Pension Policy and Governmentality in China: Manufacturing Public Compliance

Wang, Yan Book — Published Version Pension Policy and Governmentality in China: Manufacturing Public Compliance Provided in Cooperation with: LSE Press, London School of Economics and Political Science (LSE) Suggested Citation: Wang, Yan (2022) : Pension Policy and Governmentality in China: Manufacturing Public Compliance, ISBN 978-1-909890-89-3, LSE Press, London, https://doi.org/10.31389/lsepress.ppc This Version is available at: https://hdl.handle.net/10419/273991 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Rapid economic growth is often disruptive – so how has the Chinese state been able to maintain compliance from its people while pursuing an exceptionally fast social and economic transformation? This crucial question is explored through detailed analysis of the trajectories, policy rationale, and eff ects of China’s pension reforms, demonstrating how statecraft shapes the ways in which citizens ascribe credit and responsibility for pensions protection to themselves, the state, and other actors. In Pension Policy and Governmentality in China, Yan Wang shows that China’s governmentality for manufacturing compliance is hybrid, organic, and dynamic. Targeted allocation of benefi ts, policy experimentation, propaganda, and knowledge construction are all used to shape public expectations and to justify state rule. An original contribution to the study of legitimation in modern states, this book demonstrates how, when active counter-conduct is confi ned, individuals may choose cognitive rebellion and falsify their public compliance. Dr Yan Wang is a Research Fellow in the School of Public Policy at the London School of Economics and Political Science (LSE), and will join the Department of Sociology at Lancaster University as a Lecturer (Assistant Professor) in Digital Sociology in 2023. Pension Policy and Governmentality in China Manufacturing Public Compliance Yan Wang “Drawing on theorists such as Weber and Gramsci whose work emphasises the importance of public consent and utilising the Foucauldian concept of governmentality, the book argues that coercion alone is insuffi cient to build consent and that more sophisticated (and dynamic) tactics are required … this is a most timely look at the importance of social policy in terms of Chinese governance.” Dr Mel Cousins, School of Social Work and Social Policy, Trinity College Dublin “Yan Wang has provided the go-to book on China’s pension system, tracing its changes over time and its eff ects on legitimacy in the present day. It’s an important addition to the comparative conversation on the welfare state.” Dr Rory Truex, Department of Politics, Princeton University Manufacturing Public Compliance PENSION POLICY and GOVERNMENTALITY in CHINA Yan Wang PENSION POLICY and GOVERNMENTALITY in CHINA Yan Wang Pension Policy and Governmentality in China Manufacturing Public Compliance Yan Wang Published by LSE Press 10 Portugal Street London WC2A 2HD press.lse.ac.uk Text © Yan Wang 2022 First published 2022 This book is based on the author’s doctoral thesis: http://etheses.lse.ac.uk/4111/. The research was approved by the LSE Research Ethics Committee (ref. 1264) Cover design by Diana Jarvis Cover photo: ‘Three older women on bench from behind’ by hanohikirf / Alamy Print and digital versions typeset by Siliconchips Services Ltd. ISBN (Paperback): 978-1-909890-88-6 ISBN (PDF): 978-1-909890-89-3 ISBN (EPUB): 978-1-909890-90-9 ISBN (Mobi): 978-1-909890-91-6 DOI: https://doi.org/10.31389/lsepress.ppc This work is licensed under the Creative Commons Attribution 4.0 International License (unless stated otherwise within the content of the work). To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ or send a letter to Creative Commons, 444 Castro Street, Suite 900, Mountain View, California, 94041, USA. This license allows for copying any part of the work for personal and commercial use, providing author attribution is clearly stated. The full text of this book has been peer-reviewed to ensure high academic standards. For full review policies, see http://press.lse.ac.uk. Suggested citation: Wang, Y. 2022. Pension Policy and Governmentality in China: Manufacturing Public Compliance. London: LSE Press. DOI: https://doi.org/10.31389/lsepress.ppc. License: CC BY To read the free, open access version of this book online, visit https://doi.org/10.31389 /lsepress.ppc or scan this QR code with your mobile device: To my beloved family and friends About the author Dr Yan Wang is a Research Fellow in the School of Public Policy at the London School of Economics and Political Science (LSE), and will join the Department of Sociology at Lancaster University as a Lecturer (Assistant Professor) in Digital Sociology in Jan 2023. Her research seeks to understand the issues of state legitimacy and social policy, public opinion, and political communication. She is especially interested in how actors’ agency shapes authoritarian governmentality and the realisation of the public’s social rights, and how and why public opinion changes during the state-society interactions. Dr Wang has won several research grants, including from the Research Infrastructure and Investment Fund (RIIF) at LSE, and a British Academy Small Grant. Her recent publications include articles in the Journal of Chinese Political Science and elsewhere. Dr Wang received a BS in Sociology (2013) and a MA in Management (2015), from the Tsinghua University, and a PhD in Sociology (2020), from the London School of Economics and Political Science. Contents List of figures and tables xi Glossary xv Introduction 1 Chapter 1. Manufacturing compliance with ‘rule by design’ 9 Chapter 2. Who gets what and how: governance based on subpopulations 27 Chapter 3. Who deserves benefits and why – constructing fairness, pension expectations, and subjectivity 49 Chapter 4. Maximising support for pension reform using policy experimentation, and the potential to backfire 91 Chapter 5. Falsification of ‘manufactured compliance’ and wider legitimation and governmentality issues 129 Chapter 6. Pension issues, state governmentality, and falsified compliance in a comparative perspective 181 Appendix A. Data explanation and model validations 201 Appendix B. Data replication codebooks 235 Bibliography 241 Glossary anti-corruption campaign: the large-scale anti-corruption campaign initiated after the 18th National Congress of the Communist Party of China in late 2012. baoxian: insurance. campaign against the ‘four styles’: campaign against formalism, bureaucracy, hedonism, and extravagance in bureaucrats. This was promoted along with the anti-corruption campaign. CCP: Chinese Communist Party. cishi: the ‘CI SHI’ system, or ‘feudal prefectural governor’ system in China, which was originally established in the Qin and Han dynasty (around 202 BC) and continued to be used (with brief interruptions) until the Republic of China period in the early 1900s. CNKI: China Knowledge Resource Integrated Database. common sense: concept used by Gramsci, ‘Every social stratum has its own “common sense” and its own “good sense”, which are basically the most widespread conception of life and of man.’ (Gramsci 1971, p. 326). compliance: an umbrella concept of used in this book to cover many possible sources of legitimacy built up through individuals’ expectations and judgement based on their experience and the information in existing studies. It covers both positive/active consent and the passive or even coerced acceptance against state authority of the population; capturing the ‘willingness to defer to political authority regardless of the reasons’, or ‘with multiple motivations’ (Grimes 2008; Levi 1997). conduct of conduct: from Foucault’s theory of state power. ‘Perhaps the equivocal nature of the term conduct is one of the best aids for coming to terms with the specificity of power relations. For to “conduct” is at the same time to “lead” others (according to mechanisms of coercion which are, to varying degrees, strict) and a way of behaving within a more or less open field of possibilities. The exercise of power consists in guiding the possibility of conduct and putting in order the possible outcome. Basically power is less xvi Pension Policy and Governmentality in China a confrontation between two adversaries or the linking of one to the other than a question of government.’ (Foucault 1982, p. 789) consent: concept used by Gramsci, who argued that ‘consent’ from the dominated population can be generated by the civil society. Specifically, ‘1. The “spontaneous” consent given by the great masses of the population to the general direction imposed on social life by the dominant fundamental group; this consent is “historically” caused by the prestige (and consequent confidence) which the dominant group enjoys because of its position and function in the world of production. 2. The apparatus of state coercive power which “legally” enforces discipline on those groups who do not “consent” either actively or passively.’ (Gramsci 1971, p. 12) corpus: the collection of text data. danwei: work units. daode: ethics. DDD: the difference-in-difference-in-differences model, or triple difference model. decommodification: key feature for modern social welfare policies which is among the main indices for social rights measurement. The idea is that social welfare policies provide the possibility that entail citizens to ‘freely, and without potential loss of job, income, or general welfare, opt out of work when they themselves consider it necessary’ (Esping‐Andersen 1990, p. 23). defined contributory (DC) pension scheme: occupational pension schemes where participant’s own contributions and participant’s employer’s contributions are both invested and the proceeds used to buy a pension and/or other benefits at retirement. The value of the ultimate benefits payable from the DC scheme depends on the amount of contributions paid, the investment return achieved less any fees and charges, and the cost of buying the benefits. A DC scheme has a set contribution for the employee and a set contribution for the employer. delicacy social management: an idea of social governance promoted in China, which intends to manage the society in a modern, scientific, and rational way. DID: difference-in-differences model. DMR: Dirichlet Multinomial Regression Topic Model. dual system: description of the different schemes and return rates of pensions for government employees and enterprise employees. DVs: dependent variables. EE: enterprise employee. falsification: a convenient word to describe the situation of people constructing their public images/attitudes/preferences, which may not be exactly the same as their private images/attitudes/preferences. fazhan: development. Glossary xvii fully funding individual accounts: a pension reform conducted in early 2000s in China, to further clarify the division between the pooling of individual and social accounts, and to cover the deficit in individual accounts (State Council 2000). governmentality: this is a concept from Foucault, an approach that deciphers the state power with a focus on its technologies, governance objects and calculation (Foucault, Davidson, and Burchell 2008). Foucault recognised three dimensions of meaning for governmentality as state rationality. Governmentality refers to the ‘ensemble formed by institutions, procedures, analyses and reflections, calculations, and tactics that allow the exercise of state power … having the population as its target, political economy as its major form of knowledge, and the apparatuses of security as its essential technical instrument’, ‘the tendency, the line of force, that … has led to the development of a series of specific governmental apparatuses and the development of a series of knowledges’, and ‘the process, or rather, the result of the process by which the state of justice of the Middle Ages became the administrative state in the fifteenth and sixteenth centuries and was gradually governmentalized’ (Foucault 2009, p. 109). Essentially it is about how the state – through information management, benefit distribution and coercion, among many other approaches – maintains its rule over the population, and its rationale of doing so. Great Firewall: China’s system of internet censorship. hukou: a family registration programme in China that serves as a domestic passport, regulating population distribution and rural-to-urban migration. hybrid pension: a hybrid pension scheme is one which is neither a full defined benefit scheme nor a full defined contribution scheme, but has some of the characteristics of each. In a defined contribution scheme, the member generally bears the full risk (of paying higher costs or receiving reduced benefits) if investment return or pension costs are not as good as expected. In a defined benefit scheme, the employer usually takes that risk and pays higher contributions in order to maintain the agreed level of benefits. In hybrid schemes, the risk can be shared between the employer and employees, and in some cases the government as well. ILO: International Labour Organization. involution: the notion of ‘involution’ is used in the book to capture the increasingly obvious trend in Chinese political attitudes. In a situation where the state is a central source of power, the state can construct the knowledge (political or social) in whatever form is most suited to maintaining its power. Individuals who have limited access to alternative explanations of social facts are less likely to interpret the environment in a different way. In the long term, even when citizens are provided with a new possibility, they are not capable of accepting a different version of the story, or reaching out on their own initiative (as shown in Y. Chen and Yang 2019). Social xviii Pension Policy and Governmentality in China knowledge is still increasing in society; however, it increases without questioning past or current stories, and therefore is leading to an involution of the population’s ‘hard knowledge’ (Kuran 1997). iron rice bowl: a description of jobs in state-owned enterprises and public sectors before the 1980s. State-owned work units in urban China before 1978 were subsidised by the state so as to provide not only jobs to individuals, but also to generate from cradle to grave pensions, housing, education, and healthcare to employees and their dependents (Lu and Perry 1997). jiuye: employment. lagged time: time differences in pension burden across different enterprises, and in claiming pension benefits across different generations. laid off: become unemployed. When the economic reform started in the 1980s, numbers of SOEs were re-structured and hundreds and thousands of employees were laid off (Gu 1999, 2001). laodong: labour. LDA: Latent Dirichlet Allocation. legitimacy: the right and acceptance of an authority. The approach to conceptualising ‘legitimacy’ can be from the standpoint of legalism and normativity, as in issues of legality, or a particular normative definition of justice or procedure (e.g., Lord and Beetham 2001; Smoke 1994). It can also be analysed from the standpoint of belief, following Weberian notions (Weber 1978). LI: can be interpreted as ethics, manners and rules. LoR: locus of responsibility. metadata: data of the text in addition to the content, such as publication date and author. methodologically holistic and positive: focuses on the structure and involvement of the institution and interprets social facts in their historical context in order to understand the reasons for their emergence or change (Durkheim, Catlin, Mueller, and Solovay 1938). methodological individualism: social phenomena must be explained by showing how they result from individual actions, which in turn must be explained through reference to the intentional states that motivate the individual actors (Weber 2017). meticulous governance: similar term to delicacy social management. multi-layered pension system: pension schemes proposed by the World Bank. NDC: Notional Defined Contribution. nine-nine-six (996): work schedule of from 9 am to 9 pm, six days a week. OECD: Organisation for Economic Co-operation and Development. old-age insurance: similar term to pension, but a more general term referring to pension insurance. one study, one action: phrase meaning ‘study Xi’s thoughts, be a qualified youth league member’. Glossary xix one-child policy: China’s birth control policy which was introduced in 1979 and modified in the mid-1980s to relax regulations in rural areas (Scharping 2013). opening up and reform: China’s economic reform, launched in 1978. others: otherness, ‘others’ may be other individuals, social groups, or certain institutions. This is crucial for identifying the formation of groups and group boundaries. paradigm: a typical example or pattern of something; a pattern or model. PAYG: pay-as-you-go pension system, referring to state pension systems that are funded by contributions from current workers (rather than by individual past contributions from current beneficiaries). PAYG-DB: pay as you go-defined benefit. PEE: pension plan for enterprise employees. pension/social insurance: a pension that requires individual contributions. People’s Daily: the official newspaper (mouthpiece) of the Central Committee of the Communist Party of China and the biggest newspaper group in China. PGE: pension plan for government employees. PI: public institution. pilot: referring to the regions that joined the experiment as a treated group in policy experimentations. pluralistic ignorance: one possible mechanism of falsification in public opinion, such as when individuals may overestimate or underestimate the conditions of public opinion. policy experiments: one type of policy instrument, where in many cases the central government led the promotion of a new policy for a certain population in some selected regions. Sometimes, local authorities would initiate policy experiments. political status: referring to hukou, occupation, party membership and so on in China. PoS: parts of speech. PPIE: pension plan for public institution employees. PRC: People’s Republic of China preference falsification: interchangeable in the book with ‘falsified compliance’, taken as the hidden discontent with and disdain for the authorities in people’s voiced consent. PRR: pension plan for rural residents. PUR: pension plan for urban non-salaried residents. QTA: the quantitative text analysis. red-base: places such as Yan’an, Jing Gangshan, etc., where the CCP originated, or where some historical event has taken place. renmin gongshe: people’s communes, the highest administrative levels in rural areas of China. xx Pension Policy and Governmentality in China retirement allowance: full state-funded pension. retrenchment: a ‘risk privatisation’ process in welfare states, and the emergence of a hybrid welfare system in which individuals are given flexibility but increased responsibility for handling the various social risks related to their personal lives, such as ‘unemployment, death of a spouse, retirement, disability, childbirth, [and] poverty’ (Hacker 2002, p. 245). RMB: official currency in PRC. rongren: tolerance. shengchan: production. shouru: income. social coordination: one approach to solving funding issues of pension contributions proposed by the Chinese government. For instance, the enterprises that performed better and had fewer retirements were encouraged (later on, required) to pay into a social pool, which would be used to relieve the enterprises that had got into difficulties. Social Darwinism: the theory that human groups and races are subject to the same laws of natural selection as Charles Darwin perceived in plants and animals in nature. social protection: a general type of social welfare, which consists of policies and programmes designed to reduce poverty and vulnerability by promoting efficient labour markets, diminishing people’s exposure to risks, and enhancing their capacity to manage economic and social risks, such as unemployment, exclusion, sickness, disability, and old age. social rights: proposed by Marshall, this is the idea that it is people’s social right to receive the benefits brought by social protection, assistance, and insurance (Marshall 1964). It refers to rights to welfare and resources, such as the rights to health, education and a dignified level of social and economic wellbeing, regardless of economic standing (Plant and Jones 1991). socialisation of social welfare: a process during the welfare reform in China, in which the state, the market, the sectors of society, and families share welfare responsibility and the work units no longer take on the welfare function; joint responsibility in the practical method of funding, service provision and social welfare regulation. socialised self: self-regulated and self-motivated, a ‘socialised individual’ who can take care of him/herself while contributing to society and the general good. SOE: state-owned enterprises. State Council: the state council of PRC, this is the executive branchof thecentral government, executive body of theNational People’s Congress, and the highest organ of state administration. statecraft: various tactics used by the state to generate public compliance. STM: structural topic model. stratification: a key measurement proposed in Esping-Anderson’s model for classifying the type of welfare states, which takes in the social structure and Glossary xxi the concern of redistribution (Esping‐Andersen 1990). It describes the way in which social policy mediates and shapes societal inequalities. Study (XI) Stronger Country: a Chinese app primarily designed to teachXi Jinping Thought. It is strongly promoted by government institutions and party members are encouraged to download the app. subjectivity: this philosophical term relates to ideas of consciousness, agency, personhood, philosophy of mind, reality, and truth. supervised method: supervised machine learning, which infers a function from labelled training data consisting of a set of training examples. three security lines: social security system with Chinese characteristics promoted by the government, including a basic guaranteed living system for those no longer working in an SOE, intended to safeguard basic living standards, pay social insurance contributions and encourage re-employment; second, an unemployment insurance system, which provides unemployment benefits and actively encourages re-employment; and third, guaranteed minimum living standards for all urban residents whose family income per capita is lower than the prescribed level (K. W. Chan and Buckingham 2008). three-pillar system: pension model proposed by the World Bank in the 1990s. trial and error: a commonly used description of policy experimentation, also, ‘crossing the river by feeling the stones’. two studies, one action: phrase meaning ‘study the party regulations, study the leader’s speech, be a qualified party member’. typology: types, modes, classifications. unit-based self: relying on the work unit and ultimately the state. unsupervised method: unsupervised machine learning, automatically find interesting activity patterns in unlabelled data. US-China trade war: trade war between US and China, which began in 2018 and intensified in 2019. Wechat: a Whatsapp-like app used in China. Weibo: a Twitter-like online social media platform. WTO: World Trade Organization. xiagang: laid off. xiao: filial piety, caring for the elderly, not only caring for one’s own parents but also showing love, respect and support for all older people. xiaofei: consumption. Yuan: unit of RMB. zengzhang: increase. zhigong: enterprise employee. Zhongnanhai: current residence of the top leaders of the Party and central government (such as Central Politburo Standing Committee members). zhongyong: middle course. Introduction Since the beginning of the 19th century Chinese society has been subject to a number of momentous transformations. It has experienced imperial aggrandisement, state collapse, imperial occupation, political revolutions, transformations in its class structure, and integration into global markets. Important as each of these has been in Chinese history and in determining the shape of contemporary society, their salience has been matched by the current effort of the authorities to engineer an unprecedented social and economic transformation since 1978. China’s economic reform and opening up without loss of control raise fundamentally important issues for social scientists. The specific social and economic challenges during the transitional process – changes in economic endowment, ideological foundation, and social (re)distribution – provide a key opportunity for social scientists to re-examine theories of social change and regime domination. For instance, in Huntington’s discussion of modernisation and political order, he argued that a mismatch between social modernisation and institutional modernisation tends to produce social frustration and political instability (Huntington 2006). China’s ‘reform and opening up’ process strongly stimulated economic growth and social modernisation, and gave rise to an increasing demand for public participation. This transformation in the market, recalling Polanyi’s description of ‘great transformation’ (Polanyi and MacIver 1944), incurred an accelerated commodification of human capital, natural resources and other non-market values, such as social practices, family functions, and so on. However, the corresponding political institutions did not provide adequate channels for public participation. As shown in Figure I.1, the rapid development of the Chinese economy1 is in sharp contrast to the generally static nature of its political institutions.2 In addition to the gross increase in size of the economy, the economic structure has also changed; more diverse ownership of economic entities has emerged and contributes to the economic growth. How to cite this book chapter: Wang, Yan. 2022. Pension Policy and Governmentality in China: Manufacturing Public Compliance. London: LSE Press, pp. 1–8. DOI: https://doi.org/10.31389/lsepress.ppc.a. License: CC BY 8 Pension Policy and Governmentality in China Meng, K. (2018). China’s Pension Reforms: Political Institutions, Skill Formation and Pension Policy in China. Routledge. https://doi.org/10.4324%2F9781351061667 Moore, B. (1966). Social Origins of Democracy and Dictatorship. USA: Beacon. Nathan, A. (2003). ‘China’s changing of the guard: Authoritarian resilience’. Journal of Democracy, vol. 14, issue 1, 6–17. https://doi.org/10.1353/jod .2003.0019 Pan, J. (2020). Welfare for Autocrats: How Social Assistance in China Cares for Its Rulers. USA: Oxford University Press. https://doi.org/10.1093%2Foso %2F9780190087425.001.0001 Polanyi, K.; and MacIver, R. M. (1944). The Great Transformation. USA: Beacon. Yan, X. (2017). How Is China Maintain Stable? Observation and Thoughts from Fieldwork [Zhongguo Heyi Wending: Laizi Tianye de Guancha yu Sikao]. Hong Kong: Joint Publishing. CHAPTER 1 Manufacturing compliance with ‘rule by design’ Any power relation reveals the need of the dominant side to maintain its authority. The power relation between the state and the individual, which has been a fundamental topic in political sociology and political science, is a typical dominant–dominated power relationship. For a modern state, its survival requires order, stability, and effective governance – all demanding considerable cooperation from its population. While differing in content, several classical studies share the logic that the legitimation of the government must secure enough cooperation to maintain the system’s capacity to withstand shock and failure (Giddens 1981; Lipset and Man 1960; Tilly 2017). From legitimacy, the subordinates in a power relationship can obtain moral grounds for cooperation and obedience, while the poweror authority-holder gets the right to expect their compliance. Under certain conditions the population renders to the authority cooperation and obedience – which provide enough stability for the authority to continue in being, withstanding some shocks and failures. At a point in their interactions with the state the public may change its compliance. People who have suffered during the interactions may choose to resist or withdraw compliance, for instance, because of the state’s misbehaviour, or because they are dissatisfied with social policies. Many political sociologists argue that the essential theme of state ‘ruling’ is the issue of consent. The Weberian tradition regards the legitimation of the government as an essential factor for sustaining people’s compliance, since the state’s rule fundamentally depends on consent, rather than any fear induced by coercion alone. States cannot do whatever they want and expect their citizens to acquiesce. Constant and crude coercion may cause a revolution from below and eventually state collapse. As Weber (1978) pointed out, consent from the public is necessarily rooted in people’s belief. The consent How to cite this book chapter: Wang, Yan. 2022. Pension Policy and Governmentality in China: Manufacturing Public Compliance. London: LSE Press, pp. 9–26. DOI: https://doi.org/10.31389/lsepress.ppc.b. License: CC BY 10 Pension Policy and Governmentality in China that signals the people’s active acceptance and compliance may take the form of ideological trust or spontaneous supportive actions. However, beliefs are not arbitrary and can be manufactured by exogenous powers. Gramsci’s hegemony theory (Gramsci, Hoare, and Nowell-Smith 1971) suggests that the interaction between the population and the state is never a simple, watertight match. People’s seemingly ‘spontaneous’ attitudinal or behavioural consent results from some invisible and subtle ideological, cultural, or institutional infiltration from the state. In other words, in the state–society relationship the state can reconstruct and use people’s knowledge, politics, and even daily lives in a way that favours the authority. The other side of the story is that resonating beliefs must be rooted in the experiences of individuals (Gramsci, Hoare, and Nowell-Smith 1971). Any inquiry into consent must grapple with both the expectations of citizens and the behaviour of states, and the fit of both with one another. Thus, although the state generally holds more institutional resources than individuals do, it is not all-powerful; its legitimacy can be earned or lost. A ‘legitimacy crisis’ ensues when the regime is finding it difficult to manage the equilibrium between state governance and the people’s expectations. The idea of ‘governmentality’ answers the questions of whom to govern and how to govern. As a form of state rationality, governmentality can be recognised as a power relation between ‘man and things’ (Foucault 2009, p. 97); state governance therefore involves governing ‘a sort of complex of men [and women] and things’ (Foucault 2009, p. 96). The objects of the governance include not only people and their complex relationship with things (such as resources) but also things’ relationship with other things, such as customs, dependency, habits, and so on. The instruments that a government uses (such as statistics and biopolitics) to achieve a better well-being of the population (such as the population’s wealth, longevity, and health) involve and act directly on the population itself. State governmentality can be identified as institutions, regulations, and procedures. It can also be identified as a tendency of changing forces which reveals the state’s sophisticated understanding of the people, and thus the efforts that it makes to figure out proper ways of managing changes and consents, and the process of knowledge development (Foucault 2009, p. 109). Through the dynamic governmentality that has been designed and adjusted by the state, the boundary of the state, the boundary between being ‘public’ and being ‘private’, is revealed. A thorough investigation of governmentality will unpack not only the way that governmental activities unfold their effect but also how the individual’s subjectification process is shaped by statecraft, and how the risks of resistance from the public affect statecraft in return. Unpacking governmentality, therefore, is important in addressing the legitimation and issue of ruling by the modern state. In empirical investigations of governmentality, which indicators are appropriate in identifying the nature of the state’s power over the people? Many concepts have been used to describe the ruling of authority over the people Manufacturing compliance with ‘rule by design’ 11 and could indicate the success of the state’s governance. Commonly used terms include the ones mentioned above: ‘legitimacy’, ‘consent’, or other concepts such as ‘political trust’ and ‘compliance’. The conceptual details of such notions are also debated by different scholars. For instance, the approach to conceptualising ‘legitimacy’ can be from the standpoint of legalism and normativity, as in issues of legality, or a particular normative definition of justice or procedure (e.g. Lord and Beetham 2001; Smoke 1994). It can also be analysed from the standpoint of belief, following Weberian notions (Weber 1978). As well as drawing on the Weberian and Gramscian tradition of understanding state rule and highlighting the individual’s rationale of ‘believing’ and ‘consent’, my work also takes account of the Foucaudian ‘governmentality’ that the state uses to maintain its rule and investigates its underlying rationality. What then is the best way of integrating these approaches into one analytical model? Compliance, which covers both positive/active consent and the passive or even coerced acceptance of state authority by the population, seems to be a good indicator to measure the state’s rule over its people in general. In this sense, the concept of compliance is more appropriate than consent in capturing the ‘willingness to defer to political authority regardless of the reasons’, or ‘with multiple motivations’ (Grimes 2008; Levi 1997). In the following section I further discuss the theoretical model constructed on the inclusive measurement of ‘compliance’. The statecraft used in different regimes consists of choices built from an authority’s own political, social, and economic scenarios. Compared to their democratic counterparts, authoritarian states have more of a reputation for using oppression and violence in sustaining their authority. But no regime that relies only on oppression and violence can stabilise its governance and maintain its resilience in the long term (Gramsci, Hoare, and Nowell-Smith 1971). In his study, Dimitrov (2013) implied that, in a mature communist regime, the use of repression actually declines, and patronage is distributed to a wider segment of the population, citing the Soviet Union after the death of Stalin. If this is so, an authoritarian regime also needs to employ various strategies and skills to manufacture ‘spontaneous’ consent from the public. Moreover, such a regime in a prolonged transformational stage, like China over the past 40 years, requires significant public compliance to maintain overall stability. 