Nan Zou Bakkeli

The Price of Inequality Privatisation, Health and Wellbeing in China

© Nan Zou Bakkeli, 2016

Series of dissertations submitted to the Faculty of Social Sciences, University of Oslo No. 611 ISSN 1564-3991

All rights reserved. No part of this publication may be reproduced or transmitted, in any form or by any means, without permission.

Cover: Hanne Baadsgaard Utigard. Print production: Reprosentralen, University of Oslo.

Table of Contents Acknowledgements____________________________________ v Introduction__________________________________________ ix Part I _______________________________________________17 Chapter One. The Cat that Catches Mice ________________ 19 Privatising China ________________________________ 20 The healthcare system ___________________________ 26 Health inequality________________________________ 29 Stress, genes and epidemiological transition _______ 33 Wellbeing _______________________________________ 36 Inequality by geography __________________________ 40 Chapter Two. Theories: Feeling the Stones ______________ 45 Market transition theory _________________________ 47 Occupation, education and the hukou system _______ 49 Income inequality-health hypothesis ______________ 51 Epidemiologic transition _________________________ 57 Wellbeing and inequality _________________________ 59 Chapter Three. Methods and Data ______________________ 65 Data ____________________________________________ 66 Analytical framework ____________________________ 69 Causality _______________________________________ 79 Chapter Four. Summaries of the Articles ________________ 85 Article 1: Income Inequality and Privatisation ______ 86 Article 2: Income Inequality and Health ___________ 87 Article 3: Older Adults’ Mental Health _____________ 88 Chapter Five. Conclusion: Emerging Challenges _________ 89 Theory-led analytical frameworks _________________ 91 Healthcare and welfare support ___________________ 95 Privatising public hospitals _______________________ 97 The active social volcano? ________________________ 99 Future studies _________________________________ 101 References __________________________________________ 103

iv

Part II _____________________________________________ 129 Article 1: Income inequality and Privatisation__________ 131 Article 2: Income Inequality and health _______________ 169 Article 3: Older Adults’ Mental Health_________________ 181 Tables Table 1. Mean life satisfaction in China 1990s-2000s _____ 38 Table 2. Overview of the articles ________________________ 86 Figures Figure 1. Age and wellbeing, CHNS 2006-2009. _________ 40 Figure 2. Theories and framework of analysis ___________ 46 Figure 3: Income inequality, income and mortality ______ 55 Figure 4. Map of survey regions ________________________ 69 Figure 5. Adapted Coleman-diagram ___________________ 72 Figure 6. Gini coefficients by counties and waves ________ 75

Acknowledgements I am deeply grateful to my supervisors for their excellent guidance throughout my PhD research. Professor Gunn Elisabeth Birkelund (Department of Sociology and Human Geography, University of Oslo) gave me great mentoring and constant support. She helped me to position my research in light of nuanced theoretical and methodological discussions in sociology. She has read numerous drafts and provided insightful comments and detailed advice. Her efforts have doubtlessly improved the quality of the work, and her commitment, encouragement and warm support is a continuing source of inspiration and motivation. Without her, this work would never have been possible. I would hence like to express my deepest gratitude to Gunn: Thank you! My

co-supervisor,

Professor

Torkild

Hovde

Lyngstad

(Department of Sociology and Human Geography, University of Oslo) has shared his extensive knowledge on doing statistical analysis. He has helped me throughout the PhD process not only with his methodological expertise, but also by engaging in theoretical discussions with me and helping me to understand phenomena in a sociological perspective. He generously taught me how to let data talk, and how to incorporate sociological theories into one’s research interest. I appreciate his insightful comments, corrective feedback, and deep and thorough reading of the manuscripts. He has helped me to push my limits and approach

vi

questions in ways that I never would have done without him. I owe Torkild my gratitude. I owe much to my co-supervisor Jon Pedersen (Fafo Institute for Labour and Social Research), who has generously shared his knowledge and expertise of China with me. He has helped me out with everything from elaborating theoretical ideas, questioning my use of methods, discussing the updated empirical evidence from the Chinese context, to correcting my grammar and punctuation. I have benefited enormously from his continuous encouragement, vast knowledge, clarity, insightful feedback and valuable comments. A special thanks is owed to Professor Jon Ivar Elstad (Centre for Welfare and Labour Research, Oslo and Akershus University College). He is my previous supervisor and a dear friend. He has shared his thoughts, reflections and extensive knowledge about health, well-being and inequality with me. He has patiently replied to e-mails, and discussed and answered numerous long questions. I am sincerely grateful for all the help I have received from Jon Ivar – thank you! I am also thankful to my employer, the Department of Sociology and Human Geography at the University of Oslo, for providing a great academic community and offering help for practical matters. I thank my dear colleagues at the department, who have been supportive and amazing discussion partners. They have made the department a fantastic place to work at. Thanks to Tore Witsø Rafoss, Jens Lunnan Hjort, Tone Maia Liodden, Heidi Grundetjern, Emma Arnold, Nicolai Borgen, Øyvind Wiborg, Marielle Stigum Gleiss, Heidi Østbø Haugen, Sveinung Legard, Ferdinand Andreas Mohn, Jørn Ljunggren, Eirin Pedersen, Anja Beate Sletteland, Håkon Larson, Are Skeie Hermansen, Anne May Melsom, Kirsten Ulstrud, and so many others. I am fortunate to have you, not only for the effort you have invested in creating an excellent workplace, but also as good friends, with

vii

whom I have shared many interesting conversations and unforgettable times. I would also like to thank the AKS (Analytical Quantitative Sociology) network for the great number of seminars, workshops, courses and lectures it has offered. Thanks to Torkild Lyngstad and Gunn Birkelund, who put much effort into coordinating and arranging the AKS, and thanks for giving me the opportunity to participate in the network. The meetings with the AKS group gave me important inspiration and ideas, and introduced me to excellent sociologists and people. It meant a lot to me. I am grateful for my family and friends for their patience and support. I thank my mother, Teng Lili, for her love and care – thank you for the light and warm food you had waiting for me on the days I came back home late from work; thank you for your patience and for your support for your daughter who went in search of education and a new life in a foreign country. And to my deceased father, Zou Sichen: Dad, I know you would be proud of me. To my cats, Missy Fornøyd and Maki: Thank you for giving me something to look forward to everyday after a long working day, and the silent but most affirming support for my work you gave me by sitting by and on my keyboard. Last but most importantly, I want to thank my dear husband Vidar Bakkeli, for his patience, encouragement, help and love. It is wonderful to know that someone is always there and supporting me, no matter how hard it is. I am lucky to have Vidar, not only as a loving and supportive husband, but also as my closest friend with whom I could share everything, and the most talented social researcher to discuss work and studies with. You are the greatest joy in my life, Vidar; thanks for everything. Nan Zou Bakkeli Oslo, April 2016

Introduction In the period since the economic reforms started in the early 1980s, China has experienced tremendous growth, and its society has continuously been in a state of rapid transition. The structural changes have emphasised economic development, in terms of the introduction of market-oriented social policies and the development of infrastructure. The market economy has gradually replaced the planned economy of the Mao era, and newer keywords such as efficiency, incentive, profit, market and capital have replaced the keywords of the planned economic system such as class struggle, ‘proletariat dictatorship’ and ‘the iron rice bowl’. The market reforms have had an ambiguous impact on Chinese society. On the one hand, millions have been lifted out of poverty, people’s living standards have been dramatically improved, household and individual income has increased greatly, and the average annual GDP growth has been around 10% for over three decades. Following Deng Xiaoping’s slogan, ‘to be rich is glorious’, China now has more than a million millionaires (measured in USD) and more than 200 billionaires (Whyte, 2014). On the other hand, income inequality has increased, creating a more divided society. There are environmental problems on a large scale, and problems with widespread corruption. Many indicators point to a massive increase in income inequality. The Gini coefficient – an indicator measuring the degree of inequality

x

in a society – reached .49 in 2008, placing China in the group of countries with the most unequal distributions in the world (NBSC, 2013). According to the China Family Development Report 2015, the household incomes of the top 20% of households in China is 19 times higher than that of the bottom 20% (NHFPC, 2015a). Higher income inequality creates a society in which people’s life spheres are drifting away from each other. Children from rich urban families live in villas, have their food served by nannies, wear customised clothes from luxury brands and are driven to school by private drivers. Poor rural and migrant families are struggling for food, basic medical care and schooling. In some areas, villas and slums lie close together, only separated by high walls and guards, but representing two different worlds of living. In 2012, Goldman Sachs reported that China’s luxury consumption accounted for one-fourth of the global market, and a report by Bain & Company stated that in 2014 about one-third of the amount spent globally on luxury products came from Chinese consumers (D’Arpizio et al., 2014; Zhan and He, 2012). At the same time, the proportion of China’s population below the World Bank’s poverty line is higher than that of other countries with a comparable per capita income (World Bank, 2003). A major gap exists between rural and urban China. According to the National Bureau of Statistics of China, 70 million rural citizens currently live in poverty, amounting to 7.2% of the whole rural population (NBSC, 2015). A recent report by Xinhua News Agency (Xinhua News, 2015) portrayed the life of the rural poor in Central and Western China. It stated that people from Daliangshang Prefecture in Sichuan Province live in thatched sheds with their livestock and cannot afford to eat meat more than three times a year. In Xiji County, Ningxia Province, people have to travel about 40 kilometres twice a month to get drinking

xi

water. In Guru Village, Guizhou Province, 1,100 out of 1,200 people are illiterate.1 Despite the economic growth in the last decades, the Chinese government is facing enormous challenges connected to income inequality. According to the World Value Survey, the proportion of the Chinese population strongly supporting the argument ‘we need larger income differences as incentives’ fell from 32% in 1990 to 4% in 2012.2 Despite China’s rising income, the level of wellbeing has in fact diminished: In the 1990 World Value Survey, 27.5% of people reported being ‘very happy’, but this number fell to 11.5% in 2012 (Brockmann et al., 2008). In the 2014 Gallup Healthways’ Country Well-being Ranking, China ranked number 127 out of 145 countries (Gallup-Healthways, 2015). Income inequality is not only a matter of social justice and equity, but may also affect people’s physical and mental health. According to the Wilkinson hypothesis, higher inequality may lead to negative factors such as lower social cohesion and more mental stress, and therefore leads to worse health (Wilkinson, 1992, 1996, 1999). It is widely recognised that China has achieved great improvements in population and public health, such as a dramatically lower mortality rate, increased life expectancy and extensive immunisation coverage (Chan et al., 2008). However, much of the improvements in health were achieved before the economic reforms, and China is now experiencing several emerging public health problems (Chan et al., 2008). One study showed that nearly one in five adults had suffered from at least one mental disorder in the past month; that is roughly 173 million Chinese people (Phillips et al., 2009).

See detailed report from the Xinhua News Agency (2015): Survey on Poverty: Facing China’s Poor. 2 See the Online Data Analysis from World Values Survey, the perceptions about income inequality (2012 data variable coded V96; 1990 data coded V250): http://www.worldvaluessurvey.org/WVSOnline.jsp 1

xii

At the same time, market reforms meant that profit-making was prioritised. Publically funded mental health rehabilitation facilities were closed down, merged, or down-sized to smaller-scale psychiatric hospitals. For example, in Shanghai, there was at least one mental health rehabilitation facility in each district before 1990. By June 2004, the number of facilities had decreased by 62% (Liu et al., 2011). There are also many ongoing trends that cause health problems. With millions of cigarette smokers, China has become the largest tobacco consuming country in the world; obesity has increased in the population; about 10% of the population is infected with Hepatitis B; the country has experienced a progressing HIV epidemic; tuberculosis (which was previously under control) has broken out again; and in some underdeveloped areas, cholera and other epidemic diseases have been spreading (DRC, 2005; Pei and Rodriguez, 2006; Sun et al., 2002; WHO, 1999). There are also serious malnutrition issues among rural children: the stunting rate among children was 22% in 1998, and in some poor provinces, it was as high as 46% (see Park and Wang, 2001). Furthermore, there are huge rural-urban differences: the prevalence of underweight among children under five years old was 1.8% in cities in 2005, but 8.6% in rural areas (Chang et al., 2006). In the western part of China, the prevalence of stunting and underweight among children below three years old was 24% and 22.4% respectively (Zeng et al., 2003). The number of Chinese people suffering from cardiovascular diseases is also increasing. According to the World Health Organisation, about 230 million Chinese people (one in five adults) have a cardiovascular disease. The percentages of deaths associated with cardiovascular diseases in urban and rural areas are 20.9% and 17.9% of China’s total numbers of deaths per year respectively (WHO, 2015). We observe that increasing health problems and increased income inequality have both emerged during the period of

xiii

economic reform. Does this mean that the Wilkinson hypothesis is confirmed for China, or is it mostly due to the epidemiological transition? This dissertation is concerned with income inequality in China and asks two central questions. First, does the increased income inequality in China have something to do with the marketoriented economic reforms? Second, what consequences does income inequality have for people’s health and well-being? The dissertation consists of three articles on income inequality in China, and answers three related research questions. First, what is the relation between the process of privatisation and increasing

income inequality?

