Providing Public Goods in Transitional China

P rov i di ng P u bl ic G o od s i n Tr a nsi t iona l Ch i na

This page intentionally left blank

P rov i di ng P u bl ic Go od s i n Tr a nsi t iona l Ch i na

Ton y Sa ic h

PROVIDING PUBLIC GOODS IN TRANSITIONAL CHINA

Copyright © Tony Saich, 2008. All rights reserved. First published in 2008 by PALGRAVE MACMILLAN® in the US—a division of St. Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Where this book is distributed in the UK, Europe and the rest of the world, this is by Palgrave Macmillan, a division of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN-13: 978–0–230–60951–8 ISBN-10: 0–230–60951–1 Library of Congress Cataloging-in-Publication Data Saich, Tony. Providing public goods in transitional China / Tony Saich. p. cm. Includes bibliographical references and index. ISBN 0–230–60951–1 1. Human services—Government policy—China. I. Title. HV418.S24 2008 361.60951—dc22

2008012347

A catalogue record of the book is available from the British Library. Design by Newgen Imaging Systems (P) Ltd., Chennai, India. First edition: October 2008 10 9 8 7 6 5 4 3 2 1 Printed in the United States of America.

Con t e n t s

List of Tables and Figures

vii

Preface

ix

Acknowledgments

xi

List of Abbreviations The Sample Survey of Citizens’ Attitudes about Chinese Government Performance Public Goods Regimes and Social Welfare Provision in China Two Welfare Provision, 1949–1979 Three The Provision of Public Goods during the Reform Period Four Challenges in the Health System Five Dealing with Public Health Crises and Pandemics Six Building a New Social Insurance System Seven Poverty Relief and Social Assistance Eight Providing Social Welfare: States, Markets, and Civil Society

xiii xv

One

1 23 43 71 103 127 159 185

Notes

209

Bibliography

213

Index

231

This page intentionally left blank

L i st of Ta bl e s a n d Fig u r e s

Tables 1.1 Public and Private Provision of Services 1.2 Public and Private Expenditure on Healthcare of Selected Countries, 2004 1.3 Social Indicators for Selected Countries 2.1 Comparative Indicators for China’s Social Development, 1950 to 1980 3.1 China Human Development Index Comparison 3.2 Social Indicators for Selected Asian Countries 3.3 Ratio of Central-Local Revenues and Expenditures 3.4 Central and Local Government Expenditure on Social Welfare, 2005 4.1 Comparative Health Indicators for Selected Countries 4.2 Structure of Health Expenditure, 1980–2005 4.3 Medical Insurance Coverage, 1998 and 2003 4.4 Rural-Urban Health Indicators, 2005 4.5 Xiaokang Indicators and Targets for Selected Years 5.1 Reported Incidence and Death Rate of Infectious Diseases for Selected Years 5.2 Incidence Rate for Selected Contagious Diseases, 2006 5.3 Source of HIV Infections 6.1 China’s Urban Social Insurance Schemes 6.2 Participation in Different Urban Insurance Schemes 6.3 Social Security Subsidies, 1999 and 2005 6.4 Basic Pension Coverage, 2005 6.5 Pension Coverage in Selected Provinces, 2005 6.6 Insurance Coverage for Respondents in Our Survey, 2005 7.1 Official Rural Poverty Statistics

7 15 16 41 47 48 60 62 76 77 78 80 84 105 105 114 132 133 139 148 149 150 162

viii / list of tables and figures

7.2 7.3 7.4 7.5

Recipients of the MLSS and Other Relief Provision of the MLSS in Selected Municipalities Rural Poverty and Relief Support Regional Distribution of Minimum Living Standard Support, 2005 8.1 Chinese Residents’ Satisfaction with the Lowest Level of Government Service and Performance by Place of Residence, 2005 8.2 Government Service Satisfaction/Importance Matrix Template 8.3 Social Organizations and Civil Noncommercial Organizations

175 178 180 180 188 189 197

Figures 3.1 Organization of Urban Government 8.1 Citizens’ Level of Satisfaction with Different Levels of Government 8.2 Government Performance Matrix, 2003 8.3 Government Performance Matrix, 2005

57 187 190 191

P r e fac e

This book analyzes how the provision of basic goods and services has been affected by the economic reforms that the Chinese leadership launched in earnest in 1978. As is the case in most socialist regimes, China’s welfare indicators before reform were good given the level of economic development—this is so despite the huge loss of life caused by the Maoist-inspired movements of the Great Leap Forward and the Cultural Revolution. Although the “barefoot doctors” were much derided by the post-Mao leadership, much of this denunciation was political and for many rural areas there was a basic referral network in place. The administrative structures that supervised much of daily life also allowed for preventive healthcare and education programs to be put in place. Provision was basic, and availability and quality depended on where one lived. This picture changed dramatically with economic reforms. The leadership does not seem to have considered the impact on the provision of public goods and services that economic reform would bring. First, with the dismantling of the collective institutions in the countryside and subsequently with the curtailment of workplace benefits in urban China, social support weakened or entirely evaporated for many. The market did not provide an adequate solution; with weak institutions of civil society, levels of coverage began to drop and access to key services became more dependent on income than ever before. This book looks at how China’s leaders have responded to this new set of challenges. Although General Secretary Hu Jintao and Premier Wen Jiabao have committed the regime to deal with issues of social justice and have pledged to provide equitable access to all citizens, important experimentation began under the leadership of Jiang Zemin and Zhu Rongji. Already from the early 1990s experiments were launched to restructure medical care and pension provision, and these intensified in the mid-1990s as the reform of the state-owned sector began to bite. The reforms to date raise interesting questions about the proper role of the state in the delivery of social welfare services and about how this role

x / preface

should be augmented by the provision of services through the market and by civil society actors. It is clear that the state cannot provide all the services that it promises, but it has been unclear how much authority it is willing to cede to nonstate organizations. In this sense, the study touches on the broader question of changing state-society relations and may provide some insights into broader questions of reform of the Chinese body politic. It is hard for the Chinese Communist Party that claims to represent the whole of society to provide access to certain categories of citizens while denying it to others. This raises fundamental questions about the meaning of citizenship in contemporary China. It also provides the opportunity for nonstate actors to expand their sphere of activity under the rubric of supporting state-mandated provision of basic services and through the provision of services to groups that are either underserved or not served at all by state agencies. Clearly, the situation in China is changing quickly but it is hoped that the trends identified in this book will be relevant for some time to come. The specific figures may change and policy will evolve, but the fundamental challenges outlined in the subsequent chapters and the Chinese leadership’s thinking about how to address them will remain more or less the same.

Ac k now l ed gm e n t s

First, I would like to thank Steven Kennedy and Anthony (Toby) Wahl at Palgrave for encouraging me to submit this manuscript to them for possible publication. Toby provided strong support to ensure that this book would see the light of day. I am also very grateful to the anonymous reader who reviewed the entire manuscript and made a number of helpful suggestions to improve its presentation and structure. Of course, any remaining flaws and faults remain my responsibility alone. I have been muddling my way along to writing this book for a number of years. I started from an interest in the increasing role of civil society in China and the limitations that are imposed on its capacity to act effectively. Much of my thinking around this question was formed during the years that I worked as the Representative for the Ford Foundation in China (1994–1999). Looking at the role of civic institutions in providing public goods and social welfare seemed a natural fit; after moving to Harvard Kennedy School, I began to research into this in more detail. I would like to thank students who have attended my courses on China for the feedback they have given me on some of the ideas in this book. I would also like to thank the Chinese government officials who participated in our training programs for provincial-level officials. I discussed the ideas present in Chapters One and Eight and received interesting feedback. I was dragged into work around issues of public health in China by a number of my colleagues from Harvard, especially Dean Barry Bloom, Professor Arthur Kleinman, Dr. Joan Kaufman, and Dr. Yuanli Liu. Through discussions and joint workshops and publications, I have benefited from their wisdom on a number of issues relevant to this book. I have also benefited from discussions with other colleagues: Professor Mary Jo Bane, Dr. David Dapice, Dr. Arnold Howitt, Professor Dutch Leonard, Professor Mark Moore, and Dr. Jay Rosengard. I still owe a debt to my former colleagues at the Ford Foundation, especially Stephen McGurk who is very knowledgeable on social welfare issues in the mid-1990s. I relied on discussions with Arthur Holcombe (former head of UNDP in China) for

xii / acknowledgments

improving my understanding of his work in Tibet and of poverty alleviation policy more generally. Over the last couple of years, I have been a Changjiang Scholar at the School of Public Policy and Administration, Tsinghua University, and benefited enormously from my interactions there. I would especially like to thank Professor Lan Xue and Ms. Bo Meng for creating such a welcoming environment for me. I have benefited from the advice of many Chinese colleagues such as Mr. Jiange Li, Professor Xiulan Zhang, Mr. Mai Lu, Mr. Zhenyao Wang, Mr. Qunshan Zhang, Professor Zuoxue Jin, Dr. Heng Quan, and Professor Biliang Hu. I owe a particular debt of thanks to Mr. Yue (Victor) Yuan and his colleagues at the Horizon Market Research Company (especially Shen Min). They helped me design and administer the survey that we conducted in 2003, 2004, and 2005. Their help and support was fantastic and they also helped with providing the first cut of the data. In addition to the work on the survey, I have learned much from Victor about how to interpret development in China over the years. Nancy Hearst, librarian of the Fairbank Center Collection in the Fung Library at Harvard, kept me on the straight and narrow as usual. Her sharp eye ensured a number of stylistic and presentational improvements and her help with finding references and compiling the bibliography was invaluable. I just hope that she does not decide to retire before I do. T S January 2008

List of Abbr e v i at ions

ABC ADB AIDS ARV CC CCP CDC CDRF CHED CSW DRC EBF FBD FBIS-CHI FEER GDP GONGO HDI HIV HPI IDU KSG MCH MCT MLSS MOA MOCA MOH MOLSS MSM NGO

Agricultural Bank of China Asian Development Bank Acquired Immune Deficiency Syndrome Antiretroviral Therapy Central Committee Chinese Communist Party Center for Disease Control China Development Research Foundation Center for Human and Economic Development Commercial Sex Worker Development Research Center (of the State Council) Extra-Budgetary Fund Former Blood and Plasma Donor Foreign Broadcast Information Service—China Far Eastern Economic Review Gross Domestic Product Government-Organized Nongovernmental Organization Human Development Index Human Immunodeficiency Virus Human Poverty Index Intravenous Drug User Kennedy School of Government Maternal and Child Health Mother-to-Child Transmission Minimum Living Standard Scheme Ministry of Agriculture Ministry of Civil Affairs Ministry of Health Ministry of Labor and Social Security Men Who Have Sex With Men Nongovernmental Organization

xiv / list of abbreviations

NPC NSB OECD PRC PSU ROSCA SARS SCMP SOE STD TVE UNDP UNICEF WHO WTO

National People’s Congress National Bureau of Statistics Organisation for Economic Co-operation and Development People’s Republic of China Public Service Unit Rotating Savings and Credit Association Severe Acute Respiratory Syndrome South China Morning Post State-Owned Enterprise Sexually Transmitted Disease Township and Village Enterprise United Nations Development Programme United Nations Children’s Fund World Health Organization World Trade Organization

Th e Sa m p l e Su rv e y of Ci t i z e ns’ At t i t u de s a b ou t Ch i n e se Gov e r n m e n t P e r for m a nc e

In the text “our survey” refers to surveys conducted in 2003, 2004, and 2005. Here we use the 2003 survey as exemplary.1 The findings and analysis of the survey are based on the results of a purposive stratified survey of 3,967 respondents conducted by Horizon Research Group in the fall of 2003.2 Respondents, aged 16 through 60, were selected from three administrative levels: city, township, and village. At the city level, seven sites were chosen based on their geographic location, average per capita income, and population. The sites varied in all three variables, representing lower-middle income, middle income, and upper income individuals, as well as western, eastern, northern, and southern populations of China. Within the cities respondents were randomly selected through the household registration lists using the KISH method. This uses household area sampling according to a “face sheet” or table with fractional representation of each adult (Kish: 1949; Kish: 1965). At the township and village levels, because of the large average size of families, respondents were selected randomly through the closest birthday method. No fewer than 250, 150, and 100 respondents were identified for each city (1750 total), town (1050 total), and village (800 total) respectively. Consequently, the sample has an urban bias, resulting in respondents with higher age ranges and, in some cases, higher income levels than the corresponding regional averages. In the analysis stage, the results were weighted to compensate for both urban bias and relative population size. Thus, the final weight for cities in terms of weighting for urban sites was 0.5008 while that for villages was 2.6487. With the exception of those in the age range of 16 to 19, the demographic profile mirrors reasonably well the national range. The survey forms contained around 50 questions, with slight distinctions for the rural and urban respondents to address minor discrepancies in income measurements and employment responses. Interviewers recorded multiple choice, binary, and free-response answers. Care was taken to record respondent demographic data when available, such as profession,

xvi / sample survey of citizens’ attitudes

gender, age, household registration, educational attainment, employment status, private/public enterprise or NGO employer, income range, etc. All interviews were conducted by professional staff trained and employed by Horizon Research Group. The response rate was around 80 percent for all three years. By design, the sample does not include migrants or the majority of ethnic minorities living in China. Using the household registration system does not capture migrant communities. This is a disadvantage, given the high numbers of migrants who are now moving into Chinese cities in search of work. However, it does allow us to record data on formally registered citizens’ attitudes toward different kinds of services and to permit comparison across rural and urban China. In most cases, migrants lack legal access to public goods and services and their responses, therefore, create bias in the survey findings. Similarly, ethnic minorities residing in autonomous regions live under varying policy frameworks, rendering a comparison of government performance between regions difficult at best and misleading at worst. The questionnaire was conducted in the two municipalities of Beijing and Shanghai and the five cities of Guangzhou, Wuhan, Chengdu, Shenyang, and Xi’an, with Nantong (Jiangsu) added in 2005. Seven small towns (districts or counties) were covered: Zhuji in Shaoxing (Zhejiang), Changle in Fuzhou (Fujian), Beining in Jinzhou (Liaoning), Xinji in Shijiazhuang (Hebei), Linxiang in Yueyang (Hunan), Pengzhou in Chengdu (Sichuan), and Xingping in Xianyang (Shaanxi). Seven villages under these were chosen, as well as Feng Shuling village under the jurisdiction of Wuhan (Hubei), to make eight villages in total.

Ch a p t e r O n e P u bl ic Go od s R e g i m e s a n d Soc i a l We l fa r e P rov i sion i n Ch i na

The reform program in China since 1978 has left no sector of state and society untouched. Although the economic reforms have raised the standard of living for the vast majority and shifted China along the road to a market economy, they have also resulted in a significant restructuring of the provision of public goods and services that has affected every citizen. State and collective institutions in rural and urban China that previously carried much of the welfare burden have been dismantled and policymakers have struggled to devise new policies and institutions to carry the burden. As a result, new forms of inequality have emerged, with a dramatic rise in the disparity of service provision between rural and urban China and an abandonment of the compact for cradle-to-grave social welfare for the privileged working class and government employees. The fiscal decentralization that has been a key component of the economic reforms has contributed further to the new inequalities. Many localities have found it difficult to secure a sound funding base for the provision of public services. As in other transitional economies, Chinese policymakers have found it difficult to design new welfare systems, mostly because significant institutional change is inherently slower and more complex than macroeconomic stabilization and liberalization measures (see Kornai: 1997; Nelson: 1997; on the problem in transitional economies generally, see Bird and Wallich: 1993). It took Chinese policymakers some time to start building a new system to replace the old one that was frayed in the urban areas and had collapsed in rural China (Lindbeck: 2006, p. 47). The leaders placed their faith in rapid growth raising all boats and in hopes that the market would provide the necessary services in sufficient quantities. In certain respects, by the late 1990s, the social welfare system in China had begun to resemble that of other countries in East Asia where the

2 / providing public goods in transitional china

state has been reluctant to take on too great a responsibility for fear of creating a culture of dependency and diverting resources from investment in economic development. Policy in China for social welfare provision and public goods more generally also resembles a worldwide trend away from an expansion of the welfare state. We see a retrenchment of service provision, with the market, civil society, and individuals taking on a greater role, and with the state strengthening its role as a regulator rather than as a direct service provider. From the late 1990s and accelerating with the new leadership that took over in 2002, policy has emphasized a strategic role for the state to provide minimum guarantees and to serve as a provider of last resort. Programs such as rural pension provision are extremely ambitious for a developing country. These shifts in service provision entail profound consequences for the relationship between state and society, with an increasing role for nonstate providers in the allocation of goods and services. They require a significant reorientation of the state’s role and also have consequences for what it means to be a citizen in contemporary China. As citizens become less reliant on the state to provide key services, this may impact on their expectations and appreciation of government and how they view accountability. The provision of specific social welfare goods and services to certain sectors of the population, while curtailing them or even denying them to others, has significant consequences for the meaning of citizenship. In a system that still brandishes its socialist credentials, it is difficult to justify leaving certain sectors of the population to the mercy of the market’s vicissitudes or family support. At the same time, it is hard to justify the large inequalities that have arisen regionally and between urban and rural China as an integral part of the reform program (see Dollar: 2007; Khan and Riskin: 1998, 2001). This conundrum has not been lost on the leadership under General Secretary Hu Jintao and Premier Wen Jiabao who consolidated power from 2002. Although China’s achievements in social development have been impressive, especially for a low-income country, progress slowed in the 1990s and new challenges arose as a consequence of reforms. As a result, the Hu-Wen leadership has focused on achieving balanced growth that is more equitable in distribution and that will produce a comfortable (xiaokang) society by the year 2020. The new leaders have eschewed the almost exclusive focus on GDP growth rates that dominated much of the 1990s and are now trying to develop a more comprehensive framework for development. They hope this approach will address the concerns of those who have not benefited so well from reforms (many farmers and grain producers), new groups that straddle the now artificial urban-rural divide (migrant families), and the new urban poor who are products of reforms (pensioners and laid-off state enterprise workers). The leadership is

public goods regimes and social welfare / 3

struggling to weave together the frayed and fragmented strands of welfare into a more coherent whole where government replaces the workplace as the guarantor of minimum services. Yet, it wants to achieve this objective without failing into the trap of “welfarism.” Standards of living have been raised for the vast majority and, for those able to afford it, there is a wider range of education, health, and other service providers from which to choose. However, there are still many who cannot afford the higher costs of public services and there is often no service provider available for the absolute poor. Access to public services is more dependent on income now than in the past and this has created new inequalities and has exacerbated old ones. This book looks specifically at how the delivery of social welfare as a specific type of public good has been affected by the reforms to analyze how the relationship between the state, market, and civil society has been changing. With an increased role for the market to allocate goods and a necessary but reluctantly tolerated use of nongovernmental service providers, what kind of public goods regime is emerging in China? We seek to answer this question through case studies of healthcare provision, dealing with pandemics, social insurance, and social assistance. Public Service Delivery and Public Service Regimes Development theory has undergone a number of cycles in its thinking about the role of the state in providing public goods and services. The traditional model, in theory, claimed that the government decided on priorities (with varying degrees of public input), mobilized the resources, and often provided the service through government agencies. It is not surprising then that the reforms through the 1980s and 1990s sought to roll back the state’s heavy engagement in service provision (for an early critique, see Hayek: 1944). It was presumed that the market would “work its magic” with a shift from “getting the policies right” to “getting the prices right.” Most of the debate during this period became a discussion of state versus markets. However, the spectacular growth of certain countries in East Asia led analysts to revisit the appropriate role of the state in development. This was combined with the thinking of those such as Amartya Sen (1999) who questioned whether a focus on GDP growth rates alone was a sufficient objective of development. Sen suggested that governments needed to think about the quality of growth and whether increasing GDP improves the basic livelihood of citizens. Under the banner of “Development as Freedom,” Sen called for development that would enhance people’s capabilities to function

4 / providing public goods in transitional china

in the modern world. This was reflected in the UNDP’s measurement of the Human Development Index to complement the fixation on assessing a country’s GDP growth rates with an interest in the human capabilities approach. Equitable access to health and basic guarantees against the vicissitudes of the market and life shocks are important components to enhance human capabilities. Such perspectives lead us to reflect on fundamental questions concerning the changing role of the state in providing welfare and how different levels of government can combine with market-driven approaches and the institutions of civil society to enhance citizens’ access to available welfare services and their abilities to improve them. Choice becomes an important element for citizens in deciding on the kinds of services they would like to enjoy and how to access them (Prabhaker: 2006; Fung and Wright: 2003). Current policy trends promote an approach that reduces the state’s role as a direct provider and strengthens its role as an enabler or guarantor of services. Thus, we see the emergence of descriptions such as “the enabling state” (Gilbert: 2002) or “the social investment state” (Giddens: 2000). Importantly, the approach that focuses on capabilities and the use of the Human Development Index raises fundamental questions about equity in development and how the government can ensure access to all its citizens rather than simply concentrating resources on those who are better connected or organized. Development in Japan, South Korea, and Taiwan has shown that the posited trade-off between growth and equity that many see as inevitable may not necessarily exist. If there are properly designed policies for basic education and healthcare, economic growth can be maintained even as poverty is reduced and relative equity is maintained. Experience in these countries also suggests that good development is impossible without an effective state. But this should not be confused with a strong, overbearing, and interventionist state. Thus, although it may appear as if we are simply putting old wine in new bottles, there has been learning and we are reaching a more mature understanding of the ways in which governments can partner with other institutions in the marketplace or civil society to maximize the public good. There has been a gentle upward spiral rather than a never-ending, flat circle. Although the Chinese governmental structure might seem far removed from that of the United Kingdom or New Zealand, there are the same kinds of pressures that have resulted in what is termed “new public management” impacting on institutions and policy (Kamarck: 2004; Kettl: 2005). First, there is a focus on economic liberalization, with governments trying to cede more functions to the market combined with the diversification of state ownership. The state’s role has shrunk through devolution and de

public goods regimes and social welfare / 5

facto privatization, while the range of state intervention in the economy has narrowed, with policy instruments focusing more on macroeconomic controls and less on direct administrative controls (on the politics of privatization, see Feigenbaum et al.: 1998). Since the mid-1990s, China has accelerated its policy to reduce the scale of the state-owned sector in the economy. This has led to the contracting out of many government services and the conversion of many government public service units (shiye danwei) to either for-profit or not-for-profit organizations. Second, while reducing state functions there has been concern to improve the quality of those functions that remain (see Jiang Zemin: 2002). Thus, we see an emphasis on administrative reform of the core state functions that seek to cut the size of government bureaucracies while trying to make its agencies more efficient, modern, and responsive to citizen demand. This entails extensive experimentation in fields such as budget and financial reform, regulatory reform, and customer service initiatives. It is clear that even in the most marketized of systems, government is crucial in providing some support for those risks that cannot be left to the market or that arise because of market failure and that cannot be covered by the institutions of civil society (Moss: 2002; Grout and Stevens: 2003). This is particularly important in an environment where the market is distorted and is replete with information asymmetries and also where civil society is weak. Starting from the 1990s, we see a continuous set of experiments in China to put new systems in place and to figure out the correct role of government in the provision of social welfare. However, this has created the challenge of managing economically dysfunctional risks without holding back economic progress (Moss: 2002, p. 6). Which risks government chooses to manage and which risks it chooses to leave to other institutions will vary over time and will be dependent on the priorities of the political elites and the prevailing social mores. For example, while we see worldwide a strong shift toward the private delivery of many public services, the continued stress on socialism in the Chinese Communist Party (CCP) rhetoric has left the party open to attack for abandoning vulnerable groups to the vagaries of the market. As we shall see in Chapter Four, this has particularly been the case with respect to healthcare provision in poor rural areas. Given the role the state and workplaces have played in the provision of public goods in China, it is surprising how little attention the government paid in the early years of the transition to analyzing the provision of such services. Much of the policy development during the first two decades of reform, with the exception of a few areas such as urban pension reform, was by default rather than by design. Since the late 1990s and certainly since 2002–03, there has been a concerted attempt to define a more coherent

6 / providing public goods in transitional china

role for the government in service provision. Similarly, little attention was paid to understanding markets and how they allocate resources. In some ways, there was a reification of markets during the 1990s and their capacity to allocate goods effectively. In addition, we see the logic and thinking of the market penetrating the psyche of individual Chinese and it is a powerful principle. As Mark Moore (2001, p. 307) has noted: “Eventually, the market becomes not only a way of organizing the economy of a society but also a way for society’s members to think about what they should value as individuals, how they might combine together to produce valuable social results, and how lines ought to be drawn in society between the private and the public, the individual and the collective, and the voluntary and the obligatory.” In this book we adopt commonly used definitions for public and private goods. A pure public good is entirely nonrival and nonexcludable in terms of its consumption. Thus, there is no rivalry between consumers for the good; no matter how many people consume the good, the cost of production does not vary. Nonexcludability means that once produced, no one can be excluded from its consumption (see, for example, Samuelson: 1954; Foldvary: 1994). By contrast, pure private goods are produced and distributed through the market. They are by nature excludable and can be consumed or transferred only with the consent of their owners (Sargeson: 2002, p. 6). In reality, of course, there are very few goods that explicitly fall into one category or the other and this is complicated by the fact that while the state might define a service as a public good (for example, basic education in China), the method of financing and/or provision may require the involvement of nonpublic agencies. The World Bank (2005a, pp. 37–38) suggests a series of steps in the process of decision making to define the role of government in service delivery. First, one needs to decide on the scope and the priority of the intervention. Government intervention can be justified on a number of grounds such as market failure, the priorities of the political and social elites, or on grounds of producing a more equitable outcome for development. This would justify the provision of certain core public goods such as security or basic education or might be used to protect against the dangers of natural monopolies or asymmetric information. Once the scope of government engagement has been decided there are a number of questions that follow. Who should provide the service? Should it be government agencies and, if so, which level of government should provide it? If the state decides that it should provide the service, this leaves open the question of whether the state should provide only the funding or the service as well. The alternative is to gauge whether the service can be entrusted to a profit or not-for-profit organization or whether the funding

public goods regimes and social welfare / 7 Table 1.1

Public and Private Provision of Services Provision Public (Government)

Private

Financing Public (Taxation)

Private (Voluntary)

Government provides both the service and funding. Free to consumer at point of use. [A] Private payment for governmentmandated or publicly provided service. [C]

Private provider of publicly funded service. [B] Service purchase directly from private supplier or indirectly through private insurance. [D]

Source: Based on Bloom et al.: 2001, p. 184; Ramesh: 2002, p. 148.

can be entrusted directly to the consumer, for example, in the form of vouchers. If the service is contracted out, then the government will need to strengthen its role as a regulator. Thus, the government has three main mechanisms at its disposal to deal with different kinds of market failures: the funding and provision of the service by government agencies; the regulation of governmental and nongovernmental organizations for service provision; and financing through taxation for direct and indirect service provision (see table 1.1). This provides a useful matrix for thinking about the provision and financing of public services. Those services that fall under Category A in table 1.1 can include most income assistance programs and access to primary education in rural China. If no payment is to be made by the individual, the services can be extended to include other public services, including medical care. Services that fall under Category B are those that have generated the most debate as they involve the contracting out of government services to nongovernment entities, or to government entities that have won a contract through some process of competitive bidding. As Ramesh points out (2002, p. 149), this is the approach most often favored by economists for providing public services that are deemed desirable but that the market might not voluntarily provide in adequate quantities. However, this process of contracting out publicly funded services works best for activities that implement public policy rather than for those that involve the making of public policy and where the service is not a “core” function (policing or the courts) but a service function (trash collection, general maintenance, catering). If it is not a part of policymaking and it is “noncore,” then it can be considered ripe for privatization (KSG: 1995). Such services are not natural monopolies

8 / providing public goods in transitional china

but are open to competition, and performance and management can be easily measured (Kettl: 2005, p. 50). This has become an increasingly common approach adopted by local governments in China. This category can also include the use of education vouchers or scholarships provided by the government, but the individual is at liberty to choose where such vouchers may be used. Services under Category C also cover a wide range, from mandated insurance schemes to attending state-run universities or visiting national parks. Obviously, for the latter two examples there is no compulsion to avail oneself of the service; although it may be subsidized, in some cases heavily so, it still requires a private contribution. For example, China has introduced a pension system of workplace-based contributions from both the employer and employee at a fixed rate and from a voluntary third pillar. This category also includes point-of-service user charges, such as co-payments for medical visits. User fees and other forms of a private contribution are designed to prevent problems of free-riding or to prevent people from overconsuming scarce goods, while, at the same time, allowing the state to recover some of its investment. Category D is a market-based solution for the provision of goods that are excludable and divisible. This category is not the primary focus of this study, with the exception of cases where the privatization of welfare services has become common, for example, healthcare, where this is often the de facto case. Since the mid-1990s, China has undertaken significant reforms of housing, trying to turn into a private good what for many urban Chinese used to be a public good. The attempts to commercialize housing are important, as this is clearly a case where a private good has been provided by the state or its surrogates at noneconomic prices. Savings from further reducing this kind of government support and from propping up the state-owned sector can then be used to fund those areas that have clearer public goods attributes. However, the program of commercialization has also encountered the challenge of providing low-cost housing to those who cannot afford the new market prices. Last but not least, it is important to point out that some approaches to the provision of public services cut across a net public-private divide. For example, education vouchers or food stamps can be used to purchase services either from government-run agencies or from private institutions. An East Asian Model of Public Service Delivery? As China has moved away from its socialist planned economy and the provision of virtually all goods and services through the state or its

public goods regimes and social welfare / 9

surrogates (workplaces and communes), a plurality of organizational forms has emerged for providing welfare services, often by default rather than by design. Under the slogan of the “socialization” (shehuihua) of services, it has moved further than many OECD countries along the road to privatization of welfare services. Yet, what kind of a public goods regime is emerging in China? Can one distill any distinct trends in this process of transition? In some ways, China has come to resemble its neighbors in East Asia, and some might argue that it is becoming a part of the family of “oriental welfare states.” There is some truth to this, if we leave aside for the moment the issue of whether there is an oriental model of welfare provision. But China has also been prey to some of the same pressures of changing demographics, a limited financial base, and the globalization of production that have forced changes on most OECD countries. Some writers have seen an East Asian approach to public service provision emerging as a support for the model of state-led development (Wade: 1990; Amsden: 1989). In this approach, the provision of welfare and other public services is shaped in such a way as to promote the overriding policy objective of rapid industrialization and economic growth (Sargeson: 2002; Goodman et al.: 1998). Welfare is thus subjugated to economic development (Deyo: 1992). In fact, some have gone further to argue that East Asia’s rapid economic growth can be attributed in part to the absence of the kinds of welfare support that breed dependency and divert public funds from more “productive” investment (Dai: 1998; Mahatir and Ishihara: 1995). The first to outline a distinctive approach to welfare that might be labeled East Asian was Ezra Vogel (1980) in his work on Japan. Most subsequent analyses have provided embellishment of Vogel’s basic outline. He notes that the Japanese system provided “security without entitlement” and delivered excellent social outcomes in terms of life expectancy, general health indicators, and education. It was important that there was a lack of “welfare dependency” and low government spending by relying on the family, the company, and the voluntary sector to provide necessary support. This meant that taxation could be kept low, the welfare bureaucracy kept to a minimum, and a professional class of welfare specialists was restricted and curtailed from promoting policies to expand the welfare apparatus. This also kept citizen expectations low regarding what they could expect from their government in terms of direct provision of public services. Especially in urban China, a major challenge for policymakers has been that most citizens have expected the government to provide a very extensive array of goods and services, including holidays and cinema tickets. Breaking this psyche of dependence has required a significant effort.

10 / providing public goods in transitional china

Subsequently, a number of analysts have sought to distill the essence of what might constitute a distinct East Asian model of service provision that would complement the economic model of the developmental state, or some variant thereof (see, for example, Jones: 1990, 1993; Goodman et al.: 1998; Tang: 2000; Aspalter: 2006; Ku and Fine: 2007). Pulling these analyses together we can highlight a set of four key, interrelated features. First and foremost, welfare policies are subordinated to the demands of economic development. The state was not expected to get itself embroiled in the kind of expensive and expansive welfare programs that were seen as characterizing the welfare states of Western Europe. Goodman et al. (1998, p. 15) refer to “developmental welfare systems” that seek to limit social expenditures and to design social insurance systems that can also provide financial investment in industry and infrastructure. Holliday (2000) refers to this system of subordinating welfare to economics as “productivist welfare capitalism” and has made the strongest claim that it represents a distinct type of welfare regime. He posits that state policy is focused on growth and that all other policies are subordinate to meeting the economic objectives. There is a general acceptance of the efficacy of such a “trickle-down.” Rapid economic growth and maximizing employment opportunities are seen as the main ways to produce a better and more prosperous life. One can argue that China enjoyed the latter during the pre-reform period and the former during the reform period, but it has not yet been successful in putting the two together. Second, not surprisingly, levels of expenditure are relatively low in comparison with those in the West. This leads to a perception that social security schemes and welfare are more properly the concerns of the family and the individual. However, this does not preclude investment in certain areas that are highlighted as important for high-quality economic growth (Haggard and Kaufmann: 2004). Thus, on the whole, the financial commitment to education is significant and there has been experimentation with polices to provide effective healthcare, especially preventive care. Third, with the state playing a crucial but minimalist role, there is a strong emphasis on the family and other social networks to be the welfare providers. This is reinforced by the low expectations that citizens have from the state in the delivery of welfare services, and there is a tendency toward nonstatutory welfare (Jones: 1993, p. 213). Access to a wide range of public and welfare services is not seen as a key component of citizenship and there is a weak sense of the right to general welfare and not much of a sense that vulnerable groups and individuals should look to the state for support (Holliday and Wilding: 2003, p. 167). Croissant (2004, p. 512) notes that the burden of social security was left outside

public goods regimes and social welfare / 11

of the state realm and has remained within the purview of the “market, the family, and charitable organizations.” This results, according to Tang (2000, p. 135), in a mistrust of social welfare that pervades the economic tigers of the East Asian region. Fourth, the systems represent the preferences of the socioeconomic elites and as a result are “business-friendly.” They are elitist by design and until recently have resisted populist pressures for the expansion of goods and services. In particular, elites have resisted the idea that state-supported welfare programs should be used to promote the goals of social justice and significant redistribution (Goodman and Peng: 1996). This has meant that services may be concentrated in groups considered important to the stateled development approach and has reinforced the accompanying authoritarian polities. However, this does not mean that they have been negligent of distributive objectives. Rather, these have been achieved through means other than state-directed welfare redistribution programs. In fact, one of the features of East Asian development has been the relatively equal pattern of income distribution, a feature that for various reasons China does not share. The lack of strong labor unions and political parties with a social democratic philosophy has furthered this elitist tendency (Lin: 1991). In fact, Gould (1993) sees Japan’s company-based welfare model as a deliberate attempt to prevent the emergence of strong unions and allied political forces that might push for more expansive welfare programs. Given the variations across the region, some have questioned whether one can really talk of one single model of an East Asian welfare regime. Even Holliday (2000) discusses three subsets within the overall category of “productivist welfare capitalism”: facilitative (Hong Kong); development universalist (Japan); and developmental particularist (Singapore). Croissant (2004, pp. 505–507) also identifies three trends of analysis: those that emphasize cultural factors; those that focus on the public management of social risk; and those that follow more closely Esping-Andersen by concentrating on the structures that produce national welfare regimes. That said, the general characteristics do seem to hold across the region and, at a superficial level, China shares these similarities. However, one can question the extent to which these features are really unique to East Asia or ask whether they are simply a product of a particular phase of development. Certainly, they stand as a counterpoint to fully developed, post–Second World War European welfare systems. But reliance on the family and community and low expectations of government provision were features of European polity in earlier periods. For the United States, Moss (2002) traces how the social priorities shifted over two centuries from growth to security. With the process of urbanization, people left behind the traditional security of agricultural life with easy access to farm produce. This

12 / providing public goods in transitional china

caused a breakdown in the support role played by extended family networks. Slowly, but surely, welfare regimes emerged to deal with these new challenges. Many of the same challenges also confront East Asia in general and China in particular. It also appears that as authoritarian polities have given way to more democratic politics, welfare provision has expanded. This has been especially noticeable in South Korea and Taiwan (Aspalter: 2001, 2002; Croissant: 2003). In addition to pressures that may flow from increased political participation, three other challenges also confront these systems, all of which are relevant to China. First, the systems have been premised on high levels of, if not full, employment. The pressures that have led to enterprise restructuring in countries such as Japan and South Korea are beginning to undermine this approach, especially where “corporatist welfare” (Esping-Andersen: 1990) has been typical. Second, all these societies are now aging rapidly, with dramatic consequences for demographic structures and dependency ratios. This will intensify the financial pressures on both the government and the enterprise and may challenge the viability of the system. Third, the process of economic globalization, with the changes in the global production process, is forcing cutbacks in the welfare benefits that enterprises can provide. The gaps in welfare provision that these trends will create can be filled only by the state. As a result, Croissant (2004, p. 520) says, it is not surprising that discussion of an “Asian model” of welfare, especially in the post-Asian financial crisis period, as an alternative to the welfare provision in developed economies has become less attractive. From Welfare State to Enabling State? The provision of public services in China shares the same challenges that confront not only the countries in East Asia but also those within the OECD. The general shifts in provision are insightfully addressed by Gilbert (2002). He argues that across the OECD countries we are witnessing a shift from an ideal-type Scandinavian model of social welfare to a market-oriented Anglo-American version. He refers to the emerging system as an “enabling state,” with policies framed by a selective approach to the private delivery of provisions designed to promote labor force participation and individual responsibility. This replaces the traditional approach of publicly delivered benefits designed to protect labor against the vicissitudes of the market. Gilbert (2002, p. 16) calls this “public support for private responsibility.” It also entails a shift from a system based on the notion of social rights to a system based on social obligations. To quote

public goods regimes and social welfare / 13

from Gilbert (2002, p. 44), “the enabling state is a more market-oriented approach that targets benefits that promote labor force participation and individual responsibility—the purchase of service contracts, social welfare transfers in the form of vouchers and cash instead of in-kind provisions.” This replaces delivery by government agencies, direct expenditures on welfare, and transfers in the form of services. Gilbert (2002, pp. 48–49) outlines four major shifts that are taking place, all of which are relevant in varying degrees to the changes underway in China. The first, as implied, is a more market-oriented approach, with a shift from public provision to privatization. This includes the process of the contracting out of services from public departments to private organizations and the promotion of mechanisms, such as the dispersion of cash vouchers for education. The second is a policy framework that is designed to encourage labor force participation and to get people back into work as expeditiously as possible rather than by solely protecting labor against the vicissitudes of market forces. Thus, rather than providing unconditional benefits, there is a selective use of incentives and judicious application of sanctions. Third, this implies a shift from a system of universal entitlements to a system of selective targeting in welfare provision. This entails a shift from an emphasis on avoiding stigma to an emphasis that seeks to restore or promote social equity. Fourth, Gilbert maintains that this emerging system is based on solidarity of membership, focusing on a cohesion of shared values and civic duties rather than on a solidarity of membership based on citizenship, focusing on shared rights. Of course, the welfare state and the enabling state represent ideal types or—perhaps, more correctly—two ends of a spectrum, with most countries ranging somewhere between the two in terms of service provision. Whereas the United States would be the closest approximation to the enabling state with its emphasis on the obligation to work, the containment of costs, and service privatization, in recent years most nations have been shifting toward this end of the spectrum. China has followed this trend with the adoption of the values and the methods of the market to allocate public services and with the concomitant rise in the contracting out of services and de facto privatization. This has been accompanied by the devolution of the provision of public services to the local levels of government and often to the communities themselves. As noted earlier, privatization of welfare services has been expanding in China, although the process has often been by default rather than by design. Ascoli and Ranci (2002, pp. 6–10) make an important distinction between two different types of privatization. The first is demand-driven participation, with the express intent of curtailing the further expansion of state activity while promoting measures that increase private demand

14 / providing public goods in transitional china

for private services. This includes the use of vouchers and tax breaks or allowances for the purchase of services from private vendors. It overcomes the traditional distinction between funding (government responsibility) and provision (that may include private providers). The second approach is supply-driven, with the state maintaining most of its current responsibilities but trying to modify the mix of regulatory tools. Privatization only relates to supply and does not assume growth in the private demand for public services. Under this approach, most service delivery functions are privatized and competition is introduced for procedures to govern the transfer of government funds to private agencies. This has been common in China with the promotion of the pluralization of delivery of public services and the expansion of the contracting out of services to both for-profit and not-for-profit providers. Each of the two models contains advantages and disadvantages. Demand-drive participation is favored by those who would like to empower the choice of individuals over the paternalistic choices of the “nanny-state.” Its core is the market and the market allocation of resources. The role of government is to enhance the purchasing power of individuals and to safeguard individual choice. However, this approach might not help with equity since such an approach is liable to favor those who are already better placed in society; unless remedial measures are taken, it may lead to the marginalized and vulnerable groups being excluded from core services. The supply-driven approach is based on “the recognition of the existence of private resources to be used fully in the public interest with a transparent system based on the rules of competition” (Ascoli and Ranci: 2002, p. 9). Here the government retains considerable control through the regulatory process that governs the funding of service providers. Whereas the demand-driven approach rewards the capacity of providers to draw in new clients and innovate to meet new demands, the supply-driven approach rewards those providers that help the government meet its objectives in terms of providing services. The objective of the latter is to create a “public market” for social services and this privatizes supply. The Relevance to China We see a number of the trends outlined above at work in welfare provision since the reforms began in China. There has been wide-ranging experimentation, a pluralization of channels for service delivery, and an “individualization” of the social rights of citizenship (Wong and Flynn: 2001; Ferge: 1997). The resultant inequalities of public service provision are something that the CCP leadership under General Secretary Hu Jintao and Premier Wen Jiabao is committed to dealing with.

public goods regimes and social welfare / 15 Table 1.2 Public and Private Expenditure on Healthcare of Selected Countries, 2004 Country

China India Indonesia Malaysia Philippines South Korea Thailand Vietnam

Total Percentage of Public and Private Expenditure Public

Private

38 17.3 34.2 58.8 39.8 52.6 64.7 27.1

62 82.7 65.8 41.2 60.2 47.4 35.3 72.9

Per Capita Expenditure on Health at International Dollar Rate

276.7 91.4 118.2 402.3 202.8 1134.6 293.3 184.1

Per Capita Government Expenditure on Health at International Dollar Rate

105.1 15.8 40.5 236.6 80.7 596.5 189.7 49.8

Source: WHO, at http://www.who.int/whosis/database/core/core-select-process.cfm?countries=all&ind icators=nha.

First, in both the pre-reform and reform eras, welfare policy has been subordinate to the demands of economic development, with policy structured to encourage greater participation in the workforce and to reduce the burdens of the vulnerable populations on the state. The Chinese government has been unwilling to spend significant resources on what it sees as unproductive investments. This was reinforced by the adoption of a quasi-Stalinist approach to industrial development that did not see key public goods as comprising a part of the productive forces (Lu Mai: 1999). Thus, like its neighbors in East Asia, China spends relatively little on the provision of public goods and services, and out-of-pocket expenses for healthcare and education have been rising rapidly (table 1.2). Private expenditures are not as high as those in India, but they are equivalent to those in Indonesia and are appreciably higher than those in the more developed East and Southeast Asian countries. Per capita government expenditure on health at the international dollar rate compares favorably with lower-income countries in the region but lags well behind the more developed economies. In other East Asian regimes where spending on welfare has been low, there has been extensive investment in social development, especially to ensure the health and education of the population (Tang: 2000). According to Gough (2000, 2002), this emphasis was an integral part of the policies to support nation building, the creation of legitimacy, and productive investment. In China, government spending on health is

16 / providing public goods in transitional china Table 1.3 Country

China Indonesia Malaysia Philippines Singapore South Korea Thailand

Social Indicators for Selected Countries Under-5 Infant Mortality Per 1000, 2004 31 38 12 34 3 6 21

Infant Mortality Per 1000, 2004 26 30 10 26 3 5 18

Government Education Spending, as % of GDP, 2000–2004 2.2 0.9 8.0 3.2 3.7 (2000) 4.6 4.2

Government Health Health Expenditure, Spending, as Per Capita % of GDP, PPP US $ 2003–2004 2.0 1.1 2.2 1.4 1.6 2.8 2.0

278 113 374 174 1156 1074 260

Source: UNDP (2006), at http://hdr.undp.org/hdr2006/statistics/indicators/.

roughly comparable to that in other Asian middle-income countries but spending on education lags and is far behind the Asian tiger economies (table 1.3). Government spending on health as a percentage of GDP was 2 percent in 2003–2004, the same level as in 1990. This is better than most other Asian countries, with the exception of the tiger economies and Malaysia (2.2 percent). Educational outlays were 2.2 percent in the 2000–2004 period, well below that in all other countries in the region, with the exception of Indonesia. Spending has dropped from 2.3 percent in 1990 and falls well below the government’s stated intent. Throughout the 1990s, China had set a spending priority for education equivalent to 4 percent of GDP by 2000 and this again is one of its key development goals for 2010. As a result, the average time spent in school is significantly lower in China and drop-out rates before completing the required nine-year basic schooling are higher than in other comparable Asian countries. However, statistics for 2006 show a more significant rise in education spending of 3.01 percent of GDP mainly because of the increased spending on the nine-year compulsory education program. Yet even with this, over one-third of provincial governments failed to spend adequately on the program (SCMP, January 1, 2008). Despite the relatively comparable spending on health, China’s indicators for under-5 infant mortality are worse. The rate for China in 2004 was 31 per 1000 live births, while for Malaysia it was 12, Thailand 21, and South Korea 6. China was closer to the Philippines (34), where indicators were declining, and Indonesia (38). Second, this underinvestment by the government has been accompanied by an emphasis on self-reliance and resolving problems in situ

public goods regimes and social welfare / 17

rather than looking to the higher levels of government for relief. During the pre-reform period, self-reliance was focused on the collective and the family, but now it is more likely to be seen as the responsibility of the family alone. Only very recently have there been limited attempts to spread a rights-based consciousness that sees access to a wider range of public goods and services as a key component of citizenship. What has dominated has been the promotion of a caricature of quasi- Confucianism that stresses self-help, the support of the family, and the avoidance of seeking state help, except as a very last resort. In fact, Jones (1993, p. 198) renames her categorization of East Asia welfare regimes from “oikonomic” to “Confucian welfare states.”1 Further, she proposes (1993, p. 214) that such regimes might be thought of as “household economy” welfare states run in the style of a traditional Confucian extended family. Nelson Chow (1987) was one of the first to highlight the importance of the family and kinship networks for social welfare support. Third, this has meant that the market and alternative suppliers have begun to play an increasing role in service provision. The form of de facto privatization that has been dominant in China has been supply-driven, but there has also been experimentation with demand-driven approaches. However, China’s leaders place greater trust in the market (even though this is viewed with suspicion by some) than in genuine nongovernmental providers. Despite the natural Leninist resistance to an autonomous sphere of civil society actors, a controlled sector providing vital public service functions has emerged, with the church or temple becoming the key points for reciprocity for many. Fourth, social welfare provision has reflected the bias of the official ideology, with preferential treatment for those living in urban areas and employed in state-owned enterprises as well as for government officials. These key groups were seen as crucial to the industrialization drive and as important constituents for the CCP. The extensive array of benefits through the workplace compensated in part for the low wages paid in the pre-reform era. Such a reliance on the provision of public services through the workplace has provided an important policy challenge for the leadership, especially from the mid-1990s, as it has had to roll back benefits without undermining regime support. The kind of company welfare that existed in China and still exists to some extent has parallels to the company-based programs in Japan and South Korea. Esping-Andersen (1990) dubs this “corporatist welfare.” It is even more pervasive in China. In fact, the heavy reliance on the workplace for service provision is a major distinction from other Soviet-style systems. It was at this level that most welfare was managed, with local government agencies only playing a residual role. One of the major challenges of reform has been to weave these fragmented

18 / providing public goods in transitional china

systems into a more coherent whole by transferring welfare obligations to local state administrative agencies. This discussion suggests two major differences with other systems in East Asia. First, it is difficult to speak of simply one system for public service delivery in China, and second, the system is marked by even greater inequalities of service provision. Rural and urban China have operated under quite distinct systems, with the urban being consistently privileged over the rural. Even within urban China, the system is fragmented as those in government employment or working in state-owned enterprises enjoy types of support that are not available to those working in other sectors of the economy, especially in the informal sector that is home to increasing numbers of migrant laborers. In Esping-Andersen’s term (1990), it is a truly “Bismarkian” system, with relatively privileged sectors, underprivileged sectors, and totally excluded sectors. This system by its very design has institutionalized inequalities in a way that is absent from other systems in East Asia. The main division is between rural and urban, but there are also significant regional variations. One major legacy of Maoism is the concept of self-reliance, the idea that each locale should make do with its own resources and request only limited support from other levels of government. Although this has often been written of as a strategy for production, the worst excesses being the backyard steel furnaces in the Great Leap Forward (1958–1960), it was also a strategy for consumption. Localities could consume only those goods and services that they were able to produce locally. Thus, whereas intra-locale inequalities might have been limited, inequalities of service provision across locales could be quite significant. This has had a lasting impact on the reform program to date. As the main providers of public services, subnational levels of government in China have expenditure responsibilities that are quite out of line with international practice (World Bank: 2002, p. ii). The Maoist notion of “self-reliance” reinforced the idea that each locality should minimize “dependence” on support from higher levels. Through the reform period, the local levels of government have retained major responsibility for financing infrastructure and providing social welfare. The Structure of the Book This book seeks to look at how the provision of social welfare has changed under reform in China. Chapter Two looks at the pre-reform system for the provision of public goods and services and the institutional features that have impacted on the reform era. Given its low level of economic development, the Chinese system performed well in terms of improving basic life

public goods regimes and social welfare / 19

indicators. It also had a reasonable infrastructure that was facilitated by the collective system in the rural areas. After reviewing the progress of the system, we look at the structures that evolved as a result of party ideology and policy choice. Important were the bifurcation of service supply for rural and urban China and the state bias in China’s development model. When combined with the above-noted preference for self-reliance and local delivery of services, this resulted in significant inequality of provision that has framed the possibility for subsequent reforms. The heavy burden on state enterprises and collective rural structures meant that the system could not be maintained once experimentation with market-based reforms began. The financial weakness of these institutions meant that a “service gap” appeared before new structures could be developed by local administrations. In particular, we shall look at the workplace-based structures for providing services. Chapter Three provides an overview of how economic reforms impacted on the provision of public goods and services. First we look at how the pre-reform structures shaped policy choice and the degree to which there was path dependence (North: 1990). During the reform period, China has undergone multiple transitions that have affected welfare policy. The shift from an agrarian society to an urban one has been accompanied by a transition from a centrally planned economy to one that is more directly influenced by market forces. As a result, China’s policymakers can no longer rely on traditional rural households and land to provide security, and they have to think about more effective integrated urban networks of services. The transitions have also meant that the old model of workplace-based welfare has unraveled. In addition, China is experiencing an important demographic transition at a low level of development from a young to a quickly aging society. Finally, there is the shift in emphasis from the collective to the individual as the focus for choice and action. This is having a major impact on the way individuals, especially those in urban China, view the state and markets and how these changing perceptions affect their choices about service provision. We can identify three phases in policy toward the provision of social welfare. Initially, little was done to provide a new infrastructure for public service delivery but starting around 1993– 1994, significant local experimentation was encouraged, leading to new frameworks being put into place from the late 1990s. The leadership under Hu and Wen has sought to regularize and expand these frameworks, and local governments have an increasing burden for welfare provision. We examine how the de facto decentralization and incentive system affects provision. Chapters Four through Eight provide case studies of different kinds of welfare services and how their delivery and provision have changed

20 / providing public goods in transitional china

under reforms. Chapters Four and Five look at different aspects of public health, with a particular emphasis on the rural areas. Citizens might expect the government to provide access to basic healthcare and protection from unexpected shocks due to pandemics. However, access and protection are determined by where you live, who you work for, and your level of income. If you live in a rural area, are engaged solely in agriculture, or are employed in the informal sector, and have a low income, you are pretty much on your own or dependent on your family and kinship structures for help. Healthcare provides a good example of this. The drop in provision for those in poor, rural areas and for new categories of the urban poor has been used by critics of the CCP’s adoption of market-based solutions to argue the case of “market failure” and to push for greater state intervention. Given its ideological predisposition, this has been difficult for the leadership to ignore and there has been experimentation that the government hopes will compensate for such failure. With reform, the incentives for medical providers have encouraged them to provide expensive, curative procedures at the expense of the cheaper, preventive medicine with which China had been very successful during the pre-reform period. Progress in a number of social indicators has slowed and even reversed for the vulnerable populations—China ranks as one of the worst countries in the world in terms of fairness of financial contributions to healthcare. The weaknesses and vulnerabilities of the health system were revealed by the outbreak of SARS in 2002–2003 and the current concerns over avian flu have continued to raise questions about China’s capacity to deal with public health crises. This issue and how policy has evolved to deal with HIV/AIDS form the content of Chapter Five. Chinese policy on HIV/AIDS has witnessed a remarkable evolution in recent years, with it now adopting most international best practices. However, significant problems remain in terms of implementing and maintaining these policies that significantly disadvantage the more marginal and vulnerable populations. Chapters Six and Seven look at the development of policy for social protection through an investigation of social insurance and social assistance (for the distinction, see World Bank: 2006, pp. 149–50). Social insurance systems fundamentally affect citizens’ opportunities by providing them with a safety net and are designed primarily to mitigate the negative effects of unemployment and insufficient income in old age (Moss: 2002, p. 214). Thus, the groups that the state chooses to provide with such protection reveal much about its political and development priorities. Risk is mitigated for key groups in the state and government sectors, whereas migrant workers and others in the informal sector, as well as those

public goods regimes and social welfare / 21

in rural China, are largely excluded. This is an area that current policy is trying to redress. In urban China, policy has followed that of East Asia, with attempts to build contribution-based, fund-managed social insurance systems for unemployment and pension systems, rather than “payas-you-go” systems financed out of taxation (White: 1998, p. 186). These systems have been important to the CCP since they are predominantly urban-based and protect politically important groups. The policy has been regressive in terms of equity. Chapter Seven deals with social assistance that provides help for those who have no other means of support and may include redistribution to the poor from those who are better off. The Chinese government has introduced a number of important redistributive programs but, on the whole, they have not been very effective and have not provided substantial redistribution either in terms of region or households. The most important policy initiative has been the introduction of minimum living standard support that was introduced first for the urban areas and subsequently extended to rural China. Chapter Eight concludes using unique survey data to analyze how citizens view government performance in terms of providing public goods and services. This provides an important perspective on the demand for public services to supplement the government’s supply-driven approach. This is followed by looking at what the changing nature of service delivery tells us about the relationship between the state, market, and civil society in contemporary China. Although the state has asserted its role as a provider of minimum guarantees, it cannot fulfill the demands of all its citizens. Either by design or default, the market and institutions of civil society will have to take on a greater responsibility for service delivery. China is undergoing a dual transition caused by “normal” economic development and the transition from a command economy to one where market forces play a greater role. We shall draw lessons from this dual transition for the provision of social welfare.

This page intentionally left blank

Ch a p t e r Two We l fa r e P rov ision, 1949 –1979

After taking power in 1949, the CCP gradually began to build a basic system for providing public goods. The domestic stability following decades of conflict and revolution allowed economic reconstruction, and the growth that followed enabled China to perform creditably in comparison with other large nations in Asia. In the field of public health, for example, China was able to shake off its tag as the “Sick Man of Asia” and eradicated a number of contagious diseases while bringing others under control. In fact, by the mid-1970s China’s rural public health system was hailed by the WHO as a model for others to follow. Policies put in place basic guarantees for most citizens and for the urban elite they provided a network of welfare benefits that were well above China’s level of economic development. The series of five-year plans that were a feature of China’s form of Soviet development provided for major programs to control the supply of water and to build the basic road and rail infrastructure. Welfare policy and indeed policy in general reflected the interests of those favored by the ideology: government workers, the industrial working class, and soldiers. Others were left to deal with access as best as they could using communal resources. The peasants in whose name the revolution had been fought had to rely on any surplus provided by the collective to cover their welfare needs. This provided a basic guarantee for most and consequently we see a major improvement in preventive medical care, for example, but also the system led to highly unequal provisions and vulnerability to swings in political moods. The worst example of the latter was the launching of the Great Leap Forward (1958–1960) that resulted in a major famine. How does one assess the significant advances in public health during the 1950s when some 25 to 30 million excess deaths were caused by bad policy choices at the end of the decade? In this chapter, we look first at the guiding framework for the provision of public goods before the reform era started in the late 1970s and

24 / providing public goods in transitional china

then review the key institutional structures that provided them: the urban workplace (danwei), the rural collective institutions, and the residual support for those who fell outside of the regular system. Then we look at the achievements of this system in comparative perspective with reference to other nations in East Asia as well as to other large nations in Asia, such as India and Indonesia. Guiding Framework for the Provision of Welfare Services China shares two basic points of departure with all other countries in terms of how its system for welfare provision is structured. First, the ideology and value system provide the basis for decisions about who gets what level of welfare support and for how long, and what kinds of inequalities are acceptable. Second, the structure of the economy and the level of economic development affect the kind of welfare choices that can be made. Public goods provision, including welfare, is a crucial part of the institutional framework of the economy and attempts are made to coordinate the two to meet the regime’s objectives. China has often been compared to other Confucian states in East Asia as relying in the first instance on the family for provision of welfare (Chow: 1987; Jones: 1990, 1993). In traditional rural China, this is not surprising as there was no alternative, with the exception of the temple or the clan, in those rural communities that might have been able to produce a surplus for redistribution. Dependency on the family, by and large, was stressed by the CCP so that the state could preserve its funding for investment in developing the productive forces. Despite its revolutionary pedigree, occasional attacks on the family, and the development of new collective and state institutions, the new regime maintained the tradition that people sought refuge first within their family and only subsequently from the new institutions. During the height of the Great Leap Forward, there were attempts to replace the family with collective living arrangements but these were quickly abandoned as utopian experiments. Mao Zedong, the chairman of the CCP, was insistent that the development of welfare would not be at the expense of production. Throughout its history, the CCP has denounced “welfarism” and “relief ” within the labor movement as actually retarding development. As Dixon (1981, p. 6) has noted, the system that developed was not overly concerned with the politics of redistribution and was based on increasing the total amount of production that would provide incremental relief. Mao Zedong, mindful of the influence of social democratic

welfare provision, 1949–1979 / 25

politics, was wary of “relief ” that claimed to benefit the people but that actually, in his view, only harmed their long-term interests by retarding economic development. As the CCP victory neared in 1949, the leadership made it clear that the revolution had been fought to benefit the (still to be created) industrial proletariat. Party work was to shift its focus from the countryside back to the cities. A city work conference in the summer of 1948 laid the basis for this shift, with calls to concentrate work on developing the state sector of the economy while acknowledging that the Communists were “extremely poor” at managing economic enterprises (see Saich: 1996, pp. 1204, 1346–64). However, the working class in the major cities would be crucial to the consolidation of CCP power, yet the party had had no effective contact with the working class for 22 years. Thus, considering it as a part of establishing credibility on occupying the cities, the party did view establishing basic social safety nets as important (Bray: 2005, p. 103). Initially, the trade unions under Li Lisan’s leadership played a major role in establishing and running the new system of support. Li had been an early labor activist and had run the party briefly in 1930–31, promoting an aggressive policy of uprisings for which he was quickly purged. In the newly liberated areas, the unions moved to set up a labor insurance scheme (Lieberthal: 1980, pp. 72–73), with each enterprise union collecting contributions from both employers and employees. It was the union that managed these funds and distributed them to the workers as per assessed need. According to Bray (2005, p. 104), this workplace-based system was a “temporary measure” since 1948, but it was formalized in February 1951 with the passing of the “Labor Insurance Regulations.” The regulations outlined the kinds of services to be funded by a contribution of 3 percent of the total monthly wage bill from the enterprise. This was paid into a Labor Insurance Fund. The municipal trade union organization was given 30 percent of the fund for management and distributional issues. The remainder was placed with the enterprise trade union for use within the workplace. As Bray (2005, pp. 108–10) notes, the creation of an autonomous realm for trade unions to protect and promote workers’ welfare that could possibly conflict with the party and enterprise management was a cause of concern for other leaders. As a result, at the December 1951 National All-China Federation of Trade Unions (ACFTU) Conference, Li Lisan was criticized for neglecting his role as the leader of the party and for promoting the short-term economic interests of the workers at the expense of the long-term development goals of the party. He was removed from his position and, most important for our discussion, decisions on the allocation of public goods

26 / providing public goods in transitional china

and services were made by the enterprise leadership. Trade unions were effectively neutered as promoters and defenders of workers’ interests in the workplace. The obsession with boosting production figures has never been seriously challenged by a need to divert resources away from immediate investment in production. Those who have proposed countervailing policies have been denounced for promoting “welfarism” and—as in Li Lisan’s case and that of his successor, Lai Ruoyu—have usually lost their posts as a result. Mao’s rejection of “welfarism” has been adhered to by his successors during the reform period. In February 1980, the Zhejiang Daily stated that we must on the one hand oppose the reactionary fallacies of building socialism in poverty, and so forth . . . and on the other oppose the idea of realizing a so-called welfare state in China now. Our principle has always been: “one inch of increased production, one-tenth of an inch of additional welfare” . . . . We must maintain the bonus system but must not hand out excessive bonuses. We must adhere to the principle “to each according to their work” and “more pay for equal work,” but must also take care of the entire country and our neighbors.

