424297),(1998,0.433194),(1999,0.419878),(2000,0.450718), (2001,0.480608),(2002,0.508861),(2003,0.523507),(2004,0.548993), (2005,0.572488),(2006,0.587098),(2007,0.615491),(2015,0.666667), (2030,0.714912) ) ) Units: Dmnl Fraction of equities held by institutions such as mutual funds, insurance companies, etc. (see ICI.org). Fraction of Equities Held by Institutions
Dmnl
0.8
0.4
0 1970
1980
1990
2000 Date
2010
2020
2030
Fraction of Institutional Customers = INTEG( change in fraction of institutional customers , 0.25) Units: Dmnl This is a number that relates to the percentage of shares traded by institutional customers at the NYSE. It ranged around 25 percent in 1970.
282
Appendix
"Fraction of largest foreign competitors allowing some e-trade" = WITH LOOKUP( Time , ([(1970,0)-(2030,1)],(1970,0),(1976,0),(1977,0.0454545), (1982,0.0909091),(1986,0.181818),(1987,0.227273),(1988,0.363636), (1989,0.5),(1991,0.681818),(1994,0.772727),(1996,0.909091), (1997,0.954545),(2000,1),(2030,1) ) ) Units: Dmnl See the graph on page 88. fraction of time at NBBO = WITH LOOKUP( relative spread of NYSE , ([(0.82,0)-(1.22,1)],(0.82,0.99),(1.22,0.01) ) ) Units: Dmnl This variable expresses the effect of the spread on the trade execution time. A relative spread of 1.02 equally distributes the shares at the NBBO between the NYSE and remaining market. This little shift of the graph to the right represents the fact that the NYSE is more consolidated since it is a single stock exchange whereas the remaining market consists of several exchanges. Fraction of Time at NBBO 1
fraction
0.75 0.5 0.25 0 0.750
0.850
0.950 1.050 relative spread of NYSE
1.150
1.250
fractional change in valuation of floor culture = ( effect of profitability on culture + effect of market quality on culture ) * "REF. FRACTIONAL CHANGE OF VALUATION PER YEAR" Units: Dmnl/Year GRASSO EFFECT STRENGTH = 1 Units: Dmnl Expresses by what factor the Grasso scandal increased the ref. fract. inertia decrease.
Appendix
283
GRASSO SCANDAL DURATION = 0.5 Units: Year Duration of uncertainty and turbulence from scandal. HALF SPREAD = 2 Units: Dmnl Number of half spreads contained in the spread. indicated fraction of institutional customers = effect of capital distribution on customers Units: Dmnl indicated NYSE market share = ( "NYSE market share from NBBO (cons.)" * ( 1- EFFECT OF REGULATION ) + "NYSE market share from NBBO (fragm.)" * EFFECT OF REGULATION ) * market share adjustment Units: Dmnl Market share from NBBO, moderated by adjustments from speed and market qualtiy. indicated power of floor firms = effect of market quality on power * effect of institutional customers on power * "REF. POWER OF FLOOR FIRMS" Units: entity Inertia = INTEG( institutionalization - inertia decrease , INI INERTIA ) Units: consistency unit Inward-orientation of thinking, cognitive inertia, … inertia decrease = Inertia * "REF. FRACT. INERTIA DECREASE" * EFFECT OF GRASSO SCANDAL * effect of change on inertia Units: consistency unit/Year Management team turnover, unlearning, …
284
Appendix
INI CUSTOMER ORIENTATION = 0.1 Units: Dmnl The initial customer orientation represents the minimum amount of attention that the NYSE management attributes to its customers. INI MARKET SHARE = 0.8747 Units: Dmnl INI INERTIA = 0.9 Units: consistency unit Initial value = effect of (ref. fract. consistency decrease / ref. fract. institutionalization) = 0.9 INITIAL TIME = 1970 Units: Year institutionalization = "REF. FRACT. INSTITUTIONALIZATION" * Inertia * limiting effect on institutionalization Units: consistency unit/Year Growth of inertia, e.g. by cultural institutionalization, learning, etc. limiting effect on institutionalization = WITH LOOKUP( Inertia , ([(0,0)-(1,1)],(0,1),(0.2,1),(0.4,0.99),(0.6,0.9),(0.75,0.75),(0.9,0.5), (0.97,0.25),(1,0) ) ) Units: Dmnl This effect counteracts the reinforcing institutionalization loop. The more the organization is consistent, the more it slows consistency growth down. Limiting Effect on Institutionalization
limiting effect
1 0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 Inertia
1
Appendix
285
Liquidity Algorithms = INTEG( development of liquidity algorithms , 0) Units: algorithms Liquidity Algorithms allow the floor to participate also in electronic trades. This mathematical formulation chosen allows Liduidity Algorithms to rise and to decline. Even in extreme situations, their value stays in reasonable bounds. Therefore, a fuzzy min and max formulation is not chosen here. market quality adequacy gap = DESD ADEQUACY OF MARKET QUALITY - pcvd adequacy of market quality by customer Units: Dmnl Difference between desired and actual adequacy. "market quality from sp. participation" = "effect of sp. participation on market quality" * "REF. MARKET QUALITY" Units: Dmnl Quality attribute of NYSE floor trading. This is what stock exchanges used to compete on. In particular it includes price quality, but also volatility, quoted depth (volume) at each liquidity point, etc. Market quality may also adjust market share upwards. market share adjustment = "wt. on time vs. spread among all customers" * effect of time to execution on market share + ( 1- "wt. on time vs. spread among all customers" ) * "market quality from sp. participation" Units: Dmnl Upwards or downwards adjustment of market share, independent of the part of market share which is set by the time at the NBBO. "no of non-institutional customers" = "TOTAL NO. OF CUSTOMERS" * ( 1- Fraction of Institutional Customers ) Units: entity Normalized number of private or retail customers. number of institutional customers = Fraction of Institutional Customers * "TOTAL NO. OF CUSTOMERS" Units: entity Normalized number of institutional customers.