Second, does rising

income

inequality have a negative impact on people’s physical health? Third, what is the relation between income inequality and subjective well-being after we have controlled for individual income and other relevant social and individual factors? In the first article, ‘Income Inequality and Privatisation: A Multilevel Analysis Comparing Prefectural Size of Private Sectors in Western China’, I explore how privatisation in the western provinces is related to income inequality and individuals’ income through education, occupation, and the household registration system (the hukou system). I draw upon market transition theory, and compare the size of privatisation in different prefectures in order to see if inequality varies with the degree of privatisation. The article includes both macro and micro factors when analysing privatisation and inequality. I find that income inequality is negatively associated with privatisation, and that education, occupation and hukou play important roles in explaining this relation. In the second article, ‘Income Inequality and Health in China: A Panel Data Analysis’, I examine whether income inequality in China is related to individuals’ physical health outcomes. The study is an attempt to distinguish the effect of individual income

xiv

on health at the micro level, and income inequality at the macro level. I find that when using different physical measures to measure individual health, income inequality does not have a significant impact on individuals’ risks of having health problems. This result is robust when changing between different indicators for income inequality. In

the

third

paper,

‘Older

Adults’

Mental

Health

in

Transitional China: A Sociological Study of the Relationship Between Income Inequality and Subjective Wellbeing’, I explore the relationship between income inequality and subjective wellbeing for older adults aged 60-90 in China. The ‘Easterlin Paradox’ shows that rising income is not necessarily connected to rising happiness, and that relative income is important for happiness. The study is a test of the Easterlin Paradox in the Chinese context. I find that a higher level of inequality is related to lower wellbeing for older adults, and that the effect of individual income on well-being varies between different income groups. For the richer urban groups, individual income is positively correlated with better wellbeing, while income has little association with wellbeing for older rural adults, or poorer urban people. The dissertation is organised as followed. In the next chapter, I discuss the contextual background in China. The focus is on the process of market transition and privatisation after the economic reforms from the beginning of the 1980s. We will also look more closely at the development of income inequality during this period, as well as the rising concerns about the outcomes of income inequality. Central concepts like privatisation, marketisation, income inequality, the healthcare system and individual physical health and well-being are discussed. In Chapter Two, the focus is on theoretical discussions. First, I look at the market transition theory and different stratifying mechanisms in the Chinese market economy that may generate

xv

income inequality, such as education, labour market, and the household registration system. Further, I present theories concerning the inequality-health relationship. Finally, I discuss theories concerning wellbeing, happiness and life satisfaction, and their relation to income inequality. Chapter Three gives an overview of the data and methods used in the research articles. The analytical frameworks that have been used in the thesis are presented here. This includes the macro-micro

linkage,

and

discussions

about

regional

heterogeneity. Since all three articles are based on quantitative analysis, there will be some discussions about different statistical techniques as well as a general discussion about causality. Chapter Four presents summaries of the three articles in the dissertation. Chapter Five draws upon the articles’ findings and discusses the findings with respect to newer challenges in China. The

dissertation

answers

the

following

questions:

What

contributes to income inequality in a transitional China? To be more concrete: how does the process of marketisation, defined as the measure of privatisation, influence income inequality? Further, how does rising inequality affect individual health outcomes in terms of physical health and well-being? The final part of the dissertation (Part II) consists of the three articles in the PhD research: 1.

Income inequality and Privatisation: A multilevel analysis comparing prefectural size of private sectors in Western China.

2.

Income Inequality and Health in China: A Panel Data Analysis.

3.

Older Adults’ Mental Health in China: A Sociological Study of the Relationship between Income Inequality and Subjective Wellbeing

Part I

Chapter One

The Cat that Catches Mice ‘The aim of socialism is to make all our people prosperous, not to create polarization. If our policies led to polarization, it would mean that we had failed; if a new bourgeoisie emerged, it would mean that we had strayed from the right path.’ - Deng Xiaoping, March 7, 1985 3

The economic reform period in China started in December 1978, led by Deng Xiaoping. This year marked a turning point in the transition, in which the planned economic system was changed into a market economy, and in which the state gradually withdrew from its responsibility to ensure employment, welfare support and the distribution of wages and goods. Unlike in the Soviet Union, the process of reforming Chinese political economy was performed much more gradually (Naughton, 2007). It was carried out in both urban and rural areas, and covered a wide range of fields: agriculture, industry, manufacturing, construction services, technology, science and research, and the welfare system. The

core

elements

of

the

economic

reform

were

the

This was a speech given by Deng Xiaoping during the National Science and Technology Working Conference in 1985, titled ‘Unity Depends on Ideals and Discipline’. The speech is contained in the Selected Works of Deng Xiaoping (Vol. 3). (Literature Editorial Board of the CPC Central Committee, 1993a)

3

20

Chapter One

decollectivisation of the agricultural sector, the marketisation and privatisation of the state sector and the decentralisation of state control (Naughton, 2007). Almost from the very beginning, the reform measures led to rapid economic growth. The per capita disposable income of urban households increased from 343 yuan in 1978 to 24,564 yuan in 2012, and the per capita net income of rural households soared from 134 yuan to 7,917 yuan in the same period (NBSC, 2015). From 1978 to 2012, the ratio of food consumption to total expenditure (the Engel’s coefficient) for Chinese households shrunk from .58 to .36 in urban areas, and from .68 to .39 in rural areas (NBSC, 2015). 4 Taken together, these figures show the radical improvements in living standard that Chinese people have experienced. In addition to creating rapid economic growth and improved living standards, economic reforms have also led to increased income inequality. According to official Chinese figures, the nationwide Gini coefficient in 2012 was .47, having gone down from a peak value of .49 in 2008 (NBSC, 2013). The level of inequality in China has surpassed that of the U.S. (.41), thus making China one of the most unequal countries in the world, compared to Latin-American countries such as Colombia (.54), Brazil (.53) and Mexico (.48) (The World Bank, 2013).

Privatising China The concept of privatisation can be defined as a process in which public and state property is converted into private assets, and the ownership is partly or completely transferred into private control (Roland, 2008). Several factors can work together to affect how The Engel’s coefficient ranges from 0 to 1. A larger number means a higher proportion of income spent on food, and a lower standard of living in a country. See e.g. United Nations Statistics Division (United Nations Statistics Division, 2005). 4

The Cat that Catches Mice

21

privatisation works. These include the type of goods and services that are to be privatised, the parties involved in the process of privatisation and the transition cost incurred by institutions (Araral 2009). Similarly, the processes of privatisation can also impact the outcome of privatisation, as can the economic, political and historical context of the region in which these processes take place (Birdsall and Nellis 2003). Privatisation is only one dimension of the massive changes brought about by economic reforms in China but it is an essential aspect of the market transformation that has taken place. There are several reasons for looking closer at the process of privatisation

in

China.

First,

unlike

the

general

term

‘marketisation’, privatisation refers to a concrete measure in the transitional economy. It involves a set of legal, policy and behavioural underpinnings – not just economic institutions, but also underlying political and social infrastructures. Second, privatisation may have an overall impact on individuals through different

institutions.

For

example,

economic

institutions

distribute and redistribute income, goods and resources to people through labour market and welfare services. The distribution of goods may have individual consequences for employment, living standards,

health,

happiness

and

life

satisfaction.

Third,

privatisation has a strong political and ideological aspect, especially in a previously communist country like China. In fact, due

to

the

communist

and

socialist

ideology,

the

term

privatisation was little used in the official documents and announcements. Instead, the authorities preferred to use words such as ‘marketisation’, or ‘opening-up the market’. Furthermore, the process of privatisation has not only affected political discourses, but also has an impact on individuals’ daily experiences and perceptions, for example concerning their sense of equality, equity and justice. This may have a great influence on individuals’ mental health and wellbeing.

22

Chapter One

There is huge variety between countries when considering privatisation policies, laws, and ownership structures (see e.g. Roland 2008; Sclar 2001). Privatisation can happen in different ways: by reducing state provisions, cutting state subsidies in different ways and to different degrees, and reducing state regulation (Le Grand and Robinson, 1984). In China, the state has always been essential, in terms of ownership, regulation and the overall the overall power it has over other spheres in the society (Zeng, 2013). But we may still observe a transfer from public control to private management, and to some degrees a transfer to private ownership. The process of privatisation in China started in the agricultural sector. Before the 1980s, the collective served as the basic accounting unit; in these collectives, people worked together on the pooled land. Income was distributed to households, and the major part of income was given in the form of grain (Naughton, 2007). The collectives were responsible for redistributing both goods and welfare provisions. Rural citizens lived a rather secure life – they were fed, clothed and housed. But this form of production was criticised for being inefficient and unsustainable, and it gave agricultural labourers few incentives. In the reform, individual pieces of land were contracted to farm households, who took over the management of agricultural production. In return, they gave a certain amount of procurement and tax grain as a quota to the government. They kept the rest of the output and decided themselves whether to keep it, use it or sell it on the local open market (Chow, 2007). At the end of 1983, the contracting of land to households was rapidly introduced throughout rural China and became nearly universal (Naughton, 2007). Later on, farmers were allowed to rent out or sell their piece of land. The land was thus privatised in terms of usage rights, but not in terms of ownership.

The Cat that Catches Mice

23

Agricultural production began to surge, and the increase in grain output freed up the agricultural labour force by reducing the time spent in farming. Non-agricultural sectors in rural areas, known as town and village enterprises, emerged. They soon bloomed and proved ideal for absorbing the excess workforce from agricultural production (Whyte, 2010a). These enterprises were originally collectively owned, but from the mid-1990s, they were privatised and transformed into privately owned enterprises. The trend of privatising town and village enterprises soon swept across rural areas throughout the country. In 2003, a total of 136 million people were employed in town and village enterprises. Only 9% of them worked in collective or state-owned enterprises, while the rest of them worked in enterprises under private ownership, individually run businesses, stock cooperatives, jointly operated enterprises and limited liability companies (Naughton, 2007). Before the reform period, people in urban areas were organised in different work units, based on the state enterprises they were employed in. Like the rural collectives, these work units were responsible for people’s work, lives and welfare. When reform started in the 1980s, the dual-track system was introduced in urban areas, inspired by the agricultural reforms. State enterprises were given greater autonomy to decide and manage their own production. They had to fulfil certain defined tasks, paying the state a specified tax, and had to keep the profits for bonuses or new investments. By the end of 1987, about 78% of all state enterprises were organised according to the contract system (Wu, 2005). In 1994, the government decided to allow for the conversion of converting

state enterprises into shareholder

corporations,

thereby paving the way for private ownership of former state enterprises (Gan et al., 2008). During the 15th National Congress of the Chinese Communist Party, Jiang Zemin announced his

24

Chapter One

intentions to reform the large-scale enterprises by (1) creating multiple ownership forms and ‘even transnational businesses, through the link of capital’; and (2) the ‘reorganisation, reunionisation,

merging,

leasing,

contract-management,

stock

cooperation, sell-out’ of the smaller-scale state enterprises (The Central People’s Government of the PRC, 2007). The central guideline for a massive wave of privatisation in urban areas followed the policy principle of ‘keeping the large and let the small go’. Privatisation was carried out in different ways: through mixed ownership (in some large enterprises the state still had the majority

share,

while

other

enterprises

were

completely

privatised, with all shares sold to private investors), management buyouts, auction sales, mergers and acquisitions, bankruptcy declarations and the formation of new companies with changed ownership, as well as lease contracts and joint venture arrangements with foreign capital (Garnaut et al., 2005). As a result, in the period from 1994–2005 alone, the number of people employed in the state enterprises was reduced by 62%, from 78 to 30 million (Naughton, 2007). To summarise, by 2005 the share of the total workforce employed in the agricultural sector was still large, but significantly reduced, from 69% in 1978 to 47% in 2003. The state enterprises and the town and village enterprises also decreased greatly. In 2003, 4% of the entire workforce was employed in state enterprises (reduced from 14% in 1978), and 2% in publicly owned township enterprises (compared to 6% in 1978). By contrast, the private sector grew rapidly. The size of the urban private sector grew from almost nothing in 1978 to 13% of the total urban labour force in 2003, corresponding to 21 million people. Nationwide, the private sector in 2003 amounted to 36%, approximately 268 million people (Naughton, 2007). According to the National Bureau of Statistics of China, in 2014, the number of urban people employed in state enterprises was a total of 68 million,

The Cat that Catches Mice

25

approximately 17% of the entire work force in urban China (NBSC, 2015). 5 Supporters of privatisation often argue that a freer market, combined with a profit motivation, will lead to new initiatives and competition

that in

turn

result

in increased

managerial

productivity, effectiveness and profit and a better quality of products and services. From this perspective, privatisation is viewed as a rational and logical response to the problems of increasing state expenses, inefficient state-owned enterprises and the negative consequences of state control, for example, in terms of limiting human initiative and creativity (Collyer, 2003; Starr, 1988). The same debate also took place in China. To cite a famous slogan by Deng Xiaoping, ‘It does not matter if it is a yellow cat or a black cat, as long as it catches mice’. However, the reforms carried out in order to achieve economic development included measures that contrasted with the socialist ideology. There has been huge and stable growth, reductions in poverty and rising living standards. However, in this process, the country has become more unequal. Income disparities have doubled; urban-rural differences and regional inequalities have increased. Compared to pre-reform era, the quality of institutions and bureaucracy has not been efficiently improved despite the privatisation of state enterprises. Other unintended consequences have also emerged, such as health problems and lower wellbeing. In the following section, I will look closer at the privatisation of healthcare services, as an example of privatisation in China, before moving on to look at how privatisation, inequality, health outcomes and wellbeing are connected.