In the same year, Deng Xiaoping reiterated this view. He noted that although China should increase education spending since it was among the 20 countries in the world that spent the least on education relative to GDP, the country should not become a welfare state because this was impractical given its current stage of poverty. He noted that “developing production without improving the people’s livelihood is not right” but calling for an improved livelihood without “developing production is not right and cannot be attained.” However, before market-based reforms were introduced, ideology did act as a constraint on too great an increase in inegalitarian distributional policies. Before 1978, a policy shift to a more inegalitarian wage policy and to an openly inegalitarian development strategy was more difficult to pursue. The radical group that coalesced around Mao’s wife, Jiang Qing, during the Cultural Revolution and that was later denounced as the “Gang of Four” and arrested and sentenced to lengthy prison terms clearly resisted any such moves. Particularly important were the writings of two of the members (Yao Wenyuan: 1975; Zhang Chunqiao: 1975) who used the notion of “bourgeois rights” and the persistence of capitalist factors such as commodities and differential wages to argue that the reproduction of capitalism was possible. To prevent a new bourgeoisie from taking power, they argued that there would have to be frequent “cultural revolutions” to eradicate the remaining capitalist factors; in one of their memorable

welfare provision, 1949–1979 / 27

phrases, they claimed that they would rather have a “late socialist train” than “a capitalist one that ran on time.” This is not to suggest, however, that the state was not instrumental in restructuring the provision of welfare post-1949 and that the policies and institutions did not enshrine differential access. George and Manning (1980), in their study of social welfare in the Soviet Union, suggest a definition for a socialist social policy that can be applied to the provision of public goods more generally. They posit that policy rests on the fundamental premise that the private market has been largely or completely abolished and that the means of production are in the hands of the government. This gives the government—in China, in reality, it is the CCP—enormous discretionary power over decisions related to the provision of public goods in comparison with capitalist systems where individual insurance schemes may predominate and market choices determine the provision of physical infrastructure. In theory, social policy can be used to promote ambitious goals of social engineering, even more so than in the post-Second World War North European welfare states as there should be no fundamental conflict of interests between the various sections of the community and particularistic interest groups. In practice, it was the CCP leadership that made the crucial decisions as the pattern of its path to power meant that it enjoyed significant relative autonomy from all forces and classes in Chinese society. Initially, however, the move to nationalize and collectivize assets was gradual. The CCP’s main aims in 1949 were to revive the war-ravaged economy and to eliminate the remaining domestic opposition. This argued for a more conciliatory policy toward groups such as major industrialists who may have been viewed as class enemies. This need for a conciliatory policy was compounded by the shortage of properly trained, administrative, managerial, and technical personnel who could manage a modern, urban, industrial sector. With priority given to economic recovery, it was necessary to ensure that the available scarce resources were not wasted. This foreclosed the immediate introduction of a full-scale socialist transition strategy. The “moderate” mood was summed up in the slogan of “three years of recovery and ten years of development.” However, while a mixed economy was encouraged, only the CCP-controlled state apparatus was capable of providing any real coordination. This allowed the CCP to transform the mixed economy to its own advantage without a major disruption in production and distribution. The state took control over both ends of the production process, providing the industrial enterprises with their raw materials through the national ministries and placing orders with the private entrepreneurs for processed and manufactured goods. The state was, therefore, able to control what went in and what

28 / providing public goods in transitional china

came out. As soon as privately owned enterprises were tied up in this way, the CCP began to promote the creation of joint state-private enterprises. This movement reached a peak in 1954 and was gradually extended into a program to “buy-out” the private owners who were paid interest on their shares at a rate determined by the state. With the increasing dominance of state-owned enterprises in urban China, new forms of public goods provision were also developed. In the countryside, transformation was even more dramatic. The traditional social and power relationships were broken up by land reform. Land reform had the advantage of forcing an identity of interest between the peasantry and the CCP by redistributing land to the rural households. However, it was never likely that the CCP would tolerate a householdbased farming system for long and by the mid-1950s the process of reorganizing the rural sector into larger collective units began and the role of rural markets was curtailed. By the mid-1950s, CCP policy clearly saw no significant role for the market in allocating goods and services. This not only had a detrimental effect on the quality of both rural and urban life but also led to the eradication of intermediary organizations that operated within the market economy and in the space between the local state and family. In the countryside and the cities, the CCP vigorously sought to suppress lineages, clans, local place associations, guilds, and other organizations that might have presented a moral alternative or different organizing principle to the state. A major consequence of these transitions was that the state had to develop new mechanisms for delivering public goods and services that would not place too great a burden on the state’s weak financial base. New regulations from December 1950 required private enterprises to commit 15 percent of their net after-tax profits to a “workers’ relief and premium fund” to cover basic contingencies.1 The situation was similar in the growing number of joint state-private institutions. What these contingencies would cover and the actual amount that would be set aside were to be negotiated by the trade union in the enterprise. It came to resemble facilities in state-owned enterprises, with health centers, dormitories, recreational facilities, and nurseries. In fact, union activism meant that many enterprises dedicated much higher levels of their profits to welfare support. It would have been very difficult for factory owners to resist such pressures in the political climate of the early to mid-1950s. However, as noted above, this union activism was soon nipped in the bud lest unions would compete with the CCP for workers’ loyalty. As the private and joint state-private sectors were absorbed into the stateowned sector, for most in urban China the provision of key public goods and services became the responsibility of the workplace. They provided a

welfare provision, 1949–1979 / 29

range of benefits and services that were nominally at least supervised by the trade unions, but the Eighth Party Congress (1956) placed control over these services in the hands of the enterprise party committees. Not surprisingly, through the early 1950s there was an increase in workplace welfare expenditures, but even this was not to the liking of all workers concerned. By early 1957 reforms had led to a decline not only in real income and benefits but also in elite urban workers’ inputs into decision making. This resulted in major unrest in cities such as Shanghai (Perry: 1993) that must have caused concern among the central leadership. Major institutional changes for welfare provision had also taken place in the countryside, with the new cooperatives taking on many of the responsibilities that fell outside of the realm of the family. Although it is true that direct relief was limited to those who could not be helped by the family, the process of collectivization did dramatically change access to public goods for many people in the countryside by providing rudimentary public health, education, and irrigation projects. One of the benefits of the development of collective structures in the countryside was that some basic guarantees were provided to those who would have had no recourse other than to rely on charity should the household not be able to support them. The risk pool was gradually broadened in this process, although it remained ineffectively small for major provisions, such as pensions and healthcare coverage for significant illness. Under Mao, the welfare system was seen as subservient to the demands of the economy and to the pursuit of socialism. In practice, this meant that social policy was closely tied to a development strategy that kept the rural and the urban separate and granted privileged provision to the urban and industrial sectors over the rural and agricultural sectors. At the same time, the organization of the collective in the countryside and the inconsequentiality of cost meant that for its developmental level, rural Chinese enjoyed good preventive healthcare and basic education systems. Although they were subsequently derided in the official press, the paramedics of the Cultural Revolution—known as “barefoot doctors”—do seem to have provided decent vaccination programs and preventive care for many of those who would not have received them in other developing countries. To a certain extent, this contributed to redressing the urban bias of the system. The Chinese system for social welfare administration enjoys two distinctive features. First, as in other areas, there has always been a high degree of decentralization. Even during the high tide of the collective period, what one got varied largely in accordance with where one lived. Second, there was a strong emphasis on collective self-reliance and mutual aid. The first port of call was the family and then the collective in the countryside or the neighborhood committee in the urban areas.

30 / providing public goods in transitional china

Welfare needs, and social policy more broadly, were met at the level of the urban workplace, the urban neighborhood committee, and the rural people’s collectives and then the villages. These three networks, depending on party policy and direction from higher administrative levels, were responsible for deciding not only the appropriate level of welfare support but also, more importantly, the financing and delivery of the services. This provided the population with a reasonable measure of security and education in what was a surplus labor, low-income economy. For much of the time, the stringency of the household registration system prevented large-scale rural-to-urban drift; in the Mao years, China’s cities were remarkable for their lack of shantytown development and migrant laborers. Self-reliance has been and remains an important principle for the provision of public goods in China. Administratively, this requires that the basic-level administrative organizations develop capacity in line with their own resources rather than relying on handouts or redistribution payments from higher levels. In practice, it means that individuals might be provided with minimum levels of subsistence, but for the rest they relied first and foremost on the family and then borrowed from village funds in the countryside before receiving actual welfare. The CCP sets down general guidelines, such as a minimum wage or the need for unemployment relief, but it does not apply a standardized rate or system across the country. In terms of welfare provision, “self-reliance” was a strategy for consumption as well as production. Thus, there was considerable variation in the types of services and the quality of the services available as well as in the incomes that people could earn. This led to a large urban-rural divide in terms of the provision of public goods. With industry, especially light industry, heavily concentrated in a few geographic locations, consumption was confined to a few basic goods and the level of consumption remained low. Consumption was tailored to local production. During the Maoist years, policy emphasized austerity and the promotion of tightening one’s belts in the present in return for a Communist nirvana in some unspecifiewd point in the future. Such exhortations were accompanied by policies to restrict high incomes and to narrow income gaps. These were backed up by a program of rationing of key goods such as grain, cotton, and light industrial goods. The fact that most of these ration coupons were for the locality in which one lived and worked made travel difficult. A collectivist ethos was promoted and citizens were led to believe that they would get more out of the collective by putting the collective interests above their own individual wants and desires. With no market to speak of and local distribution of goods and services, it was easier to maintain this ethos before the economic reforms began.

welfare provision, 1949–1979 / 31

The principle of self-reliance was an important part of the CCP system as it developed after 1949 and it has remained an important component of public policy up until the present. Its origins lie with the experiences of the CCP in the period before 1949 when the party operated a number of precarious base areas scattered across northern and central China. Under these conditions, government and military organizations were encouraged to produce as much as possible to meet their needs without external help. As the financial situation was tight, this extended to many units providing basic subsistence goods in lieu of wages. Even here though, there was differentiation between the way peasants were treated and the way those in industrial departments were treated. The first soviet established by the CCP, the Jiangxi Soviet (1931– 1934), stated in its founding constitution that the purpose of the soviet regime was to improve the “living conditions of the working-class.” Specifically, the constitution mentioned establishing an eight-hour day, a minimum wage, and social insurance and state assistance for the unemployed. For peasants, the constitution offered the abolition of feudalism and the redistribution of the land of all landlords to poor and middle peasants, with an ultimate view to nationalization (Saich: 1996, p. 554). These biases were reflected throughout the pre-1949 experience. One CCP base established in East Guizhou in 1934 was very generous to its workers, offering a two-month maternity leave on full pay, a 30-minute breast-feeding break every other hour, four days off per month, and medical insurance covered by employers (Saich: 1996, p. 593). Finally, by way of example, the 1940 administrative program for the JinCha-Ji (Shanxi-Chahar-Hebei) Border Region stated that disaster relief and salvation were to be organized on the basis of the “noble national traditions” of mutual aid and cooperation at the county/village levels. By contrast, industrial workers were to be protected in an eight-hour working-day wage labor system, half of which was to be paid in kind. Benefits included five weeks of paid maternity leave, and the young and women were forbidden from being employed in work that could be considered harmful to their health (Saich: 1996, p. 935). The creation of a dual rural-urban society was solidified by the household registration system (hukou) that ensured state resources were channeled primarily to the cities at the same time as substantial portions of the rural surplus were transferred to urban industry, the military, and other state priority projects. The post-1949 origins of the system derived from the desire to relocate many of the refugees in the urban areas back to the countryside. The program proved successful because it was voluntary but also because the state was able to offer land and/or money for leaving the cities. In addition, the state did not announce that it would effectively close

32 / providing public goods in transitional china

the cities to its rural population (Cheng and Selden: 1997, pp. 28–29). This rural-urban division remained policy until the reforms that began in the late 1970s started to break down the barriers. As the country moved from restoration to reconstruction of the economy, the CCP began to adopt more specific regulations about the need to control migration flows. Urban residents, following the 1953 census, were issued with registration books, and directives to control villageto-city movement were sent out. Finally, in June 1955, regulations were promulgated for a permanent system of household registration that covered both the urban and the rural areas. Importantly, the new regulations made movement from rural to urban areas extremely difficult and even strengthened monitoring of movement within the countryside and from city to city. Whereas the Great Leap Forward saw one last major exodus to the urban areas, in 1960 the household registration system was invoked to return people to the countryside. As Cheng and Selden (1997, p. 45) remark, the system as it evolved from 1960 onward was quite distinct for both China and socialist systems in general. What had been a system to monitor population flows now became a system to control population. Men were permanently registered in their place of birth, and women in the place of their husbands’ birth. It “established and reified a permanent spatial hierarchy of positions that were transmitted across generations.” Permission to migrate up the spatial ladder from rural to urban or from a small city to a major metropolis was rarely granted. This locked the population into vastly different socioeconomic structures in terms of remuneration and the provision of public goods and services. Except during the Cultural Revolution when many urban dwellers were “sent down” (xiaxiang, a significant phrase that reveals much about how the state viewed the countryside) to the rural areas, most Chinese seldom traveled and knew little about the world outside of their own neighborhoods. The system was reinforced not only by the registration controls but also by the elaborate system of ration coupons for grain and other basic goods that were place-specific. Unless one had national grain coupons, which were reserved only for very special Chinese and for foreigners, one could buy food only in one’s own administrative jurisdiction. The lack of an open urban food market meant that it was difficult to migrate spontaneously. Urban Organization of Welfare Provision In urban China, the workplace became the main place not only for employment but also for those lucky enough to find employment there

welfare provision, 1949–1979 / 33

for the provision of a whole range of public goods and services. The workplace system was the defining system for urban provision and remained so in many ways well into the reform period. The local governments dealt with those who were not part of the government or enterprise systems. Although the workplace became a key institution in post-1949 China, it did have antecedents in earlier periods. Certainly, workplace-type structures operated in the base areas that the CCP controlled before 1949 (Lu Feng: 1989, 1993; Lü and Perry: 1997). Although the Jiangxi Soviet (1931–1934) had operated a more traditional system of central allocation, this was not feasible in the fragmented base areas in the early 1940s. The Japanese invasion and the increasing stranglehold that Nationalist forces were exerting on the base areas meant that production was decentralized and work organizations had to think about being as self-sufficient as possible in terms of production and the provision of services. This established a system where there was not only complementarity between workplaces and the central authorities but also possible competition as workplaces would seek to hoard goods or misreport their production in order to be able to reinvest more in their own development. Peng Zhen, one of the leaders of the Jin-Cha-Ji Border Region, suggested in 1940 that they implement a system of paying half of the wages in kind (Peng: 1940 in Saich: 1996, p. 935). Thus, as Lü and Perry have noted (1997, p. 32), the legacy of this base area experience informed the structure of enterprise development in two ways. First, there was a tendency for workplaces to maintain a degree of independent fiscal power and to favor their own collective self-interests under a centralized financial system. Even in the 1990s, walking out of the offices of the Translation Bureau of the Central Committee, members would often comment with envy about the apartment block opposite that belonged to the Central Organization Department. They were envious of the extra resources it had even though theirs was also a central party organization. Second, the workplaces played a significant role in improving the livelihood of its own members, especially in times of economic difficulty. This refines the prevailing view that the workplace was a system developed in the 1950s as a response to the need to restrict rural-urban migration and control the new urban population (Henderson and Cohen: 1984; Whyte and Parish: 1984). In fact, Bray (2005) traces cultural antecedents to traditional Chinese views of space and organization, and Bian (2005) argues not only that the Communist experiences were important but also that the structure of the Chinese workplace originates in the Nationalist era as a result of the crisis prompted by the Japanese occupation of Manchuria and its subsequent invasion of the rest of China.

34 / providing public goods in transitional china

These earlier experiences help us understand why the leadership and many workers responded to these kinds of workplace-based structures in the post-1949 period. In terms of providing public goods, Naughton (1997, pp. 171–76) suggests that the system did not really come together in its full form until the mid-1960s. He highlights three important factors. The first was the curtailment and then gradual elimination of labor markets and job mobility, something distinct from other Soviet-style systems. Individual enterprises lost the power of hiring, with some 95 percent of first jobs being allocated by the state (also see Bian: 1994). Of course, for nearly everybody, the first place of work was also the last place of work. Second, the state planning system for prices ensured that there was a substantial financial surplus in most enterprises, some of which could be held back for reinvestment in public goods. Third, there was a program of administrative simplification that included the shifting of some heavy industry to the hinterland. When the Nationalists relocated industry inland to ensure their safety from the Japanese invaders, they had to provide workers with housing and a basic social welfare infrastructure. In the early to mid-1960s when Mao Zedong feared a Soviet attack, he also moved industry inland to provinces such as Guizhou and Sichuan. Rather than relying on an expanded government apparatus, welfare services were built up within the enterprise. As many writers have noted, these workplaces formed their own “small society.” In post-1949 urban China, the workplace became the key institution for ensuring social control (Walder: 1986; Lü and Perry: 1997), and welfare provision played a key role in this. Housing was allocated through the workplace as were welfare benefits, holidays, education for children, and later permission for when and how many children to have. Lü and Perry (1997) define five basic features of the prototype workplace: it controlled personnel, provided communal facilities, operated independent accounts and budgets, had an urban or industrial role, and was in the public sector. The system eschewed horizontal contact among workers, students, and farmers, thus contributing to a system of vertically defined control and the cellularization of society for many functions (Shue: 1988). Although the cellular structure of Chinese rural society was long apparent (Skinner: 1964/65), the CCP’s pre-1949 experiences led to the adoption of a cellular structure for society as a whole in the post-1949 period. The cellularization of life as reflected in the workplace system was a core component of the CCP’s cell structure when it operated as an underground, revolutionary party. Horizontal contacts were denied because of the fears of betrayal or discovery that might lead to destruction of the organization as a whole.

welfare provision, 1949–1979 / 35

This cellular structure meant that although there could be relative equality within the workplace, it could conceal great interworkplace inequality in the provision of goods and services. The system enshrined a hierarchy of benefits and quality of life. Most important of all, employment in this SOE sector meant that one was not a rural dweller and thus was entitled to register as a nonagricultural household ( fei nongye hukou). This registration guaranteed access to a number of benefits denied to the peasants, such as secure employment, affordable housing, inexpensive medical care, and retirement income (Lü and Perry: 1997, p. 3). Housing provision was one aspect of workplace support that made the system distinct from the system that operated in the Soviet Union (Naughton: 1997). Before the economic reforms began, the workplaces controlled 90 percent of urban public housing, even though local government managed most of the investment. The system of housing provision was strengthened by the adoption of a Soviet-style plan in the mid-1950s. The massive investment in building an urban, industrial base that was part and parcel of the adoption of Soviet-style planning meant that there had to be an accompanying expansion of the provision of public goods and services. Rather than setting up an elaborate government administrative mechanism, Chinese policymakers relied on their earlier experiences of workplace provision and decided to attach infrastructure and service provision to each individual project (Bray: 2005, p. 111). Thus, each budget for a new industrial complex included funding for the construction of houses, schools, clinics, and other necessary infrastructure. The First Five-Year Plan (1953–1957) (1956, pp. 195–96) notes that of the 46 million square meters of new housing, 33 million square meters were to be allotted to projects directly affiliated with state-owned enterprises or governments departments. The workers in the SOEs and in the government bureaucracy formed the elite within the privileged urban population. Within the SOE sector, the provision of goods and services was very uneven, depending on the wealth and status of the individual enterprise. Employment in a large Shanghai state enterprise would provide one with better housing, schooling for children, and retirement prospects than work in even a large factory in a small city in the hinterland. Labor mobility was positively discouraged and workers were most likely to spend their entire lives in the workplaces to which they were assigned on graduation from school or college. In fact, an employee had to get permission from the workplace to change jobs, giving great power to the personnel departments of the workplace. This power of control was strengthened by the fact that a dossier (dang’an) on each worker was kept by the personnel department. This dossier contained not only biographical details but also information about political attitudes

36 / providing public goods in transitional china

and performance in party and mobilization campaigns. Later, biological information on women’s menstrual cycles was also kept so that familyplanning quotas could be implemented better. The notion of danwei as defining one’s identity was so pervasive that on answering the phone or on meeting someone for the first time, almost always the first question would be “Which danwei do you belong to?” The reply would help the inquirer to gauge whether the interlocutor was of superior, equal, or inferior status. In fact, when the reforms began, a number of young urban Chinese who desired to shock put their affiliation on the ubiquitous name card as “No danwei.” This was seen as a sign of rebelliousness and nonconformity to existing norms and structures. Reforming this structure was a major challenge for policymakers, especially beginning in the early 1990s. The existence of the workplace-based system for providing so many public goods meant that the state did not develop a complex social security apparatus or retirement fund. Local governments had to cover support only for those within the jurisdiction who were not employed in a SOE or with the government or those who had no family members to fall back on. For those in the best workplaces, many services were free or largely subsidized and cost very little. The need to adapt to gradually increasing costs and to purchasing the services in the marketplace caused a shock for the urban population. For those who did not have a workplace to fall back on, the CCP developed an urban-based welfare system. However, this system was not generous and, as we have noted above, Mao Zedong urged a gradual approach and warned against developing a psychology of dependency on government handouts. This was understandable given the problems of displacement, inflation, unemployment, and poverty that the CCP had faced when it took over urban China. The urban welfare expenditures administered by the People’s Relief Administration skyrocketed. Eckstein (1962, p. 39) has estimated that in 1952 some 40 million yuan was spent on unemployment relief alone. Despite the mobilization of neighborhood groups, same place associations, and other relief organizations, in early 1954, some 1.2 million city dwellers were said to be still receiving welfare relief from the government (Dixon: 1981, p. 159). To deal with this problem, the CCP set up a district-based geographic system to complement the workplace system. Below the district level of government, the CCP established a system of residents’ committees (jumin weiyuanhui) under which citizens were organized into small groups of residents. Among their various duties, these committees were charged with looking after public welfare under the direction of the district authorities. In addition to mediating local disputes, the main area of work was

welfare provision, 1949–1979 / 37

to mobilize support for those within the community who could not take care of themselves. To this end, the committees could dispense subsistence allowances for the unemployed by using funds allocated by the municipal or provincial authorities. The committees had no reliable financial base but were allowed to raise “voluntary” donations, supplemented by small user charges, to cover the costs of their activities. The committees also played an important role in public surveillance and in trying to ensure compliance with political campaigns and in spotting any deviant behavior. In fact, their roles were intertwined with those of the public security apparatus as regulations from December 1954 called on public security substations to play an active role in assisting with welfare work. As Dixon (1981, p. 162) has noted, this combination of welfare and public security work meant that welfare assistance could be used as a reward for good behavior and withheld from those who came from “bad” families or who were engaged in heterodox thinking. Mass organizations such as the local branches of the Women’s Federation were also entrusted to help with welfare work. In the countryside, the collectives took on the role of the urban workplace, albeit with far less support. The first step away from household farming was the formation of mutual-aid teams (1952–1954) that comprised up to 20 households cooperating over the course of the agricultural year. Labor was pooled but agricultural land, tools, and animals remained in private hands. This phase was followed by the formation of lower-stage agricultural producers’ cooperatives (1954–1955), voluntary associations of some 30 households that pooled not only labor but also land and tools and derived income based on their original input. In February 1953, the CCP Central Committee (CC) cautiously announced that these agricultural units had a responsibility to provide welfare assistance to needy members, but that the allocated funds should not exceed 1–5 percent of the annual income of the entity. A national survey of agricultural producers’ cooperatives showed that the lower-stage coops were allocating 1.4 percent of their net income to welfare funds whereas the higher-stage coops were allocating 1.6 percent (Dixon: 1981, p. 190). The higher-stage agricultural producers’ cooperatives, which were developed during 1956–1957, comprised between 100 and 300 households with income now dependent on earned work-points. These cooperative structures were soon supplanted by the formation of the communes during the Great Leap Forward. The communes were set up not only to deal with agricultural production but also to carry out other economic, political, and military work and to deal with health and social welfare. These became the rural equivalents of the urban “small society” workplace. Initially, 24,000 communes were established, but that number

38 / providing public goods in transitional china

was increased to 74,000 during the 1962 retrenchment that followed the excesses of the high-tide of the Great Leap. During the high-tide of the Great Leap, communes were expected to take over from the family the role of the main welfare provider and to supply a basic range of free goods to commune members. This was soon discovered to be unrealistic. By the end of 1958, the free supply of goods and other necessities made up about 50 percent of the peasants’ incomes (Dixon: 1981, p. 197). As the disastrous consequences of the Great Leap became apparent and the commune experiment was moderated, the role of the family was reemphasized and experiments such as the communal mess halls and kindergartens were abandoned. The cooperatives resumed the role of providing collective support, with the responsibility for welfare administration gradually devolving to the brigade and then to the production team.2 Social insurance and collective welfare were to be covered by funds deducted from the total income of the production team. The specific amount was left open to discussion but it was not to exceed 2–3 percent. To prevent the kind of abuse of funds that had occurred previously, the regulations stipulated that use of the funds should be decided at a general meeting and not by a cabal of officials. This meeting was also to decide on the disbursement of funds to those who were unable to work or who endured particular hardships. At the same time, the regulations forbade the production brigade and the commune from raiding the funds for “several years to come.” The brigade was responsible for helping manage the team’s affairs as well as for dealing with education and public health. One innovation introduced during the Cultural Revolution was the establishment of a voluntary, commune-based cooperative medical service. This replaced the previous method of covering the costs from the team’s welfare fund. The expansion of those in need and the growth of paramedics (barefoot doctors) during the Great Leap and the Cultural Revolution meant that the old system was no longer viable. Participants were expected to pay a small annual premium of between 1 and 2 yuan. According to Prybyla (1981, p. 265), for a family of five that was exceptionally welloff and that had three people working, this would amount to 0.5 percent of the gross income from commune work. For the same family in a very poor production team, this would amount to 8.3 percent of the gross income. There was also a co-payment fee of a few cents when healthcare was sought. This was waived for those in dire need. In addition to these payments by the consumers, the production brigade paid a contribution of up to 1 yuan out of the welfare fund for each participating member. In addition, the production brigade set aside funds from its welfare and construction fund.

welfare provision, 1949–1979 / 39

The commune also set aside monies from its welfare fund and from the reserve fund to help cover capital expenditures. The system was intended to be self-financing to the maximum extent possible and to rely on local resources. However, some transfers from the state were made available for the construction of hospitals or clinics or for the provision of medicines and medical equipment that may not have been available locally. This fit with the Maoist exhortation to practice “self-reliance.” The burden on the state was further reduced by attempts to keep costs down. The paramedics were trained in situ and although they did receive remuneration for their medical work, they also received work-points for their contribution to agricultural production. The attendant health workers received work-points from the team, thus minimizing extra financial investment. Despite this rudimentary coverage, the costs to the peasant household could be considerable. If a patient had to be treated in a hospital outside of the brigade, treatment costs could still be sizable (Prybyla: 1981, p. 266). A Balance Sheet of the Mao Years There is a general perception that the Chinese system up until 1978 performed well in terms of providing basic public goods and services, certainly given the low income level. As with other state socialist systems that operated in a low-wage economy, income was supplemented by state subsidies to key groups, and the state apparatus was used to deliver key goods and services without necessary regard to cost. For example, in the mid-1970s, China received frequent praise from international organizations for the level of healthcare provided, especially in the rural areas. In fact, China’s barefoot doctor approach was held up as the model for community-based healthcare at the WHO Alma Alta Conference in 1978, when the slogan “Health for all by the year 2000” was adopted (Saich and Kaufman: 2005, p. 179). However, actual data are very hard to locate and many of the assessments have been by inference. In fact, in 1959, the statistical system ceased to function altogether as a result of the disaster of the Great Leap Forward. Whatever successes there may have been, they were all but impossible to maintain once policy shifted to abandoning the collectives and forcing the SOEs to be more financially responsible. However, evidence also suggests that the collective-based model of public goods provision in the countryside may not have been tenable over the long term irrespective of the other reforms. According to Du Ying (2000, pp. 36–37), by the mid-1970s, the majority of communeand brigade-run health cooperatives had already collapsed or existed in name only. He attributes the lack of sustainability to a limited capacity to raise funds, poor financial supervision, and abuse of the system.

40 / providing public goods in transitional china

In particular, peasants were unwilling to entrust cadres to manage their funds, and many officials took advantage of the system to gain priority access for their family and friends. Such abuses resulted in a moral crisis and a breakdown in trust. Thus, Du sees the main cause of the collapse of the rural medical cooperatives as lying with the structural logic of the system rather than with the economic reforms and implies that collapse was inevitable. As noted earlier, service provision was dependent on location. The urban-rural bias was pronounced, but even within the countryside there was great variation. As with other polities, the medical system favored urban-based, curative medical care, and the greatest concentration of medical personnel was in the cities. Before the Cultural Revolution, the doctor-to-population ratio in the cities was 1:5000 and in the rural areas it was 1:8000. The statistics for doctor-to-population ratios in 1965 placed China behind Mexico, India, and the Philippines, while the ratio was one-third that of Japan’s at the end of the nineteenth century (Prybyla: 1981, pp. 25–51). Financing was heavily biased in favor of the urban areas, accounting for around 80 percent of total government expenditures in 1966; 80 percent of the physicians too were working in urban China. These were the kinds of imbalances that the Cultural Revolution was meant to address, with attempts to shift more funding to the rural areas and to “send down” qualified personnel to work in the countryside. These imbalances meant that most of the health gains prior to the Cultural Revolution came from the peaceful environment that followed the three decades of war and dislocation and the improvements in public hygiene often through coordinated campaigns. Participation in the rural medical schemes was voluntary and participation rates often varied. In one commune near Guangzhou, for example, only between 3,000 and 6,000 households out of 15,200 participated. As in other schemes, the healthier and more prosperous tended to stay out of the collective systems, leaving them to the poor and the sick. This put greater financial strains on the system. These inequalities and problems notwithstanding, progress was clear. The population rose from 575 million in 1952 to 962.6 million in 1978, with the birth rate dropping from 37 percent to 18.25 percent after having reached a peak of 37.88 in 1965. The death rate dropped from 17 percent in 1952 to 6.25 percent in 1978. There was a marked rise to 25.43 percent in 1960 that reflects the consequences of the policies of the Great Leap Forward (NSB: 1983, pp. 6, 105). The Great Leap famine caused the cumulative excess mortality to reach 11 percent of the population in Sichuan and almost 6 percent in Guizhou and Anhui (Naughton: 2007, p. 71). The various population rates for the early 1950s resembled

welfare provision, 1949–1979 / 41

those of India and Indonesia in the early 1950s, and China’s annual population increase of 1.9 percent was the same as that of Bangladesh and slightly lower than comparable countries such as India (2.1 percent) and Indonesia (2.3 percent) (Eberstadt: 1979, p. 5). The variance between urban and rural China was clearly apparent. The birth rate for urban China fell from 42.45 percent in 1954 to 13.56 percent in 1978, and the death rate dropped from 8.07 percent to 5.12 percent in 1978, with a peak in 1960 of 13.77 percent. In rural China, the birth rate dropped from 37.51 percent in 1954 to 18.91 percent in 1978, and the death rate dropped from 13.71 percent to 6.42 percent, with a peak of 28.58 percent in 1960, revealing that the famine hit hardest in the rural areas. The disastrous policies of the Great Leap Forward caused a major disruption in China’s population growth. Bannister (1984) calculates that life expectancy may have dropped by almost one-half, from 50 to 25 years. The Great Leap Forward notwithstanding, life expectancy rose from around 40 years to 62–63 years, putting China well ahead of India and Indonesia and on a par with countries such as South Korea and Sri Lanka but behind Malaysia and Taiwan. Statistics for infant mortality rates tell the same story, with China outperforming other large Asian countries such as India and Indonesia and roughly on par with South Korea and Sri Lanka, but trailing Malaysia and Taiwan (for details, see Eberstadt: 1988, pp. 136, 154, see table 2.1). In comparison with the other two large Asian countries, India and Indonesia, China was performing well, but it was not performing as well as the economies that would come to be known as the Asian tigers. The tigers’ economic growth, generated by their market-based policies and export orientation combined with judicious government intervention, allowed them to develop social Table 2.1 Comparative Indicators for China’s Social Development, 1950 to 1980 Country

China India Indonesia Malaysia South Korea Sri Lanka Taiwan

Life Expectancy at Birth

Infant Mortality Rates

1950

1960

1970

1980

40 32 36 52 48 58 55

25 41 40 57 52 62 65

61 47 45 66 63 65 69

65 52 51 70 66 68 72

Source: Eberstadt: 1988.

1950 175 190 166 95 116 81 91

1960 284 157 145 65 93 66 42

1970

1980

70 133 112 40 52 52 27

42 129 99 25 38 38 22

42 / providing public goods in transitional china

indicators that were the equal of—and in some cases, even better than— China’s state-socialist approach. China showed similarly impressive results for state investment in infrastructure. In the rural areas, the proportion of irrigated farmland increased from around one-quarter to about one-half and the use of fertilizer per hectare tripled (Eberstadt: 1979, p. 11). However, productivity remained low despite these impressive inputs. Grain was the most important product for most in China throughout this period; it was only in 1957 that grain levels of production reached those of the early 1930s before the Japanese invasion and the ensuing political turmoil. In 1957, grain production reached 195 million tons but it dropped to a low of 143.5 million tons in 1960 before rising steadily to 304.8 million tons in 1978. Over this time, the population rose from some 500 million in the early 1930s to 574 million in 1952 and 962.6 million in 1978. This means that in terms of productivity per capita there was a drop of some 7 percent from the early 1930s to 1978. Clearly, the existing rural production system was proving ineffective and required increasing levels of resource input. This may account for the pressure to shift to a household system of production to increase incentives. This reversal to the household had major consequences for the provision of public goods and services. In the next chapter, we turn to public service provision in the post-1978 period.

Ch a p t e r Th r e e Th e P rov ision of P u bl ic Go od s du r i ng t h e R e for m P e r iod

The economic reforms that began in the late 1970s have had a major impact on welfare provision. Although standards of living for most have been raised substantially as reflected in improved social indicators, this improvement has been accompanied by increased vulnerabilities for many in the population. In the countryside, the return to household-based farming hastened the unraveling of the collective structures that had provided public goods and services. This is reflected in the rising numbers of people without medical insurance and in medical costs becoming a major cause of poverty. It took longer for reform to impact on the welfare structures in urban China, but through the 1990s as enterprises were forced to focus on profits, they began to abandon their high-cost benefits and subsidies. While existing systems collapsed or frayed, new categories of citizens emerged with different demands on the state. The two most notable groups are the migrants who have helped drive urban development but who enjoy little access to urban social services, and the new urban poor who are the products rather than the beneficiaries of reform. Clearly neither of these two groups could be taken care of through a workplacebased system. Reforms have changed notions of entitlement, with access to services much more tightly tied to financial capacity than in the past. The linkage of service provision to ability to pay has produced new inequalities and exacerbated old ones. In particular, there has been a dramatic rise in the disparity between welfare provision in rural and urban China and an abandonment of the old cradle-to-grave social compact for the privileged urban working class. This has presented the CCP leadership with policy challenges to devise new institutions and mechanisms for service provision. The changes also raise fundamental questions about citizenship and entitlement. To whom does the state have an obligation

44 / providing public goods in transitional china

to provide welfare? And at what level? The challenge for the central leadership has been to shift from a system of severely frayed safety nets for the urban working class and government officials to a clearer articulation of what China’s citizens can expect in terms of welfare guarantees. An ideal set of policies would address the dualistic development strategy that privileges the urban population over the rural population. The literal fencing-off of the rural from the urban during the Mao years made it easier for city dwellers to enjoy their privileged position. Even though the household registration system remains, reforms have driven huge holes in this fence and as a result it is increasingly difficult to justify this privileged position politically. The market-driven economy that has created a huge influx of rural migrants to the cities has inextricably linked the two worlds; migrants and city dwellers now often share the same neighborhood. The inequities of providing education and medical services to one group while denying them to the other have become a source of tension. Thus, despite the progress, there have been high costs for some; for the absolute poor, there may be no service provider available. Unregulated privatization is no solution. As the World Bank (1998) has noted, unless the field is leveled and effective redistributive mechanisms established to compensate the poor, privatization heightens the inequality of opportunity, with access to key public goods and services being more dependent on income than ever before. Given the major institutional changes that were unleashed, it is perhaps surprising how slow policymakers were to respond. In this respect, China resembles other transitional economies that were also slow to respond to the new social needs (Kornai: 1997; Nelson: 1997). With the focus on economic growth and structural adjustment and stabilization, far less attention was paid to the social costs. This relative neglect was compounded by a lack of understanding about what markets can and cannot provide and by the absence of effective civil society institutions that could cushion the blow for the vulnerable populations. Insufficient policy attention persisted through the 1980s and into the first half of the 1990s. In the early 1990s, the central government began to focus on financial reforms and a more thoroughgoing reform of the state sector to prepare China for its WTO entry. It was clear that the social costs of restructuring the SOEs and the attendant rise in unemployment could be significant and potentially destabilizing to the sociopolitical system as a whole. Consequently, the central leadership began to pull together local experiments with reform into a comprehensive policy framework for the privileged groups in urban China. This involved transferring responsibility for social welfare from the

public goods during the reform period / 45

workplace to local governments. However, the reformed system still left most people in rural China and those working in the informal sector to their own devices. A third phase was launched when Hu Jintao and Wen Jiabao came to power in 2002–2003. From the start, their rhetoric implied that they wished to redirect more resources to those who had not benefited so well from the reform program to date. There was a wider recognition that some of the problems of social development would not be solved by growth alone. This realization led to the development of a more coherent policy framework based on identification of vulnerable groups that were provided with targeted support. Policy is moving from provision of shortterm safety nets to developing a more integrated, comprehensive system. The new policy agenda was presented at the October 2006 Sixth Plenum of the Sixteenth Central Committee. The plenum, remarkable for its focus on social development, drew policy together under the slogan of “building a harmonious society.” Among the policies most relevant for our study were attempts to reduce income inequality, improve access to healthcare and education for migrants and those in the rural areas, improve and extend the social security system, and invest more heavily in basic infrastructure for the western and northeastern regions of the country and for the rural areas in general. This chapter first provides some basic data on social progress during the reform period. Second, we analyze the main trends that have impacted on service provision. It is important to understand how public goods are financed; in China, this responsibility falls heavily on the subnational level of government. Thus, third, we review the financial situation and conclude by looking at the incentives for local officials and governments to provide public goods and services. Social Development in the Reform Period (1978–Present) Although a number of factors have altered or weakened the state’s capacity to ensure that welfare provision has kept pace with economic growth, this needs to be placed in context. In recent decades, China has ranked well in terms of the human development index (HDI) adopted by the United Nations Development Programme (UNDP) as a composite measure of social development.1 China’s HDI rose from 0.527 in 1975 to 0.628 in 1990 to 0.730 in 2000 and 0.777 in 2005. These figures reveal the significant impact of the rapid economic growth of the 1990s. However, we also see the relative neglect of China’s social development as compared with GDP

46 / providing public goods in transitional china

growth over the 1990s. In comparison with other countries, China’s HDI ranking declined in the 1990s, from 87 in 1990 to 104 in 2001 before improving in later years to 81 in 2005. The difference between China’s HDI and GDP per capita (PPP in US dollars) also narrowed rapidly in the 1990s because of the rapid economic growth and increased per capita income. In 1993, China’s HDI rank was 41 places above its GDP per capita rank, but by 2001 its HDI rank was negative two. The pro-poor rural policies of Hu Jintao and Wen Jiabao to address the glaring inequalities appear to be having a visible effect, as the ranking gap between HDI and GDP per capita rose to plus five in 2005 (UNDP: Human Development Report, various years and 2007). We do see the importance of overall economic growth, however, as the HDI improvement was very strong throughout the 1990s but less so from 1975 to 1990. China at 0.777 in 2005 outperformed the other two main populous nations in Asia—India (0.619) and Indonesia (0.728)—but still lagged slightly behind countries such as Thailand and Malaysia. China was on track to achieve most of the targets of the Millennium Development Goals by 2015, with the exception of HIV/AIDS control, promoting gender equality, and providing safe drinking water for the rural population (UNDP: 2004). China’s overall HDI ranking is comparable to a high-middle level of human development, but there was significant regional variation. Five of China’s western provinces had an HDI ranging from 0.650 to 0.5921, putting them in the middle-development category. Even Tibet, with the lowest ranking (0.5921), was above the cut-off for low development (0.5). By contrast, three of the major municipalities in China (Beijing, Shanghai, and Tianjin) had an HDI of over 0.800, placing them in a high level of development and at the approximate level of South Korea (Hu Angang: 2004, p. 8; UNDP: 2004, p. 3; see table 3.1). It is interesting to note that the regional disparities for the HDI index are lower than those for per capita income, attesting also to the relatively good social development. One of the most notable features of China’s reforms has been the massive reduction in the number of people living in absolute poverty. Hundreds of millions have benefited from the economic reforms. According to UNDP figures, only 9.9 percent of the population in 2005 was living with an income below one dollar per day and 34.9 percent was living with an income below two dollars per day. In 2005, China ranked number 29 on the UNDP’s Human Poverty Index,2 but China’s HPI minus its income poverty was negative 3, thus indicating that there is still a certain degree of underperformance. Comparatively this was much better, however; in 2004 the difference was minus 14 (UNDP: 2006,

public goods during the reform period / 47 Table 3.1 China Human Development Index Comparison (for 2003, unless otherwise stated) Country or Region Shanghai South Korea Beijing Tianjin Guangdong Jiangsu Heilongjiang Malaysia

HDI Figure

Country or Region

HDI Figure

Country or Region

HDI Figure

0.91 0.90 0.88 0.86 0.81 0.81 0.79 0.79

Thailand China 2004 China 2003 Shaanxi Henan Sichuan Indonesia Vietnam

0.78 0.77 0.76 0.75 0.74 0.73 0.70 0.70

Yunnan Guizhou India Tibet China 1990 China 1975 China 1960

0.66 0.64 0.60 0.59 0.57 0.52 0.29

—

—

Sources: Hu Angang: 2004; UNDP: 2004.

2007). However, these figures and Chinese official poverty statistics only refer to rural China. Now there are also around 30 million urban poor, this is excluding the migrant population. Further, a 2007 assessment of China’s economy by the World Bank based on a different assessment of what a dollar can buy in China suggests that the numbers in poverty may be considerably higher, perhaps nearing 300 million rather than 100 million. Such statistics reveal that despite tremendous progress, China confronts major policy challenges that are relevant for our study. First, a significant group of rural poor have not responded to the policy measures, market openings, and the benefits of “trickle down.” Second, a very large group is vulnerable to an economic downturn and inclined to recidivism. Third, a smaller but rising number of urban poor are the products rather than the beneficiaries of reform. This argues for public policy based on a better understanding of who is poor and why. China’s social indicators have continued to improve throughout the reform period and, for the most part, they place China in the middle of the middle-income countries (table 3.2). China has had better social indicators than the large Asian countries of Indonesia and India but has lagged behind the tiger economies such as Malaysia, South Korea, and Thailand. Life expectancy by 2000–2005 had risen to a respectable 71.5 years. As noted in Chapter One, public spending on health was roughly comparable to that in other middle-income Asian countries but was significantly below Malaysia and Thailand and even Vietnam. In terms of investment and performance, the reform period has maintained the rural-urban dualism and has actually led to an increase in inequalities. Chapter Four examines this in more detail for the health sector, but the inequalities exist across the board. The rural-urban income

Table 3.2 Social Indicators for Selected Asian Countries Country

China India Indonesia Malaysia Philippines South Korea Thailand Vietnam

HDI Rank

HPI Rank

HDI-GNP Infant Rank Mortality Rate Per 1000

2006

2004

2004

81 126 108 61 84 26 74 109

26 55 41 15 31 — 19 33

2 0 0 0 2 –1 2 2

1970 85 127 104 46 90 43 74 55

Under-5 Mortality Rate Per 1000

Public Spending on Education as % of GDP

2004

1970

2004

1991

2002–04

26 62 30 10 34 5 18 17

120 202 172 70 90 54 102 87

31 85 38 12 34 6 21 23

2.2 3.7 1.0 5.1 3.0 3.8 3.1 1.8

2.0* 3.3 0.9 8.0 3.8 4.6 4.2 —

HDI Human Development Index

HPI Human Poverty Index HDI-GNP Rank Human Development Index minus Gross National Product per capita * Author’s estimate Source: UNDP statistics at http://hdr.undp.org/en/statistics.

Public Spending on Health as % of GDP

Life Expectancy

GDP Per Capita PPP

2003–04

1970–75 2000–05

2004

2.0 1.2 1.1 2.2 1.4 2.8 2.0 1.5

63.2 50.3 49.2 63.0 58.1 62.6 61.0 50.3

71.5 63.1 66.5 73.0 70.2 76.9 69.7 70.4

5,896 3,139 3,609 10,276 4,614 20,499 8,090 2,745

public goods during the reform period / 49

gap is much larger than either the intra-rural or intra-urban income gaps. During the first few years of reform until 1984, the rural-urban income gap declined as a result of the spread of the household responsibility system and the increase in the state purchasing price for key rural products. This situation changed in the mid-1980s as policy attention began to focus on developing the urban economy. As a result, the rural-urban income gap rose steadily to 2.2 in 1990, 2.86 in 1994, 2.79 in 2000, and 3.2 in 2005. If the urban subsidies are included, the gap would be even greater. According to the International Labour Organization, China’s rural-urban gap is the largest in the world. This gap is even greater in China’s poorest provinces. Thus, although the gap is 2.26:1 in Shanghai, it is 4.34:1 in Guizhou and 4.09:1 in Gansu. Such income differentials contribute to significant differences in all social indicators, as will be discussed in subsequent chapters. Factors Impacting on the Provision of Public Goods and Services Hussein (2002) describes three concurrent transitions that have impacted on the demands for social welfare provision. Two are economic transitions and the third is the demographic transition from a young to an aging population. The first is a “normal” economic transition, with many in the population shifting from low-productivity agricultural labor to employment in the higher productivity manufacturing and service sectors. Parallel to this transition is the shift from a centrally planned economy to one that is more influenced by market forces. To these three transitions we can also add a fourth: the transition from a belief in the collective to a focus on individual wants and desires. As collective institutions no longer are key welfare providers, individuals are relying on their resources and those of their extended personal networks or have begun to purchase services in the marketplace. The move out of agriculture has been important to improve livelihoods, but the population shifts have presented challenges in terms of integrating the new populations into a better framework for public goods provision. With the return to family farming under the household responsibility system, a large pool of labor was released to work in rural small-scale industry as well as to provide the migrant labor to underpin the urban construction and services boom. At their height, township and village enterprises employed some 120 million workers. The 2000 census noted 144 million (12 percent of the population) living away from their registered abode. Of these, 79 million were long-distance migrants, up from 7 million in 1982 and 22 million in 1990. In a province such as

50 / providing public goods in transitional china

Guangdong, migrants make up 25 percent of the population, whereas in Shanghai they constitute over 10 percent of the population (Naughton: 2007, pp. 129–30). Uprooted from the land, these migrants have not been effectively integrated into the new or preexisting systems for health and education, for example. The pace of urbanization has accelerated and over the next two decades it will increase further. The hope is to move between 300 and 500 million people from the rural areas to towns and cities by 2020, making an urban population of around 800 million (Financial Times, November 27, 2003). In fact, urbanization is seen as the best way to provide a long-term solution to the problems of inequality that stem primarily from the rural-urban differences. The main policy debate has been over how to manage this urbanization process. The debate is divided over whether the focus should be on developing major megalopolises as migrant destinations along the coast or on furthering more controlled development based on small town expansion in the countryside. The latter is more in line with the traditional ideas of sociologist Fei Xiaotong. At the Seventeenth Party Congress (October 2007), it was stated that that “city clusters with mega-cities at the core” are described as new poles of economic growth. This suggests that the program of urbanization will move ahead despite the continued commitment to policies to “build a new socialist countryside.” The 2000 census calculated an urban population in China of 455.94 million (36.09 percent).3 This was a 9.86 percent increase over 1990, the year of the previous census. However, getting an accurate count of China’s urbanization is very difficult because of different counting systems that are used (for an excellent review, see Chan: 2007). If we take the census figure in comparative terms, we see that today China is “under-urbanized” as a result of the previously controlled urban flow and the household registration system. Other transitional societies have higher rates of urbanization—such as Hungary (64 percent) and Russia (73 percent)—as do the remaining socialist countries—Cuba (75 percent) and North Korea (59 percent) (Zhou and Ma: 2003, p. 177). Zhou and Ma (2003, p. 196), using a constant basis for calculating the urbanization rate rather than the varying basis of the Chinese census, come up with an urbanization figure of 36.25 percent for year-end 2000. Over the next ten years, this figure is expected to increase to 40 percent at current rates of growth. But with current policy favoring urbanization, it is likely to be higher, perhaps reaching the 55–60 percent range (Wang Mengkui, quoted in Financial Times, November 27, 2003). Given the level of industrialization, the urbanization rate in China should be around 60 percent. Interestingly, on the sidelines of the Seventeenth Party Congress, Guo Shuqing, chairman

public goods during the reform period / 51

of the Board of the Construction Bank of China, commented that if, following the World Bank criterion, we did not consider as farmers all those who worked in the cities for at least three months of the year, then China’s urbanization rate was already 60 to 65 percent. This is a significant difference with major consequences for effective government investment and service provision. Meeting the urbanization goals will present major challenges for the government in terms of investment in urban infrastructure and planning. It will also present significant challenges for job creation and the provision of social welfare. The programs for infrastructure development will provide employment opportunities; however, whether the service industries will be expanded sufficiently to deal with this increased urbanization and will be able to accommodate the estimated 150–200 million surplus laborers in rural China remains to be seen. In rural areas, the shift to jobs with higher productivity, combined with the move away from a centrally planned economy, has propelled China to average annual growth rates of around 10 percent. However, the economic transition has also forced financial sobriety on the previously economically inefficient state-owned enterprises. SOEs have had to shed many of the welfare functions that they undertook in the prereform period as well as to lay off a considerable number of workers. Official figures for registered unemployment show a rise from 3 percent (6 million) in the mid-1990s to just over 5 percent by end-2005, despite employment in the SOE sector dropping by about 50 million. These figures exclude those who have not registered as unemployed, including the large number of workers laid off (xiagang) by still-functioning SOEs. Thus, the real unemployment rate is much higher although difficult to calculate but it may have increased from 6 percent in 1993 to 12 percent by 2000 (Giles et al.: 2005; Knight and Xue: 2005). Scholars at the Rand Corporation (Wolf et al.: 2003) estimate an open and disguised unemployment rate of around 23 percent, or 170 million. The rural changes and the SOE reforms have led to a breakdown in the role of the collective in the life of citizens, especially with respect to the provision of public goods. Socialism is based on the premise that individuals will draw more benefits by being a part of the collective in a vertically organized, centrally planned economy rather than by acting on their own individual initiative. By contrast, horizontally organized markets place the emphasis on individual choice. The transitions underway in China today have resulted in many more citizens having to rely on their own resources to access welfare. As noted above, access to education and healthcare are more dependent on income than ever before, and families are contributing greater percentages of their income to purchase health and education services. This rise in individual choice,

52 / providing public goods in transitional china

by both default and design, will have a major impact on the way in which citizens expect to receive their services and also on the demands they make in terms of the quality and accountability of their service providers. It is important to understand the demographic transition that China is undergoing as this will have a fundamental impact on social welfare needs. First, the population is aging significantly and China will be one of the first societies to grow old before it grows affluent. This will have significant consequences on savings, which will be lower (Cheng: 2003), dependency ratios, and on the need to increase spending on medical care. Overall, China will age 13.8 years during the first half of this century, as opposed to the United States which will age 3.6 years (Hewitt: 2004, p. 103). The Chinese population will continue to age as the country industrializes and urbanizes, with the consequence that there will be an even greater need to maintain rapid and sustainable economic growth. At the end of the twentieth century, China officially entered the aging stage in terms of the internationally recognized criterion, with 10 percent of its population over 60 years of age. By the end of 2006, there were 149 million people over 60, accounting for 11.3 percent of the population (China Daily, December 18, 2007). With the elderly growing at 3.2 percent per annum, five times the total population growth, the number of those over 60 will be 248 million by 2020 and 437 million (over one quarter of the population) by 2050. According to the United Nations, the number of the oldest among the old (those over 80 years of age) will also increase from 11.5 million in the year 2000 to 27 million in 2020 and 99 million in 2050. This group that comprised 13 percent of the elderly population in 2000 will comprise 30 percent of the population in 2050, growing faster than any other cohort (Zeng Yi et al.: 2002, p. 251). The aging population will present serious policy challenges in terms of dependency ratios and pension obligations that the state will be required to meet. The structure of the population will be what Chinese researchers refer to as a “4-2-1 family,” with four grandparents, two parents, and one child. Children will have no siblings and they will have to carry a heavy burden of support. The aging population will also require a significant increase in medical costs. In reviewing the situation in the United States, Torrey (1992, p. 382) estimates the cost of long-term care for this group to be 14.4 times higher than that for the 65–74 age group. The medicare costs are 77 percent higher than those for those aged between 65 and 69. In a city such as Shanghai, the aging is particularly severe, having gone from one of the youngest cities in the world at the time of the CCP

public goods during the reform period / 53

victory in 1949 to one of the oldest in the twenty-first century. When the CCP took power in Shanghai, only 3.6 percent of the population was over 60 years of age, but 9 percent were 60 when reforms began and the total was set to peak at 32 percent in 2030 (Zhang Kaimin: 2001).4 Those over 80 years of age will peak at 16.3 percent of the total Shanghai population. By 2030, the Shanghai elderly will be 1.67 times more numerous than those under the age of 16. This means that during the 30 years of reform, Shanghai has undergone a demographic transition that took 140 years in France and 85 years in Sweden. The impact is obvious: with many inner-city schools closing down or becoming boarding schools for non-Shanghai residents, and with many people worried about who will look after them in old age. Some policy advisors have somewhat facetiously suggested that there should be a financial incentive to increase family size! These demographics will have major consequences for dependency ratios and pension obligations. In 2004, official figures showed that the total dependency ratio was 38.63, with that for the elderly at 11.87, but the ratio of the working to nonworking population was dropping fast, at 6:1 in 1991 and anticipated to be 2:1 by 2020 (Lin and Zhai: 1996, p. 349). However, as Hussein has perceptively pointed out, an exclusive focus on the dependency ratio of the elderly, which has been the focus of the debate around pension costs, is misleading. This ignores the large economic plus that comes from the declining dependency ratio of children (26.76 in 2004) that will bring benefits before the aging costs really begin to impact heavily. By 2020, it is calculated that children (0–14 years of age) will have dropped from 40.4 percent of the population in 1964 to 19.3 percent (Hussein: 2002, p. 1828). The UNDP calculates a slightly steeper drop to 18.5 percent by 2015 (UNDP: 2006). It is important for policymakers to keep in mind that there is great variation in the dependency ratios across China and thus some policy flexibility will be necessary. The 2004 total dependency ratio varied from a low of 27.14 percent in Beijing to a high of almost 50 percent in Guangdong (49.39) and Guizhou (49.37). The highest child dependency ratio is in Guizhou at 38.11 percent, with the lowest in Shanghai at 11.6 percent. Conversely, Shanghai has the highest elderly dependency ratio at 20.31 percent, with Qinghai (8.32) and Ningxia (8.28) the lowest. The northwestern and northeastern provinces tend to have the lowest elderly dependency ratios, whereas the poorer provinces in the northwest and southwest, such as Guizhou (38.11), tend to have the highest child dependency ratios, an exception being Guangdong (37.57) (NSB at http://www/statsgov.cn/tjs/ ndsj/2007/index.htm). Furthermore, the nature of support for these two groups differs (Hussein: 2002, pp. 1829–30). For children, much of the

54 / providing public goods in transitional china

financial cost and care falls on the household, with the state picking up a large part of the external education costs, for example. For the elderly, the household carries the main burden, especially in the rural areas. In the urban areas, the picture is different, with the workplace and the local state carrying the pension burden for many. Some 71 percent of retirees are from the SOE sector (Hussein: 2002, p. 1830). In addition to aging, another adverse consequence of the familyplanning policy is the distortion in the male-female ratios. Essentially, there are three ways to meet the strong demand for male offspring: have more births, engage in female infanticide and generally discriminate against the girl child, or carry out induced abortions following prenatal sex identification. In some counties, the reported discrepancy between female and male children is alarmingly large. However, it is not at all clear how reliable these figures are. For example, one early study shows that an underreporting of births accounts for between 50 and 70 percent of the differential sex ratio at birth (Zeng et al.: 1993). The same study also claims that female infanticide accounts for 5 percent of this differential (see also Li: 1992). Part of the discrepancy may be accounted for by underreporting, but clearly not all. The current census shows that the ratio of males to females is 106.74:100, resulting in 41.27 million more men than women. However, at birth the ratio is 119.92:100 and by age 4 it is 120.17:100. Jiangxi and Guangdong have ratios of 138:01:100 and 137.76:100 respectively, with rural Guangdong at a rate of 143.7:100 (Zhang Yi: 2003, pp. 66, 73). The ratio for those in the 10–15 age group is much lower, at 112.43:100 (NSB at http://www/statsgov.cn/tjs/ndsj/2007/ index.htm). This clearly indicates a strong trend toward a further imbalance in the ratio over the next 10 to 15 years and beyond. These ratios will mean that at least 1 million men per year will not be able to find a marriage partner (Zhu et al.: 1997, p. 88), and more recent figures suggest even higher numbers. One assessment calculates that there may be as many as 100 million Chinese bachelors by the year 2020 (SCMP, August 25, 2003, Internet edition). Such skewed ratios will have significant consequences for family structures and also for the capacity of families to take care of the elderly, perhaps requiring the state or civil society organizations to take on greater responsibilities. The consequences of these transitions will be dealt with in detail in Chapters Four, Six, and Seven that look at medical care, social insurance, and social assistance. The primary concern for the leadership as it began to think about reform of welfare provision was to ensure that the urban industrial working class and government employees would be cushioned against the shocks of transition, especially as SOE reform intensified in the mid-1990s. As we shall see in subsequent chapters, this has entailed a

public goods during the reform period / 55

significant shift from workplace provision to local government provision and individual responsibility. For the privileged urban dwellers, experiments with new kinds of social support began in the early to mid-1990s and became codified by the end of the century. Thus, healthcare reforms were initiated in 1994 and were drafted into a new comprehensive scheme to be put into operation by the end of 1999; a new pension system for enterprise workers was drawn together in 1997; and a scheme for minimum subsistence relief was established at the end of 1999 based on experiments that began in Shanghai in 1993. Although there was consistent mention of rural needs—for example, in 1994 the “five guarantees” policy for rural households in need was reiterated, as was the need to restore the rural cooperative medical system—little was achieved in practice. The extension of minimal welfare benefits beyond the industrial working class had to await Hu Jintao and Wen Jiabao’s ascent to power in 2002–2003. The need to provide better support to those who have not fared so well from reforms is encapsulated in the slogan put forward in 2004 by Hu to “construct a socialist harmonious society.” Since taking power, Hu and Wen have sought to extend some benefits to excluded groups such as migrants and have tried to reduce the costs for basic services for the rural population. Thus, since 2003, there has been a major push to extend coverage of rural cooperative medical insurance; since 2002, experimentation with rural pensions was stepped up; and in 2007 the leadership pledged to extend minimum living support payments across rural China. A number of measures were introduced in 2002–2003 to improve the lot of migrant workers, who, in theory, had been brought into pension schemes in 1999. In June 2004, it was announced that migrant workers should be able to take out industrial injury insurance. To reduce the burden on the farmers, Premier Wen first announced a cap on rural fees and levies at 5 percent, converting them into a tax and later abolishing the agricultural tax. The abolition of the agricultural tax was part of an active policy to invest more heavily in rural physical and social infrastructure. To relieve the burdens on townships and villages, greater responsibility for funding health and education services was moved up to the county level. In March 2005, Premier Wen announced that education fees would be waived for poor rural households in the 592 officially designated poor counties. By 2007, it was hoped that no rural students would have to pay the many miscellaneous fees levied by local schools. Also, the Ministry of Education announced that it would spend 50 billion yuan to help students from poor families with support for college and vocational schooling (China Daily, July 3, 2007).

56 / providing public goods in transitional china

The central and local governments are committed to contributing to the rural health insurance scheme, and in 2004 a medical aid scheme was launched to cover major bills targeting poor households in west China. New investment in rural infrastructure is to be directed to projects that will improve the quality of life—the focus is first on safe drinking water, livestock, and clean energy supplies such as marsh gas and hydroelectric power, and on ensuring that all townships are accessible by asphalt or cement roads. The shifts in welfare provision have also impacted on urban administration. The two most significant changes affecting urban administration have been the shedding of social welfare and a number of other obligations by the workplace and the influx of large numbers of migrants into the cities in search of work. Those employed by the SOEs, and particularly those laid-off in the process of restructuring, have lost many of their former benefits and receive inadequate coverage from the government. The migrants never received any benefits and are not integrated effectively into those urban services that do exist. Given that most services were provided through the workplace, government infrastructure was relatively weak at the local levels, especially below the district level. The street offices and the residents’ committees were not equipped to deal with major welfare support and services. As a result, there have been experiments to create new organizational forms at the grassroots level to provide better social support. The most important is the program of community construction (shequ jianshe) that the Ministry of Civil Affairs (MOCA) has been promoting since the mid-1990s (figure 3.1). (This account draws on Bray: 2005; Derleth and Koldyk: 2004.) To develop more universal and comprehensive welfare benefits to replace the fragmented workplace-based system, the MOCA experimented with a number of community-based models.5 The MOCA first put forward these ideas in November 2000 and was supported by a joint document of the Office of the CCP Central Committee and the Office of the State Council that called on all government and party committees to set up these new structures. This call was a clear acknowledgment that the old administrative system could not meet the new demands. The shequ are based on the residents’ committees under the street offices, but they are larger in area and have a wider scope of obligations. They are asked explicitly to take over the social welfare tasks that previously had been the domain of the workplace or the residents’ committee and the street office. The MOCA clearly favored the model developed in Shenyang, home to many SOEs and thus hard hit by the reforms of the mid- to late 1990s.6 The residents’ committees were judged to be too

Provinces (27) and Major Municipalities (4)

Cities at Prefecture Level (283)

Counties (2862) Cities (374)

Towns*

Residents’ Committees***

Urban Districts (852)

Townships*

Street Offices#

Towns*

Townships*

Villagers’ Committees**

Community Committees/Resident Committees***

Residents’ Committees***

Villagers’ Committees**

Figure 3.1 Organization of Urban Government Notes: * There were a total of 35,472 towns and townships at the end of 2005, comprising 19,522 towns, a drop of 361 from 2004, and 15,951 townships, a drop of 1,500 from 2004. In 2004, 956 towns and townships were merged or cancelled. ** There were 629,079 villagers’ committees at the end of 2005, an increase of 3,932 from 2004. *** There were 79,947 urban residents’ committees at the end of 2005, a drop of 70 from 2004. # There were 6,152 street offices in 2005, an increase of 248 over 2004. Sources: NSB: 2007, at http://www/statsgov.cn/tjs/ndsj/2005/indexeh.htm and Administrative Division, China Facts and Figures 2005, at http://china. org.cn/english/en-52 2005/index.