1.1 Transitional scenarios in China and the state’s ‘rule by design’ China’s transition is one form of the large-scale institutional changes in communist regimes that started in the late 20th century. The process of de-Stalinisation, liberation, and democratisation involved the deformation of central planning and a transition of state socialism. In these post-communist states, the newly emerged market has led to a change in the distributional principles and in the rate of return on financial, productive, and human capital, 12 Pension Policy and Governmentality in China and has also altered the social structure. Meanwhile, the boundaries between state, market, and society have been redrawn and negotiated through interactions between different sectors. Challenges to the legitimacy of the Chinese government have followed one another since the economic reforms of 1978. Economic development has increased people’s income, and also changed the public’s expectation of the state–individual relationship. People’s loyalty towards state authority has become more complicated. On the one hand, the benefits brought by economic development were able to help the state to buy public support; on the other, economic development brought people a keener consciousness and more demands for other rights, and therefore threatened the public’s acceptance of the state’s authority. Moreover, the transition process extended from the economy to society: 40 years of development in the economy has led to an expansion of social inequality along with the entrenchment of a social class system. The transitional scenarios brought by marketisation, privatisation, and institutional reform have led to some fundamental changes in distribution, redistribution, and social justice. In addition to the large-scale economic-social reforms, the transition of socialist regimes features another key point, due to the special ideological foundation and historical legacy of China. As a communist government, the authorities rely heavily on the ‘red ideology’ in the political culture, which was also one of the main sources of the CCP’s legitimacy during the civil war period and the early years after the founding of the People’s Republic of China (PRC). With markets penetrating every aspect of social life, communist ideology has also been reformed in the post-Mao era, particularly in recent years to fit the new social and economic scenarios. For instance, the illusion of government omnipotence persists and coexists with mixed attitudes to liberal notions. Many people buy into the official story that individual merit is highly encouraged, while ‘waiting, relying on, and demanding’ (government help) is criticised (Gallagher 2011). The spirit of selfless devotion and equality from the revolutionary period has been undermined. In other words, the transitional process also shows a de facto shift away from the revolutionary communist legacy that originally gave power to the Party: a state-controlled planned economy, semiuniversal welfare provision and a communist ideology. Social welfare provision is one of the most important areas of change, partly because it can be directly perceived by the public. In this area there has been a noticeable trend, which has led to some degree of public discontent owing to the fear of loss of social benefits and an increased welfare burden on some social groups. Before the economic reforms of 1978, China was experiencing the ‘creation of a socialist egalitarian society promising a relatively stable livelihood at the expense of economic development’ (Leung and Xu 2015, p. 33). Work units (danwei) acted in urban areas as administrative social integration sections, as well as providers of public goods (Lu and Perry 1997). Urban work units provided not only jobs for life but also pensions, housing, education, and health care to employees and their dependants. More than 80% of the urban Manufacturing compliance with ‘rule by design’ 13 labour force was covered by the danwei system (Leung and Wong 1999). At the stage of state socialism the state’s patriarchal role underwrote a collective welfare mechanism that collectively secured social rights (Xie 2016). Corresponding to the economic reform, the state promoted social reforms that helped to shift the state’s welfare burden and boost efficiency. The ongoing reform in welfare provision threatens lifetime employment, pensions, health and the housing system in urban areas (Li and Zhong 2009; Wong and Ngok 2006). Very many employees of state-owned enterprises were laid off during the marketisation process. The newly established basic health insurance scheme also required contributions from individuals and employers. Furthermore, the hospitals started to employ market-competition principles in their operation. Individual workers found that their total welfare contribution accounted for quite a large proportion of their salary (Ringen and Ngok 2017). Many of these reform plans were understood to add up to a withdrawal of governmental responsibility and an emphasis instead on the roles of the market and the family. The schemes do not treat equally members of different age groups, occupation groups, and political identity groups; some enjoy more benefits than others. Facing all the dramatic social and economic changes described above, the state needed to spend some time on generating public compliance to prevent a serious crisis of legitimacy and even regime change caused by bottom-up revolts against the transformation. Theoretically speaking, the skills in the state’s governance toolkit include propaganda, repression, absorption, allocated benefits, and so on. In practice, what we can observe of the state’s efforts to manufacture compliance is different ‘strategies’ that involve either one or several skills. For instance, in promoting certain policies, a state could combine economic benefits with propaganda (comparable to the ‘social construction of target populations’ (Schneider, Ingram, and DeLeon 2014)), while coercion and information censorship could also be used to attain its goal. The toolkit of maintaining compliance for an authority also changes depending on the various challenges in different periods. For example, in Mao’s era, when the party class identified struggle as the main problem in China, an ideological campaign was the central approach to producing loyalty and consent. In the ‘opening up and reform’ period, many scholars have argued that the Chinese authorities has used performance-based legitimacy and maintains its rule by providing material security (Zhao 2001). Some scholars have highlighted the institutional features and tricks in China’s governance. For instance, Andrew Nathan used the term ‘authoritarian resilience’ to describe the situation whereby the Chinese government reconsolidated itself in the midst of the political instability and potential governance crises. He attributed the authority’s resilience mostly to the ‘institutionalisation’ of the state, such as the normalisation of succession politics within the Communist Party leadership and the meritocratic promotion of bureaucrats (Nathan 2003). Scholars such as Yan (2017) have emphasised that the capacity of the state to absorb and assimilate is one of the main instruments that 14 Pension Policy and Governmentality in China keeps the state and the party stable. Perry (2017) contended that, by strategically using symbolic resources such as the traditional culture, the state shapes the higher education institutions and wins the allegiance of social elites. Other descriptions in existing studies include ‘a balancing act involving the supply of carrots and sticks’ (Gallagher and Hanson 2009), a ‘guerrilla policy style’ (Perry and Heilmann 2011), ‘nationalism ideological articulation’ (Bernstein 2013; Gries 2004), and so on. One common issue in existing studies is that statecraft is treated as a set of static and isolated skills rather than as a comprehensive, sophisticated design. The role of ‘the ruled’ – the people or population – is also missing or under-emphasised in shaping the specific governmentality. In this book I argue that the Chinese state uses a strategy that is hybrid, organic, and dynamic to respond to the potential crisis brought closer by social and economic transformation, and to generate public compliance even though it drifts away from the communist legacy. In particular, I emphasise that public compliance is not only acquired through buying off the public with governmental performance and transferred benefit but is also manufactured through an ideological foundation, such as nationalism, which has been rebuilt by the authority. China’s authoritarian governance has been an active process of ‘rule by design’ that has constantly adapted to new social and economic situations, especially since the notorious repression in the late 1980s. On the one hand, the state monitors and captures public expectations and adjusts its own strategies to meet them; on the other, the state intentionally shapes the public’s expectations and manufactures compliance to keep its reforms working. For instance, in the field of social welfare, where the process of privatisation might cause severe discontent, the authority has employed a mixture of retrenchment and generosity and designed diverse schemes for different social groups in order to effectively ‘divide and govern’. Moreover, the state has carefully employed propaganda skills using traditional culture and the ideological legacy of the socialist period, in order to legitimise its choices and engender consent from the public. This research enriches the discussion of authoritarian resilience by highlighting the active consent role of the public and the constraints that it imposes on the effectiveness of the state’s governance and production of legitimacy. 1.2 Manufacturing compliance in the state–individual interaction States provide security, resources, solidarity, and identity through benefit allocation, propaganda, education, and many other approaches, in order to shape public expectations and justify their rule; while individuals can update their knowledge about the state from personal benefits, public policy, and current society (their peers) and decide whether to stay loyal or rebel. To better understand the logic of the state’s rule and its choices of various forms of statecraft in dealing with its population, I first decipher the state’s governmentality, Manufacturing compliance with ‘rule by design’ 15 moving from the Foucauldian theories to formal models and empirical analyses. The second subsection below explains the analytical logic of my work with a semi-modelled clarification. I adopt a holistic viewpoint, integrating across both consent-oriented statecraft and coercion-based statecraft, and highlighting the state’s strategic selection of statecraft based on its objectives and constraints. Thus, the analytical paradigm of my research uses an interactive model with two actors’ strategies and behaviours involved: the state authority and the population.1 Coercion or consent: why coercion alone cannot do the work Gramsci argued that the state is an entire complex of practical and theoretical activities with which the ruling class not only justifies and maintains its dominance but manages to win the active consent of those over whomever it rules (Gramsci, Hoare, and Nowell-Smith 1971). The state is a combination of dictatorship and hegemony in which the state power relies on both political society (force) and civil society (consent), and these choices match up with using either force and coercion against its population, exercising dictatorship; or using consensus building, cultural hegemony, and maintaining moral and intellectual leadership (Kohli, Shue, and Migdal 1994). A coercion-based approach is very efficient in maintaining short-term stability and long-term fear. However, consent-based governance enables the state to enjoy more compliance and the ‘expression of these subaltern classes who want to educate themselves in the art of government’ (Gramsci, Hoare, and Nowell-Smith 1971, p. 126). In the spectrum of coerced compliance and voluntary compliance, authoritarian states have the reputation of using oppression and violence to sustain their rule. Therefore, in the context of an authoritarian regime, an inevitable question arises: why would the state need to manufacture consent if it can instead just use coercion or force to stay in power? The questions ‘why can’t coercion alone do the work’ and ‘why is manufacturing active compliance from the population important for rulers in authoritarian regimes?’ have been studied in many empirical and theoretical works. Empirically, we do see some authoritarian regimes, or regimes at an authoritarian stage, showing significant use of coercion and force in their rule. For instance, mass terror, repression and indoctrination were frequently used in totalitarian periods by Stalin, Mao, Pol Pot, and military regimes such as Franco’s Spain and Pinochet’s Chile. Compliance generated by fear was common in these regimes. But any regime that relies only on oppression and violence cannot stabilise its governance and maintain its resilience in the long term. The integrated global economy and the development of the internet have also made mass killing costly and less attractive for autocrats. In recent years, we have observed more non-democratic regimes using sophisticated statecraft to hold on to power. For instance, some imitate the format of democracy and 16 Pension Policy and Governmentality in China hold elections to obtain normative legitimacy, while their election processes exhibit wholesale bribery, illegal competition, and information manipulation. Wedeen’s case of Asad’s cult in Syria argued that rhetoric and symbolic display reduces the need to rely on sheer repression as a mechanism of control (Wedeen 1999). Some studies have also employed formal models to demonstrate the unsustainability of using violence in non-democratic regimes and justify an authority’s choice of hybrid statecraft to maintain compliance. By combining the manufacture of consent with coercive tools, the state can maintain a stable hegemonic position vis-à-vis the population. Repression/violence are among the most extreme ways to crush protest/revolution and alter public opinion through physical coercion (Gregory, Schröder, and Sonin 2006). However, repression is not a once-and-for-all solution. In Kricheli, Livne, and Magaloni’s (2011) working paper, they presented a formal model of protest under authoritarianism. Their two-period signalling model showed that, although regimes which are more repressive in the first period can better deter civil opposition, they are more likely in the second period to experience a cascade of power since protests’ information-revealing potential is maximised in these regimes. In addition, the work of Acemoglu, Ticchi, and Vindigni (2010), Svolik (2012), and Egorov and Sonin (2011) showed that the authority may experience a greater threat from its military allies once the period of acute repression is over. Thus, in most cases, the authority tends to consider integrating different tactics in preventing revolution and gaining compliance after rational assessment by the governed. Scholars citing theoretical and empirical evidence have identified many specific combinations. Wintrobe (1990; 2007) modelled two instruments – repression and loyalty – that dictators used to stay in power, dividing such regimes into four categories – tinpots, tyrants, totalitarians, and timocrats – according to their different objectives and correspondingly invested instruments. For instance, totalitarians always aim to maximise their power; therefore, they combine high repression with a capacity to generate loyalty, while tinpot dictators prefer to maximise their own benefits under the constraint of minimum power, so their investment is low on both counts. Although theoretically feasible and easily achieved, eliciting compliance through a redistribution of benefits – irrespective of whether the distribution is to the ruling alliance or to the remaining population – needs credible commitment from the authority (e.g. Boix and Svolik 2013; Gehlbach and Keefer 2011; Myerson 2008). In nondemocratic regimes such commitment is often fragile without relevant institutional guarantees. When the state can make only fragile commitments, those who benefit from distributions will discount the effectiveness of its effort to allocate benefits to ensure compliance. In addition to buying compliance through material means, a more sophisticated way of generating voluntary compliance is to wisely use information and manipulate through censorship, guidance by propaganda, or knowledge Manufacturing compliance with ‘rule by design’ 17 construction. For instance, a state authority may allow free social media so as to obtain the information about the population that the state needs, even though the information may also be used by society to coordinate its protest (Egorov, Guriev, and Sonin 2009). To avoid a potential backfire from the free media, the authorities may also actively send out biased signals through their own propaganda to mislead the public in evaluating the state’s capacity (Edmond 2013). Chen and Xu’s work (2017) presented a new view: that allowing people’s information communication in society actually helps the authorities to obtain material and prevent coordinated revolt from the public. To return to the main question of this subsection, how useful are these sophisticated tactics compared to simple violence? Guriev and Treisman’s work in 2015 proposed a comprehensive argument about the different ways in which modern dictators could help themselves survive. In a game of political leaders trying to convince citizens (some of whom are informed elites) of their competence, a ‘dictator can invest in making convincing state propaganda, censoring independent media, co-opting the elite, or equipping police to repress attempted uprisings’ (Guriev and Treisman 2015, p. 4). The authors showed that the portfolios of states’ techniques differ with the competence of the leader and result in multiple equilibria while: violence either is a last resort when all else has failed, or is used sparingly when it is possible to conceal it, since competent dictators do not need to use repressions, and reverting to repressions immediately reveals the dictator’s incompetence to the public and ultimately results in his [or her] downfall. (Guriev and Treisman 2015, p. 33). To summarise, violence alone cannot do the work of maintaining governance for an authority which wants to rule in the long term. The modern state enjoys considerable choices of statecraft that enable it to stay in power without much challenge from the population. Generating compliance: a two-actors model and the state’s options The ultimate goal for the state is to remain in power, in other words to ensure that the population is compliant. Putting the population’s reaction on a continuous scale, the state would prefer sincere support (active consent) to forced obedience, while the public’s collective non-compliance is more dangerous for the state than a single individual’s public non-compliance. There are cases where an individual’s private non-compliance goes unnoticed or is tolerated by the authority as long as it does not turn into public non-compliance or, worse still, public collective non-compliance. The state as ruler enjoys resources (such as economic or organisational resources) that could be used to achieve its objectives, but it may not master every detail of the population or its views. 24 Pension Policy and Governmentality in China Egorov, G.; Guriev, S.; and Sonin, K. (2009). ‘Why resource-poor dictators allow freer media: A theory and evidence from panel data’. American Political Science Review, vol. 103, issue 4, 645–668. https://doi.org/10.1017 /s0003055409990219. OA: https://ssrn.com/abstract=898888 Foucault, M. (2009). Security, Territory, Population: Lectures at the Collège de France, 1977–78 (M. Senellart ed.). Palgrave Macmillan. Gallagher, M. E. (2011). Contagious Capitalism: Globalization and the Politics of Labor in China. Princeton University Press. Gallagher, M.; and Hanson, J. K. (2009). ‘Coalitions, carrots, and sticks: economic inequality and authoritarian states’. PS: Political Science & Politics, vol. 42, issue 4, 667–672. https://doi.org/10.1017/s1049096509990096 Gehlbach, S.; and Keefer, P. (2011). ‘Investment without democracy: Ruling-party institutionalization and credible commitment in autocracies’. Journal of Comparative Economics, vol. 39, issue 2, 123–139. https://doi .org/10.1016/j.jce.2011.04.002 Giddens, A. (1981). A Contemporary Critique of Historical Materialism: The Nation-State and Violence. University of California Press. Gramsci, A.; Hoare, Q.; and Nowell-Smith, G. (1971). Selections from the Prison Notebooks of Antonio Gramsci. Lawrence and Wishart Limited. Gregory, P. R.; Schröder, P. J.; and Sonin, K. (2006). ‘Dictators, repression and the median citizen: An “eliminations model” of Stalin’s Terror’ (Data from the NKVD Archives). https://doi.org/10.2139/ssrn.948667 Gries, P. H. (2004). China’s New Nationalism: Pride, Politics, and Diplomacy. University of California Press. Grimes, M. (2008). ‘Consent, political trust and compliance: Rejoinder to Kaina’s remarks on “Organizing consent”’. European Journal of Political Research, vol. 47, issue 4, 522–535. https://doi.org/10.1111/j.1475-6765.2008.00774.x Guriev, S.; and Treisman, D. (2015). How Modern Dictators Survive: An Informational Theory of the New Authoritarianism. Retrieved from https://doi .org/10.3386%2Fw21136 Han, R. (2015). ‘Defending the authoritarian regime online: China’s “voluntary fifty-cent army”’. The China Quarterly, vol. 224, 1006–1025. https://doi .org/10.1017/s0305741015001216 Huang, H. (2015). ‘Propaganda as signaling’. Comparative Politics, vol. 47, issue 4, 419–444. https://doi.org/10.5129/001041515816103220 Kohli, A.; Shue, V.; and Migdal, J. S. (1994). State Power and Social Forces: Domination and Transformation in the Third World. UK: Cambridge University Press. Kricheli, R.; Livne, Y.; and Magaloni, B. (2011). Taking to the Streets: Theory and Evidence on Protests under Authoritarianism. Paper presented at the APSA 2010 Annual Meeting Paper. Kuran, T. (1991). ‘Now out of never: The element of surprise in the East European revolution of 1989’. World Politics, vol. 44, issue 1, 7–48. https://doi .org/10.2307/2010422 Manufacturing compliance with ‘rule by design’ 25 Le Bon, G. (1897). The Crowd: A Study of the Popular Mind. Fischer. Leung, J. C.; and Wong, H. S. (1999). ‘The emergence of a community-based social assistance programme in urban China’. Social Policy & Administration, vol. 33, issue 1, 39–54. https://doi.org/10.1111/1467-9515.00130 Leung, J. C.; and Xu, Y. (2015). China’s Social Welfare: The Third Turning Point. John Wiley & Sons. Levi, M. (1997). Consent, Dissent, and Patriotism. Cambridge University Press. Li, B.; and Zhong, Y. (2009). ‘How did China’s transitions impact people’s welfare benefits in the reform era?’ Journal of Contemporary China, vol. 18, issue 62, 813–829. https://doi.org/10.1080/10670560903174606 Lipset, S. M.; and Man, P. (1960). The Social Bases of Politics. USA: Johns Hopkins University Press. Lohmann, S. (1993). ‘A signaling model of informative and manipulative political action’. American Political Science Review, vol. 87, issue 2, 319–333. https://doi.org/10.2307/2939043 Lohmann, S. (1994). ‘The dynamics of informational cascades: The Monday demonstrations in Leipzig, East Germany, 1989–91’. World Politics, vol. 47, issue 1, 42–101. https://doi.org/10.2307/2950679 Lord, C.; and Beetham, D. (2001). ‘Legitimizing the EU: Is there a post-parliamentary basis for its legitimation?’ JCMS: Journal of Common Market Studies, vol. 39, issue 3, 443–462. https://doi.org/10.1111/1468-5965 .00298 Lorentzen, P. (2014). ‘China’s strategic censorship’. American Journal of Political Science, vol. 58, issue 2, 402–414. https://doi.org/10.1111/ajps.12065 Lu, X.; and Perry, E. J. (1997). Danwei: The Changing Chinese Workplace in Historical and Comparative Perspective. Me Sharpe. McCarthy, J. D.; and Zald, M. N. (1977). ‘Resource mobilization and social movements: A partial theory’. American Journal of Sociology, vol. 82, issue 6, 1212–1241. https://doi.org/10.1086/226464 McMillan, J.; and Zoido, P. (2004). ‘How to subvert democracy: Montesinos in Peru’. Journal of Economic Perspectives, vol. 18, issue 4, 69–92. https://doi .org/10.1257/0895330042632690 Meyer, D. S.; and Staggenborg, S. (1996). ‘Movements, countermovements, and the structure of political opportunity’. American Journal of Sociology, vol. 101, issue 6, 1628–1660. https://doi.org/10.1086/230869 Myerson, R. B. (2008). ‘The autocrat’s credibility problem and foundations of the constitutional state’. American Political Science Review, vol. 102, issue 1, 125–139. https://doi.org/10.1017/s0003055408080076 Nathan, A. (2003). ‘China’s changing of the guard: Authoritarian resilience’. Journal of Democracy, vol. 14, issue 1, 6–17. https://doi.org/10.1353/jod .2003.0019 Perry, E. J. (2017). ‘Higher education and authoritarian resilience: The case of China, past and present’. Harvard-Yenching Institute Working Paper Series. OA: https://dash.harvard.edu/handle/1/30822717 26 Pension Policy and Governmentality in China Perry, E. J.; and Heilmann, S. (2011). ‘Embracing uncertainty: Guerrilla policy style and adaptive governance in China’. Mao’s Invisible Hand: The Political Foundations of Adaptive Governance in China. Ringen, S.; and Ngok, K. (2017). ‘What kind of welfare state is emerging in China?’ Towards Universal Health Care in Emerging Economies, 213–237. Springer. https://doi.org/10.1057%2F978-1-137-53377-7_8 Schneider, A. L.; Ingram, H.; and DeLeon, P. (2014). ‘Democratic policy design: Social construction of target populations’. In P. Sabatier and C. Weible (eds), Theories of the Policy Process, 3rd edn, 105–149. Simon, J. (2014). The New Censorship: Inside the Global Battle for Media Freedom. Columbia University Press. https://doi.org/10.7312%2Fcolumbia %2F9780231160643.001.0001 Smoke, R. (1994). ‘On the importance of policy legitimacy’. Political Psychology, 97–110. https://doi.org/10.2307/3791441 Svolik, M. W. (2012). The Politics of Authoritarian Rule. Cambridge University Press. https://doi.org/10.1017/CBO9781139176040 Tilly, C. (2017). ‘Coercion, capital, and European states, AD 990–1990’. Collective Violence, Contentious Politics, and Social Change, 140–154. Routledge. https://doi.org/10.1080/714005469 Weber, M. (1978). Economy and Society: An Outline of Interpretive Sociology, vol. 1. University of California Press. Wedeen, L. (1999). Ambiguities of Domination: Politics, Rhetoric, and Symbols in Contemporary Syria. University of Chicago Press. https://doi.org/10.720 8%2Fchicago%2F9780226345536.001.0001 Wintrobe, R. (1990). ‘The tinpot and the totalitarian: An economic theory of dictatorship’. American Political Science Review, vol. 84, issue 3, 849–872. https://doi.org/10.2307/1962769 Wintrobe, R. (2007). ‘Dictatorship: analytical approaches’, The Oxford Handbook of Comparative Politics. UK: Oxford University Press. https://doi.org/10.109 3%2Foxfordhb%2F9780199566020.003.0016 Wong, L.; and Ngok, K. (2006). ‘Social policy between plan and market: Xiagang (off‐duty employment) and the policy of the re‐employment service centres in China’. Social Policy & Administration, vol. 40, issue 2, 158–173. https:// doi.org/10.1111/j.1467-9515.2006.00482.x Xie, Y. (2016). ‘Understanding inequality in China’. Chinese Journal of Sociology, vol. 2, issue 3, 327–347. https://doi.org/10.1177/2057150x16654059. OA: https://europepmc.org/article/MED/29854420 Yan, X. (2017). How Is China Maintain Stable? Observation and Thoughts from Fieldwork [Zhongguo Heyi Wending: Laizi Tianye de Guancha yu Sikao]. Hong Kong: Joint Publishing. Zhao, D. (2001). ‘China’s prolonged stability and political future: Same political system, different policies and methods’. Journal of Contemporary China, vol. 10, issue 28, 427–444. https://doi.org/10.1080/10670560120067126 CHAPTER 2 Who gets what and how: governance based on subpopulations Looking at observable public policies in order to capture the tactics used by the state in maintaining its legitimacy is complicated, especially as regards social policies that change frequently and differ across various social groups. However, the governmentality logic underlying whom to govern and how to govern is shown in the design of governmental programmes, in the way that social problems are defined and divisions/distinctions are established, and in the different kinds of knowledge produced to shape people’s ideologies. Here and in the next chapter, I probe into the different designs of policy experiments and schemes, examine the techniques intentionally used by the Chinese government to shape the public’s expectations and manufacture public consent, and discuss the rationale behind these practices. In China’s current, massive transition, social welfare provision is one of the most important areas that can be directly perceived by the public. It is also an arena in which we can vividly see a close interaction between the state and the public. For instance, there may be a ‘trade-off’ between the opportunities and challenges of promoting a welfare reform in pensions or health care. Going ahead affects how the state manages its budget, whether it can increase economic efficiency through privatisation and so on. At the same time, it also brings challenges due to the fact that some sectors will disappear during the process, and the allocation of social benefits will face a possible major reshuffle. To best capture the details from a specific area of social policy, my empirical work in this chapter focuses on the case of major pension policy changes during the social welfare reform conducted by the Chinese government after the ‘reform and opening up’ in 1978. I look intensively at social policies relating to welfare for the elderly (such as old-age insurance, pensions, and so on) because this area is one that can reveal the state’s purposeful design in modifying the distribution of public goods and balancing the role of the government and How to cite this book chapter: Wang, Yan. 2022. Pension Policy and Governmentality in China: Manufacturing Public Compliance. London: LSE Press, pp. 27–48. DOI: https://doi.org/10.31389/lsepress.ppc.c. License: CC BY 28 Pension Policy and Governmentality in China the public in welfare provision. The existence of multiple programmes – such as basic social security, old-age insurance, and pension plans – also suggests that social policies relating to old-age benefits are in an area where many negotiations are possible. People are less likely to give up their existing benefits; thus, the state needs to devote greater effort to the changes that it plans. The key questions I answer in this chapter are ‘How is access to welfare resources distributed among different social groups, and how does the distribution change along with the reforms?’ In the reformed welfare system of China, the social rights to income and social security are now more frequently defined on an individual basis than they were in the previous system. They used to be defined collectively through work units (danwei, a place of employment) and people’s communes (renmin gongshe, the highest administrative levels in rural areas). Unlike the old system, in the new system, the issue of who gets what and how they get it has gradually become connected to individuals’ endowments, such as their social status, political status, social capital, and so on. From the late 1990s, the government promoted various pension scheme reforms, under which people with different occupations have their own distinctive accessibility to various welfare benefits. For instance, the reform of basic pension scheme for enterprise employees distinguishes the individual contribution rate levied on different age groups, a scheme that was not applied to employees from other sectors such as public institutions before the 2010s. Identifying how and why the state distributes certain benefits but not others, and to some groups/regions but not others, may help to answer the questions regarding the state’s overview of population-based governance and its tactics of resource differentiation. To address the question of resource differentiation, I trace the social schemes related to old-age benefits in China from the early 1950s (with a focus on the period after 1978’s ‘reform and opening up’) and sort out comparisons among the different policy schemes that were available to people of different political status (different urban/rural areas or hukous,1 occupations, party membership, and so on). In the 2010s there were several parallel pension plans, because distinct schemes were devised to cover people of different status. There are pension plans tailored for government employees, for employees of public institutions, employees of enterprises, urban non-salaried residents, and rural residents. The political status that combines residential and occupational difference is crucial for Chinese people seeking access to these differentiated welfare benefits. 2.1 How do entitlements differ: differentiation in benefit entitlement Welfare benefits, as a type of scarce resource (or public good), can be used by the state in exchange for loyalty, forming coalitions and breaking down a population’s capacity to coordinate any critical collective action. In theory, the distribution of welfare benefits can be the result of the authority’s intentional Who gets what and how: governance based on subpopulations 29 manipulation, negotiations between interest groups, and the influence of sociohistorical factors – for instance, demographic and geographical features that may respectively lead to the varied distribution of health issues, and change the distribution of certain health care benefits. Therefore, it is difficult to image a welfare regime in which everyone enjoys exactly the same social benefits – we are more likely to see various welfare schemes and programmes tailored for different social groups and regions, some enjoying more social benefits than others (Frazier 2010). In his book The Divided Welfare State: The Battle over Public and Private Social Benefits in the United States (2002), Hacker described the coexistence of public and private social benefits that are unequally distributed to different populations in the USA. Some programmes are fully funded by public expenditure, and some programmes benefit the recipients through tax breaks and credit subsidies. There are also programmes that are promoted and regulated by the government while they are completely funded by private contributions – normally from higher earners. So China is not unusual in having multiple welfare schemes and programmes. However, the mechanism that leads to the coexistence of multiple programmes may not be the same for regimes with distinctive political institutions and paths. In the USA, Hacker identified the timing and sequence of policy development during which the public–private social benefits were constructed through political debates. Business interests played a central role in shaping the landscape of welfare distribution. In an authoritarian regime such as China, the central government has enjoyed far more power in the process of policymaking and resource allocation. Even in a typical policy experiment, which is assumed to be more flexible and open-ended than policies announced as one-off decisions, the central government would set the direction and guidance. Provincial and local governments would enjoy only a conditional and limited autonomy in localising and reinterpreting the details of the policies. So in China multiple social programmes coexisting is more a consequence of centralised policymaking. So, if the central government operates under constraint from a limited budget, how has it designed the differentiation of social rights and what were its main criteria? In the existing studies of China’s society, one important standard that was identified in the differing accessibility of resources was political status (Bray 2005; Lu and Perry 1997; Q. Zheng 2015). The difference of political status mainly resulted from the hukou system (Cheng and Selden 1994). People in China were born with the political status of being either rural or urban residents of their local area and were not supposed to move freely across regions. The rural/urban status, or hukou, was ‘political’ in the sense that it largely determined an individual’s access to initial public resources such as education (since schools are mostly localised resources), and local elections occurred at the village and district level. Thus it was constructed on residential variation. When individuals got a job, they entered into an occupation-based status, joining the staff of the government, staff of a public institution, staff of 30 Pension Policy and Governmentality in China an enterprise, working as a peasant, and so on (M.-k. Lee 2000; Xie, Lai, and Wu 2009). These status categories determined certain other public resources that an individual could access, such as pension plans and health care plans (Wang and Chai 2009). To give an example, as the economic reform proceeded, many rural residents left the countryside to search for temporary jobs in cities. However, even though they worked in the cities, their hukou status was still that of rural residents and they were ineligible for public goods such as education, housing, and pension benefits in the cities where they lived and worked (Y. Song 2014). In other words, people with different endowments did not enjoy the same public resources in China. Moreover, the unevenness of benefit allocations across different types of public resource schemes was distinctive. For instance, Cai and his colleagues (2018) identified the differentiation and polarised pension benefit allocation in China, using public transfer data. They showed that the public transfer in pension benefits was more concentrated in the top quartile of earners than was the case for those in education and health care. Those from the top quartile aged 75 received an annual pension of RMB 12,029 on average, and from the second quartile received around RMB 3,460 per person, while the elderly from the bottom quartile received less than RMB 189 a year (Cai, Feng, and Shen 2018). The variation of welfare benefits in China’s case did not merely apply to differentiation in one policy scheme; rather, it was systematically designed to provide different social groups with different programmes in order to maintain a balance of loyalty as well as avoid the risk of any hostile coalition. The formation process of such segregation entailed less political conflict than occurs in representative political systems, and showed more sense of ‘overview’ or ‘seeing the bigger picture’ from the ruler’s standpoint. It would seem quite in order if certain groups were privileged or sacrificed to improve ‘the general good’ or to reduce ‘the general cost’. The question that then arises is ‘who is to be sacrificed and who is to be kept safe in the era of transformations, and what tactics will the state use if public discontent is stirred up over the inequality?’ 2.2 Pension reforms in China: a de-synchronised story The allocation of welfare benefits is recognised as a useful tool for showing the generosity of the authority and buying loyalty from the public (Haggard and Kaufman 2008; Karshenas and Moghadam 2006; Rimlinger 1971). If budget constraints occur in a given period, one reasonable assumption regarding the allocation of welfare benefits is that the government will distribute the limited fiscal capacity to the social group that costs least per unit (that is, per person, or per group of people), and can maximise the gains for the state. But how to decide which group to favour and which to give up? There are two common assumptions about the objectives of the state: to maintain broad social stability, and to maintain active compliance, such as support or consent. The Who gets what and how: governance based on subpopulations 31 former is more basic, while the latter is more desirable, but they are in general not mutually exclusive. If the priority/objective for the government is ‘political stability’, then the resource will be allocated to those who enjoy the greatest negotiation and bargaining power. For maintaining compliance objective, the public’s consent is fundamental to the state’s legitimacy; it relies heavily on people’s perceptions that social justice and equality are being considered (at least shown superficially by state attempts to redistribute benefits) and the belief that the authority will meet the public’s expectations. Hence the central authority does its best to meet (or seem to meet) the public’s needs in order to get loyalty in return when the authority senses pressure or risks over legitimation. Under the assumption of consent manufacture and maintaining compliance, the state is more likely to redistribute public resources to social groups who are relatively disadvantaged. I develop my hypothesis about the government’s choice in welfare allocation on the basis of careful consideration regarding these two assumptions, and take into account the nature of authoritarian authority – which has an instinct to share power and benefits within its inner power circle. A strategic resource differentiation, therefore, is a reasonable hypothesis: the state will uphold favouritism towards core elites, but wield the weapon of redistributive equality when necessary, and only to a certain extent. Meanwhile, the government employs other tactics such as persuasion by propaganda, and gradual experiments to keep the general public opinion under control. To support this argument I draw evidence from institutional analysis and statistical comparison of China’s pension reforms. The institutional analysis focuses on variations in timing, direction, and content of reforms, so as to collectively present a whole image of strategic differentiation in welfare benefit allocation by the government. To precisely identify the different entitlements to benefit across pension schemes, I collected statistical data from the China Labour Statistical Yearbooks, Local Fiscal Statistical Yearbooks, National Statistical Bureau Dataset and other datasets (such as G. Zheng 2016),2 and compared variations including the eligible population, participating population, coverage rate, pension benefit (per person per year), incremental rate of pension benefit, and so on. In the 2010s there were five parallel pension plans,3 each tailored for one group of people – government employees, public institute employees, employees of enterprises, urban non-salaried residents, and rural residents. The fragmented pension scheme was compatible with the categories of political status, which originated from the socialist planning economy that locked the whole populace into a registered permanent category of residence (rural or urban) and a work unit (in an SOE, government post, or elsewhere) (Whyte 2012). From Table 2.1, we can identify the differences in accessing the public resources between the distinctive social groups. In the present pension system, only government employees enjoyed a full fiscal-funded retirement payment (Pozen 2013). Before the 2000s and 2010s reforms of the scheme, public institution employees used to enjoy a similar political status and comparable welfare 32 Pension Policy and Governmentality in China benefits to those of government employees. Afterwards they were now confined to a hybrid pension system in which individuals shared the responsibility for making pension contributions. Enterprise employees before the economic reform had their workplace unit and the government as accountable welfare providers. Nowadays, however, the pension plan for enterprise employees shares this responsibility between government, the enterprise, and the individual themselves. The pension plans for urban non-salaried residents and rural residents were similar; both became defined contribution pension insurance schemes that enjoyed a certain (slim) subsidy from the government compared to other types of employee (Wang, Béland, and Zhang 2014). Still, before the late 2000s, these groups had no systematic pension plan at all. As the main embodiment of state authority, government officials with formal contracts enjoy the most generous payments after they retire. There is solid evidence from previous studies that have demonstrated the welfare privileges of government-related work units compared to the enterprise units and urban residents compared to migrant workers and rural residents (Chan and Buckingham 2008; Chan and Zhang 1999; Selden and You 1997). Right after the Table 2.1: Comparison between different pension schemes in China (in 2015) Scheme recipients Scheme name Nature Funding source Government employees Pension Plan for Government Employees (PGE) Government-funded pension Government Public institution employees Pension Plan for Public Institution Employees (PPIE) Government-funded pensionsocial insurance Government government and individual shared Enterprise employees Pension Plan for Enterprise Employees (PEE) Enterprise-funded pensionsocial insurance Enterprise (government takes ultimate accountability)  government, enterprise, and individual shared Urban nonsalaried residents Pension Plan for Urban Nonsalaried Residents (PUR) Social insurance Government and individual shared Rural residents Pension Plan for Rural Residents (PRR) Social insurance Government and individual shared Source: author. Note: Arrows in the table indicate reforms of the pension schemes. Who gets what and how: governance based on subpopulations 33 foundation of the PRC, in 1955, a non-contributory pension scheme was issued for government officials. The employees were not required to contribute to the pension pool since the whole fund was subsidised by the state budget. The income replacement rate for retired government employees could reach 90% or 100%, according to the ‘Tentative Retirement Regulation for Government Employees’ issued by the State Council in 1956. The ‘Notice of Salary Reform for Government and Public Institution Employees’ issued in 1993 specified that the in-service salary for a government worker would consist of four parts: the duty wage, rank wage, basic wage, and seniority wage. A retired government employee could enjoy their full previous basic wage plus the seniority wage. The retiree also received some subsidy from the government in line with the duty wage and rank wage based on their working years: the longer the employee worked, the higher replacement rate that they got after retirement. Starting from 2006, when the ‘Civil Service Law’ was issued, the salary system for government employees changed to a two-tier structure: duty wage and rank wage, plus the corresponding allowances. Since the reform, a retired government employee can still enjoy a replacement rate above 80% of their previous income. Public institution (PI for convenience in the following discussions) employees, such as schoolteachers, doctors in public hospitals, employees of radio or TV stations, and so on, are also generally thought to be people who enjoy an ‘iron rice bowl’. The pension plan for PI employees was exactly the same as that for government employees before the 2006 salary reform: officially, the PI employees were subsidised by the state budget in full or in part (depending on the nature of the work units). The 2006 salary reform specified that the retirees of public institution no longer enjoyed the allowance and performance wage, though the pension was still much better than that of people in other occupations such as enterprise employees. The generous pension plan for government and public institution employees has caused great discontent in other social groups4 for quite a long time. Under huge pressure from society, the government claimed to have changed the dual track of one pension plan for government public institution employees and another for the remaining social groups in early 2008. The reform plan launched in 2008 (as a policy experiment) turned out to be a proposal that differentiated the pension benefits of PI employees from those of government employees: the old non-contributory pension scheme was changed into a defined contributory pension insurance scheme. Started in 2008 as a pilot policy, it was then implemented across the nation in 2014; all PI employees from then on had to add their own contribution to their future retirement fund.5 In Figure 2.1, I show the chronological changes in the pension scheme reforms for government employees and public institution employees. Before the economic reform in the 1980s, only state-owned enterprises (SOEs) were allowed to exist in the PRC. During that time their employees enjoyed high social and economic status, and had an enterprise-funded, 40 Pension Policy and Governmentality in China Table 2.2: Pension benefit for government employees and public institution employees Panel a. Data from statistical yearbooks Persons participated at the year end (millions) Revenue and expenses of pension fund (100 million Yuan) Total Workers Retiree Revenue Expense Balance 1999 7.62 6.42 1.19 93.2 61.8 89.3 2000 11.31 9.77 1.53 189.8 145.4 186.1 2001 12.78 10.68 2.09 253 204.4 233.2 2002 14.58 11.99 2.58 387.8 340.1 364.5 2003 16.25 13.22 3.03 470.6 405.9 441.7 2004 16.74 13.46 3.27 529.9 470.9 475.7 2005 17.72 14.09 3.62 601.6 545 534.3 2006 19.09 15.12 3.96 677.2 609.4 619.8 2007 19.02 14.92 4.09 823.6 811.3 633.2 2008 19.39 15.04 4.35 940.1 882 690 2009 19.83 15.24 4.59 1,070.3 1,007.8 751.8 2010 20.72 15.79 4.93 1,201.1 1,145 818.1 2011 21.08 15.95 5.13 1,409.9 1,339.3 888.5 2012 21.54 16.20 5.34 1,638 1,553.3 973.3 2013 21.68 16.12 5.56 1,831.7 1,729 1,076.9 2014 21.78 15.98 5.79 2,004.2 1,907.4 1,173.7 2015 22.37 16.32 6.05 2,727.7 2,671.8 1,229.6 Source: China Labour Statistical Yearbook (2016). public pension budget fund. Thus, the coverage rate in Table 2.2 can probably be better calculated by the number participants in public institution employee schemes divided by the number of public institution employees and government employees combined. Either way, it is clear that the pension benefit for government and public institution employees is generous and is largely covered by the state budget. The pension plan for enterprise employees (PEE) Enterprise employees experienced the most frequent modifications of their pensions during the whole reform period, and Table 2.3 presents the changes Who gets what and how: governance based on subpopulations 41 Panel b. Self-complied data of pension benefits Total employees (millions) Salary p/y (weighted) Benefit p/y (Yuan) Increase rate Coverage rate Replacement rate 1999 38.60 5,150 20% 2000 38.52 8,738 9,481 84% 29% 109% 2001 38.31 9,886 9,766 3% 33% 99% 2002 37.77 11,670 13,152 35% 39% 113% 2003 37.95 13,459 13,382 2% 43% 99% 2004 38.38 14,897 14,374 7% 44% 97% 2005 38.89 16,884 15,042 5% 46% 89% 2006 39.45 19,317 15,357 2% 48% 80% 2007 40.04 21,854 19,802 29% 48% 91% 2008 40.71 26,645 20,247 2% 48% 76% 2009 41.13 30,932 21,957 8% 48% 71% 2010 34,248 23,210 6% 68% 2011 26,106 13% 2012 29,049 11% 2013 31,085 7% 2014 32,897 6% 2015 44,125 34%   Sources: Local Fiscal Statistical Yearbooks; Author. Note: The data shown in the table are all rounded up, therefore may have slight differences with the read numbers. in their pension benefit and rate of increase. It shows the combined statistical data of pension benefits for employees of SOEs and other types of employment (such as self-employed persons) from the China Labour Statistical Yearbook. In addition, Figure 2.2 shows the enterprise employees’ pension coverage rate, benefit per person per year, and replacement rate data from the report by Zheng (2016). Compared to government employees and public institution employees, enterprise employees enjoyed far lower pension benefits, considered both in terms of their raw benefit per person per year and the rate of increase. The average benefit per person per year for government employees and public institution employees in 2015 was above 44,000 yuan, while for average enterprise employees it was around 27,000 yuan (or 28,236, in Zheng’s calculations) even Table 2.2: (Continued) 42 Pension Policy and Governmentality in China Table 2.3: Pension benefit for enterprise employees and others Persons participating at the year end (millions) Revenue and expenses of pension fund (100 million Yuan) Total Workers Retiree Revenue Expense Balance Benefit p/y (Yuan) Increase rate 1989 57.10 48.16 8.93 146.7 118.8 68 1,330.1  1990 61.66 52.00 9.65 178.8 149.3 97.9 1,547 16% 1991 67.40 56.53 10.86 215.7 173.1 144.1 1,592.8 3% 1992 94.56 77.74 16.81 365.8 321.9 220.6 1,914.4 20% 1993 98.47 80.08 18.39 503.5 470.6 258.6 2,558.6 34% 1994 105.73 84.94 20.79 707.4 661.1 304.8 3,179.2 24% 1995 109.79 87.37 22.41 950.1 847.6 429.8 3,781.9 19% 1996 111.16 87.58 23.58 1,171.8 1,031.9 578.6 4,375.5 16% 1997 112.03 86.70 25.33 1,337.9 1,251.3 682.8 4,940.1 13% 1998 112.03 84.75 27.27 1,459 1,511.6 587.8 5,542.6 12% 1999 117.22 88.59 28.63 1,871.9 1,863.1 644.2 6,505.9 17% 2000 124.86 94.69 30.16 2,088.3 1,970 761 6,530.9 0% 2001 129.04 97.33 31.71 2,235.1 2,116.5 818.6 6,674 2% 2002 132.78 99.29 33.49 2,783.6 2,502.8 1,243.5 7,472.8 12% 2003 138.81 103.24 35.56 3,209.4 2,716.2 1,764.8 7,636.4 2% 2004 146.78 109.03 37.75 3,728.5 3,031.2 2,499.3 8,029.7 5% 2005 157.15 117.10 40.05 4,491.7 3,495.3 3,506.7 8,726.9 9% 2006 168.56 126.18 42.38 5,632.5 4,287.3 4,869.1 10,114.9 16% 2007 182.34 136.90 45.44 7,010.6 5,153.6 6,758.2 11,341.5 12% 2008 199.51 150.83 48.68 8,800.1 6,507.6 9,241 13,368.1 18% 2009 215.67 162.19 53.48 10,420.6 7,886.6 11,774.3 14,746.8 10% 2010 236.34 178.22 58.11 12,218.4 9,409.9 14,547.2 16,191.6 10% 2011 262.84 199.70 63.14 15,484.8 11,425.7 18,608.1 18,095.8 12% 2012 282.71 213.60 69.10 18,363 14,008.5 22,968 20,270.1 12% 2013 300.49 225.64 74.84 20,848.7 16,741.5 27,192.3 22,367.3 10% 2014 319.45 239.32 80.13 23,305.4 19,847.2 30,626.3 24,766.9 11% 2015 331.23 245.86 85.36 26,613.2 23,140.9 34,115.2 27,108.2 10% Data sources: China Labour Statistical Yearbooks; Author. Who gets what and how: governance based on subpopulations 43 though their on-duty salary was around 56,000 yuan before retirement. In other words, enterprise employees would have experienced a steeper drop in living standards after retirement. Moreover, although the coverage rate increased after 1997, the replacement rate went in the opposite direction: it dropped from 80% in the late 1990s to 50% in the late 2000s, as Figure 2.2 shows. The pension plan for urban non-salaried residents (PUR) and rural residents (PRR) These two pension plans both started quite late on, and involved very limited payments. Table 2.4 shows the pension benefit change for rural residents from Zheng’s report in 2016, and demonstrates that their pension scheme had a far lower coverage rate and a low replacement rate. The average (mean) pension benefit per person per year for rural residents was about 1,000 yuan, which is far less than their urban counterparts received (e.g. enterprise employees got 20,270 yuan in 2012 on average). In terms of the value of benefits, the replacement rate in the rural residents’ pension scheme was around 13% in 2015, while urban enterprise employees enjoyed a replacement rate of 50% (2015) and government/public institution employees enjoyed 67% (2010, weighted). To better understand the strategic aspects of benefits differentiation in China, Table 2.5 shows the changing population numbers of the different social groups from 1987 to 2009.9 Though one cannot point to solid evidence Figure 2.2: Pension benefit for enterprise employees (averaged), 1995–2015 Data source: Zheng (2016). 44 Pension Policy and Governmentality in China supporting the hypothesis, a reasonable strategy for a government wanting to reduce its burdens would be to push the largest working population – enterprise employees – towards the open market. And, when a government needs to show generosity and appease discontent over inequality, it could give a modicum of ‘alms’ to a large but less advantaged population – rural residents – in exchange for more loyalty at relatively low cost. According to my preliminary interviews in summer 2016, many rural residents, especially the older ones, were very content with receiving any subsidy from the government. This was mainly because they had enjoyed few benefits from the state budget beforehand and most of the old people were being cared for by people from their families or the local community. Any cash subsidy from the government seemed like a bonus to them, especially in view of their relatively low income compared to urban employees. Conclusions The empirical evidence suggests that the strategy of differentiation in the government’s welfare allocation has been to sacrifice the social benefits for employees of state-owned enterprises (SOEs) first, and later on the employees of public institutions, so as to reduce the state’s fiscal burden. At the same time, modest or slender welfare benefits were provided for social groups that could be bought off with minimal benefits. During the process, government officials (including the core elites of governance power) consistently enjoyed the most generous social benefits. The various policy shifts of different pension schemes strongly support these propositions, especially the retrenchment policy reform for SOE employees; the expanded coverage for rural residents and migrant workers from rural areas; and the reform of the contribution ratio reform in the pension scheme for public institution employees. The segmented resource allocation in China’s social welfare reform has favoured the core elites, while it distributed limited fiscal capacity to the social groups that cost least per person. Table 2.4: Pension benefit for rural residents Expense (100 million Yuan) Recipients (millions) Benefit p/y (Yuan) Income per year (Yuan) Replacement rate 2011 598 87.60 683 6,977.3 9.8% 2012 1,150 130.75 880 7,916.6 11.1% 2013 1,348 137.68 979 8,895.9 11% 2014 1,571 143.13 1,098 9,892 11.1% 2015 2,117 148 1,430 10,772 13.3% Data source: Zheng (2016). Who gets what and how: governance based on subpopulations 45 Table 2.5: Raw population numbers by type of pension scheme recipient Government employees (millions) Public institution employees (millions) Enterprise employees (millions) Permanent rural residents (billions) 1987 8.05 20.32 103.76 8.16 1988 8.43 21.01 106.62 8.23 1989 8.85 21.75 105.64 8.41 1990 9.13 23.01 106.97 8.41 1991 8.46 1992 8.49 1993 9.86 24.26 114.36 8.53 1994 10.43 26.53 115.60 8.56 1995 10.22 27.36 115.42 8.59 1996 10.50 28.23 113.47 8.50 1997 10.62 29.05 110.68 8.41 1998 8.31 1999 10.68 27.92 82.69 8.20 2000 10.73 27.78 77.59 8.08 2001 10.70 27.60 73.34 7.95 2002 10.54 27.23 72.07 7.82 2003 10.71 27.24 71.72 7.68 2004 10.92 27.46 72.60 7.57 2005 10.96 27.93 75.14 7.45 2006 11.11 28.33 77.68 7.31 2007 11.30 28.74 80.19 7.14 2008 11.56 29.14 81.20 7.03 2009 11.83 29.29 84.12 6.89 Data sources: China Labour Statistical Yearbooks; National Statistical Bureau; and author. Notes 1 China’s ‘hukou’ system is a family registration programme that serves as a domestic passport, regulating population distribution and rural-to-urban migration. For decades it was a tool for social and geographic control, 46 Pension Policy and Governmentality in China enforcing an apartheid structure that denied farmers the rights and benefits that were enjoyed by urban residents. 2 A more detailed discussion of the codebook is available in the appendix. 3 In China’s case, the full state-funded pension is literally called the ‘retirement allowance’, and a pension that requires individual contributions is called a ‘pension/social insurance’. 4 People’s Daily Report, 2012, ‘Unfair dual track of old-age care: [dramatically] different treatment for people with different status’. 5 China News, 2014, ‘Track convergence of pension scheme: 30 million public institution employees will start paying into social insurance from July 1st’. https://perma.cc/2MCS-FJ4L [accessed 2018-11-12]. 6 This pension plan was later applied to employees of all enterprises, regardless of ownership. 7 More information on this policy experimentation is available in Chapter 4. 8 Owing to a highly scattered and incomplete condition of welfare yearbooks, I tried my best to collect all available data from various sources. There are certain places where the data do not match very well since the statistical bureau stopped publishing many indexes (such as the population and pension benefit for government employees after 2009) and often changed the grouping methods. Thus, I appended notes to each table to clarify the sources and calculations. More information on the codebook is available in the appendix. 9 The government stopped publishing the employee number and related conditions of government units and public institution units from 2009. References Bray, D. (2005). Social Space and Governance in Urban China: The Danwei System from Origins to Reform. Stanford University Press. Cai, Y.; Feng, W.; and Shen, K. (2018). ‘Fiscal implications of population aging and social sector expenditure in China’. Population and Development Review, vol. 44, issue 4, 811–831. https://doi.org/10.1111/padr.12206 Chan, K. W.; and Buckingham, W. (2008). ‘Is China abolishing the hukou system?’ The China Quarterly, vol. 195, 582–606. https://doi.org/10.1017 /s0305741008000787 Chan, K. W.; and Zhang, L. (1999). ‘The hukou system and rural-urban migration in China: Processes and changes’. The China Quarterly, vol. 160, 818–855. https://doi.org/10.1017/s0305741000001351 Chao, C.