286
Appendix
"NYSE Fraction of E-Trade" = INTEG( "change in fraction of e-trade" , 0) Units: Dmnl Fraction of fully automated trading at the NYSE. NYSE Fraction of Floor Trade = INTEG( - "change in fraction of e-trade" , 1) Units: Dmnl Fraction to which trades are executed manually on the floor. NYSE Market Share = SMOOTH3I ( indicated NYSE market share , TIME FOR CHANGING MARKET SHARE , INI MARKET SHARE ) Units: Dmnl The NYSE’s fraction of total U.S. consolidated share volume in NYSE-listed issues. "NYSE market share from NBBO (cons.)" = WITH LOOKUP( fraction of time at NBBO , ([(0,0)-(1,0.8)],(0.01,0.01),(0.07,0.05),(0.25,0.27),(0.5,0.54),(0.75,0.77), (0.78,0.79),(0.99,0.8) ) ) Units: Dmnl The better the relative spread, the higher the fraction of time at the NBBO, i. e. the fraction of time the NYSE displays the national best bid and offer. The line rises below proportionally in the very beginning so as to account for the fact that an exchange that displays bad prices most of the time has difficulties to attract a sufficient depth of liquidity at the best price. A critical mass of liquidity is necessary for an exchange to attract volume. market share from NBBO
NYSE Market Share from NBBO (cons.) 1 0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 fraction of time at NBBO
1
287
Appendix
"NYSE market share from NBBO (fragm.)" = WITH LOOKUP( fraction of time at NBBO , ([(0,0)-(1.22,0.6)],(0.01,0.01),(0.07,0.05),(0.123,0.115),(0.2,0.19), (0.35,0.28), (0.65,0.43),(0.8,0.49),(0.99,0.52) ) ) Units: Dmnl Both NBBO calculations follow a highly similar graphical shape, except that for the fragmented market, it levels off on a lower level. Due to fragmentation of orders, one exchange is not able to dominate the market to the extent at which this was possible before. market share from NBBO
NYSE Market Share from NBBO (fragm.) 0.6 0.45 0.3 0.15 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 fraction of time at NBBO
1
NYSE spread = "REF. SPREAD" * effect of relative trading volume on NYSE spread * "effect of inst. customers on spread" Units: $/share The spread is the difference between the bid and the asking price. NYSE time to execution = "NYSE Fraction of E-Trade" * "TIME TO EXECUTION E-TRADE" + NYSE Fraction of Floor Trade * TIME TO EXECUTION FLOOR TRADE Units: second/trade The time it takes to execute a trade, i.e. the time between order submission and execution. NYSE trading volume = NYSE Market Share * TOTAL U.S. SHARE VOLUME IN NYSE- LISTED ISSUES Units: share/Year The trading volume can be used to measure the model’s fit to data.
288
Appendix
openness to change = 1 - Inertia * confidence effect of market share * "REF. OPENNESS PER INERTIA" Units: Dmnl Readiness to change that is limited by inertia, but may be inhanced in the case of a performance threat. pcvd adequacy of employability = specialist participation / desired specialist participation Units: Dmnl pcvd adequacy of market quality by customer = "market quality from sp. participation" / "Desired Market Quality From Sp. Part. by Customers" Units: Dmnl pcvd adequacy of market share = NYSE Market Share / desired market share Units: Dmnl pcvd adequacy of profitability = proportional floor earnings per share traded / desired earnings per share Units: Dmnl "pcvd pressure for more e-trade" = "total pressure for more e-trade from customers" * Customer Orientation Units: pressure unit The management team's biased perception of institutional pressure for more e-trade. pcvd pressure for more floor trade = total pressure for more floor trade from customers * Customer Orientation + total pressure for more floor trade from floor * ( 1 - Customer Orientation ) Units: pressure unit The management team's biased perception of pressure for more floor trade.