58 / providing public goods in transitional china

small to operate effectively, while the street offices were too large to function as an effective grassroots organization. The MOCA referred to the new organizations as community residents’ committees (shequ jumin weiyuanhui). As Bray notes (2005, p. 185), the party acted to replace one form of collectivity with another. Rather than allowing people to interact individually with government agencies and the market, these new organizations were to take over the collective aspects of work and service provision that formerly had been provided by the workplace. Thus, they are responsible to ensure the state’s guarantees to provide minimum support to those in need; take care of vulnerable populations such as the elderly, unemployed, and children; manage urban sanitation and healthcare; enforce party policy such as family planning and maintaining social stability; and help with public security work. The committees fall under the authority of the street offices. Although the shequ can raise some funds from the services they provide, for most activities they are dependent on budget appropriations from the street offices. However, the development of these new organizations has not taken off as planned, in part because the agenda is too ambitious for the available staff. Most shequ have only between three and six full-time workers who are not particularly well paid and whose work does not have high prestige. In addition, the organizations suffer from a common problem that the higher levels of government expect them to undertake many functions but do not provide an adequate budget. In Qingdao, 160 functions of the street office were divided among the shequ, government-funded service centers, and charitable organizations (Derleth and Koldyk: 2004, p. 763). Last but not least, there is general bureaucratic inertia as China already has a complex set of organizations in the urban administration and it is difficult to restructure them to integrate effectively with these shequ. Thus, Derleth and Koldyk (2004, p. 772) conclude that “active” shequ are the exception rather than the rule. The Financing of Public Goods and Services Changes in the fiscal system have had a major impact on the provision of public goods and services in general, and on welfare provision in particular. Most public goods and services are provided by subnational levels of government and consequently they have expenditure responsibilities that are quite out of line with international practices (World Bank: 2002, p. ii). The central authorities in China have transferred a much greater percentage of expenditure responsibilities to local government than is

public goods during the reform period / 59

normally the case. The Maoist notion of “self-reliance” reinforced the idea that each locality should minimize “dependence” on support from higher levels and this has continued in the reform era. In fact, with the introduction of fiscal contracts in 1988, the central government formally ended its responsibility for financing local expenditures and expanded the role of local government from only providing services to also financing services. This delinked expenditure assignments from revenue-sharing considerations, later regularized in the Budget Law (World Bank: 2002). Unlike in many other countries, these transfers do not play an equalizing role, as richer areas often receive proportionately more (Mountfield and Wong: 2005, p. 97). This division of fiscal responsibility, with a far more decentralized approach, is quite different from that in other countries in the region (White and Smoke: 2005, p. 10). While in China the functional allocation for education, health, and social welfare was a local responsibility, as in Indonesia, for Thailand the responsibility is shared between the Center and the province, and in Vietnam between the province and the locality. As noted above, subnational expenditures as a share of total public spending are very high, at around 70 percent. This is higher than that of other countries within the region; for example, in Vietnam it is 48 percent, Indonesia 32 percent, and Thailand 10 percent (Mountfield and Wong: 2005, p. 86). Two factors mentioned earlier have combined with this de facto decentralization to redistribute incentives within the bureaucracy. The first is the diversification of the economy and the abandonment of collective structures in the countryside that initially affected the state’s capacity to raise adequate revenue. Second, the tightening financial constraints on SOEs have led them to shed their former welfare roles and to shift these roles to the underfinanced local governments. Before the economic reforms began, it was relatively easy for the state to extract revenue from the communes in the countryside and from the SOEs in the cities. Most agricultural and industrial products were sold at state-fixed prices, and the banking sector simply acted as the state’s cashier for its development priorities. Twenty years later, the situation has changed dramatically, with major consequences for the role of the state and its fiscal capacity. Market reliance has increased dramatically as has the nonstate sector of the economy. The central government has found it much more difficult to establish a sound tax system to raise revenue from the fast-growing nonstate sector. Despite the reforms, the SOE sector is still the largest source of government revenues. The importance of stateowned or collective industry as a revenue base for all levels of government has encouraged investment in enterprise development and a bias against investment in social development.

60 / providing public goods in transitional china

In the first phase of the reforms from 1980 to 1993, the Chinese government operated a fiscal contracting financial system. This was institutionalized in 1988. Under this system, local governments made a lump-sum remittance to the Center and the provinces were responsible for meeting expenditure requirements from the revenues that they retained. The objective was to give more power to local governments to finance their own priorities, subject to certain budget constraints. The system gave local governments a powerful incentive to encourage economic expansion but limited the Center’s capacity to benefit sufficiently from this expansion (Yang: 1994, p. 74). A province-collecting, Center-spending fiscal regime was transformed into an essentially self-financing regime for both the Center and the provinces (Zhang: 1999, p. 121). Revenue-sharing was delinked from revenue needs, thus placing local governments on a self-financing basis for the first time, something that was subsequently codified in the 1994 Budget Law (Wong and Bird: 2008, p. 432). As a result, the central government’s share of fiscal revenues declined from 34.5 percent in 1985 to 22 percent in 1992 (Su and Zhao: 2006, p. 4) and as a percentage of GDP, central government revenues dropped dramatically during the reforms from a starting-point of 36 percent to a low of 11 percent before beginning a slow revival to 17.5 percent in 2005 (calculated from statistics in NSB: 2007). This severely restricted the central government’s redistributive capacity and meant that local governments were left with more funds than in the past for development priorities. The resultant system was highly inequitable as provinces retained increasing amounts of funds, enabling the wealthier provinces to fund more programs. By 1993, the subnational share of total fiscal revenue reached 78.9 percent of the total (Su and Zhao: 2006, p. 13). By contrast, expenditures by the central government had dropped from 47.4 percent to 28 percent in 1993 (table 3.3). Thus, rather than a decline in state capacity under reforms, as some have suggested, there was a realignment Table 3.3 Ratio of Central-Local Revenues and Expenditures Year

Government Revenue

Government Expenditure

Central

Local

Central

Local

1978 1993 1994 2000 2005

15.5 22.0 55.7 52.2 52.3

84.5 78.0 44.3 47.8 47.7

47.4 28.3 30.3 34.7 25.9

52.6 71.7 69.7 65.3 74.1

Source: NSB: indexeh.htm.

2007,

at

http://www.stats.gov.cn/tjsj/indsj/2006/

public goods during the reform period / 61

between the Center and the localities, with the localities controlling far greater amounts of revenue than they had previously. To redress this situation and to provide the Center with greater fiscal capacity, reforms were introduced in 1993–1994 that were intended to raise the budget to GDP ratio and the ratio of centrally collected revenue to total budget revenue (Zhang: 1999, p. 131). The Center hoped to raise its share of state revenue to at least 60 percent, with 40 percent as central expenditure and 20 percent that could be used as grants to local governments. This required clarifying the division of spending responsibility, how taxes would be divided between the Center and the localities, and how to divide the administration for the collection of local taxes. This restructuring was successful, with the share of centrally collected revenue rising from 22 percent in 1993 to 55.7 percent in 1994, and it has stayed at roughly this level since reaching 52.3 percent in 2005 (table 3.3). Although the Center’s take still falls short of the 60 percent, the improved ratio, together with the general rise in revenues, has provided the Center with an enhanced capacity to support those policies it deems important. Thus Premier Wen Jiabao has been able to announce more support in the countryside for free elementary schooling, collective healthcare schemes, and greater infrastructure investment. However, it is important to note that although the balance of revenues has shifted in favor of the Center, the expenditures by local governments have hardly changed since the 1993–1994 fiscal reforms. In 1978, 47.4 percent of government expenditure was by the central government and 52.6 percent at the subnational level, but by 1993 this had shifted to 28.3 percent and 71.7 percent respectively. Subsequently, local government expenditures continued to account for around 70 percent of total spending, amounting to 74.1 percent in 2005. This has meant that local governments are more reliant on central government transfers and raising their own revenues to meet their obligations. Subnational levels of government provide 94 percent of spending on education, 98 percent on health, and 99 percent on pensions and relief funds for social welfare (table 3.4). Local government finance has become even more difficult since the government converted all fees into the single agricultural tax in 2002, followed by the special products tax and slaughter tax. The financial pressure has meant that economically developed and more resource-constrained localities share an emerging general imperative to derive from one’s own sources of revenue to cover centrally mandated obligations. The resultant fiscal inequalities that arise from this system are a major cause of the significant variation in the provision of public goods and services. Financial pressures lead to preference for a development plan that maximizes short-term revenue extraction over longer-term needs and

62 / providing public goods in transitional china Table 3.4 2005

Central and Local Government Expenditure on Social Welfare,

Spending Category

Education Health Pension and Relief Funds Social Security Subsidiary Expenses Support for Underdeveloped Areas

Total (100 m. yuan)

Central (100 m. yuan)

% of Total

Local (100 m. yuan)

% of Total

3974.83 1036.81 716.39 1817.64

244.85 21.26 5.34 236.71

6.16 2.05 0.75 13.02

3729.98 1015.55 711.05 1580.93

93.84 97.95 99.25 86.98

195.42

6.66

3.41

188.76

96.59

Source: NSB: 2007, at http://www.stats.gov.cn/tjsj/indsj/2006/indexeh.htm.

is disinclined to distributional and welfare priorities. The main concern of government at all levels is to increase revenues rather than think about the correct role of the government. The concern over revenue generation is exacerbated by the fact that despite fiscal decentralization, the central government has retained control over the policy agenda. The Center sets many tasks that must be implemented by local governments, and most of these are unfunded mandates. Cities at the prefecture and county levels should cover all expenditures for unemployment insurance, social security, and welfare, whereas in most other countries the central government will cover social security and welfare, with education and health shared between the localities and the Center. The county is being developed as a fiscal center in rural China and it is beginning to eclipse the township. This expansion of responsibility has been conducted under the slogan of “taking the county as the lead” (yixian weizhu); spending at the county level now accounts for about one-quarter of all public expenditures. Most important was the decision in 2001 to shift the fiscal responsibility for teachers’ salaries to the county from the township (People’s Daily, June 14, 2001). The expenditure responsibilities for the townships are similar to those of the counties, although they often have a weaker financial base and they carry the heaviest load for social spending. The county and the township together account for 70 percent of budgetary expenditures for education and 55 to 60 percent for health (World Bank: 2002, pp. 34, 94, 111). In Xiangyang county, Hubei, budgetary contributions in 2002 to finance education amounted to 40.6 percent of total expenditures. Of the governmental contribution, the township provided 84.6 percent and the county provided 15.2 percent, the remainder coming from the province (0.2 percent) (Han: 2003, p. 12). Prior to the 2001 reform, education constituted

public goods during the reform period / 63

the largest expenditure for township budgets—approximately two-thirds of township expenditures nationwide. A study in 2002, just as the new reform was to be implemented, discovered that 78 percent of compulsory education costs were covered by the town/township and village, with a further 9 percent covered by the county, and 11 and 2 percent by the provincial and central governments respectively (Kennedy: 2007, p. 49). Despite the lightening of the burden, the township remains responsible for subsidies for teachers and also for the maintenance and upkeep of the school premises. Villages, which are not a formal level of government, have significant expenditure responsibilities even though they have no independent fiscal powers. They have inherited many of the obligations of the old collective economy, such as salaries, care for the elderly, and even support for health and education (Wong: 1997, p. 174). These responsibilities drive the village and township leaderships to seek various off-budget revenues from user fees and other unsanctioned levies to support their activities. For example, in three counties surveyed by the Development Research Center (DRC) of the State Council (Han: 2003), expenditures exceeded revenue, thus increasing the need to raise larger off-budget revenue. Nationwide, extrabudgetary funds may total 20 percent of GDP; whereas in the three counties surveyed by the DRC, they ranged from 30 percent of total income (in Xianyang, Hebei province) to 69 percent (in Taihe, Jiangxi province). The use of these extra-budgetary funds and self-raised funds (zichou zijin) clearly increased as a result of the 1994 tax reforms. For example, the valueadded tax is split in a 3:1 ratio between the Center and the localities. Prior to the reform, Guizhou province had derived fully 45 percent of its revenue from liquor and tobacco. The Center now takes much of this revenue. Despite the rapidly rising social welfare demands, the two major causes of growth in government expenditures are capital spending and administrative outlays. The OECD has calculated that in the 1998–2003 period, capital expenditures contributed 31 percent of the total growth in expenditure and administrative outlays 21.1 percent (OECD: 2006, p. 23). The latter burden falls most heavily on local governments that provide 82.1 percent of administrative spending; they provide only 55.6 percent of capital investment (OECD: 2006, p. 47). Many local administrations have great difficulty in meeting their salary bills. The budgets of local governments are often referred to as “eating budgets” (chifan caizhang) as they cover only basic essentials and salaries for bloated staff numbers (Shih et al.: n.d.). Even in major cities local governments have difficulties in meeting their administrative expenses. Again we see enormous variation across China. Large municipalities such as Beijing and Shanghai are almost self-sufficient in terms

64 / providing public goods in transitional china

of their capacity to fund their activities: Beijing covers 88.6 percent of its needs and Shanghai 89 percent. By contrast, poorer provinces such as Qinghai and Guizhou can cover only half or less of their obligations, at 39.3 percent and 53.7 percent respectively, and Tibet covers only 12.9 percent (Su and Zhao: 2006, p. 15). In Guizhou, many counties were unable to meet payroll obligations (Wong and Bird: 2008, p. 455). The tax reforms of 1993–1994 allowed the richer provinces to retain more revenue, thus exacerbating the inequalities. Wong and Bird (2008, p. 437) have calculated that from 1992 to 1998 the ratio of provincial per capita fiscal expenditures in Shanghai grew from 2.5 to 3.9 times the national average, in Beijing it grew from 2.2 to 2.7, and in Guangdong from 1.1 to 1.4. By contrast, in Henan it dropped from 0.44 to 0.42, in Gansu from 0.8 to 0.6, and in Hunan from 0.53 to 0.49. Not surprisingly then, per capita fiscal capacity shows great variation. In 2005, Shanghai’s revenue per capita was 7,972 yuan, over 16 times that of Guizhou at 489 yuan. Shanghai’s per capita expenditure was 9,259 yuan, as against 1,396 yuan in Guizhou, a 6.6 differential (NSB: 2007 at http://www.stats.gov.cn/english). These figures reveal Guizhou’s reliance on transfers. Dollar (2007, p. 11) estimates an even higher differential (of eight times) of per capita public spending between the richest and the poorest province. By contrast, in the United States the poorest state has about 65 percent of the revenue per capita of the average state, and in Brazil the richest state has 2.3 times the revenue per capita of the poorest state. The problems are exacerbated at the subprovincial level by provincial retention of a high level of the revenues generated within the province. For example, the per capita fiscal revenue of the Yunnan provincial government is 14 times that of the county-level governments, and in Guangdong it is seven times. Dollar (2007, p. 14) estimates that the richest county’s per capita spending is 48 times that of the poorest. Together, the provinces and municipalities account for 70 percent of all subnational fiscal revenue, whereas the counties and townships, where the need is great, account for only 30 percent (Su and Zhao: 2006, p. 22). This means that the fiscal system is highly regressive and local governments in poorer areas often have no alternative other than to eliminate services or to extract high fees from the residents. But even high taxes may not enable local governments to meet their obligations, a situation that has been described as “predatory fiscal federalism” (Shih et al.: n.d.). This situation has led to a search by local governments for stable revenue sources. In the 1980s and 1990s, the financial pressures and the removal of agriculture as a viable financing source for local government contributed to the expansion of locally owned enterprises, especially

public goods during the reform period / 65

township and village enterprises (TVEs), as these were seen as the most stable source for local income. This was the case irrespective of whether the locality concerned was relatively wealthy or poor. As Albert Park and his colleagues have noted, “heightened pressure on revenue-starved local governments may have led to overinvestment in revenue-generating industrial enterprises, encouraged bureaucratic predation of enterprise resources and regional protectionism, and diverted attention away from long-term development strategies” (Park et al.: 1996, pp. 751–752).7 By the mid-1990s, the TVEs had become a burden for many local governments and a large-scale movement of privatization began. The need for funding led most local governments to collect a wide range of sanctioned and unsanctioned fees and levies. Once this practice came under increased scrutiny, local governments had to use the land under their jurisdiction to raise more revenue, often converting agricultural land to higher-priced commercial or residential land use. The advantage of these off-budget revenues is that they do not fall under the 1993–1994 tax-sharing agreement with the higher levels of government. By their very nature, these funds are hard to calculate, but Christine Wong (2000) has estimated that extra-budgetary funds (EBFs) amounted to 8 to 10 percent of the GDP, compared to an official budget of 14 percent. In a later study, Wong and Bird (2008, p. 447) calculated that the EBF and extra-budgetary activities of government amounted to between 19 and 29 percent of total public sector spending. This confirms the view that public sector spending has not actually declined under the reforms. Rather, what has declined is the amount controlled by the Ministry of Finance. Beginning in the late 1990s, the central government tried to rein in these fees and levies, some of which were illegal and many of which caused farmer discontent. The first step was to convert the various fees and levies into taxes and to reduce the level to a maximum of 5 percent of farmers’ income. Thus, in Beijing municipality, the percentage of “other revenue” in the budget declined from 42.4 percent in 1998 to 32.3 percent in 2000, whereas “on-budget revenue” rose from 39.4 percent to 47.4 percent. The formal “off-budget” revenues remained relatively constant at 18.2 percent and 20.3 percent respectively (Su and Zhao: 2006, p. 25). In our survey, during the period from 2003 to 2005 taxes as a percentage of villagers’ income dropped from 6.5 percent to 5.1 percent. In 2003, citizens ranked tax management by local government as one of the five services with which they were most dissatisfied. Interestingly, in the 2005 survey this level of dissatisfaction remained for those living in major cities, towns, and townships; villagers, on the other hand, did not rank it. This may attest to the success of Premier Wen’s policies to reduce the tax burden.

66 / providing public goods in transitional china

However, it is possible that the reduction and then the abolition of the consolidated agricultural tax may have had a negative impact on the provision of welfare services. Local governments increasingly relied on such revenues to provide services. In poor and remote communities where marketization has barely begun and where the scope of economic activity will always remain limited, local treasuries have little recourse other than the elimination of services. In many poorer parts of China, rural medical health schemes were abolished and access to schooling drastically reduced. For those services that they could provide, poorer regions often relied on the EBFs. In the mid-1990s, Guizhou province derived fully 80 percent of its educational funding from such sources.8 Park and his colleagues calculated that in Shaanxi, 86 percent of provincial consolidated revenues and 89 percent of expenditures were from extra-budgetary funds (Park et al.: 1996, p. 767). Whereas villagers’ committees in poorer areas might be more concerned with how to raise revenues to cover basic welfare requirements, richer villages preside over an extensive income from local enterprises and make decisions regarding village investment in road building, hospital development, and in creating other such amenities. Preliminary evidence suggests that the abolition of the agricultural tax, unless compensated for by adequate transfers, may indeed have a negative impact on service provision if there are no major cost savings. Research by Kennedy (2007) in Shaanxi reveals that there was a sharp decline in the provision of educational and medical services, with poorer townships heavily dependent on the county government for revenue and acting as no more than an “administrative shells,” unable to offer basic services. He shows that counties and townships have suffered only a small drop in the number of hospitals and medical personnel, whereas villages have suffered a precipitous decline. Kennedy (2007, pp. 53–54) shows that between 2001 and 2003, the number of doctors and healthcare workers in Shaanxi villages declined by 23 and 82 percent respectively. The number of village clinics dropped by 25 percent. Christine Wong has also pointed out that the abolition of the miscellaneous education fees and their replacement by transfers to cover elementary education costs may actually end up causing a shortfall in income for local governments. The subsidy from the central government covers only a portion of the revenue that local governments used to derive from various education-related fees and levies. The subsidy is 140 yuan for elementary school students and 180 yuan for junior middle school students. However, before the new program began, the costs were much higher; in one of the schools she visited, the cost was 1000 yuan per student (Wong: 2007, pp. 14, 17).

public goods during the reform period / 67

The other major impact of fiscal reform is the continued decline of the township. Although some reformers have called for the abolition of the township as a formal level of government, the preferred policy seems to be to promote a withering away of the township through mergers and conversion into towns (zhen). In 2004, the number of township jurisdictions was 43,258, down from 72,153 in 1985 and 47,136 in 1995. Of the jurisdictions in 2004, almost half were towns (19,883). In Guangdong, towns already form the majority, at 1,145 out of 1,585 township-level jurisdictions. In poorer provinces such as Guizhou and Xinjiang, townships still predominate (691 towns out of 1,543 and 229 out of 1,009 respectively; http://www.stats.gov.cn/english/). The merger of townships has led to the elimination of accessible services for many since the schools and clinics are closed in those townships that lose their appellation. In areas where distances are large such as Tibet, this means that the former township community may entirely lose its previous access to education and healthcare (discussion with Arthur Holcombe, president of Tibet Poverty Alleviation Fund, April 2007). The Incentives for Local Governments The need to focus on short-term revenue generation led local governments during the first two decades of reform to place a lower priority on social development, something reinforced by the political contract system that prioritizes investment and economic growth. Although there has been greater autonomy for local officials, those at the township level and above are dependent on the approval of officials one level above for their career advancement. The work of a number of researchers has argued persuasively that we need to complement a pure political economy approach that views state agents as revenue-maximizers with an understanding of the political incentives generated by the cadre responsibility system, the political contracting system, and the performance contracts (gangwei mubiao zerenshu) that govern the work of local governments and officials (Rong et al.: 1998; Edin: 2000; Whiting: 2001). Performance contracts do not discourage economic development, far from it, but they make it clear to local officials that economic development is just one element in a complex set of tasks that local officials are required to carry out, including maintaining social order, delivering taxes to higher levels of government, and maintaining family-planning quotas. Multiple principal-agent relationships operate between the different levels of government. These multiple vertical relationships complicate comprehensive planning and there are frequent conflicts between vertical and horizontal lines of command. The township has little or

68 / providing public goods in transitional china

no power to decide on personnel arrangements, financial revenue and expenditures, and the allocation of materials to the county branches in the township. Contract targets are divided into a mixture of priority, hard and soft targets. Priority targets are set nationwide and usually are more political or policy-oriented in nature, including, for example, the maintenance of social order and meeting family-planning quotas. Hard targets primarily concern economic outputs set by the county for the township, including meeting tax revenues and attaining or exceeding predetermined growth levels. Soft targets tend to cover social development such as health and education provision and environmental protection. Clearly meeting hard and priority targets are critical, as failure will limit promotion possibilities and result in loss of bonuses. The priorities for task fulfillment lead to the neglect of social development in favor of meeting economic growth and investment targets. Performance assessment varies somewhat from place to place, but Zhao Shukai has outlined a fairly standard contract for village officials. Evaluation is based on a 1000-point scale for which the vast majority of points (400–500) are allocated for success in collecting revenues, with a further 200 points for maintaining social stability, and 100 points for meeting family-planning targets. This means that very little attention (200 points) is paid to social policy targets or public infrastructure projects (Zhao: 2004, p. 220). Not surprisingly, the contracts can produce a number of perverse outcomes and help explain why officials often pursue unpopular policies with such zeal. One survey of 89 villages in Fuquan county (Guizhou province) found that village leaders spent 80 percent of their energy and 70 percent of their time on “the most disliked” administrative affairs (Li Xueju: 1994). However, the system does have the potential to refocus incentives should the leadership so desire. Under Premier Wen Jiabao there have been attempts to shift the work focus of local officials to pay more attention to social development. A new nationwide performance appraisal system is being developed based on trials in Qingdao, Shandong province. The experimental system tries consciously to shift evaluation away from only measuring whether local officials are satisfying the demands of higher authorities to looking at whether public service goals are being attained (People’s Daily Online, “33 Indexes Evaluate Government Performance,” August 2, 2004). The financial system and the incentives for local officials have serious consequences for the nature of local government in terms of service provision. Not only is there significant inequality of provision, but also the level of autonomy and the resource base dictate varying

public goods during the reform period / 69

relationships between the local government and the society. The variety of local outcomes is often “path-dependent” on the economic structure that existed on the eve of reform, and the reforms have often reinforced existing cleavages (see Saich: 2006b). In areas that lack adequate resources or that depend on a single product for income, the local authority is often predatory on the local population (see Guo: 1999). However, when the locality is so poor that no resources can be squeezed out of it, the government may simply be paralyzed and offer no services whatsoever (see O’Brien: 1994; Wang: 1999). By contrast, wealthier areas may be able to offer a relatively good social infrastructure for their citizens, such as in southern Jiangsu or Shandong where the strong TVE system allowed the local state to reinvest in the local community under what Oi (1992) has termed “local state corporatism.” The differing relationships all have consequences for how welfare is provided and at what level. Concluding Comments The economic reforms have wrought major changes in the way social welfare is provided. After an initial period of neglect, from the early to mid-1990s extensive experimentation took place to put in place new systems of welfare delivery. In urban China, the new systems have transferred provision of most benefits from the workplace to the local government. The experimentation has been accelerated since Hu Jintao and Wen Jiabao took over the CCP and state leadership in 2002–2003 as they have tried to extend coverage to previously excluded groups and vulnerable populations. In this attempt they have been aided by the more buoyant state revenues of recent years. In 2007, central government expenditure in health rose 86.8 percent, primarily because of the expansion of the rural cooperative health scheme, and education expenditure rose 39.5 percent with support for compulsory education (Xinhua News Agency, December 17, 2007). However, what you get still depends on where you live and work. Despite the attempts by the Hu-Wen leadership to moderate the inequalities in service provision, the inequalities are unlikely to be remedied in the short to medium term. There are also new challenges that need to be dealt with, such as caring for the aging population and constructing effective support structures for the rapidly swelling urban population. Spending on social welfare remains low and it is clear that the government cannot be the sole provider but must develop better regulatory frameworks for market provision and more effective partnerships with civil society organizations. To meet the government’s

70 / providing public goods in transitional china

objective of “putting people first” and developing a “service-oriented government” ( fuwuxing zhengfu), more accountable practices and institutions need to be developed. We examine these challenges in more detail through a study of healthcare, dealing with pandemics, social insurance, and social assistance.

Ch a p t e r Fou r Ch a ll e ng e s i n t h e H e a lt h Syst e m

The development of the health sector in China encapsulates many of the issues raised in earlier chapters and reveals an evolving complex interaction between states, markets, and civil society. After the CCP took power in 1949, the subsequent restoration of order and establishment of collective institutions with strong public health functions led to tremendous improvement in China’s indicators for healthcare provision and control of communicable diseases. This chapter looks at the former, whereas Chapter Five looks at disease control. Public spending on healthcare is low, preventive health spending is lower than that for curative care, and only a minority has access to effective medical insurance. There is a significant variation in access to and quality of care between the rich and the poor and between urban and rural dwellers, variations that have grown significantly with the reforms. The reform period has also presented the leadership with new challenges. First, the fiscal decentralization, which we discussed in Chapter Three, has impacted on the availability and quality of services. There has been a dramatic shift in the way that healthcare is financed. Second, the increased marketization of the rural and urban economy has meant that the costs of healthcare that were formerly carried by the collectives in the countryside or by the state-owned enterprises and government offices in the cities have now devolved to individuals, many of whom do not have insurance coverage. Third, the attempts to regulate prices created unintended effects that motivated healthcare providers to provide expensive drugs and to encourage invasive, high-tech procedures. This chapter first examines the general performance of China’s health system; second, we look at the challenge of healthcare insurance provision in urban China, followed by that in rural China. Performance of China’s Health System The healthcare system as it operated before the reforms began delivered very tangible benefits to the population and even subsequently;

72 / providing public goods in transitional china

despite the many criticisms, general health indicators have continued to improve. However, progress with key indicators has slowed, there is considerable variation depending upon the observed population; and in poor rural areas, some indicators have even declined. One report by the DRC of the State Council even noted that, on the whole, medical reform over the last twenty years had not been successful (2005).1 The criticism and the slowing indicators have led the Chinese leadership to increase funding significantly to improve healthcare and access. Minister of Health Chen Zhu was reported in January 2008 to have said that the government would remove all inequalities in basic healthcare by 2020 and provide the same access for all citizens. He stated that “universal entitlement means fair entitlement. All citizens, regardless of age, gender, occupation, location and income level, should enjoy the same rights” (SCMP, January 8, 2008).2 Over 2006–2007, the government health budget increased by almost 90 percent and it has committed to further increases of between 1 and 1.5 percent of GDP. In 2005, government spending was 0.84 percent of GDP. A major challenge now exists to make sure that these extra funds are invested wisely. After 1949—despite the fact that there was a shortage of doctors, a heavy urban bias in personnel distribution, and a shortage of resources— the health system functioned respectably and made considerable progress in key indicators. The low cost or free medical services combined with the promotion of major public health objectives by government and collective structures were the key reasons for this progress. In rural China, healthcare was organized through the collective structures of the communes, brigades, and teams. At the grassroots level, “barefoot doctors” (paramedics) provided basic services and were able to refer patients up through the brigade and commune clinics and hospitals. These “barefoot doctors” were chosen from within the local communities and given about three months’ training. As other rural laborers, they were allocated daily work-points and thus had no particular incentive to leave their work even if they could. By the mid-1970s, a basic rural medical cooperative insurance scheme, a program that had been popularized from 1968 on, was in place that covered much of the rural population. The insurance scheme was developed based on experiments conducted around Shanghai in 1958. With the production brigade as the primary accounting unit, the scheme was funded by small individual contributions, was assessed on a yearly basis, and took into account usage of funds. Funding also came from collective welfare funds and subsidies from higher levels. The costs of treatment, medicines, and outpatient services were kept very low by the central government. The impact of this on a country that had suffered decades of war and dislocation should not be underestimated. Although the system was biased in

challenges in the health system / 73

favor of the urban areas, for the first time the majority of China’s rural dwellers had access to basic healthcare. In urban China, an essentially Soviet-style medical system was put into operation, but with one important exception. Instead of the service being provided by the local government, workers received their medical care through the enterprise. Employees in government organizations, including universities, were funded through the state budget for medical costs, whereas those in state-owned enterprises were funded by enterprise welfare funds, thus setting up “mini-welfare states” (Gu: 2001). The government employee system was set up in 1951 and was fully operational by 1956. The enterprise welfare fund was set up in 1951 but initially operated as a social insurance system, with enterprises contributing to a labor insurance fund administered by the trade union organization. When the trade unions were abolished during the Cultural Revolution, the system ceased and free healthcare became part and parcel of the benefit package of working in a SOE (Gu and Zhang: 2006, p. 54; for a detailed history, see Dixon: 1981). However, the enterprise-based system did not provide a hard budget constraint as ultimately the expenditures and revenues all flowed in and out of the government coffers. “Moral hazard” was rife throughout the system, with people not only charging their families to the system but also purchasing household goods on a prescription from the doctor. For example, some would be allowed to purchase a refrigerator as they needed to keep their food cold to prevent sickness, and others would be allowed to purchase a television set to help them relax (discussion with vice minister of the DRC, July 2006). The new system did lend itself to the promotion of health education and the eradication of certain endemic diseases. Unlike in the United States or China during the reform period, the health system was not prey to capture by healthcare providers and highly trained professionals who seek to set prices and services in a way that is detrimental to the interests of the population at large. As in other areas of life in post-1949 China, campaigns were a major feature and many were mobilized through the “patriotic health movement” to improve both public awareness about common diseases and to eliminate pests (Horn: 1969; Sidel and Sidel: 1973). For example, these campaigns were very successful in reducing, if not eliminating entirely, the problem of schistosomiasis. A major national effort was launched in 1955 to change the habitat of the snail, limiting its ability to transmit the disease; this effort was backed up by educational endeavors to teach peasants about the life cycle of the disease and how to prevent transmission (Anson and Sun: 2005, pp. 59–62). It is estimated that an incredible 7.68 million person-days were invested in environmental change to eliminate the snail hosts between 1974 and 1980 (Anson and Sun: 2005, p. 61). However,

74 / providing public goods in transitional china

official claims need to be looked at with caution. When I was working on the Yangzhou People’s Commune (Jiangsu province) in the spring of 1977, officials prohibited us from swimming in the local water ostensibly for our own protection as they were not sure whether we were good swimmers. Later, it became clear that the real reason was that schistosomiasis was still endemic in the local waters, despite official claims that it had been eradicated. A foreign exchange student getting a disease that did not exist would clearly be an embarrassment for the local officials. In 2005, official statistics stated that there were still 448 endemic areas, with a population of 67.5 million and 689,000 patients (http://www.moh.gov.cn/open/2007/ tjts/P49.htm). As noted, Chinese health indicators continued to improve after reforms began, but with some significant variation. Average life expectancy increased from 35 before 1949 to 67.9 in 1981 and 72 in 2005. Infant mortality declined over the same period from 200 per 1000 before 1949 to 34.7 per 1000 in 1980 and to 19 per 1000 in 2005. China’s development has meant that by the 1980s it was undergoing the same epidemiological transition that had occurred in the developed economies, with infectious diseases giving way to chronic diseases (heart problems, cancer, etc.) as the leading causes of death (Blumenthal and Hsiao: 2005, p. 1166). The slowing of progress in the 1990s is clearly seen with respect to the healthcare indicators. According to the World Health Report 2000, China ranked 188 out of 191 countries in terms of fairness in financial contributions, 144 for the overall performance of the health system, and 139 in terms of healthcare expenditure per capita in international dollars. Although China ranked above most African countries, it ranked below other large developing countries such as India, Bangladesh, and Indonesia. In terms of health quality, China ranked better (61 out of 191), but this may have been because of the residual impact of the old collective medical system (WHO: 2001, p. 152). There is also evidence that with the collapse of the collective support structures, the reforms of the 1980s and 1990s also impacted on basic indicators. Infant mortality actually rose according to official figures, from 34.7 per 1000 in 1980 to 37 per 1000 in 1992. UNICEF has calculated that it was 52 per 1000 in 1998, with a rate four times higher in the poorer regions of China (UNDP: 1998). The rapid growth from the mid-1990s caused the subsequent drop. Government spending on welfare has been low in international terms, even for a middle-low income country. Although investment has grown strongly in absolute terms (annual budgetary increases of 14.2 percent), this has been below the growth in total government revenue (17.5 percent), indicating that social investment had not kept pace with overall increases in government

challenges in the health system / 75

revenue (UNDP: 2004, p. 4). The structure of spending has changed as well, with individuals and nonstate agencies contributing a higher proportion. Government spending on health is roughly comparable to that in other Asian middle-income countries. As a percentage of GDP, it was 2 percent in 2003–2004, the same as it was in 1990. This is better than most other Asian countries with the exception of the tiger economies and Malaysia (table 4.1). The health expenditure per capita is far higher than that in the two other most populous nations in Asia, at $276 compared with $113 in Indonesia and $82 in India. Despite the relatively comparable spending on health, China’s indicators for under-5 mortality are worse than those for countries such as Malaysia and Thailand. The rate for China in 2004 was 31 per 1000 live births, whereas for Malaysia it was 10 and for Thailand 18. In 2005, the under-5 mortality rate in China dropped to 22.5. The same is true for infant mortality rates. China’s indicators are good for a lower-middleincome country but are about average for a middle-income country. Public expenditure as a percentage of total health spending was lower than average for both lower-middle-income and middle-income countries. In China, a high percentage of health expenditure is private (3.6 percent of GDP) as opposed to public (2 percent); this ratio of private spending is only matched by India (3.6 percent) and Vietnam (3.9 percent). Two other features of China’s healthcare spending should be mentioned. First, spending programs have a strong urban bias and, second, personal spending on health has been rising sharply. China’s total health expenditure in 2003–2004 rose to 5.8 percent of GDP, but the growth was largely attributable to the strong increase in personal spending. Between 1991 and 2000, accumulated government spending on health in rural areas was only 15.9 percent of total government spending (69 billion yuan) and although total health spending increased by 50.7 billion yuan over this period, the increase for rural areas was only 6.3 billion, a 12.7 percent increase. Rural government spending fell to 6.59 percent from 12.54 percent of total spending (Han: 2004, p. 18). In 2005, per capita health expenditure for rural China was 318.5 yuan (up from 38.8 yuan in 1990), whereas that for urban China was 1132.8 yuan. Rural spending for healthcare was onehalf of the national average, but that for urban China was almost twice the national average (http://www.moh.gov.cn/open/2007/tjts/zp113.htm).3 Effectiveness is reduced as funding is focused on the county and township levels, even though 60 percent of outpatient services are provided by village clinics or the private sector. Only 25 percent of these services are provided by the township hospitals and only 14 percent at the county and the township level (World Bank: 2002, p. 125). The skewed nature of the financing structure is something that the Hu Jintao-Wen Jiabao leadership is trying to remedy by allocating more resources to rural healthcare.

Table 4.1

Comparative Health Indicators for Selected Countries

Country

China India Indonesia Malaysia Philippines South Korea Thailand Vietnam

Life Expectancy at Birth, 2004

All

Male

Female

71.9 63.6 67.2 73.4 70.7 77.3 70.3 70.8

70.2 62.1 65.3 71.1 68.6 73.7 66.7 68.8

73.7 65.3 69.2 75.8 72.8 80.9 74.0 72.9

Public Private Expenditure H ealth on Health, Expenditure, % of GDP, as % of GDP, 2003–04 2003–04

2.0 1.2 1.1 2.2 1.4 2.8 2.0 1.5

Notes: IMR—Infant Mortality Rate MMR—Maternal Mortality Rate PPP—Purchasing Power Parity Source: UNDP, at http://hdr.undp.org/hdr2006/statistics/.

3.6 3.6 2.0 1.6 1.8 2.8 1.3 3.9

Health IMR Per MMR Under-5 Births Physicians Expenditure, 1000 Live Adjusted Mortality Attended by Per Per Capita Births, Per 100,000 Rate Per Skilled Health 100,000, PPP $ 2004 2004 Live Births, 1000, Personnel %, 2004 2000 2004 1996–2004

278 82 113 374 174 1074 260 164

26 62 30 10 26 5 18 17

56 540 230 41 200 20 44 130

31 85 38 12 34 6 21 23

96 43 72 97 60 100 99 85

106 60 13 70 58 157 37 53

challenges in the health system / 77 Table 4.2

Structure of Health Expenditure, 1980–2005

Government Expenditure as % of GDP Government Expenditure as % of THE Social Expenditure as % of THE Personal Expenditure as % of THE THE as % of GDP Per Capita Health Expenditure Urban (yuan) Rural (yuan) Ratio of Urban to Rural Expenditure

1980

1990

1995

2000

2005

— 36.2

2.3 25.1

2.39 18.0

1.71 15.5

1.77 17.9

42.6 21.2 3.17

39.2 35.7 4.03

35.6 46.4 3.54

25.5 59.0 4.62

29.9 52.2 4.73

— — —

158.8 38.8 4.09

401.3 828.6 1122.8 112.9 209.4 318.5 3.54 3.95 3.53

Note: THE—Total Health Expenditure Source: Ministry of Health, at http://www.moh.gov.cn/open/2007tjts/P113.htm.

Government budgetary expenditure stood at 17.9 percent of total healthcare spending in 2005, down from 36.2 percent in 1990. This was a slight revival from a low of 15.5 percent in 2000 (table 4.2). However, the formal budgetary allocation alone underestimates government spending as a percentage of total healthcare expenditure. Chinese statistics include the category of “social spending” that contains a large component of state-owned enterprise contributions to health insurance and related health spending. This amounts to a further 25–30 percent, depending on one’s calculations. With the SOE reforms, this expenditure category also dropped from 42.6 percent in 1980 to 29.9 percent in 2005. This would give a combined “government” expenditure on health of 47.8 percent for 2005, down from 78.8 percent in 1980. Wang Longde (2006), the vice minister of health, calculated a lower rate of public expenditure of 33.7 percent within the total health expenditure. Both figures are much higher than government spending in India (17.8 percent) but are generally lower in international terms. The higher figure (47.8 percent) approximates the average for lowermiddle-income countries at 49.4 percent (World Bank: 2003, pp. 92–94). By comparison, in Western Europe government spending ranges from 80 to 90 percent. Government spending in China approximates that in the United States—45.6 percent in 2003. Through the 1990s, “social spending” dropped from 39.2 percent to 25.5 percent. Although personal spending increased from 35.7 percent to 59 percent of total healthcare expenditures, in 2005 personal spending declined to 52.2 percent. In the rural areas, we see a particularly dramatic shift as farmers assumed greater responsibility. In 1991, farmers’ input to healthcare spending was 80.73 percent, but this rose to 90.15 percent in

78 / providing public goods in transitional china Table 4.3

Medical Insurance Coverage, 1998 and 2003 (percentages) Nationwide

Basic Insurance Government Insurance Labor Insurance Cooperative Insurance Other Commercial Insurance Self-payment

2003

1998

8.9 1.2 1.3 8.8 1.4 7.6 70.3

— 4.9 6.2 5.6 5.0 1.9 76.4

Urban 2003 30.4 4.0 4.6 6.6 2.2 5.6 44.8

1998 — 16.0 22.9 2.7 10.9 3.3 44.1

Rural 2003

1998

1.5 0.2 0.1 9.5 1.2 8.3 79.0

— 1.2 0.5 6.6 3.0 1.4 87.3

Source: Ministry of Health, at http://www.moh.gov.cn/open/2007tjts/P25.htm

2000 (Han: 2004, p. 18). However, this seems to have been the peak. The 1998 National Health Service Survey estimates that 87.3 percent of the rural population had to pay their own medical costs, but in 2003 this dropped to 79 percent (table 4.3). This drop might show that the revived rural medical insurance schemes were having a limited effect. Also, the ratio of per capita health expenditure between urban and rural China has varied from about 4:1 to 3.5:1. The ratio declined during the first half of the 1990s only to increase again to 3.95:1 in 2000. However, the increased attention paid to rural health during the first half of the new decade brought the ratio down again to 3.53:1. Although greater responsibility for funding has shifted to individuals, health costs have been rising. The unit cost of an outpatient visit at a county hospital in 1998 was roughly four times higher than the cost in 1993.4 Official statistics show that costs for in-patient per capita medical expenses rose from 1,668 yuan in 1995 to 4,669 yuan in 2006. The cost of treatment in a comprehensive Ministry of Health hospital rose from 5,027 yuan to 12,434 yuan and costs at a county-level hospital rose from 881 yuan to 2,241 yuan. The percentage of medical expenses spent on drugs and medicines dropped from 55.4 percent to 39.5 percent in Ministry of Health hospitals and from 53.7 percent to 44.3 percent in county-level hospitals, but these expenditures were still higher than those for examination and treatment (http://www.moh.gov.cn/ open/2007/tjts/P18.htm). Not surprisingly, we see government support for hospitals declining as a percentage of revenue, as health institutions are required to cover more of their own basic expenses. In 2002, government funding covered only 10.2 percent of the costs of government-run hospitals, down from 15 percent in 2001 and about 30 percent between 1970 and 1980; whereas in 2002, 44.17 percent came from user fees

challenges in the health system / 79

and 43.04 percent came from drug sales. Also, government financing is biased toward the urban areas and most funding goes to hospitals. Of the total 2002 health expenditure, 67.7 percent went to hospitals (50.5 percent to urban hospitals), and only 7.3 percent to health centers (UNDP: 2005, p. 58). This is a problem because most of the need is at the lower levels of the system. Total healthcare spending increased 52-fold from 1980 to 2004, from 14.3 billion yuan to 759 billion yuan (866 billion yuan in 2005), whereas citizens’ spending rose 133-fold. Over this same period, urban disposable income rose 18-fold and rural disposable income 14.3-fold (Wang: 2006). Given the above statistics, it is not surprising that the percentage of healthcare consumption in the annual household budget has been increasing. For urban families, it rose from 2.5 percent to 7.4 percent; for rural families, it rose from 2.4 percent to 6 percent (Wang: 2006). In our sample survey, in 2005 medical costs amounted to 7.46 percent of household income for those living in cities, 7.25 percent for those living in towns and townships, and 7.85 percent for those living in villages. For those in the lowest income category in rural China, medical expenses, excluding the costs of drugs and medicine, amounted to 22.14 percent of household income. This shows a marked deterioration from the 1990s when one study shows that in poor areas annual medical expenses per capita rose from between 2 to 3 percent of total income in 1990 to 8 to 11 percent of total income in 1998 (Liu et al.: 2000). The consequences of these trends are becoming increasingly visible. When combined with the increasing income and regional inequalities, we see a very varied picture of access to and quality of health provision. The World Bank (2005b) notes that although Beijing spent 360 yuan per capita, Shanghai 220 yuan, and Tianjin 150 yuan, all other provinces and equivalents spent under 100 yuan, with Anhui, Hunan, Jiangxi, Chongqing, Guizhou, and Henan each spending under 50 yuan per capita. In fact, one-quarter of all government health spending occurs in the rich municipalities and provinces of Beijing, Shanghai, Zhejiang, and Jiangsu. The difference in the number of hospital beds per 1000 between the richest and the poorest regions rose from 3.1 in 1982 to 4.2 in 2001 (Li and Zhu: 2004, p. 10). In fact, although the rural share of hospital beds increased from 40 percent in 1965 to 60 percent in 1975, during the period from 1982 to 2001 the total number of rural hospital beds dropped from 1.22 million to 1.02 million. The number of beds in township health centers dropped from 0.78 million in 1980 to 0.69 million in 2006 (http:// www.moh.gov.cn/open/2007/tjts/P91.htm). Over the period from 1980 to 1990, the number of beds per 1000 rural residents dropped from 0.95 to 0.81, and then 0.80 in 2000 and 0.76 in 2003 (CHED: 2005, p. 5). The

80 / providing public goods in transitional china

number of beds in urban China rose from 0.83 to 1.96 per 1000 residents. Between 1975 and 2006, the number of rural doctors and assistants fell from 1.5 million to around 900,000. This was a revival from 1985 when the number was only 643,000 (Li and Zhu: 2004, p. 11). Although in 2006 there were 5.2 medical personnel per 1000 residents in urban China and 1.9 doctors per 1000 residents in the countryside, the ratios were only 2.2 and 0.68 respectively. The number of rural village doctors and health workers per 1000 residents decreased from 1.79 in 1980 to 1.11 in 2006 (http://www.moh.gov.cn/open/2007/tjts/P100.htm and P109.htm). We also see a regional variation, with east China having 1.81 doctors and 1.38 nurses per 1000 population in 2006 and central China and west China having 1.39 and 0.99 and 1.40 and 0.90 respectively. Beijing had the highest ratio of doctors and nurses at 4.4 and 3.8 per 1000 and Guizhou the lowest at 1.05 and 0.61 per 1000 (http://www.moh.gov.cn/open/2007tjts/ P98.htm). Between 1990 and 2000, the difference in life expectancy across China rose from 3.5 to 5.7 years. We also see the inequities in the microdata. Thus, in the relatively wealthy Zhejiang province infant mortality per 1000 live births was about 20, whereas in poor Guizhou it was 60. This mirrors the findings in a study of health conditions in 30 poor counties that found an infant mortality rate of 52.3 per 1000 live births compared to a national average 21.6 for the rural areas. The rate of maternal deaths during childbirth was 216.8 per 10,000, as compared to a rural average of 114.9 per 10,000 (Meng and Hu: 2000, p. 67). In 2005, the under-5 mortality rate per 1000 live births was 10.7 in urban China and 25.7 in rural China, and infant mortality rates increased to 9.1 in urban China and to 21.6 in rural China. The maternal mortality rates per 100,000 live births were 25 and 53.8 respectively (table 4.4). In 2000, the under-5 mortality rate per 1000 live births in Beijing and Shanghai was under 10; in Yunnan, Chongqing, and Jiangxi, it was over 40. In Guizhou province, life Table 4.4

Rural-Urban Health Indicators, 2005

Category

Total

Urban China

Rural China

Neonatal Mortality Per 1000 Live Births Infant Mortality Per 1000 Live Births Under-5 Mortality Per 1000 Live Births Maternal Mortality Per 100,000 Live Births Medical Personnel Per 1000 Citizens Doctors Per 1000 Citizens

13.2 19.0 22.5 47.7 3.59 1.25

7.5 9.1 10.7 25.0 5.18 1.88

14.7 21.6 25.7 53.8 2.18 0.68

Source: Ministry of Health, at http://www.moh.gov.cn/open/2007tjts/P39.htm.

challenges in the health system / 81

expectancy is 13 years less than that in Shanghai, maternal deaths during childbirth are ten times greater, and under-5 child mortality is five times higher (Jonathan Watts, The Guardian, September 14, 2007). Both infant mortality and maternal mortality rates are closely correlated with prenatal care and attended safe delivery, two preventive services that have been adversely affected by the privatization of healthcare in rural China (Saich and Kaufman: 2005). With the disbandment of the rural collective institutions and with SOEs under increasing financial pressures, it is not surprising that medical insurance coverage has declined. Coverage through the rural collective medical system dropped from almost 80 percent in 1979 to only 2 percent in 1987 before improving to 6.57 percent in 1997 (“Zhongguo nongcun”: 2000, p. 21). Since 1997, there was a gradual improvement, with an increase in minimal coverage and the renewed promotion of the rural cooperative medical insurance scheme beginning in 2003. Since this revival, coverage has spread rapidly, with official figures claiming that coverage had risen in rural areas from 10.8 percent in 2004 to 76 percent by September 2007 (730 million farmers, China Daily, January 7, 2008). In our survey for 2005, 57.5 percent reported no medical coverage whatsoever, but this was a major improvement on the 75.4 percent in our 2003 sample. The figure was 34.6 percent in the major cities, as compared with 68.4 percent in the villages. The Third National Health Survey states that in 2003, 44.8 percent in urban China and 79 percent in rural China had no medical coverage. This was up for urban dwellers in 1993 when coverage was at 27.3 percent and down for rural China when cover was at 84.1 percent. Equally alarming is the fact that in our 2003 survey, 72.8 percent had no insurance for catastrophic illness and they had no intention to purchase any. Medical costs are a significant cause of poverty. A Ministry of Health survey reveals that among the rural poor, 22 percent attributed their poverty to unmet medical needs, whereas another study suggests that 30 to 40 percent of the urban poor are in poverty because of unmet medical problems (Li and Zhu: 2004, p. 16; Qiu Yulin: 2003). The 2003 National Health Survey revealed that the percentage of farmers who had sunk into poverty because of illness rose from 21.6 percent in 1998 to 33.4 percent in 2003. The new rural cooperative medical scheme is intended to address these two problems of catastrophic illness and poverty caused by medical bills. In 2006, the scheme covered 50 percent of China’s counties and by the end of 2007 it was intended to cover 80 percent of the counties and reach full coverage by the end of 2008. As a result of the costs and lack of effective insurance, fewer people are able to seek medical help when it is needed. According to the 2003 National

82 / providing public goods in transitional china

Health Survey, 22.1 percent of low-income families and 13.5 percent of other income groups who needed to see a doctor did not do so because of costs. Also, 14 percent of rural residents and 10 percent of urban residents left hospital without receiving treatment (UNDP: 2005, p. 60). Of those who left early, 34 percent of urban residents chose to leave, and of these, 53 percent did so for financial reasons. In the rural areas, 47 percent requested to leave hospital, and of these 67 percent wished to leave because of financial problems. According to the 2003 National Health Survey, 38.2 percent did not seek medical help because of financial reasons, up from 35.4 percent in 1998. The urban average was 36.4 percent and the rural average was 38.6 percent. In our survey, the figures were somewhat lower, with 25.1 percent of respondents stating that cost prevented them from visiting a doctor and 15.3 percent reporting unable to stay in hospital when needed (for those in rural areas, the numbers are 25.1 percent and 17.2 percent respectively). As noted earlier, the kinds of service that one can expect to receive depends on location and local government resources. Yanzhong Huang (2004) has looked at the relationship between the level of rural industrialization and investment in people’s welfare. Basically, when local governments have a surplus to disburse, they are more likely to address public health needs. By contrast, when there is a weak industrial base, local governments have to levy extra fees on their inhabitants to provide healthcare. Knowing that the residents are likely to resent and resist this, the local leadership is more likely to cut services. Huang (2004, p. 377) concludes that the more powerful the local state is, the better the public health provision will be. Huang equates power with financial capacity, largely from income drawn from rural industrialization. Thus, in Jiangyin (in Jiangsu, a province famous for its township and village enterprises), in 1985 50 percent and in 1992 over 97 percent were covered by some kind of medical insurance (when the cooperative medical system only covered 5.3 percent nationally). By contrast, in Jichun, Hebei province, where there was a poor level of industrial development, the cooperative medical system collapsed. Jichun moved to revive the system only after 1996 when it became clear that this was a national and provincial policy intent. However, according to Huang, the program functioned more like a poverty alleviation program than a health insurance scheme. One final important factor affecting the kind of healthcare available is the change in incentives for healthcare providers. Some analysts have felt that the problems stem from the de facto privatization of health facilities in the villages (UNDP: 1998, pp. 36, 38), but the change in incentives is more important than the question of ownership

challenges in the health system / 83

of the health facilities (Eggleston et al.: 2006).5 Not only was financing of most healthcare decentralized, but also from 1981 healthcare facilities were instructed to cover recurrent costs from user charges (with the exception of those of staff). Beginning in the mid-1980s, preventive care facilities were charged on a fee-for-service basis. The private practice of healthcare was legalized in 1985 and by 1989 healthcare providers were transformed into fee-for-service organizations and active competitors in the healthcare market. In 1999, the proportion of government funding in total revenue for hospitals under the Ministry of Health had fallen to 9.9 percent and in 2003 it fell further to 7.8 percent. The percentage of revenue from service provision rose from 42.2 to 46 percent and that from drug sales dropped slightly from 46.8 percent to 43.7 percent (Gu and Zhang: 2006, pp. 65–66). Also, government support was not related to performance-based criteria but rather to the number of staff and beds, thus skewing the incentives (Eggleston et al.: 2006, p. 7). Central government pricing policies to ensure affordable healthcare had a dramatically unintended effect (Blumenthal and Hsiao: 2005, p. 1166). In order to make services available for the entire population, prices for basic and noninvasive care were set below cost. By contrast, more complex surgery and high-tech procedures were set above cost to offset this. This gave health providers the incentive to overprovide in expensive curative, invasive surgery. This was compounded by a markup of between 15 and 20 percent on drug sales. Not surprisingly, this has led to the overprescription of drugs. Extraordinarily, half of Chinese healthcare spending is on drugs, compared to 10 percent in the United States (Blumenthal and Hsiao: 2005, pp. 1167–68) or 15 percent in Europe. In 2006, drugs comprised 50.5 percent of the total costs (128.7 yuan) for outpatients and 60.8 percent of the total costs for inpatients (total costs based on 1992 yuan). A World Bank study of village clinics found that less than 1 percent of drug prescriptions were necessary and that 20 percent of expenditures for appendicitis and pneumonia were unnecessary. Such statistics indicate the need for a more concerted policy effort in the next phase of development if China wants to maintain its former impressive achievements in public health. The problems outlined above have motivated the leadership to focus on a more balanced and equitable growth strategy and to tackle major health challenges, such as that presented by HIV/AIDS. In addition to a policy approach that emphasizes equity in development, the government has established a set of 16 indicators by which it will measure progress toward building a “comfortable society” (xiaokang shehui) by 2020 (table 4.5). These indicators are broader than the usual economic indicators that China has used to

84 / providing public goods in transitional china Table 4.5

Xiaokang Indicators and Targets for Selected Years

Indicator

2002

2010

2020

GDP: Trillion Yuan Low High

1.024

1.764 1.838

3.311 3.616

Population in Billions

1.285

1.38

1.48

GDP Per Capita (Rmb)

8000

12800–13320

22370–24430

In US Dollars (1:8.28)

962

1546–1610

2790–2950

Percentage of Urban Population

39.1

48.0

60.0

Percentage of Workforce in Third Sector

27.7

38.0

51.0

Education Spending as Percentage of GDP

2.3

4.0

4.5

University Students as Percentage of Age Group

7.8

12.0

20.0

Average Life Expectancy

71.8

73

74

Doctors Per 1000 Population

1.6

2.0

2.5

Urban Disposal Income (Rmb)

7703

12300

22000

Rural Net Income (Rmb)

2476

3830

6860

Urban Per Capita Living Space (Square Meters)

20.8

24.5

30

People’s Average Electricity Consumption (KWh)

140

330

712

Engels Coefficient (Percentage)

43

33

30

Urban-Rural Income Gap

3.1:1

3.2:1

3.2:1

Poverty-Wealth Gap (Gini Coefficient)

0.458

0.45

0.40

Source: Author’s construction from various documents.

measure its progress. Thus, for example, life expectancy is expected to rise from 71.8 to 74 years and the number of doctors per 1000 population to rise from 1.6 to 2.5. Whether these targets can be achieved is open to debate, but such indicators are important in alerting officials to the fact that their future work performance may not be judged solely on GDP growth and the ability to maintain social order and to keep population growth under control. In the following two sections we look in more detail at health policy, primarily with respect to insurance coverage in both urban and rural China.

challenges in the health system / 85

Healthcare Coverage in Urban China The pre-reform medical system for urban residents could not survive the shift to a more market-oriented economy. As in a number of other areas, it was a fiscal crisis that pushed reforms forward. The former system as it operated was rather akin to the National Health Service in the United Kingdom, with free care and state financing, either directly or through the enterprises, and public organizations delivering the services. This was complemented by a patient self-payment component for services from publicly-run facilities (Gu and Zhang: 2006, pp. 52–53). But by 1988 this system had become too expensive to maintain, as the annual growth of medical care grew 40 percent, having increased from 26 percent the previous year. Over the next few years, spiraling costs were reined in at 18 percent in 1991, but they edged up again in 1993–1994 with the serious attempts at reform (China Labor Statistical Yearbook: 1998, pp. 40, 505). The rise in costs was caused by hospitals and clinics seeking to become profitable. Although these costs were rising, enterprises had less money to dedicate to medical care for their employees. For example, in 1992 the Shenyang Internal Combustion Factory was able to provide its 4,500 workers and pensioners with almost-free medical care. This cost the factory around 70,000 yuan per month. By the end of 1997, the factory had not been able to pay wages for 10 months and had a monthly healthcare budget of between 5,000 and 7,000 yuan (FEER, October 16, 1997, p. 64). During the first phase of reform, not only did free healthcare coverage not expand but it already began to decline. In 1978, 57.4 percent enjoyed some kind of free healthcare and although the absolute numbers grew, coverage rates declined from a high of 59.6 percent in 1981 to 52.8 percent in 1994 (Gu and Zhang: 2006). Despite this, profitable enterprises were still able to cover costs and pay them in a timely fashion. Yet, even in these enterprises there could be frustration, with the enterprises signing contracts with specific hospitals that the patients were first required to visit, even if it was not the most suitable for their specific ailment. During the 1990s, recourse to private medical schemes was not only alien to Chinese socialist culture but also too expensive for those most in need. In 1997, the least expensive private insurance available in Shanghai to those over age 50 but under age 65 was 600 yuan a year, covering up to 8,100 yuan of hospital costs. But for the poorest 10 percent of the population—the most vulnerable such as the retired, unemployed, and workers at effectively bankrupt SOEs—this would take a big cut out of their 1997 annual income of 3,785 yuan.