According to the National Bureau of Statistics of China (2015), in 2014, 63 million people were employed in the urban state-owned enterprises and 5 million in the urban collective-owned enterprises; the total number of urban employed persons is 39 million. 5

26

Chapter One

The healthcare system The process of privatisation has not only reformed the labour market, but has also affected other social spheres. One example is the changing structure of welfare regime and the social security system. In the pre-reform period, workers in urban areas received income, in-kind goods (e.g. food, clothing), housing, child care, schooling, medical care, pensions, etc. from their employers almost for free. In other words, the state effectively guaranteed urban citizens’ welfare through the work units. In the countryside, rural collectives had the same function, but the welfare services were less extensive compared to those in cities. Nevertheless, the system provided basic welfare needs, as well as food, clothing, healthcare and minimal living expenses for vulnerable groups. In the 1980s and 1990s the state gradually reduced its provision of life-long employment and delivery of welfare services. Therefore, when state enterprises were merged, closed down, and privatised, people not only lost their jobs but also their access to welfare services, since these had been provided by the employers. Similarly, when rural collectives collapsed during the economic reform, welfare provision for rural people also vanished. Take the healthcare system as an example. In the pre-reform era, there was relatively wide provision of basic healthcare services in China (Tang et al., 1994). The state and work units were primarily responsible for offering health care services in cities, and the healthcare services for rural citizens were organised as part of the integrated rural collective commune system. Although rural health benefits were less institutionalised than those in urban areas (Knight and Song, 1999; Wagstaff and Lindelow, 2008), the gap was relatively small, and the general level of social welfare was considered high (Gao, 2008; Leueng, 2005). Contrary to what many believe in the West, this model of healthcare was widely accepted as successful and it was

The Cat that Catches Mice

27

understood to contribute to better national health (Djukanovic et al., 1975; Sidel, 1982, 1993). Although the system was far from perfect – there were services of poor quality, and negative changes towards the end of the Cultural Revolution, for example, the discontinuation of medical training at universities – the healthcare system was mainly characterised by its wide coverage, well-organised primary preventive healthcare, and ‘efficient patient referral system for the treatment of health problems’ (Chan et al., 2008; Liu and Yi, 2004). Overall, the market-oriented economic reform that took place from the early 1980s led to a dramatic reduction in healthcare insurance coverage. In urban areas, healthcare provision was based on which sector people were employed in. Many lost their coverage when they lost their jobs in state enterprises, and only employees with close family members employed in the public sector and state-owned enterprises remained insured (Leueng, 2005). By 2003, about 65% of urban residents had to pay for healthcare themselves, and self-financed medical spending had increased 13 fold compared to 1990 (Mok et al., 2010). In rural areas, the de-collectivisation of agriculture communities led to reduction of public services, and households became responsible for themselves. As a result, the healthcare system collapsed. There was a fall in health service coverage from about 90% of the rural population in 1980 to 5% in 1985 (Liu and Cao, 1992). One of the major challenges facing the Chinese healthcare system after the economic reforms is that the health services have become heavily hospital-oriented, while primary care has been given low priority (Chan et al., 2008; Liu, 2004). Before the economic reforms, healthcare delivery was based on a three-tier healthcare system. In urban areas, primary healthcare was provided by the community health units and district hospitals, while municipal and provincial hospitals focused on tertiary medical services. In rural areas, village clinics were responsible

28

Chapter One

for the primary tier, township public hospitals served as the second tier, and county hospitals provided the highest tier for inpatients. However, as a result of the localisation and privatisation of health services during the economic reforms, local governments now only provide limited subsidies for health services. Primary care has almost ceased to function. In urban areas, 80% of people visit hospitals for even minor illnesses. In rural China, clinics do not provide preventive care without reimbursement (Chan et al., 2008). The process of decentralisation of welfare provision was combined

with

a

reduction

in

health

expenditure

and

privatisation. Total national health expenditure dropped from 28% in 1978 to 14% in 1993, and health expenditure in rural areas fell from 20% to 2% (Bloom and Gu, 1997). In 2002, the central government required each township to

provide only

one

government-run hospital, while the rest was contracted out or privatised (Meng, 2005). Healthcare providers such as hospitals and clinics were expected to become ‘independent accounting units of profit-making entities with independent management’, and a revenue-based bonus system was used to increase productivity (DRC, 2005). As a result, healthcare services have been transformed from being institutions responsible for addressing public healthcare needs, to ‘revenue-maximising organisations’ (Meng, 2006; Tang and Meng, 2004). In order to increase revenue, many hospitals have invested in expensive medical equipment to charge more; they are also recommending unnecessary treatments and selling more expensive medicines. The government has attempted to regulate drug and healthcare cost, but they have also set prices for new, high-end services above the cost regulation, and have allowed a 15% profit margin on drugs. This has prompted hospitals to encourage doctors to

The Cat that Catches Mice

29

prescribe more and costlier medicines, increase expensive hightech services, and avoid using cheaper medicines covered under regulation (Chan et al., 2008; Meng, 2006; Yip and Hsiao, 2008). Among patients who caught a cold, 75% were prescribed antibiotics (Yip and Hsiao, 2008). Aware of these growing challenges, from the beginning of the 2000’s the Chinese central government launched several new social programs to extend medical insurance. In urban areas, medical insurance was aimed at covering all urban workers and intended to function as a unified system across different work units (Liu, 2002). In rural areas, a new medical scheme, launched in 2003, was implemented to cover the whole rural population (Yip and Hsiao, 2008). From 2005 to 2006, coverage was increased from 600 to 1,433 counties, and by 2008 the programme had been implemented in around 95% of all rural counties in China (Zhang et al., 2010).

Health inequality Notwithstanding these attempts to improve the welfare system, in the middle of the 2000s, the quality of public health care actually worsened. Medical costs became even more expensive, and reimbursements were low. Many poorer people were thus excluded from access to basic healthcare (Chan et al., 2008). About 13% of Chinese households had to pay more than 40% of their disposable income on health, and half of inpatient costs were paid out-of-pocket by an average patient (Ouyang, 2013). The system was fairly unequal, with rural residents in the lowest income quintile spending the highest share of their income on healthcare (Yip and Hsiao, 2009). Further, real income per capita in 2003 was almost double that of 1993, but health expenditure per admission/visit had quadrupled in the same period. On average, the cost of one inpatient hospitalisation was 70-80% of an individual’s annual income in 1993. In 2003, this

30

Chapter One

had increased to be, on average, twice as high as an individual’s annual income (Yip and Hsiao, 2009). As a result, illness has become a major contributor to poverty: more than 20% of those living under the poverty line were in this situation because of disease or injury (Liu et al., 2003). Not surprisingly, because of the high costs of healthcare services, healthcare use has declined (Hu et al., 2011; Mao, 2007). Income has become a dominant factor in healthcare utilisation: in rural and urban China, 24% and 16% of the population respectively refused to be hospitalised due to financial burdens in 2003. Among those hospitalised, 32% of them discharged themselves against medical advice because they could not afford to stay (Yip, 2010). Overall, the reforms have widened health inequalities. In 2005, life expectancy was 6 years longer for urban citizens compared to rural citizens, and 15 years longer for people living in eastern China compared to those in western China (Chen, 2012). The rural to urban ratio of infant mortality rate has increased from 1.5 in 1981 to 2.1 in 1995 (Zhang and Kanbur, 2005). Rural mortality was 30% higher than urban mortality among older people (Zimmer et al., 2007), and the maternal and under–five mortality rates in rural areas were about three times greater than those in urban areas (WHO and DRC, 2005). In some rural areas, infant mortality even rose in the 1990s (Liu et al., 1998). At the same time, primary healthcare provision failed in its function to provide preventative care. On average 120,000 new cases of resistant tuberculosis have been reported each year (Shan, 2013). In 2012, there were 7 million cases of infectious disease that caused 17,000 deaths (DPES, 2005). Some of the previously nearly controlled infectious diseases, such as schistosomiasis, began to spread again (Blumenthal and Hsiao, 2005). Increasing health inequalities have not been the only consequence of the economic reforms. While the reforms from

The Cat that Catches Mice

31

1980s opened up new possibilities for some people, and created diverse but stratified labour markets, income inequality has also increased between households and individuals. The gap between China’s rich and poor has become one of the highest in the world, surpassing even that of the U.S. (Xie and Zhou, 2014). Based on data from the UNU-WIDER, the Gini coefficient in China was around .30 in 1980, but had nearly doubled to .55 by 2012 (Xie and Zhou, 2014). Further, the study showed that income inequality in China had far exceeded the average for other countries at a similar level of economic development. According to a newer report from the University of Michigan, China’s Gini coefficient was higher than the official figures given by the National Bureau of Statistics in China. In China there are now over one million people with personal wealth over 10 million yuan; there are 67,000 people with assets above 100 million yuan, and about 400 billionaires and billionaire families (Flannery, 2015; The Economist, 2015). The Hurun Report, a monthly magazine ranking and analysing the wealthiest individuals in China, interviewed 551 wealthy Chinese citizens who had personal assets of more than 10 million yuan in 2013. According to this, the rich individuals surveyed spent on average 1.77 million yuan a year on personal consumption, accounting for only 3% of their average wealth (Hurun Report, 2013). Earlier research has shown that the impact of income inequality on health is strongest for the poorest groups (Fang and Rizzo, 2011; Yang and Kanavos, 2012). According to the World Bank, 6.3% of the Chinese population still lives under the poverty line of $1.25 a day (PPP) (The World Bank, 2015). The vicepresident of the State Council Leading Group Office of Poverty Alleviation and Development, Zheng Wenkai, announced in 2014 that there were 82 million people living under China’s official poverty line of ¥2,300 a year (approximately $1 a day), or 200 million people according to the World Bank standard of $1.25 a

32

Chapter One

day (Zhang, 2014). By the international standard, people living in poverty accounted for approximately 15% of the total population in China (Zhang, 2014). Using night-time light as a proxy for poverty

measurement,

a

study

conducted

by

Norwegian

researchers suggested that there was no significant poverty reduction in China from 2005 to 2010 (Almås et al., 2014). One direct way of measuring the impact of poverty on people’s health is to look at malnourishment. During 2005 and 2007, 130.4 million Chinese people were undernourished, accounting for 10% of the total population (FAO, 2010). The prevalence of anaemia among children aged 6-12 months living in rural China was 35% in 2005, and the corresponding number was as high as 28% when urban children of the same age were included (Chang et al., 2007; MoH, 2012). Some studies reported prevalence rates of anaemia for children living in rural areas in China ranging from 20% to 60%, implying more than 10 million children were affected (Chen et al., 2005; Miller et al., 2012; MoH et al., 2005). A report from UNICEF referred to an estimation of 12.7 million stunted children in China – that is the same size as the population of Tokyo (Liang, 2013). The same report found that in poor rural areas in central and western China, one out of 10 children under 5 years old were stunted. In Qinghai Province, the prevalence of anaemia among children aged 6 months to 2 years old was above 70% (Liang, 2013). As

recently

as

2015,

malnutrition

among

adults

and

children/adolescents in China was estimated to be 6% and 9% respectively (NHFPC, 2015b). Childhood anaemia may also impair cognitive development, school performance and work outcomes, and also cause lower socioeconomic status throughout the life course (Haas and Brownlie, 2001; Lozoff et al., 2006; Stoltzfus, 2001; Yip, 2006).

The Cat that Catches Mice

33

Stress, genes and epidemiological transition The numbers of Chinese people diagnosed with cardiovascular diseases, hypertension and obesity increased during the reform period. The Chinese Centre for Disease Control and Prevention estimated that 200 million Chinese people were diagnosed with hypertension in 2010, about one in five adults (CGMH, 2011). Accordingly, above 40% of all deaths each year were caused by cardiovascular and cerebrovascular diseases; half of them were caused by high blood pressure (CGMH, 2011). Some official sources have shown that about 24% of those aged 15 or older in China had hypertension. Among youth and adolescents, about 15% of them

were diagnosed

with

high blood pressure,

and

approximately 30% of them were diagnosed with obesity (The Central People’s Government of the PRC, 2013). Previous

studies

have

recognised

occupational

and

psychological stress, which is a complex biopsychosocial situation, as a major health hazard worldwide; it contributes to different diseases, such as depression (Clays et al., 2007; Park et al., 2009), cardiovascular disease (Byrne and Espnes, 2008; Collins et al., 2005), abnormal blood pressure (Carroll et al., 2011; Matthews et al., 2004) and even obesity (Bose et al., 2009; Torres and Nowson, 2007). Similarly,

one

of

the

major

factors

contributing

to

cardiovascular diseases in China was recognised to be lifestyle and work pressure (Gong et al., 2006). The rising degree of competition, increased income inequality and labour market stratification in China may also have an impact on people’s stress levels, and may cause health problems. Studies of different occupation groups suffering from stress, such as nurses (Wu et al., 2010, 2007), teachers (Sun et al., 2011), offshore oil workers in the Chinese state-owned oil company (Chen et al., 2008), or intellectuals (Tian and Wang, 2005), showed that many Chinese

34

Chapter One

people were suffering from labour-market-related stress at different levels. Diabetes has also become an important public health challenge in China. The prevalence of Type 2 diabetes and impaired glucose tolerance was reported to be about 1% in a survey carried out in 1986 for urban citizens aged 25-74 in northeast China (Li et al., 1996). The numbers for the above-mentioned diseases increased to 2.5% and 3.2% in 1994 respectively (Pan et al., 1997). Some other reports found the prevalence of diabetes in the adult population to be much higher than the previously reported results. According to the International Collaborative Study of Cardiovascular Disease in Asia, in 2000–2001, the prevalence of diabetes for Chinese men and women aged 35–74 were as high as 5.2% and 5.8% respectively (Gu et al., 2003). Furthermore, the rate in northern China was higher than in southern China (7.4% compared to 5.4%), and higher in rural areas than in urban areas (7.8% compared to 5.1%) (Gu et al., 2003) 6. Some have argued that when considering non-communicable diseases, familial factors and inherited genetic variants may be more relevant in Chinese people than in Europeans (Ma et al., 2014). Although genetic pre-dispositions may be an important factor

in

the

development

of

cardiovascular

diseases,

environmental factors are also of major importance (see e.g. Cosselman et al., 2015; O’Toole et al., 2008). Epidemiological studies have identified several risk factors associated with the development of cardiovascular diseases, such as overweight and lifestyle (Chan et al., 1994; Manson et al., 1991; Ohlson et al., 1988), air pollution (Cosselman et al., 2015; Miller et al., 2007), environmental noise (Rosenlund, 2005) and the social-cultural context (see e.g. Chow et al., 2009), etc. A newer study by Yang et al. (2010) found that the age-standardised prevalence of diabetes was higher among urban residents (11.4%) than among rural residents (8.2%), while the age-standardised prevalence of pre-diabetes for urban residents was lower than rural residents (14.9% vs. 16%). 6

The Cat that Catches Mice

35

Non-communicable diseases such as type II diabetes are highly preventable (Chiasson et al., 1998; Eriksson et al., 1999). However, the management rates for such diseases in China are very low: less than half of those who were diagnosed by hypertension and diabetes in China were treated by health care providers in 2008. Because of a lack of prevention information, about 30% of those who had hypertension were not aware of their condition before diagnosis, and 54% had never received blood pressure tests by health providers (Meng and Tang, 2013). This may be a reflection of the Chinese healthcare system: despite the fact that the central government in China has started to focus more on wellbeing and human development, local governments have still prioritised economic development (Meng and Tang, 2013). Another argument is that China is facing a new epidemiological transition. This transition is defined by a shift in the burden of disease to non-communicable diseases. Because of economic growth and poverty reduction, the transition is being accelerated by a shift from malnutrition to over-nutrition (Popkin et al., 1993). The population faces a new set of health problems, including diseases of affluence, the impact of smoking and drinking, hypertension, environmental pollution-related health risks and the rise of infectious diseases (Yang et al., 2013; Zhou et al., 2016). Of the 8.3 million deaths in China per year, 7 million are due to non-communicable diseases (Yang et al., 2013). Epidemiologists have pointed out that the leading risk factors for disabilityadjusted-life-years in 2010 in China were cardiovascular diseases, cancers, low back pain, and depression. Other risk factors included dietary risks, high blood pressure, tobacco, and air pollution (Yang et al., 2013). In The Lancet, Zhou et al. (2016) investigated how patterns in cause of death had changed from 1990 to 2003 in 33 provinces in China, with 240 cases. They show that during this period, life

36

Chapter One

expectancy had improved, the death rate had fallen by almost a third, and that the major causes of mortality had changed. In 1990, the most important causes of death in most provinces were lower respiratory infections and preterm birth complications. In 2013, the leading causes of death were stroke, ischaemic heart disease, chronic obstructive pulmonary disease and lung cancer. Mortality due to infectious diseases, diarrhoeal disease and lower respiratory infections has fallen substantially. But deaths from various cancers have not been reduced, and some have even increased,

such

as

prostate

and

pancreatic

cancer

and

mesothelioma. Throughout the country an increased proportion of deaths were caused by road traffic injuries throughout the country, and in many provinces there has also been a lack of progress against HIV (Zhou et al., 2016). At the same time, challenges related to urbanisation and widening

social

and

regional

disparities

persist.