Appendix
289
pcvd pressure from customers = ( "total pressure for more e-trade from customers" + total pressure for more floor trade from customers ) * Customer Orientation Units: pressure unit Biased perception of pressure from institutional and non-institutional customers. pcvd pressure from the floor = total pressure for more floor trade from floor * Floor Orientation Units: pressure unit Biased perception of pressure from floor. Power of Floor Firms = INTEG( change in power of floor firms , "REF. POWER OF FLOOR FIRMS" ) Units: entity The degree of influence of the floor. "pressure for more e-trade per inst. customer" = "dissatisf. with time per inst. customer" * "REF. PRESSURE PER DISSATISF. UNIT" Units: pressure unit/entity Pressure or desire for quicker and more electronic trading per institutional customer. pressure for more floor trade per customer = dissatisfaction effect of market quality on pressure * "REF. PRESSURE PER NON.INST. CUSTOMER" Units: pressure unit/entity Resistance and dissatisfaction pressure per stakeholder due to dissatisfaction with the extent of market quality offered by the NYSE. proportional floor earnings per share traded = floor earnings per share handled * specialist participation Units: $/share Floor earnings per share, adjusted by the extent of specialist participation in trading. It gives a better measure of total earnings. "REF. COMMISSION PER SHARE" = 1 Units: $/share Traditional, fix commission per share.
290
Appendix
"REF. FRACT. CHANGE IN CUST. ORIENT. P.A." = 0.03 Units: Dmnl/(Year*pressure unit) The management team's general flexibility of attention. It may represent the degree to which the organization 'looks outside' and seeks information on important stakeholders. "REF. FRACT. CHANGE IN TRADING PER PRESSURE P.A." = 0.02 Units: Dmnl/(Year*pressure unit) The NYSE's general propensity to react to perceived pressure. It may also represent the degree of decentralization or employee empowerment. "REF. FRACT. INERTIA DECREASE" = 0.15 Units: Dmnl/Year Due to the statement of a NYSE employee that people were grown from within, I assume that only half of them came from ousite the organization. Thus the assumed external turnover rate is half of that of the finance and insurance industry (30.0 % / 2 = 15.0 %). "REF. FRACT. INSTITUTIONALIZATION" = 0.3 Units: Dmnl/Year Institutionalization grows by a fraction of 0.3 of current inertia per year. Since ref. institutionalization is higher than ref. consistency decrease, the organization becomes inert over the years. "REF. FRACTIONAL CHANGE OF VALUATION PER YEAR" = 0.12 Units: Dmnl/Year This variable allows floor culture to diminish only slowly with an average delay of about 8 years. "REF. LIQUIDITY ALGORITHMS" = 1 Units: algorithms "REF. MARKET QUALITY" = 1 Units: Dmnl
Appendix
291
"REF. OPENNESS PER INERTIA" = 1 Units: Dmnl/consistency unit Openness per difference to maximum possible inertia of 1. "REF. POWER OF FLOOR FIRMS" = 100 Units: entity The reference power of floor firms equals the total power of customers—allowing for two theoretically equally powerful groups. Their power balance shifts endogenously over time. "REF. PRESSURE PER DISSATISF. UNIT" = 1 Units: pressure unit/dissatisfaction unit Pressure or customer desire for quicker trading and more electronic trading per institutional customer. "REF. PRESSURE PER NON.INST. CUSTOMER" = 0.5 Units: pressure unit/entity Pressure from dissatisfaction per non-institutional customer is half the pressure of institutional customers since non-institutional customers are less powerful. "REF. RESISTANCE PRESSURE PER FLOOR FIRM" = 1 Units: pressure unit/entity "REF. SP PARTICIPATION" = 0.1 Units: Dmnl Maximum of 10 percent. "REF. SPREAD" = 0.22 Units: $/share It relates to the traditional value of the year 1970. "REF. VALUATION OF FLOOR CULTURE" = 1 Units: valuation unit Maximum.