-m.; and Dickson, B. (2003). Remaking the Chinese State: Strategies, Society, and Security. Routledge. Cheng, T.; and Selden, M. (1994). ‘The origins and social consequences of China’s hukou system’. The China Quarterly, vol. 139, 644–668. https://doi .org/10.1017/s0305741000043083 Who gets what and how: governance based on subpopulations 47 Frazier, M. W. (2010). Socialist Insecurity: Pensions and the Politics of Uneven Development in China. USA: Cornell University Press. Gu, E. X. (1999). ‘From permanent employment to massive lay-offs: the political economy of “transitional unemployment” in urban China (1993–8)’. Economy and Society, vol. 28, issue 2, 281–299. https://doi.org /10.1080/03085149900000006 Gu, E. X. (2001). ‘Dismantling the Chinese mini-welfare state?: Marketization and the politics of institutional transformation, 1979–1999’. Communist and Post-Communist Studies, vol. 34, issue 1, 91–111. https://doi.org/10.1016 /s0967-067x(00)00025-8 Hacker, J. S. (2002). The Divided Welfare State: The Battle over Public and Private Social Benefits in the United States. Cambridge University Press. https:// doi.org/10.1017%2Fcbo9780511817298 Haggard, S.; and Kaufman, R. R. (2008). Development, Democracy, and Welfare States: Latin America, East Asia, and Eastern Europe. Princeton University Press. Huang, X. (2014). Social Protection under Authoritarianism: Politics and Policy of Social Health Insurance in China. Columbia University. Karshenas, M.; and Moghadam, V. M. (2006). Social Policy in the Middle East: Economic, Political and Gender Dynamics. UK: Palgrave Macmillan. Lee, M.-k. (2000). Chinese Occupational Welfare in Market Transition. Springer. https://doi.org/10.1057/9780333982549 Lu, X.; and Perry, E. J. (1997). Danwei: The Changing Chinese Workplace in Historical and Comparative Perspective. Me Sharpe. Pozen, R. C. (2013). Tackling the Chinese Pension System. USA: Paulson Institute. Ratigan, K. (2017). ‘Disaggregating the developing welfare state: Provincial social policy regimes in China’. World Development, vol. 98, 467–484. https://doi.org/10.1016/j.worlddev.2017.05.010 Riedmüller, B. (2008). Private Pensions Versus Social Inclusion?: Non-state Provision for Citizens at Risk in Europe. Edward Elgar Publishing. Rimlinger, G. V. (1971). Welfare Policy and Industrialization in Europe, America and Russia. USA: Wiley. Selden, M.; and You, L. (1997). ‘The reform of social welfare in China’. World Development, vol. 25, issue 10, 1657–1668. https://doi.org/10.1016/s0305-750x (97)00055-7 Song, Y. (2014). ‘What should economists know about the current Chinese hukou system?’ China Economic Review, vol. 29, 200–212. https://doi.org /10.1016/j.chieco.2014.04.012 State Council. (1997). Decision on Establishing a Unified System of Basic Pension Insurance for Enterprise Employees (State Council [1997] No.26). Retrieved from http://www.gov.cn/ztzl/nmg/content_412509.htm. State Council. (2000b). The Notice of Issuing the Pilot Program of Urban Social Security System (State Council [2000] No.42). Retrieved from http://www .gov.cn/xxgk/pub/govpublic/mrlm/201011/t20101112_62507.html. 48 Pension Policy and Governmentality in China State Council. (2005). Decision on Perfecting Basic System of Pension Insurance for Enterprise Employees (State Council [2005] No.38). Retrieved from http://www.gov.cn/zwgk/2005-12/14/content_127311.htm. State Council. (2009). New Rural Old-age Social Insurance Plan (State Council [2009] No. 32) Retrieved from http://www.gov.cn/zwgk/2009-09/04 /content_1409216.htm. Wang, D.; and Chai, Y. (2009). ‘The jobs–housing relationship and commuting in Beijing, China: the legacy of danwei’. Journal of Transport Geography, vol. 17, issue 1, 30–38. https://doi.org/10.1016/j.jtrangeo.2008.04.005 Wang, L.; Béland, D.; and Zhang, S. (2014). ‘Pension fairness in China’. China Economic Review, vol. 28, 25–36. https://doi.org/10.1016/j.chieco .2013.11.003 Whyte, M. K. (2012). ‘China’s post-socialist inequality’. Current History, vol. 111, issue 746, 229–234. https://doi.org/10.1525/curh.2012.111.746.229. OA: https://online.ucpress.edu/currenthistory/article-pdf/111/746/229/392449 /curh_111_746_229.pdf Xie, Y.; Lai, Q.; and Wu, X. (2009). ‘Danwei and social inequality in contemporary urban China’. Research in the Sociology of Work, vol. 19, 283. https://dx .doi.org/10.1108%2FS0277-2833(2009)0000019013. OA: https://www.ncbi .nlm.nih.gov/pmc/articles/PMC2828673 Zheng, G. (2016). Evaluation of China’s Social Protection Policies. Retrieved from https://www.euchinasprp.eu/images/documents/Component1Cn/2017 -assessment-eport/EvalSSCn.pdf Zheng, Q. (2015). Identity Acceptance and Production Politics: Research on Labour Relations in Changes of State-Owned Enterprises. China: Beijing Book Co. CHAPTER 3 Who deserves benefits and why – constructing fairness, pension expectations, and subjectivity What kind of truth and knowledge about pension benefits is produced and promoted by the Chinese state when it tries to persuade the public about the many policy experiments and reforms of recent times? The question is complicated by an underlying epistemological assumption here, that what we know about society is never entirely consistent between individuals or uniform over time. Sociologists have long discussed the process of ‘socialisation’ during which every individual gradually internalises norms and ideologies about society. The state is among the major powers that provide inputs for the socialisation process. It is equipped with various tools to intervene in political socialisation, such as education, the public media, and legal systems (Glasberg and Shannon 2010). Using them the state can produce certain truths and knowledge that shape the way that people understand their society and where they can expect to fit within the social ‘blueprint’. While accomplishing considerable socio-economic transitions, the state cannot simply cut benefits to a certain group of people in order to minimise the budget or ease its burden. The government needs to keep a minimal degree of public consent for its own survival and stability, employing additional tactics to help promote major institutional changes that hurt someone’s existing interests (as almost all pension reforms will do, and certainly in China). Institutionalism generally supposes that the state has a sophisticated understanding of its subordinate population (though it is not necessarily 100% correct), and uses this knowledge to try to figure out the optimal or proper ways to manage changes in public compliance. The state, therefore, can decide to manipulate people’s emotions through shaping, sculpting, and mobilising the options, expectations, and daily lives of individuals. Government and governance not only signal the How to cite this book chapter: Wang, Yan. 2022. Pension Policy and Governmentality in China: Manufacturing Public Compliance. London: LSE Press, pp. 49–90. DOI: https://doi.org/10.31389/lsepress.ppc.d. License: CC BY 56 Pension Policy and Governmentality in China Topic 25 ‘Rural Migrants’ top words: Highest probability: peasant, rural area, worker, urban-rural, agriculture, city FREX: peasant, lose land, urban-rural, land, migrant, city Lift: Dujiangyan, deep water, whole scale, Pujiang, Xinyang Score: peasant, rural area, urban-rural, lose land, agriculture, worker, rural Topic 2 ‘Commercial Insurance’ top words: Highest probability: insurance, company, invest, bank, market, China FREX: client, life, annuity, life insurance, company, business Lift: be clever, actuary, collusion Score: insurance, life insurance, company, life, annuity, client, bank Notes: Some Chinese words have several meanings or act as several parts of speech (PoS); I include word extensions inside brackets. ‘Dujiangyan’, ‘Pujiang’, and ‘Xinyang’ are all the names of places. Table 3.1: (Continued) Topic 29 mainly addresses pension fees. There are some other clusters or topics that do not seem directly related to the pension reform. For instance, Topic 30 (Letter/Visits) addresses the theme of appeals, issues and the government. Topic 15 (State Budget) discusses fiscal problems, the budget, and expenditure. Topic 7 (Community Care) covers key words such as people, difficulty, life, disability, warm, street-level administration, and so on. To give specific examples of relevant discourse (all of which I have translated), I selected some excerpts that best describe the feature of certain topics, based on my understanding of the full corpus (with a full reading of the whole sample text) and the instruction of the key topics (with unsupervised models). The isolated Commercial Insurance (Topic 2) in the figure addresses the promotion of purchased rights to cover social risks. In relevant documents here the social risks for each individual are treated as capital, are calculable, and are avoidable once complete preparations have been made. A selected letter to the newspaper editor from a former worker read: Dear Editor: Both my partner and I have retained our posts in our original work unit while [our] salary payment has been suspended. We have started a small private business since our ‘iron rice bowl’ has disappeared. Therefore, we have quite a strong sense of (potential) risks and want to invest in some basic insurance to avoid future problems. I know there are several insurance companies and many kinds of insurance. Can you tell me what I should pay attention to when choosing insurance? (‘What we should pay attention to when choosing personal insurance’, Reader’s letter, 1997-05-28) The editor replied with a long discussion of the difference between various types of commercial insurance, addressed from the angle that people who make plans for their own and care about preparing for risks in life (especially those who have been laid off due to the SOE reform) should be warmly Who deserves benefits and why – constructing fairness 57 encouraged. In another document, ‘commercial insurance’ is identified as the stabiliser of society: The social attributes of commercial insurance make it capable of adding up to and supplementing the government’s social insurance scheme. (Its existence) is good for reducing people’s anxieties and it functions as ‘the stabilizer of the society’. (‘Bring out the role of commercial insurance as “the stabilizer of society”’, Zeng Yujin, 2004-09-18) Figure 3.1: Correlations of topics Notes: Each topic in the figure is labelled with one or two key words for easier identification, but note that each topic contains far more complicated information than these labels. 58 Pension Policy and Governmentality in China Another special topic not directly related to either cluster 4–20–23 or cluster 9–19–29 but worth highlighting here is Topic 16 (Birth Control). This refers to the ‘one-child policy’ and birth control, correlating with Topic 8, which covers the keywords rural officials, village, and towns. It also correlates distantly with Topic 5 (Old-Age Care), which addresses the social group of elderly people through Topic 7 (Community Care). To see why these topics show up here, I move on in the next section to discuss how social welfare policies were used as instruments of other contemporary major reforms. 3.2 Pension reforms as instruments of broader socio-economic reforms Policies and their promotions are never isolated or arbitrary; they are proposed and framed in a way that is consistent with their socio-economic conditions. In particular, welfare policy is not just concerned with the allocation of social benefits but also extends to employment policy, tax policy, and demographic policy. After sorting out the content and correlations of the core topics, it is important to connect the topics and their longitudinal variations with the major socio-economic reforms initiated by the government. The interaction or dialogue between welfare policies and other policies at the same time shows the comprehensive and sophisticated design of governance being used. It also shows how the details of a welfare policy are shaped by their social and economic circumstances, looking at eligibility standards and subsidy methods. My chief objects in the text analysis are urban enterprise employees (especially SOE employees), and rural residents, key groups for reasons set out in Section 3.1. Thus, in this section I explore the way that pension reforms interact with the other reforms imposed on these two social groups. To situate the text analysis, it is worth recapping a little on Figure 2.1 (showing the chronological schedule of pension reforms). The ‘Decisions on Economic Reform’ were issued in 1984 and the ‘Enterprise Bankruptcy Law’ was promulgated in 1986. In 1993, the ‘Decisions on Constructing [a] Socialist Market Economy’ accelerated the process of economic reform following Deng’s visit to southern China. SOEs formed the backbone of China’s economy during the central planning era, so their transformation was the most prominent of the changes in China’s enterprise system made in tandem with other institutional and policy reforms (Garnaut, Song, and Fang 2018). The marketisation of the SOEs was followed by the laying-off of millions of workers. The official total of redundancies increased from 3 million in 1993 to 17.24 million in 1998 (Cai 2002; Jefferson and Rawski 1994). The numbers of laid-off workers brought challenges to social stability and pressure towards pension reform. According to the OECD’s report, the unemployment rate in urban areas increased from 7.6% to 12.7% between 1995 and 2001 (OECD 2005). Thus, the government needed to persuade people of the need for the reform and tell them what the Who deserves benefits and why – constructing fairness 59 employees of enterprises (SOEs in particular) should expect in future in social benefits from the government. In addition to the reforms of the state-owned enterprises, the emergence and exponential growth of private enterprises following the economic reform also brought challenges to the existing pension system for enterprise employees. In 2003, private firms (about 3 million domestic private enterprises and 24 million sole proprietorships) comprised 59% of the economy (OECD 2005). Their employees were not covered in the traditional socialist pension system, and the central government was unlikely to provide ‘cradle to grave’ social protection, as in the old system. Therefore, the authority needed to push for the acceptance of a new welfare system for employees of enterprises (whatever their ownership), one where individuals shared the responsibility for pensions. Figure 3.2 presents the change of expected topic proportions in the text analysis for the themes of economic reform (Topic 20), SOE reform (Topic 4), pension reform (Topic 19), and laid-off workers (Topic 9). I also highlight the critical events during the reform. The green dashed vertical lines here show the dates of critical events for the economic reforms – specifically, ‘Decisions on Economic Reform’ (1984), ‘Enterprise Bankruptcy Law’ (1986), and ‘Decisions on Constructing [a] Socialist Market Economy’ (1993). The blue dashed vertical lines show the dates of critical events for the pension reforms of enterprise employees – namely, ‘Decision on Pension Insurance Reform’ (1991), ‘Basic Pension Insurance Scheme’ (1997), the pilot programme of fully funded individual accounts for the reform of pensions for enterprise employees (issued in 2000), and ‘Improving the Basic Pension System’ (2005). The first chart in Figure 3.2 shows the changing trend of topics to do with both the SOE reform, marketisation and the reform of pensions for enterprise employees. The incidence increased in the 1980s and reached a high point in the early 1990s. By contrast, the topic of macro economy growth (shown in the second chart) did not show strong fluctuations in its topic proportions. A reasonable understanding is that the documents are mainly about pensions, and the issues of macro economy growth have relatively weak presence here. The topic of institutional change in enterprise pension reform (shown in the third chart) has a similar pattern of responding to the key events of pension reforms. In sharp contrast, the issue of laid-off workers (shown in the fourth chart of Figure 3.2) was not highlighted until 1997, when the lines on the chart shot up, at a late stage of the marketisation and the pension reform for enterprise employees after 1993. How does the content of these topics dialogue with and echo each other? In the following section, I consider some typical examples of each topic, then discuss the governance logic behind the discourse usage. The economic reform itself was described by the state as a desired future for everyone in Chinese society. Topic 20 (Economic Reform) refers to the macro of economic growth and emphasised the urgency of the ‘overall situation’ in 60 Pension Policy and Governmentality in China Figure 3.2: Topic proportions by year: economic reform and pension reform for enterprise employees Who deserves benefits and why – constructing fairness 61 Figure 3.2: (Continued) 62 Pension Policy and Governmentality in China current society in terms of development, social stability, and long-term vision. For instance, a 2003 article argued: Dealing with the relations between reform, development, and stability correctly, and resolving the important issues during the process, are crucial for the progress of socialism with Chinese characteristics, in order to attain the goal of building a society prosperous in every sense, for the long-term advancement of all kinds of socio-economic affairs. (‘Making the effort to resolve the crucial questions of the overall situation’, Shi Zhongxin, 2003-09-04) As a main component of economic reform, the reform of the SOEs was described as the most efficient and effective way of achieving a ‘prosperous society’. From the texts within Topic 4 (SOE Reform and Marketisation), the main points of promoting the efficiency of the reformed SOEs was tied up with economic performance and statistics. For instance, in a passage of the text introducing an example model of SOE reform, the discourse read: When we liberate the small SOEs we cannot use a unified mode of ‘one size fits all’, or just leave them on their own. Here is a summary of the Xinle city’s practices for liberating small SOEs. The reform started in the second half of 1992, when the debt-asset ratio of eight branches of SOEs (including industry, light industry, and retailing, etc.) was close to 100%, and the scale of loss was about 63%. The nation’s fiscal capacity was almost exhausted. Three years after the liberalisation of the small SOEs, the profits and taxes from the SOEs had increased dramatically. [This precedes a long statistical justification] (‘Using multiple strategies, coordinating the government and enterprises, Xinle’s reform for small SOEs is alive and stable’, Wang Qingxian, 1996-07-18) Along with the SOE reform, an important task of the pension reform was to help the state and enterprises ease the burden of funding employees’ pensions. As the state started to reform for enterprise employees in the early 1990s, official discourse warmly commended the necessity and importance of this reform: The decision from the State Council pointed out that the reform of the enterprise employees’ pension insurance system was an important move protecting the lives of retirees and maintaining social stability. It’s also very important for reducing the burden on the state and on enterprises, advancing the reform of the economic system, and guiding domestic consumption in a rational way. This work [of reform] is highly policycentred and will affect many aspects of socio-political life. All levels of government need to enforce leadership, follow the spirit of this decision, design concrete implementation plans based on the local situation and promote the reform in positive and steady ways. (‘The State Council Who deserves benefits and why – constructing fairness 63 made the decision to reform the pension insurance system for enterprise employees; [we will] gradually construct a system which combines basic pension insurance, enterprise-supplemented insurance and employees’ personal savings’, 1991-10-10) And, in a summary text about welfare reform, the difference between the desired new hybrid welfare system (with shared responsibilities) and the unwanted old system (social protection based on the work unit) was stated as a welcome change that fitted the overall trend of the social and economic reforms: The old system, unit-based, fully covered, exclusive and inefficient, in which the state takes full responsibility, has been replaced with a new multilayered system of shared responsibility, socialised basic protection and inclusive schemes. A new welfare system which fits the requirements of a socialist market economy is now built. (‘The new opportunity for deepening the social welfare reform brought about by the advent of WTO membership’, Zhang Yitian, 2002-03-23) The SOE reform left many employees either experiencing being laid off or facing the risk of becoming unemployed. High levels of unemployment can pose many risks to social stability. In response, the central government proposed a ‘social security system with Chinese characteristics’ based on ‘three security lines’ (2002): • a basic guaranteed living system for those no longer working in an SOE, intended to safeguard basic living standards, pay social insurance contributions, and encourage re-employment; • an unemployment insurance system, which provides unemployment benefits and actively encourages re-employment; and • guaranteed minimum living standards for all urban residents whose family income per capita is lower than the prescribed level (Chan and Buckingham 2008). In the official discourse addressed to laid-off workers, the government drew attention not only to the importance and correctness of the reform but also to the point that the government would never abandon this group of people, but still encouraged them to make a new life, if they could. [In a visit to Liaoning] Zhu Rongji [then premier] pointed out that the policies of ‘encouraging amalgamation, regulating bankruptcy, redirecting laid-off workers, cutting jobs and boosting efficiency, [and] implementing a re-employment scheme’, and of constructing basic protection for the SOEs’ laid-off workers, pension insurance and unemployment insurance, secure minimum living standards, and so on, have shown themselves to be absolutely correct. These policies play an important 64 Pension Policy and Governmentality in China part in deepening the reform of the SOEs, advancing the adjustment of the institutions, and maintaining social stability. Hence, they should be carried out comprehensively. The path of pension reform in Liaoning is absolutely correct and we should persevere unwaveringly. (‘Zhu Rongji emphasised during his visit to Liaoning, that [we] need to pursue to the end the path of pension reform and use our best efforts to improve the state of employment and re-employment’, Liu Siyang, Meng Huan, 2002-07-25) All these discourses refer either to economic efficiency, the national interest, and well-being for everyone, or simply to the competence of the government, serving the intention of carrying out economic reform while preserving social stability. Another good example of social policy as an instrument of reform is the changes made in the rural pension plan, which was not as well designed or widely implemented as the pension reform for employees of urban enterprises. As the discussion of differentiation in Chapter 2 showed, this plan started as a policy piloted in scattered local regions in the late 1980s and early 1990s, initiated and promoted by local governments. The central government at this stage only acted as cheerleader. The text analysis identified that Topic 16 (Birth Control) was related to the rural pension scheme. To see why this was, consider a typical example of the text promoting the rural pension policy but connected to the one-child policy: The priority of controlling birth is in rural areas, which is also one of the most difficult work [in one-child policy implementation]. But Zhejiang’s practice tells us that the difficulties can be overcome. Recently we investigated 100 households in 50 villages from 10 counties in the company of officials from the Zhejiang Birth Control Association. Our experience made us feel that their work here has made renewed progress … The way that this progress was made benefited from several skills. First of all, they promoted the birth control policy in every household. The association members paid many visits to households, bringing educational materials specifically designed to combat the difficulties in each of them, combining reasons of different kinds and clarifying misunderstandings with detailed information. For instance, the chair of the birth control association from Hengdu village, Anji County, visited all the villagers and helped them to calculate possible future expenses like this: parents who already have a girl must, if they have a second child who is a boy, pay at least 150,000 yuan when he marries and at least 200,000 yuan for his education. But if the household participates in the pension insurance programme they need to pay less than 30,000 yuan in order to receive a pension of 800 yuan when they get old [60 for males, 55 for females]. This calculation totally persuaded the villagers. Many couples of childbearing age said that, even if their first child was a Who deserves benefits and why – constructing fairness 65 girl, they wouldn’t try to have a second child. (‘Birth control birth while enjoying a pension is a good idea’, Zhao Xiangru, 1988-12-17) Thus, pension benefits are being presented here as an exchange for the willingness of rural residents to practise birth control. Another example reads as follows: Currently more than 800 counties have started to pilot the pension insurance system, and about 7 million [people] are participating. Peasants are thrilled and praise the pension insurance scheme, saying ‘The party and the government have given us a dutiful son’, ‘[it] is a nursing home without fences’. (‘Our social welfare is going ahead vigorously’, Chen Hong, 1992-10-06) The one-child policy was introduced in 1979, modified in the mid-1980s to relax regulations in rural areas (Scharping 2013), and finally replaced by a ‘second child’ policy that encouraged couples to have more children at the end of 2015. In order to promote the policy, the government from the 1980s allocated five yuan per month to families with only one child (the amount gradually increasing in step with the economic development). It also gave these households a ‘one-child glory certificate’, with which the parents could enjoy longer maternity leave, subsidised social insurance, and so on. The one-child limit was most strictly enforced in densely populated urban areas, where the government found it easier to grant benefits and impose punishments (such as heavy fines or mandatory contraception) through the citizens’ work units. The situation in rural areas was more difficult to manage. For rural residents, one extra child meant a bigger labour force on a smallholding or farm, and the government had no way of enforcing punishment on all who ‘deserved’ it. Thus, rural residents were now allowed to have two children, especially those whose first child was a girl (so long as they waited for some years after her birth). Since it was difficult to enforce birth control among rural citizens, the government had to use other means such as subsidies and praise. Social policy hereby was effectively used to encourage birth control and safeguard gender balance. (China has a long-term problem in families preferring to have boys rather than girls, causing significantly more men than women in the population.) Figure 3.3 presents the chronological changes in the expected topic proportions for Topic 16 (Birth Control), which was high from the early 1980s, dropped in the mid-1980s and increased again in the early 1990s. However, since the 1990s the overall trend has declined. The interactions between pension reforms and their parallel social and economic policies, and the ways in which the pension reforms were used as instruments for other reforms, demonstrate that the state authority’s governance in China is dynamic, reflective, and experimental. The increased proportion of references to pension reform for enterprise employees when the policy was being introduced, plus the later steep rise in attention to the laid-off workers, 72 Pension Policy and Governmentality in China tion reaches the point when industry becomes the main engine, we need to coordinate the industry–agriculture relationship by having the state increase its support and protection of agriculture in addition to the (existing) market intervention, so we can convert from the model of agriculture-feeding-industry to that of industry-repaying-agriculture. Experience from other countries shows that when the process of industrialisation and urbanisation accelerates, the whole economy attains the phase of industry-repaying-agriculture. If we strengthen and repay agriculture, the national economy will achieve industrialisation and modernisation in a healthy way. Otherwise, if we still exploit and ignore the agriculture, it will lead to agriculture lagging behind, growing inequality and a wider urban–rural gap. Moreover, it will sharpen social conflict and lead to social instability and retrogression. (‘Industry repays agriculture and the urban supports the rural: how to support the rural areas and agriculture in the new scenario’, Han Jun, 2005-11-18) When rural migrants flooded into the cities on a large scale in the urbanisation process, the question of pensions for them and, more generally, social welfare for them, became a knotty problem. Owing to the existence of the hukou system, rural residents had only restricted or even no use of schools, hospitals, and other public facilities in urban areas – with many depending on private payment solutions instead. In the 1990s and 2000s, when the number of migrant workers surged, the absence of a proper welfare scheme for them led to many social unrests. However, if the migrant workers had been allowed to enjoy the same welfare system as the urban residents, the competition for resources would have been fierce and that would have caused discontent from urban residents. Thus, the government tried to persuade society, mainly the urban residents, that there were good reasons for these migrant workers to receive some more social benefits and fiscal support. As we might expect, one of the most obvious and frequently used reasons was that the migrant workers contributed to the construction of the cities: Among the manufacturing, construction, mining and service industries (such as domestic service, catering and so on), rural migrants constitute more than half of the employees. As one important part of the labour force, they deserve fair treatment. This is not only a necessary action defending the migrant worker’s legitimate rights, but also directly relates to a defence of social justice and fairness … Compared to urban workers, rural migrant workers receive low wages (which sometimes can’t even be paid on time) and take on dirty and dangerous jobs, while enjoying none of the social welfare benefits [of urban workers]. They contribute to the growth of cities, yet they enjoy none of the convenience enjoyed by urban employees in their daily lives and suffer greatly in such areas Who deserves benefits and why – constructing fairness 73 as children’s education … Treating them fairly requires the government to serve them faithfully, consider their troubles, improve their employment situation, and protect their legal rights with efficient rules. Urban employees have quite different situations from rural migrants; thus some existing policies may not be appropriate for them. The government therefore should be aware of the extreme mobility of rural migrant workers and design policies that fit their needs. Luckily, the central government is already working on the issue of migrant workers. Early this year the ‘State Council’s guidance on solving migrant workers’ issues’ was examined and passed by the State Council. (‘Treat rural migrant workers with fairness and justice’, Bai Tianliang, 2006-03-27) In addition to such text examples, Figure 3.4 shows that the text analysis data identify the increasing proportions concerning the fairness and social justice topic (Topic 23). Comparing the patterns over time for Topic 23 with references to the rural migrants shows a resemblance. References to social justice rose gradually up to the late 1990s, while references to rural migration remained stable. From 1999 on, references to rural migration increased sharply and the Topic 23 proportion also soared a little more slowly (with both showing a marked dip and recovery in 2007–8). The emphasis on social justice and fairness thus increased with the increased visibility of rural migrants, as well as its connection to certain social problems (such as rural migrants) from the topic model. As noted above, Topic 23 refers to the fair distribution and redistribution in the macro issue of institutional reform. Figure 3.4: Topic proportions by year: social justice and rural migrants 74 Pension Policy and Governmentality in China Redistribution between different generations The nature of social insurance also brings up the problem of redistribution between different generations, especially for pensions. For trust in the pension system to work, the government needed to raise funding from current employees, particularly because the state was trying to transform China’s previous ‘payas-you-go’ system into a mixed two-tier system comprising both social and individual accounts. The cost of the social transition had to be met by someone. What kinds of reason were provided by the government to justify the redistribution of social benefits between different generations? The first and most convenient tool was highlighting traditional Chinese culture. Filial piety (xiao 孝 ) is one of the most important and prestigious merits in Chinese traditional culture. Conventionally, caring for the elderly is undoubtedly the responsibility of their children. Filial piety is a virtue that means not only caring for one’s own parents but also showing love, respect, and support for all older people. Hence, in the official discourse, a reconstruction of family and piety culture is generously praised: A recent competition for ‘Star of filial piety’ which selects the person who has been most filial to parents and dutiful to the elderly, is quite unique and deserves our praise. The ‘Star of filial piety’ contest, literally, has the core merit of ‘filial piety’. For instance, one of the winners, Han Shihe, an employee from Hangu Saltworks, is taking care of his grandmother, parents-in-law, and aunt with all his heart and all his strength … Appreciation of filial piety is definitely the main trend in our society and matches the willingness of most of our population. Although there are defiant peoples who refuse to care for their parents, they are not the mainstream. More importantly, their attitudes and actions are despised by society as a whole. Once their cases are exposed, they are denounced and disdained by the whole world. Respecting and caring for the elderly is natural in its essence. Just like parents’ obligation to bring up their children, adult children have the responsibility of taking care of their elders. The pension welfare system in our country is currently incomplete, caring for the elderly still need to be based in families. Thus, filial piety is particularly important [for us]. (‘Complimenting the “Star of filial piety”’, Chen Fei, 1997-11-10) In addition to cultural factors, the state stressed that older people also deserve respect and benefits from society because they have already given their time and effort to it. In addition, the younger generation should see its own future from the situation of the generation before theirs. Moreover, rationally speaking, the ‘time difference’ will help the social welfare system operate better. Here are some examples of these messages: Today’s elderly were hard at work yesterday. They contribute to the development of our society in different degrees. Therefore, they deserve Who deserves benefits and why – constructing fairness 75 to share the material and spiritual benefits brought by the country’s development. Today’s youth will tomorrow become the elderly. Today when they see the whole society treating old people fairly, they won’t be worrying about their fate [in the future]. They will devote more to their work in the prime of life. (‘About elderly issue’, Hong Tianguo, He Liangliang, and Zhang Da, 1982-06-22) And: Social welfare reform needs to take good advantage of the ‘time difference’. [It should include] more people in the social insurance scheme … because most of these people are still young or middle aged and the promise (of a pension) can only materialise 20–30 years ahead. By the time they get old, the social insurance trust will have been accumulated to quite a size and will certainly be able to afford pension and health insurance for these people.1 In this case, the new social insurance scheme, the realisation of past promises, and the social insurance trust will form a strategic triangle and the lapse of time will be valuable. Taking full advantage of the time difference, our social welfare system will be on the winning side. Because the gap in time will give us enough time to reduce holding shares and realise state-owned assets,2 increase the funding of the national social insurance trust, and boost the confidence of the participants in our social insurance scheme. (‘Building a new platform of social insurance’, Gao Shusheng, 200304-08) The logic in the discourse on redistribution was fairly simple, revolving around contributions and rewards, and rights and obligations. The old enterprises deserved to be funded by the new enterprises through the coordinated social pension funding, because they had contributed to the start-up of industrialisation. The rural areas deserved to have transferred public finance because they had contributed to the development of urban areas. Rural migrants deserved fair treatment (more social benefits) because they had contributed to urbanisation and the construction of cities. The elderly deserved better care because they had contributed to society when they were young. The maxim ‘Someone deserves reward because they contributed’ can also be translated as ‘Someone should be rewarded only if they contributed’. Social rights are distributed to anyone who has contributed to society. In this way, the rationalised subjectivity of ‘I am/for everyone’ and ‘everyone is/for me’ was highlighted in the politics of redistribution. This inclusiveness, which was always conditional on ‘contributions’, was especially obvious when taking the ‘lagged time’ into account. We can only imagine our future according to the current redistribution system; therefore, everyone is tightly constrained inside the game of ‘producing and giving’. 76 Pension Policy and Governmentality in China 3.4 A renewed state–individual relationship: the ‘socialised self’ Pension reforms, for whatever social group, are also a move to reconstruct the relationship between the state and the public, as some texts with policy content indicated: Except for some special programmes (such as insurance for occupational injury and childbirth) for which individuals do not need to pay fees according to the law or international practice, the fundraising responsibility for all other social insurance programmes should be shared by the state, enterprises and individuals. Enterprises and individuals should pay the insurance fee, while the government provides a fiscal subsidy under exceptional circumstances. Raising funds from individuals is not only helpful for expanding the funding source of social insurance but also beneficial for increasing labour’s awareness of social insurance. (‘Speeding up and deepening the reform of social insurance system’, 1997-05-20) Accordingly, in this section I investigate the issue of ‘shaping citizens’ expectations’ of state action in the official discourse. What should citizens expect to get from the government when faced with social risks such as unemployment, illness, and ageing? What is the proper relationship between citizens and state regarding welfare responsibility? I generated the classification data by dividing the documents into these subcategories: • ‘locus of responsibility’ • ‘praise’, or ‘denounce’ • ‘national conditions’, and • ‘international experiences’ according to their content. The locus of responsibility includes five types of responsibility allocation regarding care for the elderly in general (such as raising funds for pension fees, daily caring, and so on) as follows: ‘no clear direction of locus’, ‘state/party’, ‘individual/family’, ‘enterprise’, and ‘social coordination’. Since many documents contain more than one topic and sometimes refer to various kinds of message, my coding of the locus of responsibility is based on the most obvious direction of responsibility that can be perceived or identified from the content. Classifying the tone of a document as one of ‘praise’ or ‘denounce’ is useful as these are covariates in addressing the issue of ‘what has been endorsed by the government’. The categories of ‘national conditions’ and ‘international experience’ are important in identifying whether the whole document emphasises the local situation of China or Chinese characteristics Who deserves benefits and why – constructing fairness 77 or draws lessons from overseas experience. The main results in the following discussion use the Naïve Bayes classifier; in Appendix A3, Figure A.8, I present the flowchart of the coding and validation of different classifiers’ performances. Table 3.2 shows the descriptive data for each category from the classification results. Documents that are categorised as ‘praise’, ‘denounce’, emphasising ‘national conditions’ or ‘international experiences’ are coded as present (1) in the data, or not present (0) otherwise. Unfolding the locus of responsibility: topic-based promotions The promoted locus of responsibility regarding elderly care in general (such as fundraising to meet pension fees, daily caring, and so on) varied according to topic. Figure 3.5 presents on the x-axis the expected proportions of each topic in different covariate levels of responsibility. The results are quite intuitive. Within the first chart for Topic 4 (SOE Reform and Marketisation), it is clear the related documents are more likely to be coded as promoting the role of social coordination in solving the problem of care for the elderly, while the topic proportion is smaller in the class of government/party responsibility. For Topic 19 (EE Pension Plan), shown in the third chart, the pattern is quite similar. Although the main target of the EE pension reform was to ease the burden for SOEs and the government, there was not much significance in the signals from the official discourse showing that the government was promoting only individual responsibility, or at least that could not be directly perceived by the readers. By contrast, the topic that refers to the laid-off workers, shown in the second chart, has a much higher proportion at the covariate class of Category Not present Yes present Yes % Praise 2,996 394 12 Denounce*3,343 47 1 Stress national conditions 3,278 112 3 Stress international experiences 3,282 108 3 Notes: ‘praise’ and ‘denounce’ measure the tones of the articles in general; ‘national conditions’ and ‘international experiences’ measure whether the article shows highlights of national conditions or international experiences. *The variation of the variable ‘denounce’ is small and not significant. Accordingly, the results of covariates in the analysis below does not include the statistics of ‘denounce’-related results. They are presented in Appendix A3. Table 3.2: Descriptive statistics of document categories 78 Pension Policy and Governmentality in China Figure 3.5: Topic by covariate: different types of responsibility emphasised Who deserves benefits and why – constructing fairness 79 Note: The x-axis shows the proportion of each class of responsibility. Figure 3.5: (Continued) 80 Pension Policy and Governmentality in China Figure 3.6: Some other topics by covariate: different types of responsibility emphasised Who deserves benefits and why – constructing fairness 81 Note: The x-axis shows the proportion of each class of responsibility. Figure 3.6: (Continued) 88 Pension Policy and Governmentality in China ‘indiscriminate egalitarianism’. The goal of the social welfare system is to defend social fairness and share the fruits of development with all the members of the society … Second, we cannot simply insert the principle of efficiency from the market economy into the social welfare system … we need to prevent potentially bad outcomes brought by some irrational marketisation of the welfare system. Third, we cannot exclude migrant workers and rural residents by treating the welfare system as the special possession of urban residents. However, we cannot ignore the current stage of our national development and try to pursue a universal welfare system too fast. We need to gradually build a unified welfare system through diverse, multilayered institutional arrangements based on the principle of fairness and interest-sharing. (‘Social welfare construction in the harmonious society’, Zheng Gongcheng, 2005-09-16) Conclusions This chapter and Chapter 2 have addressed one crucial segment in answering my overarching question about how the modern state maintains compliance from the governed in a period of rapid social and economic transformation, and how the logic of its governmentality changes accordingly. In these two chapters, I have investigated the logic of ‘whom to govern’ and ‘how to govern’ through a decomposition of the design, implementation, and promotion of the pension reform in China, paying special attention to the way that social problems are identified, the way distinctions are defined, and the knowledge that is produced to persuade the public. At each point in history the promotion of different key topics in pension reform closely interacted with and served the most salient issues in the social and economic reform. For instance, the documents that addressed the topics of laid-off workers and their pension plans was closely connected to the SOE reform. Official discourse pushing demographic control described the pension plan in rural areas as a ‘son’ that could ‘take care of’ the elderly. The propaganda also imported the discourse of social justice and harmony when urbanisation and migration were flourishing. The rationale of reconstructing public knowledge and expectations of the redistribution of social benefits, and the allocation of welfare responsibility between the state and individuals mainly focused on the reiteration of the principles of: ‘contribution and rewards’ and ‘rights and obligations’. (Technically speaking, persuasions that related to enterprise employees – regardless of the socially coordinated solution to pension fundraising, or the individual’s responsibility to contribute to the pension scheme – were more likely to use the discourse of profit-oriented ‘efficiency’.) Meanwhile, the messages related to rural residents and other urban residents – such as birth control, rural migrants, home-based elderly care and so on – were more likely to borrow moral arguments from traditional Chinese culture. Who deserves benefits and why – constructing fairness 89 Individuals’ personal lives, therefore, were socialised, and initiated in a broader system rather than the previous unit-based and localised system. Everyone’s personal interest was materialised and combined with the performance of the whole society, the whole state. Only by contributing to others could people receive rewards. The subjectivity was reconstructed among persons who were directly or indirectly involved in the production process as self-motivated, self-regulated, and self-sufficient in caring. More interestingly in the official discourse is the frequent absence of ‘action initiators’; the suggestions, appeals, and logical persuasions are commonly unidentified but appeal to ‘us’: ‘we should’, ‘ we need to’, ‘our target/job/tasks’, and so on. The anonymity of initiators, in fact, imposes an effective substitution on the readers, and turns them into an ‘activated-self’, and then a ‘responsible-self’. The analysis prompts a follow-up question: ‘to what extent were the state’s advocacy tactics effective in changing public opinion?’ In the next chapter, I use causal inference to empirically unlock the mixed effect of the government’s strategies of combining experimentation and propaganda in a specific policy reform, which was conducted through several waves of policy experimentation in the late 1990s and early 2000s. Data from social surveys provide empirical evidence on changing public attitudes, especially among the target social group for this specific policy reform, under the influence of policy experimentation and related official propaganda. Notes 1 Ironically, a piece of news reported on 12 April 2019 was that, with the current payment rate, the accumulated surplus pension fund would run out by 2035 according to the ‘Actuarial Evaluation of China’s Pension’ published by the Social Insurance Research Center in the Chinese Academy of Social Science. https://perma.cc/L2SG-C6MG 2 To subsidise pension funds, in the late 2000s China also resorted to another means: selling off stock shares and privatising some of its state-owned enterprises (Frazier 2010). References Cai, Y. (2002). ‘The resistance of Chinese laid-off workers in the reform period’. The China Quarterly, vol. 170, 327–344. https://doi.org/10.1017 /s0009443902000219 Chan, K. W.; and Buckingham, W. (2008). ‘Is China abolishing the hukou system?’ The China Quarterly, vol. 195, 582–606. https://doi.org/10.1017 /s0305741008000787 Dean, M. (2010). Governmentality: Power and Rule in Modern Society. Sage publications. 90 Pension Policy and Governmentality in China Foucault, M.; Davidson, A. I.; and Burchell, G. (2008). The Birth of Biopolitics: Lectures at the Collège de France, 1978–1979: Springer. Frazier, M. W. (2010). Socialist Insecurity: Pensions and the Politics of Uneven Development in China. USA: Cornell University Press. Gallagher, M. E. (2011). Contagious Capitalism: Globalization and the Politics of Labor in China. Princeton University Press. Garnaut, R.; Song, L.; and Fang, C. (2018). China’s 40 Years of Reform and Development: 1978–2018. ANU Press. Glasberg, D. S.; and Shannon, D. (2010). Political Sociology: Oppression, Resistance, and the State. SAGE Publications. Grimmer, J.; and Stewart, B. M. (2013). ‘Text as data: The promise and pitfalls of automatic content analysis methods for political texts’. Political Analysis, vol. 21, issue 3, 267–297. https://doi.org/10.1093/pan/mps028 Grün, B.; Hornik, K.; and Grün, M. B. (2018). Package ‘topicmodels’. Hopkins, D. J.; and King, G. (2010). ‘A method of automated nonparametric content analysis for social science’. American Journal of Political Science, vol. 54, issue 1, 229–247. https://doi.org/10.1111/j.1540-5907.2009.00428.x. OA: http://nrs.harvard.edu/urn-3:HUL.InstRepos:5125261 Jefferson, G. H.; and Rawski, T. G. (1994). ‘Enterprise reform in Chinese industry’. Journal of Economic Perspectives, vol. 8, issue 2, 47–70. https://doi.org /10.1257/jep.8.2.47 Lane, D. (2007). ‘Post-state socialism: A diversity of capitalisms?’ Varieties of Capitalism in Post-communist Countries, 13–39. Springer. https://doi .org/10.1057%2F9780230627574_2 Li, J. (2019). Package ‘tmcn’. Lucas, C.; Nielsen, R. A.; Roberts, M. E.; Stewart, B. M.; Storer, A.; and Tingley, D. (2015). ‘Computer-assisted text analysis for comparative politics’. Political Analysis, vol. 23, issue 2, 254–277. https://doi.org/10.1093/pan/mpu019. OA: https://dash.harvard.edu/handle/1/38057808 OECD. (2005). Organisation for Economic Co-operation and Development (OECD) Annual Report. Retrieved from https://www.oecd.org/about /34711139.pdf Roberts, M. E.; Stewart, B. M.; and Tingley, D. (2014). stm: R package for structural topic models. Journal of Statistical Software, vol. 10, issue 2, 1–40. https://doi.org/10.18637/jss.v091.i02 Roberts, M. E.; Stewart, B. M.; and Tingley, D. (2016). Navigating the local modes of big data. Computational Social Science, vol. 51. https://doi.org/10 .1017%2Fcbo9781316257340.004 Scharping, T. (2013). Birth Control in China 1949–2000: Population Policy and Demographic Development. Routledge. https://doi.org/10.4324%2F9 781315027777 Svallfors, S. (2007). The Political Sociology of the Welfare State: Institutions, Social Cleavages, and Orientations. USA: Stanford University Press. https:// doi.org/10.11126%2Fstanford%2F9780804754354.001.0001 CHAPTER 4 Maximising support for pension reform using policy experimentation, and the potential to backfire With the emergence of market orientation, the socialist and post-socialist countries were confronted by the challenge of conducting proper reforms to their welfare systems and retrenching the demands that their legacy arrangements imposed on the state. However, because the Chinese public had not reached a consensus regarding the relative shares of the state and the individual in welfare responsibility, movements contradicting the previous image of the state–individual relationship were potentially controversial. In addition to its propaganda for change, reviewed in Chapter 3, an essential element of the Chinese government’s statecraft was a strategy of regional experimentation with shared welfare responsibilities. Regional pilot schemes were deployed to facilitate the public understanding of shared welfare responsibility on the pretext that the public had not reached a social consensus. Meanwhile, the official propaganda of local governments emphasised government omnipotence, which assisted them in managing public faith in regime capacity and governance. In this chapter, I take advantage of a quasi-experimental pilot policy in China, referred to as the ‘pension insurance pilot scheme in urban areas’, to explore the case of hybrid responsibility of welfare provision. This empirical analysis offers a counterfactual analysis of the effect of policy intervention and official propaganda on the attitude of welfare responsibility allocation and regime support of the general public by taking advantage of the pilot policy launched by the central government in selected provinces. With the help of two nationwide surveys (‘Chinese Attitudes toward Inequality and Distributive Injustice’) conducted in 2004 and 2009, I collected over 5,000 randomly pooled cross-sectional data results for residents in eight treatment and 12 control provinces. How to cite this book chapter: Wang, Yan. 2022. Pension Policy and Governmentality in China: Manufacturing Public Compliance. London: LSE Press, pp. 91–127. DOI: https://doi.org/10.31389/lsepress.ppc.e. License: CC BY 92 Pension Policy and Governmentality in China The empirical results show that the pilot policy launched by the Chinese central government significantly affected the citizens’ understanding of shared responsibility and privatised social risks in general and that the public apparently accepted the underlying accentuation of the individual’s responsibility for pension contributions. The length of time that the policy has been in force amplified the attitudinal change in individuals’ perception of the government’s role. Local official propaganda, which emphasised the image of an ‘omnipotent government’, moderated the treatment effect of the pilot scheme on the public’s understanding of shared responsibility. In other words, local propaganda helped the regime to maintain the public’s faith in the government’s capacity and responsibility for social welfare provision, while gradually implementing the experimentation policy. Beyond the attitudinal change toward individual welfare responsibility, I also find that the influence of policy propaganda interfered with the pilot policy, as shown by the contradictory finding about the public’s political trust. Controlling for other factors, short-term exposure to the local propaganda – which praised the government’s efforts to respond to people’s expectations and improve people’s living standards – increased the public’s confidence in institutions. However, the disjunction of the policy content, which set out to share with individuals the responsibility for welfare – and content of the propaganda actually backfired on the authorities regarding institutional trust in the long term. Thus, the strategies used by governments were influenced by the conditional and practical aspects of policy promotion and the local governments’ handling of the perceived role of the government for public. A mismatch between the policy content and propaganda details is likely to be noticed by the public – especially by members of the target population (enterprise employees) who are more likely to expect a ‘big government’ that can take care of their social risks – and to weaken their support for the institution. 4.1 Risks in the pension reform and the statecraft of policy experiments As a special form of society, state socialism adopts a distinct structure of institutions and rules regarding development, production relations, and welfare provision (Polanyi and MacIver 1944). The basic principle of state socialist regimes is that material resources are distributed through central planning and a system of political identification. Take the example of China. Before the reform and opening up in the late 1970s, the state organised and governed individuals through work units (danwei) in urban areas and people’s communes (renmin gongshe) in rural areas. However, the redistributive principle posed serious challenges to governance continuity. In particular, collective ownership hindered production efficiency, while the offering of incentives for productive Maximising support for pension reform using policy experimentation 93 improvements and the scarcity of resources cultivated a sense of ‘manipulated equality’ among privileged groups (Kornai 1992; Szelenyi 1978). The unsustainability of such institutionalised settings pressured state socialist countries in the late 20th century to pursue market reforms (Szelenyi and Szelenyi 1994). The emergence of market power during the reform led to the change in distributional principles, the return rate of capital and human capital, social structures, and so on (Bian and Logan 1996; Nee 1996). At the same time, the boundaries of the state, market, and society in these state socialist countries were redrawn and negotiated across different sectors, especially in Eastern Europe, eastern and south-eastern Asian countries, and the former Soviet Union. Szelenyi and Kostello (1996) argued that market competition, as opposed to state mechanisms, began to play a greater role in Eastern Europe in 1980–89 and in China after 1985. Subsequently, among the East European countries after 1989, the privatisation of public enterprises became a key state policy. The transitional process caused profound changes in the social welfare provision in China. Market-induced competition led the state and urban collective enterprises to reduce or renege on pensions, medical costs, and housing for employees (Guthrie 2012; Song and Chu 1997). The responsibility for welfare provision shifted from state institutions to society (including families) and this scheme was identified by the Chinese central government as the ‘socialisation of social welfare’ (State Council 2000a). This special concept captures the change from state-led welfare provision to a welfare system with multiple contributors, in which the state, the market, the sectors of society, and families all share welfare responsibility, and the work units no longer take on the welfare function. To match the socialist market economy, the government promoted joint responsibility as the practical method of funding, service provision and social welfare regulation (Li and Zhong 2009; Wong and Ngok 2006). This process was similar to the ‘retrenchment’ and ‘risk privatisation’ process in Western welfare states and the emergence of a hybrid welfare system there, in which individuals were given flexibility but increased responsibility for handling the various social risks related to their personal lives, such as ‘unemployment, death of a spouse, retirement, disability, childbirth, [and] poverty’ (Hacker 2002, p. 245). From the mid-1970s onward, welfare states faced rising unemployment rates, high levels of inflation and low economic growth, along with demographic changes that made the social policy increasingly costly and unsustainable. Welfare systems were burdened with extensive problems over ‘low-wage, low-skill labour with low work incentives’, which were severe in the case of ‘welfare without work’ (Anderson and Ebbinghaus 2011; Pierson 2001). Meanwhile, ‘big government’ in welfare provision was recognised as over-committed and underperforming (Mashaw 2006). In these crises, Western policymakers decided to address the new social and economic situation by adopting a neoliberal modernisation strategy (Hall 2001). The welfare regimes in Western Europe, therefore, opted for the retrenchment of welfare provision, including: 94 Pension Policy and Governmentality in China policy changes that either cut social expenditure, restructure welfare state programs to conform more closely to the residual welfare state model, or alter the political environment in ways that enhance the probability of such outcomes in the future. (Pierson 1994, p. 17) In addition to the cutbacks in public spending and other fiscal rearrangements, certain countries started to reconsider citizens’ social right in order to legitimise the welfare reforms. For instance, rather than the universal entitlement to social rights for every qualified citizen, welfare support should be provided for the population that truly needs it (Béland 2005; Cox 2001). Models that offered a choice, such as the contracting-out and welfare-mix model, were popular in developed countries and also used frequently in less developed countries. For instance, the private pillar in the pension system was adopted in the 2000s in countries such as Chile, Mexico, and Uruguay, with varied proportions of public to private (Mares and Carnes 2009). In East Asia, countries such as Korea and Malaysia also expected a fast-expanding private market in social services (Gough 2001). Haggard and Kaufman (2008) located their discussion in the context of the global economic crisis and identified attempts to retrench social policies in Latin America and Eastern Europe driven by the economic crisis and liberalisation, leading to extending a hybrid welfare system in which social welfare responsibilities were shared between the state, the market, and individuals (Benish, Haber, and Eliahou 2017). Individuals now faced long working lives and redefined responsibility in a state-led welfare-mix system where they enjoyed enhanced ‘flexibility’ and increased ‘responsibility’. As the 2010s began, welfare reform in China gradually stopped its previous pursuit of marketisation and socialisation. The official discourse about social policy now focused on expanding ‘welfare coverage’ to rural and non-salaried urban residents. Certain scholars have labelled this new trend in the welfare format ‘state capitalism’ or a ‘state paternalistic capitalism’ (Gao, Yang, and Li 2013). In this chapter, I focus on the period of ‘welfare socialisation’ in the 1990s and 2000s, when the main pension reform for enterprise employees was aligned with the marketisation of state-owned enterprises (SOEs). This period was an important stage in which the ideas of ‘a welfare system with hybrid contributors’ and ‘shared responsibility for facing social risks’ were developed and actively promoted by the state. Only after the reconstruction of the public’s understanding of shared responsibility could the state be confident in diffusing its attempts to expand welfare coverage without reincurring huge burdens similar to those in the state socialist period. Existing studies have carefully examined the details of the pension policy in this period, such as the return rate, coverage, and return on investment of pension trusts (for instance, Li and Ge 2010; Li and Wang 2009; Lin and Ding 2007; Yang, Wang, and Zhang 2010). However, the effects of the transitional process on social beliefs, especially public perceptions of the state–individual Maximising support for pension reform using policy experimentation 95 relationship and the function of institutional dependency in shaping people’s political attitudes, have not been fully investigated. The rhetoric of ‘socialism’ itself describes the state or state-managed public bodies acting as sources of social welfare – in schemes similar to a social contract – with the cost of low salaries and limited social mobility (Haggard and Kaufman 2008). For instance, work units in urban China before 1978 were subsidised by the state so as to provide not only jobs to individuals but also to generate pensions, housing, education, and health care to employees and their dependents (Lu and Perry 1997). In the rural areas, funds were also allocated for the basic education and medical support or health care of residents (Wong 2005). The socialist institutional setting cultivated a strong image of an omnipotent government, in which the state was expected to superintend the social welfare of the public, especially among urban enterprise