86 / providing public goods in transitional china

Significant experimentation began in November 1994, with pilots in Zhenjiang (Jiangsu province) and Jiujiang (Jiangxi province), in what was referred to as the “two-Jiang model.” Both cities had populations of around 2.5 million (for details, see World Bank: 1997a, pp. 56–59; Yu and Ren: 1997, pp. 448–50; Gu and Zhang: 2006). Pilots were also set up in Shanghai and Shenzhen, and their success encouraged the government to extend the “two-Jiang” experiments to two additional cities or prefectures in each province, thus making a total of 50 sites. The experiments were based on a mixture of individual accounts and social pooling. They combined a compulsory insurance scheme with public institutions as the providers and compulsory savings to be held in an individual medical account along the lines of Singapore’s insurance program. Workers contributed 1 percent of their annual salary, with the enterprise providing an additional 10 percent. These funds went to newly set up insurance centers that committed them to individual and public medical insurance accounts. The split was roughly 50–50. The costs were also shared on the demand side; if a worker fell prey to a serious illness, the individual account was drawn on first. Once this account was exhausted, the worker paid up to 5 percent of his/her annual salary for treatment and then cost sharing kicked in, with the government covering from 80 to 95 percent of other costs. However, for major illnesses the individual liability was considerable and was enough to drive an average family into poverty. The results were good in as far as they went. Coverage for those in loss-making SOEs increased, the costs of expensive medical tests were cut, the annual rate of growth in aggregate hospital expenditures declined by 23 to 28 percent, and inpatient and outpatient visits declined (World Bank: 1997a, p. 58). However, a number of shortcomings were also observed. First, the total coverage rate was low, as it covered only 11.5 percent of urban employees. Second, with the exception of in Zhenjiang, employees from nonprofit and administrative units did not participate. Third, the program was highly fragmented and there was no capacity for portability. This last factor also undermined the capacity for improvements in labor mobility. As a result, in January 1998 the government proposed and further elaborated later in the year a new scheme to be put into operation by the end of 1999. This required the establishment of a “basic health insurance scheme” for all cities, and all enterprises and employees were expected to join. This included nonstate enterprises and those working in foreign enterprises and joint ventures. Under the new scheme, the employee contribution was raised to 2 percent of the annual wage and the employer contribution rate was 6 percent of the total wage bill. The employee’s

challenges in the health system / 87

contribution and 30 percent of the employer’s contribution were paid into the individual account and the remainder was paid into the pooled account. The individual accounts were to be used for minor medical bills, whereas the funds in the pooled accounts were to be used for major treatments that included hospitalization (SCMP, January 8 and 25, 1999; Gu: 2001; Gu and Zhang: 2006, pp. 58–60). The pooled accounts, used only at designated hospitals, paid out a minimum of 10 percent of the patient’s annual wage and a maximum of 400 percent. Any amount over this was to be covered by commercial insurance. The designation of a maximum was expected to reduce costs. The new system allowed local governments to set regulations for individual accounts. Administration of the pooled funds was put under the department of labor and social security of the local government. This led to the potential for corruption and misuse of funds. Importantly, the new system excluded the provision of medicines from hospital provision. As we have seen, one of the main ways for hospitals to generate revenues had been through the sale of drugs. Because basic pay for doctors was low, they generated further income not only from “donations” from patients who wished to be treated, but also from kickbacks from pharmaceutical companies and from the purchase of unnecessary medical equipment. Thus Beijing has more than 70 CT scanners, more than the total in England, and Nanjing hospitals have three times as many CTs as London (SCMP, January 25, 1999). The new providers and pharmacies were to be accredited by a mixture of local labor and social security bureaus, employers’ organizations, trade unions, healthcare providers, and designated experts. The new system was intended to unify the myriad existing different health schemes into a standardized program. The compulsory universal scheme was expected to provide those enrolled with the financial means to meet their health costs without causing massive financial distress to the family. However, resistance to the reforms has persisted. In addition, it is clear that many hospitals have not ended their sales of medicines. In May 2000, Chinese officials were still complaining that many hospitals derived half their income from this source, claiming that in some of the smaller hospitals this revenue comprised 70 to 80 percent of hospital income (SCMP, May 29, 2000, Internet edition). China’s leaders hope to introduce competition among hospitals, allowing patients to choose rather than being allocated to a specific hospital, and to improve efficiency. Former vice premier Li Lanqing has even suggested that public tenders for drug provision might be introduced. However, a vice governor of Guangdong province has suggested that reforms are not moving as well as planned and have been resisted by cash-strapped hospitals. She said that many hospitals are more concerned with the financial bottom-line than with the

88 / providing public goods in transitional china

patient’s health and are even renting their space and name to unqualified clinics (SCMP, June 14, 2000, Internet edition). It is not surprising that in our survey from 2003 to 2005, medicine and drug provision was rated as being in urgent need of attention by residents in urban China (see Chapter Eight). The central government has made a concerted push to expand coverage even as it recognizes that universal coverage will be beyond current capacity. The coverage rate in this new system has risen steadily, from just 1.4 percent in 1993 as the local experiments were beginning, to 7.9 percent in 1999 when the national program was put into place, and then to 25.8 percent in 2001 and 39.8 percent (124.4 million people) in 2004 (Gu and Zhang: 2006, p. 60). This reveals that despite considerable reforms and central government promotion, the new medical insurance scheme has not come to dominate for urban employees. As noted above, the number of people with no health coverage has risen from 27.3 percent in 1993 to 44.8 percent in 2003. We see similar low coverage rates in our sample survey conducted over the 2003–2005 period. In 2005, 34.6 percent of those in major cities had no coverage whatsoever (the rate had been 45.7 percent in 2003, and only 19.1 percent had coverage for major medical treatment). The percentages for small towns and townships were 63.8 percent and 9.6 percent respectively. In our sample survey for 2005, we see a marked increase in the use of commercial insurance in one form or another. In the cities, 51.8 percent of the sample had commercial medical insurance. It is not surprising that respondents ranked health insurance as one of the five services with which they were most dissatisfied and that they felt was in urgent need of improvement. The new system suffers from the same kinds of problems as the trial experiments. It has been difficult to persuade those government employees who have relatively guaranteed tenure to shift to the new system, just as it has been hard to persuade those working in the nonstate sector such as the foreign-owned and private sectors to enroll in the new system. Many of the poorer enterprises have found it difficult to make their payments and there has often been collusion with local governments to reduce or avoid payment obligations. Because local governments have been concerned about the possible social unrest that might come about due to enterprise bankruptcy, they have often turned a blind eye to enterprises not making their payments. The greatest problem, however, is that the system is designed for the working elite and consciously excludes the marginalized social groups such as the unemployed and those working in the informal sectors of the economy, especially the migrant workers, who have fueled China’s urban

challenges in the health system / 89

boom. These groups remain just as vulnerable as ever. This need has led to a new round of policy experimentation. Although the wages of migrant workers are lower than those of urban residents, their main problem is the lack of access to social services and welfare facilities. Those migrants who are unable to transfer to an urban resident permit (hukou) are not properly integrated into any system and this reached the policymakers’ attention only in the late 1990s (see Fan: 2007; Cai: 2003). If accommodations are provided for the migrants, they tend to be very rudimentary, either in the form of segregated dormitories, tents, or temporary shacks. The sanitary conditions are poor, creating significant problems as migrants generally do not have health coverage. Those who do have coverage probably will only be covered for direct work-related injuries and not for infectious diseases or other sicknesses. This places a burden on the already weak rural social infrastructure as when one gets sick the only way to avoid expensive urban treatment is to return to the village to be looked after by the family. It is not surprising that a survey conducted by Fudan University found that only 7.6 percent of migrants were satisfied with their social status in the city (China Daily, January 7, 2008). The leadership under Hu Jintao and Wen Jiabao put together experimental reforms in the first State Council document of 2003 that acknowledged the problems of migration but confirmed that they were an inevitable part of China’s progress. In March 2001, the central government decided to promote reforms in small towns based on prior experiments. The reform of the household registration system had begun in 1997 for 450 small towns (Yu Depeng: 2002, p. 379). This allowed residency in small towns and townships for all those from rural areas who could demonstrate legal employment and a place to live. Although restrictions have subsequently been eased for some larger cities, major municipalities have still sought to control the flow of migrants. Importantly, in November 2002 migrants were given the political status of being a part of the working class. In January 2003, the State Council confirmed that the unfair restrictions on migrants were to be lifted and they were to be accorded equal treatment with urban residents when applying for work. Their wages were to be paid in full and on time and their living and working conditions were to be improved. Perhaps most significantly, urban education departments would have to recognize schools for migrant children and provide them with equal education access. It was left up to local governments as to whether they would implement these measures, thus many of the demands are more honored in the breach than in the observance. However, the official recognition is important. Also, as Fan (2007) points out, there is the usual clear state bias in the awarding of an urban registration. Skilled and urban migrants are far more likely to be granted

90 / providing public goods in transitional china

permanent residence in the city than the less qualified, rural migrants. This will perpetuate the gradual emergence of an urban underclass that will not have access to quality medical services and other welfare facilities in the urban areas. The fiscal buoyancy of recent years (in the first half of 2007, the Ministry of Finance stated that revenues were up 30.6 percent over the same period in 2006) has allowed the leadership to experiment with extending medical coverage. In 2007, the central government was able to allocate 1.3 billion yuan to subsidize local governments’ urban healthcare (an increase of 170 percent from 2006), while the budget of local governments increased by 30 percent to 1.76 billion yuan (Xinhua News Agency, December 13, 2007). In July 2007, Premier Wen announced that over the next three years experiments would be developed to provide medical insurance to primary and secondary school students, adolescents, the elderly, and the disabled. This will be provided through the Urban Residents’ Basic Medical Insurance Scheme. Experiments began in September 2007 in 79 pilot sites and an evaluation was to follow at the end of the year. The scheme was launched on a voluntary basis, with the amount of both the premiums and claims said to be “moderate initially” (SCMP, July 25, 2007, Internet edition). With the attitude of the top leadership now clear that healthcare reform is a priority, local officials have become more serious about implementing suggested reforms. Experimentation has also been launched for a medicare-style network in urban centers with an expansion in the role of community healthcare centers to support the basic medical insurance scheme. Beijing and Shanghai were the first two municipalities to launch this program with a plan in 2008 to cover 5 percent (250,000) of the working population. The basic medical insurance scheme should cover all those in need in Beijing in 2008 and in Shanghai by 2010. Elderly without medical insurance and with minor ailments or chronic conditions (such as diabetes or hypertension) are encouraged to seek help at community healthcare centers where they will be given a discount on outpatient service fees. The Beijing Municipal Government was also thinking about a separate program for its 70,000 disabled residents and the 80,000 said to be living on the minimum living subsistence support (China Daily, January 2, 2008). Healthcare Coverage in Rural China In contrast with the urban areas, most people in the countryside were left to depend on the family and, in times of desperation, on collective funds for healthcare. However, this does not mean that reforms have not benefited the countryside—one of the most remarkable effects has

challenges in the health system / 91

been to lift more than 200 million people out of dire poverty. Yet, it is also true that inequality rose and collective rural health services were paralyzed by the initial reforms. New policies were necessary to resolve these problems. Although we are dealing with the urban and the rural separately in this chapter, it is worth bearing in mind that the progress of each is linked to that of the other not only in terms of the number of migrants who are now moving to the urban areas but also in terms of the skewed distribution of resources that has been discussed above. Urban elites are likely to resist any significant transfers of funds to the rural areas and thus the new leadership will have to rely primarily on a redistribution of the new funds that have become available with China’s economic progress. Income alone does not account for all the inequality. As we have seen, the cost of healthcare takes a higher percentage of rural disposable income than it does of urban income, while insurance is less readily available. Although some inequality may have been inevitable with the transition, policies and/ or policy inaction have heightened the polarization. In the initial period of reform, healthcare access vanished for the most vulnerable populations, and regressive social policy exacerbated the situation as the poor were paying more in taxes and fees and the rich were receiving net subsidies. On taking power, the Chinese Communist Party set about to put in place a basic structure for healthcare in rural China. Given the scale of the problems, it was impossible to focus on individual healthcare needs through a massive expansion of delivery services, although there was impressive progress here as well. Instead, a major focus of the post-1949 general health policy that was important for rural areas was disease control and health education. The promotion of these programs was helped by the collectivization of agriculture. Collectivization meant not only that the collective organizations could absorb much of the cost of preventive health and education campaigns, but also that the rural population could be more easily mobilized for health campaigns. On the whole, it did not matter whether they were working in the fields or clearing away stagnant water in which malaria-ridden mosquitoes might breed. They would get the same number of work-points for each day worked and any medical costs would be deducted at source, often in kind, from the collective income. A combination of prevention and health education was adopted as part of a four-pronged policy adopted at the First National Health Conference in August 1950 (Wong and Chiu: 1998). Other key features of the policy were that public ownership and provision should predominate, and traditional Chinese medicine should be combined with Western medicine. This policy led to a decline in private medical practice. For

92 / providing public goods in transitional china

example, by 1962 only 3 percent of all health workers in Shanghai were in private practice, down from 56.5 percent in 1950 (Anson and Sun: 2005, p. 12). During the phase of collectivization of agriculture, the three-tier structure of rural healthcare and a rudimentary insurance system was put into place. Clinics were set up at the village (brigade) level, hospitals at the township (commune) level, and more comprehensive hospitals at the county level. This three-tier structure has remained fairly constant with only little variation. The rural cooperative medical scheme began with a number of local experiments that were then adopted as national policy in 1960 when the ministries of health, agriculture, and finance combined to adopt draft regulations. This led to 90 percent of the brigades (villages) establishing such schemes. Funding for the schemes drew from three financial streams: contributions of brigade members (up to 2 percent of the household income, between 1.5 and 3 yuan, deducted directly by the collective); and brigade welfare funds and public welfare funds held by the commune. This allowed for the provision of healthcare to individuals at a subsidized rate. The basic system showed considerable variation from brigade to brigade depending on their particular financial circumstances. Although initial reforms in urban China left healthcare basically intact, reforms in rural China heralded immediate and dramatic changes for both the healthcare providers and also for families in terms of contributions and insurance coverage. With increased income, those who could afford it tended to avoid what they saw as lower-quality village clinics and even township hospitals and they preferred to go to county-level hospitals. The change in the ownership structure of village networks and the nature of the incentive system that arose as a result of these changes gave rise to significant challenges (UNDP: 1998, pp. 36, 38). Thus in terms of national health spending, whereas the collective schemes accounted for 20 percent in 1978, by 1993 they accounted for only 2 percent (UNDP: 1998, p. 37). In addition, as discussed above, the incentives for these health providers also changed. The rural healthcare apparatus has a three-tier structure. As of 2006, at the village level there were 864,168 health professionals, of whom 3.68 percent had a college education, with a further 103,863 licensed physicians and assistants. There were 583,209 village clinics covering 83.7 percent of all villages, up from 66.6 percent in 1985 but down slightly from a peak of 89.8 percent in 2005. Essentially, they provide a referral system to the higher levels. Importantly, they charge a fee for their services and are also allowed to charge for medicines that they prescribe. In the villages many health workers work for themselves and the number of private clinics has risen rapidly. The number of healthcare

challenges in the health system / 93

personnel per 1000 of the rural population has dropped from 1.55 in 1985 to 1.05 in 2005. The buildings they use are often unsanitary, the staff are poorly trained, and there is little incentive to upgrade facilities. Not surprisingly, diagnostic capabilities are extremely weak, leading to incorrect diagnosis or the lack of recognition of “new” diseases such as SARS or HIV/AIDS. As in the urban system, the sale of medicines is the most lucrative form of income for these local health workers who can receive high fees especially for Western pharmaceuticals. Three months of training is sufficient to be able to prescribe drugs. Not surprisingly, preventive medicine has been eschewed in favor of more expensive curative approaches. This increases the financial burden on the rural household. The choice in the poor areas is stark if major illness strikes. Either wait and die or incur financial burdens that will take decades to pay off. In a very interesting study of rural medical provision in Hebei, Anson and Sun (2005) note the differences in income between those who have remained in group practice and those who have set up private practice. Their survey shows that the income of village doctors, especially those in group practice, has not kept pace with that of other villagers and this must be an incentive to give up medical practice. Indeed, all village doctors also engaged in agricultural or other work to supplement their income. This was especially the case for those who were self-employed. Patients’ fees comprised almost all the income generated by private village doctors and most of the income for those in group practice, but only 30 percent of the income of those employed in government-owned clinics (Anson and Sun: 2005, p. 101). At the township level, in 2006 there were 39,975 health centers that carried out basic medical tasks, such as simple surgery and treating infections. The number of centers has varied over the reform period. In the early years of reforms from 1980 until 1985, the number dropped by 14.5 percent and there was a further 4.5 percent decline by 1993. In 1994–1995 measures and incentives were introduced to improve the quality of provision at these centers and their number increased again by 15 percent (Anson and Sun: 2005, pp. 45–46). However, over the entire period from 1980 to 1994–1995, the number of township health centers fell from 55,413, a drop of 27 percent. In recent years, the absolute decline has been influenced by the merger of townships throughout parts of China. The statistics for villages with a clinic and for the number of beds and personnel at township hospitals reveal the variations between the more economically developed areas and the less developed western part of the country. At first glance, the figures for the village clinics seem counterintuitive. In 2005, although the national average for villages with a clinic was 83.7 percent, for east China it was 77.9 and for west China

94 / providing public goods in transitional china

88.3. Major municipalities in the east such as Beijing (68.3) and Tianjin (60.8) and rich provinces such as Shandong (63.5) and Zhejiang (46.1) had low village coverage. This is because many of the villages have been incorporated into the urban areas, and healthcare provision is more readily available in townships or urban areas. By contrast, provinces in the poorer northeast (Liaoning, Heilongjiang) and some poorer provinces in the central and western regions (Jiangxi, Chongqing, Henan, Yunnan, Ningxia, and Qinghai) had full village coverage. Not surprisingly, village coverage was low in Tibet (59.6 percent), with only Xinjiang (53.1) ranking lower. One of the most important indicators of healthcare access is the availability of beds and personnel at the township level. Shanghai in 2005 had 4.49 township beds and 3.81 medical personnel per 1000 of rural population, whereas Guizhou had only 0.43 and 0.60 respectively. Generally, the east is better served than the west, with 0.92 beds and 1.35 medical personnel per 1000 as compared to 0.70 and 0.91 respectively. Insufficient state funding means that the extended family provides the necessary support for many of those in the countryside, but changing demographics have led to the realization that over the long term the family cannot be the sole provider. The Ministry of Health called for 8 percent of rural budgets to be spent on healthcare, but actual resource allocations fall far short of this. The weaknesses of the rural health system have resulted in strong criticism, both domestically and internationally. The comment by Dr. Marcel Roux, China head of Médécins Sans Frontières, just before the new push for healthcare reform was launched is striking: Healthcare is better in Africa, for sure. There, people are organized and there are good African physicians and health workers but in China, they don’t have the knowledge, the structures or people to make it work (SCMP, October 19, 1999, Internet edition).

Such criticisms prompted the Chinese leadership to try to improve conditions. The first step was to reconfirm the 1994 decision to restore the cooperative medical system. In this respect, China has suffered from “state withdrawal” and the lack of public funding, training, and regulation. Evidence clearly shows that where there is a functioning cooperative medical system, utilization rates of the medical services increase, especially the demand for clinic care and hospitalization. This is one clear case where the “public good” argument would appear to apply and the government needs to tighten the regulatory framework to ensure that guidelines on health are followed and that in poor areas there is better provision by the central government. The vital role of the government in providing basic healthcare

challenges in the health system / 95

in the rural areas has now been acknowledged by China’s senior leaders. The minister of health and the party secretary of the ministry, following a national health conference, stated clearly that market forces should not be allowed to dominate service provision. In their view, inequality and ignoring the needs of the countryside would be the result of too strong a reliance on market forces. The weakening of the public interest function of the health sector, in their view, would overlook the “principle of social benefit and equity” (reported in SCMP, January 2, 2008). One of the first experiments with the cooperative medical scheme, covering 26 villages in two counties, took place in Sichuan in 1989–1990 (World Bank: 1997a, pp. 49–50). Premiums were set at 1.5 percent of average income and those insured could freely visit village and township facilities, but visits to the county level were allowed only in cases of emergency or on referral from the township. Participation rates were high: 90 percent, with a re-enrolment of 95 percent after the first year. Administrative costs were kept low at only 8 percent of total reimbursements. The need for some kind of catastrophic insurance was clearly shown as 11.5 percent of the covered population used 70 percent of total health expenditures. For such schemes to work in poor areas, it is necessary to diversify contributions away from the household and to use village social welfare funds and government allocations for poverty relief. One study of 30 poor counties in the mid-1990s found that household contributions to community health funds made up about 50 percent, with about 20 percent coming from the village funds and 16 percent coming from the government (personal communication). These experiments were summarized in an article in July 1994 (People’s Daily, July 2) that proposed community-based healthcare schemes for the rural areas. The system was voluntary, with each community organizing its own collective financing for basic healthcare. In line with the experiments to date, the funding base was to be diversified and priority was to be given to preventive services. The chaotic state of rural healthcare continued to attract the leadership’s attention. In December 1996, a National Conference on Health was convened to try to fix the broken system. The change of tone was remarkable as the leadership shifted from presenting its healthcare system as a shining example to other developing countries to one of concern about its collapse (discussion with participants at the conference; Cailliez: 1998, pp. 42–43; for the joint CC and State Council Decision, see Jiankang ribao February 18, 1997). The central government was clearly alarmed at the social consequences of the financial decentralization and decollectivization and it tried to put forward a coherent policy to restore standards and access to healthcare. First, it declared that

96 / providing public goods in transitional china

spending in the national budget would be raised from 2 to 5 percent, something that has not yet been achieved. Further measures sought to revive preventive healthcare and public hygiene awareness through education. The proportion of the government budget for spending on preventive care had dropped from 23 percent in 1978 to 18 percent in 1994 (Hu and Jiang: 1998, p. 192). Village doctors were to receive a pay boost to bring them in line with government officials and to stop their reliance on kickbacks and other nonsanctioned revenues. The fact that the need to build a cooperative medical system was reiterated once again in 1996 suggests that little progress had been made. Variation and noncompliance were widespread and the frequent requests by the Center to reduce financial burdens on farmers were interpreted by some as meaning that they should not have to contribute to medical schemes. In 1996, the Ministry of Health undertook a study of the rural cooperative medical care experiments and, in 2003, based on this, the government issued a set of ideas about how to move forward. The new rural cooperative medical scheme is defined as voluntary public medical insurance. The household forms the basic unit for participation, with the fee set at 10 yuan. The government would contribute a subsidy of 20 yuan, which was increased to 40 yuan in 2006. This payment is divided equally between the central and the local government. The amount of government support was increased again to 80 yuan starting in 2008. In addition, it was suggested that the individual contribution might be raised to 20 yuan. The program has expanded rapidly and by 2007 covered fully 20 out of China’s 31 provinces, which amounted to 730 million farmers (76 percent of the total). In addition to the medical scheme, the government has also launched schemes for financial aid for poverty-stricken families. By 2005, over 11 million such payments had been made (Wang: 2006). Finally, on August 29, 2006 the “Construction and Development Plan for the Rural Health Service System” was promulgated. This called for the central government to invest 21.7 billion yuan to construct health facilities, with 14.13 billion yuan dedicated for township hospitals and 4.9 billion yuan for county-level hospitals to improve outpatient facilities and to buy key equipment. However, little is allocated to the villages. The objective is that by 2010, 8 percent of the county budget will be dedicated to healthcare and each village will have a clinic. A number of problems still afflict the cooperative health system. In the individual accounts there is now an allocation of 50 yuan per year (90 or 100, starting in 2008), but in 2005 the rural per capita medical expense was 318.5 yuan (http:www.moh.gov.cn/open/2007tjts/P113.htm). Van Dalen (2006), using data from the 2004 China Health and Nutrition

challenges in the health system / 97

Survey, shows that health insurance is not affecting the demand for healthcare in any significant way. The problem is clear—insurance does not offer real protection against high healthcare expenditures. Thus, we see a continuation of many people declining health service provision when it is needed. Second, the scheme also suffers from those problems normally associated with voluntary insurance schemes. Many of those who are better off or consider themselves healthy do not participate. However, many poor households forgo treatment because of the high costs, meaning that those receiving reimbursements tend to be concentrated in the higher-earning households. The system has also been prey to coercion. In western China, it is considered an important political achievement to increase participation in the scheme—thus some peasants have been forced to join, or local governments have required officials and teachers to pay fees to boost the participation rates (CHED: 2005, p. 20). The counties carry the costs of the medical insurance schemes and this leads them to seek ways to deflect costs. Many design conservative plans that limit coverage and benefits. Some shrink the risk pool by devolving responsibility to the township or by limiting the payout level or the kinds of treatments that can be covered. Concluding Comments Public health, especially the delivery of services to poor rural areas and the provision of health education, would appear to be a case where the public goods argument applies. The central government needs to tighten the regulatory framework to ensure that guidelines on health are followed and that key services and provision in poor areas are available at central government expense. The central government would be well served to be the provider and supporter of public health, ensuring more equitable access, rather than focusing its efforts on subsidizing the salaries of professionals in the health system. The central government has acknowledged the persistence of major problems and has begun to dedicate more resources to health provision and education. However, increased resources alone will not be sufficient to resolve these problems and there is the additional challenge of making sure that the new funds are not wasted. As we have seen, the incentives for health providers are badly skewed in favor of expensive, curative care and the provision of drugs. In terms of meeting equity challenges that the new leadership has made central to it policy agenda, the World Bank (2006, p. 144) has suggested that a healthcare financing system should have two goals: to provide affordable access to a basic package of goods and to provide financial protection against the costs of

98 / providing public goods in transitional china

catastrophic illness. In 2005, according to our survey, only 14.4 percent of respondents had coverage for major illness, of whom 28.6 percent were in major cities and 18.9 and 8.2 percent in smaller towns and villages respectively. In China, as in other low-income countries, out-of pocket expenses are the main form of expenditure for healthcare, thus further marginalizing the poor. The Asian Development Bank (2002, p. 3) has shown that out-ofpocket expenses in China tend to be higher than in other countries in the region. In China these expenses are over 70 percent, whereas in Korea they are 43 percent and in Malaysia 42.4 percent. Expenses in China are even higher than in Indonesia (47.4 percent) and the Philippines (49.1 percent). Reducing these out-of pocket expenses is difficult and requires not only enhanced fiscal transfers from the central state but also concerted political will. In order to reduce charges for the poor, effective ways must be found to increase the size of the risk pool and to improve the pre-payment mechanisms. The insurance schemes that have been introduced will help, but they are unlikely to increase usage to the kinds of levels that are necessary. In addition, where voluntary, these schemes suffer from the same kinds of problems as such schemes elsewhere. One major problem in the fiscal system is the tension that arises from the concentration of resource spending on urban over rural health. This has strong historical roots—one of the underlying criticisms by Mao Zedong when he launched the Cultural Revolution was the urban bias of the health system. The urban bias will not be eradicated entirely, but recent measures by the leadership to improve funding and make a major investment in rural China will help. China has lacked a powerful interregional transfer payment system. County or township fiscal revenues as a percentage of national revenue deceased from 31.6 percent in 1993 to 19.7 percent in 2000, whereas the share of county or township expenditures only decreased from 31.4 percent to 26.4 percent (UNDP: 2005, p. 75). Although it is impossible to expand health coverage to all citizens in the short term, intermediate steps might be taken to expand coverage to the families of urban employees and try to bring the unemployed into a program that would have government funding. This would expand coverage beyond the formal sector of the economy, but there would remain the need to integrate migrant workers more effectively. In addition, the government could provide a greater subsidy to the new rural cooperative medical scheme, with full funding in the poor areas. Certainly the central government has the necessary fiscal capacity. There remains the problem, however, of whether this will improve take-up rates if the gap between coverage and full payment remains as large as is currently the

challenges in the health system / 99

case. China may wish to look at experiments in other countries, such as risk pooling in Colombia, health cards in Vietnam, and Thailand’s “30 baht” universal coverage scheme (World Bank: 2006, p. 12). In particular, it is important to protect low-income groups against low-risk, high-cost occurrences through some kind of pooling as they seem better able to deal with high-frequency health shocks that are low-cost (World Bank: 2006, p. 145). Some have argued that the urban bias of spending and competing demands will prevent this. Yet, according to Professor Ba Danian from the School of Medicine, Zhejiang University, a scheme to expand urban coverage to families of the employed and unemployed, together with subsidies for rural China and full support for poor areas, would require only 150–200 billion yuan from the central government. In 2007, it was predicted that national revenues would increase by 1.2 trillion yuan over 2006 (31 percent). Thus, if the political will is there, so is the funding. There are a number of more specific policy measures that could be undertaken to help improve the situation. The Ministry of Health is a weak player in the bureaucracy and mainly represents the interests of the health professionals rather than the population at large. The weakness of the Ministry of Health is exacerbated by the bureaucratic fragmentation of the system of healthcare provision. One good example is the unintended consequences of the introduction of the family-planning system. The creation of the family-planning program and agencies throughout the administrative system drew resources away from maternal and child healthcare, yet the greater burden fell on maternal and child health workers (Saich and Kaufman: 2005). With respect to health provision at the local level, integration of the family-planning network with the general health system would be beneficial. This would have the effect of restoring funding to the impoverished maternal and child health (MCH) system and would help subsidize related services such as gynecological care and follow-up for contraceptive side-effects and problems. This topic has been the subject of much discussion and debate in recent years and was hotly debated at the November 2000 conference on Health Reform and Development. It has been identified as one for further immediate policy analysis and discussion through the China Health Development Forum launched in 2000.6 However, as of 2007 it had not been implemented effectively because of conflicting bureaucratic interests. China has been moving in the opposite direction to most of the world that has sought to integrate better health services, especially at the local levels. However, China’s engagement in international health discussions and its acceptance of the international reproductive health agenda have led some to realize that this fragmentation may not be beneficial.

100 / providing public goods in transitional china

One important unresolved question is the role of the government and the market in providing health services. Naturally, in the area of public health there is a propensity to stress the role of government, but even here the Chinese authorities have a strong state bias and in some areas an inadequate understanding of the role that markets can play. Opinion in China is divided along familiar lines of those who favor a UK-style national health system and those who favor a greater role for the market that would move the system closer to that in the United States. The former has been argued strongly by a research group at the Development Research Center of the State Council and by Professor Li Ling at the China Center for Economic Research, Peking University. The view that the problem is not with the market as such but with the policy biases has been expressed by Professor Gu Xin, Peking University (2005). In fact, on the question of the role of government in public health, especially in terms of provision for poor rural areas, all seem to agree that the government should take primary responsibility. Those who favor a stronger government role suggest that the extra funding should be allocated through the Ministry of Health. The ministry would then fund and manage public agencies. But the ministry and its public-funded agencies have little incentive to provide efficient services or to be responsive to citizens’ needs. Certainly, the Ministry of Finance seemed skeptical of this approach, with concerns about the inability to control costs, waste of resources, and inefficient allocation. One major area of contention is over the question of providers. The group from the DRC argues for state control over most medical institutions and for planned prices. In so far as market measures are used, they do not concern providers as such but rather are to be used to make the performance of the institutions and their personnel more effective while under government management and ownership. They argue that this will provide greater fairness and equity. By contrast, others such as Gu Xin argue that the greatest problem is that most people do not have medical insurance and that medical services are too expensive. This is independent of the ownership system. Given the incentive structure at present, it is not surprising that hospitals encourage patients to “overconsume.” The dominance of public hospitals and the skewed incentives undermine the benefits that marketization might bring. Thus, they argue for more effective regulation and for bringing into balance the influence of consumers and providers. This would bring prices down and allow private hospitals to enter the market more effectively. In a review of the urban medical insurance system, Gu and Zhang (2006, p. 71) note that public managers are not familiar with the idea of the government purchasing services for social functions from private providers.

challenges in the health system / 101

This last point is important and points to the more general problem of the preference for state engagement over effective use of the market. Although the impact of market forces has been extensive, China’s leaders still do not trust the market to work effectively and they have a natural predisposition for the state to play a dominant role. However, as we have seen, most institutions are still government-managed, but they are failing in their capacity to deliver good public health to the majority of China’s citizens. China risks giving its citizens the worse of both worlds: the rigidity and inefficiency of the public sector combined with a poorly regulated market that offers incentives that underprovide public health for most. Adopting a more plural and effective system still remains a major challenge.

This page intentionally left blank

Ch a p t e r Fi v e D e a l i ng w i t h P u bl ic H e a lt h Cr ise s a n d Pa n de m ic s

The Chinese public health system during the Mao years was not only the first to create a near universal coverage system for rural dwellers but also an extremely successful force in controlling and reducing substantially communicable and infectious diseases. The system’s ability to combine public health programs with mobilization campaigns without regard to cost proved very successful in reducing significantly endemic diseases such as schistosomiasis and cholera. The social control and public espousal of sexual puritanism that were features of CCP rule post-1949 also led to a dramatic reduction in the extensive commercial sex trade that was a feature of pre-1949 China, especially in a city such as Shanghai. However, since economic reforms began, the government’s capacity to deal with such public health challenges has changed significantly. The breakdown of collective life has made it more difficult to conduct public education programs and to mobilize citizens in campaigns to eradicate disease vectors. These changes have led to a resurgence of certain diseases such as schistosomiasis. Social mores have changed and with people more mobile than at any time since 1949, there has been a revival of extensive commercial sex work and an increase in sexually transmitted diseases (STDs) and the emergence of HIV/AIDS as a threat. China has been very successful in dealing with communicable and epidemic diseases, which by 1998 accounted for only 5 percent of fatalities, compared with 23 percent in India (Wolf et al.: 2003, p. 45). Although tuberculosis still claims around one-quarter of a million lives per year, given its level of economic development, China has dealt well with such health problems.1 The key areas of health strategy such as public education campaigns and immunization programs have been affected by the health sector reforms discussed in Chapter Four. Continued progress in these areas relies on better community organization and government

104 / providing public goods in transitional china

engagement. Progress also requires effective intersectoral collaboration and an extensive surveillance system that can monitor disease outbreak and spread. Morbidity and mortality rates for infectious diseases dropped consistently up until 1979. For example, malaria morbidity dropped 71.55 percent between 1965 and 1979, and mortality by 66.67 percent; typhoid and paratyphoid fever dropped by 33.9 and 66.7 percent respectively; and poliomyelitis by 86 percent and 87.5 percent respectively. The last attests to the effectiveness of immunization campaigns (Anson and Sun: 2005, p. 11). Data from the Shandong Department of Health in 1992 (Public Health in China: 2005) show that immunization rates fell by 36 percent from 1979 to the mid-1980s, and Gautam (1993, quoted in Anson and Sun: 2005, p. 18) notes that with the introduction of fee-for service, immunization coverage declined from close to full coverage to 60 percent for urban children and to 33 percent for rural children. However, renewed attention and special programs have increased the immunization rate. The immunization rate for measles, for example, rose from 78 percent in 1985 to 97 percent in 1990 and has now stabilized around 95–96 percent (Anson and Sun: 2005, p. 70). Ministry of Health statistics confirm this, with immunization rates for the most common incidences for children at around 98 percent in 2006. Under reforms, progress in disease control has stagnated and in some cases has begun to reverse (table 5.1). The reported incidence of infectious diseases declined rapidly in the 1980s, from 872.33 per 100,000 in 1985 to 292.22 in 1990 and continued to decline to a low of 167.05 in 1996. Since then the incidence rate has increased, with a significant jump to 244.66 in 2004. The death rate per 100 patients was 0.22 in 2004, almost the same as in 1985 (0.23). Although diseases such as viral hepatitis and pulmonary tuberculosis remain the most prevalent (table 5.2), worryingly, sexually transmitted diseases have been increasing, from 12.32 per 100,000 in 1989 to 50.66 per 100,000 in 1998, with syphilis increasing from 0.17 to 4.31 (Chen et al.: 2000) and to 13.30 in 2006 (an annual increase of 30.7 percent, State Council AIDS Working Committee Office: 2007, p. 8). In this chapter, we look at the policy challenges raised by two different diseases that reveal the “system’s strengths” and weaknesses in dealing with communicable diseases. First, we look at how the authorities dealt with the Severe Acute Respiratory Syndrome (SARS) that originated in China and spread worldwide. SARS brought home to the leaders the weaknesses in the health system, especially in the rural areas, which are discussed in Chapter Four. This has prompted them to improve the disease surveillance

public health crises and pandemics / 105 Table 5.1 Reported Incidence and Death Rate of Infectious Diseases for Selected Years Year

Incidence Per 100,000

1985 1990 1995 1996 2000 2003 2004

872.33 292.22 176.24 167.05 197.63 192.18 244.66

Death Rate Per 100,000

Death Rate Per 100 Patients

2.00 1.15 0.34 0.34 0.26 0.48 0.55

0.23 0.40 0.19 0.21 0.14 0.25 0.22

Source: Ministry of Health, at http://www.moh.gov.cn/open/2007tjts/P43.htm.

Table 5.2

Incidence Rate for Selected Contagious Diseases, 2006

Disease

Incidence Rate Per 100,000

Viral Hepatitis Pulmonary Tuberculosis Dysentery Syphilis Gonorrhea Malaria AIDS

102.09 86.23 32.36 12.80 12.14 4.60 0.51

Death Rate Per 100,000 0.10 0.26 0.01 0.01 0.00 0.00 0.10

Death Rate Per 100 Patients 0.10 0.30 0.03 0.05 0.00 0.06 19.95

Source: Ministry of Health, at http://www.moh.gov.cn/open/2007tjts/P43.htm.

system, which is all the more important with the potential for the spread of avian flu (Saich: 2006). The spread of SARS also made the leadership aware of the potential economic impact that an outbreak of a major disease can have. Second, we look at the policy response to HIV/AIDS that has spread from discrete groups to the population at large. The development of policy to deal with the spread of HIV/AIDS reveals the progress that can be made once the importance of a particular issue is appreciated by the central leadership. However, it reveals the weakness of policy implementation by localities and how the government acting alone lacks the necessary capacity for an effective response. We conclude by drawing some general lessons about how effective the government is in providing the necessary infrastructure for public health and by looking at possible policy options.

106 / providing public goods in transitional china

Combating SARS: A Wake Up Call for China’s Health System It must have come as a surprise to General Secretary Hu Jintao and Premier Wen Jiabao that their first major policy challenge came not from the economic realm but rather from the outbreak of SARS. Although some writers expressed optimism that the struggle to contain SARS might lead to significant changes, once Hu and Wen won their “war on SARS,” dramatic change looked less likely and “old politics” began to reassert itself (this section is based on Saich: 2006). The CCP began to extol its virtues in taming the viral beast. The traditional propaganda system soon found its footing to portray the struggle in terms of patriotism and turned out little ditties, such as “Angels in White Coats,” that paid homage to the nurses and doctors of China who, moved by love of the party and concern for the people, worked tirelessly to help the people. While this suggests that the effects of SARS on the system as a whole were minimal at best, it underestimates the impact that SARS had on thinking about healthcare and more generally about investment in the countryside. As we shall see in the following section, SARS had a significant effect in pushing forward policy on HIV/AIDS and in thinking about emergency response more generally. The fact that it took from November 2002 until mid-April 2003 before the Chinese leadership took decisive action against SARS reveals certain problems with China’s administrative and reporting systems. Obviously, attention was deflected by the leadership transition that was taking place during the same period. But also, the prevailing obsession with social stability and economic development meant that there was no incentive to release information about the disease for fear that it might cause panic or slow down economic growth through reduced consumption or investor flight. In fact, the poor handling of information flows led to panic buying as well as rioting and the destruction of quarantine centers. The delay in response revealed strong bureaucratic disincentives to effective reporting. The Ministry of Health is a weak player institutionally and at the local levels it is subordinate to party organs. Thus, if local party leaders receive and transmit messages to downplay the disease, there is little that health officials can do. In addition, at the time, there was a lack of clarity about to whom to report and under what circumstances. In accordance with Chinese law, epidemics fall under the classification of state secrets; the localities do not have the power to make public comments about disease outbreaks before they have been announced by national-level authorities. There is no incentive for local officials to report bad news before they have a clear signal to do so. This encourages optimistic reporting and

public health crises and pandemics / 107

the suppression of bad news. As discussed below, the interplay of these factors led to a delay in timely reporting and exacerbated the impact of the disease, causing precisely the kind of domestic and international crisis that the system wishes to prevent. Although local researchers in Guangdong province denied that there had been unnecessary delays and that it is very difficult to detect a new virus, politics did intervene to ensure that knowledge was not shared widely enough. Throughout the spread of the disease, authorities at each level sought to deny that there was a problem until events forced them to admit otherwise. Even then, the first response was usually to assure the public that while there might have been a problem, this was no longer the case. It seems certain that Guangdong authorities knew about the disease by early January 2003 (for details, see Saich: 2006), but this news was suppressed, mainly to ensure that the Chinese New Year holiday at the end of the month would not be disrupted and consumer spending would not be curtailed. The limited press coverage that was allowed was designed to ridicule the idea that there was anything untoward happening and to make sure that there would be no public panic. The Southern Daily claimed that police had ordered operators of leading Web sites to carry only positive reports on dealing with the disease (FEER, April 10, 2003). Trying to keep a lid on things, on February 14, 2003, Guangdong party secretary Zhang Dejiang ordered provincial authorities to educate the public to “voluntarily uphold social stability, not believe in rumors, not spread rumors” and to focus on the objective of the Sixteenth Party Congress to build China into a moderately well-off society (xiaokang shehui). While the provincial authorities were trying to calm public anxieties, it is clear that by early February officials in Beijing knew of the situation in Guangdong and the Ministry of Health (MOH) even sent a team headed by a vice minister to investigate. However, the Guangdong pattern of initial denial and then rejection of the disease’s importance continued until April 20, 2003. The slow response may have derived from underestimation of its impact and the thought that it was localized in the south. However, even if more information had became available, no senior official was in a position to undertake action as they were preparing for the National People’s Congress (NPC) meeting that was due to open on March 5. With the leadership transition about to be completed, media control was tighter than usual and any negative reporting would have been inconceivable. Thus, on March 9 officials from the MOH met with the heads of major Beijing hospitals to inform them about SARS but told them that this was not for public distribution and certainly should not be repeated to the media (Pomfret, Washington Post, May 13, 2003).

108 / providing public goods in transitional china

This delay because of the convocation of the NPC was fatal in terms of preventing the spread of the disease in the Chinese capital. On March 15, three days before the NPC closed, the World Health Organization issued its first global warning about SARS but the Propaganda Department instructed Chinese media not to report it. After the congress, things began to change but it took international pressure and domestic dissent to force China’s leaders to act. It is likely that without the pressure from international organizations such as the WHO and the postponement or cancellation of high-profile events such as the World Economic Forum China Business Summit and the Rolling Stones concert, the leadership would not have changed its approach of assuring in public that all was under control while seeking internally to cut off the spread of the virus. Finally, it took the actions of a Chinese doctor to release information about the severity of the situation in Beijing to prompt action. Even then, it was not until mid-April that policy really began to change. The new sense of urgency and the notion that a new approach had to be undertaken were confirmed at a key Politburo meeting on April 17. Effectively the meeting acknowledged that China had been lying and called on officials to report periodically to the public (daily reporting had begun on April 1), not to delay reporting, and not to cover up the situation. It also called for greater international and local cooperation and the exchange of experiences on limiting the disease. It linked combating SARS to the policy line of the party congress, calling on each party committee and government organization to recognize the extreme importance of SARS work. A new task force was set up under the Beijing Municipal Government to bring about better cooperation between the civilian and the military authorities. It was acknowledged that the fragmented jurisdiction over medical facilities in the capital meant that accurate information had not been collected. Unlike at Chinese New Year, it was now recognized that it was not a good idea to have tens of millions of people crossing the country for the seven-day May Day vacation, so it was cut to one day. Supervision teams were to be sent out to investigate the situation in key provinces and special concern was raised to prevent the disease from spreading into rural areas where the health system was liable to collapse. A special medical aid fund, with costs shared by central and lower-level governments, was set up to provide subsidies for those unable to cover costs. Subsequently, it was revealed that the minister of health and the mayor of Beijing had been dismissed to take ministerial responsibility for failures. Vice Premier Wu Yi was placed in charge of the Ministry of Health and on April 23 a central command center headed by Wu was established with a budget of 2 billion yuan (SCMP, April 24, 2003).

public health crises and pandemics / 109

Following this, a number of high-profile press conferences were given that admitted shortcomings and that pledged to make information more freely available. The energetic approach of Hu and Wen and the new appointments ushered in a brief period of unexpected cooperation and openness of reporting about SARS in the media. This not only helped warn the public of the dangers and allowed them to take sensible precautions, but also indicated to other officials that they had to take SARS work seriously or face punishment. Importantly, for considering disease control in the future, this period of openness was short-lived and soon “politics as usual” was reestablished. The traditional propaganda system came to grips with the new demands. While the new campaign was geared to mobilizing people’s support for the struggle against SARS, it was accompanied by more familiar restrictions on reporting. Patriotic slogans were devised and traditional CCP language now took over. Although the language of Leninism is that of struggle, the language of the CCP is that of war, and sure enough on May 1, Hu Jintao declared that indeed China was engaged in a “people’s war” against SARS for which the “masses should be mobilized” (Xinhua News Agency, May 1, 2003). This recourse to traditional propaganda techniques and the tools of mass mobilization disillusioned many who thought that SARS might herald real change but they were effective in restoring party morale and dealing with the epidemic. Constraints on the media were soon put back in place and the limited openness around SARS did not spread to other areas. In fact, by mid-June the Propaganda Department was clearly worried that the relaxation might go too far and moved to censure some publications and called for all publications to stop reporting on sensitive topics (Pomfret, Washington Post, June 20, 2003). This included writing stories critical of how Guangdong handled SARS or reporting about the Chinese doctor who had brought the government cover-up to international and domestic attention. By the end of June, media outlets and academics had been warned not to analyze how the government had dealt with SARS. This attitude was in marked contrast with mid-May when a senior official from the Beijing Municipal Party Committee announced that “media have the right to expose any cover-up or false report on the epidemic situation” (Zhongguo xinwenshe, May 20, 2003, translated in FBIS-CHI-2003–0520). This description of the politics of SARS is important as it reveals how policymakers are likely to respond when confronted by some kind of epidemiological crisis and it provides a number of general lessons for thinking about how Chinese leaders might deal with such occurrences in the future. After outlining the situation concerning HIV/AIDS in China, we shall draw together some of these general lessons.

110 / providing public goods in transitional china

The Challenge of HIV/AIDS The heightened awareness that SARS created about the weaknesses in the public health system and their potentially destabilizing effects was used by both domestic and international advocates to encourage China’s leaders to pay more attention to the threats presented by the potential spread of HIV/AIDS. A major breakthrough occurred on World AIDS’ Day in December 2003 when Premier Wen Jiabao and Vice Premier Wu Yi (both newly appointed), who had led the policy response to SARS, visited people living with HIV/AIDS at the Ditan Hospital in Beijing and announced the “Four Frees and One Care” policy.2 Since then, a number of policy initiatives have been launched and China’s most senior leaders have kept up the public visits, the most recent being the general secretary’s meet and greet with HIV/AIDS patients on the 2007 World AIDS Day (People’s Daily, December 1, 2007).3 It is clear that since Wen’s visit, more serious policy attention has been paid to the threats posed by HIV/AIDS and the shift in central policy has been remarkable. Policy has moved from benign neglect to official acceptance and promotion of most international best practices. This is just as well as although infection rates are still low, cases have been reported in all provinces and recent trends show that the disease is developing a profile similar to that in other countries. Those infected are no longer in specific geographic regions and rather are members of the general population. The central government has recognized this fact and an institutional framework for dealing with HIV/AIDS has been set up and a number of policy initiatives have been promoted. However, implementation remains problematic as does promotion of the kind of multisectoral responses and engagement of civil society that have proved successful in dealing with the disease in other countries such as Thailand. Further, social trends such as the large influx of migrants and the more relaxed social mores make education, monitoring, and evaluation more difficult. Last but not least, it remains to be seen whether policy will be implemented responsibly by local governments. As stated in Chapter Three, the incentive system for local governments works against adequate resource allocations for preventive medical care and the promotion of health education. At first glance, the figures for HIV/AIDS infections in China do not look alarming, certainly not when compared with Sub-Saharan Africa. In fact, the 2005 MOH, UNAIDS, and WHO joint study (MOH: 2006) lowered the previous official estimate by 22 percent, from 840,000 to 650,000 infected based on claims of a more effective data gathering system.4 The study released in December 2007 made a slight upward adjustment to 700,000, with 80,000 said to be living with AIDS, and

public health crises and pandemics / 111

fewer than 100,000 cumulative deaths (State Council AIDS Working Committee Office: 2007). These revised figures have been met with skepticism by some in the HIV/AIDS community, and other estimates run up to 1.5 million. Thus, although widespread, the infection has a low prevalence (0.05 percent), with concentration in a number of provinces. However, given the recent increased policy attention, the two reports and their sponsors were at pains to point out that this was no cause for complacency as there had been around 50,000 new infections in the first nine months of 2007 and 20,000 deaths. By late October 2007, a total of 223,501 HIV/AIDS cases had been reported, up from 183,733 by October 31, 2006 (Reuters, September 9, 2007). The number of new cases reported in the first half of 2007 was equal to the number of cases reported in the whole of 2006. Cases have now been reported in all provinces and the trends confirm the spread beyond specific communities. The disease developed in three distinct categories that have begun to merge. The first is the category of intravenous drug users (IDU), who were initially concentrated in western provinces such as Yunnan and Xinjiang; the second is the category of infections among the former blood and plasma donors (FBD) in central provinces such as Henan; and the third is the category of transmissions through sexual activity with a higher concentration of commercial sex workers (CSW) in southern China (for more details on this, see Thompson: 2005). It is now clear that not only is transmission moving between IDUs and CSWs, with many drug users engaging in commercial sex to pay for drugs, but also spreading to the broader sexually active population. Assertive action now will help control rates closer to levels in Thailand (2–3 percent) rather than in South Africa (20 percent). The lower rate would still mean 25–30 million infections; Chinese officials want to keep the number of infections under 1.5 million by 2010 (State Council: 2006). Problematically, China has a number of features that are conducive to the spread of HIV/AIDS. One of the adverse impacts of the familyplanning policy as we saw in Chapter Three is the surplus of males and a population that is increasingly mobile and that will become more urbanized over the next decade. Also, Chinese people have low levels of awareness about the disease and how it is spread. For example, according to a 2004 survey conducted by the China Center for Disease Control and Prevention (CDC), over 80 percent of men who have sex with men (MSM) were “totally ignorant” about the risk of contracting HIV and “mistakenly believe they are safe from HIV/AIDS” (Kaiser Network: 2004). It should be kept in mind that in China, most MSM are married and their partners are unaware of their MSM activities. There is a rising

112 / providing public goods in transitional china

CSW community with an attendant rise in sexually transmitted diseases and low rate of condom use (see Yuan Yue: 2003). Condom use has been rising but it is still low. The rate for CSWs who never used condoms decreased from 37.4 percent in 2001 to 7.5 percent in 2006 and the rate of female CSWs who had used a condom every time during commercial sex during the previous month had risen from 14.7 percent to 41.4 percent in 2006. This still leaves 60 percent not using condoms regularly. Among MSMs, only 30 percent used a condom for anal sex and only 50 percent used one when engaged in commercial homosexual sex (State Council AIDS Working Committee: 2007, pp. 20, 8). This suggests that the potential for rapid spread exists and also accounts for why some believe that China’s official figures for HIV/AIDS underestimate the extent of the disease in China. Although the surveillance may have been improved, there remain low levels of knowledge about the disease within the health community and an unwillingness of individuals to admit having the infection because of social stigma. One survey reported that 40 percent of respondents shunned those living with HIV/ AIDS (State Council AIDS Working Committee: 2007, p. 9). The situation in central China is indicative of the confusion around the numbers. The United Nations suggested that there were 55,000 commercial and plasma donors infected with HIV, whereas AIDS expert Zhang Ke reported that in 2004 the figure for Henan, the key province in central China, could be more than 170,000. Dr. Zhang estimated that a further 130,000 in Henan were infected from hospital transfusions (Economist, January 18, 2007). These higher claims for FBD are countered by the new assessments, and at the end of 2006 official figures claimed a total of only 35,232 reported HIV cases (Xinhua News Agency, July 10, 2007). Whatever the truth of the figures, the infection rate is increasing sharply and the profile of the disease is beginning to look more like that in other parts of the world. In 2007, for the first time the number of total infections among the 700,000 HIV positives derived from heterosexual activity exceeded the number of intravenous drug users: 40.6 percent as opposed to 38.1 percent. The number of those infected from homosexual behavior was calculated at 11 percent. The number of IDU infections in the total has been falling progressively from 72.6 percent in the period from March 1985 to 2000 (Zhang Konglai: 2002, p. 306) and 44.3 percent in the 2005 survey. One survey found that 50.8 percent of drug addicts continued to share needles (Xinhua News Agency, November 24, 2006), while the 2007 survey estimates the number at 40 percent. Almost 90 percent of the current IDU infections are concentrated in seven provinces, with over 10,000 infections in each.5 The figures also show the flow through the statistics of the infections of former blood and plasma

public health crises and pandemics / 113

donors that were discovered in Henan and elsewhere in the mid-1990s at 10.7 percent of the total in 2005, up from 9.4 percent in 2003 (Han Mengjie: 2004) and at only 0.5 percent during the 1985–2000 period (Zhang Konglai: 2002, p. 306).6 The percentage had dropped back to 9.3 percent in 2007. There were no reported new infections in this category in 2005 and 2007. As late as 2004, the MOH stated that it had closed down 150 illegal blood collection and supply agencies across the country (Kaiser Network: 2007). From January 1, 2008, all blood products were to be screened for HIV and other blood-borne diseases (Xinhua News Agency, September 12, 2007). Importantly, in China more males are infected than females, the reverse of the situation in many other countries, but this is also beginning to change. Although Zhang Konglai (Zhang: 2002, p. 307) found a ratio of 5:1, the ratio has dropped rapidly to 2:1. As of the end of 2007, women accounted for 30.8 percent of all reported HIV/AIDS cases, up from 19.4 percent in 2000 and comprising over half of the infections contracted through sexual transmission (Xinhua News Agency, June 5, 2007; State Council AIDS Working Committee Office: 2007, p. 4). The ratio was as high as 9:1 in the early 1990s (Jolly and Wang: 2003, p. 4).7 The infection rate is highest among those aged 20–39—with 70 percent in 2007. Those in the 15–20 age range claimed 5.9 percent of the infections (Ru: 2006). This argues for a stronger education program for the young. The prevalence among IDUs has risen from 1.95 percent in 1996 to 6.48 percent in 2004 (Ministry of Health: 2005, p. 4). Not surprisingly, the number of IDUs is uncertain. Public Security statistics mention a population of 1 million, but others suggest a population of 5 to 6 million. Injection of drugs will remain the dominant mode of transmission in northwest and southwest China. A 2003 UNAIDS report estimated that in certain areas of Sichuan and Guangxi, HIV prevalence is 50 and 43 percent respectively, whereas in parts of Yunnan and Xinjiang it is reported to be 80 percent (Ministry of Health and UN Theme Groups: 2003, p. 10). Yili in Xinjiang is said to have a prevalence rate of 89 percent among IDUs (Thompson: 2005, p. 70). For the provinces of Xinjiang, Yunnan, and Sichuan, the overall rate was said to be over 50 percent in the 2005 Ministry of Health joint survey with the United Nations Programme on HIV/AIDS and WHO (MOH: 2006). Drug use is linked with commercial sex work, with clear evidence that female IDUs are more likely to engage in commercial sex to support their habit. In a survey of Sichuan, nearly 60 percent of female IDUs reported selling sex for money or drugs and less than 30 percent reported consistently using a condom with customers (Choi et al.: 2006). Choi and his

114 / providing public goods in transitional china Table 5.3 Source of HIV Infections (percent) Transmission Mode

2007 Survey

2005 Survey

2003

Zhang Konglai Infected in (1985–2000) 2005

Intravenous Drug Use Former Blood and Plasma Donors Sexual Transmission CSW and Clients Partners of HIV Positive Individuals MSM Transmission Mother-to-Child Transmission

38.1 9.3

44.3 10.7

60.0 9.4

72.6 0.5

48.6 0.0

51.6 — —

43.6 (19.6) (16.7)

8.4 — —

6.8 — —

49.8 — —

11.0 1.0

(7.3) 1.4

— —

— —

— 1.6

Notes: CSW—Commercial Sex Workers MSM—Men who have Sex with Men Sources: 2005 survey, MOH, UNAIDS, and WHO: 2006. 2003 figures from Han: 2004. 1985–2000 figures from Zhang Konglai: 2002.

colleagues found that the HIV infection rate among IDUs in Sichuan increased at an annual growth rate of 97.23 percent between 1991 and 2003 (table 5.3). With the expansion of the sex trade in China there are anywhere between 1 and 6 million commercial sex workers, with one estimate at 20 million (Gill et al.: 2007, p. 14). As noted, although improving, condom use is low. The CSWs most likely to use condoms are those working in high-class hotels (56.7 percent), but this figure drops to 29.9 percent for those working in bath houses, massage parlors, and barber shops and then to only 15 to 20 percent for those working in establishments ranked below that (Parish and Pan: 2006, pp. 195–97). HIV prevalence among the CSW community has increased from 0.02 percent in 1996 to 0.93 percent in 2004 (Ministry of Health: 2006, p. 4). Again the rates vary significantly. In Shenzhen, it is said to be between 1 and 3 percent (interview with scholar, January 2004); whereas in some parts of Yunnan, Chongqing, Hunan, Guangdong, Guangxi, and Sichuan, the rate is said to be over 1 percent (MOH, UNAIDS, and WHO: 2006, p. 4). The highest documented prevalence rates among CSWs are estimated to be 10 percent in Guangxi, 5 percent in Yunnan, and 3 percent in Guangdong (Kaufman and Meyers: 2006, p. 51). Although migrants have received much attention from policymakers, evidence suggests that it is the more educated and more affluent who are most likely to visit

public health crises and pandemics / 115

commercial sex workers. According to Pan Suiming (2001, p. 5), the wealthiest 5 percent of men are 33 times more likely to visit commercial sex workers than the poorest 40 percent of men. Also, managers, factory owners, and businessmen are ten times more likely to visit commercial sex workers than urban manual workers and 22 times more likely than male laborers in rural areas. This group, now referred to as mobile men with money, is the most likely bridge to bringing the disease to the general population. STIs are increasing rapidly, with an officially acknowledged count of 703,001 in 2001, up from 5,800 in 1986; among this category, the infection rate has risen 55-fold from 0.02 percent in 1995 to 1.1 percent in 2000 (Han Mengjie: 2004; Gill et al.: 2007, p. 16). However, some estimate that the figure for STIs might represent only as little as 5 to 10 percent of the actual total as many of those infected prefer to visit private clinics because of the potential ramifications of visiting a public clinic (Thompson: 2005, p. 9). A 2001 survey alarmingly found that of 800 prostitutes interviewed in Sichuan and Yunnan, two in five had never had a health check-up, and over half did not consult doctors when they contracted sexually transmitted diseases (Yuan Yue: 2003). These statistics table 5.3 provide the background for understanding the potential for the spread of HIV/AIDS and they have forced the government into action. The general lessons that have been drawn will be covered in the following section. Here we shall mention some of the most important specific policy initiatives that relate to dealing with HIV/AIDS. Greater leadership commitment to improving the disadvantaged lot and the weaknesses in the health system have led to increased funding for dealing with HIV/AIDS. A number of important financial measures have been undertaken. The new action plan introduced in 2001 raised the financial commitment from the Chinese government from 10 million yuan to 100 million yuan, with a further increase to 810 million yuan in 2004 (State Council AIDS Working Committee Office: 2004, pp. ii, 22).8 At the end of 2005, the amount of funding available for the next two years was raised to 1.5 billion yuan. This funding was to cover activities such as the free antiretroviral therapy for rural and urban poor patients and the continual expansion of the China CARES program. However, there needs to be both better oversight to ensure that local governments will spend the funds responsibly as intended and that incentives are in place so that local governments will match funds where necessary. In addition, to improve the safety of the blood banks, the government provided over 2 billion yuan (Zhang Fujie et al.: 2006). The Chinese government has also been able to rely on substantial support,

116 / providing public goods in transitional china

estimated at 2.2 billion yuan, from the international community (MOH, UNAIDS, and WHO: 2006, p. 10). Between 2004 and 2009, round three of the Global Fund is committed to provide $99 million, with a target of supporting 40,000 patients with antiretroviral (ARV) therapy for the China CARES counties in the central provinces of Hebei, Henan, Shanxi, Shandong, Shaanxi, and Anhui. Round four will provide $64 million in support for the IDU- and CSW-driven epidemics in Yunnan, Xinjiang, Guangxi, Sichuan, Guizhou, Hunan, and Jiangxi. This will cover some 40,000 to 50,000 patients with ARV over a five-year period. Round six of the Global Fund was to provide $14 million for the support of civil society organizations in dealing with infected populations and improving public awareness. As discussed below, this last round of funding has proved difficult to implement because of the problems in the development of civil society in China. These financial commitments are, of course, welcome, but by themselves they will not be enough unless incentives and other behavior change. For example, although the treatment may be free, the testing to determine whether a person has HIV/AIDS is in many cases not free, and local governments and hospitals can use this as a way to generate funds to cover their associated costs. There have been many cases of hospitals withholding relevant information from the patients and charging them for unnecessary services or for drugs that should have been free (Goodman, Washington Post, November 8, 2005, p. A01). During SARS, this was a problem when the government announced free treatment for the poor, but local governments charged fees for testing an1d various other treatments. The government first guaranteed free drugs and tests for HIV/AIDS patients in early 2004, but by June 2006 only 20,000 people were on antiretrovirals and six provinces had not even started free drug provision (Watts: 2006). The free HIV testing was expanded from 365 counties in 15 provinces in 2002 to over 2,300 counties and 3,037 sites across all provinces in 2006. The number of screening laboratories has expanded to 5,500, with 99 able to conduct confirmatory HIV tests (Wu et al.: 2007, p. 584). Costs for the testing labs are to be shared between the central and the local governments, but most of the costs for maintenance and other recurrent costs are the responsibility of the local government (Liu and Kaufman: 2006). Other best practices that have been adopted include needle exchange programs and the use of methadone (for details, see State Council: 2006; Wu et al.: 2007). By 2006, some 729 needle exchange stations had been established in 204 counties or districts in 17 provinces; by October 2007, 49,108 IDUs were said to have joined a clean needle exchange program (State Council AIDS Working Committee Office: 2007, p. 22).

public health crises and pandemics / 117

Problems exist with diagnosis since many local governments are still not effectively equipped to determine the disease properly. In fact, in Beijing, Henan, Anhui, Sichuan, and Yunnan, there are only 700 provincial- and county-level doctors trained to diagnose and treat HIV/AIDS (Liu and Kaufman: 2006, p. 81). This lack of professional personnel is compounded by ignorance among the general population about the disease and by the stigma attached to it. Much more extensive public health education is necessary for both the general population and for specific at-risk populations. A good example is the program of Snow Lotus, an independent volunteer group based in Urumqi, Xinjiang, which has launched the Sunrise Club with support from the Urumqi CDC. The club is promoting AIDS awareness and prevention at Xinjiang Agricultural University and is preparing a two-month program of outreach activities on campuses (China Development Brief, November 2, 2005). In addition to interventions that will help with awareness and prevention, the reporting system too needs to be improved further. Currently, the surveillance system is weak among the general population and little use is made of behavioral social science research that could provide good guidance to policymakers. If China’s top policymakers do not direct more attention to these areas, there is the risk that the recent policy initiatives and extra funding will not be able to make their full impact felt. Apart from the general concern about the weakness of the rural health system and the lessons learned from dealing with SARS, one other factor has contributed to the decisive policy shift. The Chinese applications for funding from the Global Fund to Fight AIDS, Tuberculosis, and Malaria were important in bringing about policy change (the following draws on Szlezak: 2005). The framework for applying to the Global Fund and the implementation of the funded programs opened up a political space for alternative voices to be heard in the decision-making process. In particular, the negotiations within China included not only government agencies but also multilateral organizations and both Chinese and international NGOs as members of the China Country Coordinating Mechanism. Most importantly, all participants have equal voting rights and have to sign off on the agreement.9 The policy shifts noted above were aided by the rejection of the first round application. Rejection caused shock waves among the relevant government circles as many felt that China was too big and too important to be ignored and would not have to take seriously the application guidelines from the international community (discussion with government officials familiar with the drafting process). The first round application focused on prevention policies for IDUs and CSWs in

118 / providing public goods in transitional china

west China, but the Global Fund felt that it paid inadequate attention to reduction programs (such as needle exchange), none of which were operating in China. As a result, in the second and third round applications, the Chinese approach changed focus to free treatment for FBDs who were seen as victims of poor practices in central China. This switch made it easier to reach a consensus. With the successful application of round three, in round four the focus moved back to IDUs and CSWs and incorporated a set of best practices that had been rejected in the initial application. However, the implementation of round six highlights the limits to which the Chinese government is capable of developing a broad engagement with civil society to deal with the challenge of HIV/AIDS. The funding for this round was meant to support the work of NGOs and community-based organizations with an NGO as the principal recipient of the funding. This has not happened and the ambitious plans presented in the proposal have not been implemented and have been reversed. The main problem is the dominance of the system by government-organized NGOs (GONGOs) and the inability of small community-based organizations to register legally. The requirement that subrecipients of funds should be legally registered tax-exempt bodies has made it all but impossible for any of the organizations in China to receive the funds. The problem of NGO registration is dealt with in more detail in Chapter Eight. The Chinese authorities decided that a GONGO, the China HIV/AIDS Association that is headed by a former senior government official, should be the principal recipient; the association, in turn, intended that its provincial branches should be the subrecipients. However, the Global Fund decided that the association did not have sufficient capacity and thus for the first two years the principal recipient would be the Chinese CDC. Not only is the CDC classified by the Global Fund as a government entity but also it was the recipient of the previous rounds of funding. The China HIV/AIDS Association was defined as the main subrecipient with ten of its provincial branches also listed as subrecipients together with two other GONGOS. None of the proposed recipients are a community-based organization or represent people living with HIV/AIDS (interviews with those involved in round six drafting; Gill et al.: 2007, pp. 18–19). The difficulty to register for NGOS, combined with the allure of funding from an international organization for quasi-official agencies, thwarted the efforts to bring civil society into playing a much greater role in dealing with HIV/ AIDS. This statist bias exists in other areas too and prevents effective state-society collaborations that have been important and successful in other countries.