The

epidemiological transition is not unrelated to social contexts. For example, Lei et al. (2010) found that hypertension affects all socioeconomic classes independently. But people with a better education living in urban China are more aware of their condition, and there is therefore a higher treatment percentage and better control among this group. Further, public healthcare services fail to inform patients of their hypertension status. Yang et al. (2013) interpreted the rapid rise of non-communicable diseases as driven by urbanisation, rising incomes, and ageing. Richard Horton, the present editor-in-chief of The Lancet, commented on China’s transition, and suggested that China needs to place a greater emphasis on building a primary healthcare system, addressing environmental threats, and reducing inequalities (Horton, 2015).

Wellbeing Psychological health is an important indicator for people’s wellbeing. In China, mental and behavioural disorders are

The Cat that Catches Mice

37

important risk factors for health, and depression was the second leading cause of disability and alcohol use disorders (Yang et al., 2013). One may expect that when incomes rise in a country – and the population therefore experiences greater living standards – the degree of wellbeing and happiness will also rise, especially at a low level of per capita income (Diener & Diener 1995). But despite China’s fast growth in the last two decades, people’s life satisfaction trajectories are in decline (Brockmann et al., 2008; Easterlin et al., 2012; Kahneman and Krueger, 2006; Knight and Gunatilaka, 2011). In the Gallup-Healthways country wellbeing ranking for 2014, China ranked at 127 out of 145 countries, with only 7.9% of the population thriving in more than three elements of the index. By comparison, in Denmark 37% of the population was thriving, in Russia 23%, and in Japan 13.5% (Gallup-Healthways, 2015). In general, economic growth seems to correlate to higher inequality in wellbeing. According to some scholars, this trend is related to a rise in unemployment and the dissolution of the social safety net; this worsening life satisfaction hits the lowest socioeconomic groups the hardest (Easterlin et al., 2012). In the World Values Survey, the proportion of Chinese people that reported being ‘very happy’ had dropped by more than a half, from 27.5% in 1990 to 11.5% in 2000; while the proportion who were ‘not very happy’ or ‘not at all happy’ rose from 15.8% in 1995 to 21.8% in 2000 (Appleton and Song, 2008; Brockmann et al., 2008). Kahneman and Krueger (2006) presented results from the Gallup wellbeing survey in China, and showed that the percentage of respondents who were satisfied or very satisfied fell by 15% from 1994 to 2005, while the percentage of people who were dissatisfied or very dissatisfied rose monotonically. A study by Easterlin and Sawangfa (2010) found that different surveys showed a similar trend – the average life satisfaction

38

Chapter One

score has fallen in China: this is from 2.82 in 1997 to 2.67 in 2004 (the Gallup survey), from 3.73 in 2003 to 3.68 in 2007 (the Asiabarometer survey), and from 6.83 in 1995 to 6.76 in 2007 (the World Values Survey). This was summed up by Knight and Gunatilaka (2014, p. 46) in the table below.

Table 1. Mean life satisfaction in China 1990s-2000s

In an interview by the Southern Daily, Sociologist Sun Liping from Tsinghua University said: ‘Happiness is relative, comes from comparing to others. What matters is not what we have, but what we have that our neighbours don’t… The Government’s first priority is not to develop economy, but to maintain fairness and justice’ (Zhao and Si, 2011).

This perspective is widely accepted in academic debates; the impact of relative position on a person’s wellbeing is seen as important. By comparing themselves to others and finding their rank in a distribution, individuals are making judgements about their overall state of wellbeing. One example of this is that ruralto-urban migrants in China often report lower average subjective wellbeing than others – even compared to rural households (Knight and Gunatilaka, 2010). It may be due to their false

The Cat that Catches Mice

39

expectations about their urban conditions and urban aspirations, but it may also relate to their disadvantaged economic position in the cities. When comparing migrant workers to urban citizens, socioeconomic inequality may be a reason for their lower scores on happiness and life satisfaction (Knight and Gunatilaka, 2010). One especially vulnerable group in China is older adults. A major challenge in today’s China is its increasing socioeconomic inequality, which may also affect the elderly. As the proportion of elderly people grows, there will be challenges relating to the wellbeing of more than 100 million people aged 65 and over (Feng and Xiao, 2007). The sharpest declines of life satisfaction with age are found in Eastern Europe and the former Soviet Union countries (Deaton, 2008). Life evaluation among elderly people in these transition countries is particularly low. Here research shows that people have a sense of having lost a system that provided pensions and healthcare, a system that was meaningful in their lives (Steptoe et al., 2015). A similar pattern can be found in China. Life satisfaction for people over 60 rapidly reduces with age (Figure 1). According to the 2006 China national survey, only about half of the urban respondents and one third of the rural respondents reported feeling happier when compared to others (Lou and Gui, 2011). Chinese people born in and before 1950s have spent more than 25 years of their lives under socialism, and experienced a time of rapid social change. On the one hand, the market reforms may have brought improvements in nutrition and living conditions that are critical to maintaining their well-being. On the other hand, they may be unprepared for the new, unpredictable, and competitive market-oriented society. As adults, they may have difficulty adjusting to the change, and their wellbeing may be compromised. Further, unlike younger generations that have grown up in an environment in which inequality is lent legitimacy,

40

Chapter One

the wellbeing of older adults may be negatively influenced by the contrast between the more egalitarian past and unequal present.

Figure 1. Age and wellbeing, CHNS 2006-2009.

Inequality by geography The economic reform in China started in the eastern coastal areas. The famous slogan from Deng Xiaoping was to ‘permit some people and some regions become prosperous first, for the purpose of achieving common prosperity faster’ (Literature Editorial Board of the CPC Central Committee, 1993b). Meanwhile, western China as a region has been lagging behind. Economic development in China has been highly imbalanced. Research shows that two of the major contributors to China’s high income inequality are regional disparities and the rural-urban gap (Xie and Zhou, 2014). The western region covers an area of 6.9 million square kilometres, which amounts to 74% of the whole national territory.

The Cat that Catches Mice

41

The population was 367 million in 2002, which was 29% of China’s total national population (Lu and Neilson, 2004). In 2002, the combined economy of western China accounted only for 17% of total GDP in China. Income per capita in the west amounted to only 40% of that in eastern China (Lu and Neilson, 2004), and the average income in the west was only half of that in east (Gustafsson, Shi and Sicular, 2008). Most of the western provinces have weak communication and underdeveloped infrastructure. Agriculture is important in western China. According to statistics from 2005, almost 68% of the workforce works in the agricultural sector (Yao et al., 2009). As a result of the huge regional differences, the Chinese government implemented the ‘Open up the West Strategy’ (Xibu Da Kaifa) at the beginning of the 2000s. The central goals of the campaign were to develop the western and central regions, increase fairness and social equity, improve political stability, alleviate relative deprivation, and reduce conflict between regional, social and ethnic groups (CCCPC, 2000). The state-owned enterprises comprise a larger share of the local economy in the western regions, compared to the coastal region. The development strategy consists of economic reform, privatisation, the promotion of foreign direct investment, and sustainable development (Chow, 2007). A central measure was the privatisation of parts of the state sector in order to achieve higher productivity and efficiency. Thus, it was a state-driven market reform. The political economist McNally has drawn attention to how the development strategy promotes important aspects of the market economy in the western provinces: ‘The Open Up the West campaign is pushing the structural transformations that are already gripping coastland provinces westward. The infrastructure for accelerated capital accumulation is being put in place, including political and economic support by

42

Chapter One

the state for market forces, property rights and private holders of capital.’ (McNally, 2004, p. 115)

A central question is how these structural changes have affected inequality, both between regions, and within western China. Some scholars have argued that intraregional inequalities in the 2000s were rather stable (Fan and Sun, 2008; Groenewold et al., 2008). Groenewold found that inequality has developed in different directions within the western region. By distinguishing between western regions, he found that inequality has fallen in the north west and increased in the south west (Groenewold et al. 2008). As to inequality between the western and eastern regions as a whole, some have found that inequality on the whole has decreased (Mao, 2011). However, there are also studies that show that inequality between the regions rose after the campaign, in terms of GDP growth and income (Wu, 2007). The different regions in China are socially, culturally and economically diverse, and there are also important variations within the provinces. Within western China, there are still important regional heterogeneities. For example, Xinjiang has had the highest economic growth and the highest GDP per capita. Other provinces have highly developed industries, such as Sichuan, Shaanxi and Chongqing. Some provinces, on the other hand, are very poor, such as Guizhou and Gansu (Goodman, 2004). Furthermore, there are important disparities between developed economic centres and lagging peripheries, and between different ethnic groups. Regional disparities are also large, and poor and rural areas face many difficulties in developing their economies, connected to social issues such as poverty, low social services and welfare insurance, unstable employment, income and distribution, food prices, etc. Therefore, it is important to take regional heterogeneity into account. Different administrative levels in China, such as the province,

prefecture,

and

county,

may

capture

different

The Cat that Catches Mice

43

characteristics inside each unit. Even within these geographical units, there may also be huge differences, both in terms of inequality, and its influence on individual outcomes, such as health and well-being. In Chapter Three, I will look more closely at the methods for controlling regional heterogeneity and eliminating certain time-invariant factors when exploring the correlation between income inequality and other variables.

Chapter Two

Theories: Feeling the Stones ‘It is true that to reform we must rely on theoretic studies, economic statistics, and economic forecasts. But it is more vital that we begin reform by conducting experiments, and by summing up our experiences at all times. That is to say, we cross the river by feeling for the stones.’ - Chen Yun, 1980 7

A famous Chinese idiom, ‘crossing the river by feeling the stones’, describes the Chinese approach to market liberalisation and reform. It is a process that involves finding a feasible way (feeling the stones) to walk along the theorised development path (crossing the river). In this chapter, I will review the theoretical debates, and look at perspectives and previous studies that may help us to understand income inequality in China. The theories used in the dissertation relate to the two major topics. The first of these is how the process of market transition, measured by privatisation, may influence income inequality. The link between privatisation and income inequality operates via different factors such as education, occupation and the household 7

From a speech given by Chen Yun at the Central Work Conference, see

(CCCPC Party Literature Research Office, 1986, p. 279).

46

Chapter Two

registration system, and in this debate market transition theory is in the centre of the theoretical discussion. The second issue is, how income inequality may have an impact on individual outcomes, such as health and wellbeing. When considering the relationship between income inequality and health, I will discuss different hypotheses, such as the income inequality hypothesis, absolute income hypothesis, and relative income hypothesis, in relation to each other. When considering the development of life satisfaction, happiness and wellbeing following the rising income inequality, I focus on theories drawn from happiness studies. The debate between the ‘tunnel effect’ and relative position hypothesis will also be discussed. The general framework of the dissertation can be seen in Figure 2:

Figure 2. Theories and framework of analysis

This chapter is organised as follows: First, I will look at market transition theory, and how marketisation / privatisation may affect income inequality through different mechanisms, such as stratifying educational attainment, occupational class diversity in the labour market and the household registration system, which automatically divides rural citizens from urban citizens in terms of eligibility, entitlement, etc. After that, I will move on to look at

47

how income inequality is further connected to individual health, drawing upon theories from the Wilkinson hypothesis, which I will discuss in light of other inequality-health related hypotheses. Finally, I will discuss literature and theories concerning wellbeing.