292
Appendix
"rel. valuation of floor culture" = Valuation of Floor Culture by Floor / "REF. VALUATION OF FLOOR CULTURE" Units: Dmnl relative spread of NYSE = NYSE spread / spread in market Units: Dmnl Determines the average time the NYSE quotes at the NBBO. relative time to execution = NYSE time to execution / time to execution in market Units: Dmnl Attribute that has become highly important for the order routing decision. It expresses the relation between the NYSE's and the market`s speed. resistance pressure for floor system per floor firm = effect of employability on resistance * effect of profitability on resistance * "REF. RESISTANCE PRESSURE PER FLOOR FIRM" Units: pressure unit/entity Resistance per floor firm due to dissatisfaction with the extent to which it is able to participate in trading. SAVEPER = 0.25 Units: Year [0,?] The frequency with which output is stored. specialist participation = "effect of floor trade on sp. participation" * effect of liquidity algorithms on participation * "REF. SP PARTICIPATION" Units: Dmnl Fraction of trades executed against money or shares of a specialist. spread in market = "REF. SPREAD" * effect of relative trading volume on market's spread * "effect of inst. customers on spread" Units: $/share Difference between the bid and the asking price at exchanges other than the NYSE.
Appendix
293
TIME FOR CHANGING MARKET SHARE = 1 Units: Year Short reaction time of one year since market share represents liquidity and routing decision and thus adjusts quickly. TIME STEP = 0.0078125 Units: Year [0,?] The time step between iterations of calculations. TIME TO ADJUST DESD MARKET SHARE = 3 Units: Year Medium adjustment time of 3 years for performance measures. TIME TO ADJUST DESIRED EARNINGS = 5 Units: Year Since earnings deteriorate, the floor gets used to a worse situation rather slowly which explains the relatively long adjustment time of 5 years. TIME TO ADJUST DESIRED PARTICIPATION = 5 Units: Year Since participation deteriorates, the floor gets used to a worse situation rather slowly which explains the relatively long adjustment time of 5 years. TIME TO BECOME CUSTOMER = 5 Units: Year There is a time delay to the indicated fraction so as to neglect shorttime changes and to account for the delay between the creation and the amendment a portfolio of securities. TIME TO CHANGE DESD MARKET QUALITY = 10 Units: Year The adjustment time is long (10 years) because market quality mainly falls and people only slowly get used to a worse situation.
294
Appendix
TIME TO CHANGE POWER OF FLOOR FIRMS = 2 Units: Year Adjustment time of 2 years means a rather quick, but not instantaneous adaptation. TIME TO DEVELOP ALGORITHMS = 1.5 Units: Year Algorithms take about one and a half years to be initiated since the gap needs to be perceived as being problematic. "TIME TO DEVELOP E-TRADE POSSIBILITIES" = 5 Units: Year Delay time between the technical development and implementation of e-trade. The delay of 5 years can be divided into the market's reaction and implementation time. "TIME TO EXECUTION E-TRADE" = 1 Units: second/trade Electronic trading is considered fast trading, and a trade is considered fast if it has sub-second speed. Therefore the time it takes to execute a trade in an electronic environment is set to 1 second. The concept of the time to execution is similar to the notion of latency. Advantages of technology are not be taken into consideration here and a constant time to executon of 1 second in electronic trading is assumed. TIME TO EXECUTION FLOOR TRADE = 9 Units: second/trade The least amount of time it takes to execute a trade manually. According to the NYSE, this is 9 seconds. time to execution in market = "fraction of e-trade in remaining market" * "TIME TO EXECUTION E-TRADE" + ( 1- "fraction of e-trade in remaining market" ) * TIME TO EXECUTION FLOOR TRADE Units: second/trade Speed of trading in the market.
Appendix
295
"TIME TO FULLY IMPLEMENT E-TRADE" = 5 Units: Year Equivalent to the time to develop e-trade possibilities for real data. "TOTAL NO. OF CUSTOMERS" = 100 Units: entity Normalized number of customers. "total pressure for more e-trade from customers" = "pressure for more e-trade per inst. customer" * number of institutional customers Units: pressure unit Pressure by the entire group of institutional customers. total pressure for more floor trade = total pressure for more floor trade from customers + total pressure for more floor trade from floor Units: pressure unit Pressure for more floor trade from non-institutional customers and the floor. total pressure for more floor trade from customers = pressure for more floor trade per customer * "no of non-institutional customers" Units: pressure unit Total pressure by the entire group of non-institutional customers for more floor trade. total pressure for more floor trade from floor = cultural multiplier of pressure from floor * resistance pressure for floor system per floor firm * Power of Floor Firms Units: pressure unit Total pressure by the entire floor for more floor trade.