public health crises and pandemics / 119

Improving the System for Dealing with Pandemics China’s experience in dealing with SARS and its ongoing policy evolution for HIV/AIDS provide some important indicators about how the public health system might respond to future challenges such as avian flu or other pandemics. The responses outlined above reveal the current advantages of the politico-administrative system for dealing with and preparing for crises and the barriers that delay or prevent a fully effective response. These weaknesses include China’s sometimes poor understanding of its global responsibilities, the general weakness of the health surveillance and reporting system, and the poor incentives for local governments to report crises promptly and effectively. Perhaps the most important lesson concerns the level of attention that the central leadership pays to an issue and the clarity of the signal that it sends to local leaders that a policy must be taken seriously. Without a free media or a civil society that can mobilize to advocate policy change and represent marginalized groups, the system is dependent on either bureaucratic interests pushing for change or individual senior leaders becoming aware of a particular problem and deciding to champion it. Given the incentive structures, local leaders are sensitive to clear signals emanating from the Center, yet they have many ways to evade central directives unless they are sure of the seriousness and the consistency with which the leadership takes on a particular issue. This was clear, for example, in former general secretary Jiang Zemin’s commitment to crush the influence of the spiritual movement Falungong, but it is less clear in the case of his dealing with HIV/AIDS. Certainly, the central leadership has promulgated a number of important regulations and set up a new institutional framework for HIV/AIDS policy, and top leaders have kept up their high-profile visits to HIV/ AIDS patients. But these visits have not been reciprocated by local leaders. This is in marked contrast to the time when Jiang Zemin publicly administered polio vaccine to children, and provincial leaders quickly followed his example. Local leaders have not publicly embraced HIV/AIDS patients as Hu Jintao and Wen Jiabao have (see China Development Brief, May 2005, p. 16). Following SARS, the central government vowed to punish local officials who cover up flu outbreaks. This was reinforced in the State Council’s emergency response guidelines adopted in January 2006 that called for emergencies or incidents to be reported to the State Council within four hours and for the public to be informed in a timely, accurate fashion (http://english.peopledaily.com.cn/200601/09/ eng20060109_233888.html). Yet it remains unclear how high a priority

120 / providing public goods in transitional china

the guidelines are for the Center in terms of enforcement, what the mechanisms are for punishment, and whether this will override the incentives for local governments to cover up an outbreak. Chapter Three describes how poorly aligned the incentives for local officials are for investing resources, time, and energy into social development in general and public health in particular. The examples of dealing with SARS and HIV/AIDS raise other general concerns with which China’s policymakers must deal. First, these kinds of epidemics have global consequences that have been poorly perceived by China’s leaders, especially those at the local level. With SARS and HIV/AIDS, it took both domestic pressure and international criticism before the Chinese leadership moved to act. Without this external pressure, China’s local leaders seemed content to try to bury the problem and to punish those domestic critics who tried to bring the problem to the attention of the global community. Government officials at each level undertook action only when they were pressured to do so. The appointment of Hong Kong’s Dr. Margaret Chan to head the WHO in January 2007 might provide an opportunity for closer cooperation between the Chinese authorities and the global agencies that are trying to coordinate a response to avian flu and other future global diseases. On the other hand, bureaucratic interests in China may override international commitments. It may also be the case that Chinese officials are unwilling to release information that they feel might embarrass Dr. Chan early on in her tenure. The international community has consistently held different views not only about SARS but also about the seriousness of the spread of HIV/ AIDS in China. This has impacted on the question of how to deal with avian flu. Dr. Julie Hall, an infectious disease expert in the WHO office in Beijing, criticized the Ministry of Agriculture (MOA) for not sharing samples of a newly discovered strain of avian flu. In her words, “unless the ministry tells us what’s going on and shares viruses on a regular basis, we will be doing diagnostics on strains that are old.” Naturally, this was denied by the ministry that is also careful to refute any suggestions that China might be the source of outbreaks outside of its borders. The ministry blamed the United States for the fact that China had not yet shared the avian flu virus, claiming that the U.S. labs had not completed the necessary import procedures. The WHO expert felt that the MOA had not yet come around to putting global interests ahead of its own. When the Chinese authorities regained their composure after SARS and began to reject the notion that they had been at fault, officials in Guangdong even suggested that the disease had not originated in China but had been present in the United States in February 2002 (Lam, at CNN.com, May 13, 2003). Following its experience with SARS and HIV/AIDS,

public health crises and pandemics / 121

the MOH has become much more globally aware and has worked more effectively with the international community. This is not the case with the MOA that is more domestically oriented and has far less international exposure. It controls access to the samples from poultry and, as noted above, has not been forthcoming. Second, China needs to improve the infrastructure at all levels that deals with crisis and information management. The initial response to crisis is denial and cover-up and once action is called for, the vertical and segmented structure of China’s bureaucracy hamper effective action. For example, leaders in Guangdong initially did not feel any urgency to provide information about SARS to Beijing, thus undermining the capacity to act effectively. Neither did the military hospitals think that it was their responsibility to inform the civilian authorities. CDC head Li Liming was even reported to have said, “If we controlled the military hospitals at the beginning, we never would have had this epidemic in Beijing” (Pomfret, in Washington Post, April 27, 2003). We see the same problems with avian flu. Although the MOH is charged with looking after the health of the public, the MOA is concerned with the farmers and their livelihoods. The MOH deals with disease spread among humans, the MOA that among animals, but neither is responsible for transmission of disease from animals to humans. According to Zhong (2007, pp. 92–93), initially both ministries shirked responsibility and became motivated only once funding was increased and which they then proceeded to fight over. It was only in 2005 that the two ministries set up a cooperative structure to develop a joint approach, including a reporting system and a joint inspection team. In the Chinese system, it is notoriously difficult to gather information across different sectors. A system that encourages cross-sectoral collaboration needs to be developed to provide comprehensive, integrated solutions. All too often different ministries or localities work in their own respective interests and undermine national policy. China often sets up ad hoc bodies that make decisions that cannot then be enforced or are subverted by agencies at the same level or at lower levels. This affects both policy coordination and policy implementation. With respect to disease outbreak, it is important to keep in mind that the MOH is an institutionally weak player and, for example, party secretaries of major provinces or municipalities such as Guangdong or Beijing would outrank the MOH minister. Furthermore, the MOA is not only new to this area but institutionally weak as well. Provincial health departments and local disease control centers have to report first to the local party authorities. A number of steps have been taken to improve the situation. In midMay 2003, China announced that it would set up an Emergency Response

122 / providing public goods in transitional china

Bureau under the State Council to act as a powerful new agency to deal with future health crises and natural disasters (SCMP, May 13, 2003, Internet edition). The bureau is modeled on the U.S. Federal Emergency Management Administration that was established in 1979 and was absorbed into the Department of Homeland Security in March 2003. Tasked with drawing up plans for dealing with future emergencies, membership of the bureau, which was put into operation in December 2005, comprises the heads of key ministries and commissions. With respect to HIV/AIDS, in February 2004, a new State Council AIDS Working Committee was set up to improve coordination. Headed by Vice Premier Wu Yi, the committee includes vice ministers of 23 key ministries and mass organizations, together with the vice governors of China’s worstaffected provinces. These kinds of organizations should help improve the information flow across the various bureaucratic institutions. However, to date the AIDS Working Committee has seldom met and its true effectiveness is thus questionable. One further key step was the promulgation on May 12, 2003 of temporary regulations for dealing with health emergencies. Overseen by Premier Wen Jiabao, these regulations were drafted in a record 16 days and approved on May 9. The six chapters and 54 articles deal with outbreaks of infectious diseases, large-scale food poisoning episodes, and other serious public health threats. They lay out general rules and guidelines for prevention and reporting. For example, the guidelines call for national authorities to be informed within four hours of a major outbreak, and provincial governments within eight hours, something reiterated for all disasters in the January 2006 emergency response guidelines. The guidelines also call for setting up Centers for Disease Control and Prevention all the way from the central government down to the county, something that adds to the fiscal burden of local governments.10 Such administrative structures help if they are backed up by strong political support and clearer guidelines for effective reporting. In November 2005, the outbreak of avian flu led the State Council to issue emergency regulations to respond to major animal epidemics. The regulations call on veterinary authorities at various levels to draft contingency plans and to outline processes for reporting. Although the regulations state that any attempt to delay or fail to report an outbreak will be severely dealt with, they also state that only “competent veterinary authorities under the State Council” can release information on major animal epidemics. They call for coordination with the forestry authorities when dealing with animal infections and for coordination with health authorities when an outbreak is likely to infect humans (People’s Daily online, at http://english.people. com.cn/20511/22/eng20051122.223094.html).

public health crises and pandemics / 123

These suggestions are reasonable, but the incentives for the various agencies are not necessarily aligned. Post-SARS, conflicting interests concerning trade in wild animals threatened to undermine any attempt to control future outbreaks of the disease. On June 3, 2003, Guangzhou municipal authorities announced that they wished to move toward centralizing the slaughter of poultry to improve hygiene in wet markets (SCMP, June 4, 2003). This followed earlier directives in Guangzhou and Shenzhen to stop the sale and consumption of wild animals. Fines ranged up to 10,000 yuan. It is thought that the most likely explanation for the origin of the disease was that it crossed from animal species to humans in southern China, the civet cat being seen as the most likely suspect. However, at the time this had not been proved conclusively and agricultural authorities and the forestry bureau objected to the ban. The forestry bureau is involved in direct regulation of the rearing and sale of wildlife, and forestry officials saw the development of this trade as a way to raise farmers’ incomes—a primary objective of the government. They were reported to have tried to lift the June ban, arguing that there was no proven link that justified its continuation (FEER, August 21, 2003). Of course, in the case of an outbreak of an infectious disease, it will be the health authorities who receive the blame and not the forestry officials. Under these circumstances, it is difficult to apply a coherent policy. The situation for reporting on major disease outbreaks remains confusing with respect to what to report, when, and to whom. Thus, more effective reporting flows must be developed. One issue much debated after the outbreak of SARS was the need to create a better system of information management. There is a fundamental tension between a political system structured to control and manage information flows and a society that is information savvy and “wired.” Such an informational asymmetry is dysfunctional when the official media report that all is under control and the online international news media tell a different story. China’s leaders claim that they have provided sufficient information about avian flu, but major incentives persist for local leaders to deny the problem or to underreport. To be fair, the underreporting is enhanced by the general weakness of the rural medical system and the poor levels of training that may mean that cases are not diagnosed properly. Chinese estimates of HIV/AIDS infections are consistently lower than international estimates, and studies of particular at-risk groups suggest higher rates than the official statistics. For example, in Yunnan, based on data of HIV prevalence and the population size of subgroups, there were an estimated 80,000 people with HIV/AIDS at the end of 2003, as opposed to official figures of 14,905.11 The death totals were around 10,000 rather

124 / providing public goods in transitional china

than the official figure of 558 (Yunnan Center for Disease Control and Prevention: 2004, pp. 3, 12). The confusion in 2005 and 2006 about the situation with respect to avian flu (H5NI) does not bode well for transparency and clarity in reporting. There was also an initial delay and secrecy surrounding the outbreak of pig-borne bacterium streptococcus in Sichuan that was reported to have killed more than 35 people (http://news.bbc. co.uk/2/hi/asia-pacific/4742319.stm). Senior leaders have recognized the problems with local government incentives and have called on local officials to pay more attention to social justice and sustainable development. Stressing that HIV/ AIDS must be taken seriously, State Council Document No. 7 (March 2004) announced that government leaders would be held accountable for their work with respect to HIV/AIDS and that this would be taken into account in their job performance assessments (State Council AIDS Working Committee Office: 2004, p. 10). This step will help if it is implemented effectively. As noted, local authorities may not report the outbreak of diseases before they are given the green light by the national-level authorities, thus encouraging the suppression of bad news. When journalists asked local Guangdong health official Huang Qingdao why they had not reported the outbreak of SARS earlier, he replied, “atypical pneumonia is not a disease we’re legally required to report, so we didn’t feel it was necessary to make it public. Now, because it has had a big social impact, we decided to make it public” (FEER, April 10, 2003). This is a point worth considering as it helps explain the predicament of local officials should they wish to take action on new diseases. According to the 1996 implementing regulations, SARS would have fallen under the category of a highest-level secret (jia lei) as it was an infectious disease that covered a wide area. As a result, it could not be disclosed until the MOH or those organs authorized by the ministry made the disclosure. This means that without ministry disclosure, any local official talking publicly about the disease is liable for prosecution. This is a disincentive for transparency, to say the least. The tight control on the media and on civil society organizations makes it all the more difficult for problems and cover-ups by local officials to be brought to wider public attention. These problems point to the need to create a new system of information management. The weakness of civil society and the suspicion with which local authorities often view NGOs undermine China’s capacity to deal with HIV/AIDS. Many of those infected or at risk are from groups that are considered to be engaged in illegal activities (prostitution or drug use) or are considered immoral (homosexuals). Although policy has shifted significantly, these

public health crises and pandemics / 125

groups are still viewed at best with suspicion and at worst as criminals. As a result, it is very difficult for government and party agencies to reach out to them. This is something that could be performed more effectively by NGOs and grassroots support groups. Public health education and care for those afflicted with diseases such as HIV/AIDS would be helped by if there were to be a greater involvement of the community and nongovernmental organizations. In countries such as Thailand that have adopted effective strategies to confront the spread of HIV/AIDS, government has forged strong partnerships with nongovernment organizations and has mobilized the resources of civil society and community-based organizations. There is evidence of increased NGO activity with respect to HIV/AIDS work in China, but senior CCP leaders remain ambivalent about the development of the NGO sector and groups such as People Living with HIV/AIDS are not involved effectively in policymaking. This ambivalence was best seen in the summer of 2007. With the party congress due to be held in October and the summer Olympics in 2008, the work atmosphere for NGOs in areas that had formally been tolerated (health and environment) became much tougher and came under more scrutiny. Yet, in July 2007, the vice minister of the MOH, Wang Longde, noted that grassroots organizations were more able to reach the vulnerable groups than government agencies. He described these organizations as “an indispensable force in the war against the deadly disease [HIV/AIDS]” (China Daily, July 18, 2007). This message was reinforced by Peter Piot, the head of UNAIDS, in a September speech to the World Economic Forum in Dalian when he stressed the need for the government to give rein to NGOs and to enroll the support of companies (SCMP, September 11, 2007). Also in July 2007, in Henan, a province notorious for its repression of AIDS activists and for its attempts to cover up the real situation, an AIDS Prevention and Treatment Association was established as an NGO to help those infected with HIV/AIDS. Its members included HIV carriers as well as people living with HIV/AIDS, lawyers, teachers, journalists, and entrepreneurs (Xinhua News Agency, July 10, 2007). Both domestic and international NGOs as well as government-backed NGOs are taking on important roles in dealing with HIV/AIDS. A directory of NGOs published in June 2005 lists 101 civil society organizations and projects working in the areas of HIV/AIDS prevention, treatment, and advocacy (China Development Brief, July 2005, p. 1). These range from well-established groups such as the China Association of STD/ AIDS Prevention and Control to small groups providing help and care for children or working with “same sex love groups.” In Yunnan province,

126 / providing public goods in transitional china

traditional mass organizations such as the Women’s Federation have been very active and the provincial Red Cross has conducted peer education among young people, organizing 1,500 training courses for some 30,000 people (Yunnan Center for Disease Prevention and Control: 2004, p. 26). In addition, international NGOs such as Save the Children UK have been very active in Yunnan, Henan, and Anhui, working with AIDS orphans. Together with the Ministry of Health, the Clinton Foundation launched a service fellowship to place clinical experts in areas affected by HIV/AIDS that includes one-month training in the United States (China Development Brief, November 2, 2005). People Living with HIV/AIDS have established a number of self-support groups in Xinjiang, Shaanxi, Shanxi, Guizhou, Shanghai, Beijing, Guangdong, and Sichuan (State Council AIDS Working Committee Office: 2004, p. 13). This review of China’s treatment of contagious diseases reveals both the strengths and the weaknesses of the current system. Still, if mobilized, China’s apparatus can be a very powerful force that can organize resources to deal with major health challenges. However, there are significant incentives and institutional barriers to recognizing the emergence of health challenges that need to be overcome. This is an area where the role of the government needs to be clear and where incentives need to be aligned to make sure that institutions at all levels pursue the public good.

Ch a p t e r Si x Bu i ldi ng a Ne w Soc i a l I nsu r a nc e Syst e m

The restructuring of the social security system clearly reveals the shift under reforms from a system that was based on the workplace to one that is based on government provision, with a significant and increasing role for the individual, family, and market. In urban China, the main concern of China’s leaders has been how to move from a system of enterprise-based welfare to one where the government is the main provider and upholds minimum support levels. This reform has involved dismantling the hierarchy and privileges that existed within the planned economy and the smoothing out of the inequalities that existed between the workplaces under the plan and those outside of the system. A major challenge exists to develop incentives and mechanisms to include the private sector and take the decision of whether to integrate the large migrant populations into the urban system. If the answer is yes, which services should be provided and how? The result of this is a diversification of service providers in urban China, with a reluctant acceptance of commercial insurance, a demand for greater cost recovery rather than highly subsidized provision and for increasing space for nongovernmental organizations to operate. In rural China, the effective collapse of the collective rural structures has meant the disappearance of formerly available support networks outside of the family and a struggle to think about new mechanisms under which basic guarantees can be provided to rural dwellers. The structural adjustment programs promoted by international financial institutions and the World Bank in the 1970s and 1980s came at high social cost in those countries that adopted them (Stiglitz: 2002). In response to the high costs, the World Bank launched a new poverty agenda (1990) that tried to balance its prior emphasis on economic growth and fiscal prudence with the provision of basic public services and safety nets to protect the most vulnerable. The necessity for such support was

128 / providing public goods in transitional china

reinforced when the Asian Financial Crisis and the remedies that were suggested to provide a cure exposed the weakness of social protection in a number of countries that had appeared economically successful (Cook et al.: 2003). China has experienced similar costs during economic transition, and World Bank thinking has influenced its reforms. The experiences of other countries provided lessons for the Chinese leadership to think about as they underwent the deepening of reform beginning in the mid-1990s. Shedding 50 to 60 million jobs in the state sector would clearly leave many vulnerable unless better safety nets were designed to complement the expanding social insurance programs. However, such programs by definition are seen as short-term solutions to longer-term problems. As Cook notes (2003, p. 71), such programs are “designed to absorb the shock of entitlement collapse (loss of employment); to deliver relief rather than development; support short-term consumption rather than reduce long-term poverty or vulnerability and to deal with symptoms rather than causes.” Chapters Six and Seven look at the new architecture that the government has been putting in place for social security. In the 1980s, especially in the rural areas, many were left without the minimum support that had been guaranteed through the collectives. In the urban areas, as reform began in the latter part of the 1980s and with the gathering pace through the 1990s, many found themselves without the cradle-to-grave support that they had come to expect from the workplace. These new vulnerabilities heightened the fear of unrest, reinforced by the demonstrations of 1989, and led to extensive experiments to put in place a new system that would knit the fragmented pockets of support into an integrated whole. The experiments included attempts to bring excluded sectors of the urban population into a unified structure, but it still left those in the informal sector of the economy and the rural areas lacking a truly effective support system. Social protection systems provide a cushion against not only the shocks of transition but also against temporary isolated shocks to household stability, such as unemployment, sickness, and against longer-term impacts such as disability. For the long-term unemployed and the elderly without other means of support, they provide a safety net with minimum guarantees. The World Bank, among others, classifies social protection into two types of interventions (2006, pp. 149–50). The first comprises contributory schemes (social insurance) that have the task of managing risks by smoothing an individual’s income over time when confronted by specific difficulties. This would include pensions, disability allowances, and health and unemployment insurance benefits. The second is transfers from general tax revenues (social assistance) and includes fiscal

building a new social insurance system / 129

redistributions from the wealthier to the poorer and cash-in-kind programs. This aid might be supported by specific labor regulations to meet government objectives, such as minimum wage regulations, child labor laws, and mandatory retirement ages. China operates both systems. This chapter covers social insurance, with a particular emphasis on pensions as this is exemplary. For urban China, where there has been a separate pension system with higher benefits for party and state officials and government employees, we focus on the revamped pension system for workers and staff in SOEs and on the attempts to integrate other workers into one system. Chapter Seven deals with traditional poverty relief and social assistance. Not surprisingly, spending on the former exceeds the latter. According to the World Bank, spending in East Asia on social insurance amounts to around 3.9 percent of GDP, whereas spending on social assistance is around 1.8 percent. In OECD countries, such spending is 13 percent and 2.4 percent respectively (World Bank: 2006, p. 149). This chapter looks first at social protection in urban China before looking at the development of a restructured pension system, followed by a review of attempts to build a new rural system. Social Protection and Pension Reform in Urban China As we have seen in Chapter Two, before economic reform began, social protection was centered on the workplace, a system that worked relatively well when there was no informal sector to speak of and when most people were employed through government agencies or state-owned enterprises. A community-based relief scheme was set up to deal with those who were not covered through the workplace system, and special benefits and work opportunities were provided for those who suffered from physical disabilities. The move to a more market-influenced economy revealed the high costs that welfare provision had placed on the SOEs and the increasing incapacity of this sector to bear these costs. Research by the Labor and Social Security Research Institute found that by 1998, 58 percent of the total SOE payroll comprised social security expenses, compared with 19 percent in collective enterprises and 18 percent in private enterprises (in Nielsen et al.: 2005, p. 1762). In addition, the existing institutional structure inhibited the further development of a labor market. Labor mobility was highly restricted by the fact that pensions, medical care, and, most importantly, housing belonged to the workplace. With benefits not portable, life-long employment was reinforced and an effective labor market was naturally discouraged. One of the most important days of the year was when enterprises announced the distribution of new

130 / providing public goods in transitional china

housing units or reallocations for the following year. People would time their weddings to coincide with this distribution, and others would rush back from overseas trips so as not to be left out. Many families adopted the strategy of “one family, two systems”: one member would work in the state sector to ensure maximum benefits while the other would work in the private sector where financial rewards were higher but there was no housing or other significant benefits. Not surprisingly, the system was prey to ever-increasing demands. Although the workers did not contribute directly to benefits and paid little in rent, demands for increased benefits were continual bargaining points in the workplace. Before stricter financial discipline was imposed, it was relatively easy for enterprise leaders to cede to this in return for industrial peace (Walder: 1986). Yet this also produced a culture of dependence in the workplace that proved relatively hard to break. But the system was changed when tighter fiscal discipline was imposed by market forces. Although there was reform experimentation beginning in the mid-1980s, it was in 1998 that Premier Zhu Rongji announced an end to the enterprisebased, cradle-to-grave care that the Chinese industrial working class and government employees had grown up with and had come to expect. Policy thrust was to ensure greater individual responsibility through contributions to pension, medical plans, and other insurance and the privatization of workplace housing stock. At the same time, the old “iron rice bowl” of permanent employment was smashed. Basically, the policy objective was threefold: first, to reduce the welfare burden on the SOEs and to redistribute it across enterprises and to individuals working in other ownership categories; second, to try to provide equal rights and levels of protection across all ownership categories; third, to establish linkage between the contributions an individual makes to his/her own benefits and what he/she actually receives (Nielsen et al.: 2005, p. 1763). Certainly the burden on enterprises has been reduced, but the burden has not been picked up fully by other sectors of the economy. Policymakers have tried to extend the benefits beyond the state sector but with only limited success. The new policies have caused spending on social security to increase rapidly as a percentage of GDP, but it remains low in comparative terms. In 2005, spending, including social insurance and social assistance, still amounted to only 3.3 percent of GDP, up from 1.9 percent in 1997 and 0.9 percent in 1990. This is well below the East Asian average of 5.7 percent. Of this figure, pensions comprise by far the largest category at 2.2 percent in 2005, up from 0.7 percent in 1989 (Salditt et al.: 2007, p. 23). Given future obligations, this percentage will have to rise significantly. A Rand study suggests that by 2025, government expenditures on pension, health, education, and interest payments on

building a new social insurance system / 131

government debt will have to rise to 15.5 percent from the 7.5 percent in 2000 (Crane et al.: 2005). In actual amounts, the rise from 1990 to 2004 represents a significant increase from 16.9 billion yuan to 495.1 billion yuan. Insurance costs amounted to 404 billion yuan in 2005; since 1989, revenue has been higher than expenditures, thus allowing a build-up of funds (Salditt et al.: 2007, p. 23). Table 6.1 outlines the kinds of social insurance schemes that China operates. There are five main insurance schemes for urban workers: pensions, healthcare, unemployment, industrial accident, and childbirth. The two largest items of social insurance are pensions and health insurance. In 2005, pensions made up 73 percent of the total insurance funds, and health insurance comprised 20 percent. All schemes are based on a mix of employer and employee contributions. Again, the most significant is the pension scheme that has a 20 percent payroll contribution from the employer and an 8 percent contribution from the individual employee. The industrial accident and childbirth schemes rely solely on employer contributions (0.5–2 percent and 1 percent respectively). This social insurance system is supplemented by the social assistance programs described in Chapter Seven. Unemployment insurance coverage has actually been dropping in recent years. In 1995, it covered 43 percent of employees, but it dropped to a low of 37 percent in 1998, before climbing again to 45 percent in 2000 and then undergoing a gradual decline to 40 percent in 2004 (China Labor Statistical Yearbook: 2004, p. 571). Not surprisingly, the coverage rates for industrial injury, a voluntary scheme, are much lower (table 6.2). The rate for industrial injury was 26 percent in 2004, up from 21 percent in 1998. By contrast, coverage for childbirth insurance has increased significantly since it was formalized. In 1997, it covered 43 percent of all eligible employees, but in 2004, it topped 100 percent at 108, indicating that some outside of the formal sector were also covered (Gu: 2006, p. 5). Medical insurance is dealt with in more detail in Chapter Four. Here we shall focus on the most important of the systems—pensions. The Urban Pension System China does not have a current pension problem, but unless it acts soon, demographics will turn unfavorable between 2010 and 2015 and a potential policy headache might develop. This has meant that in terms of policy priorities, the leadership has recognized that it has a problem, but not necessarily one of the most urgent to be solved. However, as discussed below, procrastination on pension reform will carry very high costs in a decade or so. This was recognized by the head of the People’s Bank of China in the

Table 6.1

China’s Urban Social Insurance Schemes

Social Insurance Scheme

Beneficiary

Entitlement

Pension

Those officially retired Each month 20 percent of average social wage and 1/120 of from urban enterprises accumulated amount in personal account

Health

Employer Contribution (Percent)

Employee Contribution (Percent)

20

8

All participating urban Pooled funds cover hospitalization and outpatient costs for certain employees major diseases; individual accounts cover general outpatient services—to a capped amount

6

2

Unemployment

Urban employees who To a maximum of 24 months dependent on period of contribution. have paid one full year, Lower than the minimum wage, higher than urban minimum are involuntarily unem- subsistence ployed, and are formally registered

2

1

Industrial Injury

Participating enterprise Medical costs, injury and disability subsidies, pension rights, and employees injury or disability nursing charge

0.5–2

0

Childbirth

Participating enterprise Medical costs for childbirth, monthly allowance during maternity employees leave

0.7–1

0

Source: White Paper on Labor and Social Security, 2002, at http://www.china.org.cn/english/2002/Apr/31774.htm.

Table 6.2 Participation in Different Urban Insurance Schemes Year

1995 1998 2000 2004 2005

Urban Employment (Millions)

Retirees Proportion to Working Population

Basic Pension Insurance

Basic Health Insurance

Industrial Injury Insurance

No. (Millions)

Coverage (%)

No. (Millions)

Coverage (%)

No. (Millions)

190.4 216.2 231.5 264.8 273.3

1:4.8 1:4.4 1:3.5 1:2.3 1:2.1

87.4 84.8 104.5 122.5 131.2

45.9 39.2 45.1 46.1 48.0

82.4 79.3 104.1 105.9 106.5

43.2 37.0 44.9 39.9 38.9

26.2 37.8 43.5 68.5 84.8

Source: China Labor Statistical Yearbook: 2006, pp. 8, 534, 543, 546, and passim. Figures for 1998 from Gu: 2006, pp. 5–7.

Coverage (%) 13.8 17.4 18.8 25.8 31.0

Maternity Insurance No. Coverage (Millions) (%) 15.0 27.8 30.0 43.8 54.1

25.5 59.0 68.0 103.7 125.1

134 / providing public goods in transitional china

comments he made in November 2006 when he called for the country to address the problem while it still had the financial buoyancy to do so. He wanted a policy that was “innovative, bold, and generous” and suggested that those with adequate savings should be encouraged to buy pensions and that those who could not afford them should be helped with loans. He also recognized the problem of low returns on pension investments—only 2–3 percent at a time when real wages were rising by about 10 percent. Consequently, he suggested that greater use be made of capital markets (Financial Times, November 21, 2006). Currently, China’s pension assets amount to only 5 to 7 percent of GDP, compared with 33 percent in South Korea and 50 percent in Japan. With the aging of the population, China’s pension assets will have to be raised significantly. The challenge for China is whether it can grow sufficiently rich before it grows old. During the Cultural Revolution, the enterprises inherited responsibility for the management and payment of pensions. This was because the trade unions, which had previously managed pensions, were dismantled. Current policy is developing a more integrated system out of the former fragmented parts, together with the introduction of market elements, albeit rather gingerly. Once reforms began to bite and many SOEs sank deeply into the red in the 1990s, enterprises began to run into problems with meeting their pension obligations, thus making reform necessary. Evasion of pension obligations became more common over time and compliance in many cities dropped from 90 percent at the beginning of the 1990s to 70 to 80 percent by 1995 (World Bank: 1997b, p. 2). The World Bank (1997b, p. 3) also calculated that total pension reserves in 1995 were less than 1 percent of GDP and the accumulated surplus by the end of 1996 was insufficient to cover pension obligations for even six months. Further, as we saw in Chapter Three, future demographics present significant challenges. In 1995, there were 10 workers for every pensioner, five among urban workers, but there would be only three by 2050, with some even suggesting that the ratio may be as low as 2:1 by 2040 (Jackson and Howe: 2004, p. 6). These shifting demographics are having a strong impact on contribution rates. Whereas they represented only 3 percent of payroll when China began its system in 1951, they rose to 20 percent by the mid-1990s. Unless something is done, by 2033 Chinese estimates suggest that contribution rates will rise to around 40 percent of payroll (Kennedy School of Government: 1999). The ratio of retirees varies across China. Major industrial cities such as Shanghai, where the family-planning program has been particularly successful in reducing the birth rate, are aging fast. The number of people over 60 was 18 percent in 1999, up from 3.6 percent in the

building a new social insurance system / 135

early 1950s, and was set to peak at around 30 percent in 2030 (SCMP, September 21, 1999, Internet edition). Industrial provinces such as Liaoning in the northeast also have a very high percentage of retirees. Whereas Shanghai has 5.72 percent of all retirees, it has only 1.36 percent of China’s total population. Liaoning has 7.5 percent of all retirees but only 3.23 percent of the population. The attempts to reform the enterprise pension system were drawn together in a State Council document of 1997 that was influenced not only by domestic experimentation but also by input from World Bank staff. Experimentation had begun over a decade earlier when the central leadership began to turn its attention away from rural reform to urban reconstruction. In 1978, regulations formalized the Cultural Revolution practice of placing responsibility for pensions with the workplace. The financial impact was dramatic, enhanced by the dramatic drop in the ratio of workers to retirees. In October 1984, the CCP Central Committee passed the “Decision on the Reform of the Economic Structure” that chronicled the problems that beset the urban sector and offered a more thoroughgoing reform than the piecemeal experimentation that had taken place previously. Considerable opposition to reforms remained and there was concern about social unrest, causing a slowdown of reforms in 1985–1986 and 1989. A core concern was to make SOEs more economically responsible. Removing the economic burden of social costs was a key element in this. Thus, in 1986 reforms were introduced to become effective on October 1 that contained the seeds of the 1997 nationally adopted system. The reforms attempted to provide a more secure funding stream and introduce the idea of individual responsibility by asking all new SOE employees to contribute up to 3 percent of their basic wage to social insurance agencies. This was to be supplemented by an employer contribution of 15 percent of the pre-tax wage bill. As with reforms in other areas, there was widespread experimentation and much local variation. Yet by the end of 1991, all counties and municipalities had set up their own social insurance agencies, covering approximately two-thirds of SOE workers, to oversee pension funds. The following year the experiments were extended to collective enterprises (Salditt et al.: 2007, p. 16). This change began to break the psychology of worker dependence on the workplace and laid the foundation for more experimentation through the 1990s. The basic framework for pension reform was developed further in the 1991 State Council “Resolution on the Reform of the Pension System for Enterprise Workers.” The resolution proposed bringing all workers in the state system into a uniform, three-pillar pension system. The first pillar provided a basic pension with contributions from the

136 / providing public goods in transitional china

worker, the enterprise, and the state. The second pillar was funded by the enterprise from its profits, and the third pillar was built on a voluntary account funded by the individual worker and payable as a lump sum on retirement. The various experiments were pulled together in the 1997 State Council Document Number 26 (“On the Establishment of a Unified Basic Old Age Insurance System for Enterprise Staff and Workers”). This set out the new system that was to be adopted nationwide by the year 2000. However, in 2000 there was still considerable variation and only in December was the system put into experimental effect in Liaoning province. The system called for the unification of public pillar benefits, the standardization of the size of individual contributions, and unified management of funds (Zhao and Xu: 1999, p. 1). The new pension system also constructed individual pensions from three pillars. The first pillar is a defined-benefit public pillar for redistribution. This is funded by a payroll tax of 20 percent drawn from pre-tax enterprise revenues and guarantees a replacement rate of 20 percent of the average wages at the time of retirement, if a minimum of 15 years of work is contributed. The second pillar is a mandatory-funded, defined contribution pillar for each worker. This is funded through a payroll tax of 11 percent, initially comprising both enterprise and individual contributions. Subsequently, this was amended to include only individual contributions and was reduced to 8 percent. On retirement, the worker is to receive a monthly payout that equals the account balance divided by 120, the factored annuity (State Council Documents 35 and 36: 2006). This assumes a life expectancy of 70 and a are made, then this pillar is expected to provide a replacement rate of 38.5 percent. The third pillar is a voluntary supplemental pillar managed by each enterprise separately or through an insurance company. The first two pillars are intended to provide a replacement rate of around 60 percent, the figure that the World Bank suggests as representing a realistic target. It aims to bring China in line with practice elsewhere in the world. Previously, Chinese pensions were very generous in relative terms by providing a replacement rate of 80 to 90 percent. This was already untenable for many enterprises with the pay-as-you-go system that predated reform. The only source for paying the pensions of retired workers was current operating funds. The lack of a unified system and the regional variance in contribution rates led some enterprises to withhold payment. For example, the Sichuan iron and steel producer Panzhihua joined the local pooling system in 1986 but in 1992 the iron and steel sector began its own pooling system (11 industrial sectors were allowed to form a pooling system) with lower rates. As a result, Panzhihua refused to pay the higher rates in the geographic pool and accumulated a debt to the local pension

building a new social insurance system / 137

fund of 300 million yuan (Zhao and Xu: 1999). This practice was quite common. A number of further steps were taken to provide the policy and institutional structure for the new system, and experimentation continued. First, in the March 1998 restructuring of the State Council, a new Ministry of Labor and Social Security was established to provide more effective coordination of social insurance and to oversee implementation. The creation of this ministry provided an administrative agency throughout the hierarchical system to oversee pension management. Although its focus is primarily urban, the new ministry took over responsibility for the limited rural pension program from the Ministry of Civil Affairs. Second, in August 1999, a new State Council document building on the previous reforms was issued. The 1997 document had called for pooling of pension accounts at the provincial level, but now it was specified that the provinces should have a unified contribution rate and a unified management of funds by the year 2000. It also sought to clarify who took what responsibilities at the local level. Most importantly, it confirmed that responsibility for the collection and distribution of pensions was removed from the enterprises themselves and handed over to municipal social insurance agencies. To ease the way toward provincial management, each province was to set up a “readjustment fund” to backstop pension obligations and to iron out inequalities among the various municipalities within the province. In a major disincentive for local officials, the document announced that any surplus would be disposed of at the provincial level and most of the money would be invested in central government bonds. Third, in January 1999, the State Council issued regulations to expand the contribution base to take into account the diversified urban economy. By the end of 1997, although 93.9 percent of SOEs were in social pension programs, only 53.8 percent of urban collectives and 32 percent of joint ventures, private enterprises, and the self-employed were in such schemes. Migrant workers were also to be brought into the program, and the 11 industrial sectors that had their own social pooling systems were to be folded into the provincial system. This expansion of the system was to be completed by the end of June 1999, increasing the number of workers participating in the pool by 26 million (31 percent) to reach a total of 110 million (Zhao and Xu: 1999). The rationale for this was obvious. First, if the nonstate sector could escape the payroll tax, then workers would have even fewer incentives to stay in the SOEs and would move to other sectors where their monthly costs would be lower. Second, the state needed increased participation to be able to cover the bills of the current retirees and the large numbers who would retire from the SOE sector in the coming years.

138 / providing public goods in transitional china

In making this transition, the government needed to deal with the thorny question of how and when to fund the individual accounts as part of the new pension plan. This meant that policy needed to strike a balance between the three goals of meeting the obligations of current pensioners, not placing too heavy a tax burden on the workforce, and minimizing the payment requirements for future generations. So, fourth, to act as government guarantor, in September 2000, the National Council for Social Security was set up to oversee the National Social Security Fund. The purpose of the fund is to provide a long-term strategic reserve that can cover future social security needs. However, it is unlikely that it can currently cover any major expenditures and it is not likely to be able to do so until after 2010 (Salditt et al.: 2007, p. 18). In 2000, Liaoning province was chosen as the pilot for what was initially intended to be a three-year experiment. The choice of Liaoning was not surprising given that the northeast provinces had suffered the most from the deindustrialization of the 1990s. Unemployment was far higher than the national average and very little new employment was being attracted to help those laid-off or forced into early retirement. The pilot was intended to realize the 1997 policy objectives by ensuring funding of the individual accounts and by expanding effective coverage and collection. Again benefits were to be managed independent of the workplace (Xinhua News Agency, July 8, 2001). According to the then provincial governor, five million people had to pay the pensions of 6.88 million as well as support 2.68 million retired workers (SCMP, March 11, 2002). In 2004–2005, as a part of Premier Wen Jiabao’s policy to “revitalize the northeast,” the pilot was extended to Jilin and Heilongjiang. In 2004, the three northeast provinces received nearly one-quarter of the central government’s social security subsidies. By the end of 2005, individual accounts were said to be fully funded in Jilin and Heilongjiang, but this was with a huge financial subsidy from the central and local governments (Salditt et al.: 2007, p. 18). Despite the huge investment from the state to maintain this system, experiments were extended to an additional eight provinces in 2006.1 Problems with the Pension System Policy has been clarified dramatically in the last few years but implementation still lags and a number of serious problems continue to exist with current policy. First, encouraging a unified rate is one thing, but enforcing compliance is another. Most provinces have acquiesced in allowing the municipalities to set their own rates based on current obligations rather than risking a raise in the rate. Yet, in many instances, rates were already above the stipulated 20 percent, especially in the old industrial areas of the

building a new social insurance system / 139 Table 6.3 Social Security Subsidies, 1999 and 2005 Region

China Beijing Shanghai Chongqing Liaoning Jilin Heilongjiang Guizhou Sichuan

% of % of Ratio of Social Security Total SOE Employees to Subsidy Population Employees Retirees Per Capita (Yuan) 2005

2005

2005

100 1.2 1.4 2.1 3.2 2.1 2.9 2.9 6.3

100 2.9 1.9 1.9 4.7 2.8 4.6 2.3 4.9

3.0 2.4 1.9 1.9 2.3 2.5 2.4 2.6 2.4

1999

2005

27.5 27.2 30.0 36.3 117.7 77.9 109.6 15.6 19.0

120.9 163.2 405.4 154.1 391.0 386.2 287.5 68.0 98.0

Source: Salditt et al.: 2007, p. 30 and National Bureau of Statistics, at http://www.stats. gov.cn/tstj/indsj/2006/indexeh.htm.

northeast and Sichuan. For example, in Jiamusi (Heilongjiang province), the contribution rate was often above 33 percent of the total wage bill (Kennedy School of Government: 1999). As a result, in 2000, the State Council raised the contribution rate for enterprise payments to the social security fund to 30 percent but no longer required enterprises to pay into personal accounts, thus raising employee contributions to 11 percent, but later dropping them again to 8 percent. There is also variation in the level of central subsidy to the various provinces (table 6.3). For the most part, the subsidy reflects the distribution of SOEs from the pre-reform era that was concentrated in the northeast and inland provinces such as Sichuan and Guizhou. The inland provinces constituted an important part of the “Third Front” that was formed when Mao Zedong decided to relocate some heavy and nuclear-related industry in the 1960s following the Sino-Soviet split. Although the impact of the pilots in the northeast is evident in subsidy per capita, it is less remarkable in the inland provinces. There is also a lack of transparency with local budgets and there is no mechanism to hold the local administrations truly accountable. Also, as we saw in Chapter Three, many local governments lack adequate revenue in their formal budgets and are driven to seek off-budget revenues. In this context, large pension funds look very tempting. The decentralized functioning of the fiscal system has not helped. Pension funds can be major sources of investment funds for the localities and thus there has

140 / providing public goods in transitional china

been opposition to find ways to make it difficult to integrate pooling at a higher administrative level. A 1998 survey revealed that local governments had diverted over 10 billion yuan from social security funds to projects that had nothing to do with pensions. Since then, the situation has become worse. According to the head of the National Audit Office, 7.1 billion yuan of social security funds have been misappropriated, 4.78 billion yuan since 2000 and 2.34 billion yuan before then (SCMP, June 28, 2007). Corruption scandals concerning social security funds have become common. In the summer of 2006, they became entwined with elite party politics when the party secretary of Shanghai, Chen Liangyu, was dismissed from his position (Xinhua News Agency, September 26, 2006; Saich: 2007a). A social security fund scandal had been brewing in Shanghai for several years, earlier claiming the scalps of a number of lower officials. But it was only in the run-up to the Sixth Plenum of the Sixteenth Central Committee (October 2006) that General Secretary Hu Jintao felt it necessary to remove Chen, who was seen as a protégé of former General Secretary Jiang Zemin and as a key member of what is referred to as the “Shanghai Clique.” Chen’s position began to look tenuous when the head of the Shanghai Municipal Labor and Social Security Bureau was sacked in August for lending money from the fund to projects in which he had personal connections. The municipality’s social security fund was said to total 10 billion yuan, of which over 3 to 4 billion yuan was siphoned off into a variety of property projects and a toll road (SCMP, September 29, 2006). It was revealed that Shanghai officials had illegally funneled 6.32 billion yuan into the stock market between 2003 and 2006. This included 100 million yuan that had been diverted from the municipal social security fund (SCMP, June 21, 2007). Local governments have resisted more centralized control over the management of pension funds, and this scandal again led to suggestions that the central government should centralize the management of such locally run funds. The fact that the scandal broke in Shanghai does not bode well for other localities as Shanghai had been considered a relatively well-run program. In 2005, the city paid out about 35 billion yuan to retirees, with the objective of covering 98 percent of the population with “basic” social security by 2010 (SCMP, September 27, 2006). A major problem is that at the local levels the Bureau of Labor and Social Security is both the administrator and operator of the funds. There are no semiannual or annual reports on how the funds have been used or invested. Prior to the Shanghai scandal, misuse of pension funds had been reported in Henan, Heilongjiang, Hunan, Sichuan, and Zhejiang (China Daily, September 7, 2006). Many of those in a financially healthy pool see no incentive to being merged at a higher administrative level that would include a town with

building a new social insurance system / 141

high levels of unemployment and a weak financial base. There has been resistance to joining provincial pools and some towns have refused to contribute to the adjustment funds. If there is no alternative, the perverse incentives encourage the spending out of any surplus and even the running up of a deficit before joining. When 11 industrial-sector pools were merged with the provincial-level pools, industries moved to spend the surplus, raise benefits, and encourage earlier retirement. The same problems have arisen when county-level pools have been merged into provincial pools. In 1993, of all provinces, Guangdong had the largest reserve in provincial redistributive funds, but by the end of the decade all of the funds had been spent out. Enterprises have also been reluctant to contribute, even those enterprises that have the funds to do so. By the end of 1999, nonpayment by enterprises amounted to 38.3 billion yuan, with over 200 enterprises owing in excess of 10 million yuan (SCMP, December 16, 1999, Internet edition). Lack of payment has been the source of a number of demonstrations and unrest and the government has usually moved quickly to make up payments. At the end of 1999, a joint statement by the Ministry of Labor and Social Security, the State Economic and Trade Commission, the Ministry of Finance, the China Securities Regulatory Commission, and the State Industrial and Commercial Bureau announced new penalties for defaulters. These included a ban from listing on stock markets, setting up joint ventures and subsidiaries, establishing new branches, and expanding their scope of business. However, as with other admonitions, the impact has been limited, as evidenced in the statement by the minister of labor and social security in June 2000 that nonpayment would not be tolerated. He acknowledged that another 1.45 billion yuan had been added to the unpaid total in the first five months of the year and that payment delays had occurred in 19 provinces (Xinhua News Agency, June 22, 2000). The move to integrate the nonstate sector into the program has met with mixed results as there has been ambivalence not only by companies but also by local authorities about imposing what amounts to an extra 30 percent cost on these businesses through the payroll tax. A number of provinces have allowed cities to set lower rates for the private sector. In 1999, officials in Qingdao (Shandong province) lowered the contribution rate for private companies from 25 percent to 11 percent of payroll, with an additional 7 percent to be paid by the workers (Kennedy School of Government: 1999). The variations have continued. In Shanghai, the employer has to pay 27 percent of the previous year’s average wage of the workforce into the individual pension accounts, whereas in Shenzhen the company has to pay only 9 percent (Salditt et al.: 2007, p. 8). This is not

142 / providing public goods in transitional china

surprising as Shenzhen has thrived on the expansion of private industry and foreign-invested companies. Before the special economic zone was opened to experimentation with economic reform, it was a small village across the border from Hong Kong, replete with rice paddies and water buffaloes. Its leadership has consistently taken a pro-business position. By contrast, Shanghai has a significant legacy of heavy industry from before the economic reforms began. Before the corruption scandal was exposed, Shanghai was thought to have a relatively competent administration and had been able to generate sufficient revenues for its social security programs. It is in the process of issuing a social insurance card for all residents that will consolidate unemployment insurance, health insurance, and pension payments in a computerized system. The scheme is administered by the Bureau of Labor and Social Security and operates independent of all enterprises in the municipality. This scheme is more likely to be successful than the pilot scheme being implemented in the northeast, which will rely heavily on central government handouts. However, few provinces have the personnel resources and financial capacity of Shanghai. Progress to date then has been mixed. The major question for the future is the pension system’s long-term financial viability. In part, this depends on how high a priority the central leadership places on funding pension reform. The answer appears to be high, but not high enough given other pressing issues. Resolution will also be affected not only by increasing returns on investments from the social security fund, but also by improving coverage, raising the retirement age, and thinking about the level of benefits paid out. The scandals that have been exposed, with local authorities losing significant amounts of pension funds in bad investments, have raised the issue of the people’s trust in the government to manage their money effectively. As one senior official muttered to me at an international meeting on pension reform, “Well, would you trust the Chinese government with your money?” By 2006, the system was at least receiving more in revenues than it was paying out, but the demographics discussed in Chapter Three mean that this could easily change. In 2006, incoming pension funds amounted to 631 billion yuan with payments of 489.7 billion yuan. The accumulated funds amounted to 548.9 billion yuan (Xinhua News Agency, December 5, 2007). Despite this, perhaps the greatest problem of all is that the system remains one of pay-as-you-go, despite the new pillars that are to be constructed to provide greater long-term viability. “Real money” is not being accumulated as there is nothing in the individual accounts. All along, local governments had been drawing from the individual accounts and in 1995 the State Council actually

building a new social insurance system / 143

legislated that this practice was acceptable. This means that current pension obligations are still being financed out of current revenues, and financing for the future is not being accumulated. Thus, to all intents and purposes, the individual accounts are purely notional and as these people begin to retire, they will have to have their pension obligations covered out of general revenues. By about 2025, the system could be overwhelmed as contributions from the contemporary workforce will not meet pension obligations. There is also the tendency to set contribution rates at a low level in order to reduce future pension obligations; if rates of return are lower than the opportunity cost of the capital, evasion and noncompliance will follow (Zhao and Xu: 1999). While we know that the implicit pension debt has risen, no one knows for sure its amount. Throughout the 1990s, pension expenditures as a percentage of GDP tripled from 0.8 to 2.4 percent (Frazier: 2006, p. 49). The Chinese government has been greatly concerned about social instability, and indeed one of the overriding mandates for local officials is to maintain social order at all costs. Unrest around the question of unpaid pensions has become a common phenomenon, so higher levels of government will often step in to bail out local authorities to payoff pension debts (Hurst and O’Brien: 2002). Between 1998 and 2001 alone, the central government transferred 117 billion yuan in subsidies (Frazier: 2006, p. 49), and per capita subsidies from the government have risen from 27.5 yuan in 1999 to 120.9 yuan in 2005. There are various estimates of the implicit pension debt. The World Bank (1997a) has calculated a range of 46 to 69 percent of GDP, while Wang et al. (2001) estimated it to be 71 percent in 2000 and Dorfman and Sin (2000) suggested a figure of 94 percent. Sin (2005) later revised this estimate to 141 percent of GDP in 2001. With the exception of Sin’s figures (2005), these estimates are actually low in comparative terms since the pension coverage is limited to a relatively small percentage of China’s total labor force; for example, in the United States, it is 113 percent. In fact, if the central government acts soon and adopts adequate measures, the fiscal costs should be manageable (Ma and Zhai: 2001). A number of suggestions for improvement have been made, but a greater use of the market is the most promising. First, as a short-term measure, budget expenditures have been increased. At the March 2003 NPC meeting, the finance minister announced a 38.6 percent increase in the social security budget to help those in difficult circumstances and to head off unrest. However, strong competing budgetary demands make this difficult to maintain over the long term. Second, there was an aborted attempt to consider the implicit debt part of the national debt and to raise funds through the sale of state assets, a program formally endorsed by the State Council in June 2001. It

144 / providing public goods in transitional china

was swiftly abandoned in October 2001 by which time share values had dropped by 1000 billion yuan. Current thinking is to expand a specific social security tax (now operating in 16 provinces), but a better long-term strategy would be to open up the pension funds to a market-driven approach and to move away from centralizing pension pool administrations and try to reduce evasion and noncompliance by administrative means (see Zhao and Xu: 1999). Of course, for the private sector to take over management of pension funds would not only require a psychological leap for China’s leaders, but would also demand that a proper regulatory infrastructure be put in place. The transition costs could be reduced considerably by creating a fully funded system, with the public pillar over time drawing on general fiscal reserves and the individual accounts managed by a private and decentralized system. There are some positive signs, and economic necessity will probably force a faster pace of market-driven change. As a result of a reported pension deficit of 41 billion yuan in 2000, the Ministry of Labor and Social Security, together with the Boshi Management Company, announced that perhaps as much as 15 percent of pension funds could be invested in stocks, with another 10 percent invested in treasury and corporate bonds (China Daily, May 25, 2001). This marked a significant shift away from allowing investment only in bank deposits and government bonds. Most important in this respect was the establishment in September 2000 of the National Council for Social Security to oversee the social security fund, with investment management delegated to qualified asset managers—with the exception of investment in bank deposits and the purchase of government bonds in primary markets. The fund is expected to cover the shortfalls in provincially managed pension funds, where returns between 2000 and 2004 were too low, with an aggregate yield of 11.82 percent, which was higher than inflation but lower than the 13.66 percent return on treasury bonds. This was also well below the real rate of income growth, which provided a further disincentive for individuals to invest in a pension fund. The need to improve on the 2 to 3 percent return has led to cautious experimentation. According to Xinhua News Agency (March 29, 2005), the fund yielded a better return of 3.1 percent in 2004, but this was still short of what is calculated as the required rate of return of 5 percent (Salditt et al.: 2007, pp. 42, 37). The international norm is around 6 percent. The improved return was aided by the move in early 2003 to change the way in which the fund could be invested. An initial fund of 30 billion yuan was released for investment in stocks, with the promise that more local funds would follow. This presented challenges for the investment skills of the fund managers and, as a result, fund management was divided into

building a new social insurance system / 145

two. The safe investments (bonds etc.) were kept in the hands of the fund, but investments in the stock market were taken over by 10 domestic fund managers, producing improved results. In 2005, the director-general of the National Council for Social Security announced that the then 230 billion yuan fund would achieve a yield of around 5 percent, some 2 percentage points higher than its previous best results. The “risky” investment portfolio comprised about 10 percent of the total social security fund. Building on the initial regulations for domestic fund managers, it was announced in June 2007 that the China Banking Regulatory Commission and the Ministry of Labor and Social Security (MOLSS) had drafted plans for pilot pension fund management companies within commercial banks. The banks (along with the foreign banks that are licensed to operate in China), however, unlike the insurance companies, lack custodial experience. One additional reform was the establishment of enterprise annuities. Enterprise annuities hold a great deal of potential for future development but they are currently held back by the lack of clear regulations combined with complex oversight by numerous government bodies, including the MOLSS, the China Banking Regulatory Commission, the China Securities Regulatory Commission, the China Insurance Regulatory Commission, and the State Administration of Taxation. The annuities will provide the framework for the second pillar of the three-pillar system. In August 2005, the MOLSS granted four types of licenses for the first batch of 37 companies to conduct enterprise annuity business. Over the short term, the introduction of enterprise annuities has heightened the inequality of pension provision as it has been the major SOEs in the power, energy, telecommunication, banking, and manufacturing sectors that have been providing annuities. As of May 2007, 24,000 companies had joined the scheme, with a total amount of about 100 billion yuan. At the end of 2005, the scheme covered 9.24 million employees in these enterprises (Salditt et al.: 2007, p. 41). There remains a significant potential for growth. Shanghai alone had 4.3 million enterprises in 2005, and the World Bank estimates that the annuities should grow to $1.8 trillion by the end of 2030. This would make China the third-largest scheme in the world, and this new pension market will generate at least $29 billion for trustees, custodians, and investment managers (SCMP, May 28, 2007, Internet edition). The lack of transparency has led to misuse of funds. In 2006, it was reported that 3.2 billion yuan had been misused. Such abuse has led to reforms in the management of the funds, with local governments instructed to turn them over to licensed financial institutions by the end of 2007. The Shenzhen Municipal Government was the first to have a single entity to manage enterprise annuities when it agreed to

146 / providing public goods in transitional china

transfer the city’s 2.3 billion yuan of assets to Ping An Insurance and China Merchants Bank. The largest amount of funds under management, 15 billion yuan, is in Shanghai. In April 2007, Shanghai set up a new pension management firm, Changjiang Pension Insurance, to manage the funds of its eleven major enterprises, including Bao Steel and the Shanghai International Group (SCMP, May 28, 2007, Internet edition). In perhaps the most significant step toward fund investment, in October 2006, the National Council for Social Security signed overseas investment partnerships with two global investment trustees: Northern Trust Corporation and Citigroup Inc. These agreements will allow the council to move its investments beyond low-yielding government bonds and wildly unpredictable Chinese stocks. The initial overseas investment in stocks and bonds was scheduled to be about 8 billion yuan. In total, the amount for overseas investment cannot exceed 20 percent of the total fund managed by the council. However, at the end of 2006, this represented a healthy 45 billion yuan. Raising coverage levels has remained a persistent problem. Obviously, this is a major issue in rural China, but increasing participation from the nonstate sector in the urban economy has also presented challenges. Although formal sector employees may well be enrolled in all five schemes, the self-employed can enroll only on a voluntary basis. There have also been attempts to extend some kind of social insurance to migrant workers who have been excluded from virtually all of the social insurance schemes. In June 2004, the MOLSS issued a circular calling for the participation of migrant workers in industrial injury insurance. Priority was to be given to those working in high-risk occupations, such as construction work and mines (Zhao: 2006, p. 7). A major disincentive for migrant workers is the lack of portability of the pension funds. If migrant workers move back to the rural areas or to a different administrative jurisdiction in search of work, it is highly unlikely that they will be able to take the savings (these, in any case, would probably be nominal at best) in their individual accounts or make a claim against their local pension funds. This problem, affecting the flexibility of the labor market, means that in old age the migrants will be reliant on their own savings and/or family support. As a result of these concerns, in June 2007, a vice minister of the MOLSS announced that a new pension plan had been drafted to try to incorporate the 140 million migrant population. Those with stable jobs were expected to join the pension plan where they worked; for the others, an individual account will be set up, the contents of which will be transferred to their hometowns each

building a new social insurance system / 147

year. Employers and employees will both contribute, with the employee contribution capped at 5 percent of the monthly salary. These individual accounts will follow the migrant from job to job. Self-employed migrants will also be allowed to join the scheme if they pay the premium (SCMP, June 12, 2007, Internet edition). This is an important step but, as in so many other areas, the main concern is implementation. Profit margins are low in the areas where migrants work and it is unlikely that employers will want to increase their business costs. Also, there is a major question as to whether there is an administrative capacity to manage such a complex system effectively. Last but not least, there is the question of trust. There are many stories of migrants being cheated out of their due wages and benefits, thus it may be difficult for them to trust the new system to deliver on its promises. By the end of 2006, 49.9 percent of the 283.1 million urban employees were contributing to the pension scheme (table 6.4). Despite all the regulations and publicity given to the new system since its introduction in 1997, this is a rise of only 7.5 percent; although in absolute terms, it is an increase of 75.3 million workers. As a percentage of the total urban population, the percentage was the same in 2005 as it was in 1995 (31 percent). We can see the shift between those contributing and those receiving, causing the need to increase contributions. In 1990, contributors were 84.3 percent of the combined total, but this dropped to 75.7 percent in 1998 when the percentage stabilized. Despite the rise in participant percentages, coverage for pension recipients for those over 60 years of age actually dropped 3.7 percent to 19.8 percent between 2000 and 2005 (Salditt et al.: 2007, p. 27). Thus full coverage is still a long way off. A 2000 survey by the China Research Center on Aging found that just over 70 percent of the urban elderly were covered by some form of pension and that for those who had a rural household registration, it was only 14.96 percent, as opposed to 77.39 percent for those with a nonrural registration. These statistics reveal the traditional bias of the system. In fact, if one includes the rural population into the statistics for 2005, only 13.4 percent of China’s population received a pension. There was also a regional variation, with 72.59 percent of the urban elderly covered in the eastern region of China, 67.51 percent in central China, and 66.66 percent in the west. Table 6.5 reveals the provincial variations. The high rate of coverage in Liaoning (96.3 percent) and the other two northeast provinces is indicative of the special attempts that have been made to run pilot schemes there. The relatively well-developed economy of Shanghai also facilitates high coverage rates (88.1 percent). There was a significant gender dimension to the coverage as well, with 84.7 percent of males among

Table 6.4 Basic Pension Coverage, 2005 Year

Urban Employees (millions)

No. of Contributors (millions)

Recipients (millions)

Contributors and Recipients

Contributors as % of Urban Population

Contributors as % of Urban Employees

1990

170.4

52.0

9.7

61.7

20.4

30.5

1995

190.4

87.4

22.4

109.8

31.2

45.9

1997

207.8

86.7

25.3

112.0

28.5

42.4

1998

216.2

84.8

27.3

112.0

26.9

39.2

2000

231.5

104.5

31.7

136.2

29.7

45.1

2005

273.3

131.2

43.7

174.9

31.1

48.0

2006

283.1

141.3

46.4

187.7

n/a

49.9

Source: China Labor Statistical Yearbook 2006, pp. 541, 542 and http://www.molss.gov.cn/gb/news/2007–05/18/content_178167.htm.

building a new social insurance system / 149 Table 6.5 Region

China Beijing Shanghai Chongqing Liaoning Jilin Heilongjiang Guizhou Sichuan

Pension Coverage in Selected Provinces, 2005 Total Urban Employment (millions) 273.3 7.36 6.12 3.54 8.65 4.14 6.77 2.82 8.02

Active Contributors (millions) 131.2 3.65 5.39 1.89 8.33 3.25 5.46 1.32 5.63

Contributors as % of Urban Employees 48.0 49.6 88.1 53.4 96.3 78.5 80.6 46.8 70.2

Source: China Statistical Yearbook 2006, pp. 13, 544.

the elderly receiving some form of pension, whereas the figure for females was only 56.48. In west China, only 46.42 percent of females had coverage as opposed to 66.66 percent of males. Our surveys between 2003 and 2005 found a similar situation with respect to coverage. Pensions had by far the largest enrollment rates among the various insurance schemes, ranging from 39.4 percent coverage in major cities to 22.2 percent coverage in smaller towns and townships where they were not compulsory. Not surprisingly, coverage was only 8.3 percent in the villages (table 6.6). Participation rates in almost all categories in 2005 fell sharply. Although only 33.2 percent of city residents had no insurance coverage, this figure rose to 78.6 percent in the villages. Whereas 14.2 percent of city residents had free medical services, the corresponding figure for rural areas was only 0.7 percent. The one category that was way out of line is maternity insurance. It is unclear why this should be the case from our survey. The only two categories where enrollment rates were higher for those living in small towns and townships than for those in the cities was workplace or joint workplace and individually purchased comprehensive life insurance. For the pension system to thrive it is important to get younger people covered to offset the numbers of those who will be retiring. In our survey, however, the numbers enrolled increase with age. Thus, whereas 13.5 percent in the 16–30 age group were enrolled, 20.5 percent in the 46–50 category were enrolled. However, we can see the impact of the newer insurance schemes where enrollment rates are higher for younger age groups. Thus, 9.1 percent of the younger age groups are enrolled in the unemployment insurance scheme, as opposed to 5.2 percent of the oldest

150 / providing public goods in transitional china Table 6.6

Insurance Coverage for Respondents in Our Survey, 2005 Total No.

Cities

Small Town/ Township

Village

%

No.

%

No.

%

No.