Market transition theory Market transition and income inequality have been discussed in studies of post-socialist countries (cf. Bandelj and Mahutga 2010; Oberschall 1996; Stark 1991). Nee’s (1989, 1991, 1992, 1996) market transition theory prompted a wider debate about how changes in the reform period have affected social stratification and income distribution in China. According to the theory, the growing market sector challenges the position of the public sector. Market transition theory proposes that as the market grows, market mechanisms will gradually replace political mechanisms with regard to income distribution (Nee and Cao, 2005). An important legacy from the planned economy is the state redistribution system and the power connected to privileged positions in this system. This has also continued to be important in reform China. However, the growth in the market and the private sector creates changes in the distribution of power and income in China. The hierarchy in the public sector is based on political capital and positions, while the main principles in the market sector are human capital, entrepreneurship, control of capital and investments. Individuals with market power will gradually gain more advantages, while the advantages of individuals in the public sector will gradually be reduced. This

development

of

market

transition

has

important

consequences for income inequality. But how are market transition and income inequality related? This relationship is often discussed in economic studies focusing on the market’s impact on development. Some argue that the process of marketisation and privatisation in China, as an important aspect

48

Chapter Two

of the opening-up of China’s economy, will increase efficiency and further promote development. Inequality in China is expected to rise initially, but will later be reduced through a ‘trickle-down’ effect (Holbig, 2004; Li and Luo, 2007, 2010). Supporters of marketisation and privatisation often argue that a freer market, combined with a profit motivation, will lead to new initiatives and competition. These will in turn result in increased managerial productivity and effectiveness, greater profit and a better quality of products and services. From this perspective, privatisation is viewed as a rational and logical response to the problems of increasing state expenses, inefficient state-owned enterprises and the negative consequences of state control, for example, in terms of limiting human initiative and creativity (Collyer, 2003; Starr, 1988). However,

some

other

scholars

argue

that

the

market

mechanism in itself is increasing income inequality (Chen, 2010a, 2010b; Zhou and Zhao, 2004). Market transition theory, on the other hand, argues that inequality in the initial phases will be reduced due to the loss of power among political elites, but that it will later rise again due to the emergence of new economic elites in the form of entrepreneurs (Nee, 1996; Nee and Matthews, 1996). The reason for this is as follows: under state socialism, the state sector controls the distribution of wages, income, and goods. The advantaged position of being employed in the state sector is reduced when there is a higher degree of privatisation, while the private sector acquires a higher economic position in the market economy (Bian and Logan, 1996). The process of market transition provides an alternative for entrepreneurs and other people who did not have positions in the redistribution system. The income gap between the dominant state sector and the previously suppressed private sector would therefore decrease when the market sector process of privatisation is more widespread. However, later on, as the market sector gradually becomes more

49

important, this sector has considerable potential for generating higher inequality levels than before the reforms, particularly when the sector has been legitimized and actively supported by the state, as was the case in China from the mid-1990s and onwards (Sun, 2008). Several factors work together to affect how privatisation works. These include the type of goods and services that are to be privatised, the parties involved in the process of privatisation, and the transition cost incurred by institutions (Araral 2009). Similarly, the processes of privatisation can also impact the outcome of privatisation, as can the economic, political and historical context of the region (Birdsall and Nellis 2003).

Occupation, education and the hukou system The relation between market transition and income inequality in China is a macro-level linkage. In order to look more closely at the underlying factors that bind these two together, we may draw upon different institutional perspectives; these refer to occupation, education and the household registration system in the Chinese labour market. Both labour market structure and occupational class have been important contributors to higher wage inequality in the Chinese market transition process. Increasing stratification of the labour market has taken place in the reform period. Privatisation can lead to a more stratified occupational structure, both by creating increased advantages for skilled workers, and by increasing the number of service sector jobs that are characterised by low-income and insecurity (Gaetano and Jacka, 2004; Schucher and Hebel, 2006; Zhang, 2002). Occupations – whether defined as political or market-oriented capital – also have a significant effect on income inequality (Bian and Logan 1996). Knight and Song (1995, p. 105) have found that increased wage differences in urban China were caused by occupational differences in China’s economic reform.

50

Chapter Two

Park et al. (2004) argued that rising rates of return to skilled labour was one of the most important factors contributing to increasing wage inequality in China. Others have pointed out that the working class and agricultural class in China have faced greater disadvantages in the market-oriented economic reform period (Blecher 2005). In addition, education raises an individual’s income level both directly and indirectly as it improves his or her chances of getting a well-paid job (Knight and Song, 1995, p. 105). According to human capital theory and theories of skill-biased technological change, individuals with higher education levels are better rewarded in more market-oriented and privatised societies because of the need for higher-skilled labour (Becker, 2009; Lemieux, 2008). This pattern seems to have developed in China as well. Rising urban income inequality is connected to rising rates of return to education, and the work units (danweis) are still central for the distribution of resources (Zhou, 2000). A growing private sector significantly influenced income levels in the reform period (Zhou, 2000). Related specifically to the Chinese context, market transition theory emphasises increasing rates of return to human capital and decreasing rates of return to political capital in the transitional Chinese society. In other words, the higher the degree of privatisation, the higher the rate of return to education (Nee and Cao, 2005, p. 47). This trend has been found in several studies. In an analysis of urban retrospective panel data, Zhou found that there are significant increasing rates of return to all levels of education in urban private sectors (Zhou, 2000, p. 1163). Using 1995 and 2002 CHIP data, Gustafson et al. (2008) concluded that the higher rates of return to education are one of the most important contributors to increased income inequality in China. Yet another study concluded that rates of return to education increase when there is a higher degree of marketisation, especially in less-developed and

51

low-income provinces (Li 2003). The impact

of market

transition and

privatisation on

differences between urban and rural citizens continues to be debated.

Economists

often

argue

that

the

process

of

modernisation and privatisation will contribute to reducing regional inequality in China through the ‘trickle-down’ effect (Dollar, 2007). However, many empirical studies argue that the market transition process is increasing urban-rural disparities in China (Gustafsson, Shi and Sicular, 2008, p. 23; Wahl, 1998). These disparities may be reflected at the individual level in the household registration system, hukou system, in China. Hukou status is given to each individual at birth. It systematically separates agricultural and non-agricultural citizens and their different welfare rights. Usually, only residents with local hukou registration have access to welfare benefits in their area, while people living outside their registered residency area do not have access to the same rights as local residents. This has been especially problematic for rural-to-urban migrants, who are struggling for their rights to healthcare, pensions and to obtain schooling for their children. In recent years, the state has launched several measures to reduce rural-urban disparities and increase welfare support for people with a rural hukou. However, the focus remains on economic development as a solution for overall

social

problems;

the

envisaged

solution

involves

privatisation measures and an opening up to private actors and sectors.

Income inequality-health hypothesis After discussing factors that may contribute to a higher degree of income inequality, we now move on to look at potential outcomes of income inequality. One of the direct measures of the quality of human life is health. Due to better living standards and

52

Chapter Two

technological

development,

general

population

health

has

improved worldwide. However, some indicators suggest that health development in China has slowed down after the economic reform, and some health performance indicators have even dropped. At the same time, inequality is rising. This relation between income inequality and health outcomes has been examined in different social settings. Previous studies have identified different ways in which income inequality is connected to health. The earlier studies focused on absolute individual income as having the most important positive effect on health (Gravelle, 1998; Preston, 1975; Rodgers, 1979). Wagstaff and van Doorslaer (2000) refer to this as the absolute-income hypothesis, in an article in which they reviewed literature on the effect of income inequality on population health. According to this hypothesis, an individual’s absolute income directly influences her health, and the curve is increasing and concave down. In the same review, Wagstaff and van Doorslaer (Wagstaff and van Doorslaer, 2000) identified other pathways through which relative income may influence health. Three pathways are central to the debate: (1) income differences to the population mean (the relative-income hypothesis), (2) individuals’ ranking and relative position

in

the

income

distribution

(the

relative-position

hypothesis) and (3) income distance to poverty (the deprivation hypothesis). The income inequality hypothesis says that an individual’s health is directly affected by income inequality. The question was initially raised by Preston (1975), who observed that life expectancy was more disassociated from income growth. He suggested that position in the income distribution might explain the relation between health and income. Rodgers (1979) found the link between income inequality and life expectancy by studying 50 countries, and elaborated on the idea of aggregating a curvilinear relation between individual income and health to a negative

53

macro-level association between income inequality and population health. Wilkinson (1992) further developed the income inequalityhealth hypothesis by studying three international datasets, and found a direct link between income inequality and health in developed countries, which was independent of individual income. This was explained by psychosocial mechanisms, in which more egalitarian societies promote greater social integration, lower stress and better health for individuals. Unequal societies, on the other hand, instead promote problems connected to low social status, poor social relations and social cohesion, low investment in social capital and harmful health behaviours such as smoking and comfort food consumption (Cohen et al., 1997; Marmot et al., 1991; Wilkinson, 1996, 1999, 2000). Although some of the literature has contradicted nonetheless

the

income

considerable

inequality support

for

hypothesis, this

there

hypothesis.

is By

reviewing 155 articles based on epidemiological studies, Picket and Wilkinson (2015) found that 131 of them supported or partly supported the hypothesis that income inequality harms health causally. In a meta-analysis that includes about 60 million subjects from 9 cohort studies and about 1.3 million subjects in 19 cross-sectional studies, Kondo and colleagues identified a ‘modest’ impact of income inequality on health. It indicated that potentially 1.5 million deaths could be averted in 30 OECD countries by reducing the Gini coefficient to below .3 (Kondo et al., 2009). In addition to the psychosocial environment interpretation, some studies emphasise a ‘pollution effect’ of income inequality on health that occurs due to contextual factors. Accordingly, income inequality

affects

public

health

negatively

by

influencing

structural, social and political patterns (Subramanian and Kawachi, 2004). For example, inequality may cause residential and spatial segregation. It may also affect the process of policy-

54

Chapter Two

making and welfare politics. By reviewing multilevel studies that include

individual

income,

income

inequality

and

health,

Subramanian and Kawachi (2004) observed that geographic scale is important. Studies based on state-level units, where it is easier to identify the political mechanisms and contextual impacts, tend to find an inequality-health relationship, A similar point has been made by Wilkinson and Pickett (2006). They argued that one of the reasons researchers do not find evidence for the income inequality hypothesis is that they do not use sufficiently large analytical units to capture the social class differences and salient social heterogeneity in a society. This approach has been much debated. Gravelle (1998) described the relation between income inequality and population health reported by Wilkinson as a statistical artefact, and claimed that the aggregate relation is caused by the underlying concave relation between individual income and health (see Figure 3). Deaton (2003) questioned the importance of income inequality when considering health. He proposed that it might be more interesting to look into how income, not income inequality, affects health. Similarly, Deaton (2003) argues that the relationship between income and mortality does not necessarily change with economic development. Income inequality itself is not a major determinant of population health; it is low income that matters. According to Deaton, income inequality is important only because it is correlated with poverty, and the only exception is the case of homicide, where income inequality itself affects mortality. Lynch and colleagues conducted a systematic review of 98 aggregated and multilevel studies about the relation between income inequality and health in 2004. Among the 98 studies, there were 33 that found negative associations, 25 providing mixed evidence and 40 with positive associations. Of the studies that found positive associations, the strongest evidence came from the U.S. (Lynch et al., 2004).

55

Figure 3: Income inequality, income and mortality (Source: Gravelle 1998)

Additionally, a literature review from Wagstaff and Doorslaer (2000) concluded that there was strong support for the absoluteincome hypothesis, little support for the relative-inequality hypothesis, and no support for the income-income hypothesis. This interpretation, which claims that an aggregate association between income inequality and health is only a reflection of the nonlinear concavity between income and health at the individual level, is often referred to in the literature as the individual income interpretation (Lynch et al., 2000). There is one approach that combines individual and structural factors when explaining the effect of income inequality on health, called the ‘neo-materialist’ approach. Instead of focusing on only income or income inequality, the neo-materialists emphasise the importance of individuals’ resources and experiences from the material world and how they may have an impact on individual health (Lynch et al., 2000). Lynch and colleagues (2004) called

56

Chapter Two

attention to the structural, political and economic processes that may generate inequality. They pointed out that individual income is determined by people’s education, skills and efforts, and that inequality is generated by structural, political and economic processes. They stated: ‘…the total effect of income inequality on health reflects both a lack of resources held by individuals, and public underinvestment in the human, physical, health, and social infrastructure’ (Lynch et al., 2004). Further, they pointed out the differences between income inequality and income is between ‘a characteristic of a social system’ and ‘characteristics of individual person’. They also noted: ‘there may be different determinants at the individual and population levels; sometimes the factors that cause sickness individuals are different from those that cause sick population’ (Lynch et al., 2004). Coburn (2004) brought class, neo-liberalism and welfare regimes into the discussion, in order to understand how inequality may affect health in a historical and cross-sectional way. He viewed neo-liberalism as a new phase of capitalism that widens class differences and generates higher income inequality. According to him, it has also contributed to unequal access to both welfare benefits and health-relevant resources. The neo-liberal policies are further correlated with the balance between welfare regimes and the market. A higher degree of neo-liberalism promotes a less generous welfare state, which leads to higher income inequality, more unequal access to social resources and welfare, and greater poverty. When welfare regimes are scaled back, market mechanisms become more dominant; health related resources such as healthcare services depend more on individual income and wealth. This may lead to a lower degree of social cohesion and trust, and may also have a negative impact on health and well-being (Coburn, 2004).