employees, who were expected to be cared for ‘from cradle to grave’ (i.e. with generous welfare benefits). The legacy of this socialist policy was a population with a strong sense of state dependency, attachment to the welfare state and organised stakeholders favouring the welfare setup (Cook 2013). However, the cognitional inertia of the public’s assumptions about welfare responsibility arising from socialist traditions may pose considerable challenges when neoliberal reforms come to be implemented. For instance, Cook (1993) found that workers from the former Soviet Union expressed discontent and to a certain extent jeopardised state legitimacy when the social contracts that used to guarantee their economic welfare broke down. Denisova et al. (2012) analysed data from a survey held in 2006 involving 28,000 individuals from 28 post-socialist countries. They found that transition-related difficulties influenced respondents’ support for privatisation reforms, given the preference for state agency and concerns regarding the legitimacy of privatisation. Evidence from East Europe suggested that, when ‘literally over a single night, all the things that had been taken for granted were no longer valid’, many individuals suffered from ‘serious identity crisis’ (Ekman and Linde 2005, p. 357). Apart from cognitional dependency, two other factors weakened welfare reforms that were intended to rebalance state–individual accountability. First, when the market principle was introduced in social and economic matters, the ruling party in China did not relinquish the claim of communist ideology. Perry (2007; 2017) indicated that its resilience in power was endorsed by its skilled employment of the communist revolutionary legacy and of symbols from traditional culture. Slogans were presented, such as ‘serve the people’ and ‘the party represents the benefit of the overwhelming majority of the people’. By combining the destiny of the party and the welfare of the people, these repeated and solemnly vowed ideological claims in fact aided the formation of the ‘common interest’ of society and became an important element of the ‘inertial thinking’ of the public when it had to face external changes. In other words, people in former socialist states are more likely to treat the government as the bearer of ultimate liability for all social and economic problems. But, at 96 Pension Policy and Governmentality in China the same time, they are more likely to accept social or policy changes ‘for the sake of the common interest’. Second, the shadow of other previous public policies also shaped the expectations of the people. For instance, the demographic ‘one-child’ policy posed a specific quandary for the care of older people in China because its demand for the obligatory compliance of the people in reducing the number of their descendants went against perceptions of children as being the main form of old-age support in the traditional context of filial piety. So, to gain leverage and make a morally and politically fair request, the government had to take charge of elderly welfare there. Despite this fairness rhetoric, urban residents suffered more than those in rural areas, because the one-child policy was strictly enforced in urban areas. To sum up, China’s welfare reforms during the ‘socialisation process’ in the 1990s and 2000s may have induced a failure of consensus between the state’s conduct and individual perceptions regarding the role of government in welfare provision. The difference may also have endangered state legitimacy by the public’s sense of betrayal. To avoid a legitimacy crisis caused by the above consensus gap, the Chinese government needed to seriously consider the public’s expectations and anticipate their feedback in its policymaking process, especially during dramatic transitions in the social welfare provision. A parallel statecraft of social policy experimentation and official propaganda was created to address the consensus gap and resolve the dilemma of combining privatised social risks and shared responsibility during the pension reform process. In China, policy experimentation is a core tool of an incremental policy process, a matter of ‘crossing the river by feeling the stones’. Such experimentation is very important for maintaining state legitimacy while avoiding radical policy changes at the national level. In a huge state like China, a process of conducting moderate and manageable policy changes allows enough space for the central government to learn from ‘trial and error’ (Heilmann 2008b). Policymakers of the central government could determine the types of experimentation to undertake, look at the results, and decide what aspects of successfully implemented experimentation could be adopted at the national level (Heilmann 2008a; Mei and Liu 2014; Zhu and Zhao 2018a). Provinces, cities, or regions could participate in the selection of pilot sites. Within the process of pilot policymaking, local governments could benefit from conditional and limited autonomy if the central policies permitted. Apart from the unified policy guidelines proposed by the central government, local authorities could localise and reinterpret the policy details on the basis of their specific local conditions (H. Huang 2013). In such a structure, local governments could also employ strategies, such as the construction of rhetoric, enhancing certain policy elements to popularise the policies when they sense potential obstruction from the public. I propose that policy experimentations in the field of social welfare were used as a tactic for dialogue with the general public to demonstrate the legitimacy of a reform. The idea of experimentation as a dialogue tactic differs from the Maximising support for pension reform using policy experimentation 97 conventional understanding of policy experimentation in the theoretical spectrum. The traditional wisdom on policy experimentation has mainly focused on policymakers at the central and local levels, but has paid little attention to the recipients of social policies – the public. Conventional theories either highlight the autonomy of local authorities, arguing that decentralised federations contribute to economic leapfrogging (Montinola, Qian, and Weingast 1995; Weingast 1995) or accentuate the full control of the central government, whose experimentation serves to demonstrate that policies are workable (Heilmann 2008b), identify errors (Zhu and Zhao 2018a), or delimit competition (Cai and Treisman 2006). I emphasise that the government can integrate the public’s expectation into the policymaking process around policy experimentation and employ a feed-forward effect (Schneider and Sidney 2009) to minimise any undesired impacts on society. If the central government pushes for an acrosscountry reform that contradicts its omnipotent stature (which is generally interpreted by the public as being a ‘caring and accountable’ government), then the difficulties for the public of identifying and accepting the new situation can be considered risky, given the cognitional dependency discussed above. By distributing policy changes in selected regions through policy experimentation, the government can measure and test the potential feedback and the limits of the public in turn, and contain the public’s expectations about the general process of policymaking. Moreover, policy experimentation provides effective channels for the government to further influence public opinion by engaging individuals in dialogue. The central government can wield power via several mechanisms. First, the experimentation depends upon the fiscal division between China’s central and local governments, where the central tier controls the fiscal resources and allocates transfer payments to local governments, and the local governments are motivated to accomplish or improve pilot programmes so as to gain financial resources and cope with welfare expenditure (Zhu and Zhao 2018b). This interactive loop helps the central government find the most appropriate policy instrument to develop as the much broader version of the experimental reforms. Moreover, the central government can establish its legitimacy by modestly implementing the reform process (Zhu and Zhao 2018b). Second, the general public may change its perceptions, mainly through an experiential process, in the context of external experimental interventions. Gradual reforms based on policy experimentation instil in the public the belief that a new policy may be cancelled if it does not work. In the regions chosen to pilot the new welfare policy, the public may have derived benefits and observed deficiencies – but with less anxiety. Therefore, the central government can promote new policies without the need to obtain large-scale prior approval from the general public as a whole; rather, it can persuade people by informing them of the pros and cons of the policies as the experimentation proceeds. An incremental pilot scheme facilitates the building of public attitudes, in which the most preferred outcome corresponds to the policy design, especially when a 104 Pension Policy and Governmentality in China experiment created by the pilot scheme for pensions for enterprise employees in the 2000s. The main research interest is the effectiveness of statecraft regarding changes in individuals’ attitudes to the locus of responsibility (LoR) of certain welfare provisions, and regarding the trust given to political institutions across regions and periods. Individual perceptions of welfare responsibility are sensitive to changes in social policies (Im and Meng 2015; Lü 2014) and are of great importance in understanding political support in general. As a market-oriented policy that resonated with the privatisation of the SOEs in the late 1990s, the pension insurance reform had as one of its goals the sharing with individuals, the market, and society of the responsibility for social pension insurance contributions. For the central government, the ideal micro-level outcome of the reform would be that the target population should recognise their responsibility as individuals for pension contributions, thereby achieving a sustainable system of pension contributions for future retirees. Accordingly: Hypothesis 1: The implemented pilot experimentation of the basic pension insurance reform increased the popular acceptance of individual responsibility for elderly care. The longer the public experiences the pilot experimentation, the more intensely they become affected by and adapted to the reform. The new emphasis on incorporating individual responsibility in welfare provision differed from the egalitarian–socialist period practice stressing the duty of the state to provide elderly care in the form of either pensions or social insurance. So, local governments were strongly motivated to take precautions to offset the potential negative effects of the policies – the anxiety of the public (especially the target population) over losing the state’s support/benefit. Meanwhile, the local governments did not want to be blamed for the reform, as they would have been if the public distributed blame/trust in its usual hierarchical fashion, given the special central–local relationship in China. A countervailing stratagem against this risk was for the local governments to use official propaganda to send messages in the course of the basic pension insurance reform emphasising the omnipotent role of the state. I found qualitative evidence of the propaganda efforts of local official newspapers focusing on the public’s faith in the government’s devotion to duty and responsibility. Drawing on these arguments, I test the following hypothesis: Hypothesis 2: Policy propaganda emphasising the omnipotent role of the government offset the pilot policy’s effect on public perception and shifted the people’s perception to governmental responsibility. I recognise the short-term effectiveness of official propaganda in maintaining the public’s faith. However, in the long term, individuals will be likely to distinguish the real target of the policy and even to resist its implementation (Chen Maximising support for pension reform using policy experimentation 105 and Shi 2001; H. Huang 2018; Kennedy 2009). In the case of the pension insurance policy reform, although its description in the official media highlighted the government’s efforts to improve the public pension system, individuals reportedly had to increase their individual contributions to their pension, and once in retirement they encountered difficulties in claiming benefits. People who conscientiously planned their monthly income and expenditures were deeply influenced by the implications of governmental retrenchment as part of the policy design. Thus, in the event, the reform dramatically changed the individuals’ disposition of their salary and their expectations of risks as they aged, social welfare and the state–individual relationship. This contradiction between policy propaganda and policy experimentation may over time undermine the public’s confidence and trust on government institutions. Thus, I propose the following hypothesis regarding the concurrent effect of the propaganda and pilot policy on the public’s political trust: Hypothesis 3: In the short term, local official propaganda regarding the pilot policy increased the public’s support for the regime. However, in the long term, local official propaganda regarding the pilot policy can reduce the public’s support. To measure the dependent variables (DVs) here I draw on two rounds of household surveys called ‘Chinese Attitudes toward Inequality and Distributive Injustice’, which were conducted by teams led by Marty Whyte and Mingming Shen in 2004 and 2009. The two surveys used randomised spatial sampling under the global positioning system (GPS sampling). The sample pool of the national adult population involved respondents aged 18–65. The total number of observations from the two surveys is over 6,100, as shown in Table 4.1. Since the three provinces in north-eastern China started the pilot policy before 2004, we dropped the samples of these three provinces from the dataset, thus constraining the analytical samples within the window of the two surveys (2004 and 2009). This modification left the analysis with 5,280 observations from 20 provinces. The main independent variable in this study is the different waves of pilot policy. The full pilot policy started in 2001–3 and expanded in 2006 and 2008, respectively involving three, eight, and two pilot provinces. Figure 4.1 shows the provinces in the three waves. The construction of treatment variables is explained in the model identification section. The two core questions in the survey that were used to construct the DVs for hypothesis testing are as follows: (DV for Hypothesis 1): ‘Between the government and the individuals, who should take greater responsibility for elderly pension provision?’ (DV for Hypothesis 2): ‘Do you trust the central/provincial/local governments?’ (asked as separate questions)’ 106 Pension Policy and Governmentality in China Figure 4.2 presents the provincial variation of the dependent variable – the locus of (welfare) responsibility from two rounds of survey data. In addition to the dependent variables, the two-round survey also provides demographic information about the respondents. These variables are then used as control variables in our models and comprise (among others): age, gender, educational Table 4.1: Descriptive data of two rounds social survey respondents Province Survey year Total samples Pilot year Pilot wave 2004 2009 Shandong 486 453 939 2006 2 Hubei 251 291 542 2006 2 Heilongjiang 239 212 451 2003 1 Guangxi 242 206 448 — 0 Liaoning 205 183 388 2001 1 Anhui 177 184 361 — 0 Guangdong 164 181 345 — 0 Shanghai 233 87 320 2006 2 Beijing 121 136 257 — 0 Henan 122 110 232 2006 2 Shanxi 112 104 216 2006 2 Xizang 94 108 202 — 0 Zhejiang 90 111 201 2008 3 Yunnan 137 63 200 — 0 Jiangsu 88 107 195 2008 3 Fujian 87 71 158 — 0 Shaanxi 67 73 140 — 0 Jiangxi 42 84 126 — 0 Hunan 54 68 122 2006 2 Hainan 58 54 112 — 0 Hebei 58 44 102 — 0 Ningxia 25 37 62 — 0 Samples (total) 3152 2967 6119 Notes: Martin Whyte served as the principal investigator for the project, which also involved Albert Park (Oxford University), Wang Feng (University of California-Irvine), Jieming Chen (Texas A&M University-Kingsville), Pierre Landry (Yale University), and Shen Mingming (Peking University), with Jie Yan, Tianguang Meng, and Chunping Han as research assistants. The initial project was held in 2004. Maximising support for pension reform using policy experimentation 107 Figure 4.1: Visualisation of three waves pilot policy Figure 4.2: Provincial variations of the dependent variable: locus of responsibility perception 108 Pension Policy and Governmentality in China attainment, marital status, party membership, household income level, and residential registration (hukou) status. To measure the independent variable of local official propaganda, I collected newspaper articles from the ‘China Knowledge Resource Integrated Database’, 1 which covered data beginning in 2000. The official newspapers published by the local provincial party committees were selected, because these were the provincial party newspapers that generally highlighted propaganda information from the provincial government. Moreover, the propaganda rhetoric of provincial party newspapers can help to construct public opinion in the provinces. Officials’ attitudes to the current welfare policy that are revealed in provincial newspapers can be spread and appear in other media platforms across a province. So people who do not read or subscribe to official provincial newspapers are also informed about such attitudes. The collection of data involves keyword searching and manual selection. To capture the intensity of propaganda regarding the pension insurance reform, I collected articles containing the exact name of the pilot policy (e.g. ‘fully funding the individual accounts’) to construct the variable ‘policy propaganda’, and I used the ratio of the variables rather than absolute numbers. Figure 4.3 presents the provincial variation of local official propaganda efforts (as the article rate) on the pilot policy. Moreover, I calculated the accumulated ratio in three or five years ( ( ) å å = ( ) t t Article of Pilot Policy Total News Articles Ratio , where t equals three or five years before the two survey years of 2004 and 2009, respectively) to capture the long-term effects of the propaganda. To compare the provincial-level covariates between the treatment provinces and the rest of the country, for the period covering 2000 to 2010 I collected from the National Bureau of Statistics2 provincial-level statistical data on social and economic variables – which can influence the possibility that certain provinces will be selected as pilot provinces and the public’s perception of pension insurFigure 4.3: Provincial variation of local official ‘policy propaganda’ efforts Maximising support for pension reform using policy experimentation 109 ance. The selected social and economic variables included regional economic performance, demographic characteristics, fiscal revenue and expenditure distribution, the implementation and participation rate of pension insurance, and so on. The nature of the two rounds of survey data permits us to adopt the difference-in-differences (DID) model for estimating the average policy effect on individuals through counterfactual inference. The detailed justification of the counterfactual DID design is presented in Appendix A, Section A4, including parallel trends and individual and regional balance. In the DID model, I define the treatment group as all the samples from the provinces that participated in the pilot policy, with other provinces used as the control group. The baseline model is to estimate the difference between the treatment area and the control area before and after the policy experimentation. ab b b b e =+ + + + ++ 1 23 4it i t i t it i it LR Pilot Post Pilot Post X P (1) where LRit denotes the individuals’ attitudes regarding the LoR of pension insurance; Postt is a dummy variable that equals 1 for year 2009 and 0 for year 2004; Pilotj is the treatment variable that equals 1 for samples in the piloted provinces and 0 otherwise; the β3 of the interaction term between Pilotj and Postt is the average treatment effect on individuals; Xit is an array of control variables that is employed to capture minor imbalances in demographic factors that can interfere with the outcome of interest; and Pi is a a dummy variable for the provinces. Given that the pilot sites were selected at the provincial level, I included a dummy variable Pi for provinces to ensure that the selections did not lead to an overestimation of the treatment effects. By taking advantage of the condition that the policy was targeted at urban enterprise employees, I further analysed the occupational and residential differences by using different subsamples. In addition to the dualistic treatment-or-control variable, I also coded a continuous variable denoted by Duration to capture the gradual feature of the policy implementation by substituting the variable Pilot. This variable on Duration corresponds to the length of time that each treatment province had experienced the new pension arrangements by 2009, the time that the post-treatment survey was conducted: so its value was set at 0, 1, or 3. Duration also captures the slight policy differences between two different waves.3 The model including ‘Duration’ as the explanatory variable is similar to Model (1). ab b b be =+ + + + ++ 1 23 4 it i t i t it i it LR Duration Post Duration Post XP (2) where Durationi is the length of time for which each sample experienced the pilot policy. 110 Pension Policy and Governmentality in China In investigating the mixed effect of the pilot policy and the local official propaganda, I further constructed the difference-in-difference-in-differences (DDD, or triple difference) model as follows: ab b b b bb b be =+ + + + ++ + + ++ 1 23 4 5 6 7 8 ijt i t j i t i jt j i t j ijt i ijt LR Pilot Post Propaganda Pilot Post Pilot Propaganda Post Propaganda Pilot Post Propaganda X P (3) where ‘propagandaj’ represents the individuals’ direct exposure to the pilot policy. The coefficient β7 of the interaction of pilot effect (Pilot × Post) and propaganda thus caught the concurrent effect on the outcome variable. I contained the same control variables of Xit and province dummy Pi To test the short-term and long-term effects of the local policy propaganda on the public’s political trust, I employed the question in the 2009 survey measuring the level of Chinese citizens’ trust in the central government, provincial government, and local government (county or district). (There were no questions related to political trust in the 2004 survey – so I can use only the 2009 survey for data on political trust.) Short-term and long-term propaganda were identified with the accumulated ratio of articles that contained the exact name of this pilot policy in the previous year, three years and five years. I constructed the following model by using the interaction between pilot policy and local propaganda to capture the marginal effect of official propaganda on the public’s political trust in the treatment provinces: ab b b be =+ + + + ++ 12 34 ij i j i j ij i ij Trust Pilot Propaganda Pilot Propaganda X P (4) where the marginal effect of local official propaganda on political trust was calculated as ( ) ( ) bb ¶=+ ¶23 ij i j Trust Pilot Propaganda (5) Thus, the coefficient β2 + β3 × 1 indicates the estimated marginal propaganda effect on the public’s political trust in the pilot provinces. Here, the dichotomous variable Pilotj can be replaced as the continuous variable Durationi (to be discussed in the empirical section), which then turns the measurement into the marginal effect of propaganda on political trust for one additional year. Using Durationi helps us identify the long-term and short-term effects of propaganda in spite of the stepwise pilot policy. I present the DID regression results of the public’s attitude to the LoR on pensions by using Pilot, Post and the interaction between Post and Pilot (pilot effect), along with other control variables, in Table 4.2. Clustered standard errors at the provincial level are reported in parentheses. The coefficient of the Maximising support for pension reform using policy experimentation 111 pilot effect (Post × Pilot) shows a significant positive effect (0.126), which indicates that the policy in the treatment provinces after the pilot increased the public acknowledgement of individual responsibility on pension insurance. After controlling for province, year and individual demographic variables, the effect remains positive (0.091) but insignificant. This mixed effect is partly addressed in the following section by discussing the propaganda effect. The interaction between Duration and Post (duration effect) has a significant and positive effect (0.073) on people’s LoR attitude after controlling for the demographic factors, provinces and year dummies. This result indicates that the people in the provinces who experienced longer pilot policy experimentation had higher levels of acceptance of individual responsibility on pension insurance. Thus, our Hypothesis 1 is supported. As shown in Table 4.3, after controlling for the demographic features and province dummies, policy propaganda was shown to have a contrary effect on the public’s LoR conditioning in pilot situations. In other words, people affected by the pilot policy were likely to have a higher impression of governmental responsibility when exposed to stronger local official propaganda. The result from the decomposed subsample indicated that the ‘pulling back’ function of the policy propaganda was significant for the target populations of Table 4.2: Where people see the locus of government responsibility for pensions by policy effects Variables Treatment vs control Policy duration Pilot effect (DID) 0.126** (0.060) 0.123* (0.068) 0.091 (0.068) Duration effect (DID) 0.085*** (0.020) 0.083*** (0.023) 0.073*** (0.023) Demographic controls No Yes Yes No Yes Yes Provincial dummies No No Yes No No Yes Year dummies Yes Yes Yes Yes Yes Yes Observations 4,921 3,790 3,790 4,921 3,790 3,790 R-squared 0.025 0.069 0.148 0.027 0.071 0.150 Note: *p<0.1, **p<0.05, ***p<0.01. The table presents ordinary least square (OLS) results. Clustered standard errors at the provincial level are reported in parentheses. The estimates of treat, post, duration, demographic controls – which include age, age squared, gender, educational attainment, marital status, party membership, household income level, and hukou status – are not reported. The estimates of constants, provincial dummies, and year dummies are not reported either. 112 Pension Policy and Governmentality in China the reforms, namely, the enterprise employees (−0.161) and the public sector employees (−0.552) in urban areas. Meanwhile, rural residents who had largely been alienated from the policy were weakly influenced by the propaganda, as shown by the slight increase in their confidence (0.018) on the omnipotent role of the government. Hypothesis 2 is thus supported. Along with the pilot policy, local governments’ official propaganda in defending the governmental image in welfare provision (as well as the justification of joint responsibility to a certain extent) attenuated people’s faith – especially that of the target population of enterprise employees – in the ‘glorious government’. In other words, official propaganda acted as a moderator for the treatment effect of the pilot scheme and kept the government from ‘losing face’ (faith). 