%

Basic Social Pension Basic Social Unemployment

721

17.5

383

39.4

123

22.2

215

8.3

282

6.8

158

16.2

42

7.6

82

3.2

Housing Accumulation Fund

243

5.9

185

19.0

49

9.0

9

0.4

Self-purchased Life Insurance

226

5.5

87

9.0

46

8.3

93

3.6

Retirement Fund Free Medical Service

227 203

5.5 4.9

176 138

18.1 14.2

40 48

7.2 8.6

11 18

0.4 0.7

Basic Social Industrial Injury

178

4.3

71

7.3

16

2.9

92

3.5

Workplace Comprehensive Life

149

3.6

63

6.5

41

7.5

45

1.7

78 1.9 52 1.3 212 5.1 2640 64.0

34 23 28 323

3.5 2.4 2.9 33.2

25 4 43 269

4.5 0.7 7.8 48.7

Workplace and Self Joint Comprehensive Life Basic Social Maternity Unclear None

19 0.7 25 1.0 141 5.4 2048 78.6

group; 6.8 percent in the 16–30 age group are insured for industrial injury as against 2.8 percent in the 46–60 age range. As one would expect, those with higher incomes in the urban areas are better insured. Only 35.6 percent of those in the highest income category have no insurance whatsoever, whereas 56.9 percent of those in the lowest income category do not have any coverage. Two other solutions that have been proposed to deal with China’s pension debt problem are to raise the retirement age and to adjust the replacement rate. Currently, the retirement age is extremely low at 50 years for women in production, 55 for women in offices, and 60 years for men. In practice, the retirement age is often even lower; the China Daily (November 29, 2006) has reported a retirement age of only 51.2 years. Life expectancy is now at 73.33 years for women and 69.63 for men, this leaves a long period of time to be covered by a pension system that is not designed for it. Around 20 years have to be covered at an almost 60 percent replacement rate. The retirement age was set low to deal with the problem of job creation for new entrants into the labor force and for those shaken out of the SOEs during

building a new social insurance system / 151

the 1990s. Frazier’s survey (2006) of Beijing and Shanghai suggests that there will be opposition to changing this policy to allow later retirement. Not only has there been opposition to raising the retirement age, but also to making the retirement age equal for men and women. In fact, in December 2006, a vice minister of the MOLSS ruled out any possibility of raising the pension age (People’s Daily online, December 22, 2006). But demographics and fiscal capacity will most likely force a change in such a categorical veto. Citizen Perception of the Pension System The use of market mechanisms and the notion of personal contributions as described above are a significant change for the way the pension system has operated. The president of the Development Research Center of the State Council described the introduction of private accounts as “in and of itself revolutionary” (discussion, June 2006). Pre-reform structures appear to influence how people think about who should be the insurance provider. Our survey in 2003 supports the work of others that shows that there remains a statist predisposition on the part of many when thinking about the provision of public goods. Reflecting the old workplace system, a much higher percentage of respondents in the cities expected the workplace to provide insurance. In major cities, the workplace was still the major provider, accounting for 32.1 percent; 45.1 percent of respondents stated that they would prefer the workplace to provide insurance. In the cities, there was also relatively little enthusiasm for making use of the market to purchase commercial insurance. Only 18.4 percent said that they preferred this method, and only 14 percent of individuals had purchased insurance themselves. By contrast, in the villages there was greater recognition that outside support was likely. Only 11.7 percent of those living in the villages had insurance provided through the workplace; 36.6 percent said they preferred to purchase insurance from a commercial company. In the villages, perhaps reflecting memories of the collective era, 51.5 percent thought that local government should provide insurance, as compared to 36.4 percent in the cities. This is interesting given that the policy we have reviewed above tries to shift the responsibility for more social welfare such as pensions and medical insurance from the workplace to the local government. It would appear that urban sentiment has not yet caught up with this trend. For those who bought insurance, trust was the most important factor (59.8 percent), followed by service (26.2 percent) and cost (22.9 percent). There has been little survey work on citizens’ perceptions of changes in welfare, but what has been conducted reveals similar results. Nielsen

152 / providing public goods in transitional china

et al. (2005), in their survey of perceptions of social security in Shanghai (conducted in November 2003), find that, on the whole, there is a relatively good level of satisfaction with social protection and a surprisingly high level of trust in the capacity of the government to provide for those in old age, a finding that is echoed by a survey conducted by Frazier in Beijing and Shanghai (2006; survey conducted in spring 2005). Nielsen and her colleagues note that citizens in Shanghai, rather than taking on a greater responsibility themselves, still see the state as the major provider for welfare. Wong and Lee (2001), a few years earlier, found a similar outcome in Shanghai when people ranked government as the most important provider of individual economic and social security, followed by the workplace, and the individual in the last place. They also found that 68.5 percent of the respondents disagreed or strongly disagreed with the notion that retirement was a personal issue and that one had to plan ahead. With respect to medical expenses, 73.1 percent disagreed with the idea that it was reasonable to be responsible for one’s own medical expenses. Nielsen et al. found that in the period from 2001 to 2003, 47.3 percent of respondents felt that there had been some improvement in the level of social protection coverage, but 35 percent felt that there had been no change. This conflicts with the view expressed by many that with the reduction in SOE capacity, there would be greater dissatisfaction in urban China and among workers. The advantage of Nielsen’s survey is that it covers all urban workers and not just those in the SOE sector. This would suggest that any negative feelings that the SOE workers had were offset by the positive feedback from the new groups that were now benefiting from some kind of coverage, thus lending support to the government’s attempts to expand the coverage base. On the specific question of pensions, 35.1 percent were confident that their coverage would be sufficient for their basic living needs in the future. A further 37.1 percent were ambivalent, with 22.9 percent either not confident or extremely unconfident. Not surprisingly, the findings show that those who felt there has been a decrease in social protection were those groups who had been disadvantaged by the reforms that sought to shift more responsibility from the government and enterprise to the individual. These include those who have low personal or household incomes, the unemployed, the retired, or those living away from the richer coastal regions. Frazier’s survey (2006), carried out in Shanghai and Beijing, found that a substantial majority of respondents, regardless of employment status, age, and gender, supported universal pension coverage and a strong state role in the provision. Majorities of 90 percent or more supported the idea that pensions are a basic right of citizens and equally high percentages

building a new social insurance system / 153

agreed that a pension law should provide greater assurance that citizens will receive pensions. Respondents expressed a relatively high degree of confidence in the future of China’s public pensions and in the ability of the government to finance such pensions in the foreseeable future. Given the scandals that have surrounded the misuse of pension funds, it is perhaps surprising that respondents expressed high levels of trust in the local and central government agencies that administer pensions. It is interesting to speculate whether such strong opinions would have been held in Shanghai after the purge of Party Secretary Chen Liangyu for pension fund–related scandals. By marked contrast, respondents across the spectrum rated commercial insurance forms as consistently untrustworthy. Only 24 percent in the Beijing sample and 14 percent in the Shanghai sample felt that commercial insurance firms were relatively trustworthy or very trustworthy. Interestingly, there does seem to be a generational factor at work here as respondents under the age of 40 (46 percent in Shanghai and 35 percent in Beijing) found commercial insurance firms to be relatively trustworthy or very trustworthy in how they handled pensions. Rural Pensions Whereas the problem in urban China has been to reconstruct an existing system, the challenge in rural China has been to build a new system in its entirety. This is a daunting task and, if achieved, will be an amazing achievement for a developing country. The prevailing consensus is that attempts to cover rural populations with social insurance have failed pretty much everywhere in developing countries (Gillion et al.: 2000). As Jütting (1999) has concluded, the extension of government-based social security systems to include rural populations and approximate universal coverage has been unsuccessful. Although there has been a central government commitment to experimentation and progress with weaving together a new pension scheme for urban China, there has been prevarication, lack of direction, and even opposition to state involvement in pension provision in the countryside. However, the changing structure of work in the countryside and the demographics clearly indicate that something needs to be done, but what exactly that is is less clear. The traditional approach to support in the rural areas is to rely on the family with land, in the form of a private plot, as a guarantee for security. The back-up for those without such security was a system of local government interventions in the form of disaster and poverty relief and support for the “five-guarantee” (wubao) households (for details, see Chapter Seven). The changing demographics have led policymakers

154 / providing public goods in transitional china

to realize that over the long term the family cannot be the sole provider. The impact of family-planning policies has made smaller nuclear families the norm for many and out-migration has skewed community demographics. In villages with high out-migration, it is quite common to see only the old and the very young, with women predominantly engaged in farming activities. Although many peri-urban and wealthier areas are still counted as rural in terms of household registration, they have become heavily industrialized with a predominance of township and village enterprises (TVE). The TVEs employed 142.7 million rural workers in 2005, as opposed to the 299.7 million still engaged in primary agricultural activities (http://www.stats.gov.cn/tjsj/inds/2006/ indexeh.htm). The TVE employees work in an environment more suitable for urban-style pensions. In addition, there are large numbers of migrants, some of whom have lost contact with their rural roots as they spend longer periods of time away. A patch of rural land is unlikely to help them in old age. Absent any central government direction, rural localities began to experiment with new forms of old-age provision and these were pulled together by the Ministry of Civil Affairs (MOCA).2 By early 1987, the MOCA won approval to oversee the experiments with rural pilot schemes and it proposed six models in five provinces. On the basis of this work, in 1991, the MOCA gained the support of the Ministry of Labor and the State Commission for Restructuring Economic Systems (now both defunct) to issue a joint document giving the ministry oversight of rural pension work and to set up the Temporary Office for Rural Social Insurance Pensions. In April 1991, experiments began in five counties and in October they were extended to 52; by late 1992, there was experimentation in 1,028 counties, including almost the whole of Jiangsu (ADB: 2002, pp. 15–16). The heavy involvement of Jiangsu is not surprising given the numbers of TVEs and in the south of the province the development of what Jean Oi (1992) terms “local state corporatism.” The MOCA’s objective was ambitious and, if realized, would have been unprecedented for a developing country. The intent was to provide coverage for all rural dwellers between the ages of 20 and 59, that is, not only those engaged in off-farm activity. On turning 60, individuals would be entitled to receive annuities based on the total amount that had been accumulated in personal accounts. The pension was to run for a minimum of 10 years. The county was selected as the basic level for implementation, and the accumulated funds were to be invested in special government bonds. The schemes operated through voluntary individual contributions supplemented by collective payments that were accumulated in individual accounts owned directly by the individual. Responsibility for asset

building a new social insurance system / 155

accumulation and right of control rested with the rural household; within each jurisdiction, there was a single unified scheme that facilitated labor mobility sectorally if not geographically. The problem was that the operational and financial management of these funds remained with a single government agency at the county level. The program expanded gradually; by the end of 1998, voluntary rural pension schemes organized by local governments, with combined assets exceeding 14 billion yuan, covered 82 million farmers in all provinces excluding Tibet. Not surprisingly, the average contributions were very low (starting at 4 yuan per month) and average payments were only about 75 percent of the nationally designated poverty line of 560 yuan. There were 557,000 beneficiaries. According to the ADB, progress in accumulating funds was slowing over time because of the slow expansion. Of the 3.3 billion yuan increase in 1997, one-third was from investment income rather than from annual contributions, which, as noted above, were low. Also, the administrative costs were high, at 27.9 percent of the contribution income, as opposed to the 6 percent proposed by the MOCA (ADB: 2002, p. 18). The idea of rural pensions came under increasing suspicion and criticism at the political center. There was a general sense that the scheme was both premature and risky. The concentration of the program at the county level has meant that transparency and accountability are weak. This lack of accountability has resulted in a number of cases of fund mismanagement, either through embezzlement by local leaders or through unwise investments that have frittered the money away. A 2001 survey by the China National Democratic Association in Hunan found that 28.4 percent of the province’s rural pension fund had been misused or embezzled. In Sichuan, 100 million yuan of the 648 million yuan could not be accounted for (http://www.caijing.com.cn/newcn/English/Cover/2006–06-26/14027. shtml). However, whether this is any worse than the misuse of urban pension funds is hard to say. What seems to have primarily concerned the central leadership was the potential fiscal consequence if it had to bail out the system, and such fears were heightened after the Asian Financial Crisis of 1997. In the late1990s, it became increasingly clear that rural pensions were not a policy priority. In late1997, a special group was set up to review the future development of the entire insurance industry in China and 18 months later its findings were delivered. In the meantime, in 1998 when the Ministry of Labor and Social Security was established, responsibility for rural pensions was transferred from the MOCA to a new Department of Rural Social Insurance. This effectively downgraded the importance of rural pensions as the new department was

156 / providing public goods in transitional china

weak and the ministry had as its main priority the development of urban-based systems. A report delivered to the State Council declared that conditions were not ripe for an extensive pension scheme to be carried out in rural China. As a result, Zhu Rongji, the then premier, who had appeared hostile to the notion of rural pensions, announced that any schemes where conditions were suitable should be commercialized. Commercialization led to decline, confusion, and mistrust of the local government schemes. Coverage dropped from 75 million in 1997 to 54.42 million in 2005 (China Labor Statistical Yearbook: 2006, p. 561); it dropped further to 53.74 million in 2006, from 15.4 percent to 12 percent, a slight revival from the 11 percent of 2004. The accumulated fund was 31 billion yuan, with pensions paid out to 3.02 million people (China.org.cn/English/China/190397.htm). One innovation clearly linked the extension of pensions in the rural areas to the CCP’s intent to limit family size. Although the idea was initially vetoed by Zhu Rongi, the then premier, it has since been accepted and expanded. From 2002, the State Family Planning Commission began experiments in 23 provinces and equivalent jurisdictions covering 1.35 million retirees to provide financial support if they had only one child or two daughters. Those families that complied were to be granted 600 yuan per annum starting at the age of 60 (Xinhua News Agency, October 16, 2006). The prevarication by the central government and the clear preference for a commercial system, if any, have created a problem of trust among the local population. The successful MOCA experiments were told to be commercialized, and others were instructed to be “cleaned-up” or “rectified.” What this meant in practice in many instances was that programs were shut down, with the initial contributions (if the farmers were lucky) returned. For example, in experiments in Sichuan in 1993, farmers were asked to contribute 200 yuan and were promised 30 yuan per month once they turned 60. Yet, in 2004, the initial lump-sum was simply returned to them without interest. Some one million farmers were said to have been treated in this way in 2003–2004 (http://www.Caijing.com.cn/newcn/ English/Cover/2006–06-26/14027.shtml). Were the experiments in the 1990s to develop a rural pension system so problematic? Certainly they were bold in comparative terms and would have proved a policy success that other developing countries have been unable to emulate. The 2002 ADB report and a 1995 evaluation by an experts’ group partly funded by the Dutch insurance group ING are more positive than China’s central government. The 1995 evaluation thought the experiments realistic and the framework adequate and concluded that the pensions had had a positive effect on economic development, population

building a new social insurance system / 157

control, and social stability. The evaluation also supported the MOCA’s view that the most viable option was to develop an integrated plan for workers in township and village enterprises and for those still engaged in agriculture. The ADB report (2002: pp. 88–89, 97) acknowledged that there were problems, but that there was little solid evidence to “support the far-reaching conclusions against the scheme.” The report stated that there was no need to abandon the experiments and implied that the decision was essentially political. The ADB report rejected what it saw as the leadership’s benign neglect and supported the view that a solely commercial scheme was simply not feasible. It preferred a continued expansion of the pilot schemes and in a very interesting analysis proposed that European experience suggested that it was neither too early nor necessarily too risky to introduce a social insurance scheme, especially in the wealthier areas. The report showed that several European countries did, in fact, start such a scheme at an equally early stage of development or even at a far lower level of development. A social insurance scheme could be introduced differentially across China. It is clear that the family and land will not be able to provide adequate support in the future. With the restricted growth of grassroots organizations and NGOs, the state and the market will have to combine to carry the burden of providing guarantees for old age in the rural areas. There is no country in the OECD, for example, that does not end up pumping large sums of money into either agriculture or pension support or both. It is unlikely that China will be able to avoid the same fiscal fate. The ADB (2002, pp. 137–38) proposed the same kind of dualism that operates in Europe. The TVEs and other employment-based sectors would operate a system akin to the pension scheme in urban China and this would include the richer farmers. For those still engaged in agriculture, there could be a system of individual contributions and earnings-related benefits but with heavy subsidies from the government. For the poorest, the traditional systems and new Minimum Living Standard Scheme, to be discussed in Chapter Seven, will remain essential for some time to come. Concluding Comments The urban and rural pension systems are dealing with fundamentally different problems and reveal the CCP’s prioritization of support for the urban working class and those employed in government over rural dwellers. Given China’s level of development, one could argue that the urban pension scheme has been overgenerous in terms of benefits and that the rural system has been neglected. However, China is not unusual in this neglect and actually has done more than other developing countries in

158 / providing public goods in transitional china

trying to put a system in place. In large measure, the pension pilots can be more correctly characterized as organized savings plans. The ambivalence of the central government about how far it should be involved, if at all, given the experiences of other countries is understandable. In urban China, the challenge has been to smooth out somewhat the inequalities of pension benefits and to integrate new sectors into the system. In rural China, it has been a case of constructing an entirely new system. The system of differential entitlements between urban and rural inhabitants and between government and party officials and workers leads the ADB (2002, p. 42) to classify China’s system as closest to Esping-Andersen’s (1990) conservative welfare regime. Over time, the two systems need to be integrated better and, where possible, fully integrated. This would break down China’s traditional rural-urban divide and begin the development of a social insurance system based on citizenship and need rather than on status and residence. Given the changing demographics, off-farm employment, and migration, this would amount to no more than keeping up with socioeconomic trends.

Ch a p t e r Se v e n Pov e rt y R e l i e f a n d Soc i al Assista nc e

The new programs for social assistance, marking the beginning of a shift from traditional approaches to relief to a modern welfare system, have been developed to support those gaps left by the formal social insurance programs outlined in Chapter Six. The main program is the means-tested program for social protection to provide minimum income support first in urban China and now in rural areas as well. This program supplements the traditional targeted poverty relief programs. In addition to the general income support for all those who fall below a defined income minimum, there are programs of targeted support to defined social groups, such as disabled veterans or families of revolutionary martyrs. These programs provide cash or payments in kind to specific support such as healthcare benefits, worker retraining programs, or provision of low-cost housing. To run these programs effectively, it is important to understand properly who is poor and vulnerable and why? Before economic reforms began, social protection was very limited as most, if not all, of the urban population was protected through the support schemes that operated through the workplace. Social assistance was very much a residual category for those urban poor who fell outside of these formal systems, such as disabled service personnel, the handicapped, or those referred to as belonging to the “three nos.” Similar programs operated in the countryside for what was referred to as the “five-guarantee households.”1 The level of support to these groups was left to the local authorities and thus there could be considerable variance both in terms of who was covered and also in terms of the level of support. As we have seen in Chapter Six, at the present time the social insurance programs not only do not cover all, but also they are heavily biased toward the urban areas. Although growing more quickly in recent years, the social assistance category has been a residual one in the urban areas. From the mid-1990s on, social assistance schemes began to receive more attention. This heightened

160 / providing public goods in transitional china

attention derived from the recognition that the social insurance schemes that had been developed still left significant sections of the population exposed to risk, for example, elderly living alone or the new urban poor, such as laid-off state workers. The need has been greatest in the rural areas, but coverage is low. In fact, the average social security expenditure per capita in urban China is 10 times that in the rural areas (UNDP: 2005, p. 62). Although the number receiving traditional forms of support has been declining, support under the Minimum Living Standard Scheme (MLSS) has been increasing rapidly. There are a number of categories for social assistance, but policy has been to try to consolidate the payments into a means-tested system. Thus, in urban China, the various categories are being consolidated into the urban MLSS. Coverage is for urban low-income or non-income households that have the requisite urban household residence permit. With funding coming from local government sources, entitlement is based on the household level of income. The situation is more complex in the countryside but there is also an intent to move to a rural MLSS. This was restated as central government policy at the January 2007 rural work conference. In addition to this new program, there is also a “five guarantees” program for rural households. These programs are funded by a mixture of government funds and funds raised within local rural communities. Before analyzing the new programs of social assistance, we shall look first at the dimensions of poverty and the traditional programs for poverty alleviation. Traditional welfare relief has been the major mechanism for the government to provide support to poor and vulnerable groups when in crisis or under economic stress. These are after-the-fact interventions that try to smooth out temporary shocks to the household that might depress income so much so that long-term livelihood may be affected. They differ from social insurance schemes such as pensions that are geared for predictable long-term challenges of retirement and old age. The programs of welfare support complement the informal coping strategies that people draw on, such as using up savings, pooling resources within the family, borrowing from kinship or other networks within the community, or selling off household assets. These coping strategies may be insufficient to deal with all situations and cannot cope with covariate risk when an entire community may be affected by drought or flooding, for example. To deal with these challenges, the World Bank has promoted the notion of “social risk management.” Holzmann and Jorgensen (1999, p. 3) describe social risk management as “public measures intended to assist individuals, households, and communities in managing income risks in order to reduce vulnerability, improve consumption smoothing, and enhance equity while contributing to economic development in a participatory manner.”

poverty relief and social assistance / 161

The Dimensions of Poverty Writing about social protection in China, Sarah Cook (2003, p. 68) provides a useful threefold distinction between being poor, dependent, and vulnerable. In China the definition of being poor has traditionally been reserved for those who lived in absolute poverty in the rural areas. This meant not having sufficient food to eat or clothes to wear. However, over the last ten years, Chinese authorities have begun to recognize new groups of poor who are the products rather than beneficiaries of reform, such as the new urban poor. Dependence is taken to mean those who do not have independent labor power, such as children, the elderly, or the disabled. If they lack family support, they may receive residual support through the Ministry of Civil Affairs. They are recognized as legitimate recipients of state support. Those who are vulnerable have received little support from official policy, but they constitute a growing group of China’s citizens. They most likely live just above the poverty line, but they may be pushed below that line by shocks such as major illness or industrial accident. They are likely victims of gradual or sudden insecurity. China’s approach to dealing with poverty has undergone significant evolution during the reform period. For rural areas, the initial policy interventions, which were not well targeted or focused, were based on funding to poor counties. This geographic targeting, which was based on the county until 2001, meant that poor households often did not benefit from funding that was disbursed. Prior to this, rural poverty reduction focused on the development of poor regions as a whole, with the expectation that “trickle down” would work. Urban areas were not considered at all, despite the growing presence of urban poverty. Although the number of poor declined dramatically, the effectiveness of specific poverty interventions was increasingly questioned by domestic and international critics alike. These criticisms led to a significant shift in policy. Already from 1997, microcredit programs were being expanded to reach poor households and from 2001, with the completion of the “Eight-Seven Poverty Reduction Plan,” policy was focused on poor villages, even though key poverty reduction counties were still designated. Under the Hu Jintao-Wen Jiabao leadership, new programs have been introduced to provide more focused support for education, health, and infrastructure. Support for the poorest in rural China was brought more into line with that in the urban areas through the introduction of the MLSS. Unlike rural China, policy for urban China did not suffer from the problems of regional targeting and was household-focused, first through the elaborate system of benefits provided by the workplace and later through the revamped social insurance schemes backed up by the MLSS for the very poor.

162 / providing public goods in transitional china

Poverty alleviation is one area where the market, rather than government intervention, has clearly had a major impact. The return to household farming and the effective abandonment of the rural collective structures meant that the pent up energies of the rural population were released, with farmers able to cultivate and sell their produce on the market. This, combined with the state’s decision to raise the purchasing price for staple goods such as cotton and grain in the early 1980s, led to a significant rise in farmers’ incomes, pulling tens of millions out of poverty. Producer prices had been held artificially low so as to transfer resources from the rural to the urban areas and to contribute to the program of urbanization and industrialization. The numbers of those in absolute poverty dropped from 250 million in 1978 to around 125 million in the mid-1980s. By contrast, the state interventions to deal with poverty alleviation that were introduced from the mid-1980s were less successful in raising significant numbers out of poverty. Poverty reduction slowed considerably until the pro-growth policies of 1993 took effect. However, from 1996 until 2001, poverty reduction slowed once again and in 2003 the number of absolute poor rose for the first time (table 7.1). This led to renewed attention to interventions, focusing on the household, to support education and health. The rate of those in absolute poverty shows significant variation across the various provinces in China. As one would expect, the highest percentages are in the west of the country. In 2005, Qinghai had the highest rate of absolute poverty with 11.5 percent, followed by Guizhou (9 percent), Gansu (8.1 percent), and Tibet and Shaanxi (both 7.2 percent). The heavy investment in Tibet seems to have brought benefit as in 1998 the poverty Table 7.1

Official Rural Poverty Statistics

Year

Poverty Line (Yuan/Person)

Rural Poor (Millions)

1978 1985 1990 1995 2000 2002 2003 2004

100 206 300 530 625 627 637 668

250.0 125.0 85.0 65.4 32.1 28.2 29.0 26.1

Percentage of Rural Poor 30.7 14.8 9.5 7.1 3.4 3.0 3.1 2.8

Note: The percentages do not tally with the figures given for the official rural population in the Statistical Yearbooks. Using these figures would produce slightly higher percentages. Source: UNDP: 2005, p. 80.

poverty relief and social assistance / 163

rate was 19 percent. Only Inner Mongolia had a poverty rate in 2005 that was higher than the rate in 1998 (6.6 percent as opposed to 6.4 percent) that was caused by the problems resulting from the death of livestock in 2000 and 2001 (CDRF: 2007, p. 39). In 2001, the poverty rate had risen as high as 13.3 percent. The nature of poverty has changed under the reforms. During the Mao years the strict household registration system separated rural from urban China and the bloated workforce in the state-owned sector, with its relatively high welfare provision, kept absolute urban poverty to negligible levels. But migration patterns have made this divide more porous. The migrant workers are fueling the urban economic boom and also remitting significant amounts of income to their rural homes. In the cities, harder budget constraints have caused large numbers to be laid off in the SOE sector. Official statistics show that employment in the SOE sector and collectively owned enterprises fell by 47.3 million and 23 million respectively from 1995 to the end of 2005. The greatest gain in urban employment has been in the private, foreign, and joint venture sector, which has risen by almost 30 million. This suggests a cumulative job reduction of 40 million. Not surprisingly, as noted, the real unemployment rate is probably three times the official rate of 4.2 percent (end-2005, China Statistical Yearbook: 2006, p. 142). This rate is down slightly from the highest official rate of 4.3 percent in 2003 and well below the official rate of 5.3 percent in 1978. The highest unemployment rate is in Liaoning province, at 5.6 percent (2005); the lowest rate is in Beijing at 2.1 percent. This employment realignment puts a greater strain on welfare budgets and creates newly disadvantaged groups. Not surprisingly, the number of urban poor has increased. By the end of 2005, there were some 22.3 million recipients of the MLSS, up sharply from the number enrolled at the turn of the century. But such figures, not including the migrant population, clearly underestimate urban poverty. A 1999 survey by the National Bureau of Statistics revealed a 15.2 percent migrant poverty rate, some 50 percent higher than that in the local communities where they lived (UNDP: 2002, p. 31). A survey conducted by the Chinese Academy of Social Sciences that was reported in 2002 analyzed in detail the situation of poor urban families in Shanghai, Wuhan, Tianjin, Lanzhou, and Chongqing. The survey found that between 60 and 90 percent of poor families (ranging from 64 percent in Tianjin to 92 percent in Chongqing) could not afford tuition and miscellaneous expenses charged by schools. There were also significant drop-out rates, ranging from 7 percent in Shanghai to 27 percent in Wuhan. Similar problems exist with respect to receiving medical care. Between 30 and 60 percent of those in poor families had a chronic or hereditary disease, the percentages

164 / providing public goods in transitional china

ranging from 33 percent in Shanghai to 61 percent in Tianjin. Some 50 percent of those surveyed in Shanghai and 70 percent in Tianjin were deterred from receiving medical attention because of costs. However, unlike those classified as poor in rural China, the survey revealed that the urban poor could cover the basic necessities of food, clothing, and housing. For the first two decades of reform, policymakers did not focus on urban poverty as they assumed that the urban poor would be taken care of by the SOEs or government agencies where they worked, with a residual role for the Ministry of Civil Affairs to deal with the “three nos.” It was only with the big shakeout of the state system in the mid-1990s that urban poverty became a major policy concern. However, urban poverty had existed well before the 1990s. Khan and Riskin (1998), using grouped income data, suggest that the urban poverty head count fell from 20 percent in 1981 to 13 percent in 1985 to only 5 percent in 1991, before rising again (Khan and Riskin: 2001) to 8 percent in 1995. The general growth in the late 1990s caused this figure to drop again to 2.2 percent in 2002 (CDRF: 2007, p. 48). This drop was almost certainly aided by the introduction of the MLSS from 1998. Although new categories of urban poor have been created by the reforms, China’s spectacular success in reducing those in absolute poverty in rural China slowed during the latter part of the 1990s as it became clear that the remaining 23.6 million (end-2005) living in absolute poverty would not be pulled out of poverty by relying on higher GDP growth alone. Official statistics for poverty alleviation show over 200 million have been lifted out of absolute poverty since the reforms began. One reason for the slowing rates of the decline in absolute poverty and the slight rise in poverty observed in 2003 is the problem of recidivism. Although 14.6 million were pulled out of poverty in 2003, another 14.4 million fell back below the poverty line. A closer inspection reveals a more complex picture and less success than Beijing claims for the direct state intervention. China’s official statistics (using $0.66 per day in purchasing power parity dollars) underestimate the real level of rural poverty. Using the World Bank norm of $1 per day reveals a total approaching 200 million still in poverty in 2000 (22 to 24 percent of rural dwellers and 18.5 percent of the population as a whole). A $2 per day standard would render a total of 53.7 percent of the total population. These totals put China roughly on a par with Indonesia (15.2 and 66.1 percent respectively) and considerably better off than India (44.2 and 86.2 percent respectively) (World Bank: 2003, p. 59). In May 2004, the Chinese authorities released a figure of 56.17 million living below the $1 per day threshold in 2003 (SCMP, May 4, 2004), whereas former World Bank president Paul Wolfowitz stated during an October 2005 press conference in Beijing that there were still 150 million living in absolute poverty in China (Xinhua News Agency, October 18, 2005). The Rural Survey

poverty relief and social assistance / 165

Team of the National Bureau of Statistics (2004) calculated the number at 85.17 million based on the $1 per day standard. These are now referred to as the “low-income population,” as opposed to the “absolute poor.” Although the number of those in rural China earning between 200 and 400 yuan per capita per annum remained stable, those earning under 100 yuan per capita per year rose from 0.31 percent of the absolute poor in 2000 to 0.65 percent in 2005 (for a good review of the problems in counting China’s rural poor, see Wang: 2004). Further, most official Chinese surveys of poverty do not cover non-income aspects of poverty and wellbeing. As we have seen in Chapter Four, in terms of medical access, this can be significant. The National Bureau of Statistics in 2000 carried out a survey of 170,000 administrative villages in designated poor counties. The survey discovered that 20 percent lacked acceptable roads, 8 percent lacked electricity, 65 percent had no phones or tap water, 14 percent lacked primary schools, and 25 percent had no clinics or doctors. What these figures show is that despite tremendous progress, there is still a significant group in the countryside that has not been helped by government policy, a very large group living just above the poverty line who are still very vulnerable, and a smaller but growing group of urban poor who are the creation of the reforms. The basic view of the government has been that people are poor because of physical disadvantage (such as living in remote areas) or lack of reform. The Jiang Zemin–Zhu Rongji leadership seemed to follow the “trickle-down” notion, giving primacy to rapid economic growth accompanied by limited targeted interventions. As inequality became a political concern in the late 1990s, a more proactive “Develop the West” policy was promoted. The leadership recognizes that this is insufficient and thus would like to enhance mechanisms for redistribution while pursuing the urbanization drive. However, policies of redistribution raise questions about targeting, whether such funds will be used optimally, and whether it can work against development of the most productive parts of the economy. Traditional Poverty Alleviation Mechanisms The slowdown in poverty alleviation was partly inevitable as those who could be helped quickly by the reforms had moved out of poverty and it was primarily the more intractable cases that remained. The majority of these extremely poor communities are in inhospitable environments, often in fragmented upland villages with poor infrastructure and with little if any agricultural surplus to market. In fact, for many in this group the introduction of the market reforms had a negative effect as the farmers did not benefit from the increase in prices and access while the costs of inputs

166 / providing public goods in transitional china

and consumer durables rose. This situation convinced the Chinese government that a massive push was necessary to raise the group out of poverty. As a result, in 1984, it set up a special agency under the State Council, the Leading Group for Poverty Alleviation, with poverty alleviation offices at the county and township levels to coordinate policy. It invested a huge amount of funds into this objective, some 274.6 billion yuan from 1986 to 2004 (in budget support and subsidized loans) and devised a number of specific policies targeted at the counties where it thought the poor were situated. From 1986, the central government and the provincial authorities began to identify “poor counties” that would receive targeted poverty alleviation interventions. For even a moderately poor county, it was beneficial to be listed as one of the counties to receive benefits and dispensations; therefore, intense lobbying took place both at the national and the provincial levels. Of the initial 258 officially designated national poor counties in 1986, only one-third (83) actually met the criterion of below 150 yuan in rural net income per capita in 1985, indicative of the intense politicization of the process. Another 82 had incomes between 130 and 200 yuan and 93 had average incomes between 200 and 300 yuan (Wang: 2004, p. 20). Included on the list were counties that had been part of the pre-1949 CCP revolutionary base areas, counties that contained ethnic minorities, and pastoral counties. By 1988, there were 328 nationally designated counties in addition to 370 provincially designated poor counties. The number of national counties increased to 331 in 1993; after a major overhaul in 1994, the total was expanded to 592. This expansion took place despite the fact that the total number of poor had fallen, according to official statistics, from 120 million in 1985 to 80 million by 1993. In addition, two of the ten poorest counties in China were not among the nationally designated counties (Nyberg and Rozelle: 1999, p. 97). A new poverty line was set at 320 yuan in 1993 prices, but only 326 of the counties met this criterion (Wang: 2004, p. 22). In addition to the fact that targeting has clearly not been very effective, the poverty alleviation program has two main objectives that sometimes clash. One objective is to help the poor and the second is to promote economic development. Many of the absolute poor live outside of the nationally designated counties, so the distribution of funds within the poor counties is not effective in supporting the poorest households. The National Bureau of Statistics estimates that one-third of the rural poor live outside of the 592 counties, and other studies suggest that the number may even be as high 50 percent (UNDP: 2002, p. 35). The leakage of funds through misappropriations and bad investments and use by the non-poor households have meant that effectiveness is low. Within the counties, the

poverty relief and social assistance / 167

money is not properly directed to poor households; instead it is usually divided out evenly between poor and non-poor townships. The UNDP calculated that some 55 to 75 percent of poverty alleviation funds did not reach poor households (UNDP: 1998, p. 102). Thus, spreading the money evenly over the 200 million residents in these counties where only 21 million of the absolute poor lived in 1998 meant that the benefits to the poor were outweighed tenfold by distribution to the non-poor (World Bank: 2000, pp. xvi–xvii). For many local leaders, poverty alleviation funds are regarded as general development resources and thus a useful supplement to local state income. With the lack of transparency, the funds can often be siphoned off to support pet projects. This was especially a problem in the early and mid-1990s before central policy shifted back to targeting the household. The World Bank (2000, p. xxiv) has calculated that in 1992 and 1993, around half of all poverty alleviation-subsidized loans were lent to industrial enterprises. A survey by the Ministry of Finance comes to similar conclusions, finding that the majority of subsidized loans were made to large-scale enterprises or were used for infrastructure. In 2002, it seems that of the 750 million yuan in poverty alleviation loans in Jiangxi, only 150 million went to households. According to Wang (2004, p. 48), this diversion of loans to non–poverty alleviation targets became more serious with the commercialization of the Agricultural Bank of China and the closure of most township bank branches. By the mid-1990s, this strategy of copying the successful counties in developing township and village enterprises had run its course and many were in severe financial difficulty. This may have contributed to the decision in the mid-1990s to again direct subsidized loans to the household. There are three main programs for poverty alleviation: the subsidized loan program managed by the Leading Group and the Agricultural Bank of China (ABC) (responsibility was returned to the ABC in 1998 from the Agricultural Development Bank that was set up in 1994); the food-forwork program managed by the former State Development and Planning Commission (now the National Development and Reform Commission); and the budgetary grant program. As suggested above, the impact of these programs to help the poorest has not been significant and there have been serious distortions in implementation.2 In 1986, the central government increased by 33 percent the amount of development capital for the Ministry of Finance to direct to counties for investment in poor areas. The budgetary grants totaled 2.7 billion yuan in 1997 (Park et al.: 1999, pp. 21–22) and were complemented by a number of other special funds and tax breaks for poor areas. These earmarked funds have been subject to diversion to cover local government operating costs,

168 / providing public goods in transitional china

especially salaries, and general budgetary pressures have often made it difficult to raise the amounts needed. Increasingly important has been the compulsory education project. Overseen by the Ministry of Education and implemented by the county-level offices, the project was set up as a separate fund in 1996 to improve basic schooling. During the first five years of the program, no subsidies were given to students but between 2001 and 2005 about 10 percent of the total budget was allocated to student support. This should increase with Premier Wen Jiabao’s announcement that fees should be waived to support children from poor families in rural areas and, when possible, exemptions should be extended to all rural children. This policy has actually resulted in a reduction of funds available to local governments as the amount they receive from the central government is less than the amount they used to raise in fees from rural families. Beginning in the mid-1980s the Chinese government relied primarily on such kinds of targeted transfers as the main mechanism for alleviating rural poverty. The main problem is that targeting to date has not been effective (Wang: 2004; Park et al.: 2002; Fan et al.: 2002). Paradoxically, the most dramatic decline in poverty slowed to a halt in the mid-1980s, just as China was setting up a poverty alleviation program (UNDP: 1998, p. 86). In fact, in a centrally planned economy the simple act in the late 1970s of increasing state purchasing prices for staple goods by 30 to 50 percent followed by the introduction of market incentives did more than any other single policy to move people out of poverty. Recent work suggests that the main determinant of the exodus from poverty has been the economic growth more broadly and that there is no statistically significant relationship between poverty incidence and poverty investment, thus rendering poverty investment ineffective (Rozelle et al.: 1998). Fan and his colleagues have shown that, of all investment programs, government-targeted antipoverty loans have had the least impact on poverty alleviation (Fan et al.: 2002). Problematically for current policy, as poverty levels fall, it is more difficult to pull the remaining people out of poverty by relying on economic growth alone. This is not to say that there has been no effect at all. Research by Park and his colleagues (2002) suggests that designated poor counties experienced higher growth than would otherwise have been the case, but the impact of targeting declined in the 1990s and the growth was insufficient to prevent the counties from falling farther behind. A long-term study of the impact of investment in the World Bank-financed southwest program produced mixed results (Chen et al.: 2006). The study found a sizable and statistically significant impact on mean household income in participating villages during the disbursement period. But there was a smaller impact on consumption. The study concluded that four years after disbursement

poverty relief and social assistance / 169

ended, both project and nonproject villages had experienced sizable economic gains but with only a modest net gain to mean income attributable to the project. The “food-for-work” program began in 1984 and was initially conceived of as a program to use the stocks of surplus grain procured by the government after the 1984 harvest. However, it evolved into a program to finance much of the rural road construction, agricultural infrastructure, and the maintenance of water systems. In theory, this program should have been more successful as it bypassed the local budget departments, thus reducing the possibility of a diversion of funds. Yet this does not mean that the benefits reached those most in need. Driving along rural mountain roads one is struck by how frequently one comes across small teams working on road maintenance with crude tools and doing much of the heavy work by hand. In fact, much of the main road system in the mountainous areas is in remarkably good shape. However, this highlights a problem with the program objective. The objective is to improve the infrastructure for the community as a whole and not to maximize employment opportunities for the poor. The expectation is that road construction will enhance the capacity for growth within the region and that the “trickle down” benefits will help the poorest. The resources provided by the central or higher-level authorities are expected to be matched by the local authorities, but this is rarely the case. Unlike in India, which runs a similar program, labor participation in China is rarely voluntary. In fact, in a number of counties that I have visited they counted the labor of the farmers as their financial input, and this labor goes unpaid by the local authorities! This has been a major problem for the poor as it effectively amounts to yet another tax burden. Zhu and Jiang (1996, pp. 75–107), who carried out research in Ningxia and Shandong, discovered that 40 percent of the surveyed households did their work without receiving payment. It is probably also the case that local governments have simply used these projects to substitute for construction projects that they should have carried out with their own funds. The subsidized loans are run through the ABC, now supposedly a commercial bank, at heavily discounted rates of around 3 percent (since 2002) loaned over a three- to five-year period. But over time, the effectiveness of these loans in reaching the poor has declined. Such cheap capital is attractive to local governments and wealthier individuals for their own projects and much of the money is captured by those for whom it is not primarily intended. In addition, since banks are now under pressure to cut their losses, there are further disincentives for the ABC to loan to poor households (World Bank: 2000, p. xiv). The costs of administration in dealing with scattered households often living in inhospitable terrains are high;

170 / providing public goods in transitional china

given the low enforced interest rates and on-time repayments of below 50 percent, such loans are not profitable. Also, as land is contracted out from the collective to the household, the farmers are not able to offer their land as collateral on the loans. These factors incline the ABC officials to lend to what they believe will be money-making projects, usually involving enterprise development. The problems that have been highly associated with these approaches of regional targeting and poorly directed, subsidized loans have led to a rethinking of policy. First, there has been an increased interest in the use of microcredit that can reach poor households directly. Second, there has been a shift of emphasis from county to village and household. In 2001, the central government focused poverty alleviation policy on 148,000 designated poor villages that were said to be home to 80 percent of those in poverty (Xinhua News Agency, October 6, 2006). Third, there has been a move to provide minimum subsistence support for the rural poor, similar to the support that began operations in the urban areas in the second half of the 1990s. The 1996 decision by the Leading Group to focus attention on loans to poor households has led to a boom in interest in microcredit schemes. In 1997, a number of provinces began to authorize the use of poverty alleviation funds for such schemes and in 1998 the Leading Group emphasized that microcredit should be expanded to all provinces. Small-scale lending schemes have been very successful in other Asian countries, not only in targeting the poor but also in building up the lending infrastructure and reaping high rates of return, well above those experienced in China. But local groups in China have been highly innovative, and programs targeting poor rural women have been run by the local offices of the Poverty Alleviation Bureau and the Women’s Federation. Programs modeled on those of Grameen Bank in Bangladesh or Cash Poor in Malaysia have been introduced by international organizations such as the UNDP and private organizations such as the Ford Foundation. In 1998, it was also decided that the loans should be distributed through the local branches of the ABC rather than through the poverty alleviation offices in order to ensure better supervision and repayment. By the end of 2001, the ABC had issued 3.8 billion yuan to 2.3 million poor households and 10.6 million people (Wang: 2004, p. 29). But this structure reduced incentives for the poverty alleviation offices to ensure a good functioning system, and the branches of the ABC have little incentive to manage the program. These institutional factors and the enforced low lending rate mean that microcredit schemes have not provided the “magic wand” that the Leading Group was hoping for and will not be as successful as their counterparts elsewhere in Asia. The interest rates cannot cover operating costs, which

poverty relief and social assistance / 171

in turn will make it impossible to keep such programs sustainable. Despite these problems, there have been renewed attempts under Premier Wen to expand microcredit schemes. A major problem with implementation arises from the way Chinese officials view such programs. Essentially, officials see them as a welfare program rather than as one that can be financially sustainable. The traditional top-down approach of planning discourages the kinds of local and community-based networks that are necessary to allow such lending programs to prosper. Such thinking is hard to shift as is state policy on low interest rates, and without such changes the programs will not be successful and may create significant financial problems. Research by Park and Ren (2001) has demonstrated that internationally run programs tend to function better than government schemes. Average repayment rates in government schemes are declining to 50 to 70 percent, well below the threshold for them to be viable. Unger (2002) points out that there are also structural problems in trying to apply a Grameen-type lending program in China. In Bangladesh, where the model was developed, there are a large number of landless poor women who engage in small trade and production and Grameen primarily lends to this kind of producer. The loan allows these women to increase their stock and to make daily sales to help meet their weekly repayments. In China, by contrast, the poor are farmers in marginal, mountainous areas who do not have regular earnings but rather have to wait until the harvest is in or until their livestock has been sold off to make repayments. There are no opportunities for lenders to diversify their sources of income by engaging in nonagricultural pursuits. Also, village-based lending opens up the possibility for nepotism and cronyism at the local level as traditional power structures reassert themselves. Unger estimates that in China 40 percent of loan officers in 2002 were either village heads or village party secretaries and all were male. This has necessitated some experimentation with the Grameen model to suit local circumstances in China. One interesting set of experiments has taken place in Tibet under the auspices of a program supported by the Tibet Poverty Alleviation Fund, an international NGO headed by a former head of the UNDP in China.3 The program operates in two main areas, Lhoka (Shannan) prefecture and Nakchu (Naqu) prefecture. In Lhoka, an area of sedentary agricultural villages, with few modifications the program has been successful with close to 100 percent payback rates in all but one village. The most significant innovation was the introduction of monthly rather than weekly repayments because of transportation difficulties. Of course, the administrators are not members of the NGO itself but rather are drawn from the ranks of township governments, and they have to struggle with a government-imposed 3 percent interest rate

172 / providing public goods in transitional china

and a 3 percent management fee. By contrast, the program in Nakchu has required more significant experimentation. The prefecture is home to a scattered population with poor weather, limited access to any kind of market, and a nomadic population. Because of resistance to the idea of individual liability, larger loans are made to the village as a whole. Given that the products that the lenders can sell are meat and hides, it is impossible to set a weekly or even monthly repayment schedule and thus loans are to be repaid every six months. Also, government officials from the township have a much more limited role in the program than is the case elsewhere. This kind of innovation is becoming more common in other areas of poverty alleviation work as well. The government has gradually begun to acknowledge not only the role of the market but also that of NGOs and community participation that have proved successful in other countries. Poverty alleviation work is an area where the government seems willing to accept a role for NGOs. In a major departure from past practice, the 10-year plan for poverty alleviation adopted in 2001 explicitly states the need to bring NGOs on board to help implement government projects in poor areas (State Council: 2001). A meeting convened by the China Foundation for Aiding Poor Areas in October 2001 calculated that incomplete statistics suggested that resources mobilized by the NGO sector amounted to between 18 and 28 percent of the total funds for poverty alleviation in the second half of the 1990s (information from participants). Since its establishment in 1989, the foundation had raised over $60.4 million in cash and kind to help the rural poor in central and western China (China Daily, August 4, 2001). Similar organizations with close party and state ties have also been important in fund raising for welfare projects. Perhaps the best known is the Youth Development Foundation set up by the Communist Youth League, which launched Project Hope to build primary schools and to provide scholarships for poor children. Project Hope has provided aid to all the officially designated poor counties in China; in 1996, its funding for primary school construction accounted for 8.8 percent of the county-level nationwide budget for capital construction for education (National Research Centre: 1998, pp. 3, 5). In February 2006, an interesting experiment was announced by the ADB for village-level NGO-government collaboration. The ADB selected six NGOs to participate in government-funded poverty alleviation work in three counties in Jiangxi province. In addition, one international NGO (Heifer Project International) was selected and officially endorsed by the Jiangxi Poverty Alleviation and Development Office. The six organizations were to work with the China Foundation for Poverty Alleviation, the implementing agency of the pilot project. State funds cannot be used to pay NGO running costs, thus the support from

poverty relief and social assistance / 173

the ADB (China Development Brief, February 25, 2006, at www: chinadevelopmentbrief.com/node/493). A major problem with these poverty alleviation strategies is the fragmented nature of the implementing agencies. A more comprehensive and participatory approach to rural development is needed. Current policies are implemented by different, vertical bureaucratic hierarchies, with little attempt to integrate them effectively. Further, there is very little consultation with farmers about what they actually want in terms of help and little effort to build with them sustainable participatory institutions. The paternalism of CCP rule is evident, with many local officials convinced that the farmers do not know what is in their best interests. The phrase “their cultural quality is too low” (suzhi tai di) is frequently used by local officials to justify why they do not ask farmers for inputs on the design and implementation of projects. In an important new departure for China, in June 2006, the State Council Leading Group Office for Poverty Alleviation and Development together with the World Bank launched a two-year $8 million program to promote villagers’ participation in local development. Sixty administrative villages in Guangxi, Inner Mongolia, Shaanxi, and Sichuan with 100,000 poor farmers were chosen for inclusion. The “Community-Driven Development Program” is expected to improve the targeting of poverty alleviation funds by allowing poor people to manage funds in pursuit of their own priorities. The poor communities are responsible for managing program funds and for implementing small-scale infrastructure and public service improvements (Xinhua News Agency, October 6, 2006). To resolve the problems of poverty and rural social policy in general, the Chinese government needs to develop a pro-poor approach that integrates social and economic development policy that is more participatory and inclusive. Obviously, urbanization, if well conceived, will resolve much of the problem, but in the meantime the structure of public finance hugely disfavors the rural areas. The main thrust of economic policy is to achieve rapid growth, often by shifting resources away from the poor and then returning some funds through limited poverty alleviation interventions. Rural net taxes have been highly regressive, so in 1995 the top decile of the rural population appropriated 133 percent of all net subsidies, whereas the two poorest deciles paid nearly half of all net taxes (UNDP: 2002, pp. 37–38). Investment patterns, as we have seen, favor the coastal and urban areas. However, any major reallocation of resources is liable to be politically unacceptable. Wang (2004, pp. 56–57) concludes that although poverty alleviation programs may have stopped the poor areas from falling even farther behind, the efficiency of the investment has decreased and the impact of other investments has been greater. He concludes that investments in

174 / providing public goods in transitional china

agriculture, education, and health would be more effective than the local government preference for investment in industry in poor areas. In addition to increasing the effectiveness of design and delivery of government programs, it is also necessary to create conditions that enable the poor to cast off poverty on their own, which is often a matter of ensuring that they are able to secure access to and control over productive resources and credit. Strengthening the rights of farmers and rural communities to manage and derive benefits from the natural resources on which they depend is one of the keys to securing sustained improvements in the lives of the rural poor. The Minimum Living Standard Scheme The development of the Minimum Living Standard Scheme (MLSS) marks a clear step in the direction of a modern welfare state and a shift away from a traditional approach of alms and charity for specially identified groups. China resembles other developing countries in the sequencing of social insurance and social assistance programs. By contrast, the OECD countries developed social assistance programs, such as the poor laws, before putting in place social insurance schemes (Overbye: 2005, p. 312). On the whole, social insurance schemes tend to favor those who are in the formal sector of the economy and those who are better off in society. To deal with others who are not so fortunate, the World Bank (1994) suggests that social assistance should be the “first pillar” of social protection in all countries. Social assistance schemes can more effectively extend social protection to the poor than can the formal social insurance schemes. These kinds of schemes can also reach not only the urban poor but also those in the rural areas who are vulnerable. As we have seen, the social insurance programs are not only unable to cover all at the present time but also heavily biased toward the urban areas. The social assistance category has been a residual in the urban areas, but it has been growing more quickly in recent years. From the mid-1990s on, social assistance schemes have received greater policy attention. This derived from the recognition that the social insurance schemes that had been developed still left significant sections of the population exposed to risk, for example, the elderly living alone or the new urban poor such as the recently laid-off state workers. The need has been even greater in the rural areas but again coverage is low. There are a number of categories of social assistance but policy has been to try to consolidate payments into a means-tested system. The traditional system of welfare relief as it developed in the urban areas over the 1960s and 1970s was designed to catch those who fell outside of the formal system either on a long-term or a temporary basis. Those receiving regular support in

poverty relief and social assistance / 175 Table 7.2

Recipients of the MLSS and Other Relief (millions)

Type of Support

2001

2002

2003

2004

2005

2006

Urban MLSS

11.70

20.64

22.46

22.05

22.34

22.41

Urban Recipients of Temporary Relief

2.14

3.21

3.48

2.85

2.34

—

Rural MLSS

3.05

4.01

3.67

4.88

8.25

15.93

—

0.51

1.15

1.65

0.50

—

0.81

0.90

7.93

9.14

10.67

—

—

—

1.74

2.29

3.00

—

18.00

16.78

20.10

17.20

13.60

—

—Of Whom “FiveGuarantee” Households Number Receiving Traditional Relief (Rural) —Of Whom Five-Guarantee Households Rural Recipients of Temporary Relief

Source: National Bureau of Statistics, at http://www.stats.gov.cn/tjsj/ndsj/2006/htm/W2329c.htm; MOCA, at http://admin.mco.gov.cn/news/content/recent/2007523/22309.htm.

urban areas are referred to as the “three nos.” In addition, there was support for families in difficulty, laid-off workers, and people such as disabled veterans who were given support under various regulations. According to the Ministry of Civil Affairs that oversaw these programs (MOCA: 2000), some 600,000 to 800,000 people per year received regular urban relief and over 2 million received temporary relief. This reached a peak of 3.5 million in 2003 before dropping to 2.3 million in 2005 (table 7.2). However, as the ministry document went on to state “the traditional social relief system, with its limited coverage, low standard, and unstandardized working system fails to meet the demands of the socialist market economy and calls for immediate reform and improvement.” The shake-up of the SOEs created new categories of urban poor to be added to the traditional groups. In urban China, these various categories are being consolidated into the urban MLSS, covering urban low-income or non-income households that have the requisite urban household residence permit. With funding from local government sources, entitlement is based on the level of income of the household. The situation is more complex in the countryside, but there is also an intent to move to a rural minimum income scheme. This was restated as central government policy at the January 2007 rural work conference. Traditional social assistance programs were focused on the “five-guarantee” households. The number of people receiving temporary relief dropped from 18 million in 2001 to 13.6 million in 2005, and the number of “fiveguarantee” households was 4.68 million at the end of 2006. It is clear that where possible government policy is to shift people onto an MLSS and, as

176 / providing public goods in transitional china

a result, the numbers rose from 3.05 million in 2001 to 15.93 million at the end of 2006. Because the level of support to those identified is left to the decision of the local authorities, there may be considerable variance both in terms of who was covered and also in terms of the amount provided. We begin with a review of MLSS development in urban China. Prior to experiments in the early to mid-1990s, those who could work were not eligible to claim public assistance benefits even if their income was insufficient. If they were not working, they were expected to exhaust other available resources first, including support from the immediate and extended family. In 1992, there were 190,000 urban residents who received social assistance (0.06 percent of the urban population). The amount of assistance totaled 87.4 million yuan and the per capita benefit was 38 yuan per month (Leung: 2006, p. 191). This system was clearly inequitable, heightened vulnerability, and provided no real relief from economic shocks. As urban restructuring moved ahead, it became clear that a better-conceived system needed to be put in place. As in so many other areas of reform, initially the program was not imposed by the Center but rather it developed from local experimentation. In 1993, Shanghai was the first city to reform the urban assistance programs, with a particular focus on setting up a minimum living standard threshold to provide the basis for a means-tested approach. Under this experimental system, all households whose incomes fell below the government-set poverty line were eligible to claim benefits in cash and kind. The local government was required to set aside sufficient financial resources to cover the program. After this experiment was discussed at the tenth conference of the Ministry of Civil Affairs in 1994, the ministry proposed expanding the system across China’s urban areas. In September 1997, the State Council issued a circular on establishing a nationwide system by the end of 1999. This was formalized and standardized in the October 1, 1999 “Regulations on Guaranteeing Urban Residents’ Minimum Standard of Living.” The document was issued to coincide with the fiftieth anniversary of the founding of the People’s Republic and by way of commemoration, the benefit levels were raised 30 percent. In fact, by September 1999, 667 cities and 1,638 towns had set up the new system, some three months ahead of the original schedule. Beginning in 2000, the number of people supported by this new system began to expand significantly. This development marked a radical change in thinking about support. The prior necessity to test for work ability was abolished and entitlement was decided solely on the basis of the financial situation of the household. The poverty line itself was set by the local administration based on cost of living in the city and thus could vary significantly. The amount actually received is the difference between the family per-capita income

poverty relief and social assistance / 177

and the estimated minimum living standard (White Paper: 2001). In a further reform in 2001, it was announced that all the other types of social assistance that operated in the urban areas would be rolled into the MLSS. The “three nos” category received the full amount of support. To ensure that the new program did not dampen the commitment to find work, the level of support was to be below the minimum wage, pension level, and unemployment benefits. Thus, in 2003, the minimum wage per month was 490 yuan, unemployment insurance benefits were 329 yuan to 419 yuan, and the minimum pension benefit was 466 yuan, whereas the minimum living standard allowance was 290 yuan (Leung: 2006, p. 191). The impact of the program can be seen by the rapid expansion in the number of those receiving subsidy. Recipients rose from 0.8 million in 1996 to 4 million in 2000 when the system was promoted nationwide and the number of recipients stabilized at 22.34 million in 2006. In 2006, the monthly average eligible payment was 169.6 yuan per person, but the amount actually paid out averaged 82.9 yuan, up 14.7 percent on 2005. The total amount of finance provided was 22.1 billion yuan, up from 19.1 billion yuan in 2005. In 2005, the central government provided a subsidy of 11 billion yuan to the localities. By contrast, the numbers of those on other forms of public assistance in urban China dropped from 1.2 million in 1996 to 0.2 million in 2000 and to none in 2004. In 2004, some 4.1 percent of urban residents were receiving the new social assistance (Gu: 2006, pp. 8–9). With the minimum level of support set by the local government, the poverty threshold varies from locality to locality. Local governments take into account not only this poverty line but also the financial viability of the program, and this affects the amount that might be available for citizens in hardship. Not surprisingly, major cities in the east have higher levels of support than those in the central and western regions of the country. In 2000, as the system was expanded, Shenzhen and Xiamen had the highest subsistence standards of 319 and 315 yuan respectively. Major cities such as Guangzhou (281 yuan), Shanghai (280 yuan), Beijing (273 yuan), and Tianjin (241 yuan) had relatively high levels, and the lowest levels were in Huhehot, Nanchang, and Yinchuan at 143 yuan. The payments can be adjusted on an annual basis and in August 2007 the MOCA announced that the MLSS could be raised by 15 yuan if local governments could afford it. This was to counteract the inflation of daily commodities, especially pork. As Vice Minister Li Liguo acknowledged, such price hikes hit low-income families most severely (Xinhua News Agency, August 7, 2007). By 2005, the highest level of support was in Guangzhou (330 yuan), with the lowest levels in Yining (165 yuan) and Shenyang and Guiyang

178 / providing public goods in transitional china

(170 yuan). The average level for eastern China in 2005 was 254 yuan, whereas that for cities in central China was 199 yuan and in west China 187 yuan. More important is the actual payout, which was very high in Beijing at 234 yuan (table 7.3) and lowest in the northwest—Xining and Yinchuan (66 yuan). Beijing also had the highest payout rate for the MLSS as a percentage of the average monthly income (18 percent) and as a percentage of the monthly costs of food necessities (62.9 percent). The lowest rates were, not surprisingly, in the southwest and northwest together with Wuhan in central China. The percentage of payout to monthly income was as low as 6.5 percent in Wuhan, 6.8 percent in Nanchang, and 7.3 percent in Yinchuan. In terms of percentage of food costs, the lowest were 32.1 percent in Chengdu, 34.2 percent in Wuhan, and 34.4 percent in Yinchuan. The regulations allow for certain adjustments to be made for special circumstances. Thus, there can be a 20 percent increase for the elderly who live alone and a 30 percent increase for the severely disabled (Leung: 2006, p. 191). Also, provinces have made varying decisions about the kinds of things people may own or do when assessing suitability for support. Thus, in Guizhou, households with a mobile phone, a motorcycle, a computer, or air conditioning are not eligible, whereas in Hunan

Table 7.3 City

Beijing Guangzhou Shanghai Shenzhen Xiamen Huhehot Nanchang Yinchuan Xining Guiyang Wuhan Chongqing Chengdu Shenyang

Provision of the MLSS in Selected Municipalities 2004 (Rmb)

2005 (Rmb)

290 300 290 344 265–315 180 165 170 155 156 220 185 178 156

300 330 300 — — 190 190 180 165 170 — 195 210 170

Actual Payout, 2005 234 165 149 185 109 100 68 66 66 78 77 77 72 70

Average Monthly Income, 2004 1303 1278 959 1504 1149 1007 1004 899 859 873 1188 907 940 938

Actual Payout Actual Payout as % of as % of Average Monthly Food Income, 2004 Expenses, 2004 18 12.9 15.5 12.3 9.5 9.9 6.8 7.3 7.7 8.9 6.5 8.5 7.7 7.5

62.9 45.8 36.0 51.4 36.9 53.5 34.7 34.4 36.3 40.4 34.2 28.4 32.1 40.4

Source: Tang and Zhang: 2006, pp. 168–69 for 2004 and MOCA, at http://www.mca.gov.cn/artical/ content/SDB_B2/0051125172845.html for 2005.

poverty relief and social assistance / 179

no one who invests in the stock market is eligible (Gustaffson and Deng: 2007). The MLSS is in the process of being expanded into the countryside and will come to replace traditional programs such as the “five-guarantee” households that receive support if they are unable to earn income and have no other visible means of support. As a result, we see a sharp spike in those receiving public assistance in rural China, from 0.1 percent of the population (1.1 million) in 2000 to 1.9 percent (14 million) in 2004 (Gu: 2006, p. 11). The rural work conference in January 2007 pledged to extend this program throughout rural China; by mid-2007, it covered a total of 20.7 million people (Xinhua News Agency, July 31, 2007). In fact, in June 2007, MOCA vice minister Li Liguo declared that the rural MLSS formed the “foundation for a nationwide social security system” and that it would be possible to cover all the rural poor by the end of 2007 (Xinhua News Agency, July 31, 2007). This attempt to expand the program rapidly has caused controversy about whether or not a “universal” minimum support system is feasible. Those who prefer not to extend support argue that farmers have land and housing to provide a minimum guarantee. They fear that if minimum support is provided some may give up farming altogether. However, as we have seen above, land no longer provides an adequate guarantee and large numbers are vulnerable in rural China. Yu Jianrong (2006) estimates that there are 64.3 million impoverished in rural areas and within this group there are 23.6 million with annual incomes below 683 yuan who do not have sufficient food and clothing. This group includes the elderly, sick, and orphaned who have no means of support. In Yu’s opinion, and it would appear that this is now the government’s view as well, these people should be provided with the MLSS. Yu has calculated that the extra funding required would amount to only about 10 billion yuan and this is easily affordable for the state. As currently structured, the scheme reflects the funding bias of the urban over the rural areas. In fact, in 2003, even in a municipality such as Shanghai, the poverty line was 290 yuan in the urban districts but only 150–183 yuan in the rural districts (Leung: 2006, p. 191). In 2003, the average threshold for eligibility for minimum living support in urban areas was about 2,040 yuan whereas in rural areas it was 650 yuan, a difference of 3.1 times (Xue et al.: 2006). In 2005, the rural MLSS covered only a relatively small percentage of those in hardship. The number in rural China estimated to be in absolute poverty was 23.65 million, of whom 19.81 million were receiving some kind of relief (around 80 percent of the total). However, of these, only 8.25 million (34.88 percent) received the rural MLSS (table 7.4). This means that 65 percent of those in poverty were not

180 / providing public goods in transitional china Table 7.4

Rural Poverty and Relief Support

Year

Numbers in Absolute Poverty (Million)

2001 2002 2003 2004 2005

29.27 28.20 29.0 26.1 23.7

Numbers % of Those Receiving in Poverty Support (Million) Who Are Covered 3.85 4.98 11.6 14.0 18.9

13.16 17.65 40.02 53.72 79.99

Number Receiving the Rural MLSS (Million)

% of Those Receiving the Rural MLSS

3.1 4.1 3.7 4.9 8.3

10.41 14.46 12.66 18.7 34.88

Source: Yang et al.: 2006, p. 38.