57

Epidemiologic transition The theory of epidemiological transition, developed by Abdel Omran, focuses on ‘the complex change in patterns of health and disease, and on the interactions between these patterns and their demographic,

economic

and

sociological

determinants

and

consequences’ (Omran, 2005). According to this theory, mortality is characterised by three historical phases: the ‘age of pestilence and famine’, the ‘age of receding pandemics’, and the ‘age of degenerative and man-made diseases’ (Omran, 1971). In addition, the renewed decline in mortality from ischaemic heart disease, which started around 1970, has sometimes been referred to as a fourth stage of epidemiologic transition, the ‘age of delayed degenerative diseases’ (Olshansky and Ault, 1986). According to the theory, different countries in the world were categorised into three models. The transition started early in Western Europe and North America, which was called the ‘western’ or ‘classical model’ of the epidemiologic transition. The shift started later in some other countries (such as Japan and Eastern Europe); this was called the ‘accelerated model’. The transition started latest in many third world countries, called the ‘delayed’ or ‘contemporary model’ (Omran, 1971). The decline in mortality is closely linked to the process of modernisation, and there are three major categories of disease determinants:

the

eco-biologic

determinants,

socioeconomic,

political and cultural factors, and medical and public health determinants (Omran, 1971). While socioeconomic development was recognised as the most important factors for improving living standards and reducing mortality in the western model, medical technologies and public health were considered more important for the accelerated and contemporary model. One may argue that income inequality is a component in and unintended consequence of the economic transition, and that it further mirrors the ongoing epidemiological transition. However,

58

Chapter Two

Wilkinson (1994) demonstrated the relation between income inequality and life expectancy in wealthier industrialised countries which have passed through

the epidemiological

transition, and where the cause of mortality has changed from infectious diseases to chronic diseases. Observing that lifeexpectancies had increased in countries with lower income inequality, while countries with higher inequality had fallen behind in terms of life-expectancies, Wilkinson argued that whereas material deprivation leads to poverty and infectious disease, social disadvantage provokes stress and chronic disease (Deaton, 2003; Wilkinson, 1992, 1996). Similarly, Gaylin and Kates (1997) criticised epidemiological transition theory for being overly optimistic about the demise of infectious diseases, and argued that it ignored the importance of social inequalities. They suggested that socioeconomic inequality should be brought to the centre of the analysis, and that epidemiologic differences between population subgroups should be taken seriously. Cook and Dummer (2004) attempted to apply the model of epidemiologic transition to China. They documented that before 1949, infectious disease and natural catastrophes such as flooding and drought were the major causes of mortality in China, corresponding to Omran’s ‘age of pestilence and famine’. From the 1990s, China experienced an important overall decline in infectious diseases of poverty and underdevelopment, such as cholera, dysentery, hepatitis, and typhoid. In the last decade, China has experience increases in death rates from chronic diseases, including cancers and heart disease. These shifts reflected Omran’s model quite well. However, Cook and Dummer emphasised that epidemiologic transition in China is segmented, with large variations between rich and poor, and between urban and rural areas, as well as an emerging ‘medical poverty trap’

59

that reinforces the widening inequalities of access to healthcare services. They wrote: ‘The health situation in China reflects a new, late stage, epidemiological transition phase, where the transition from diseases of poverty to diseases of affluence has not reflected smoothly the economic and development transition. Consequently, China faced health issues related to an aging and increasingly affluent population, combined with problems caused by rapid urbanization, emerging and re-emerging infectious diseases and widening inequality in health and health care’ (Cook and Dummer, 2004, p. 341).

Wellbeing and inequality Research on subjective wellbeing started in psychology in the 1960s, with an empirical focus on self-reported happiness and life satisfaction. The terms ‘subjective wellbeing’, ‘happiness’ and ‘life satisfaction’ are often used synonymously (Frey, 2008; Veenhoven, 1984, 2008, 2012). Psychologist Diener (1984) referred to different approaches to defining wellbeing. A normative definition of wellbeing is based on a value framework of desired happiness. Such a definition implies that happiness and wellbeing is relative to some standard that may be different in different cultures and contexts (Coan, 1977; Tatarkiewicz, 1976). Another way of defining subjective wellbeing asks how people evaluate their lives; this approach stresses the standards that people consider as constituting a good life (Diener, 1984; Diener et al., 1999). A third approach emphasises positive experiences and involves everyday discourses of happiness (Bradburn and Caplovitz, 1965). Diener and colleagues define subjective wellbeing as ‘a person’s cognitive and affective evaluation of his or her life’ (Diener et al., 2002). When discussing the definition of subjective wellbeing, Diener (1984) made three observations: First, the term is not only connected to objective conditions or moral judgements, but most importantly, it involves individual’s subjective evaluations of their

60

Chapter Two

own life satisfaction. Secondly, there are positive measures that indicate wellbeing. Third, subjective wellbeing includes a ‘global assessment of all aspects of a person’s life’ (Diener, 1984). Further, psychologists view this term as a broad category with different components, including people’s emotional responses (positive or negative affect), domain satisfactions and life satisfaction (Diener et al., 1999). There has been increasing interest in subjective wellbeing within sociology as well. In the article ‘Sociological theories of subjective

well-being’,

Veenhoven

(2008)

discussed

how

sociologists use the term in different ways. A sociological definition of subjective wellbeing goes beyond cognitive evaluation: it also includes an overall appraisal of life. Sociologist Veenhoven drew upon two sources of information when considering subjective wellbeing: ‘cognitive comparison with standards of the good life (contentment)’, and ‘affective information from how one feels most of the time (hedonic level of affect)’. In

contrast

to

the

wellbeing

approach

within

positive

psychology, a sociological focus on wellbeing often refers to problems

and

subjective

experiences.

Instead

of

asking

individuals how happy they feel, sociologists are more interested in what they feel, and why they feel that way. Happiness and wellbeing is not only viewed as a state or condition of mind, but is also embedded within the broader societal and institutional context. The combination of society and individual perceptions makes the conceptualisation of subjective wellbeing more analytical (Veenhoven, 2008). Further, the definition also encompasses affective experience, implying that wellbeing studies do not just look at awareness as a social construction. This, again, is connected to the questions sociologists ask: we want to know why a subjective judgement is what it is. The experience of our own wellbeing is not that of an isolated individual, but is mixed with social influences and a community-based common ground of

61

shared views (Veenhoven, 2008). Subjectivity in the study of life quality is often similar to subjectivity in the positivistic methodological approach: it is developed through interpersonal influence and communication (Baltatescu, 2007; Rahn et al., 1996). To summarise, the state of subjective wellbeing cannot be separated from one’s social setting, or the interpersonal context. Social comparison theory directly responds to how subjective wellbeing can be influenced by other people in a particular social setting. The main idea of this theory is that life satisfaction depends on social comparison: The smaller the discrepancies, the happier we are, and the higher subjective wellbeing we have. Social comparison is one of the major explanatory factors when studying how individuals’ subjective perception of their life quality is connected to rising income inequality. Many studies have focused on economic conditions and their positive effect on subjective wellbeing and happiness (see e.g. Easterlin 2001; Firebaugh and Schroeder 2009). Empirical studies have confirmed this association for Europe (Caporale et al., 2009; Di Tella et al., 2003), the U.S. (up to a certain threshold) (Kahneman and Deaton, 2010) and China (Appleton and Song, 2008; Knight and Gunatilaka, 2011), and some even suggest that the correlation is universal (Deaton, 2008; Easterlin, 2001). However, scholars have observed that subjective wellbeing is not necessarily connected to a rise in per capita income. Easterlin (1974) observed that the fast economic growth in the U.S. did not add to subjective well-being. Some scholars suggest that when inequality rises, the income of rich groups rises comparatively more, and this may explain why the degree of happiness in the U.S. has not risen with its economic growth (Fischer, 2007). Further, satisfaction comes from having more income compared to one’s peers (Easterlin, 1974, 1995, 2001).

62

Chapter Two

There are different theories and empirical studies about the link between wellbeing and unequal income distribution. Some argue that higher income inequality indicates a higher degree of economic development and greater employment opportunities, and therefore raises individual life satisfaction (Clark and Senik, 2011; Marshall and Firth, 1999). Others point out that change in socioeconomic position and psychological stress can generate a strong feeling of relative deprivation, and hence decrease individual life satisfaction (Alesina, Di Tella, & MacCulloch, 2004; Wilkinson, 1996; Zhao, 2012). One approach emphasises the ‘tunnel effect’. In a study, Alesina and colleagues (2004) found that the poor and left-wingers are less bothered by inequality in the U.S. They explained that the effect of income inequality varies according to people’s perceptions of mobility, especially when income inequality is interpreted as a signal for opportunity. When people observe upward mobility for others in the income distribution, their expectation of their own future social mobility increases, which makes them happier (Hirschman & Rothschild, 1973). Critics argue that although some people may have a higher expectation of mobility in a more unequal society, others may not. If the upward mobility happens mostly with poor people, it does not then necessarily result in increased inequality – it may even reduce inequality – and some people may even fear mobility, and thus being less happy if mobility happens (Verme, 2011). Therefore, the relation between income and wellbeing may vary for people with different levels of income. Some findings indicate that beyond a certain income threshold, further income is related to wellbeing to a lesser degree (Diener & Seligman, 2004; Di Tella & MacCulloch, 2008; Frey & Stutzer, 2002; Kahneman & Deaton, 2010; Layard, 2003). Critics argue that it is a log-linear relationship that shows a greater increment when measuring happiness for the poor than for the rich for each additional income

63

unit, instead of having a threshold or a ‘satisfaction point’ of income (Deaton, 2007; Easterlin, 2001; Stevenson and Wolfers, 2008, 2013). A social comparison perspective may also be useful for explaining this phenomenon. For example, people in the middle and higher rank of the income distribution may have a higher degree of life satisfaction when they make comparisons to poorer individuals,

because

of

their

advantageous

socioeconomic

situation. When they compare themselves to richer individuals in their own group, they anticipate being as rich as the other group members by virtue of the fact that they are members of the group (Hirschman and Rothschild, 1973). Therefore, higher incomes may indicate better wellbeing. The opposite may happen with the poorest income group. The higher their incomes, the closer they come to the boundary of a higher income group, however, this comparison with richer people makes them depressed. Even when they compare themselves with people in their own income group, they will notice that they are trapped in a poor social position.

Chapter Three

Methods and Data T-stat looks too good. Use robust standard errors— significance gone. - Keisuke Hirano

This chapter presents data, analytical framework and methods used in the dissertation. The three articles in the dissertation are based on two different surveys: Monitoring Social and Economic Development of Western China (MEDOW) and the China Health and Nutrition Survey (CHNS). The data are analysed with quantitative methods, using different statistical techniques. The choice of methods is based on considerations regarding what the thesis was aimed to capture. For example, when studying the relation between privatisation and income inequality, the aim was to bring together the macro-level structures and micro-level individual characteristics. In this case, hierarchical linear modelling is capable of addressing both macro-level linkages between privatisation and inequality, and at the same time going into depth on stratifying mechanisms such as education, occupation and the household registration system. Further, when studying what impact income inequality may have on health and subjective wellbeing, the studies were designed to take regional

66

Chapter Three

heterogeneity into account; I therefore chose panel data fixedeffects models. This section begins with a presentation of the survey data, followed by a general overview of the analytical framework. After this, I will take a closer look at the different statistical techniques used in the thesis. The chapter will end with a brief discussion of causality, correlation and mechanism.

Data Medow In paper 1 I use the cross-sectional survey data set Monitoring Social and Economic Development of Western China (MEDOW), collected in 2004-2005. The Medow survey was conducted by the Fafo Institute for Applied International Studies in Norway, in cooperation with the National Research Center of Science and Technology of Development (NCRSTD). 8 The survey includes data on population composition, health, household economy, work and employment, education, living conditions, migration, infrastructure and agriculture. A total of 167,000 individuals in 44,738 households participated in the survey, which covers 11 provinces in western China: Gansu, Xinjiang,

Yunnan,

Ningxia,

Guangxi,

Guizhou,

Qinghai,

Chongqing, Sichuan, Inner Mongolia and Shanxi. The only western province not included in the survey was Tibet (See Figure 4 for surveyed provinces). The provinces in Medow covered 5.6 million square kilometres, 58% of the total geographic area of China. The sampling process was done in several steps. Each of the provinces was divided into 18 replicate sample areas, drawn from In 2007, the NCRSTD was reorganised and renamed as the Chinese Academy of Science and Technology of Development (CASTED). 8

Methods and Data

67

14 primary sampling units based on lists from the neighbourhood committees and townships. The primary sampling units were selected by a probability sampling with inclusion probabilities proportional to the population of each unit. In the 14 primary sampling units, 16 rural households and 20 urban household were chosen. The participation rate was high, and the sample can be assumed to be representative for households and individuals in Western China. There were in total 44,738 households selected for interview, of which 41,695 households were able to participate; among these, 41,222 households completed the interviews. Of all sampled household, 94% were interviewed. Only .96% refused to participate. The reasons for non-response or incomplete interview were 1) the interviewers could not find the household location, 2) no household members were at home, or 3) that the interview was interrupted. Some of non-responses are because of confused or obviously wrong answers. The data in Medow were collected through four questionnaires: a main household questionnaire, an adult female questionnaire, a randomly

selected

individual

questionnaire

for

household

members and a community questionnaire. The household questionnaire contains rich information on the household economy,

household

infrastructure,

housing

conditions,

agricultural activities and environmental threats as well as information about individuals in the household: their gender, age, hukou-registration, marital status, education, employment status, etc. Paper 1 is based on data about individual information and the household economy from the household questionnaire. CHNS Article 2 and 3 uses survey data from The China Health and Nutrition Survey (CHNS). This is a longitudinal data set collected in 1989, 1993, 1997, 2000, 2004, 2006, 2009 and 2011. It is an

68

Chapter Three

international collaboration project between the Chinese Centre for Disease Control and Prevention and the University of North Carolina Population Centre. CHNS contains data from 72 counties in nine provinces. Thematically it covers a wide range of topics, such as income, employment, education, welfare benefits, insurance arrangements, health, nutrition and demographic factors. The CHNS data is based on a multistage, random cluster sampling procedure. Although the survey is not nationally representative, the provinces selected cover a wide and highly diversified range of regions, in terms of geography, socioeconomic conditions, and in relation to very diverse health indicators. It gives rich information on income, employment, education, insurance arrangements, health and nutrition and demographic factors. The study population is drawn from Guangxi, Guizhou, Heilongjiang, Henan, Hubei, Hunan, Jiangsu, Liaoning and Shandong. The sampled provinces host approximately 45 per cent of China’s total population (see Figure 4). Follow-up levels are relatively high. The survey followed families that moved within the primary sampling units and in some larger urban entities, but families that migrated from one community to a new one have been dropped. Since 1997, new households in the original communities and in new communities in the original provinces were added to replace the households/communities that no longer participated in the survey.9 The CHNS longitudinal data contains several variable files organised by different topics. The data used for analysis in papers 2 and 3 is from merged files containing information about education, healthcare, income, medical insurance, physical exams, and basic individual information such as gender, age, ethnicity, Please visit CHNS’ website: http://www.cpc.unc.edu/projects/china for detailed information. 9

Methods and Data

69

marital status, labour-market and employment-related variables. The variables are mainly from the adult survey questionnaires, household survey questionnaires and physical examination questionnaires. Figure 4 shows the surveyed provinces in Medow and CHNS. The green shaded regions are the provinces covered by Medow, the blue shaded regions are provinces in CHNS, and the turquoise shaded provinces were included in both surveys.