296
Appendix
TOTAL U.S. SHARE VOLUME IN NYSE- LISTED ISSUES = WITH LOOKUP( Time , ([(1970,0)-(2030,4e+012)], (1970,3.5e+009),(1980,1.4e+010), (1990,4.8e+010),(1995,1.1e+011),(2000,3.2e+011),(2005,5.2e+011), (2007,8.5e+011),(2008,1.2e+012),(2009,1.5e+012),(2011,2e+012), (2015,2.5614e+012),(2020,2.98246e+012),(2030,3.57895e+012) ) ) Units: share/Year Data is taken from NYSE Facts and Figures: Historical > Annual reported volume, turnover rate, reported trades (mils. of shares), and Market Activity > Consolidated Volume in NYSE Listed Issues. Data after 2009 is assumed. Total U.S. Share Volume in NYSE-listed Issues
share/Year
4e+012 3e+012 2e+012 1e+012 0 1970
1980
1990
2000 Date
2010
2020
2030
trading volume of the remaining market = ( 1 - NYSE Market Share ) * TOTAL U.S. SHARE VOLUME IN NYSE- LISTED ISSUES Units: share/Year The trading volume can be used to measure the model’s fit to data. Valuation of Floor Culture by Floor = INTEG( change in valuation , 1) Units: valuation unit The higher the market quality the floor is able to provide, the higher it values its own contribution and culture. The more profitable the floor is, the more it values its own way of doing things. In 1970, it is still at almost 100 percent. "wt. on time vs. spread among all customers" = ACCESS TO INFORMATION TECHNOLOGY * Fraction of Institutional Customers Units: Dmnl Importance of time among customers.
Appendix
297
Appendix E: Generic Model Equations
Attention to Stakeholders Favoring A = INTEG ( - change in attention, 1 - INI ATTENTION TO STAKEHOLDER FAVORING B) Units: Dmnl Orientation towards the stakeholders favoring the ‘old’ strategy A. Attention to Stakeholders Favoring B = INTEG ( change in attention, INI ATTENTION TO STAKEHOLDER FAVORING B) Units: Dmnl Orientation towards the stakeholders favoring the ‘new’ strategy B. change in attention = ( ABS ( pcvd pressure from stakeholders favoring B * effect of attention to B on change ) - pcvd pressure from stakeholders favoring A * effect of attention to A on change ) * "fract. change in attention per pressure p.a." Units: Dmnl/Year change in performance = ( indicated performance – Performance ) / TIME FOR CHANGING PERFORMANCE Units: performance unit/Year change in strategy = ( pcvd pressure from stakeholders favoring B * effect of B on change - pcvd pressure from stakeholders favoring A * effect of A on change ) * "fract. change per pcvd pressure p.a." Units: Dmnl/Year
298
Appendix
confidence effect of performance = WITH LOOKUP ( pcvd adequacy of performance, ([(0,0)-(1.2,1)],(0,0),(0.2,0.04),(0.4,0.14),(0.5,0.22),(0.6,0.33),(0.7,0.5), (0.8,0.75),(0.9,0.95),(0.95,0.985),(1,1),(1.2,1) ) ) Units: Dmnl Effect by which performance inadequacies increase the management team's openness to change. Minor inadequacies have less than proportional effect, but the effect on openness quickly rises before it slowly approaches the limit of a fully open organization in the case of organizational collapse. Confidence Effect of Performance confidence effect
1 0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 pcvd adequacy of performance
1
desired performance = SMOOTH ( Performance, TIME TO ADJUST DESD PERFORMANCE ) Units: performance unit Floating performance goal. desired quality A by stakeholders favoring A = SMOOTH ( quality A, TIME TO ADJUST DESIRED QUALTIY) Units: quality unit Floating goal of desired quality A. desired quality B = diffusion of B in remaining market * "REF.QUALITY B OF STRATEGY B" + ( 1 - diffusion of B in remaining market ) * "REF. QUALITY B OF STRATEGY A" Units: quality unit Expectation by customers/stakeholders.
299
Appendix
DEVELOPMENT OF STRATEGY B = WITH LOOKUP ( Time, ([(0,0)-(50,1)],(0,0),(5,0),(10,0.18),(15,0.57),(20,0.9),(22,0.97),(24,1), (50,1) ) ) Units: Dmnl Invention of strategy B. Development of Strategy B 1
Dmnl
0.75 0.5 0.25 0 0
5
10
15
20 25 30 Time (Year)
35
40
45
50
DEVELOPMENT OF STRATEGY B QUICK = WITH LOOKUP ( Time, ([(0,0)-(50,1)],(0,0),(7,0),(12,1),(50,1) ) ) Units: Dmnl Quicker invention of strategy B, or different reference group. Development of Strategy B Quick 1
Dmnl
0.75 0.5 0.25 0 0
5
10
15
20 25 30 Time (Year)
35
40
45
diffusion of B in remaining market = SMOOTH3 (DEVELOPMENT OF STRATEGY B * ( 1 - SWITCH QUICK DEVELOPMENT ) + SWITCH QUICK DEVELOPMENT * DEVELOPMENT OF STRATEGY B QUICK, TIME TO DIFFUSE B IN REMAINING MARKET ) Units: Dmnl Adoption of strategy B in market.