4.3 The concurrent effects of experiments and media campaigns on political trust Despite the image construction of a ‘caring and accountable’ government that may have effectively swayed public opinion in the short term, the case presented here shows that the divergence between the propaganda images and the benefits derived by individuals from the pilot policy would probably result in political Table 4.3: Effect of pilot policy and policy propaganda on the locus of government responsibility Variables All samples Urban samples Rural samples Enterprise employees (target population) Public sector employees (urban) Pilot effect (DID) 0.500*** (0.117) 0.557*** (0.168) 0.379** (0.176) 0.523 (0.339) 2.245*** (0.722) Pilot effect × Policy propaganda (DDD) −0.105*** (0.029) −0.184*** (0.041) 0.018 (0.043) −0.161** (0.080) −0.552*** (0.197) Policy propaganda 0.017*** (0.006) 0.045*** (0.008) −0.021** (0.009) 0.038*** (0.012) 0.029 (0.024) Policy propaganda × Pilot −0.026* (0.014) −0.066*** (0.018) 0.012 (0.021) −0.082 (0.052) 0.142 (0.140) Policy propaganda × Post 0.079*** (0.025) 0.139*** (0.036) −0.009 (0.035) 0.123** (0.061) 0.261** (0.120) Observations 3,790 1,947 1,846 729 207 R-squared 0.156 0.164 0.149 0.191 0.260 Note: *p<0.1, **p<0.05, ***p<0.01. All models include demographic controls and province and year dummies. Maximising support for pension reform using policy experimentation 113 distrust in the long term. The results of the marginal effect of propaganda on the public’s political trust in the pilot provinces are shown in Table 4.4. The coefficient of the interaction between treatment and policy propaganda indicates that local official propaganda in pilot areas significantly increased the public’s trust in local governments by 0.286 (=0.497−0.211) and in provincial governments by 0.132 (=0.298−0.166) in the short term (one year), calculated using Equation (5). The effect on the public’s trust in the central government (0.012) is not as significant as the effects on trust in the local and provincial governments. Meanwhile, in the long term (three years), official propaganda significantly but negatively affected the public’s trust in the local and provincial governments (−0.279 and −0.129, respectively). The effect on the public’s trust in the central government (−0.012) was not significant. The coefficients of accumulated policy propaganda for five years indicate a similar pattern: policy propaganda affected the public’s trust in local and provincial governments (−0.131 and −0.060, respectively) in a significant and negative way, while the effect on trust in the central government was very weak (−0.006). The effect on the central government was clearly not statistically evident, which is reasonable, considering that the statistically measured propaganda and the above discussion on pilot policy took place only at the provincial level. So Hypothesis 3 was also supported. In this result, I also distinguished a possible confounding variable to capture the self-interest factor under the policy effect: the number of older family members who needed to be taken care of (as shown in Table 4.4). The coefficients of this variable in the various models suggest that the direct self-interest factor has limited influence on the public’s political trust. A logical concern over the difference between the short-term and longterm effects of propaganda on political trust may arise if we consider the different waves of policy implementation in the treatment provinces. Therefore, I added another test to show how the effects of propaganda varied in the different waves, shown in Table 4.5. Here I replaced Pilot with Duration in Equation (4). The marginal effect of propaganda on the individuals exposed to the pilot policy for different years is calculated with β2 + β3 Durationi. The results, which are consistent with those in Table 4.4, further validate our hypothesis on the effect of incremental experimentation. For individuals in the treatment provinces that started their pilot policy in 2008, the duration of experience of the reform is one year. Therefore, the estimation of the effect of the one-year propaganda on the public’s trust in the 2008 wave of pilot provinces is −0.045 (=−0.211+0.166×1). Individuals in these provinces in the 2006 wave of policy experimentation with a three-year pilot exposure show significant and positive change in their trust on the local government by 0.287 (=−0.211+0.166×3), different from the weak negative effect on the ones under the intensity of one-year official propaganda. For provincial governments, the pattern is similar: short-term official propaganda shows 120 Pension Policy and Governmentality in China −0.562 (=−0.556−0.006) and the provincial government (from 0.206 to 0.399) using similar calculations. The results show that, with long periods of local official propaganda, the target population’s political trust in local and provincial governments regarding the pension pilot policy consistently declined, reversing from positive to negative. In general, we may conclude that the negative effect of long-term local official propaganda on local regime support does not vary among different social groups. More robustness tests, including results with order logit model, multilevel model, models with intergenerational difference, and other confounding variables can be found in Appendix A, Section A4. Conclusions Policy experimentation can help downplay controversial reforms by slowing down policy implementation and minimising the confusion and reaction of the public. In cases where the authority wishes to evaluate the potential tradeoff between opportunities and challenges of conducting social and economic reforms (social change), policy experimentation is useful to buffer the associated risks. Meanwhile, it is vital for the authority to promote its intentions, construct knowledge, and shape a population’s ideology by propaganda. Hence, a combination of incremental piloting of policy and relevant constructed propaganda helps the government to manage any chance of public discontent and build a consensus for the reform. A government always needs to make the greatest possible effort to promote a potentially controversial policy, especially during a time of socio-economic transition. After the ‘socialisation’ period, the welfare reform in China entering the 2010s exhibited substantial diversity in policy design, under increasing pressure from the people over the social justice of redistribution and the sustainability of the welfare system. The government proposed reforms such as postponing the retirement age, integrating the rural and urban pension schemes, and allowing the social security fund to be listed. However, the steady progress of these follow-up adjustments relied on a consensus over the sharing of welfare responsibility between the state and individuals. Evidence in this chapter shows that China’s pension reform for urban enterprise employees is a combined tactic of policy experimentation and official propaganda. However, although the socialisation of welfare provision relieved the government of its complete responsibility, the transition may also reduce the public’s confidence on government capacity and accountability. Therefore, the central government used policy experimentation for the pension reform as a dialogue mechanism between the state and the public for building up social consensus on a welfare system with hybrid contributors. Local governments employed official propaganda to socially construct and persuade the public by projecting an omnipotent image of the government as a way of maintaining the public’s faith about regime capacity. Maximising support for pension reform using policy experimentation 121 The empirical results based on two rounds of survey data and local propaganda data show that the pilot policy decreased the public’s perception of governmental responsibility on elderly welfare in general. Moreover, the longer that individuals experienced the pilot policy, the greater the attitudinal change was. However, along with the pilot policy, local governments’ use of official propaganda to maintain the government’s image amid welfare provision and justify shared responsibility reduced the public’s faith in the omnipotent role of the government to a certain extent. Moreover, the disjunction of policy propaganda and policy experimentation seems to have been recognised by the public and this phenomenon led to a decrease in the perceived credibility of the governments in the long term. These conclusions align well with those in many studies focusing on the transitional role of governments in welfare reforms. Despite China’s policy experimentation and propaganda both being useful at the onset, the political attitude of the public in general later seems to have been influenced by the mixed information arising from the various approaches. As in other cases, their support for the regime may lessen unexpectedly – or even reverse – in the long term. Facing the state’s well-designed statecraft, an individual’s options are limited but not null. People are capable of identifying the potential inconsistency in policy details and the propaganda, so should the state be seen to (in a broad sense) ‘go back on its word’ this might lead to loss of public confidence. Therefore, in Chapter 5 I move beyond looking only at pension reforms to further unpack the wider state–individual power interactions, from the perspective of individuals and the possibilities they have for counter-conduct. Acknowledgements Chapter 4 received strong support from Prof. Xufeng Zhu (Tsinghua University) on data collection and valuable feedback on framing, and a related collaborated article is under review with disciplinary journals. Notes 1 The dataset does not include the official newspaper data of Shandong Province and thus is complemented by another newspaper database, that of ‘Wisenews’ (http://wisenews.wisers.net) [accessed July 2016]. 2 ‘National Bureau of Statistics’ (http://www.stats.gov.cn) [accessed July 2016]. 3 The pilot policy was conducted in three waves. The full scale of funding the individual accounts by local governments and the financial subsidies from the central government varied somewhat across these three waves. 122 Pension Policy and Governmentality in China References Anderson, K.; and Ebbinghaus, B. (2011). The Varieties of Pension Governance. Pension Privatization in Europe. Beck, H. (1997). The Origins of the Authoritarian Welfare State in Prussia: Conservatives, Bureaucracy, and the Social Question, 1815–70. University of Michigan Press. https://doi.org/10.3998%2Fmpub.13913 Beetham, D. (1991). ‘Max Weber and the legitimacy of the modern state’. Analyse & Kritik, vol. 13, issue 1, 34–45. https://doi.org/10.1515/auk-1991-0102 Béland, D. (2005). ‘Ideas and social policy: An institutionalist perspective’. Social Policy & Administration, vol. 39, issue 1, 1–18. https://doi.org/10.1111/j .1467-9515.2005.00421.x Benish, A.; Haber, H.; and Eliahou, R. (2017). ‘The regulatory welfare state in pension markets: mitigating high charges for low-income savers in the United Kingdom and Israel’. Journal of Social Policy, vol. 46, issue 2, 313–330. https://doi.org/10.1017/S0047279416000593 Bian, Y.; and Logan, J. R. (1996). ‘Market transition and the persistence of power: The changing stratification system in urban China’. American Sociological Review, 739–758. https://doi.org/10.2307/2096451 Boushey, G. (2016). ‘Targeted for diffusion? How the use and acceptance of stereotypes shape the diffusion of criminal justice policy innovations in the American states’. American Political Science Review, vol. 110, issue 1, 198–214. https://doi.org/10.1017/s0003055415000532 Cai, H.; and Treisman, D. (2006). ‘Did government decentralization cause China’s economic miracle?’ World Politics, vol. 58, issue 4, 505–535. https://doi .org/10.1353/wp.2007.0005 Chen, X.; and Shi, T. (2001). ‘Media effects on political confidence and trust in the People’s Republic of China in the post-Tiananmen period’. East Asia, vol. 19, issue 3, 84–118. https://doi.org/10.1007/s12140-001-0011-3 Cook, L. J. (1993). The Soviet Social Contract and Why It Failed: Welfare Policy and Workers’ Politics from Brezhnev to Yeltsin, vol. 86. Harvard University Press. Cook, L. J. (2013). Postcommunist Welfare States: Reform Politics in Russia and Eastern Europe. Cornell University Press. https://doi.org/10.7591% 2F9780801460098 Cox, R. H. (2001). ‘The social construction of an imperative: Why welfare reform happened in Denmark and the Netherlands but not in Germany’. World Politics, vol. 53, issue 3, 463–498. https://doi.org/10.1353/wp.2001.0008 Denisova, I.; Eller, M.; Frye, T.; and Zhuravskaya, E. (2012). ‘Everyone hates privatization, but why? Survey evidence from 28 post-communist countries’. Journal of Comparative Economics, vol. 40, issue 1, 44–61. https://doi.org /10.1016/j.jce.2011.11.001 Di Tella, R.; Galiani, S.; and Schargrodsky, E. (2012). ‘Reality versus propaganda in the formation of beliefs about privatization’. Journal of Public Economics, vol. 96, issue 5–6, 553–567. https://doi.org/10.1016/j.jpubeco.2011.11.006 Maximising support for pension reform using policy experimentation 123 Donovan, M. C. (2001). Taking Aim: Target Populations and the Wars on AIDS and Drugs. Georgetown University Press. Easton, D. (1975). ‘A re-assessment of the concept of political support’. British Journal of Political Science, vol. 5, issue 4, 435–457. https://doi.org/10.1017 /s0007123400008309 Ekman, J.; and Linde, J. (2005). ‘Communist nostalgia and the consolidation of democracy in Central and Eastern Europe’. Journal of Communist Studies and Transition Politics, vol. 21, issue 3, 354–374. https://doi .org/10.1080/13523270500183512 Fan, Y. (2015). ‘The centre decides and the local pays: Mandates and politics in local government financial management in China’. Local Government Studies, vol. 41, issue 4, 516–533. https://doi.org/10.1080/03003930.2014.9 68706 Gao, Q. (2006). ‘The social benefit system in urban China: Reforms and trends from 1988 to 2002’. Journal of East Asian Studies, vol. 6, issue 1, 31–67. https://doi.org/10.1017/s1598240800000035 Gao, Q.; Yang, S.; and Li, S. (2013). ‘The Chinese welfare state in transition: 1988–2007’. Journal of Social Policy, vol. 42, issue 4, 743–762. https://doi .org/10.1017/s0047279413000329 Goode, E.; and Ben-Yehuda, N. (1994). ‘Moral panics: Culture, politics, and social construction’. Annual Review of Sociology, vol. 20, issue 1, 149–171. https://doi.org/10.1146/annurev.so.20.080194.001053 Gough, I. (2001). ‘Globalization and regional welfare regimes: The East Asian case’. Global Social Policy, vol. 1, issue 2, 163–189. https://doi .org/10.1177/146801810100100202. OA: https://eprints.lse.ac.uk/43959 Guthrie, D. (2012). China and Globalization: The Social, Economic and Political Transformation of Chinese Society. Routledge. Hacker, J. S. (2002). The Divided Welfare State: The Battle over Public and Private Social Benefits in the United States. Cambridge University Press. https:// doi.org/10.1017%2Fcbo9780511817298 Haggard, S.; and Kaufman, R. R. (2008). Development, Democracy, and Welfare States: Latin America, East Asia, and Eastern Europe. Princeton University Press. Hall, P. A. (2001). ‘The evolution of economic policy’. Developments in French Politics, vol. 2, 172–191. Heilmann, S. (2008a). ‘From local experiments to national policy: The origins of China’s distinctive policy process’. The China Journal, vol. 59, 1–30. https://doi.org/10.1086/tcj.59.20066378 Heilmann, S. (2008b). ‘Policy experimentation in China’s economic rise’. Studies in Comparative International Development, vol. 43, issue 1, 1–26. https:// doi.org/10.1007/s12116-007-9014-4 Huang, H. (2013). ‘Signal left, turn right: Central rhetoric and local reform in China’. Political Research Quarterly, vol. 66, issue 2, 292–305. https://doi.org /10.1177/1065912912443874 Huang, H. (2015). ‘Propaganda as signaling’. Comparative Politics, vol. 47, issue 4, 419–444. https://doi.org/10.5129/001041515816103220 124 Pension Policy and Governmentality in China Huang, H. (2018). ‘The pathology of hard propaganda’. The Journal of Politics, vol. 80, issue 3, 1034–1038. https://doi.org/10.1086/696863 Hynes, B. O. D.; and Hayes, N. (2011). ‘Who benefits from early childcare subsidy design in Ireland?’ Journal of Poverty and Social Justice, vol. 19, issue 3, 277–288. https://doi.org/10.1332/175982711x597017. OA: https://arrow .tudublin.ie/cgi/viewcontent.cgi?article=1034&context=cserart Im, D.-K.; and Meng, T. (2015). ‘The policy–opinion nexus: The impact of social protection programs on welfare policy preferences in China’. International Journal of Public Opinion Research, vol. 28, issue 2, 241–268. https:// doi.org/10.1093/ijpor/edv013 Keefer, P.; and Khemani, S. (2011). Mass Media and Public Services: The Effects of Radio Access on Public Education in Benin. The World Bank. https://doi.org /10.1596%2F1813-9450-5559 Kennedy, J. J. (2009). ‘Maintaining popular support for the Chinese Communist Party: The influence of education and the state-controlled media’. Political Studies, vol. 57, issue 3, 517–536. https://doi.org/10.1111/j.1467-9248 .2008.00740.x Kornai, J. (1992). The Socialist System: The Political Economy of Communism. Oxford University Press. https://doi.org/10.1515%2F9780691228020 Lawrence, E.; Stoker, R.; and Wolman, H. (2013). ‘The effects of beneficiary targeting on public support for social policies’. Policy Studies Journal, vol. 41, issue 2, 199–216. https://doi.org/10.1111/psj.12014 Leung, J. C.; and Nann, R. C. (1995). Authority and Benevolence: Social Welfare in China. Chinese University Press. Leung, J. C.; and Wong, H. S. (1999). ‘The emergence of a community-based social assistance programme in urban China’. Social Policy & Administration, vol. 33, issue 1, 39–54. https://doi.org/10.1111/1467-9515.00130 Leung, J. C.; and Xu, Y. (2015). China’s Social Welfare: The Third Turning Point. John Wiley & Sons. Li, B.; and Zhong, Y. (2009). ‘How did China’s transitions impact people’s welfare benefits in the reform era?’ Journal of Contemporary China, vol. 18, issue 62, 813–829. https://doi.org/10.1080/10670560903174606 Li, J.; and Ge, K. (2010). ‘Review of pension insurance individual account studies’. Social Security Studies, vol. 2, 21–26. Li, Z.; and Wang, H. (2009). ‘Return rate and replacement rate of individual account in basic pension insurance’. Journal of Public Management, vol. 6, issue 4, 45–51. Lin, D.; and Ding, Y. (2007). ‘New pension policy: Replacement rate comparison of new-old pension insurance policies’. Population and Economy, vol. 1, 69–74. Lipset, S. M. (1959). ‘Some social requisites of democracy: Economic development and political legitimacy’. American Political Science Review, vol. 53, issue 1, 69–105. https://doi.org/10.2307/1951731 Lü, X. (2014). ‘Social policy and regime legitimacy: The effects of education reform in China’. American Political Science Review, vol. 108, issue 2, 423–437. https://doi.org/10.1017/s0003055414000124 Maximising support for pension reform using policy experimentation 125 Lu, X.; and Perry, E. J. (1997). Danwei: The Changing Chinese Workplace in Historical and Comparative Perspective. Me Sharpe. Mares, I.; and Carnes, M. E. (2009). ‘Social policy in developing countries’. Annual Review of Political Science, vol. 12, 93–113. https://doi.org/10.1146 /annurev.polisci.12.071207.093504 Mashaw, J. L. (2006). ‘Accountability and institutional design: Some thoughts on the grammar of governance’. Public Law Working Paper, vol. 116, 115–156. https://ssrn.com/abstract=924879 Maynard-Moody, S. W.; Musheno, M.; and Musheno, M. C. (2003). Cops, Teachers, Counselors: Stories from the Front Lines of Public Service. University of Michigan Press. https://doi.org/10.3998%2Fmpub.11924 McQuail, D. (1987). Mass Communication Theory: An Introduction. Sage Publications. Mei, C.; and Liu, Z. (2014). ‘Experiment-based policy making or conscious policy design? The case of urban housing reform in China’. Policy Sciences, vol. 47, issue 3, 321–337. https://doi.org/10.1007/s11077-013-9185-y Montinola, G.; Qian, Y.; and Weingast, B. R. (1995). ‘Federalism, Chinese style: The political basis for economic success in China’. World Politics, vol. 48, issue 1, 50–81. https://doi.org/10.1353/wp.1995.0003 Nee, V. (1996). ‘The emergence of a market society: Changing mechanisms of stratification in China’. American Journal of Sociology, vol. 101, issue 4, 908–949. https://doi.org/10.1086/230784 Perry, E. J. (2007). ‘Studying Chinese politics: Farewell to revolution?’ The China Journal, vol. 57, 1–22. https://doi.org/10.1086/tcj.57.20066239. OA: https://dash.harvard.edu/bitstream/1/11595641/1/farewell.pdf Perry, E. J. (2017). ‘Higher education and authoritarian resilience: The case of China, past and present’. Harvard-Yenching Institute Working Paper Series. OA: https://dash.harvard.edu/handle/1/30822717 Pierson, P. (1994). Dismantling the Welfare State?: Reagan, Thatcher and the Politics of Retrenchment. Cambridge University Press. https://doi.org/10.1017% 2Fcbo9780511805288 Pierson, P. (2001). ‘Post-industrial pressures on the mature welfare states’. The New Politics of the Welfare State, vol. 1, 80–105. https://doi.org/10.1093 %2F0198297564.003.0004 Polanyi, K.; and MacIver, R. M. (1944). The Great Transformation. USA: Beacon. Ringen, S.; and Ngok, K. (2017). ‘What kind of welfare state is emerging in China?’ Towards Universal Health Care in Emerging Economies, 213–237. Springer. https://doi.org/10.1057%2F978-1-137-53377-7_8 Schneider, A. L.; and Ingram, H. M. (2019). ‘Social constructions, anticipatory feedback strategies, and deceptive public policy’. Policy Studies Journal, vol. 47, issue 2, 206–236. https://doi.org/10.1111/psj.12281 Schneider, A. L.; Ingram, H.; and DeLeon, P. (2014). ‘Democratic policy design: Social construction of target populations’. In P. Sabatier and C. Weible (eds), Theories of the Policy Process, 3rd edn, 105–149. 126 Pension Policy and Governmentality in China Schneider, A.; and Ingram, H. (1993). ‘Social construction of target populations: Implications for politics and policy’. American Political Science Review, vol. 87, issue 2, 334–347. https://doi.org/10.2307/2939044 Schneider, A.; and Sidney, M. (2009). ‘What is next for policy design and social construction theory? 1’. Policy Studies Journal, vol. 37, issue 1, 103–119. https://doi.org/10.1111/j.1541-0072.2008.00298.x Shi, S. J.; and Mok, K. H. (2012). ‘Pension privatisation in Greater China: Institutional patterns and policy outcomes’. International Journal of Social Welfare, vol. 21, S30–S45. https://doi.org/10.1111/j.1468-2397.2012.00875.x; https://doi.org/10.1111/j.1468-2397.2012.00875.x Shi, T. (2014). The Cultural Logic of Politics in Mainland China and Taiwan. Cambridge University Press. https://doi.org/10.1017/CBO9780511996474 Shirk, S. L. (2011). Changing Media, Changing China. Oxford University Press. Siebert, F. S.; Peterson, T.; and Schramm, W. (1956). Four Theories of the Press: The Authoritarian, Libertarian, Social Responsibility, and Soviet Communist Concepts of What the Press Should Be and Do. University of Illinois Press. Song, S.; and Chu, G. S. F. (1997). ‘Social security reform in China: The case of old-age insurance’. Contemporary Economic Policy, vol. 15, issue 2, 85–93. https://doi.org/10.1111/j.1465-7287.1997.tb00468.x State Council. (1997). Decision on Establishing a Unified System of Basic Pension Insurance for Enterprise Employees (State Council [1997] No.26). Retrieved from http://www.gov.cn/ztzl/nmg/content_412509.htm. State Council. (2000a). Opinions on Accelerating the Process of Social Welfare Socialization. Retrieved from http://shfl.mca.gov.cn/article/zcfg/200809 /20080900019761.shtml. State Council. (2000b). The Notice of Issuing the Pilot Program of Urban Social Security System (State Council [2000] No.42). Retrieved from http://www .gov.cn/xxgk/pub/govpublic/mrlm/201011/t20101112_62507.html. State Council. (2005). Decision on Perfecting Basic System of Pension Insurance for Enterprise Employees (State Council [2005] No.38). Retrieved from http://www.gov.cn/zwgk/2005-12/14/content_127311.htm. Szelenyi, I. (1978). ‘Social inequalities in state socialist redistributive economies’. International Journal of Comparative Sociology, vol. 19, issue 1–2, 63–87. https://doi.org/10.1177/002071527801900105 Szelenyi, I.; and Kostello, E. (1996). ‘The market transition debate: Toward a synthesis?’ American Journal of Sociology, vol. 101, issue 4, 1082–1096. https://doi.org/10.1086/230791 Szelenyi, I.; and Szelenyi, B. (1994). ‘Why socialism failed: Toward a theory of system breakdown—Causes of disintegration of East European state socialism’. Theory and Society, vol. 23, issue 2, 211–231. https://doi.org/10.1007 /bf00993815 Weingast, B. R. (1995). ‘The economic role of political institutions: Marketpreserving federalism and economic development’. Journal of Law, Economics, & Organization, 1–31. https://doi.org/10.1093/oxfordjournals.jleo .a036861 Maximising support for pension reform using policy experimentation 127 Wong, L. (2005). Marginalization and Social Welfare in China. Routledge. https://doi.org/10.4324%2F9780203982990 Wong, L.; and Ngok, K. (2006). ‘Social policy between plan and market: Xiagang (off‐duty employment) and the policy of the re‐employment service centres in China’. Social Policy & Administration, vol. 40, issue 2, 158–173. https://doi.org/10.1111/j.1467-9515.2006.00482.x Xie, Y. (2016). ‘Understanding inequality in China’. Chinese Journal of Sociology, vol. 2, issue 3, 327–347. https://doi.org/10.1177/2057150x16654059. OA: https://europepmc.org/article/MED/29854420 Yang, Y.; Wang, W.; and Zhang, M. (2010). ‘Operational safety of social insurance fund: Experience from fully funding the individual accounts’. Chinese Public Administration, vol. 5, 61–66. Zheng, G. (2016). Evaluation of China’s Social Protection Policies. Retrieved from https://www.euchinasprp.eu/images/documents/Component1Cn/2017 -assessment-eport/EvalSSCn.pdf Zhu, X.; and Zhao, H. (2018a). ‘Experimentalist governance with interactive central–local relations: Making new pension policies in China’. Policy Studies Journal. https://doi.org/10.1111/psj.12254 Zhu, X.; and Zhao, H. (2018b). ‘Recognition of innovation and diffusion of welfare policy: Alleviating urban poverty in Chinese cities during fiscal recentralization’. Governance, vol. 31, issue 4, 721–739. https://doi.org/10.1111 /gove.12332 CHAPTER 5 Falsification of ‘manufactured compliance’ and wider legitimation and governmentality issues Although China is experiencing tremendous social and economic changes, the Chinese government reports considerable public endorsement from many kinds of survey results – all of which suggest people’s praise of the current authority together with a strong sense of belonging and solidarity. Of course, the government uses many tactics to manufacture public consent and minimise the possibility of challenge from the governed with the case of pension reforms, as earlier chapters have shown. However, people can also reflect on and reshape the idea of the ‘state’, ‘politics’, and the ‘state–individual relationship’ in their interactions with governmental power. In other words, despite the well-designed statecraft of the government, there are risks for the authorities of falsified compliance/consent from the people. These individual-level reflections and (undisclosed) second thoughts – or ideological rebellions or modifications – play an important role in shaping the long-term expectations and superficial compliance of the general people. Many studies have highlighted the constant bubbling up of collective actions from the bottom of society in China, which obviously contradict the orderly scene on the surface. Chapter 4 also showed empirically that, while in the short term people may express contentment with a controversial policy, in the long term there is a decline in political support as the real impact of the policy on their everyday lives fails to match with the official propaganda. This situation makes China an interesting field for studying the possibilities of falsified compliance and its implications for state governmentality. Explaining why people do not do something is always complex and needs to take account of several different factors. I begin by considering the idea of ‘falsification’ by individuals of their views and what methods might help in detecting it. The second section looks at different aspects of people’s compliance, How to cite this book chapter: Wang, Yan. 2022. Pension Policy and Governmentality in China: Manufacturing Public Compliance. London: LSE Press, pp. 129–179. DOI: https://doi.org/10.31389/lsepress.ppc.f. License: CC BY 232 Pension Policy and Governmentality in China Table A.13: Case list Case Age range Gender Ethnicity Education level Occupation Place Non/local Interview time 1 Post 1985 Female Han Postgraduate Student Shanghai Non-local 2019.6 2 Post 1985 Female Han College Government Chengdu Local 2019.6 3 Post 1980 Female Minority College Public sector Shanghai Non-local 2019.6 4 Post 1980 Female Han College Public sector Nanjing Non-local 2019.6 5 Post 1980 Male Han College SOE; government Beijing Non-local 2019.6 6 Post 1990 Female Minority Postgraduate Private enterprise Beijing Non-local 2018.11 7 Post 1995 Male Han High school Private enterprise Suzhou Non-local 2019.6 8 Post 1995 Male Han Postgraduate Private enterprise Beijing Non-local 2019.6 9 Post 1980 Male Han Postgraduate SOE Beijing Non-local 2019.6 10 Post 1985 Female Han Postgraduate New media Beijing Non-local 2019.6 11 Post 1985 Female Han Postgraduate Foreign enterprise Beijing Non-local 2019.6 12 Post 1950 Male Han College Retired (private sector) Shanghai Local 2019.6 13 Post 1980 Male Han College Commercial media Beijing Non-local 2019.6 14 Post 1990 Male Han Postgraduate Student London Non-local 2019.6 Appendix A. Data explanation and model validations 233 15 Post 2000 Female Han High school Student Xi’an Local 2019.6 16 Post 1950 Male Han Junior high school Retired (SOE) Tianjin Local 2018.11 17 Post 1950 Female Han High school Retired (SOE) Tianjin Local 2018.11 18 Post 1950 Male Han College Retired (public sector) Jilin Local 2018.11 19 Post 1970 Male Han PhD Higher education Beijing Non-local 2018.12 20 Post 1970 Male Han PhD Higher education Beijing Non-local 2018.12 21 Post 1960 Female Han High school Public sector Yuncheng Local 2018.11 22 Post 1960 Male Han College Government Yuncheng Local 2018.11 23 Post 1990 Female Han College Finance sector Beijing Non-local 2018.12 24 Post 1990 Male Han College Unemployed Beijing Non-local 2018.12 25 Post 1990 Female Han Postgraduate SOE Beijing Non-local Observation 26 Post 1960 Male Han PhD Higher education Overseas Non-local Observation 27 Post 1980 Female Han PhD Higher education Overseas Non-local 2019.6 Note: The list is for both the interviews and some of the observation cases. Appendix B. Data replication codebooks Appendix B presents a brief explanation of the statistical methods used in the empirical chapters and instructions on the replication file for Chapters 3 and 4. All documents are available upon request. B1. Text analysis (Chapter 3) As explained in the main content, I use quantitative text analysis (QTA) to show the persuasive efforts of the government when intentionally producing truth and knowledge about pension benefits, social fairness, and the allocation of responsibility. The QTA part of my research included two subsections: an unsupervised structural topic model to identify the main topics in the text relating to pension reforms and supervised methods to label each document with categories that were of interest to my research. The main software used here was Rstudio (version 1.1.447) and the main QTA packages including Rwordseg, tmcn, stm, topicmodel (all available on https://cran.r-project.org). Helpful literature for further understanding the method can be found in work of Hopkins and King (2010), Grimmer and Stewart (2013), Lucas et al. (2015), and Roberts, Stewart, and Tingley (2014). I provided code and data files necessary to replicate some of the main results reported in Chapter 3. Owing to the large amount of original data, I share the data-term-matrix and metadata. Since text segment and software simulation 236 Pension Policy and Governmentality in China can be different each time, it may not be possible to replicate the exact results reported in the chapter, but, as long as the replication uses the same version of the packages, the results should be similar. It should be noted that the original data are in Chinese, and therefore the direct results (such as the generated topics) are also in Chinese. Table B.1: Codebook for Chapter 3 Variable name Variable label and measurement X Document number Title Title of the document Author Author of the document Date Date of the document Year Year Month Month Day Day Year_Month Year-month Edition_name Edition name of the document Column Column of the document Text Full text of the document Responsibility Identified welfare responsibility of the document (categorical variable), ‘no clear direction of locus’=0, ‘state/party’=1, ‘individual/family’=2, ‘enterprise’=3, ‘social coordination’=4 National.condition Whether the document emphasis national condition (Yes=1) Foreign.experience Whether the document emphasis foreign experience (Yes=1) Praise Whether the sentiment is positive (Yes=1) Denounce Whether the sentiment is negative (Yes=1) Appendix B. Data replication codebooks 237 B2. Causal inference (Chapter 4) In Chapter 4, I used difference-in-differences when examining the causal relationship between the trajectory of welfare reform from state socialism to shared responsibility and individual changes of perception. The idea of causal inference here is that the causal effect of a treatment on some outcome can be assessed. The causal effect of the treatment on the outcome for unit i is the difference between its two potential outcomes: an observed outcome in a treated situation and an unobserved outcome in a control situation (i.e. something counterfactual). However, in practice, it is impossible to observe the values of Y1i and Y0i on the same unit i. Therefore, in empirical studies, scholars rely on statistical solutions to the fundamental problem of causal inference (Winship and Morgan 1999). Commonly used approaches include randomised experiments (e.g. Fisher 1960; Gerber and Green 2012), matching and weighting on covariates (e.g. Hainmueller 2012; Iacus, King, and Porro 2012; Morgan and Harding 2006), difference-in-differences (e.g. Ashenfelter 1978), a synthetic control method (Abadie and Gardeazabal 2003), instrumental variables (e.g. Angrist, Imbens, and Rubin 1996), causal mediation analysis (Imai, Keele, and Yamamoto 2010), and regression discontinuity design (Imbens and Lemieux 2008; Thistlethwaite and Campbell 1960). The difference-in-differences method was developed from the ‘simple difference estimator’, which used the post-treatment outcome (Y11) minus the pre-treatment outcome (Y01) for the treatment group. However, owing to the existence of the time factor, the Y11−Y01 estimator is not actually the treatment effect. Therefore, the DID approach takes advantage of the control group, which did not receive any treatment of intervention, such as a policy/programme. If the assumption of a parallel trend holds, the estimator (ytreat, after−ytreat,before)−(ycontrol,after−ycontrol,before) can be treated as the treatment effect of itervention. We signed a confidentiality agreement to use the survey data (with Martin Whyte) and therefore cannot publish the original data used for Chapter 4. Instead, the code used to generate the results in the chapter is available upon request. The relevant codebook (individual level and provincial level) is presented in Table B.2. 