Table 7.5

Regional Distribution of Minimum Living Standard Support, 2005 Eastern China (%)

Urban Areas Rural Areas

26.99 77.90

Central China (%) 30.08 13.29

Western China (%) 33.09 8.81

Source: Yang et al.: 2006, p. 42.

receiving long-term systematic support (Yang et al.: 2006, p. 37). However, we do see the impact of the recent policy initiatives promoted by Premier Wen Jiabao as the coverage of rural MLSS has risen from 14.46 percent in 2002 to 34.88 percent in 2005. The rate in 2005 was 4 percent of the rural population and it is estimated to rise to 4.6 percent in 2008 and to a high of 6 percent by 2015. Not surprisingly, the urban MLSS is better spread across the country, ranging from 27 percent of those covered in the east to 33 percent of the total covered in the west (table 7.5). By contrast, the coverage in rural China in 2005 just as the program was being rolled out was concentrated heavily in the eastern areas of China that received almost 80 percent of the support. Only 8.8 percent of the support was in western China. In fact, the bias against the rural areas goes even further since when the family’s total income is calculated, all members are counted but spouses who do not have an urban registration are not included for benefits! The highest coverage rate for urban MLSS in 2004 was in Jilin province (11.25 percent) and one can see the impact of the attention paid to social welfare in the northeast as the rates for Heilongjiang and Liaoning are also relatively high (7.79 percent and 6.88 percent respectively). The northwest also has high coverage rates with Tibet, Qinghai, Ningxia, and Xinjiang all over 9 percent (CDRF: 2007, p. 149).

poverty relief and social assistance / 181

The major expansion of the program throughout both urban and subsequently rural China has had major fiscal consequences. The financial commitment from the central government has risen from 0.4 billion yuan in 1999 (26 percent of the total expenditure) to 11 billion in 2005 (57.6 percent of total expenditure). Seven of the more prosperous provinces and municipalities, such as Guangdong, Jiangsu, Zhejiang, Shanghai, and Beijing, do not receive any central subsidy, whereas the amount of support in other provinces varies from 16 percent in Fujian to over 70 percent in Gansu and Guizhou. Perhaps not surprisingly, the old industrial northeast (Liaoning, Jilin, and Heilongjiang) has received the largest amount of central government subsidy for the MLSS program (Leung: 2006, p. 193). As a result, we see contributions by the cities falling. In Sichuan, the percentages of contributions in 1999 were 30.8 percent from the central government, 5.8 percent from the province, and 63.3 percent from the cities. In 2002, this shifted to 62.8 percent from the central government, 9.9 percent from the province, and only 27.3 percent from the cities (Leung: 2006, p. 194). The amount paid out in 2005 for the urban areas was 7.6 times that paid out for rural China (Yang et al.: 2006, p. 37). Those who receive the urban MLSS tend to have lower education levels and tend not to be CCP members. In terms of age, the main cohorts of recipients are between ages 16 and 20 and 41 and 45, and those between the ages of 21 and 35 and over 51 receive the least (Gustaffson and Deng: 2007). The recipients are essentially the new urban poor, products rather than beneficiaries of economic reform. In a March 2002 survey, the Ministry of Civil Affairs determined that the program covered 13.9 million recipients. Of these, almost 49.5 percent were unemployed, 9.8 percent had insufficient income, 30.2 percent were their family members, 4.8 percent were retired, and 5.7 percent were the “three nos” (Leung: 2006, p. 194). A 2003 survey by the Chinese Academy of Social Sciences found a similar profile, with 53 percent unemployed, 26 percent children, and 12 percent elderly, chronically sick, or disabled (Tang: 2000). It is interesting that the percentage of the retired is relatively small; at the end of 2006, it was only 2.4 percent of the total. The combined category of the unemployed and those laid-off (xiagang) amounted to 34.4 percent, whereas the “three nos” amounted to 4.2 percent (MOCA, at http://admin.mca.cn/news/content/ recent/2007523122309.htm). Although presumably welcome, the amount of support provided is not yet extremely significant in terms of total household income. The amount provided by social assistance rose from one yuan in the household income in 1988 to 45 yuan in 2002 (Gao: 2006, pp. 53–54). It is more significant for those families that actually receive the support. Yet, in general, the households do not receive more than 310 yuan per capita over a full year,

182 / providing public goods in transitional china

amounting to only 10 percent of income. In household income, the largest component remains earnings, followed by pensions, and then the MLSS. However, this could still prove crucial for some families. As noted above, the support can cover a substantial percentage of monthly expenditures on food. Since the system is still evolving, these kinds of schemes run into a number of general problems in addition to the obvious inequity between urban and rural provision. First, there is the question of effectiveness of the targeting mechanism. In a country such as China where much of the labor force is engaged in the informal economy, it is difficult to determine the true household income. Only a minority of those who are eligible actually receive support. This might be a result of the lack of financial capacity at the local government level, or the ineffective mechanisms for tracking who really qualifies for support. The program relies heavily on the engagement of the local community to identify eligible citizens and to monitor implementation. Although legally local governments above the county level are responsible for providing the MLSS, the county-level office of the Ministry of Civil Affairs, the local street office, and the township government are responsible for the administration of the system and also for the identification and approval of applicants. If requested, the day-to-day administration is often taken over by the local residents’ committee. In fact, in urban areas claimants apply to the residents’ committee and their materials are verified by the street office or local township government before a final decision is made by the Ministry of Civil Affairs office at the district or county level. The involvement of the residents’ committee in assessing, monitoring, and delivering the benefits has the advantage that it keeps administrative costs to a minimum, but it does bring other problems. Such a process without clear regulations leaves itself open to arbitrariness, favoritism, and corruption. The lengthy application procedure results in applicants being deterred by the process. This might be a greater problem as the program expands farther into the rural areas where administrative oversight will be weaker. There is also the common question of whether such support provides a disincentive to reenter the labor force. Therefore, the average social assistance line is set below what an average worker would bring home if working approximately 43 hours per month. The average paid workers when working full time would earn enough to maintain themselves and 2.5 family members before their incomes would fall below the assistance line (Gustaffson and Deng: 2007). On this basis, it does not make economic sense to give up work and rely only on the MLSS. However, as Gustaffson and Deng also point out (2007), for low-income earners, the situation is different. Workers in the first decile only earn 0.51 yuan per hour, meaning that if they live

poverty relief and social assistance / 183

alone, they will have to work at least 310 hours per month in order to earn more than the minimum subsistence. Concluding Comments The Chinese leadership has begun the process of moving toward a modern welfare state by shifting its social assistance programs from benefits based on specific groups such as veterans or the disabled to one that adopts a means-tested approach that provides benefits to those in financial need. If completed, this will mark a major transformation and open the possibility for developing a social welfare system based on citizenship rather than a system that is arbitrarily defined by the leadership at the national and local levels. However, there is still a long way to go to meet this objective and the MLSS exhibits the same urban bias as other social welfare provisions in China. It has had the advantage of recognizing the existence of urban poverty among those who are products of the new economic policies and the profound demographic changes that China as a whole is experiencing. The more traditional relief and poverty alleviation programs have also undergone significant changes during the reform period. The Chinese authorities have begun to accept a limited role for the market through microcredit schemes and the gradual adoption of international best practices, such as bringing NGOs into project implementation as well as enhancing community participation. China’s statist predisposition still predominates, with most officials, presuming that they know best what the poor need in terms of support, preferring a top-down approach.

This page intentionally left blank

Ch a p t e r Eig h t P rov i di ng S o c i a l We l fa r e: Stat e s, M a r k e t s, a n d Ci v i l Soc i e t y

The provision of social welfare has undergone significant change under reform. The decline in the capacity of the workplace to deliver public goods and services has meant that local governments, with varying amounts of central government support, and nongovernment organizations have had to take on greater responsibility. In this process, the more distinctive features of Chinese welfare provision have been eroded. The state faces challenges similar to not only other developing countries but also developed economies. Debate revolves around common questions of who has the responsibility for key services, who should be the provider, and who should pay. Former sterile arguments in favor of either state delivery or markets have been replaced by more fruitful discussions about how states, the institutions of the market, and civil society can work together to provide more effective service outcomes. Although structures and policies in China have come increasingly to resemble those elsewhere in East Asia, the CCP’s commitment to socialist ideology means that it retains a state-dominant view, even if this means that the services cannot be provided effectively. In practice, the family and extended support networks continue to play a major role, especially in the countryside. However, this burden on the family will be difficult to sustain over the long term. China is facing the same demographic and economic challenges as other countries in Asia that are impelling the state to take on a greater role as a guarantor of welfare coverage. In rural China, as we saw in Chapter Seven, land does not provide a sufficient guarantee for those registered as rural households. The shifts in East Asia have led one observer (Croissant: 2004, p. 529) to write that discussion of an “Asian model” with limited state support as an alternative to a “Western-style” welfare state has lost much of its attractiveness. In China, we even see ambitious attempts, especially for a developing country, to provide basic pension cover or government-backed minimum income support to those

186 / providing public goods in transitional china

in the rural areas. These initiatives could mark a shift from a traditional approach to assistance to a modern welfare state. Yet in practice, the situation remains that what one gets depends primarily on where one lives and whether one works in the formal or informal sector. As our discussion suggests, although there has been significant change, no coherent framework has been developed in China to guide policy development for social welfare. One major issue of importance that we have not touched in detail thus far is how citizens feel about government services and how they prioritize their needs from government—the demand side of the equation. We begin this concluding chapter with this discussion. Second, we look at how the role of government needs to change as alternative service providers begin to expand and conclude by returning to the key questions raised in Chapter One about the nature of social welfare provision under reforms. Citizens’ Perceptions of Government Service Provision1 The perception that people have of service delivery by the state affects the way they assess government performance more broadly. This is even more the case for those who are poor (World Bank: 2004, p. 78). Citizens may expect from their government access to a decent basic education and primary healthcare, safe water to drink, possibly a functioning electricity system, and certain guarantees against the unexpected shocks of life. As we have seen, in theory, in China the primary burden for providing support for these services falls on local government. To date, the main approach to resolving the challenges of local government welfare service provision, suggested by both the World Bank and the Chinese government, has been to improve the revenue stream. In addition to adjusting incentives for officials, this is important but insufficient. There needs to be a thorough rethink about the role of government in providing services and about what kinds of partnerships can be formed to meet policy objectives. A better understanding of how citizens view government, what kind of services they expect, and how they prioritize them will clarify thinking about the changing role of government. Like most countries, China operates a supply-side approach to policymaking in providing public services. The central government sets public policy goals, such as providing nine years of compulsory education, reviving some kind of collective health system for the countryside, or providing some minimum funding guarantees to families in distress. These are laudable goals but they are insufficiently funded mandates, with the burden of implementation falling on local governments. Local governments

states, markets, and civil society / 187

61.3%

55.7%

75.4%

80.5%

77.0%

82.1%

50.0%

55.9%

60.0%

62.3%

70.0%

43.6%

80.0%

52.0%

90.0%

75.0%

100.0%

86.1%

have neither the finances nor adequate incentives to implement such policy directives effectively. As the Chinese saying goes, “the Central hosts the banquet, the Local foots the bill” (Cheng: 1995, p. 72). It might be beneficial to consider also the demand side of the equation, more specifically what do citizens think about the provision of particular public goods by local governments and how do they prioritize their different needs? To answer these questions, we conducted a nationwide survey together with Horizon Research Group to understand which areas of government service citizens approved of and which areas frustrated them.2 When asked about their satisfaction with different levels of government, the respondents confirmed the research of others that reveals that many residents “disaggregate” the state (O’Brien: 1996; Li and O’Brien: 1996; Bernstein and Lü: 2000). Significantly, survey respondents express high levels of satisfaction with the central government and declining levels of satisfaction with each lower level of government. This confirms the work of others (see Shi: 2001; Chen: 2004; Li: 2004). Over 80 percent of respondents expressed that they were relatively or extremely satisfied with the work of the central government (figure 8.1). Satisfaction drops to the mid-70 percent range for the provincial level and then lower for the district and county (52 percent in 2003, rising to 61.3 percent in 2005) levels and to the mid-50 percent range for the townships, villages, and street office levels. Although the levels of satisfaction are lowest at the basic level, it is interesting to note that satisfaction rose

40.0% 30.0% 20.0% 10.0% 0.0% 2003 Central

Figure 8.1

2004 Provincial

District/County

2005 Township/Village

Citizens’ Level of Satisfaction with Different Levels of Government

Source: Author’s Survey with Horizon Research Group.

188 / providing public goods in transitional china

from 43.6 percent in 2003 to 55.7 percent in 2005, perhaps reflecting the policy initiatives of Premier Wen Jiabao to improve the lot of those in the rural areas. Yet, at the same time, the level of satisfaction with the central government has declined slightly. This is important as the local government provides most services. The level of satisfaction with government falls progressively irrespective of where one lives. The declining satisfaction in rural China as one gets to the township and village that has also been observed by others holds true for urban China in our sample, although the drop in satisfaction is not as acute as that in the rural areas. At the lowest levels of government in 2005, satisfaction (both extremely and relatively) was highest for those living in major cities (66.9 percent), falling to 56.1 percent in towns and townships, and only 51.5 percent in villages. For those living in villages, this was a marked improvement over 2003 when those satisfied amounted to only 37 percent (table 8.1). These figures confirm the general impression that the quality of governance is better in larger cities than in rural China where training opportunities are limited and education levels are lower. One clear finding is that a process of urbanization would improve governance since in virtually all categories citizens in major urban areas are more satisfied than those living in small towns, townships, or villages. The process of urbanization would also facilitate the expansion of social insurance and welfare provision to a wider range of the population. To correlate the level of importance citizens attached to certain services with the level of satisfaction with local government service provision, we devised a simple matrix (table 8.2). This provides a preliminary attempt Table 8.1 Chinese Residents’ Satisfaction with the Lowest Level of Government Service and Performance by Place of Residence, 2005 Cities

Very Dissatisfied

Small Towns and Townships

Number

Percentage

51

5.2

Number

Villages

Percentage Number Percentage

32

5.8

301

11.6

Not Very Satisfied

198

20.4

145

26.3

775

29.8

Relatively Satisfied

569

58.5

263

47.6

1109

42.6

Extremely Satisfied

81

8.3

47

8.5

231

8.9

Not Stated

14

1.4

11

2.0

25

1.0

Undecided Total

59

6.1

54

9.8

163

6.3

972

100.0

552

100.0

2604

100.0

Source: Author’s survey with Horizon Research Group.

states, markets, and civil society / 189 Table 8.2

Government Service Satisfaction/Importance Matrix Template

Area D High level of importance, low level of satisfaction (Work is of poor quality and urgently needs improvement)

Area A High level of importance, high level of satisfaction (Work is of good quality and should remain a policy priority)

Area C Low level of importance, low level of satisfaction (Work is of poor quality but does not require immediate attention)

Area B Low level of importance, high level of satisfaction (Work is of good quality but is of limited importance)

Source: Author’s survey with Horizon Research Group.

to understand which areas of government service in our survey citizens approved of and which areas frustrated them. Citizen satisfaction with the provision of certain public services reveals some interesting insights that are helpful for thinking about the changing role of government and what local government should concentrate on to improve satisfaction levels. Across the three years of the survey there was little variation in the five aspects of local government work that received the highest ratings. In 2003, these were family planning, water and electricity supply, road and bridge construction, oversight of religious worship, and traffic management. The only change in 2005 was replacement of oversight of religious worship by management of middle and elementary schools. However, it should be pointed out that only family planning in 2005 received a rating of “somewhat satisfied.” Three of these five services relate to the provision of physical infrastructure whereas none relate to pressing social policy concerns. Most are services that the old central planning system was good at providing. By contrast, the five areas of government work that caused the greatest dissatisfaction varied from 2003 to 2005. In 2003, for the sample as a whole, respondents were most unhappy with combating corruption, job creation, unemployment insurance, tax management, and hardship family relief. In 2005, employment creation and medical insurance replaced unemployment insurance and tax management. These areas of dissatisfaction relate much more directly to household economic and social concerns and they are mainly problems that have arisen as a result of the reforms. When we apply the matrix to correlate the level of importance people attach to a particular service with their satisfaction with government performance, we get a list of high-importance and low-satisfaction items that is more closely related to the social and economic problems faced by households during the transition. In 2003, the areas identified by

190 / providing public goods in transitional china

respondents who thought government work was poor and in urgent need of improvement were job creation, unemployment insurance, hardship family relief, medical insurance, and environmental health (figure 8.2). We also find that the family-planning policy that enjoyed the highest level of satisfaction was deemed to have the lowest level of importance. In 2005, punishing corruption was ranked as having a high level of importance together with social order and security (figure 8.3). This would suggest that the areas that citizens really wish government to concentrate on are job creation and providing basic guarantees to protect against the shocks of the transition to a market economy. Labor and medical insurance are high priorities for all residents. In our 2005 survey, one of the local government agencies that people thought should be expanded was the Labor and Social Security Bureau. Given that there will be a limit to how much extra revenue the government can provide to finance the provision of public services, it must meet objectives by reducing costs and by focusing more clearly on the kinds of services that local governments can and should provide. In fact, in 2005, 46.4 percent of respondents said they

0.5

Medical Services Traffic Management Management of Primary/Middle School Education Cultural/Athletic Facilities Road/Bridge Construction

Medical Insurance Unemployment Insurance Hardship Family Relief

0.4

Market Management

Job Creation Environment Health

0.373

Land Management

Importance

Tax Management Attracting Investment Religious belief

Punishing Corruption

Water/Electricity Supply 0.3

Environmental Governance Public Order and Safety

2.54 Family Planning 0.2 1.9

2.1

2.3

2.5

2.7

2.9

Level of Satisfaction

Figure 8.2

Government Performance Matrix, 2003 (all respondents)

Source: Author’s survey with Horizon Research Group.

3.1

states, markets, and civil society / 191 3.6

3.5

Public Order and Safety

Punishing corruption

3.4 3.337 3.3 Importance

Management of Primary/Middle School Environmental Governance Education Traffic Management Hardship Family Relief Medical Insurance Environmental Health Road/Bridge Construction Unemployment insurance Water/Electricity Supply Medical Services Land management Tax management Attracting Investment Family Planning Market Management Cultural/Athletic facilities Job Creation

3.2

3.1

3

Religious Beliefs 2.722

2.9 2.3

2.4

2.5

2.6

2.7

2.8

2.9

3

3.1

Level of Satisfaction

Figure 8.3 Government Performance Matrix, 2005 (all respondents) Source: Author’s survey with Horizon Research Group.

would not be willing to pay increased taxes in order to receive improved services. This was down from 54.8 percent in 2003, but the number of those willing to pay increased taxes dropped slightly from 30 percent to 28.3 percent. Further, in 2005, 60.4 percent wanted to see a decrease in taxes, up from 52.6 percent in 2003. Thus, it is likely that we shall see a continuation of the trends outlined in the previous chapters, with the state claiming for itself the dominant role in service provision while, in practice, ceding more space to market and civil society institutions. If our survey is in any way representative, to satisfy citizens’ needs, government will have to facilitate the further development of alternate service providers and form new partnerships with them. Local governments alone will not be able to provide the necessary services. The survey reveals one further problem. Generally, citizen satisfaction is highest with the provision of physical infrastructure and less so with the provision of economic and social services that have an impact on the household. However, it is precisely the physical infrastructure services, such as road and bridge maintenance and the supply of water and electricity, that

192 / providing public goods in transitional china

can be, and already are, most easily contracted out. Citizens view local government as less adept at providing the kind of social support that needs to accompany the economic transition. Of States, Markets, and Civil Society To improve the quality of service provision, government will have to make more effective use of the market and not-for-profit-organizations (Saich: 2002b; 2004). As the World Bank (2005a, p. 36) has noted, the state can empower individuals by bequeathing rights and cushioning citizens against critical adversities; the market empowers them by providing choice and opportunity. Yet to ensure that the use of markets really does empower and does not result in service provision being reduced, for those who are vulnerable or in problems of monopoly or asymmetric information, the role of the state itself has to undergo a fundamental shift. First and foremost, an effective, as distinct from a strong, state structure is a precondition for any hope of successful reform. Certainly the role of the state in the economy and society will change with a narrower set of interventions and less direct administrative interference. However, an increase in state capacity is a prerequisite for an effective market to function (Polanyi: 1944). The state must adjudicate the increasingly contentious nature of economic market transactions while ensuring that there is sufficient revenue to provide minimum social services and welfare guarantees to protect those who are vulnerable in the shift to a market economy (World Bank: 1997c, pp. 4–5). Far from making governance easier, the shift away from the direct provision and administration of services makes the work of government more complex. Chinese government at all levels has added new regulatory roles to their old monopoly functions, and many, if not most, administrators are less familiar with the work of being an indirect provider or a regulator. To manage in a markedly different framework, officials need to acquire additional skills and capacities. The reform of pensions that we discuss in Chapter Six provides a good example of the fundamental challenges of such change. The government is shifting responsibility for the pensions of the privileged sector of the working class from SOEs to the government through a system of local collection, pooling, and disbursement. However, many local administrations lack the skills to manage these funds effectively and because of insufficient transparency and regulation, many have simply used pension funds as state assets to be invested in their pet development schemes. Effective management also requires a set of actuarial skills that are currently lacking. These are just two small examples of the new kinds of skills that government officials will have to learn in order to manage social

states, markets, and civil society / 193

welfare effectively. Given the problems of adjustment, it is little wonder that administrative reform is resisted by many within the bureaucracy. The need to manage a different kind of economy that is increasingly integrated globally is, however, forcing change in governing capabilities, in the forms and mechanisms of governance, and in the style of governance. The powers of government have been redistributed both vertically and horizontally. Vertically, some powers have been redistributed from the national government to international organizations such as the WTO, while at the same time responsibility for most public service provision has been decentralized to local governments. We have seen great geographical variations in service provision that derive from this decentralization. In terms of horizontal redistribution, the previous chapters have discussed a number of examples of government contracting out or ceding service provision to the market or to civil society organizations. In general, there is a shift from the old “command and control” forms of regulation to those characterized by contracts and covenants between public and private actors and to information management. Service delivery will need to be more consumer focused and guided by cost effectiveness than it was in the past. Thus the government, although directly responsible for public policymaking, will need to transform its role from a public sector provider to a service facilitator. Government performance will thus be assessed on its ability to effectively manage partnerships with for-profit and not-for-profit organizations rather than on its ability to fund and provide services (Goldsmith and Eggers: 2004, pp. 10, 21). These processes create a more complex set of government relationships, sometimes referred to as “networked government,” where decisions are made within complicated networks with pluri-centric forms of governance. This presents a major challenge for effective governance since states are organized in terms of hierarchy, whereas markets operate horizontally with multiple centers. Dali Yang (2004) provides an optimistic assessment of progress, claiming that China has undergone a “tectonic shift” in the way it conducts itself and has made progress in developing the kind of regulatory state that can oversee a market economy. We have suggested that although there may be substantial new legislation and institutional development, policy implementation remains a significant problem. Despite official claims that the CCP is moving from being a revolutionary party to being a governing party, it still has a long way to go to develop the required new and relatively independent regulatory bodies to oversee privatized sectors, especially for private natural monopolies such as energy supply, water, railroads, telecom, air transport, and broadcasting. The key problem is that many of the activities delegated to the market have been delegated by default rather than by design. We have seen

194 / providing public goods in transitional china

this clearly in Chapter Four with respect to healthcare, where expensive, invasive procedures and the issuing of, often unnecessary, drugs have come to dominate over less expensive preventive medical services that have a public goods character. Especially in the 1990s, it seemed that the leaders’ reification of the market was combined with a poor understanding of what markets could and could not manage well. The lack of attention paid to putting in place effective rules and regulations to govern the market has led to problems familiar to other transition economies, with poorly regulated user charges for services and a lack of transparency and accountability leading to corruption in the use of public funds and a decline in trust in public service agencies. Especially important for service delivery are Public Service Units (PSU, shiye danwei). With a genuine not-for-profit sector lacking in China, these organizations are required to play an important role in service delivery. There are estimated to be around 700,000 PSUs in China, down from 1 million at the turn of the century. The majority of these PSUs are education institutions, hospitals, and research institutions. They employ some 30 million people, about 41 percent of public sector employment (World Bank: 2005a, p. 1). The State Council (1998) defines them as “social service organizations established by the state for the purpose of social public benefit.” Current policy intent is to move these PSUs out of the state sector to become either for-profit or not-for-profit organizations. However, attempts to marketize them have produced unexpected results, with nearly half of their funding derived from fees that are charged, most of which goes to cover staff benefits and welfare. This has resulted in a strong incentive to distort the markets in which they operate (World Bank: 2005a, p. vii). Oversight and a regulatory framework to guide their work are very weak. The 2005 World Bank study cites the well-known example of Suqian in northern Jiangsu that was renowned for the privatization of its medical facilities. Beginning in 2000, the town privatized all 124 township and 6 county-run hospitals. However, most reports on the results have been negative, claiming that service declined as prices increased. According to the World Bank (2005a, p. 6), between 2000 and 2002, average outpatient fees rose from 90 yuan to 113 yuan in municipal-level hospitals, from 51 yuan to 143 yuan in county-level hospitals, and from 32 yuan to 35 yuan in township-level hospitals. These negative outcomes led to further experimentation to transform PSUs into not-for-profit institutions that might inspire more trust from users. A major problem in developing effective partnerships with the NGOs is that the regulatory framework for the management of the sector is unclear and seems intended to discourage organizations from developing independent of the state. We saw this with respect to the attempts to engage

states, markets, and civil society / 195

community-based organizations in the round six application to the Global Fund. Senior CCP leaders remain ambivalent about the development of civil society. Although keen to mobilize private resources from society, they prefer that the sector be developed within a highly restrictive legislative and organizational framework that ensures CCP and state control. This derives in part from the Leninist organizational predisposition that views with suspicion institutions outside of the party-state’s direct control and seeks to incorporate as much of society as possible into its networks (Jowitt: 1992). Yet the leadership is aware that the next phase of reform will shrink the role of the state in people’s lives even further. Concerns about possible loss of control have been heightened by the spate of democratic movements in former Soviet-dominated countries—the so-called color revolutions. These factors, together with the fear of social unrest that might result from the negative consequences of reform, have combined to cause the party to hold back organizational plurality. In all societies the state plays a fundamental role in structuring the nature of civil society, not only in terms of the regulatory framework but also in terms of norms and assumptions about the respective roles of state and civil society. China is no exception. It is impossible to think of civil society independent of the state. For example, the constitutions of many NGOs in the United States are influenced by the norms of the national state, whereas in countries such as the Netherlands, Canada, and those in Scandinavia, NGOs receive between 50 and 90 percent of their funding from the government. The ambivalence about the NGO sector is clearly seen in the regulatory framework that favors larger organizations, many of which have been set up by the government or party agencies. This preference is evident in the State Constitution as well as in the specific regulations for the NGO sector. Although the current State Constitution permits citizens considerable freedom of association, such freedoms are rendered vulnerable by the acknowledgment of the CCP’s leading role as outlined in the preamble. The prevailing ethos within the party that it is capable of representing all interests in society imposes severe constraints on the autonomy of civil society organizations in carrying out their activities. The 1998 “Regulations on the Registration and Management of Social Organizations” provide a clear example of the attempt to incorporate such organizations more closely with existing party-state structures (for details, see Saich: 2000). There are several key features of the legislation that are important for a traditional Leninist reordering of civil society organizations. First, all organizations must find a professional management unit that will act as a sponsor—usually referred to as the sponsoring unit and more commonly called the “mother-in-law.” After finding a sponsor and gaining its

196 / providing public goods in transitional china

approval, the paperwork for the organization is sent to the Ministry of Civil Affairs or its relevant agency. This sets up a two-tier registration system where affiliation precedes registration. In contrast with previous regulations, the 1998 regulations set more onerous requirements for the sponsor, resulting in many organizations being unwilling to take on such a role. The cumulative nature of this process is such that organizations tend to avoid formal registration, thus undermining government intent. Second, “similar” organizations are not allowed to coexist at the various administrative levels. Thus, there cannot be two welfare associations or two national charity federations. In theory, this limits representation to a smaller number of manageable units and has been used to deny registration for some groups. It ensures that the “mass organizations,” such as the All-China Women’s Federation and the All-China Federation of Trade Unions, enjoy monopoly representation and cannot be challenged by independent groups seeking to represent similar interests. Social organizations must register with the appropriate civil affairs department at the county level and up. This makes it impossible for local groups to enroll members from different areas, thus limiting the potential for the spread of grassroots organizations that could develop broader horizontal representation. The intent of this legislation is clear: it is to mimic the compartmentalization of government departments and limit horizontal linkages. It favors those groups with close government connections and discourages bottom-up initiatives. Furthermore, it keeps people with different opinions on the same issue from setting up “opposing” interest groups. Other aspects of the regulations further hamper bottom-up initiatives or initiatives by the disadvantaged or poorer sectors of society. The need to have substantial assets and the required paperwork for registration make it difficult for those groups that lack good connections and a relatively sophisticated organizational apparatus to register. In practice, these kinds of regulations encourage groups to operate underground or to find alternative mechanisms for registration. These administrative hurdles notwithstanding, the growth of NGOs has been a remarkable feature of the reform period and they are playing a greater role in social welfare provision. Indeed, there has also been a growing official recognition that certain kinds of NGOs can play a positive role to help the party meet its objectives with respect to social welfare. By the end of 2006, there were some 192,000 registered social organizations, defined as community entities composed of a certain social group with common intention, desires, and interests (MOCA: 2007 at http:// admin.mca.gov.cn/news/content/recent/2007523122309.htm). This is an increase over the previous high point of registration in 1996 when there

states, markets, and civil society / 197 Table 8.3 Social Organizations and Civil Noncommercial Organizations (thousands) 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Social Organizations

184

181

166

137

129

131

133

142

153

171

192

% increase on previous year

1.7

–1.6 –9.0 –2.1 –4.6 –1.6

3.1

6.8

7.7

11.8 12.3

Civil — Noncommercial Organizations

—

—

6

23

82

111

124

135

148

161

% increase on previous year

—

—

—

383

356

35

11.7

8.9

9.6

8.1

—

Source: MOCA: 2006, p. 6; MOCA: 2007, at http://admin.mca.gov.cn/news/content/recent/ 2007523122309.htm.

were 184,000 registered social organizations (MOCA: 2003, p. 159). In the late 1990s and in the first two years of the new century, there was a drop in the number due to an extensive review of registered organizations that led many to lose their official registration (table 8.3). The majority of social organizations (95,263) were registered at the county level (MOCA: 2006, p. 94). In 2005, not surprisingly, economically developed provinces such as Jiangsu, Shandong, and Zhejiang had the highest number of social organizations. Together with Fujian, Guangdong, and the two municipalities of Beijing and Shanghai, they had 34.3 percent of the national total. By contrast, the eight western provinces, excluding Sichuan and Chongqing municipality, had only 14.7 percent. Sichuan has 12,672 social organizations (MOCA: 2006, p. 150). A more expansive definition would include the converted PSUs and the growing number of civilian noncommercial enterprises. After the introduction of regulations to govern them in 1998, this latter category increased rapidly, from 6,000 in 1998 to 148,000 by 2005. Of these, the majority were in the fields of education and health: 75,812 and 27,179 respectively (MOCA: 2006, p. 94). These two categories also cover organizations set up by the party and state agencies to handle welfare concerns and the various organizations that have been progressively spun off from government; organizations that operate independently but enjoy government support at some level, such as the environmental group Friends of Nature; or organizations that have grown out of party-state organizations, such as Rural Women Knowing All, which takes up a range of issues facing women. The vast majority of these organizations are run by urban elites who undertake work on behalf of the groups they claim to represent, for example, migrants

198 / providing public goods in transitional china

or poor rural women. Very few have been organized at the grassroots level or work in a community-based or participatory mode. Yet, slowly by default these kinds of organizations have also been expanding. Chapter Six outlines the growth of NGOs involved in work related to HIV/AIDS, but their acceptance is still vulnerable to the whims of local officials. Activists and organizations that advocate too stridently may easily run afoul of local governments and find their activities curtailed. This has consistently been the case in Henan province where the transmittal of the disease via blood selling was first revealed. In November 2006, a meeting involving hemophiliac activists was banned by local police. Earlier, in April 2006, Shanghai police broke up a news conference of a group of hemophiliacs who claimed that they had contracted HIV/AIDS through contaminated blood transfusions (Reuters, November 25, 2006). By contrast, officials in Sichuan province called for greater involvement of NGOs in work dealing with the spread of HIV/AIDS. The deputy director of the provincial Disease Prevention and Control Bureau acknowledged that because many of those infected were sex workers or in the gay community, “it would be very difficult to reach” policy objectives by using “the traditional method—mobilizing government institutes at all levels”—as was the case in other public health crises such as SARS. As a result, he called for existing regulations to be revised and greater scope to be given for the development of the NGO sector (Xinhuanet, May 1, 2007). These two examples reveal the differing bureaucratic interests in dealing with HIV/AIDS. Whereas health officials have been generally more sympathetic to contacts with groups such as CSWs and IDUs, public security officials have often taken a tougher line, regarding them as engaged in illegal activities. Following the trend in many other countries, the idea of NGOs as service delivery agents has been gaining acceptance (Edwards et al.: 1996). The ideal is for the NGO to act as a bridge and mediator between the government and society. Essentially, the organization becomes an arm of the government, carrying out the latter’s objectives, and the organization becomes state- and donor-dependent for carrying out its activities. In a survey of various types of NGOs, carried out by the now defunct China Development Brief (Wexler et al.: 2006, p. 151), foreign donors were actually in the majority as funders for grassroots NGOs and governmental departments, and foreign donors were equal funders for national government-organized NGOs. This bridge function of NGOs is preferable for a specific task or for meeting a short-term objective such as implementing a vaccination campaign or reaching out to those in marginalized communities suffering from HIV/AIDS. However, it might not be so effective to meet the objective of longer-term sustainable development and the

states, markets, and civil society / 199

mobilization of resources within the community that neither the government nor an NGO serving as a bridge would be able to provide. Thus, it is not surprising that the vast majority of NGOs in China stress their support of government policies and that they are working to help the government meet its commitment to build a “harmonious society.” For most NGOs, their work strategy is based on maintaining close government relations to ensure their continued existence and political acceptance. Some may even nest within government bodies using official personnel and resources to carry out their programs and to profit from this special position to seek to influence policy. In the 2006 China Development Brief survey (Wexler et al.: 2006, p. 149), 9 out of 12 responding grassroots NGOs claimed to have close and cooperative relations with the government, and 17 out of 19 stated that close relations would help them accomplish their goals. Not surprisingly, all the national government-organized NGOs replied positively to these two questions. That NGOs could and should take on service delivery functions is slowly gaining acceptance in major cities, but they are still seen as second-class organizations. Shanghai provides a good example. In 1996, the government invited the Shanghai Young Men’s Christian Association (YMCA) to manage a community center in Pudong district, the first such experiment in China. By the end of 2006, the YMCA was running six such centers in Shanghai, with the government providing free premises and sometimes utilities or equipment as well. As a result, over one half of the YMCA’s projects were fully or partly funded by the local government. In October 2006, the Pudong government issued a policy circular on the funding of such organizations for public service provision. In addition to getting procurement contracts, NGOs would also be eligible to receive cash grants (or land, buildings, and equipment) from the local government as well as to receive help with payment of bank loans. Public and corporate support was also encouraged. However, implementation is inconsistent and preferential treatment for government agencies persists. Thus, each of the six centers in Shanghai has to negotiate its own arrangements and they are liable to the whim of local leaders—that is especially problematic if leaders change. With many of the public service units being converted into not-for-profit organizations, they not only compete with more autonomous NGOs but also still receive preferential treatment from the local government in terms of funding and other support. It is clear that even in a metropolis such as Shanghai, there is no clarity over the division of responsibility between the government and the NGOs in public service delivery. Interestingly, Yang Tuan, a researcher at the Chinese Academy of Social Sciences, although acknowledging Shanghai as a pioneer in this conventional role of NGO service delivery,

200 / providing public goods in transitional china

notes that for less conventional services, such as advocacy or working on HIV/AIDS, the atmosphere is less accommodating and there are more such organizations in Beijing (China Development Brief, November 2006, pp. 14–17). In rural China, community-based organizations are expanding in number and it is clear that alternatives are emerging to take over functions that were previously considered the preserve of the government. Many of these organizations are based on traditional associations such as temples or social networks, including clans and lineages. The role of these informal institutions has been explored by Hu (2007) and Tsai (2001, 2002, 2007) who show how membership of social networks becomes important given the low investment in public goods by village government. Tsai (2007) terms these “solidary groups” and shows that if officials are “embedded” in such groups they will feel more obligated to provide public goods as fellow group members can use the norms and the network to punish them. Hu (2007) shows the importance of “village trust” in setting up Rotating Savings and Credit Associations (ROSCA) in Xiangdong village, Zhejiang province (on the importance of trust in informal institutions, see Hermann-Pillath: 2000). Almost all of the villagers have participated in a community-based ROSCA that can be used for a number of different functions. For our purposes, one of the most salient functions is to finance family emergencies such as medical expenses and school fees. Those who join this kind of ROSCA tend to be from poorer families and thus the scale is rather small, with fund payments limited to 1000 yuan over three months. Poor households have set up such ROSCAs each year to cover their expenses and one villager has set up one that he has headed for six years running to cover his basic livelihood expenses. About half of the village participates in this kind of ROSCA. Such organizations are important given the problems that rural dwellers face in accessing formal financial institutions. Bad debt in the ROSCAs is rare (with a default rate of less than 1 percent) and the financial risk is very limited. Research by Tsai (2001, 2002, 2007) reaches similar conclusions that in some areas social capital substitutes for governmental performance, and officials allow social institutions to take over the provision of public goods. In her survey of 316 villages in Shanxi, Hebei, Jiangxi, and Fujian, Tsai discovered that lineage or religious organizations in 54 villages had organized public projects. She makes an interesting distinction between the role of temples and churches in enmeshing local officials in networks that provide moral compulsion to deliver better public goods. Village temples are described as “encompassing and embedding” (2007, p. 120), whereas the churches are “encompassing but not embedding,” even more so as local CCP officials are in theory banned from joining the church. She describes

states, markets, and civil society / 201

the village temples as reinforcing state authority and legitimacy, whereas the history of churches in China means that they are viewed suspiciously and as dangerously connected to foreigners. Also the temples serve only a specific community and do not necessarily make a claim of union with some broader or higher entity. However, in some areas local officials may rely on the church to provide public goods and services. In the Catholic Duan village in north China, the church organized villagers rather than ask local authorities to build a road. It was said that even if the village authorities had the money, no one would listen to them; by contrast, no one would refuse the church (Wu Fei: 1997). As a result of this and other examples, Wu concludes that local religious elites have become a strong group in the local politics of resource allocation. In his view, the penetration of state power into these communities depends not only on village officials but also on the cooperation of the local elite. “The reputation of village officials is markedly lower than that of the local religious elite” (Wu Fei: 1997, pp. 56, 62). Zhang and Baum (2004) review the Sanchuan Development Association in Qinghai province, an organization they describe as a “relatively autonomous, indigenous NGO.” They claim that this kind of organization is being formed with increasing frequency in rural China. The association is said to be only “loosely and intermittently linked to the state,” running over one hundred development projects that cover a wide range of public goods provision, social welfare, and poverty alleviation. However, it is important to note that the association probably would not have developed without the input of a foreign English-language teacher who actively raised funds for it from within the international community. These alternatives for mobilizing community and outside resources are especially important as the incentives to encourage village officials, village elections notwithstanding, to provide public goods are weak and there are very few institutionalized channels for gaining support from higher-level governments to provide public goods in the village (Tsai: 2002, p. 25). Certainly, there is no direct incentive for township officials to invest in village infrastructure. Thus, although they might finance a road to link the village to the township, they are unlikely to invest in improving communications within the village. The use of township funds will go to meeting the obligations laid down by the county-level government (Cai: 2000, pp. 801–802). As we saw in Chapter Seven, government-organized NGOs and foundations have become increasingly important for providing funding for poverty alleviation programs and for helping the government implement its policies effectively. As of the end of 2005, there were 975 foundations nationwide, of which 92 were operating at the national level and 883 at the

202 / providing public goods in transitional china

provincial level (MOCA: 2006, p. 94). By the end of 2006, the number had risen to 1,144. The confusing Public Welfare Donations Law that took effect in September 1999 favors those organizations that are close to government. It acknowledges tax relief for corporate and individual donors, but it is not specific about the levels and conditions. This means that the environment for giving will remain uncertain, with the results dependent on negotiations between the donor, the recipient, and the local tax authorities. This too will favor those large organizations with senior ex-officials in leadership positions over grassroots organizations. Presumably, this is intentional, as the central government has been consistently concerned that tax relief for donations to NGOs could result either in money laundering or in a reduction in the funds available for its own coffers. Also, given the difficult financial circumstances of a large number of local authorities, many are likely to resist tax reductions unless it is for an organization with powerful local connections. For individuals, 2 percent of salary may be offset against tax for charitable donations and this may be raised to 12 percent but there are only around 20 organizations that are approved to receive charitable donations. Despite this restrictive legislative environment, there is a growing official recognition that NGOs have a role to play in welfare provision broadly defined. Former CCP general secretary Jiang Zemin and former premier Zhu Rongji recognized the necessity for further development of NGOs at the party and state congresses of 1997 and 1998. In his speech to the Fifteenth Party Congress, Jiang stressed the need to “cultivate and develop” what he termed “social intermediary organizations” (Jiang: 1997). The 1998 plan for restructuring the State Council stated that many functions appropriated by government organs should be given back to society and handled by the new social intermediary organizations. In introducing the plan, State Councilor Luo Gan stated that “government has taken up the management of many affairs that it should not have managed, is not in a position to manage, or actually cannot manage well.” This overload detracted from the government’s capacity to carry out its work effectively. Luo thus called for social intermediary organizations to be expanded (Luo: 1998). However, it was clear that this was not to be a free-for-all but a closely managed process. More recently, Premier Wen Jiabao vowed to turn over responsibility for more activities that the government should not be engaged in to enterprises, NGOs, and intermediary organizations (China Daily, March 13–14, 2004). The monopoly control enjoyed by large mass organizations restricts the capacity of organizations to operate in all arenas of civil society and to be able to pressure government in the name of public interest. This is not unusual in East Asia where states have been dominant. Even in Japan,

states, markets, and civil society / 203

regulation before the Kobe earthquake was restrictive and the government did not pay much attention to the development of civil society organizations. In Japan and South Korea, the strong state has dominated the relatively small nonstate sphere, but in both countries they have had to adapt as their own weaknesses to provide welfare and relief became apparent. China is beginning to face the same challenges. Thus, despite the restrictive environment for civil society development, China has begun to develop the same categories of organizations that one finds elsewhere. Most fall into the category of mediatory organizations that take on roles that the state can no longer fulfil and they often act as a bridge to society or contribute to state objectives by providing services and enhancing capacity. Many of these organizations are run by urban elites who seek to work on behalf of certain groups. By contrast, although the numbers may be growing, fewer organizations appear to be developing from the grass roots up and to be genuinely community-based. During the reform period, local governments have been experimenting with service delivery through a variety of mechanisms: contracts and concessions to not-for-profit and for-profit organizations, public-private joint ventures, and informal and voluntary cooperation. The reason for the rise in such experimentation is similar to that elsewhere in the world. Most governments have insufficient funds to extend services as much as they would like or to subsidize the welfare programs that they wish to support. Second, whether true or not, there seems to be a popular perception that publicly provided services are more embroiled in bureaucracy, are of lower quality, and are liable to be mired in corruption. Indeed, there are many practical examples of goods for which people can pay (transport, piped water, or housing) that can be delivered more efficiently by the private sector. Yet, there is no necessary logic that a private monopoly will perform any better than a government monopoly. Thus, an important role for government regulation is to create and maintain competition by providing an enabling external environment. Concluding Comments As the chapters in this book have shown, the delivery of social welfare in China has changed significantly during the reform period. The old workplace-dominated model has given way to a more complex set of relationships among the workplace, local governments, the institutions of civil society (including the market), and the family. There is a clear shift to a more modern approach to welfare provision to replace traditional practices, but the transition has not been completed and no clear theory or framework has been developed to guide it. A hybrid system has emerged

204 / providing public goods in transitional china

that shares features with other systems in Asia and beyond, while retaining special features shaped by the pre-reform institutional legacies in China. The challenges provided by the shifting demographics with a rapidly aging population and a more diverse workforce, and with large numbers leaving the land in search of work in the urban areas, placed a major stress on the traditional systems of welfare support. Although slow to respond, the government is taking on a greater role in welfare support, especially in providing baseline guarantees for both the rural and urban populations. This trend follows developments elsewhere in Asia where such pressures have made it difficult for the state to push too great a responsibility for welfare onto the family. In places such as South Korea and Taiwan, the expansion of the state’s role in welfare provision has been aided by a democratization that has extended the clientele eligible for benefits and has also brought in new benefits. In China, expansion of the state’s role has been pushed by the Hu Jintao and Wen Jiabao leadership’s policy priorities and underpinned by the residual impact of the official socialist ideology. The CCP claims to represent the whole population, part of its rationale for not acknowledging political competition; under Hu and Wen, it has promoted the objective of building a “harmonious society” by the year 2020. This latter objective includes providing a basic livelihood guarantee to as much, if not all, of the population as possible. The claim to be a socialist party has made the leadership vulnerable to criticism from both the “old” and the “new” left who feel that overreliance on the market has created new inequities and has adversely affected the vulnerable and marginalized populations. We saw this particularly in the attacks on the inadequacies of the health system wherein lie the inequities of access and costs that the de facto market-based reforms have brought about. However, the severity of the impact has been heightened by the pre-reform institutional legacies, such as the household registration system that condemns urban and rural dwellers to different levels of welfare. The Maoist notion of “self-reliance” reinforces the current reality that where you live dictates what you get. In China, as elsewhere in Asia, by the late 1990s, public pension systems and unemployment schemes were becoming the norm. The question of old-age income security still vexes most countries; in this respect, China has been unusual in trying to extend its pension scheme to the rural areas. The government has moved to expand its role as provider of last resort, from support for the “three nos” in urban China and the “five-guarantee” households in the rural areas to provision that is based on a means-tested minimum income support system. Although the level of benefits varies depending on where one lives, this could represent a step toward a welfare system based on citizenship. This would amount to what T.H. Marshall

states, markets, and civil society / 205

refers to, in the context of Western Europe, as “social citizenship” that encompasses the right to economic welfare and social security. In China, unlike in Europe, this might even precede “political citizenship” (Marshall: 1950, 1964). Increasingly, voices are being raised to suggest that social welfare should help promote social justice and that fiscal transfers should play a more significant role in redistributing the fruits of reform. However, a fundamental shift is still a long way off. China’s four-legged stool of welfare support is operating with two long legs (state and family) and two truncated legs (civil society and the market), thus creating imbalance in the system and perpetuating existing inequalities. During the reform period, the state has taken on more responsibility for welfare provision at both the central and local levels. As explained, local governments in China generally play a far greater role in public goods provision than in most other countries. As the fiscal buoyancy of the central state has been enhanced, it has been able to make larger transfers from available funds to cover benefits such as schooling costs for rural students, the MLSS, and contributions to the revived rural cooperative medical insurance scheme. But it is debatable whether such transfers are sufficient to make a significant difference and in some cases central transfers may not compensate local governments for income previously derived from local fees and levies. The main burden still falls on underfunded local governments, leading to great variation in provision and even nonprovision of mandated services in some cases. CCP distrust of the market and civil society means that government involvement and control will remain significant. The capacity of the state to carry a larger role is premised on the continuation of the high rates of growth that have increased central revenues. This has meant that, to date, the government has not had to choose between investment in the economy and expanded social welfare—in recent years, it has been able to pursue both. It remains to be seen whether the CCP would maintain its commitment to expanded social welfare should there be an economic slowdown resulting in lower government revenue. The government’s role may well be crucial, but it is still limited to minimal last resort and emergency relief coverage. Thus, despite the new policies we have discussed in previous chapters, the family and extended kinship networks will continue to play a major role in welfare provision for many in the foreseeable future. This is especially the case for those in rural China where government support can be only residual at best. Oldage security will still be the responsibility of the family in most rural areas as will major medical expenses, and this will be the case even for many in urban China. Pooling family resources or establishing lending systems such as the ROSCAs may be the only ways for many families to cover such

206 / providing public goods in transitional china

expenses. Yet the changing demographics mean that overreliance on the family in the future will become increasingly problematic. With the onechild policy in urban areas and two-children policy in the countryside, there will be a heavy burden on the working children. Unless the government allows the other two legs of the intermediary organizations between the state and the family to expand their roles, it will have no alternative but to expand the government’s role even further into the funding and provision of social welfare. Not to do this will thwart the CCP’s objective of building a “harmonious society” and may well increase the chances for social instability. This may lead eventually to hard choices for government investment. Neither markets nor the institutions of civil society are fully playing their role in social welfare provision. As we have noted above, much of the expansion of the market and NGOs into welfare provision, and public goods in general, has been by default rather than by design. This has contributed to services becoming too expensive for some and those services that are available are poorly regulated and open to abuse. Welfare provision in China is prey to the same tendencies as elsewhere, not only in Asia but also in Western societies, with a mixed system for delivery slowly emerging. In this process, it is important that the state define its role clearly and create space for alternative service providers to function effectively. However, certain legacies of China’s pre-reform past and CCP ideology will influence the exact nature of the system. Most noticeable is the significant difference in service provision between urban and rural China. The privileged position of the industrial working class reinforces the differences in accessibility to services between those working in the formal sector of the economy and those working in the informal sector. Even though the cradle-to-grave benefits of the Maoist years have been eroded, social insurance provision, including medical care, is still immeasurably better for those in the formal sector than it is for other workers. The reification of the party and state, with weak countervailing pressures from society, has also entrenched a system under which party cadres and government officials enjoy the best service treatment. The CCP and its governing apparatus have been insulated from external pressures since taking power in 1949 and although the economic reforms are changing this, the system remains largely unaccountable and lacking in transparency. The party has appropriated the right to speak on behalf of those it claims to represent, initially the workers and peasants, but now the whole of society. This attitude of knowing best has been translated into a set of welfare benefits and guarantees for the bureaucracy that is unavailable to other groups in society.

states, markets, and civil society / 207

Moving forward, if the leadership wants to attain its goal of building a “harmonious society,” a more coherent philosophy for the provision of social welfare and public goods provision in general will need to be developed. Policy will need to continue pulling together the fragmented welfare systems inherited from the pre-reform system into a more consolidated system in which the government plays the role of the last resort and regulator of for-profit and not-for-profit agencies. The government should take on those risks that neither the market nor society and the family can cover. The future design of welfare programs and the expansion of benefits should be based on a notion of citizenship that begins to break the barriers that have been created between the rural and the urban areas, the informal and the formal sectors of the economy, and between those working in the bureaucracy and those outside of it. In her excellent study of migrants, Solinger (1999, p. 1) notes that “citizenship does not come easily to those outside the political community.” This does not mean that the government should rush to extend all existing benefits to all citizens at the same level over the short term. This is not only impractical but also might not make good economic sense and might not even enhance social equity. The introduction of the MLSS for both urban and rural dwellers and the extension of pensions to some rural dwellers are good examples of where Chinese policy is moving in this direction. The policy initiatives to incorporate migrants into welfare systems and services in their chosen place of work also offer the potential of developing a welfare system based on citizenship. A guiding principle for the provision of social welfare should be equity, so that citizens are provided with the opportunity to access the kinds of resources they need to enhance their capabilities to function in a modern economy and society. The inequities of access are especially apparent in healthcare. This is one area where the government has not devised an effective framework that guarantees citizens the kinds of services they deserve. Health and protection from disease (see Chapters Four and Five) can be seen as a special good as they are a prerequisite for an individual to be able to function effectively. As Anand (2004, p. 16) has written, health has both an intrinsic and an instrumental value. This makes it more important than, for example, income inequality that has only an instrumental value. Amartya Sen notes (2004, p. 21; 1985), in any discussion of social equity and justice, illness and health must figure as a major concern. Thus, although the government’s efforts to introduce social insurance schemes and to expand the MLSS are laudable, the absence of an enabling health policy undermines its attempts to build a just society. Building an effective health system should be the major component in constructing a “harmonious society.”

This page intentionally left blank

No t e s

The Sample Survey of Citizens’ Attitudes about Chinese Government Performance 1. Unless otherwise stated the details for the 2004 and 2005 surveys are the same. All surveys were conducted during fall. 2. There were 4128 respondents in 2005, and 3859 in 2004.

One Public Goods Regimes and Social Welfare Provision in China 1. Oikos is the Greek for a household or family.

Two Welfare Provision, 1949–1979 1. See “Temporary Regulations of Private Undertakings,” December 29, 1950, China News Analysis, 15 (December 4, 1953). 2. See Regulations on the “Work of the Rural People’s Communes” (Revised Draft), September 1962, in Union Research Services (ed.), Documents of the Chinese Communist Party Central Committee, Sept. 1956–Apr. 1969, vol. 1 (Hong Kong: Union Research Institute, 1971, 702ff).

Three The Provision of Public Goods during the Reform Period 1. The HDI composite index is based on three indicators: longevity as measured by life expectancy at birth; educational attainment calculated from adult literacy (two-thirds weighting) and the secondary and tertiary enrollment ratio (one-third weight); and standard of living calculated by GDP per capita. 2. The HPI reflects the distribution of progress and measures the backlog of deprivations that still exist. It focuses on three dimensions: longevity measured by the probability of not living to 40 years of age from birth; adult literacy rates that attest to knowledge; and general economic provisioning as measured by the percentage of citizens who are not using improved water sources, those

210 / notes

3.

4. 5. 6. 7.

8.

without sustainable access to improved water sources, and children who are underweight for their age. This figure does not include the 2.5 million military personnel who are classified neither as urban nor as rural. It compares to the urbanization rate in the United Kingdom during the 1880s, that in the United States in 1911, and in Japan in 1950. Wuhan displays a similar trend, with 11.23 percent of its population over the age of 60 at the end of 1998 (Wuhan: 1999). There were originally 11 designated models, to which an additional 15 were added in 2000. Other models described by Derleth and Koldyk (2004) are the Shanghai, Wuhan, and Qingdao models. These provide variance on the degree of autonomy and control of government and party over the communities. Their conclusions were based on research in a county without a financial surplus situated in Shaanxi province. The province itself has consistently run fiscal deficits exceeding 20 percent of expenditures, making local governments increasingly self-reliant in terms of meeting expenditure responsibilities. Discussion with provincial vice governor, May 1997.

Four Challenges in the Health System 1. The tenor of this report and the fact that it was reported so extensively both in the domestic and international press disturbed China’s senior leaders. It led to the DRC having to heavily edit the 2005 Human Development Report that it was preparing for the UNDP on development with equity to tone down language that was too critical of reforms. 2. Basic healthcare was defined as including: disease prevention, immunization, health education, hygiene supervision, care for infants and mothers, mental health, public health emergencies, blood collection, food safety, prevention of occupational diseases, safe drinking water as well as supplying basic drugs and proper procedures to diagnose and test chronic diseases. 3. From 1991 to 2000, the rural share of total healthcare expenditures fell from 34.9 percent to 22.5 percent (Li and Zhu: 2004, p. 10) 4. Shaoguang Wang (2003) calculates that between 1989 and 2001, the average outpatient charge increased 965 percent, while the average hospitalization charge rose 998 percent. 5. It should be noted that there are very few formal private medical hospitals. In 2005, only 10.8 percent of the 8,703 main hospitals were private (China Daily, March 22, 2007). 6. See China Health Development Forum, Beijing, November 2000.

Five Dealing with Public Health Crises and Pandemics 1. The rate is second in the world after India (5 million patients of whom 80 percent were in the countryside). From 2000 to 2005, China claims to have provided

notes / 211

2.

3.

4. 5. 6. 7. 8.

9. 10.

11.

treatment for 2.05 million lung TB patients and that at the end of 2005 it had a 79 percent detection rate and a 91 percent recovery rate (Xinhuanet, March 10, 2006). There are about 1.5 million new infections each year in China and some public health officials feel that the attention paid to avian flu and even HIV/AIDS detracts from the more immediate problems caused by TB. This refers to: free ARV drugs for AIDS patients in rural areas or those in financial difficulty in urban areas; free voluntary counseling and testing; free drugs for HIV-infected pregnant women to prevent mother-to-child transmission and testing of new born babies; free schooling for AIDS orphans; and care and economic support to the households of those living with HIV/AIDS. By September 2006, 65.7 percent of reported AIDS cases were receiving free drugs and 70 percent of HIV-infected pregnant women had agreed to receive ARVs, and 80 percent of the babies born to infected mothers received treatment (Xinhua News Agency, November 24, 2006). Most notable was Premier Wen’s visit during the Chinese New Year 2005 to HIV-positive people in Shangcai county and to the “Sunshine Home” in Xincai county, Henan province, places at the center of the government-led blood donor drives that led to significant levels of infection. It is interesting to note that one of the families that Wen visited has since suffered prejudice from neighbors who previously did not know about the infection. The data in the new survey are drawn from 329 national sentinel surveillance sites rather than the previous 194 sites and the estimates are taken from prefecture data rather than the provincial data in the 2003 survey. Yunnan, Xinjiang, Guangxi, Guangdong, Guizhou, Sichuan, and Hunan. The five provinces of Henan, Hubei, Anhui, Hebei, and Shanxi account for 80.4 percent of this population. This is a much lower prevalence rate than was thought earlier. Han (2004) mentions a 500-fold rise from 1995 to 2000, rising from a prevalence rate of 0.02 percent to a peak of 12.1 percent in 1999 before dropping to 10 percent in 2000. By way of comparison, from 2001 China allocated 40 million yuan per year for dealing with TB prevention and control; this was increased to nearly 300 million yuan in 2004 (Xinhuanet, March 10, 2006). Given the prevalence of TB, some have been critical of the greater attention that HIV/AIDS has received, which may be related to the greater international priority (see below). There are 13 Chinese government agencies, 11 multilateral agencies, 5 Chinese NGOS, and 3 international NGOs, out of a total of 54 participants. Admittedly, the number of NGOs is small, but for China this was a significant step. See “Wen Jiabao’s Speech at a Seminar on Implementation of ‘Emergency Regulations on Sudden Outbreaks of Public Health Incidents’,” Xinhua News Agency, May 13, 2003, translated in FBIS-CHI-2003–0515; and “Wen Jiabao Stresses Enforcement of SARS-Related Laws,” Xinhua News Agency, May 15, 2003. Relevant institutes under its jurisdiction include those for Communicable Disease Control and Prevention, Viral Disease Control and Prevention, Health Education, and the offices of Epidemiology, and Disease Control and Emergency Response. At the end of 2006, official statistics noted 48,951 HIV cases in Yunnan , the highest in China (Xinhua News Agency, July 10, 2007). However, it appears

212 / notes that Xinjiang may now exceed this number, with about 60,000 infections (New York Times, November 12, 2006).

Six

Building a New Social Insurance System

1. Henan, Hubei, Hunan, Shandong, Shanghai, Shanxi, Tianjin, and Xinjiang. 2. The following description is based on discussions with Stephen McGurk, officials at MOCA and MOLSS, and ADB: 2002.

Seven Poverty Relief and Social Assistance 1. The “three nos” are elderly who are alone and unable to work, have no income source, and have no person legally responsible for their support. The “five guarantees” are for food, clothing, housing, medical care, and burial expenses. 2. The following discussion is based on Saich (2004). 3. This is based on discussions with Arthur Holcombe, president of the fund, and other program staff.

Eight Providing Social Welfare: States, Markets, and Civil Society 1. This section draws on Saich: 2007a and 2007b. 2. The sample comprises a purposive stratified survey of approximately 4,000 respondents each year (2003–2005) ages 16 through 60, selected from three administrative levels: city, township, and village. By design, the sample does not include migrants or the majority of ethnic minorities living in China. The questionnaire was conducted in the two municipalities of Beijing and Shanghai and the five cities of Guangzhou, Wuhan, Chengdu, Shenyang, and Xi’an, with Nantong (Jiangsu) added in 2005. For details of the survey see p. xiv.

Bi bl io g r a ph y

Amsden, Alice H. 1989. Asia’s Next Giant: South Korea and Late Industrialization. New York: Oxford University Press. Anand, Sudhir. 2004. “Health Equity: The Concern for Equity in Health,” in Public Health, Ethics, and Equity, ed. Sudhir Anand, Fabienne Peter, and Amartya Sen. New York: Oxford University Press. Anson, Ofra, and Shifang Sun. 2005. Health Care in Rural China: Lessons from HeBei Province. Aldershot: Ashgate. Ascoli, Ugo, and Costanzo Ranci, eds. 2002. Dilemmas of the Welfare Mix: The New Structure of Welfare in an Era of Privatization. New York: Kluwer Academic. Asian Development Bank (ADB). 2002. People’s Republic of China. Toward Establishing a Rural Health Protection Scheme. Manila: ADB. Aspalter, Christian. 2001. Conservative Welfare State Systems in East Asia. Westport, CT: Praeger. ———. 2002. Democratization and Welfare State Development in Taiwan. Burlington, VT: Ashgate. ———. 2006. “The East Asian Welfare Model,” International Journal of Social Welfare, 15, no. 4: 290–301. Banister, Judith. 1984. “Population Policy and Trends in China, 1978–83,” China Quarterly, no. 100 (December): 717–41. Bernstein, Thomas P., and Xiaobo Lü. 2000. “Taxation without Representation: Peasants, the Central and the Local States in Reform China,” China Quarterly, no. 163 (September): 742–63. Bian, Morris L. 2005. The Making of the State Enterprise System in Modern China: The Dynamics of Institutional Change. Cambridge, MA: Harvard University Press. Bian, Yanjie. 1994. Work and Inequality in Urban China. Albany: State University of New York Press. Bird, Richard, and Christine Wallich. 1993. Fiscal Decentralization and Intergovernmental Relations in Transition Economies: Toward a Systemic Framework of Analysis. Washington, DC: Policy Research Dept., World Bank. Bloom, David, Patricia Craig, and Pia Malaney. 2001. The Quality of Life in Rural Asia. Oxford: Oxford University Press. Blumenthal, David, and William Hsiao. 2005. “Privatization and its Discontents: The Evolving Chinese Health Care System,” New England Journal of Medicine, 353 (September): 1165–70. Bray, David. 2005. Social Space and Governance in Urban China: The Danwei System from Origins to Reform. Stanford: Stanford University Press.