50
300
Appendix
effect of A on change = WITH LOOKUP ( Orientation to Strategy A, ([(0,0)-(1,1)],(0,1),(0.5,1),(0.75,0.95),(0.9,0.75),(0.95,0.5),(0.99,0.01), (1,0) ) ) Units: Dmnl Limit to the willingness to further react to pressure. Effect of A on Change 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 Orientation to Strategy A
1
effect of attention to A on change = WITH LOOKUP ( Attention to Stakeholders Favoring A, ([(0,0)-(1,1)],(0,1),(0.25,1),(0.5,0.95),(0.8,0.75),(0.9,0.5),(0.96,0.04), (0.98,0.005), (1,0) ) ) Units: Dmnl Limit to the willingness to further react to pressure. Effect of Attention to A on change 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 Attention to Stakeholders Favoring A
1
effect of attention to B on change = WITH LOOKUP ( Attention to Stakeholders Favoring B, ([(0,0)-(1,1)],(0,1),(0.25,1),(0.5,0.95),(0.8,0.75),(0.9,0.5),(0.96,0.04), (0.98,0.005),(1,0) ) ) Units: Dmnl Limit to the willingness to further react to pressure. The graph has the same shape as the one indicating the effect of attention to A on change.
301
Appendix
effect of B on change = WITH LOOKUP ( Orientation to Strategy B, ([(0,0)-(1,1)],(0,1),(0.5,1),(0.75,0.95),(0.9,0.75),(0.95,0.5),(0.99,0.01), (1,0) ) ) Units: Dmnl Limit to the willingness to further react to pressure. The graph has the same shape as the one indicating the effect of A on change. effect of change on inertia = WITH LOOKUP ( ABS ( change in strategy ), ([(0,0)-(0.5,7)],(0,1),(0.05,1.4),(0.1,2.4),(0.15,4.2),(0.2,5.4),(0.3,6.2), (0.5,6.5) ) ) Units: Dmnl Small changes have a less than proportional effect on consistency loss. This allows an organization to change incrementally without disruption in its internal consistency. The consistency decrease from change represents turnover rates, but it also captures changes in the people's thinking even if they remain in the organization. Effect of Change on Inertia 8
effect
6 4 2 0 0
0.100
0.200 0.300 change in strategy
0.400
0.500
302
Appendix
effect of openness on change = WITH LOOKUP ( openness to change, ([(0,0)-(1,1)],(0,0.05),(0.1,0.06),(0.2,0.1),(0.3,0.18),(0.4,0.3),(0.5,0.435), (0.6,0.63), (0.7,0.81),(0.8,0.92),(0.9,0.97),(1,1) ) ) Units: Dmnl Low openness to change may reduce fractional change to 10 percent of its reference value. The effect of openness on change is an s-shaped curve indicating that the organization quickly reacts to perceived pressure if it has a rather high openness. It becomes less responsive as openness decreases until its reactivity reaches a lower bound. Effect of Openness on Change 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 openness to change
1
effect of quality A on performance = WITH LOOKUP ( quality A, ([(0,1)-(1,1.1)],(0,1),(1,1.1) ) ) Units: Dmnl Effect that pushes performance upward proportionally to the extent to which the organization outperforms in quality A. Effect of Quality A on Performance 1.1
effect
1.075 1.05 1.025 1 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 quality A
1
303
Appendix
effect of quality A on resistance = WITH LOOKUP ( pcvd adequacy of quality A, ([(0,0)-(1.1,1)],(0,1),(0.1,0.99),(0.15,0.97),(0.2,0.93),(0.5,0.5),(0.8,0.07), (0.85,0.03),(0.9,0.01),(1,0),(1.1,0) ) ) Units: Dmnl Inversely s-shaped. Slowly approaches maximum, and slowly starts in the beginning because minor inadequacies cause less than proportional reactions. Effect of Quality A on Resistance 1
effect
0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 pcvd adequacy of quality A
1
"effect of rel. quality B on performance" = WITH LOOKUP ( "rel. quality B", ([(-0.9,0)-(1,2)],(-0.9,0),(-0.7,0.53),(-0.6,0.7),(-0.5,0.85),(-0.4,0.93), (-0.3,0.97),(-0.2,0.99),(0,1),(0.25,1.005),(0.5,1.07),(0.7,1.2),(1,1.4) ) ) Units: Dmnl Effect that may push performance upward or downward depending on the organization's achievement regarding quality B relative to the market. It is formulated as an order winning criterion. Effect of Relative Quality B on Performance 2
effect
1.5 1 0.5 0 -1
-0.60
-0.20 0.20 "rel. quality B"
FINAL TIME = 50 Units: Year Time bounds of the simulation.