Figure 4. Map of survey regions

Analytical framework The articles in the thesis use different analytical strategies. The analytical procedures were chosen in order to give good answers to the questions asked in the studies. Statistical tools were also chosen in accordance with the analytical frameworks that have been important within sociology, such as the macro-micro linkage

70

Chapter Three

adapted from the Coleman diagram (Coleman, 1986, 1994). The first paper examines the relation between privatisation and income inequality. In addition to the macro level linkage between privatisation and income inequality, the paper explores how structural influences via different social institutions, such as education and occupation, affect individual income, and how this varies between prefectures with different degrees of privatisation. It also explores how individual income, at the micro-level, is aggregated up to form income inequality. Concerning this question, it is not enough to only focus on the theoretical discussion and interpretation, but is also necessary to develop an analytical framework that can be developed into a hierarchical linear model, to be solved by statistical methods. The following section discusses this framework of macro-micro linkage. The Macro-micro linkage Social and economic processes, viewed from a macro-level perspective, can have a great impact on individuals’ income opportunities and will therefore, as a result, also impact income inequality between different social groups. An example of such a process is private investment in a geographical area. This may generate more competition in the labour market and in turn increase labour demand and thus wages. In particular, the rate of return to education and the relative rewards of different occupations may change, as wage-determination is freer in private enterprises than in state-owned enterprises. In Figure 5, four links between macro and micro actions are identified: (a) The macro-macro link, (b) the macro-micro link, (c) the micro-micro link, and (d) the micro-macro link. Inspired by Coleman’s macro-micro model of social explanation, Hedström and Swedberg (1998: 21-22) combined quantitative analytical sociology with mechanism-thinking by asking three questions: 1) how do macro-level systems affect individuals? (link b), 2) how do

Methods and Data

71

the individuals at the micro-level act and interact under macrolevel conditions? (link c), and 3) how do individuals generate macro-level outcomes? (link d). In the present context of China, the macro causal factor is the degree of privatisation, which provides different labour market opportunities for individuals in different geographical areas. However, analysis only on the macro level may create an ecological

fallacy.

Therefore,

the

micro-level

of individual

socioeconomic positions – specified as education level and household registration status – should also be included in the analysis. These factors may substantially affect individuals’ income levels; a higher educational level, higher occupation class and urban hukou status may correlate with a higher income level (cf. Gustafsson et al. 2008). The macro-macro link a demonstrates the connection between the two macro-characteristics: the degree of privatisation in prefectures and the level of economic inequality. 10 The three links b-d in the diagram represent methodological individualism. The first link between macro systems is not necessarily causal, because social systems and institutions and other macro-level conditions do not have the ability to act on their own. But the system and institutional conditions have an impact on individuals, and individuals in turn influence and change society through their actions and interactions (Veselỳ and Smith, 2008). Combining actor and structure with statistical analytical frameworks brings theory back into the analysis and takes individuals and their actions into account (Lindenberg et al., 1986; Wippler, 1978). There is thus a shift in the focus of statistical analysis from aggregated social phenomena to more individual-

Prefectures here refers to political and administrative provincial subdivisions. There are different types of prefectures, including prefecture-level city, leagues, autonomous prefectures and development zones. The official term “district” was more commonly used before 1983. 10

72

Chapter Three

oriented understandings, assisted by multiple levels of analysis of reality (Manzo, 2007, p. 43). Privatisation at the prefectural level may influence the income of individuals as a result of changes in the reward system, and by providing new opportunities in the labour market. Therefore, the economic outcome of an individual’s education level and occupational class may vary between prefectures with different degrees of privatisation.

Figure 5. Adapted Coleman-diagram

Since social explanations typically located on lower levels than observable phenomena, it is advantageous to use multilevel analysis when combining macro social systems and micro agent positions (Fararo and Butts 1999; Manzo 2007). In order to fulfil the macro-micro-macro link, Paper 1 used a multi-stage design. The first step was to look at the macro-macro link a, using a basic OLS regression model. The model can be read as ln 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 = 𝛼𝛼 + 𝜃𝜃𝜃𝜃 + 𝜀𝜀

(1)

Methods and Data

73

where P is the degree of privatisation, and ε is a disturbance term. Further, the multi-level analysis, denoted by link b and c, was performed. Link c can be written as ln⁡(𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖)𝑖𝑖𝑗𝑗 = 𝛽𝛽𝑜𝑜𝑜𝑜 + 𝛽𝛽1𝑗𝑗 𝑥𝑥∗𝑗𝑗 + 𝑅𝑅𝑖𝑖𝑖𝑖

(2)

where ln⁡(𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖)𝑖𝑖𝑖𝑖 denotes the natural logarithm of income for

individual i in prefecture j. Individual socioeconomic position (education, occupation and household registration) was expressed by⁡𝑥𝑥∗𝑗𝑗 , the prefecture-specific intercept is expressed by 𝛽𝛽𝑜𝑜𝑜𝑜 , and 𝛽𝛽1𝑗𝑗

is the prefecture-specific slope associated with ⁡𝑥𝑥∗𝑗𝑗 . 𝑅𝑅𝑖𝑖𝑖𝑖 ⁡ is the

residual of the dependent variable𝑌𝑌𝑖𝑖𝑖𝑖 , 𝑅𝑅𝑖𝑖𝑖𝑖 ⁡~⁡𝑁𝑁(0, 𝜎𝜎 2 ).

In Equation (2), 𝛽𝛽0𝑗𝑗 and 𝛽𝛽1𝑗𝑗 were further split into a fixed part

and a random part:

𝛽𝛽0𝑗𝑗 = 𝛾𝛾00 + 𝛾𝛾01 𝑊𝑊𝑗𝑗 + 𝑈𝑈0𝑗𝑗 𝛽𝛽1𝑗𝑗 = 𝛾𝛾10 + 𝛾𝛾11 𝑊𝑊𝑗𝑗 + 𝑈𝑈1𝑗𝑗

(3) (4)

where 𝑊𝑊𝑗𝑗 is the level-2 variable, 𝛾𝛾00 and 𝛾𝛾01 respectively denote

the population mean of the intercepts and the population mean of

the coefficients. 𝛾𝛾10 is the different average log income for people

with different ⁡𝑥𝑥∗𝑗𝑗 -value in prefectures with average privatisation, 𝛾𝛾11 is the difference in the impact of a one unit increase in

privatisation between people with different values on⁡𝑥𝑥∗𝑗𝑗 . 𝑈𝑈0𝑗𝑗 and

𝑈𝑈1𝑗𝑗 are the random variables with the standard properties,

representing, respectively, the group-specific part of the intercept, and the coefficient.

A main effect for privatisation 𝑊𝑊𝑗𝑗 and the cross-level

interaction 𝑊𝑊𝑗𝑗 𝑥𝑥∗𝑗𝑗 were included in the fixed part in Equation (5). (𝑈𝑈0𝑗𝑗 + 𝑈𝑈1𝑗𝑗 𝑥𝑥𝑖𝑖𝑖𝑖 ) determines the prefectures component of variance, and 𝑅𝑅𝑖𝑖𝑖𝑖 to the individual:

74

Chapter Three

𝑌𝑌𝑖𝑖𝑖𝑖 = (𝛾𝛾00 + 𝛾𝛾01 𝑊𝑊𝑗𝑗 + 𝛾𝛾10 𝑥𝑥∗𝑗𝑗 + 𝛾𝛾11 𝑊𝑊𝑗𝑗 𝑥𝑥∗𝑗𝑗 ) + (𝑈𝑈0𝑗𝑗 + 𝑈𝑈1𝑗𝑗 𝑥𝑥𝑖𝑖𝑖𝑖 + 𝑅𝑅𝑖𝑖𝑖𝑖 )

(5)

Finally, link d can be demonstrated by a simulation based on coefficients from the second step. The fitted log income is predicted by estimating the best linear predicted log income based on the estimated coefficients for 𝑥𝑥∗𝑗𝑗 :

̂ ̂ ̂ ̂ ln⁡(𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖) 𝑖𝑖𝑖𝑖 = 𝛽𝛽𝑜𝑜𝑜𝑜 + 𝛽𝛽1𝑗𝑗 𝑥𝑥∗𝑗𝑗 + 𝑅𝑅𝑖𝑖𝑖𝑖

(6)

and the Gini coefficient is further generated by converting ̂ ln⁡(𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖)𝑖𝑖𝑖𝑖 into 𝑌𝑌 𝑖𝑖𝑖𝑖 , and using the equation 𝐺𝐺̂ = 1 +

𝑛𝑛

1 2 −( ) ∙ [∑(𝑛𝑛 + 1 − 𝑖𝑖) ∙ 𝑦𝑦̂𝑖𝑖 ] 𝑛𝑛 𝑙𝑙 ∙ 𝑛𝑛2

(7)

𝑖𝑖=1

where the population in a prefecture is considered to be i = 1, 2, …, n with predicted income 𝑦𝑦̂𝑖𝑖 and the arithmetic mean income l.

By using such a stage-wise analysis, the analysis followed the

framework of the Coleman model to examine the macro-micro linkage.

The

study

included

both

structural

factors

and

individuals in the analysis. Furthermore, a multilevel analysis takes advantage of data that consists of clusters of observations made on different occasions for the same subject. Individuals are nested in prefectures. In a random slope design, we allowed the slopes of the prefecture to vary, meaning that slopes are different across prefectures. For people living in prefectures with different levels of privatisation, their educational attainment, occupational position, and household registration status may have a different impact on their income. This was also the result when we included cross-level interaction terms between the prefecturallevel characteristics of privatisation and individual level control variables of education, occupation and household registration. The

Methods and Data

75

use of hierarchical linear models combined with a stepwise design made the relation between macro and micro linkages clearer. Regional heterogeneity and confounders Paper 2 and 3 deals with the problem of regional heterogeneity. In both papers, the analytical units are on the county level. 11 As discussed in Chapter One, there is huge regional and local diversity both between and within different regions in China.

Figure 6. Gini coefficients by counties and waves

Patterns of income inequality in China over the last decades show considerable variation from region to region, and it is reasonable to assume that certain county-level characteristics influence income inequality. Even though the Gini coefficient has increased at a national level, we find both increasing and decreasing Ginicoefficients on the county level in different time periods. Figure 6 The term ‘county’ is used here interchangeably with ‘county-level units’. It includes both counties in rural areas and county-level cities in urban areas 11

76

Chapter Three

shows how the Gini coefficients have changed in the 54 counties in the CHNS. 12 We can see that the degree of income inequality changes over time, and that the differences between counties are relatively large. Some of them show an increasing trend, some are rather stable, and in some counties, the Gini coefficient has dropped. One concern here is that there are county-level factors that may influence both income inequality and health and wellbeing. We can observe some of them, but many are unobservable. In order to control for county-level heterogeneities, I used the fixedeffects model to control for all observed and unobserved stable characteristics for counties. The model is based on repeated observations from a county over time, and it estimates how the average difference between the dependent variables across years is related to the average difference between the independent variables. In the fixed effects model, we do not need restrictions on the mean and variance of the combined effect on y of all unobserved time-constant variables. A general fixed-effects equation (as used in the third paper) can be written as 𝑦𝑦𝑖𝑖𝑖𝑖𝑖𝑖 = ⁡ 𝛽𝛽0 + 𝛽𝛽1 𝑥𝑥1𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽2 𝑥𝑥2𝑐𝑐𝑐𝑐 + 𝛽𝛽3 𝑥𝑥3𝑐𝑐𝑐𝑐 + 𝛽𝛽4 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 + 𝜀𝜀𝑐𝑐 + 𝜂𝜂𝑖𝑖𝑖𝑖𝑖𝑖

(8),

where 𝑦𝑦𝑖𝑖𝑖𝑖𝑖𝑖 is the outcome variable, 𝑥𝑥1𝑖𝑖𝑖𝑖𝑖𝑖 is the individual income, 𝑥𝑥2𝑐𝑐𝑐𝑐 is the Gini coefficient in a county, 𝑥𝑥3𝑐𝑐𝑐𝑐 is the average income

in a county, and 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 is the vector of individual control variables.

𝛽𝛽0 - 𝛽𝛽4 are coefficients to be estimated, 𝜀𝜀𝑐𝑐 (i = 1 … n) is the

unknown intercept for each entity, and 𝜂𝜂𝑖𝑖𝑖𝑖𝑖𝑖 is the disturbance

term. This is also the model used in paper 3, where 𝑦𝑦𝑖𝑖𝑖𝑖𝑖𝑖 is the dependent variable of subjective wellbeing. The two error terms There are 72 counties included in the CHNS, but only 54 of them contain information from at least two waves. 12

Methods and Data

77

are different. There is a different error term 𝑢𝑢𝑖𝑖𝑖𝑖𝑖𝑖 for each

individual in each county at each point in time, and the term represents random variation at each point in time. On the other hand, 𝛼𝛼𝑐𝑐 does not have a time dimension. It only varies across

counties, and represents the combined effect on the dependent variable 𝑦𝑦𝑖𝑖𝑖𝑖𝑖𝑖 of all time-constant variables.