238 Pension Policy and Governmentality in China Variable name Variable label and measurement Urban Hukou registration (Urban=1) Gender Gender (Male=1) Age Age Minority Minority ethnic group (Minority=1) CCP_member Member of CCP (Member=1) Edu_year Years of education Age_grp Age group Edu_grp Education levels Income_log Annual income of the household (log value) Unit_Enterprise Occupational variation (Enterprise employees=1) Unit_Governmental Employee Occupational variation (Governmental employees=1) Old-age dependency (oldcare_n) Numbers of old-age person the family need to take care Treated Provinces being treated or controlled (treated=1) Duration Years each province being treated Wave Provinces in different waves of experiment Post Survey year (year 2009=1) Province_n Province code, 22 provinces in total Respon_old Locus of responsibility regarding old-age caring (attitudinal preference, individual responsibility=5) Respon_med Locus of responsibility regarding health caring (attitudinal preference, individual responsibility=5) Respon_edu Locus of responsibility regarding compulsory education (attitudinal preference, individual responsibility=5) Respon_job Locus of responsibility regarding employment (attitudinal preference, individual responsibility=5) Trust_central Political trust for central government (most trust=4) Trust_province Political trust for provincial government (most trust=4) Trust_local Political trust for prefectural/city government (most trust=4) Province name Province name province_n Province code Year_Prov Year-province panel All news All article number Table B.2: Codebook for Chapter 4 Appendix B. Data replication codebooks 239 (Continued) Variable name Variable label and measurement Related_news Identified article number relates to pension insurance/ old-age social insurance Related_news_ratio Ratio of Identified article number relates to pension insurance/old-age social insurance compare to all articles Accumulated Related_ news_ratio (3 years) Related News (2002+2003+2004)/ All News (2002+2003+2004) Related News (2007+2008+2009)/ All News (2007+2008+2009) Accumulated Related_ news_ratio (5 years) Related news (2000+2001+2002+2003+2004)/All News (2000+2001+2002+2003+2004) Related news (2005+2006+2007+2008+2009)/All News (2005+2006+2007+2008+2009) Directly related news Identified article number relates to fully funding the individual accounts pilot policy Directly related news_ratio Ratio of Identified article relates to fully funding the individual accounts pilot policy compare to all articles Accumulated Directly Related_news_ratio (3 years) Directly related news (2002+2003+2004)/ All news (2002+2003+2004) Directly related news (2007+2008+2009)/ All news (2007+2008+2009) Accumulated Directly Related_news_ratio (5 years) Directly related news (2000+2001+2002+2003+2004)/ All News (2000+2001+2002+2003+2004) Directly related news (2005+2006+2007+2008+2009)/ All News (2005+2006+2007+2008+2009) Wave Wave of pilot policy Municipal_n Number of municipalities of certain province County_n Number of counties of certain province GDP_total Province GDP (yuan) GDP_per capita Province GDP per capita (yuan) GDP_index Province GDP Index (previous year=100) Residents_n Yearend population of certain province (10,000) Urban_residents Urban population of certain province (10,000) Rural_residents Rural population of certain province (10,000) Population growth rate Population growth rate (‰) Population Population (census) Old_population Population of people above 65 (census) Table B.2: (Continued) 240 Pension Policy and Governmentality in China Variable name Variable label and measurement ODR_cens Old-age dependency rate (census) Urban_employees Urban employee population (10,000) Fin_budgetary revenue Local fiscal general budgetary revenue (hundred million) Fin_revenue Local fiscal revenue (hundred million) Fin_specific purpose revenue Local fiscal specific purpose revenue (hundred million) Fin_non-tax revenue Local non-tax revenue (hundred million) Fin_general budgetary expenditure Local fiscal general budgetary expenditure (hundred million) Fin_general public service expenditure Local fiscal general public service expenditure (hundred million) Fin_social security and employment expenditure Local fiscal social security and employment expenditure (hundred million) Fin_health and medical expenditure Local fiscal health and medical expenditure (hundred million) Saving_deposit Saving deposit (hundred million) Ur_employee participants Urban employee participants for old-age social insurance (10,000) Ur_in service employee participants Urban in service employee participants for old-age social insurance (10,000) Ur_retired employee participants Urban retired employee participants for Old-age social insurance (10,000) Pension_revenue Basic pension insurance funds revenue (10,000 yuan) Pension_expen Basic pension insurance funds expenditure (10,000 yuan) Pension_balance Basic pension insurance funds balance (10,000 yuan) Table B.2: (Continued) Bibliography Acemoglu, D.; Ticchi, D.; and Vindigni, A. (2010). ‘A theory of military dictatorships’. American Economic Journal: Macroeconomics, vol. 2, issue 1, 1–42. https://doi.org/10.1257/mac.2.1.1. OA: https://dspace.mit.edu/bitstream /1721.1/61747/2/Acemoglu_A%20theory.pdf Anderson, K.; and Ebbinghaus, B. (2011). The Varieties of Pension Governance. Pension Privatization in Europe. Asch, S. E.; and Guetzkow, H. (1951). ‘Effects of group pressure upon the modification and distortion of judgments’. Documents of Gestalt Psychology, 222–236. Bank, C. M. (2017). China Minsheng Bank Report. Barrientos, A. (2009). ‘Labour markets and the (hyphenated) welfare regime in Latin America’. Economy and Society, vol. 38, issue 1, 87–108. https://doi.org /10.1080/03085140802560553 Barrientos, A. (2013). Social Assistance in Developing Countries. Cambridge University Press. Beck, H. (1997). The Origins of the Authoritarian Welfare State in Prussia: Conservatives, Bureaucracy, and the Social Question, 1815–70. University of Michigan Press. https://doi.org/10.3998%2Fmpub.13913 Beetham, D. (1991). ‘Max Weber and the legitimacy of the modern state’. Analyse & Kritik, vol. 13, issue 1, 34–45. https://doi.org/10.1515/auk-1991-0102 Beetham, D. (1995). ‘What future for economic and social rights?’ Political Studies, vol. 43, issue 1, 41–60. https://doi.org/10.1111/j.1467-9248.1995 .tb01735.x 248 Pension Policy and Governmentality in China Hall, P. A. (2001). ‘The evolution of economic policy’. Developments in French Politics, vol. 2, 172–191. Han, R. (2015). ‘Defending the authoritarian regime online: China’s “voluntary fifty-cent army”’. The China Quarterly, vol. 224, 1006–1025. https://doi .org/10.1017/s0305741015001216 Heilmann, S. (2008a). ‘From local experiments to national policy: The origins of China’s distinctive policy process’. The China Journal, vol. 59, 1–30. https://doi.org/10.1086/tcj.59.20066378 Heilmann, S. (2008b). ‘Policy experimentation in China’s economic rise’. Studies in Comparative International Development, vol. 43, issue 1, 1–26. https:// doi.org/10.1007/s12116-007-9014-4 Hinings, C. R.; Tolbert, P. S.; Greenwood, R.; and Oliver, C. (2008). ‘Emerging inequalities in central and Eastern Europe’. The Sage Handbook of Organizational Institutionalism, 473–492. UK: Sage. https://doi.org /10.4135%2F9781849200387.n20 Hirschman, A. O. (1970). Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States, vol. 25. Harvard University Press. Holzmann, R.; and Palmer, E. E. (eds) (2006). Pension Reform: Issues and Prospects for Non-financial Defined Contribution (NDC) Schemes. World Bank Publications. https://doi.org/10.1596%2F978-0-8213-6038-5 Hopkins, D. J.; and King, G. (2010). ‘A method of automated nonparametric content analysis for social science’. American Journal of Political Science, vol. 54, issue 1, 229–247. https://doi.org/10.1111/j.1540-5907.2009.00428.x. OA: http://nrs.harvard.edu/urn-3:HUL.InstRepos:5125261 Howard, R. (1983). ‘The full-belly thesis: Should economic rights take priority over civil and political rights-evidence from Sub-Saharan Africa’. Human Rights Quarterly, vol. 5, 467. https://doi.org/10.2307/762231 Huang, H. (2013). ‘Signal left, turn right: Central rhetoric and local reform in China’. Political Research Quarterly, vol. 66, issue 2, 292–305. https://doi.org /10.1177/1065912912443874 Huang, H. (2015). ‘Propaganda as signaling’. Comparative Politics, vol. 47, issue 4, 419–444. https://doi.org/10.5129/001041515816103220 Huang, H. (2017). ‘A war of (mis) information: The political effects of rumors and rumor rebuttals in an authoritarian country’. British Journal of Political Science, vol. 47, issue 2, 283–311. https://doi.org/10.1017/s0007123 415000253 Huang, H. (2018). ‘The pathology of hard propaganda’. The Journal of Politics, vol. 80, issue 3, 1034–1038. https://doi.org/10.1086/696863 Huang, P. C. (1990). The Peasant Family and Rural Development in the Yangzi Delta, 1350-1988. Stanford University Press. Huang, P. C. (2012). ‘Profit-making state firms and China’s development experience: “State capitalism” or “socialist market economy”?’ Modern China, vol. 38, issue 6, 591–629. https://doi.org/10.1177/0097700412455839 Bibliography 249 Huang, X. (2014). Social Protection under Authoritarianism: Politics and Policy of Social Health Insurance in China. Columbia University. Huang, X. (2020). Social Protection under Authoritarianism: Health Politics and Policy in China. Oxford University Press. https://doi.org/10.1093%2Foso %2F9780190073640.001.0001 Huntington, S. P. (2006). Political Order in Changing Societies. Yale University Press. Hynes, B. O. D.; and Hayes, N. (2011). ‘Who benefits from early childcare subsidy design in Ireland?’ Journal of Poverty and Social Justice, vol. 19, issue 3, 277–288. https://doi.org/10.1332/175982711x597017. OA: https://arrow .tudublin.ie/cgi/viewcontent.cgi?article=1034&context=cserart Iacus, S. M.; King, G.; and Porro, G. (2012). ‘Causal inference without balance checking: Coarsened exact matching’. Political Analysis, vol. 20, issue 1, 1–24. https://doi.org/10.1093/pan/mpr013 Im, D.-K.; and Meng, T. (2015). ‘The policy–opinion nexus: The impact of social protection programs on welfare policy preferences in China’. International Journal of Public Opinion Research, vol. 28, issue 2, 241–268. https:// doi.org/10.1093/ijpor/edv013 Imai, K.; Keele, L.; Tingley, D.; and Yamamoto, T. (2011). ‘Unpacking the black box of causality: Learning about causal mechanisms from experimental and observational studies’. American Political Science Review, vol. 105, issue 4, 765–789. https://doi.org/10.1017/s0003055411000414. OA: http://hdl.handle .net/1721.1/84065 Imai, K.; Keele, L.; and Yamamoto, T. (2010). ‘Identification, inference and sensitivity analysis for causal mediation effects’. Statistical Science, 51–71. https://doi.org/10.1214/10-sts321 Imbens, G. W.; and Lemieux, T. (2008). ‘Regression discontinuity designs: A guide to practice’. Journal of Econometrics, vol. 142, issue 2, 615–635. https:// doi.org/10.1016/j.jeconom.2007.05.001 Inglehart, R. (2018). Culture Shift in Advanced Industrial Society. Princeton University Press. https://doi.org/10.2307%2Fj.ctv346rbz Jefferson, G. H.; and Rawski, T. G. (1994). ‘Enterprise reform in Chinese industry’. Journal of Economic Perspectives, vol. 8, issue 2, 47–70. https://doi.org /10.1257/jep.8.2.47 Jennings, M. K. (1996). ‘Political knowledge over time and across generations’. Public Opinion Quarterly, vol. 60, issue 2, 228–252. https://doi.org /10.1086/297749 Jiang, J.; and Yang, D. L. (2016). ‘Lying or believing? Measuring preference falsification from a political purge in China’. Comparative Political Studies, vol. 49, issue 5, 600–634. https://doi.org/10.1177/0010414015626450 Kamenica, E.; and Gentzkow, M. (2011). ‘Bayesian persuasion’. American Economic Review, vol. 101, issue 6, 2590–2615. https://doi.org/10.1257/aer .101.6.2590 250 Pension Policy and Governmentality in China Karshenas, M.; and Moghadam, V. M. (2006). Social Policy in the Middle East: Economic, Political and Gender Dynamics. UK: Palgrave Macmillan. Keefer, P.; and Khemani, S. (2011). Mass Media and Public Services: The Effects of Radio Access on Public Education in Benin. The World Bank. https://doi .org/10.1596%2F1813-9450-5559 Kelly, D. (2006). ‘Citizen movements and China’s public intellectuals in the Hu-Wen era’. Pacific Affairs, vol. 79, issue 2, 183–204. https://doi.org /10.5509/2006792183 Kennedy, J. J. (2009). ‘Maintaining popular support for the Chinese Communist Party: The influence of education and the state-controlled media’. Political Studies, vol. 57, issue 3, 517–536. https://doi.org/10.1111/j.1467-9248 .2008.00740.x Kohli, A.; Shue, V.; and Migdal, J. S. (1994). State Power and Social Forces: Domination and Transformation in the Third World. UK: Cambridge University Press. Kornai, J. (1992). The Socialist System: The Political Economy of Communism. Oxford University Press. https://doi.org/10.1515%2F9780691228020 Kricheli, R.; Livne, Y.; and Magaloni, B. (2011). Taking to the Streets: Theory and Evidence on Protests under Authoritarianism. Paper presented at the APSA 2010 Annual Meeting Paper. Kuran, T. (1991). ‘Now out of never: The element of surprise in the East European revolution of 1989’. World Politics, vol. 44, issue 1, 7–48. https://doi.org /10.2307/2010422 Kuran, T. (1997). Private Truths, Public Lies: The Social Consequences of Preference Falsification. Harvard University Press. Lane, C. (1984). ‘Legitimacy and power in the Soviet Union through socialist r i t u a l’. British Journal of Political Science, vol. 14, issue 2, 207–217. https://doi .org/10.1017/s0007123400003537 Lane, D. (2007). ‘Post-state socialism: A diversity of capitalisms?’ Varieties of Capitalism in Post-communist Countries, 13–39. Springer. https://doi .org/10.1057%2F9780230627574_2 Lawrence, E.; Stoker, R.; and Wolman, H. (2013). ‘The effects of beneficiary targeting on public support for social policies’. Policy Studies Journal, vol. 41, issue 2, 199–216. https://doi.org/10.1111/psj.12014 Le Bon, G. (1897). The Crowd: A Study of the Popular Mind. Fischer. Lee, M.-k. (2000). Chinese Occupational Welfare in Market Transition. Springer. https://doi.org/10.1057/9780333982549 Lessenich, S. (2010). ‘Constructing the socialized self: Mobilization and control in the “active society”. Governmentality, 312–328. Routledge. Leung, J. C.; and Nann, R. C. (1995). Authority and Benevolence: Social Welfare in China. Chinese University Press. Leung, J. C.; and Wong, H. S. (1999). ‘The emergence of a community-based social assistance programme in urban China’. Social Policy & Administration, vol. 33, issue 1, 39–54. https://doi.org/10.1111/1467-9515.00130 Bibliography 251 Leung, J. C.; and Xu, Y. (2015). China’s Social Welfare: The Third Turning Point. John Wiley & Sons. Levi, M. (1997). Consent, Dissent, and Patriotism. Cambridge University Press. Li, B.; and Zhong, Y. (2009). ‘How did China’s transitions impact people’s welfare benefits in the reform era?’ Journal of Contemporary China, vol. 18, issue 62, 813–829. https://doi.org/10.1080/10670560903174606 Li, J. (2019). Package ‘tmcn’. Li, J.; and Ge, K. (2010). ‘Review of pension insurance individual account studies’. Social Security Studies, vol. 2, 21–26. Li, L. (2010). ‘Rights consciousness and rules consciousness in contemporary China’. The China Journal, vol. 64, 47–68. https://doi.org/10.1086/tcj .64.20749246 Li, Z.; and Wang, H. (2009). ‘Return rate and replacement rate of individual account in basic pension insurance’. Journal of Public Management, vol. 6, issue 4, 45–51. Lin, D.; and Ding, Y. (2007). ‘New pension policy: Replacement rate comparison of new-old pension insurance policies’. Population and Economy, vol. 1, 69–74. Lipset, S. M. (1959). ‘Some social requisites of democracy: Economic development and political legitimacy’. American Political Science Review, vol. 53, issue 1, 69–105. https://doi.org/10.2307/1951731 Lipset, S. M.; and Man, P. (1960). The Social Bases of Politics. USA: Johns Hopkins University Press. Liu, T.; and Sun, L. (2016). ‘Pension reform in China’. Journal of Aging & Social Policy, vol. 28, issue 1, 15–28. https://doi.org/10.1080/08959420.2016.1111725 Logvinenko, I. (2020). ‘Authoritarian welfare state, regime stability, and the 2018 pension reform in Russia’. Communist and Post-Communist Studies, vol. 53, issue 1, 100–116. https://doi.org/10.1525/cpcs.2020.53.1.100 Lohmann, S. (1993). ‘A signaling model of informative and manipulative political action’. American Political Science Review, vol. 87, issue 2, 319–333. https://doi.org/10.2307/2939043 Lohmann, S. (1994). ‘The dynamics of informational cascades: The Monday demonstrations in Leipzig, East Germany, 1989–91’. World Politics, vol. 47, issue 1, 42–101. https://doi.org/10.2307/2950679 Lord, C.; and Beetham, D. (2001). ‘Legitimizing the EU: Is there a post-parliamentary basis for its legitimation?’ JCMS: Journal of Common Market Studies, vol. 39, issue 3, 443–462. https://doi.org/10.1111/1468-5965.00298 Lorentzen, P. (2014). ‘China’s strategic censorship’. American Journal of Political Science, vol. 58, issue 2, 402–414. https://doi.org/10.1111/ajps.12065 Lü, X. (2014). ‘Social policy and regime legitimacy: The effects of education reform in China’. American Political Science Review, vol. 108, issue 2, 423–437. https://doi.org/10.1017/s0003055414000124 Lu, X.; and Perry, E. J. (1997). Danwei: The Changing Chinese Workplace in Historical and Comparative Perspective. Me Sharpe. 252 Pension Policy and Governmentality in China Lucas, C.; Nielsen, R. A.; Roberts, M. E.; Stewart, B. M.; Storer, A.; and Tingley, D. (2015). ‘Computer-assisted text analysis for comparative politics’. Political Analysis, vol. 23, issue 2, 254–277. https://doi.org/10.1093/pan/mpu019. OA: https://dash.harvard.edu/handle/1/38057808 Lust-Okar, E. (2006). ‘Elections under authoritarianism: Preliminary lessons from Jordan’. Democratization, vol. 13, issue 3, 456–471. https://doi.org /10.1080/13510340600579359 Madrid, R. (2002). The politics and economics of pension privatization in Latin America. Latin American Research Review, 159–182. https://www.jstor.org /stable/2692153 Magaloni, B. (2006). Voting for Autocracy: Hegemonic Party Survival and Its Demise in Mexico, vol. 296. UK: Cambridge University Press. https://doi.org /10.1017/CBO9780511510274 Mahoney, J.; and Rueschemeyer, D. (2003). Comparative Historical Analysis in the Social Sciences. Cambridge University Press. https://doi.org/10.1017 /CBO9780511803963 Mares, I.; and Carnes, M. E. (2009). ‘Social policy in developing countries’. Annual Review of Political Science, vol. 12, 93–113. https://doi.org/10.1146 /annurev.polisci.12.071207.093504 Marshall, T. H. (1964). ‘Class, citizenship and social development’. New York, vol. 19642. Mashaw, J. L. (2006). ‘Accountability and institutional design: Some thoughts on the grammar of governance’. Public Law Working Paper, vol. 116, 115–156. https://ssrn.com/abstract=924879 Mattingly, Daniel C. (2019). The Art of Political Control in China. Cambridge University Press. Maynard-Moody, S. W.; Musheno, M.; and Musheno, M. C. (2003). Cops, Teachers, Counselors: Stories from the Front Lines of Public Service. University of Michigan Press. https://doi.org/10.3998%2Fmpub.11924 McCarthy, J. D.; and Zald, M. N. (1977). ‘Resource mobilization and social movements: A partial theory’. American Journal of Sociology, vol. 82, issue 6, 1212–1241. https://doi.org/10.1086/226464 McMillan, J.; and Zoido, P. (2004). ‘How to subvert democracy: Montesinos in Peru’. Journal of Economic Perspectives, vol. 18, issue 4, 69–92. https://doi.org /10.1257/0895330042632690 McQuail, D. (1987). Mass Communication Theory: An Introduction. Sage Publications. Mei, C.; and Liu, Z. (2014). ‘Experiment-based policy making or conscious policy design? The case of urban housing reform in China’. Policy Sciences, vol. 47, issue 3, 321–337. https://doi.org/10.1007/s11077-013-9185-y Meng, K. (2018). China’s Pension Reforms: Political Institutions, Skill Formation and Pension Policy in China. Routledge. https://doi.org/10.4324% 2F9781351061667 Bibliography 253 Meyer, D. S.; and Staggenborg, S. (1996). ‘Movements, countermovements, and the structure of political opportunity’. American Journal of Sociology, vol. 101, issue 6, 1628–1660. https://doi.org/10.1086/230869 Montinola, G.; Qian, Y.; and Weingast, B. R. (1995). ‘Federalism, Chinese style: The political basis for economic success in China’. World Politics, vol. 48, issue 1, 50–81. https://doi.org/10.1353/wp.1995.0003 Moore, B. (1966). Social Origins of Democracy and Dictatorship. USA: Beacon. Morgan, S. L.; and Harding, D. J. (2006). ‘Matching estimators of causal effects: Prospects and pitfalls in theory and practice’. Sociological Methods & Research, vol. 35, issue 1, 3–60. https://doi.org/10.1177/0049124106289164 Munro, N. (2006). ‘Russia’s persistent communist legacy: Nostalgia, reaction, and reactionary expectations’. Post-Soviet Affairs, vol. 22, issue 4, 289–313. https://doi.org/10.2747/1060-586x.22.4.289 Myerson, R. B. (2008). ‘The autocrat’s credibility problem and foundations of the constitutional state’. American Political Science Review, vol. 102, issue 1, 125–139. https://doi.org/10.1017/s0003055408080076 Nathan, A. (2003). ‘China’s changing of the guard: Authoritarian resilience’. Journal of Democracy, vol. 14, issue 1, 6–17. https://doi.org/10.1353/jod .2003.0019 Naughton, B.; and Tsai, K. S. (2015). State Capitalism, Institutional Adaptation, and the Chinese Miracle. Cambridge University Press. https://doi.org /10.1017%2Fcbo9781139962858 Nederhof, A. J. (1985). ‘Methods of coping with social desirability bias: A review’. European Journal of Social Psychology, vol. 15, issue 3, 263–280. https://doi.org/10.1002/ejsp.2420150303 Nee, V. (1996). ‘The emergence of a market society: Changing mechanisms of stratification in China’. American Journal of Sociology, vol. 101, issue 4, 908–949. https://doi.org/10.1086/230784 Ngok, K. (2013). ‘Shaping social policy in the reform era in China’. Handbook on East Asian Social Policy, 105–128. https://doi.org/10.4337 /9780857930293.00011 Noelle-Neumann, E. (1993). The Spiral of Silence: Public Opinion, Our Social Skin. University of Chicago Press. OECD. (2005). Organization for Economic Cooperation and Development (OECD) Annual Report. Retrieved from https://www.oecd.org/about /34711139.pdf Offe, C. (1987). ‘II. Democracy against the welfare state? Structural foundations of neoconservative political opportunities’. Political Theory, vol. 15, issue 4, 501–537. https://doi.org/10.1177/0090591787015004002 O’Gorman, H. J. (1986). ‘The discovery of pluralistic ignorance: An ironic lesson’. Journal of the History of the Behavioral Sciences, vol. 22, issue 4, 333–347. 254 Pension Policy and Governmentality in China Ortiz, I.; Duran, F.; Urban, S.; Wodsak, V.; and Yu, Z. (2018). Reversing Pension Privatization: Rebuilding Public Pension Systems in Eastern European and Latin American Countries (2000-18). Available at SSRN 3275228. Pan, J. (2020). Welfare for Autocrats: How Social Assistance in China Cares for Its Rulers. USA: Oxford University Press. https://doi.org/10.1093%2Foso %2F9780190087425.001.0001 Pepinsky, T. (2007). ‘Autocracy, elections, and fiscal policy: evidence from Malaysia’. Studies in Comparative International Development, vol. 42, issue 1–2), 136–163. https://doi.org/10.1007/s12116-007-9006-4 Perry, E. J. (2007). ‘Studying Chinese politics: Farewell to revolution?’ The China Journal, vol. 57, 1–22. https://doi.org/10.1086/tcj.57.20066239. OA: https://dash.harvard.edu/bitstream/1/11595641/1/farewell.pdf Perry, E. J. (2017). ‘Higher education and authoritarian resilience: The case of China, past and present’. Harvard-Yenching Institute Working Paper Series. OA: https://dash.harvard.edu/handle/1/30822717 Perry, E. J.; and Heilmann, S. (2011). ‘Embracing uncertainty: Guerrilla policy style and adaptive governance in China’. Mao’s Invisible Hand: The Political Foundations of Adaptive Governance in China. Pierson, P. (1994). Dismantling the Welfare State?: Reagan, Thatcher and the Politics of Retrenchment. Cambridge University Press. https://doi.org /10.1017%2Fcbo9780511805288 Pierson, P. (2001). ’Post-industrial pressures on the mature welfare states’. The New Politics of the Welfare State, vol. 1, 80–105. https://doi.org/10.1093 %2F0198297564.003.0004 Plant, R.; and Jones, D. (1991). Modern Political Thought. UK: Blackwell. Polanyi, K.; and MacIver, R. M. (1944). The Great Transformation. USA: Beacon. Pozen, R. C. (2013). Tackling the Chinese Pension System. USA: Paulson Institute. Qiu, G.; and Wen, Z. (2007). ‘Critical adjustment of rural migrant policy in China: towards new paradigm’. Chinese Public Policy Review, vol. 11. Ratigan, K. (2017). ‘Disaggregating the developing welfare state: Provincial social policy regimes in China’. World Development, vol. 98, 467–484. https://doi.org/10.1016/j.worlddev.2017.05.010 Riedmüller, B. (2008). Private Pensions Versus Social Inclusion?: Non-state Provision for Citizens at Risk in Europe. Edward Elgar Publishing. Rimlinger, G. V. (1971). Welfare Policy and Industrialization in Europe, America and Russia. USA: Wiley. Ringen, S.; and Ngok, K. (2017). ‘What kind of welfare state is emerging in China?’ Towards Universal Health Care in Emerging Economies, 213–237. Springer. https://doi.org/10.1057%2F978-1-137-53377-7_8 Roberts, M. E. (2018). Censored: Distraction and Diversion inside China’s Great Firewall. Princeton University Press. https://doi.org/10.23943% 2F9781400890057 Bibliography 255 Roberts, M. E.; Stewart, B. M.; and Tingley, D. (2014). stm: R package for structural topic models. Journal of Statistical Software, vol. 10, issue 2, 1–40. https://doi.org/10.18637/jss.v091.i02 Roberts, M. E.; Stewart, B. M.; and Tingley, D. (2016). Navigating the local modes of big data. Computational Social Science, vol. 51. https://doi.org/10 .1017%2Fcbo9781316257340.004 Rose, R. (2007). ‘The democracy barometers (Part I): Learning to support new regimes in Europe’. Journal of Democracy, vol. 18, issue 3, 111–125. https:// doi.org/10.1353/jod.2007.0054 Rose, R.; and Carnaghan, E. (1995). ‘Generational effects on attitudes to communist regimes: A comparative analysis’. Post-Soviet Affairs, vol. 11, issue 1, 28–56. https://doi.org/10.1080/1060586X.1995.10641393 Rose, R.; Mishler, W.; and Haerpfer, C. (1997). ‘Social capital in civic and stressful societies’. Studies in Comparative International Development, vol. 32, issue 3, 85–111. https://doi.org/10.1007/bf02687332 Rose, R.; Mishler, W.; and Munro, N. (2006). Russia Transformed: Developing Popular Support for a New Regime. Cambridge University Press. https://doi .org/10.1017%2Fcbo9780511492150 Scharping, T. (2013). Birth Control in China 1949-2000: Population Policy and Demographic Development. Routledge. https://doi.org/10.4324% 2F9781315027777 Schmidt, V. A. (2002). Europeanization and the mechanics of economic policy adjustment. Journal of European Public Policy, vol. 9, issue 6, 894–912. Schneider, A.; and Ingram, H. (1993). ‘Social construction of target populations: Implications for politics and policy’. American Political Science Review, vol. 87, issue 2, 334–347. https://doi.org/10.2307/2939044 Schneider, A.; and Sidney, M. (2009). ‘What is next for policy design and social construction theory? 1’. Policy Studies Journal, vol. 37, issue 1, 103–119. https://doi.org/10.1111/j.1541-0072.2008.00298.x Schneider, A. L.; Ingram, H.; and DeLeon, P. (2014). ‘Democratic policy design: Social construction of target populations’. In P. Sabatier and C. Weible (eds), Theories of the Policy Process, 3rd edn, 105–149. Schneider, A. L.; and Ingram, H. M. (2019). ‘Social constructions, anticipatory feedback strategies, and deceptive public policy’. Policy Studies Journal, vol. 47, issue 2, 206–236. https://doi.org/10.1111/psj.12281 Schuman, H.; and Rieger, C. (1992). ‘Historical analogies, generational effects, and attitudes toward war’. American Sociological Review, 315–326. https:// doi.org/10.2307/2096238. OA: https://deepblue.lib.umich.edu/handle/2027 .42/91764 Seidman, I. (2006). Interviewing as Qualitative Research: A Guide for Researchers in Education and the Social Sciences. Teachers College Press. Selden, M.; and You, L. (1997). ‘The reform of social welfare in China’. World Development, vol. 25, issue 10, 1657–1668. https://doi.org/10.1016/s0305 -750x(97)00055-7 256 Pension Policy and Governmentality in China Shen, X.; and Truex, R. (2018). ‘In search of preference falsification’. Working Paper. Shi, S. J.; and Mok, K. H. (2012). ‘Pension privatisation in Greater China: Institutional patterns and policy outcomes’. International Journal of Social Welfare, vol. 21, S30–S45. https://doi.org/10.1111/j.1468-2397.2012.00875.x https://doi.org/10.1111/j.1468-2397.2012.00875.x Shi, T. (2001). ‘Cultural values and political trust: A comparison of the People’s Republic of China and Taiwan’. Comparative Politics, 401–419. https://doi .org/10.2307/422441 Shi, T. (2014). The Cultural Logic of Politics in Mainland China and Taiwan. Cambridge University Press. https://doi.org/10.1017/CBO9780511996474 Shirk, S. L. (2011). Changing Media, Changing China. Oxford University Press. Shue, H. (1996). Basic Rights: Subsistence, Affluence, and US Foreign Policy. Princeton University Press. Siebert, F. S.; Peterson, T.; and Schramm, W. (1956). Four Theories of the Press: The Authoritarian, Libertarian, Social Responsibility, and Soviet Communist Concepts of What the Press Should Be and Do. University of Illinois Press. Simon, J. (2014). The New Censorship: Inside the Global Battle for Media Freedom. Columbia University Press. https://doi.org/10.7312%2Fcolumbia %2F9780231160643.001.0001 Smoke, R. (1994). ‘On the importance of policy legitimacy’. Political Psychology, 97–110. https://doi.org/10.2307/3791441 Song, S.; and Chu, G. S. F. (1997). ‘Social security reform in China: The case of old-age insurance’. Contemporary Economic Policy, vol. 15, issue 2, 85–93. https://doi.org/10.1111/j.1465-7287.1997.tb00468.x Song, Y. (2014). ‘What should economists know about the current Chinese hukou system?’ China Economic Review, vol. 29, 200–212. https://doi.org /10.1016/j.chieco.2014.04.012 State Council. (1997). Decision on Establishing a Unified System of Basic Pension Insurance for Enterprise Employees (State Council [1997] No.26). Retrieved from http://www.gov.cn/ztzl/nmg/content_412509.htm. State Council. (2000a). Opinions on Accelerating the Process of Social Welfare Socialization. Retrieved from http://shfl.mca.gov.cn/article/zcfg /200809/20080900019761.shtml. State Council. (2000b). The Notice of Issuing the Pilot Program of Urban Social Security System (State Council [2000] No.42). Retrieved from http://www .gov.cn/xxgk/pub/govpublic/mrlm/201011/t20101112_62507.html. State Council. (2002). New Rural Collaborative Health Insurance Scheme. State Council. (2005). Decision on Perfecting Basic System of Pension Insurance for Enterprise Employees (State Council [2005] No.38). Retrieved from http://www.gov.cn/zwgk/2005-12/14/content_127311.htm. State Council. (2009). New Rural Old-age Social Insurance Plan (State Council [2009] No. 32) Retrieved from http://www.gov.cn/zwgk/2009-09/04/content _1409216.htm. Bibliography 257 State Council. (2014). Urban-Rural Residents Basic Old-age Social Insurance Scheme (State Council [2014] No. 8). Retrieved from http://www.gov.cn /zwgk/2014-02/26/content_2621907.htm. Stockmann, D. (2013). Media Commercialization and Authoritarian Rule in China. Cambridge University Press. https://doi.org/10.1017/CBO9781139087742 Streeten, P. (1980). ‘Basic needs and human rights’. World Development, vol. 8, issue 2, 107–111. https://doi.org/10.1016/0305-750x(80)90019-4 Su, Z.; and Meng, T. (2016). ‘Selective responsiveness: Online public demands and government responsiveness in authoritarian China’. Social Science Research, vol. 59, 52–67. https://doi.org/10.1016/j.ssresearch.2016.04.017 Svallfors, S. (2007). The Political Sociology of the Welfare State: Institutions, Social Cleavages, and Orientations. USA: Stanford University Press. https:// doi.org/10.11126%2Fstanford%2F9780804754354.001.0001 Svallfors, S. (2010). ‘Policy feedback, generational replacement, and attitudes to state intervention: Eastern and Western Germany, 1990–2006’. European Political Science Review, vol. 2, issue 1, 119–135. https://doi.org/10.1017 /s1755773909990257 Svolik, M. W. (2012). The Politics of Authoritarian Rule. Cambridge University Press. https://doi.org/10.1017/CBO9781139176040 Szelenyi, I. (1978). ‘Social inequalities in state socialist redistributive economies’. International Journal of Comparative Sociology, vol. 19, issue 1–2, 63–87. https://doi.org/10.1177/002071527801900105 Szelenyi, I.; and Kostello, E. (1996). ‘The market transition debate: Toward a synthesis?’ American Journal of Sociology, vol. 101, issue 4, 1082–1096. https://doi.org/10.1086/230791 Szelenyi, I.; and Szelenyi, B. (1994). ‘Why socialism failed: Toward a theory of system breakdown—Causes of disintegration of East European state socialism’. Theory and Society, vol. 23, issue 2, 211–231. https://doi.org/10.1007 /bf00993815 Tang, W. (2016). Populist Authoritarianism: Chinese Political Culture and Regime Sustainability. Oxford University Press. https://doi.org/10.1093/acp rof:oso/9780190205782.001.0001 Tannenberg, M. (2017). ‘The autocratic trust bias: Politically sensitive survey items and self-censorship’. V-Dem Working Paper, vol. 49. https://doi .org/10.2139/ssrn.2980727 Thistlethwaite, D. L.; and Campbell, D. T. (1960). ‘Regression-discontinuity analysis: An alternative to the ex post facto experiment’. Journal of Educational Psychology, vol. 51, issue 6, 309. https://doi.org/10.1037/h0044319 Tilly, C. (2017). ‘Coercion, capital, and European states, AD 990–1990’. Collective Violence, Contentious Politics, and Social Change, 140–154. Routledge. https://doi.org/10.1080/714005469 Tolbert, P.; and Zucker, L. (1996). ‘The institutionalization of institutional theory’, Handbook for Organization Studies, 175–190. UK: Sage. https://doi.org /10.4135%2F9781446218556.n6