0.60
1
304
Appendix
"fract. change in attention per pressure p.a." = "REF. FRACT. CHANGE IN ATTENTION P.A." * effect of openness on change Units: Dmnl/(pressure unit*Year) Flexibility of attention. Mix of the management team's general flexibility of attention and situational factors. "fract. change per pcvd pressure p.a." = "REF. FRACT. CHANGE IN STRATEGY PER PRESSURE P.A." * effect of openness on change Units: Dmnl/pressure unit/Year Responsiveness of the strategy to pressure. Mix of the management team's general responsiveness to pressure and situational factors. fraction of stakeholders favoring B = diffusion of B in remaining market Units: Dmnl It is assumed that the market is adapted to the demands of market participants. indicated performance = "REF. PERFORMANCE" * performance adjustment Units: performance unit Inertia = INTEG ( institutionalization-inertia decrease, INI INERTIA ) Units: consistency unit Inward-orientation of thinking, cognitive inertia, … inertia decrease = Inertia * "REF. FRACT. INERTIA DECREASE" * effect of change on inertia Units: consistency unit/Year Management team turnover, unlearning, … INI ATTENTION TO STAKEHOLDER FAVORING B = 0.1 Units: Dmnl The initial attention represents the minimum amount of attention that the management team attributes to its stakeholders.
305
Appendix
INI INERTIA = 0.9 Units: consistency unit Initial value = effect of (ref. fract. consistency decrease / ref. fract. institutionalization) = 0.9 INITIAL TIME = 0 Units: Year Initial time bounds of the simulation. institutionalization = "REF. FRACT. INSTITUTIONALIZATION" * Inertia * limiting effect on institutionalization Units: consistency unit/Year Growth of inertia, e.g. by cultural institutionalization, learning, etc. limiting effect on institutionalization = WITH LOOKUP ( Inertia, ([(0,0)-(1,1)],(0,1),(0.2,1),(0.4,0.99),(0.6,0.9),(0.75,0.75),(0.9,0.5), (0.97,0.25),(1,0) ) ) Units: Dmnl This effect counteracts the reinforcing institutionalization loop. The more the organization is consistent, the more the effect slows consistency growth down. Limiting Effect on Institutionalization
limiting effect
1 0.75 0.5 0.25 0 0
0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 Inertia
no of stakeholders favoring A = "TOTAL NO. OF STAKEHOLDERS" - number of stakeholders favoring B Units: entity Normalized number of stakeholders.
1
306
Appendix
number of stakeholders favoring B = fraction of stakeholders favoring B * "TOTAL NO. OF STAKEHOLDERS" Units: entity Normalized number of stakeholders. openness to change = 1 - Inertia * confidence effect of performance * "REF. OPENNESS PER INERTIA" Units: Dmnl Readiness to change that is limited by inertia, but may be inhanced in the case of a performance threat. Orientation to Strategy A = INTEG ( - change in strategy, 1 ) Units: Dmnl Fraction to which the focal organization's strategy is oriented to the ‘old’ strategy A. Orientation to Strategy B = INTEG ( change in strategy, 0 ) Units: Dmnl Fraction to which the focal organization's strategy is oriented to the ‘new’ strategy B. pcvd adequacy of performance = Performance / desired performance Units: Dmnl pcvd adequacy of quality A = quality A / desired quality A by stakeholders favoring A Units: Dmnl
307
Appendix
pcvd inadequacy of strategy per stakeholder B = WITH LOOKUP ( "rel. quality B", ([(-1,0)-(1,1)],(-1,1),(0,0),(1,0) ) ) Units: Dmnl Stakeholders' extent of dissatisfaction with or dislike of the focal organization's strategy/offerings. Perceived Inadequacy of Strategy per Stakeholder B pcvd inadequacy
1 0.75 0.5 0.25 0 -1
-0.60
-0.20 0.20 "rel. quality B"
0.60
1
pcvd pressure from stakeholders favoring A = total stakeholder pressure for more A * Attention to Stakeholders Favoring A Units: pressure unit The management team's biased perception of stakeholder pressure for A. pcvd pressure from stakeholders favoring B = total stakeholder pressure for more B * Attention to Stakeholders Favoring B Units: pressure unit The management team's biased perception of stakeholder pressure for B. Performance = INTEG ( change in performance, "REF. PERFORMANCE" * effect of quality A on performance) Units: performance unit May represent market share, sales volume, size of customer base, etc. performance adjustment = "wt. on quality B vs. quality A" * "effect of rel. quality B on performance" + ( 1 - "wt. on quality B vs. quality A" ) * effect of quality A on performance Units: Dmnl Upward or downward adjustment of performance by quality A and B.