214 / bibliography Cai Fang, ed. 2003. Zhongguo renkou yu laodong wenti baogao: Zhuanggui zhong de chengshi pinkun wenti [Report on China’s Population and Labor: Urban Poverty in Transitional China]. Beijing: Shehui kexue wenxian chubanshe. Cai, Yongshun. 2000. “Between State and Peasant: Local Cadres and Statistical Reporting in Rural China,” China Quarterly, no. 163 (September): 783–805. Cailliez, Charlotte. 1998. “The Collapse of the Rural Health System,” China Perspectives, no. 18 (July–August): 36–43. Center for Human and Economic Development (CHED). 2005. China’s Healthcare Reforms and Challenges. Beijing, unpublished manuscript. Chan, Kam Wing. 2007. “Misconceptions and Complexities in the Study of China’s Cities: Definitions, Statistics, and Implications,” Eurasian Geography and Economics, 48, no. 4: 383–412. Chen, Jie. 2004. Popular Political Support in Urban China. Stanford: Stanford University Press. Chen, Shaohua, Ren Mu, and Martin Ravallion. 2006. Are There Lasting Impacts of Aid to Poor Areas? Washington, DC: World Bank Policy Research Paper, no. 4084 (December). Cheng, K.M. 1995. “Education: Decentralization and the Market,” in Social Change and Social Policy in Contemporary China, ed. L. Wong and S. MacPherson. Aldershot: Avebury. Cheng, Kevin. 2003. “Economic Implications of China’s Demographics in the 21st Century,” IMF Working Paper, WP/03/29. Cheng, Tiejun, and Mark Selden. 1997. “The Construction of Spatial Hierarchies: China’s Hukou and Danwei Systems,” in New Perspectives on State Socialism of China, ed. Timothy Cheek and Tony Saich. Armonk, NY: M.E. Sharpe. Chen, Xiang-sheng, Xiang-dong Gong, Guo-jun Liang, and Guo-cheng Zhang. 2000. “Epidemiologic Trends of Sexually Transmitted Disease in China,” Sexually Transmitted Diseases, 27, no. 3: 138–42. China Development Research Foundation (CDRF). 2007. Zhongguo fazhan baogao 2007: Zai fazhan zhong xiaochu pinkun [China Development Report 2007: Eliminating Poverty through Development]. Beijing: Zhongguo fazhan chubanshe. China Labor Statistical Yearbook. 1998. Beijing: Zhongguo tongji chubanshe. ———. 2004. Beijing: Zhongguo tongji chubanshe. ———. 2006. Beijing: Zhongguo tongji chubanshe. China Statistical Yearbook. 2006. Beijing: Zhongguo tongji chubanshe. Choi, S.Y.P., Y.W. Cheung, and K. Chen. 2006. “Gender and HIV Risk Behavior among Intravenous Drug Users in Sichuan Province, China,” Social Science & Medicine, 62, no. 7: 1672–84. Chow, Nelson. 1987. “Western and Chinese Ideas of Social Work,” International Social Work, 30: 31–34. Cook, Sarah. 2003. “After the Iron Rice Bowl: Extending the Safety Net in China,” in Social Protection in China, ed. Sarah Cook, Naila Kabeer, and Gary Suwannarat. Delhi: HarAnand Publications. Cook, Sarah, Naila Kabeer, and Gary Suwannarat. 2003. Social Protection in Asia. Delhi: HarAnand Publications. Crane, Keith, Roger Cliff, Evan S. Medeiros, James C. Mulvenon, and William H. Overholt. 2005. Modernizing China’s Military: Opportunities and Constraints. Santa Monica, CA: Rand.

bibliography / 215 Croissant, A. 2003. Aufholende Nachzügler? Sozialpolitik und Demokratisierung in Sud Korea, Thailand und auf den Philippinen,” Welt-Trends, 39: 89–110. ———. 2004. “Changing Welfare Regimes in East and Southeast Asia: Crisis, Changes and Challenges,” Social Policy and Administration, 38, no. 5: 504–24. Dai, Xiaohua. 1998. “East Asian Model: A Few Problems, but it Works,” Beijing Review, 41, no. 12: 7–9. Derleth, James, and Daniel R. Koldyk. 2004. “The Shequ Experiment: Grassroots Political Reform in Urban China,” Journal of Contemporary China, 13, no. 41 (November): 747–77. Development Research Center Project Group. 2005. “Dui Zhongguo yiliao weisheng tizhi gaige de pingjia yu jianyi (zhaiyao yu zhong dian)” [Evaluation and Suggestions on Reform of China’s Medical Health System (Abstract and Main Points)]. Beijing. Deyo, Frederick C. 1992. “The Political Economy of Social Policy Formation: East Asia’s Newly Industrialized Countries,” in States and Development in the Asia Pacific Region, ed. R.P. Appelbaum and J. Henderson. London: Sage. Dixon, John. 1981. The Chinese Welfare System, 1949–1979. New York: Praeger. Dollar, David. 2007. Poverty, Inequality and Social Disparities during China’s Economic Reform. Washington, DC: World Bank Policy Research Working Paper, no. 4253, June. Dorfman, Mark C., and Yvonne Sin. 2000. “China: Social Security Reform,” in Technical Analysis of Strategic Options. Human Development Network. Washington, DC: World Bank, December 13. Du, Ying. 2000. “Guanyu nongcun yiliao weisheng tizhi gaige de jidian kanfa” [Some Ideas on the Reform of the Rural Medical Health System], in Economics Research Department, Ministry of Public Health and the Institute of Development Studies, Zhongguo nongcun weisheng gaige yu fazhan guoji yantaohui [International Conference on China’s Rural Health Reform and Development]. Beijing. Eberstadt, Nick. 1979. Poverty in China. Bloomington: Indiana University International Development Institute. ———. 1988. The Poverty of Communism. New Brunswick, NJ: Transaction Books. Eckstein, Alexander. 1962. The National Income of Communist China. New York: Free Press of Glencoe. Edin, Maria. 2000. Market Forces and Communist Power: Local Political Institutions and Economic Development in China. Uppsala: Uppsala University. Edwards, Richard L., Philip W. Cooke, and P. Nelson Reid. 1996. “Social Work Management in an Era of Diminishing Federal Responsibility,” Social Work, 41, no. 5: 468–79. Eggleston, Karen, Magnus Lindelow, Li Ling, Meng Qingyue, and Adam Wagstaff. 2006. Health Service Delivery in China: A Literature Review. Washington, DC: World Bank Policy Research Working Paper no. 3978. Esping-Andersen, Gosta. 1990. The Three Worlds of Welfare Capitalism. Princeton, NJ: Princeton University Press. Fan, C. Cindy. 2007. “Migration, Hukou, and the Chinese City,” in China Urbanizes, ed. Shahid Yusuf and Tony Saich. Washington, DC: World Bank. Fan, Shenggen, Linxiu Zhang, and Xiaobo Zhang. 2002. “Growth, Inequality, and Poverty in Rural China.” International Food Policy Research Institute Research Report no. 125.

216 / bibliography Feigenbaum, Harvey, Jeffrey Henig, and Chris Hamnett. 1998. Shrinking the State: The Political Underpinnings of Privatization. New York: Cambridge University Press. Ferge, Zsuzsa. 1997. “The Changed Welfare Paradigm: The Individualisation of the Social,” Social Policy and Administration, 31, no. 1: 20–44. First Five-Year Plan for Development of the National Economy of the People’s Republic of China in 1953–57. 1956. Beijing: Foreign Languages Press. Foldvary, Fred. 1994. Public Goods and Private Communities: The Market Provision of Social Services. Brookfield, VT: E. Elgar. Frazier, Mark W. 2006. “Pensions, Public Opinion and the Graying of China,” Asia Policy, no. 1 (January): 43–68. Fung, Archon, and Erik Olin Wright. 2003. Deepening Democracy: Institutional Innovations in Empowered Participatory Governance. New York: Verso. Gao, Qin. 2006. “The Social Benefit System in Urban China: Reforms and Trends from 1988 to 2002,” Journal of East Asian Studies, 6, no. 1: 31–67. Gautam, K. 1993. Health Challenges for China. Paper presented at the International Seminar on Health Reforms under Socialist Market Economy, Beijing, June. George, Vic, and Nick Manning. 1980. Socialism, Social Welfare, and the Soviet Union. Boston: Routledge & Kegan Paul. Giddens, Anthony. 2000. The Third Way and its Critics. Malden, MA: Blackwell. Gilbert, Neil. 2002. Transformation of the Welfare State: The Silent Surrender of Public Responsibility. New York: Oxford University Press. Giles, John, Albert Park, and Juwei Zhang. 2005. “What is China’s True Unemployment?” Chinese Economic Review, 16, no. 2: 149–70. Gill, Bates, Yanzhong Huang, and Xiaoqing Lu. 2007. Demography of HIV/AIDS in China. Washington, DC: Center for Strategic and International Studies. Gill, Bates, J. Stephen Morrison, and Xiaoqing Lu. 2007. China’s Civil Society Organizations. What Future in the Health Sector? Washington, DC: Center for Strategic and International Studies. Gillion, Colin, John Turner, Clive Bailey, and Denis Latulippe. 2000. Social Security Pensions: Development and Reform. Geneva: International Labour Office. Goldsmith, Stephen, and William D. Eggers. 2004. Governing by Network: The New Shape of the Public Sector. Washington, DC: Brookings Institution Press. Goodman, Roger, Gordon White, and Huck-ju Kwon, eds. 1998. The East Asian Welfare Model: Welfare Orientalism and the State. New York: Routledge. Goodman, Roger, and I. Peng. 1996. “The East Asian Welfare States,” in Welfare States in Transition, ed. G. Esping-Anderson. London: Sage. Gough, I. 2000. Welfare Regimes in East Asia and Europe. Paper delivered at the Annual World Bank Conference on Development Economics, Paris. ———. 2002. “Globalization and National Welfare Regimes: The East Asian Case,” in Social Security in the Global Village, ed. R. Sigg and C. Behrendt. London: Transaction. Gould, Arthur. 1993. Capitalist Welfare Systems: A Comparison of Japan, Britain, and Sweden. New York: Longman. Grout, Paul A., and Margaret Stevens. 2003. “The Assessment: Financing and Managing Public Services,” Oxford Review of Economic Policy, 19, no. 2: 215–34.

bibliography / 217 Gu, Edward X. 2001. “Dismantling the Chinese Mini-welfare State? Marketization and the Politics of Institutional Transformation, 1979–1999,” Communist and Post-Communist Studies, 34, no. 1 (March): 91–111. Gu, Edward X., and Jianjun Zhang. 2006. “Health Care Regime Change in Urban China: Unmanaged Marketization and Reluctant Privatization,” Pacific Affairs, 79, no. 1 (Spring): 49–71. Gu, Xin. 2005. Health Care Marketization in Urban China. Singapore: East Asian Institute, National University of Singapore. ———. 2006. China Reweaves a New Social Safety Net. Singapore: East Asian Institute, National University of Singapore. Guo, Xiaolin. 1999. “The Role of Local Government in Creating Property Rights: A Comparison of Two Townships in Northwest Yunnan,” in Property Rights and Economic Reform in China, ed. Jean Oi and Andrew Walder. Stanford: Stanford University Press. ———. 2001. “Land Expropriation and Rural Conflicts in China,” China Quarterly, no. 166 (June): 422–39. Gustaffson, Bjorn, and Quheng Deng. 2007. “Social Assistance Receipt and Its Importance for Combating Poverty in Urban China.” Bonn: Institute for the Study of Labor, Discussion Paper Series, no. 2758, April. Haggard, S., and R. Kaufmann. 2004. “Revising Social Contracts: Social Spending in Latin America, East Asia, and the Former Socialist Countries.” Paper presented at International Symposium on Comparative Social Policy. National Chi Nan University, Puli, Taiwan, May 31. Han, J. 2003. “Public Finance Crisis in Chinese Counties and Towns: Performance, Causes, Impact and Measures.” International Conference Report on Rural Public Finance, Beijing. ———. 2004. “China: From Urban-Rural Partition to Coordinated Urban-Rural Development,” in China: Towards a Balanced Development. Beijing: China Development Forum Background Paper. Han, Mengjie. 2004. “Presentation to HIV/AIDS Public Policy Training Workshop.” Sponsored by the Kennedy School, Harvard University. Beijing, January. Hayek, Friedrich A. 1944. The Road to Serfdom. Chicago: University of Chicago Press. Henderson, Gail E., and Myron S. Cohen. 1984. The Chinese Hospital: A Socialist Work Unit. New Haven: Yale University Press. Hermann-Pillath, Carsten. 2000. Social Capital and Networks: A Proposal for Conceptual Integration. Working Paper no. 63. University of Witten/ Herdecke. Hewitt, P.S. 2004. “The Geopolitics of Global Aging,” Harvard Generations Policy Journal, 1 (Winter): 103–13. Holliday, Ian. 2000. “Productivist Welfare Capitalism: Social Policy in East Asia,” Political Studies, 48, no. 4: 706–23. Holliday, Ian, and Paul Wilding, eds. 2003. Welfare Capitalism in East Asia: Social Policy in the Tiger Economies. New York: Palgrave Macmillan. Holzmann, Robert, and Steen Jorgensen. 1999. “Social Protection as Social Risk Management: Conceptual Underpinnings for the Social Protection Sector Strategy Paper.” Washington, DC: World Bank, Social Protection Discussion Paper Series, no. 9904.

218 / bibliography Horn, Joshua S. 1969. Away with all Pests: An English Surgeon in People’s China. London: Hamlyn. Hu, Angang. 2004. “From Unevenness towards Coordination for China’s Regional Development (1978–2005),” in China: Towards a Balanced Development. Beijing: China Development Forum Background Paper. Hu, Biliang. 2007. Informal Institutions and Rural Development in China. New York: Routledge. Hu, Shanlian, and Minghe Jiang. 1998. “The People’s Republic of China,” in Social Sector Issues in Transitional Economies of Asia, ed. Douglas H. Brooks and Myo Thant. Oxford: Oxford University Press. Huang, Yanzhong. 2004. “Bringing the Local State Back In: The Political Economy of Public Health in Rural China,” Journal of Contemporary China, 13, no. 39: 367–90. Hurst, William, and Kevin J. O’Brien. 2002. “China’s Contentious Pensioners,” China Quarterly, no. 170 (June): 345–59. Hussein, Athar. 2000. “Social Welfare in China in the Context of Three Transitions.” Center for Research on Economic Development and Policy Reform. Working Paper no. 66. Stanford University, August, at http://scid. stanford.edu/publications/abstracts.html#66. ———. 2002. “Demographic Transition in China and its Implications,” World Development, 30, no. 1: 1823–34. Jackson, R., and N. Howe. 2004. “The Greying of the Middle Kingdom: The Demographics and Economics of Retirement Policy in China.” Washington, DC: Center for Strategic and International Studies Working Paper 04/2004. Jiang, Zemin. 1997. “Full Text of Jiang Zemin’s Report at 15th Party Congress,” at http://www.china.org.cn/english/features/49007.htm ———. 2002. “Report to the Sixteenth Party Congress,” at http://www.16congress. org.cn/english/features/49007.htm Jolly, Susie, and Ying Wang. 2003. Key Issues on Gender and HIV/AIDS in China. Beijing: China-UK HIV/AIDS, DFID. Jones, Catherine. 1990. “Hong Kong, Singapore, South Korea, and Taiwan: Oikonomic Welfare States,” Government and Opposition, 25, no. 3: 446–62. ———, ed. 1993. New Perspectives on the Welfare State in Europe. New York: Routledge. Jowitt, Kenneth. 1992. New World Disorder: The Leninist Extinction. Berkeley: University of California Press. Jütting, Johannes. 1999. “Strengthening Social Security Systems in Rural Areas of Development Countries.” Bonn: Centre for Developing Research. Kaiser Network. 2004. “Global Challenges: More than 80% of MSM Know Little about HIV/AIDS, Survey Shows,” Daily HIV/AIDS Report (December 16), at http://www.kaisernetwork.org/daily_reports/rep_index.cfm?hint⫽1&DR_ ID⫽27274. ———. 2007. “Global Challenges: China’s Blood Supply Not Being Monitored Properly for HIV, Needs International Assistance, Report Says,” Daily HIV/ AIDS Report (September 7), at http://www.kaisernetwork.org/daily_reports/ rep_index.cfm?hint⫽1&DR_ID⫽47352.

bibliography / 219 Kamarck, Elaine. 2004. Government Innovation around the World. Cambridge, MA: Research Programs, John F. Kennedy School of Government, Harvard University. Kaufman, Joan, and Kathrine Meyers. 2006. “AIDS Surveillance in China: Data Gaps and Research for AIDS Policy,” in AIDs and Social Policy in China, ed. Joan Kaufman, Arthur Kleinman, and Tony Saich. Cambridge, MA: Asia Center, Harvard University, pp. 47–71. Kennedy, John James. 2007. “From the Tax-for-Fee Reform to the Abolition of Agricultural Taxes: The Impact on Township Governments in North-West China,” China Quarterly, no. 189 (March): 43–59. Kennedy School of Government. 1995. “Mayor Stephen Goldsmith Organizing Competition in Indianapolis A, B, and Sequel.” Case Study no. 1269, 1270.0 and 1270.1. ———. 1999. “Pension Reform in China: Weighing the Alternative.” Case Study in Public Policy and Management. no. 1547.0. Kettl, Donald F. 2005. The Global Public Management Revolution, 2nd ed. Washington, DC: Brookings Institution Press. Khan, Azizur Rahman, and Carl Riskin. 1998. “Income and Inequality in China: Composition, Distribution and Growth of Household Income, 1988–1995,” The China Quarterly, no. 154 (June): 228–53. ———. 2001. Inequality and Poverty in China in the Age of Globalization. New York: Oxford University Press. Kish, Leslie. 1949. “A Procedure for Objective Respondent Selection within the Household,” Journal of the American Statistical Association, 44, no. 247: 380–87. ———.1965. Survey Sampling. New York: John Wiley. Knight, John, and Jinjun Xue. 2005. “How High is Urban Unemployment in China?” Journal of Chinese Economic and Business Studies, 4, no. 2: 91–107. Kornai, Janos. 1997. “Reform of the Welfare Sector in the Post-Communist Countries: A Normative Approach,” in Transforming Post-Communist Political Economies, ed. Joan M. Nelson, Charles Tilly, and Lee Walker. Washington, DC: National Academy Press. Ku, Yeun-wen, and Catherine Jones Finer. 2007. “Developments in East Asian Welfare Studies,” Social Policy and Administration, 41, no. 2 (April): 115–31. Leung, Joe C.B. 2006. “The Emergence of Social Assistance in China,” International Journal of Social Welfare, 15, no. 3: 188–98. Li, Lianjiang. 2004. “Political Trust in Rural China,” Modern China, 30, no. 2 (April): 228–58. Li, Lianjiang, and Kevin J. O’Brien. 1996. “Villagers and Popular Resistance in Contemporary China,” Modern China, 22, no. 1 (January): 28–61. Li, Minqi, and Andong Zhu. 2004. “China’s Public Services Privatization and Poverty Reduction: Health Care and Education Reform (Privatization) in China and the Impact on Poverty.” Beijing: United Nations Development Programme Policy Brief. Li, Xueju. 1994. Zhongguo chengxiang jiceng zhengquan jianshe gongzuo yanjiu [A Study of Urban and Rural Grassroots Regimes]. Beijing: Zhongguo shehui chubanshe. Li, Yongping. 1992. “Sex Ratio of Infants and Relations with Some Socioeconomic Variables: The Results of China’s 1990 Census and Implications.” Paper

220 / bibliography presented at the International Seminar on China’s 1990s Population Census. Beijing, October. Lieberthal, Kenneth G. 1980. Revolution and Tradition in Tientsin, 1949–1952. Stanford: Stanford University Press. Lin, Fude, and Zhai Zhenwu. 1996. Zouxiang ershiyi shiji de Zhongguo renkou, huanjing yu fazhan [The Trend of China’s Population, Environment, and Development in the Twenty-First Century]. Beijing: Gaodeng jiaoyu chubanshe. Lin, W.I. 1991. “Structural Determinants of Welfare Effort in Postwar Taiwan,” International Social Work, 34: 171–90. Lindbeck, Assar. 2006. An Essay on Economic Reforms and Social Change in China. Washington, DC: World Bank Policy Research Working Paper no. 4057. Liu, W., X. Zhen, and M. Wen. 2000. “Gender Analysis on Household Health Expenditure in Rural Areas of China.” Paper presented to the Conference on Rural Health Reform and Development. Beijing, November. Liu, Yuanli, and Joan Kaufman. 2006. “Controlling HIV/AIDS in China: Health System Challenges,” in AIDs and Social Policy in China, ed. Joan Kaufman, Arthur Kleinman, and Tony Saich. Cambridge, MA: Asia Center, Harvard University. Lu, Feng. 1989 “Danwei: Yizhong tequ de shehui zuzhi xingshi” [Danwei: A Typical Form of Social Organization], Zhongguo shehui kexue [China Social Sciences], no. 1. ———. 1993. “Zhongguo danwei tizhi de qiyuan he xingcheng” [Origin and Formation of the Danwei System in China], Zhongguo shehui kexue jikan [China Social Sciences Quarterly] (Autumn). Lu, Mai. 1999. “China’s Urgent Challenge: To Provide Public Goods in a Market Environment.” Unpublished paper. Lü, Xiaobo, and Elizabeth J. Perry. 1997. Danwei: The Changing Chinese Workplace in Historical and Comparative Perspective. Armonk, NY: M.E. Sharpe. Luo, Gan. 1998. “Explanation on Plan for Institutional Restructuring of the State Council: Delivered at the First Session of the Ninth National People’s Congress on March 6, 1998,” Dagong bao (March 7): B1–B2; translated in FBIS-China98–068, March 9. Ma, Jun, and Fan Zhai. 2001. “Financing China’s Pension Reform.” Paper presented at the Conference on Financial Sector Reform in China. Cambridge, MA: Harvard University, September. Mahathir, Mohamad, and Shintaro Ishihara. 1995. The Voice of Asia: Two Leaders Discuss the Coming Century. New York: Kodansha International. Marshall, T.H. 1950. Citizenship and Social Class, and Other Essays. Cambridge, England: University Press. ———. 1964. Class, Citizenship and Social Development. Garden City, NY: Doubleday. Meng, Qingguo, and Angang Hu. 2000. “Eliminating Health Poverty Ought to be a Top Strategic Priority in Rural Health Reform and Development,” Ministry of Health, Proceedings of the Conference on Rural Health Reform and Development, Beijing, November, pp. 72–83. Ministry of Civil Affairs (MOCA). 2000. Zhongguo minzheng tongji nianjian 2000 [China Civil Affairs Statistical Yearbook 2000]. Beijing: Zhongguo tongji chubanshe.

bibliography / 221 ———. 2003. Zhongguo minzheng tongji nianjian 2003 [China Civil Affairs Statistical Yearbook 2003]. Beijing: Zhongguo tongji chubanshe. ———. 2006. Zhongguo minzheng tongji nianjian 2006 [China Civil Affairs Statistical Yearbook 2006]. Beijing: Zhongguo tongji chubanshe. Ministry of Health and Joint United Nations Programme on HIV/AIDS and World Health Organization. 2006. 2005 Update on the HIV/AIDS Epidemic and Response in China. Beijing: Ministry of Health and UN Theme Group on HIV/AIDS in China. 2003. A Joint Assessment of HIV/AIDS Prevention, Treatment and Care in China. Beijing: December 1. Moore, Mark H. 2001. “The Market versus the Forum,” in Governance amid Bigger, Better Markets, ed. John D. Donahue and Joseph S. Nye. Washington, DC: Brookings Institution Press, pp. 306–39. Moss, David A. 2002. When All Else Fails: Government as the Ultimate Risk Manager. Cambridge, MA: Harvard University Press. Mountfield, Edward, and Christine P.W. Wong. 2005. “Public Expenditure on the Frontline: Towards Effective Management of Public Expenditure by SubNational Governments in East Asia,” in East Asia Decentralizes, ed. Roland White and Paul Smoke. Washington, DC: World Bank. National Bureau of Statistics (NSB). 2005 and 2007. Zhongguo tongji nianjian 2005 (2007) [China Statistical Yearbook 2005]. Beijing: Zhongguo tongji chubanshe. National Research Centre. 1998. “Evaluation of the Economic and Social Benefits of Project Hope: A Summary of Basic Conclusions.” Beijing. Unpublished Memo. Naughton, Barry. 1997. “Danwei: The Economic Foundations of a Unique Institution,” in Danwei: The Changing Chinese Workplace in Historical and Comparative Perspective, ed. Xiaobo Lü and Elizabeth J. Perry. Armonk, NY: M.E. Sharpe. ———. 2007. The Chinese Economy: Transitions and Growth. Cambridge, MA: MIT Press. Nelson, Joan M. 1997. “Social Costs, Social-Sector Reforms, and Politics in Post Communist Transformations,” in Transforming Communist Political Economies, ed. Joan M. Nelson, Charles Tilly, and Lee Walker. Washington, DC: National Academy Press. Nielsen, I., C. Nyland, R. Smyth, and C.J. Zhu. 2005. “Marketization and Perceptions of Social Protection in China’s Cities,” World Development, 33, no. 11: 1759–81. North, Douglass C. 1990. Institutions, Institutional Change and Economic Performance. New York: Cambridge University Press. Nyberg, Albert, and Scott Rozelle. 1999. Accelerating China’s Rural Transformation. Washington, DC: World Bank. O’Brien, Kevin. 1994. “Implementing Political Reform in China’s Villages,” Australian Journal of Chinese Affairs, no. 32 (July): 33–59. ———. 1996. “Rightful Resistance,” World Politics, 49, no. 1: 31–55. OECD. 2006. Challenges for China’s Public Spending: Toward Greater Effectiveness and Equity. Paris: OECD. Oi, Jean C. 1992. “Fiscal Reform and the Economic Foundations of Local State Corporatism in China,” World Politics, 45, no. 1 (October): 99–126.

222 / bibliography Overbye, Einar. 2005. “Extending Social Security in Developing Countries: A Review of Three Main Strategies,” International Journal of Social Work, 14: 305–14. Pan, Suiming. 2001. “AIDS in China: How Much Possibility is There in Sexual Transmitting?” Unpublished paper, First China Conference on AIDS/STDs. Beijing, November. Parish, William L., and Suiming Pan. 2006. “Sexual Partners in China: Risk Patterns for Infection by HIV and Possible Interventions,” in AIDs and Social Policy in China, ed. Joan Kaufman, Arthur Kleinman, and Tony Saich. Cambridge, MA: Asia Center, Harvard University, pp. 191–213. Park, Albert, Scott Rozelle, Christine Wong, and Changqing Ren. 1996. “Distributional Consequences of Reforming Local Public Finance in China,” China Quarterly, no. 147 (September): 751–78. Park, Albert, and Changqing Ren. 2001. “Microfinance with Chinese Characteristics,” World Development, 29, no. 1 (January): 39–62. Park, Albert, Sangui Wang, and Guobao Wu. 1999. “Regional Poverty Targeting in China.” Working Paper, Department of Economics, University of Michigan. ———. 2002. “Regional Poverty Targeting in China,” Journal of Public Economics, 86, no. 1 (October): 123–53. Peng, Zhen. 1940. “Jin-Cha-Ji bianqu muqian shizheng gangling” [The Current Administrative Reform in the Jin-Cha-Ji Border Region], in Saich: 1996: 934–36. Perry, Elizabeth J. 1993. Shanghai on Strike: The Politics of Chinese Labor. Stanford: Stanford University Press. Polanyi, Karl. 1944. The Great Transformation. New York: Rinehart & Co. Prabhakar, Rajiv. 2006. Rethinking Public Services. New York: Palgrave Macmillan. Prybyla, Jan S. 1981. The Chinese Economy: Problems and Policies, 2nd ed. Rev. Columbia: University of South Carolina Press. Public Health in China. 2005. World Bank China Rural Health Study. Washington, DC. Qiu, Yulin. 2003. “Yiliao fuwu yu ruoshi qunti zhichi” [Medical Services and Support for Vulnerable Groups], in Zhongguo shehui fazhan yanjiu baogao 2002 [2002 Research Report on China’s Social Development], ed. Zheng Hangsheng. Beijing: Renmin daxue chubanshe. Ramesh, M. 2002. “Social Policies as Collective Goods: Malaysia and Singapore,” in Collective Goods, Collective Acts in East and Southeast Asia, ed. Sally Sargeson. New York: Routledge, pp. 147–61. Ravallion, Martin, and Shaohua Chen. 2007. “China’s (Uneven) Progress Against Poverty,” Journal of Development Economics, 82, no. 1: 1–42. Rong, Jingben, Cui Zhiyuan, Wang Shuanzheng, Gao Xinjun, He Zengke, and Yang Xuedong. 1998. Cong yali xing tizhi xiang minzhu hezuo tizhi de zhuan bian: Xianxiang liangji zhengzhi tizhi gaige [Transformation from the Pressurized System to the Democratic System of Cooperation: Reform of the Political System at the County and Township Levels]. Beijing: Zhongyang bianyi chubanshe. Rozelle, Scott, Albert Park, V. Benziger, and C. Ren. 1998. “Targeted Poverty Investments and Economic Growth in China,” World Development, 26, no. 12: 2137–51.

bibliography / 223 Ru, Xiaomei. 2006. “Youth and HIV in China,” in HIV/AIDS and Social Policies in China, ed. Joan Kaufman, Arthur Kleinman, and Tony Saich. Cambridge, MA: Asia Center, Harvard University. Rural Survey Team of the National Bureau of Statistics. 2004. Zhongguo nongcun pinkun jiance baogao 2004 [Poverty Monitoring Report of Rural China]. Beijing: Zhongguo tongji chubanshe. Saich, Tony. 1996. The Rise to Power of the Chinese Communist Party: Documents and Analysis. Armonk, NY: M.E. Sharpe. ———. 2000. “Negotiating the State: The Development of Social Organizations in China,” China Quarterly, no. 161 (March): 124–41. ———. 2002a. “The Blind Man and the Elephant: Analysing the Local State in China,” in On the Roots of Growth and Crisis: Capitalism, State and Society in East Asia, ed. Luigi Tomba. Milan: Annale Feltrinelli, no. 36. ———. 2002b. “Social Welfare Provision and Economic Transition: Pluralism and Integration of Service Delivery.” Geneva: International Labour Office. ———. 2004. Governance and Politics of China, 2nd ed. New York: Palgrave Macmillan. ———. 2006. “SARS: China’s Chernobyl or Much Ado about Nothing,” in SARS in China: Economic, Political and Social Consequences, ed. Arthur Kleinman and James L. Watson. Stanford: Stanford University Press, pp. 71–104. ———. 2007a. “China in 2006: Focus on Social Development,” Asian Survey, 47, no. 1 (January/February): 32–43. ———. 2007b. “Citizens’ Perceptions of Governance in Rural and Urban China,” Journal of Chinese Political Science, 12, no. 1 (April): 1–28. Saich, Tony, and Joan Kaufman. 2005. “Financial Reform, Poverty, and the Impact on Reproductive Health Provision: Evidence from Three Rural Townships,” in Financial Sector Reform in China, ed. Yasheng Huang, Tony Saich, and Edward Steinfeld. Cambridge, MA: Asia Center, Harvard University. Salditt, Felix, Peter Whiteford, and Willem Adema. 2007. “Pension Reform in China: Progress and Reform.” Paris: OECD Social, Employment and Migration Working Papers. Samuelson, Paul A. 1954. “The Pure Theory of Public Expenditure,” Review of Economics and Statistics, 36, no. 4 (November): 387–89. Sargeson, Sally, ed. 2002. Collective Goods, Collective Futures in Asia. New York: Routledge. Sen, Amartya. 1985. Commodities and Capabilities. Amsterdam: North-Holland. ———. 1999. Development as Freedom. New York: Knopf. ———. 2004. “Why Health Equity?” in Public Health, Ethics, and Equity, ed. Sudhir Anand, Fabienne Peter, and Amartya Sen. New York: Oxford University Press. Shi, Tianjian. 2001. “Cultural Values and Political Trust: A Comparison of the People’s Republic of China and Taiwan,” Comparative Politics, 33, no. 4: 401–19. Shih, Victor, Mingxing Liu, and Qi Zhang. n.d. “ ‘Eating Budget’: Determining Fiscal Transfers under Predatory Fiscal Federalism.” Unpublished paper. Shue, Vivienne. 1988. The Reach of the State: Sketches of the Chinese Body Politic. Stanford: Stanford University Press. Sidel, Victor W., and Ruth Sidel. 1973. Serve the People: Observations on Medicine in the People’s Republic of China. Boston: Beacon Press.

224 / bibliography Sin, Yvonne. 2005. China: Pension Liabilities and Reform Options for Old Age Insurance. Washington, DC: World Bank, Working Paper no. 2005–1, May. Skinner, G.W. 1964/65. “Marketing and Social Structure in Rural China,” Journal of Asian Studies, 24 (1–3). Solinger, Dorothy J. 1999. Contesting Citizenship in Urban China: Peasant Migrants, the State, and the Logic of the Market. Berkeley: University of California Press. State Council. 1998. “Tentative Regulation on Public Service Units Registration,” Decree 252, October 25. ———. 2001. “The Development-Oriented Poverty Reduction Programme for Rural China,” Xinhua News Agency, October 15. ———. 2006. Action Plan for Reducing and Preventing the Spread of AIDS, 2006–2010, at http://www.unchina.org/unaids/UNAIDS%20IN%20 ENGLISH/documents/Action%20plan%20-English%20_draft_%20.pdf State Council AIDS Working Committee Office and UN Theme Group on HIV/ AIDS in China. 2004. A Joint Assessment of HIV/AIDS Prevention, Treatment and Care in China. Beijing. ———. 2007. A Joint Assessment of HIV/AIDS Prevention, Treatment and Care in China. Beijing. State Council Information Office. 2006. White Paper on the Development of China’s Undertakings for the Aged, at http://www.china.org.cn/english/ China/191990.htm State Statistical Bureau. 1983. Statistical Yearbook of China 1983. Hong Kong: Economic and Information Agency. Stiglitz, Joseph E. 2002. Globalization and its Discontents. New York: W.W. Norton. Su, Ming, and Quanhou Zhao. 2006. The Fiscal Framework and Urban Infrastructure Finance in China. Washington, DC: World Bank Policy Research Working Paper, no. 4051. Szleak, Nicole. 2005. “The Role of Knowledge-Processing in Health PolicyMaking: The Global Fund to Fight AIDS, Tuberculosis and Malaria in China: A Case Study.” Qualifying paper, Kennedy School of Government, Harvard University. Tang, Jun, and Shifei Zhang. 2006. “Tiaozheng zhongde chengxiang zuidi shenghuo baozhang zhidu” [Adjustment of the Urban and Rural Minimum Living Support System], in 2006 nian: Zhongguo shehui xingshi fenxi yu yuce [2006: Analysis and Forecast on China’s Social Development], ed. Ru Xin, Lu Xueyi and Li Peilin. Beijing: Shehui kexue wenxian chubanshe, pp. 165–75. Tang, K.L. 2000. Social Development in East Asia. Basingstoke: Palgrave. Thompson, Drew. 2005. “China Confronts HIV/AIDS,” Population Reference Bureau (June). Torrey, Barbara Boyle. 1992. “Sharing Increasing Costs on Declining Income: The Visible Dilemma of the Invisible Aged,” in The Oldest Old, ed. Richard Suzman, David P. Willis, and Kenneth G. Manton. New York: Oxford University Press, pp. 381–93. Tsai, Lily Lee. 2001. “Strategies of Rule or Ruin? Governance and Public Goods Provision in Rural China.” Paper presented at the Annual Meeting of the Association for Asian Studies, March.

bibliography / 225 ———. 2002. “Cadres, Temple and Lineage Institutions, and Governance in Rural China,” China Journal, no. 48 (July): 612–45. ———. 2007. Accountability without Democracy: Solidary Groups and Public Goods Provision in Rural China. New York: Cambridge University Press. Unger, Jonathan. 2002. “Poverty, Credit and Microfinance in Rural China,” Development Bulletin (Contemporary China Centre, ANU), no. 57 (February): 23–26. United Nations Development Programme (UNDP). 1993. Human Development Report. New York: Oxford. ———. 1998. China: Human Development Report: Human Development and Poverty Alleviation 1997. Beijing. ———. 1999. Human Development Report. New York: Oxford. ———. 2000. Human Development Report. New York: Oxford. ———. 2001. Human Development Report. New York: Oxford. ———. 2002. An Integrated Approach to Reducing Poverty in China (in collaboration with the International Labour Organization). Beijing. ———. 2004. Millennium Development Goals: China’s Progress 2003. Beijing: Office of UN Resident Coordinator in China. ———. 2005. China Human Development Report 2005: Development with Equity. Beijing: China Translation & Publishing Corporation. ———. 2006. Human Development Report 2006: Beyond Scarcity: Power, Poverty, and the Global Water Crisis. At http://hdr.undp.org/en/reports/global/hdr2006. ———.2007. Human Development Report 2007/2008: Fighting Climate Change: Human Solidarity in a Divided World. At http://hdr/undp/org/en/reports/ global/hdr2007–2008 Van Dalen, Hendrik P. 2006. “When Health Care Insurance Does Not Make a Difference? The Case of Health Care ‘Made in China.’ ” Tinbergen Institute Discussion Paper, no. 06–091/1. Vogel, Ezra F. 1980. Japan as Number One: Lessons for America. New York: Harper. Wade, Robert. 1990. Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization. Princeton, NJ: Princeton University Press. Walder, Andrew G. 1986. Communist Neo-traditionalism: Work and Authority in Chinese Industry. Berkeley: University of California Press. Wang, Longde. 2006. “To Develop an Equitable and Efficient Health System: China’s Strategies,” Presentation to the China Medical Board. Beijing, October. Wang, Sangui. 2004. “Poverty Targeting in the People’s Republic of China.” ADB Institute Discussion Paper, no. 4. Wang, Shao guang. 2003. “Zhongguo gonggong weisheng de weiji he zhuangji” [The Crisis and Opportunity in China’s Public Health]. Hong Kong: The Chinese University of Hong Kong. Wang, Xiaotao. 1999. “The Politics of Village Elections: The Case of North Village,” Issues and Studies, 35, no. 6: 167–85. Wang, Yan, Dianqing Xu, Zhi Wang, and Fan Zhai. 2001. Implicit Pension Debt, Transition Cost, Options, and Impact of China’s Pension Reform. Washington, DC: World Bank Policy Research Working Paper no. 2555.

226 / bibliography Watts, Jonathan. 2006. “AIDS in China: New Legislation, Old Doubts,” The Lancet, 367, no. 9513 (March 11). Wexler, Robyn, Xu Ying, and Nick Young. 2006. NGO Advocacy in China. Beijing: China Development Brief, Special Report. White, Gordon. 1998. “China’s Social Security Reforms,” in East Asian Welfare Model, ed. Roger Goodman, Gordon White, and Huck-ju Kwon. New York: Routledge, pp. 175–98. White Paper on Labor and Social Security. 2002. At http://www.china.org.cn/ english/2002/Apr/31774.htm White Paper on Rural China’s Poverty Reduction. 2001. At http://english.peopledaily. com.cn/features/PRpaper/pr.html White, Roland, and Paul Smoke, eds. 2005. East Asia Decentralizes. Washington, DC: World Bank. Whiting, Susan H. 2001. Power and Wealth in Rural China: The Political Economy of Institutional Change. New York: Cambridge University Press. Whyte, Martin King, and William L. Parish. 1984. Urban Life in Contemporary China. Chicago: University of Chicago Press. Wolf, Charles, Jr., K.C. Yeh, Benjamin Zycher, Nicholas Eberstadt, and Sungho Lee. 2003. Fault Lines in China’s Economic Terrain. Santa Monica: Rand. Wong, Chack-Kie, and Peter Nan-Shong Lee. 2001. “Economic Reform and Social Welfare: The Chinese Perspective Portrayed through a Social Survey in Shanghai,” Journal of Contemporary China, 10, no. 28: 517–32. Wong, Christine P.W., ed. 1997. Financing Local Government in the People’s Republic of China. Hong Kong: Oxford University Press. ———. 2000. “Central-Local Relations Revisited: The 1994 Tax-Sharing Reform and Public Expenditure Management in China,” China Perspectives, no. 31 (September–October). ———. 2007. “Fiscal Management for a Harmonious Society: Assessing the Central Government’s Capacity.” British Inter-University China Center Working Paper, September. Wong, Christine P.W., and Richard Bird. 2008. “China’s Fiscal System: A Work in Progress,” in China’s Great Transformation: Origins, Mechanism, and Consequences of the Post-Reform Economic Boom, ed. Loren Brandt and Thomas Rawski. New York: Cambridge University Press, pp. 429–66. Wong, Linda, and Norman Flynn, eds. 2001. The Market in Chinese Social Policy. New York: Palgrave. Wong, Victor C.W., and Sammy W.S. Chiu. 1998. “Health-care Reforms in the People’s Republic of China: Strategies and Social Implications,” Journal of Management in Medicine, 12(4–5): 270–86. World Bank. 1990. World Development Report 1990: Poverty. New York: Oxford University Press. ———. 1994. Averting the Old-Age Crisis: Policies to Protect the Old and Promote Growth. New York: Oxford University Press. ———. 1997a. Financing Health Care. Washington, DC: World Bank. ———. 1997b. Old Age Security: Pension Reform in China. Washington, DC: World Bank. ———. 1997c. Sharing Rising Incomes. Washington, DC: World Bank.

bibliography / 227 ———. 1998. Reform of State-Owned Enterprises and Privatisation: A Dialogue, ed. Rehman Sobhan, M.M. Akash, and T. Akram. Dhaka: Pathak Shamabad. ———. 2000. China: Overcoming Rural Poverty. Washington, DC: World Bank. ———. 2002. China: National Development and Sub-national Finance: A Review of Provincial Expenditures. Washington, DC: Poverty Reduction and Economic Management Unit, East Asia and Pacific Region, World Bank. ———. 2003. 03 World Development Indicators. Washington, DC: World Bank. ———. 2004. Making Services Work for Poor People: World Development Report 2004. Washington, DC: World Bank. ———. 2005a. China: Deepening Public Service Unit Reform to Improve Service Delivery. Washington, DC: World Bank, July. ———. 2005b. “Reforming China’s Health Care System,” Quarterly Update (World Bank Office Beijing) (November), at www.worldbank.org/china. ———. 2006. World Development Report 2006: Equity and Development. New York: Oxford University Press. World Health Organization (WHO). 2001. World Health Report: Report of the Director-General 2000. Geneva: WHO. Wu, Fei. 1997. “Zhongguo nongcun shehui de zongjiao: Huabei mouxiang nongcun tianzhujiao huodong kaocha” [Religion in Rural Society: Investigation into Catholic Activities in a Certain County in North China], Zhanlue yu guanli [Strategy and Management], no. 4: 54–62. Wu, Zunyou, Sheena G Sullivan, Yu Wang, Mary Jane Rotheram-Borus, and Roger Detels. 2007. “Evolution of China’s Response to HIV/AIDS,” Lancet, 369 (9562): 679–90. Wuhan Municipal People’s Congress Education, Science and Culture Committee. 1999. “Wuhan shi renkou laonianhua wenti ji duice” [Problems and Countermeasures of the Aging of the Wuhan Municipality’s Population]. Wuhan: n.p. Xue, Xingling, Changxi Li, Chena Lei, and Jianhua Yu. 2006. “Chengxiang shehui baozhang zhidu de chaju yu tongchou duice” [The Difference in the Urban and Rural Social Insurance System and a Comprehensive Countermeasure], Nongye guancha [Rural Investigations], no. 3: 38–44. Yang, Dali. 1994. “Reform and the Restructuring of Central-Local Relations,” in China Deconstructs, ed. David S.G. Goodman and Gerald Segal. London: Routledge, pp. 59–98. ———. 2004. Remaking the Chinese Leviathan: Market Transition and the Politics of Governance. Stanford: Stanford University Press. Yang, Yiyong, Ying Zhang, and Yan Gu. 2006. “Goujian chengxiang tongchou de zuidi shenghuo baozhang tixi” [Construct a Comprehensive Minimum Living Standard Support System], Chinese Journal of Population Science, no. 6, 36–42. Yao, Wenyuan. 1975. “On the Social Basis of the Lin Piao Anti-Party Clique,” Peking Review, 18, no. 10 (March 7). Yu, Deping. 2002. Chengxiang shehui: Cong geli zouxiang kaifang: Zhongguo huiji zhidu yu hujifa yanjiu [City and Countryside Societies: From Segregation to Opening: Research on China’s Household Registration System and Laws]. Ji’nan: Shandong renmin chubanshe.

228 / bibliography Yu, Jianrong. 2006. “Quanguo dibao jiu shi gei nongmin guomin daiyu” [National Minimum Living Support Gives Farmers the Status of Citizens], at http:// paper.cs.com.cn/html/2006–07/25/content_1835590.htm. Yu, Wei, and Minghui Ren. 1997. “The Important Issue of Enterprise Reform: Health Care Insurance System,” in The Reformability of China’s State Sector, ed. G.J. Wen and D. Xu. Singapore: World Scientific, pp. 433–63. Yuan, Yue. 2003. Zou jin feng yue: Dixia xing gongzuozhe [Underground Commercial Sex Workers]. Beijing: Zhongguo mangwen chubanshe. Yunnan Center for Disease Prevention and Control, 2004. HIV/AIDS Epidemic and Response of Yunnan. Yunnan. Zeng, Yi, Tu Ping, Gu Baochang, Xu Yi, Li Bohua, and Li Yongping. 1993. “Causes and Implications of the Recent Increase in the Reported Sex Ratio in China,” Population and Development Review, 19, no. 2 (June): 283–302. Zeng, Yi, James W. Vaupel, Xiao Zhenyu, Zhang Chunyuan, and Liu Yuzhi. 2002. “Sociodemographic and Health Profiles of Oldest Old in China,” Population and Development Review, 28, no. 2: 251–73. Zhang, Chunqiao. 1975. “On Exercising All-Round Dictatorship over the Bourgeoisie,” Peking Review, 18, no. 14 (April 4). Zhang, Fujie, Michael Hsu, Lan Yu, Yi Wen, and Jennifer Pan. 2006. “Initiation of the National Free Antiretroviral Therapy Program in Rural China,” in AIDs and Social Policy in China, ed. Joan Kaufman, Arthur Kleinman, and Tony Saich. Cambridge, MA: Asia Center, Harvard University, pp. 96–124. Zhang, Kaimin. 2001. “Shanghai renkou laonianhua de fazhan” [The Development of Shanghai’s Aging Population], unpublished paper. Zhang, Konglai. 2002. “HIV/AIDS in China,” in Health Care, East and West: Moving into the 21st Century, ed. Zong Shujie and Nelson Y.S. Kiang. Beijing: China Science and Technology Press. Zhang, Le-Yin. 1999. “Chinese Central-Provincial Fiscal Relationships, Budgetary Decline and the Impact of the 1994 Fiscal Reform: An Evaluation,” China Quarterly, no. 157 (March): 115–41. Zhang, Xin, and Richard Baum. 2004. “Civil Society and the Anatomy of a Rural NGO,” China Journal, no. 52 (July): 97–107. Zhang, Yi. 2003. “Jingzhong: Woguo nuer chusheng xingbie zai chixu shangsheng” [Warning: An Imbalance of the Sex Ratio in China], in Shehui lanpishu 2003 nian: Zhongguo shehui xingshi fenxi yuce [Blue Book of Chinese Society 2003: Circumstances and Analysis of China’s Society], ed. Ru Xin, Lu Xueyi, and Li Peilin. Beijing: Shehui kexue wenxian chubanshe. Zhao, Litao. 2006. Time to Care for Migrant Workers: New Social Initiatives in the Hu-Wen Era. Singapore: East Asian Institute, National University of Singapore, February. Zhao, Shukai. 2004. “Lishixing tiaozhan: Zhongguo nongcun de chongtu yu zhili” [An Historic Challenge: Conflict and Governance in China’s Villages], in 2004 nian: Zhongguo shehui xingshi fenxi yu yuce [2004: Analysis and Forecast of the Situation in Chinese Society], ed. Ru Xin, Lu Xueyi, and Li Peilin. Beijing: Shehui kexue wenxian chubanshe. Zhao, Yaohui, and Jianguo Xu. 1999. “Alternative Transition Paths in the Chinese Urban Pension System.” Beijing: China Center for Economic Research Working Paper.

bibliography / 229 ———. 2002. “Chinese Urban Pension System: Reforms and Problems,” Cato Journal, 21, no. 3: 395–414. Zhong, Kaibin. 2007. “Crisis Management in China,” China Security (Winter): 90–109. “Zhongguo nongcun weisheng gaige yu fazhan Beijing ziliao” [Background Materials on Rural Health Reform and Development]. 2000. In Economic Research Department of the Ministry of Health and the Institute of Development Studies, Zhongguo nongcun weisheng gaige yu fazhan guoji yantaohui [International Conference on China’s Rural Health Reform and Development]. Zhou, Y., and L.J.C. Ma. 2003. “China’s Urbanization Levels: Reconstructing a Baseline from the Fifth Population Census,” The China Quarterly, no. 173 (March): 176–96. Zhu, Chuzhu, Shuzhou Li, Changrong Qiu, Ping Hu, and Anrong Jin. 1997. The Dual Effects of the Family Planning Program on Chinese Women. Xi’an: Xi’an Jiaotong University Press. Zhu, Ling, and Zhongyi Zhang, 1996. Public Works and Poverty Alleviation in Rural China. Commack, NY: Nova Science Publishers.

This page intentionally left blank

I n de x

aging 52–53, 134–35 in Shanghai 52–53 Agricultural Bank of China 167, 169–70 AIDS Working Committee (State Council) 122 All-China Federation of Trade Unions 196 All-China Women’s Federation 196 Anhui 40, 79, 116, 117, 126 antiretroviral therapy (ARV) 116 Asian Development Bank 98, 155, 156–67 Asian Financial Crisis 127–28 avian flu 20, 105, 119–24 Bangladesh 41, 74, 170–71 barefoot doctors ix, 29, 38, 72 basic health insurance scheme 86–88 Beijing xvi, 46–47, 53, 63–65, 79–80, 87, 90, 94, 107–10, 117, 120–21, 126, 139, 149, 151–53, 163–64, 177–78, 181, 197, 200 Bismarkian (welfare) system 18 Boshi Management Company 144 Budget Law 59–60 Canada 195 Cash Poor (Malaysia) 170 census (2000) 50 Center for Disease Control and Prevention (CDC) 111, 117–18, 121

Chen Liangyu 140, 153 Chengdu xvi, 178 China Banking Regulatory Commission 145 China CARES 115–16 China Country Coordinating Mechanism 117 China Foundation for Aiding Poor Areas 172 China Foundation for Poverty Alleviation 172–73 China Health Development Forum 99 China HIV/AIDS Association 118 China Insurance Regulatory Commission 145 Chinese Communist Party (CCP) 5, 14, 17, 21, 13–25, 27–28, 30–34, 36–37, 43, 52–53, 69, 71, 103, 106, 109, 156–57, 166, 173, 181, 185, 193, 195, 204–6 Central Committee Sixth Plenum (October 2006) 45, 140 Seventeenth Party Congress (2007) 50–51 Chongqing 79, 80, 94, 114, 139, 149, 163, 178, 197 Clinton Foundation 126 Colombia 99 comfortable society (xiaokang shehui) 83–84, 107 commercial sex workers (CSW) 111–12, 114, 116–18, 198

232 / index commune 9, 37–40, 59, 72, 74, 94 Communist Youth League 172 community residents’ committees (shequ jumin weiyuanhui) 56, 58 Confucian welfare states 17, 24 Confucianism 17 corporatist welfare 12, 17 corruption, social security finds 139–40, 145–46 rural pensions 155 Cuba 50 Cultural Revolution ix, 26, 29, 32, 38, 40, 73, 98, 134–35 Deng Xiaoping 26 dependency ratios 53–54 Develop the West 165 Development Research Center of the State Council 100 developmental welfare systems 10 drug pricing policy 82–83, 87–88, 93 East Asian welfare model 1–2, 8–12, 195 Eight-Seven Poverty Reduction Plan 161 Emergency Response Bureau (State Council) 121–22 enabling state 4, 12–14 enterprise annuities 145–46 Esping-Andersen on welfare 11–12, 17–18, 158 family-planning 99, 134, 154, 156 Fei Xiaotong 50 financing public goods 58–67 five guarantees households 55, 153, 159–60, 175, 179 food-for-work program 167, 169 Ford Foundation 170 former blood and plasma donors (FBD) 111–13, 118 Four Frees and One Care 110 France 53

Friends of Nature 197 Fujian xvi, 18, 197, 200 Gang of Four 26 Gansu 49, 64, 162, 181 gender imbalance 54–55, 111 Gilbert on welfare 12–14 Global Fund to Fight AIDS, Tuberculosis, and Malaria 116–18, 194–95 government organized NGOs (GONGO) 118–19 Grameen Bank 170–71 Great Leap Forward ix, 18, 23–24, 32, 37, 39–41 Guangdong 47, 50, 53–54, 64, 67, 87, 107, 109, 114, 120–21, 124, 126, 141, 181, 197 Guangzhou xvi, 40, 123, 177–78 Guangxi 113–14, 116, 173 Guiyang 177–78 Guizhou 31, 34, 40, 47, 49, 53, 63–64, 66–68, 79–80, 94, 116, 126, 139, 149, 162, 178, 181 harmonious society (hexie shehui) 199, 206–7 health system performance 71–74 spending 74–79 urban China pre-reform 73 Hebei xvi, 31, 63, 82, 93, 116, 200 Heifer Project International 172 Heilongjiang 47, 94, 138–40, 149, 180–81 Henan 47, 64, 79, 94, 111–13, 116–17, 125–26, 140, 198 HIV/AIDS 20, 46, 83, 93, 103, 105–6, 109–26, 198, 200 Horizon Research Group xv–xvi, 187 household registration system (hukou) 30–32, 35, 44, 50, 89, 163, 204

index / 233 Hu Jintao ix, 2, 14, 45, 46, 55, 69, 75, 89, 106, 109, 119, 140, 161, 204 Hubei xvi, 62 Huhehot 177–78 human capabilities approach 3–4 Human Development Index (HDI) 4, 45–48 Human Poverty Index (HPI) 46–47 Hunan xvi, 64, 79, 114, 116, 140, 155, 178 Hungary 50 immunization programs 103–4 implicit pension debt 143–44 India 15, 24, 40–41, 46–48, 74–77, 103, 164, 169 Indonesia 15–16, 24, 41, 46–48, 59, 74–76, 98, 164 infectious diseases 103–5 ING 156 Inner Mongolia 163, 173 intravenous drug users (IDU) 111–14, 116–18, 198 Japan 4, 9, 11–12, 17, 33–34, 40, 42, 134, 203 Jiang Qing 26 Jiang Zemin ix, 119, 140, 165, 202 Jiangsu xvi, 47, 69, 74, 79, 82, 86, 154, 189, 194, 197 Jiangxi 54, 63, 79, 80, 86, 94, 116, 167, 172, 200 Jiangxi Soviet (1931–34) 31, 33 Jiangyin (Jiangsu) 82 Jichun (Hebei) 82 Jilin 138–39, 149, 180–81 Jin-Cha-Ji Border Region 31, 33 Jiujiang (Jiangxi) 86 KISH Method xv Lai Ruoyu 26 Lanzhou 163

Leading Group for Poverty Alleviation (State Council) 166–67, 170, 173 Li Lanqing 87 Li Lisan 25–26 Liaoning xvi, 94, 135–36, 138–39, 147, 149, 163, 180–81 Malaria 104 Malaysia 15–16, 41, 46–48, 75–76, 98, 170 Mao Zedong 24–25, 29, 34, 36, 98, 139 maternal child healthcare (MCH) 99 Médécins sans Frontières 94 medical costs 81–82 medical insurance coverage 78–81 men who have sex with men (MSM) 111–12, 114 Mexico 40 microfinance 170–72 Migrants 49–50, 88–90, 146–47 migration 49–50 Millennium Development Goals 46 Minimum Living Standard Scheme (MLSS) 160–64, 178–83, 205, 207 Ministry of Agriculture 120–21 Ministry of Civil Affairs 56, 58, 154–57, 175, 177, 179 Ministry of Finance 65, 90, 100, 141, 167 Ministry of Health 78, 81, 83, 94, 96, 99–100, 104, 106–08, 113, 126 Ministry of Labor and Social Security 137, 141, 144–46, 151, 155 Nanchang 177–78 Nantong xvi National Council for Social Security 138, 144–46

234 / index National Development and Reform Commission 167 National Health Conference (First, 1950) 91–92 National Health Conference (1996) 95–96 National Health Service (UK) 85 National Health Survey 81–82 National Social Security Fund 138 the Netherlands 195 networked governance 193 new public management 4–5 New Zealand 4 Ningxia 53, 94, 169, 180 nongovernmental Organization (NGO) 3, 7, 17, 117–19, 124–27, 157, 171–72, 183, 185, 194–96, 198–99, 201–2, 206, 211 our survey xv–xvi, 81, 98, 149–50, 151, 187–92 Panzhihua iron and steel producer 136–37 pension reform rural 153–57 urban 129–53, 192–93 People Living with HIV/AIDS 125–26 People’s Relief Administration 36 Philippines 15–16, 40, 48, 76, 98 poverty alleviation policy 46–47, 161–74 poverty alleviation funding 167–69 private goods definition 6 productivist welfare capitalism 10–11 Project Hope 172 public goods definition 6 Public Service Unit (PSU) 5, 194, 199 Public Welfare Donations Law 202

Qingdao 58, 68, 141 Qinghai 53, 64, 94, 162, 180, 201 residents’ committee 36, 56–58, 182 Rolling Stones 108 Rotating Savings and Credit Association (ROSCA) 200, 205–6 rural collective welfare provision 37–38 rural collectivization 28–29 rural cooperative medical scheme 38–39, 72–73, 92, 95–96 new rural cooperative medical scheme 96–97 rural healthcare coverage 90–97 rural minimum living standard scheme 175–76, 179–80 rural-urban differences 18, 48–49, 79–80 Rural Women Knowing All 197 Russia 50 Sanchuan Development Association 201 SARS 20, 93, 104–10, 116–17, 119–21, 123–24 Save the Children (UK) 126 Scandinavia 195 schistosomiasis 73–74, 103 self-reliance (and Maoism) 18, 30–31, 59 Sen, Amartya 3, 207 sexually transmitted diseases (STD) 103, 105, 115 Shaanxi xvi, 47, 66, 116, 126, 162, 173 Shandong 68–69, 94, 104, 116, 141, 169, 197 Shanghai xvi, 29, 35, 46–47, 49–50, 52–53, 55, 141–42, 145–47, 149, 151–53, 163–64, 176–79, 181, 197–99

index / 235 Shanghai Young Men’s Christian Association 199–200 Shanxi 31, 116, 126, 200 Shenyang xvi, 56, 177–78 Shenyang Internal Combustion Factory 85 Shenzhen 86, 114, 123, 141–42, 145, 177–78 Sichuan xvi 34, 40, 47, 95, 113–17, 124, 126, 136, 139–40, 149, 155–56, 173, 181, 197–98 Sino-Soviet split 139 social investment state 4 South Africa 111 South Korea 4, 12, 15–17, 41, 46–48, 76, 134, 203–4 Sri Lanka 41 State Administration of Taxation 145 State Economic and Trade Commission 141 State Family Planning Commission 156 State Industrial and Commercial Bureau 141 state-led development 9–10 state-owned enterprise (SOE) 35–36, 39, 44, 51, 54, 56, 59, 69, 73, 77, 81, 85–86, 88, 129–30, 134–35, 137, 139, 145, 150, 152, 163–64, 175, 192 subsidized loan program 167–68, 169–70 Sunrise Club 117 Sweden 53 Taiwan 4, 12, 41, 204 tax reforms (1993–94) 61, 64–65 Thailand 15–16, 46–47, 48, 59, 75–76, 99, 110–11, 125 Third Front 139 three no households 159, 175, 177, 181 Tianjin 46–47, 79, 94, 163–64, 177

Tibet 46–47, 64, 67, 94, 155, 162, 171, 180 Tibet Poverty Alleviation Fund 171–72 township and village enterprise (TVE) 49, 65, 69, 154, 157 trade unions and welfare 25–26, 28–29 Tuberculosis 103–5 two-Jiang model (Urban Medical Insurance Reform) 86 UNAIDS 125 unemployment insurance 131 United Kingdom 4, 85 United Nations Development Programme (UNDP) 4, 45–46, 53, 167, 170 United States of America (welfare system) 11–12 urban healthcare coverage 85–90 urban minimum living standard scheme 175–82 urban poverty 163–64 urban welfare provision (pre-reform) 32–39 urbanization 50–51 Vietnam

15, 47–48, 59, 75–76, 99

welfare state 12–14, 185 Wen Jiabao ix, 2, 14, 45, 46, 55, 61, 68, 69, 75, 89, 106, 110, 119, 122, 138, 161, 168, 180, 188, 202, 204 Workplace (urban) 3, 9, 17, 19, 24–25, 28–29, 30, 32–37, 43, 45, 54–56, 58, 69, 127–30, 135, 138, 149–52, 159, 161, 185, 203 World Bank 47, 51, 83, 127–29, 136, 160, 164, 168, 173–74, 186

236 / index World Health Organization (WHO) 23, 39, 108, 110, 113, 120 Wu Yi 108–9, 122 Wuhan xvi, 163, 178

Yinchuan 177 Yining 177–78 Youth Development Foundation 172 Yunnan 47, 64, 80, 94, 111, 113–16, 123–26

Xiamen 177–78 Xi’an xvi Xiangdong Village 200 Xining 178 Xinjiang 67, 94, 111, 113, 116–17, 126, 180 Xinjiang Agricultural University 117

Zhang Dejiang 107 Zhejiang xvi, 79, 80, 94, 140, 181, 197, 200 Zhejiang University 99 Zhenjiang (Jiangsu) 86 Zhu Rongji ix, 130, 156, 165, 202