308
Appendix
PERMANENTLY POWERFUL STAKEHOLDERS FAVORING A = 0 Units: entity A group that can significantly impede an organization’s operations such as change-averse employees. quality A = Orientation to Strategy A * "REF. QUALITY A OF STRATEGY A" Units: quality unit Quality of the 'old' strategy, such as price quality in trading, resolution quality in photography. This is what organizations in the respective area used to compete on. quality B = Orientation to Strategy B * "REF.QUALITY B OF STRATEGY B" + Orientation to Strategy A * "REF. QUALITY B OF STRATEGY A" Units: quality unit Achievement in the ‘new’ quality by focal organization. "REF. FRACT. CHANGE IN ATTENTION P.A." = 0.05 Units: Dmnl/pressure unit/Year The management team's general flexibility of attention. It represents the degree to which the organization 'looks outside' and seeks information on important stakeholders. "REF. FRACT. CHANGE IN STRATEGY PER PRESSURE P.A." = 0.02 Units: Dmnl/pressure unit/Year An organization's general propensity to react to perceived pressure. It may represent the degree of decentralization or employee empowerment. "REF. FRACT. INERTIA DECREASE" = 0.15 Units: Dmnl/Year Reference inertia decrease has been adapted to a rather low rate of annual turnover in order to represent an organization that accumulates inertia rather quickly.
Appendix
309
"REF. FRACT. INSTITUTIONALIZATION" = 0.3 Units: Dmnl/Year Institutionalization grows by a fraction of 0.3 of current inertia per year. Since ref. institutionalization is higher than ref. consistency decrease, the organization becomes inert over the years. "REF. OPENNESS PER INERTIA" = 1 Units: Dmnl/consistency unit Openness per difference to maximum possible inertia of 1. "REF. PERFORMANCE" = 0.5 Units: performance unit "REF. PRESSURE PER STAKEHOLDER FAVORING B" = 0.6 Units: pressure unit/entity Medium extent of pressure or desire of new stakeholder for strategy B. "REF. QUALITY A OF STRATEGY A" = 1 Units: quality unit Attribute of the new strategy. High value of quality A. "REF. QUALITY B OF STRATEGY A" = 0.1 Units: quality unit Degree to which strategy A can fulfill quality B. Strategy A has a low value of quality B. "REF. RESISTANCE PRESSURE PER STAKEHOLDER FAVORING A" = 1 Units: pressure unit/entity High extent of resistance pressure of old stakeholder for old strategy. "REF.QUALITY B OF STRATEGY B" = 1 Units: quality unit An attribute of the new strategy B. E.g. speed as the attribute of electronic trading, ability to store photos electronically, high ethical compliance, etc.
310
Appendix
"rel. quality B" = quality B - desired quality B Units: quality unit The relative quality B espresses the difference between the focal organization's quality B and what is desired by customers/stakeholders. SAVEPER = 0.25 Units: Year [0,?] The frequency with which simulation output is stored. stakeholder pressure for more B = pcvd inadequacy of strategy per stakeholder B * "REF. PRESSURE PER STAKEHOLDER FAVORING B" Units: pressure unit/entity Pressure or customer desire for more strategy B per stakeholder favoring B. stakeholder resistance pressure for more A = "REF. RESISTANCE PRESSURE PER STAKEHOLDER FAVORING A" * effect of quality A on resistance Units: pressure unit/entity Resistance per stakeholder due to dissatisfaction with the extent of quality A offered. SWITCH QUICK DEVELOPMENT = 0 Units: Dmnl Can switch on and off a different environment. TIME FOR CHANGING PERFORMANCE = 1 Units: Year Reaction time of stakeholders. TIME STEP = 0.0078125 Units: Year [0,?] The time step between iterations of calculations.
Appendix
311
TIME TO ADJUST DESD PERFORMANCE = 3 Units: Year Medium delay time. TIME TO ADJUST DESIRED QUALTIY = 5 Units: Year Long delay/adjustment time to changes in desired attributes. TIME TO DIFFUSE B IN REMAINING MARKET = 5 Units: Year Time delay between the invention and implementation of the new strategy in the market. "TOTAL NO. OF STAKEHOLDERS" = 100 Units: entity Normalized number of stakeholders. total stakeholder pressure for more A = stakeholder resistance pressure for more A * ( no of stakeholders favoring A + PERMANENTLY POWERFUL STAKEHOLDERS FAVORING A ) Units: pressure unit Total pressure by the entire group of stakeholders favoring A for more A. total stakeholder pressure for more B = stakeholder pressure for more B * number of stakeholders favoring B Units: pressure unit Total pressure by the entire group of stakeholders favoring B for more B. "wt. on quality B vs. quality A" = fraction of stakeholders favoring B Units: Dmnl Importance of strategy B and quality B among stakeholders.