A linear probability model with fixed-effects was adopted in the

second paper. In this case, the outcome variable is dichotomous. The binary outcome variable here for each observation takes values of either 0 or 1. The linear probability model predicts the probability of an event occurring, and the parameters are interpreted as a probability that varies from 0-1. This model (as used in the second paper) included both countylevel fixed effects and year dummies in order to control for county and year heterogeneity. The model is specified as P(𝑦𝑦𝑖𝑖𝑖𝑖𝑖𝑖 = 1|𝑋𝑋 = 𝑥𝑥) = 𝛽𝛽0 + 𝛽𝛽1 𝑥𝑥1𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽2 𝑥𝑥2𝑐𝑐𝑐𝑐 + ⁡𝛾𝛾1 𝑥𝑥3𝑐𝑐𝑐𝑐 + ⁡𝛾𝛾2 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 + (9), ⁡𝛾𝛾3 𝑇𝑇𝑡𝑡 + 𝜀𝜀𝑐𝑐 + 𝑣𝑣𝑡𝑡 + 𝜂𝜂𝑖𝑖𝑖𝑖𝑖𝑖

where x and X are as specified in equation (8), T is a vector of

year-dummies, 𝛽𝛽 stands for the coefficients of interest, 𝛾𝛾 stands

for the coefficients of the control variables, 𝜀𝜀𝑐𝑐 is the unknown intercept for each county unit, 𝑣𝑣𝑡𝑡 is the error for year dummies, and 𝜂𝜂 the error disturbance term.

The linear probability model is simply based on a linear

regression model with a binary dummy dependent variable. It is simple to compute, makes it easier to interpret the marginal effects (the estimation of the marginal effects are usually very

similar to the logit and probit models), and it can, for example, avoid the problem of perfect correlation in the probit model. At the same time, the linear probability model has several limitations. First, the model does not estimate appropriate structural parameters if it is non-linear. Second, it does not give

78

Chapter Three

consistent estimates of the marginal effects. Third, it does not deal with measurement errors in the dependent variable (See e.g. Horrace and Oaxaca 2006). However, the potential bias of the linear probability model will be reduced if the relative proportion of the predicted probabilities that fall outside the unit interval is smaller (Horrace and Oaxaca, 2006). In other words, the greater the proportion of predicted probability that falls between 0 and 1, the less bias and inconsistency there will be in the model. In my analysis, almost all predicted probabilities of health outcomes from paper 2 lie within this interval (0, 1), and less than 2% lie outside. Therefore, the main estimate is to a large degree unbiased and consistent. In addition, the predicted probabilities are almost identical to the linear probability model, the probit model and the logit model. The probit or logit model is not necessarily a better replacement for the linear probability model (Hellevik, 2007). As noted by Wooldridge (2002): ‘If the main purpose is to estimate the partial effect of 𝑥𝑥𝑗𝑗 on the response probability, averaged across

the distribution of 𝑥𝑥, then the fact that some predicted values are

outside the unit interval may not be very important. The linear probability model need not provide very good estimates of partial

effects at extreme values of 𝑥𝑥’. Furthermore, Wooldridge (2002) has stressed that if the model contains mutually exclusive dummy

variables (and paper 2 does contain many variables of this type),

then ‘we need not worry about fitted probabilities less than zero or greater than one.’ Two-way clustering In the second paper, when examining the relation between income inequality and probabilities of having health problems, I first applied robust standard errors clustered at individual level. Health outcomes turned out to be significant. This led to the conclusion that when income inequality increases, the probability

Methods and Data

79

of having a normal value in health is reduced, reflected in women’s

waist

to

hip

ratio,

men’s

blood

pressure

and

overweightness, and mid-upper arm muscle circumference for both genders. The analyses based on clustering at only the individual level led to heteroscedastic standard errors, the OLS estimators were no longer BLUE, the true variance and covariance were underestimated, and the confidence intervals and hypotheses tests were not reliable. In the study, the highest aggregated level in the analysis is at county-level. Cluster-robust standard errors should be computed at the most aggregated level of clustering (Moulton, 1990). When clustered only at individual level, model errors for individuals in the same region may be correlated. This leads to a violation of the i.i.d. error assumption. The problem can be corrected by applying two-way clusterrobust standard errors at both individual and county-level units. After two-way clustering, the standard errors increased, and the health outcomes became non-significant. Two-way clustered standard errors are robust to heteroscedasticity and serial correlation in the error term. Serial correlation means that observations

can

be

correlated

over

time

or/and

within

counties/cities in panel data. In the CHNS data set used in my analyses, the number of waves is not large enough for a two-way covariance model, but the number of county-level units is sufficient for this. The advantage of two-way clustering is that it allows for correlated error across the variance-covariance matrix 𝑁𝑁 ′ ̂ = ∑𝑁𝑁 ̂ 𝑖𝑖⁡ 𝑢𝑢̂𝑖𝑖𝑖𝑖 ], where I(i,j)=1 if observations are in Ω 𝑖𝑖=1 ∑𝑗𝑗=1 I(𝑖𝑖, 𝑗𝑗) [𝒙𝒙𝒊𝒊 𝒙𝒙𝑗𝑗 𝑢𝑢 the same cluster, 0 otherwise (Cameron et al., 2006).

Causality Social scientists are usually aware of situation-specific social contexts. Causality is very closely related to circumstances, and

80

Chapter Three

the causal effects may differ from situation to situation. In order to simplify causal inference, one possible solution is to implement limitations

and

restrictions.

Statistician

Donald

Rubin’s

counterfactual thinking has probably been one of the most influential studies in this field (Holland, 1986; Morgan and Winship, 2007; Sobel, 2006). He stresses that valid causal relations can be demonstrated in an experiment where the units of analysis are randomly assigned into treatment and control groups, while non-causal relations cannot be demonstrated. Therefore, we can control for causal and non-causal effects in experiments (Holland and Rubin, 1983; Rubin, 1974, 1977). When we want to evaluate the causal effect of an intervention, we can observe the outcome of the intervention on the treatment group, and the outcome of being untreated for the control group. However, if we compare between these two outcomes, we are making the mistake of comparing apples to pears. The treatment and control groups may be systematically different from each other in terms of individual attributes, choice, social position, selection bias in the design, etc. The ‘true’ causal effect should be the difference between the actual and potential outcomes of the same group. In other words, we should either compare the outcome of the treatment group as a result of being treated with the outcome of it not being treated, or the difference for the control group being unaffected compared to potentially being treated. The problem is that the potential and actual outcome cannot appear at the same time, and we cannot observe the potential outcomes when the intervention has already been implemented. Therefore, strictly speaking, we do not have a proper comparison. To solve this problem, Rubin introduced the assumption that all the units should be more or less the same. By random assignment in an experimental situation, the treatment and control groups are effectively indistinguishable. Furthermore, if a

Methods and Data

81

treatment is randomly assigned, then it is independent of other factors. The independence allows for an implied counterfactual: even though we cannot observe what would happen if the control group was affected, we can use the unaffected control group outcome as the counterfactual for the treatment group. In this way, it is possible to reach a conclusion on causal effect (Rubin, 1974, 1977, 1978). In applied panel data analysis, fixed effect regression models move

closer

towards

making

causal

inference

claims

by

controlling for potential unmeasured confounding variables that remain stable across repeated measurements, and whose effects are constant over time (e.g. Angrist and Pischke 2009). The logic of fixed-effects estimations is to create an ‘as if’ situation that consists of ‘before-and-after’ experiments and use each unit as its own comparison. There is data on income inequality both before and after it changes. By including county dummies, we are controlling for the average differences across counties in predictors (changes in income inequality and changes in health/wellbeing), and we are able to see whether variations in 𝑥𝑥 (income inequality) are related

to variations in 𝑦𝑦 (health/wellbeing). This is also to estimate the ‘average treatment effect’ for all stable characteristics of counties.

So the fixed-effects models implicitly use each county’s outcome

when not receiving the treatment as each county’s missing counterfactual (c.f. Allison 2009; Angrist and Pischke 2009). The fixed-effects model specification is that there may be some constant characteristics that make income inequality high and also produce bad health, meaning that the time-invariant unobserved county factors are correlated with income inequality. The advantage of this method is that it can control for all timeconstant characteristics of the counties, and reduce potential bias. This simultaneously also deals with another major challenge in causal inference: the ability to control for variables that cannot be

82

Chapter Three

observed (Allison, 2009). In a fixed-effects model, the treatments are received at different points in time, and by identifying and controlling for the variability across counties, one can eliminate much of the error variance (see e.g. Angrist and Pischke 2009 and Wooldridge 2002). Although this method reduces omitted variable bias, more is required to conclude that there is a causal relationship. Several major concerns should be discussed. First, although the fixedeffects model can remove both observed and unobserved timeconstant

characteristics,

it

cannot

rule

out

time-varying

unobserved heterogeneity. In other words, we cannot make sure that the relationship is not driven by a latent variable. In particular, we might think of other events in a county that may be associated both with the degree of income inequality and health/wellbeing, such as local policies. Second, the fixed-effects approach cannot capture unobserved individual differences related to selective migration. One example is that people with certain characteristics choose to stay in or move to a county with high income inequality. In this case, bad health for people observed in a county with high income inequality may be caused by other unobserved factors rather than income inequality itself. Third, there may be inverted causality. By basing the analysis only on the fixed-effects model, we cannot conclude whether there is inequality that influences people’s health or wellbeing – the opposite could be the case. Causality and correlation Former president of the Biometric Society Terry Speed states that ‘considerations of causality should be treated as they have always been in statistics: preferably not at all but, if necessary, then with very great care’ (Speed, 1990). Applied statisticians Cox and Wermuth are also very careful about using the term. They wrote:

Methods and Data

83

‘we shall… for the most part avoid the use of the word causal on the grounds that no statistical analysis of the kind we consider could on its own provide convincing evidence of causality in the strong sense just mentioned’ (Cox and Wermuth, 1996, p. 58). Among other reasons for being careful with the term that they mentioned is the dilemma of actual and potential interventions and the exclusion of unobserved responses as possible causes (Cox and Wermuth, 1996, p. 222). This dissertation does not seek to make causal inference between income inequality and, for example, privatisation, health or subjective wellbeing. In the articles, I was careful to avoid using terms like ‘causality’, ‘effect’, ‘influence’, etc. I chose to rather focus on correlations between factors that may be connected to income inequality. There are two main reasons for doing it this way. First, even though causality was not determined, we can get much information from careful analysis of correlation as well. Karl Pearson, the founder of mathematical statistics and a student of Galton, stated that ‘there was a category broader than causation, namely correlation, of which causation was only the limit, and that this new conception of correlation brought large parts of psychology, anthropology, medicine, and sociology into the field of mathematical treatment’ (Pearson, 1957, p. 159). By understanding correlation, one would ‘grasp the essence of the conception of association between cause and effect’ (Norton, 1978, p. 17). Similarly, sociologist Savage was more in favour of descriptive analysis than causal analysis. He and Burrow argued that sociologists should ‘abandon a sole focus on causality (which we are very bad at) and instead embrace an interest in description and classification’ (Savage and Burrows, 2007, p. 896). Second, descriptive statistics are valuable in themselves. They are sometimes even more accessible for ‘normal people’, since they do not require too much technical knowledge to follow the

84

Chapter Three

numbers. Abbort argues that sociology disappeared from the public mind because while the public wanted descriptions, sociologists are giving them causality: ‘… focusing on causality alone we refuse to publish articles of pure description, even if the description can be quantitatively sophisticated and substantively important’ (Abbott, 2001, p. 121). Efforts to open up ‘black-boxes’ and develop sociological theories are important tasks for sociologists. Can the question of causation contribute to theory development and the clarification of mechanisms? According to Elster (1989), a mechanism opens up a black box, and ‘provides a continuous and contiguous chain of causal or intentional links between the explanans and the explanandum’. Elster distinguishes between correlation and causation. General laws may be superior to correlation, while a causation mechanism consists of several links, in which each link ‘will have to be described by a general law, and in that sense by a ‘black box’ about whose internal gears and wheels we remain ignorant’ (Elster, 1989, p. 7). Further, statistical analysis can be used to test explanations and proper mechanisms, rather than being about the cause itself (Hedström, 2005). A social mechanism is here described as ‘a constellation of entities and activities that are organized in such a way that they regularly bring about a particular type of outcome’ (Hedström, 2005, p. 25). Hedström clarified on what level mechanisms are situated. Regarding the use of statistical tools, he argued that they give possibilities with regards to how individuals and social outcomes are linked in complex social processes. Therefore, it is important to integrate empirical analysis with background material on the social, political and historical context of China. While the background is necessary to give a contextual understanding of the society, the analytical framework is able to provide an appropriate balance between theory and method.

Chapter Four

Summaries of the Articles The three articles in Part II of this dissertation focus on income inequality in China. The first article is about the impact of market transition on income inequality, while the second and third article examine the impact of income inequality on health and wellbeing. An overview of the empirical focus, theoretical context and methods used in each article is presented in Table 2. The first article provides an outline of the market transition process in China. Drawing upon stratifying institutions such as education, occupation and the household registration system, the article asks how the process of market transition in China, measured by the degree of prefectural privatisation, has influenced income inequality. The second and third articles are about the impact of income inequality on quality of life. The second article addresses what impact income inequality has on individual physical health, measured by blood pressure, waist-hip ratio, upper-arm muscle circumference, and overweight. The third article goes beyond the physical measures, and moves towards individuals’ experiences of their wellbeing. It asks whether a higher degree of income inequality is associated with lower subjective wellbeing for Chinese older adults, and examines whether the correlation between inequality and wellbeing varies between different income groups.

86

Chapter Four

Article 1. Income inequality and privatisation: A multilevel analysis comparing prefectural size of private sectors in Western china 2. Income Inequality and Health in China: A Panel Data Analysis

3. Older Adults’ Mental Health in Transitional China: A Sociological Study of the Relationship Between Income Inequality and Subjective Wellbeing

Theoretical perspectives Market transition theory

Methodological approach Multilevel analysis, coefficientbased simulation

Findings

Income inequalityhealth hypothesis

Linear probability models, county-level fixed-effects and year fixedeffects Regression models, county-level fixed-effects

Health indicators are not significantly associated with income inequality.

Literature from happiness research, life satisfaction and wellbeing studies

Table 2. Overview of the articles

The degree of privatisation is negatively correlated with income inequality.

Higher income inequality is associated with lower subjective wellbeing. Higher individual income is positively correlated with subjective wellbeing, but only for richer urban citizens.

Article 1: Income Inequality and Privatisation This article explores whether a higher degree of privatisation is associated with larger income inequality in western provinces in China. The article draws upon market transition theory, and looks at transitional processes in China, in which the previously state-owned enterprises have been transformed into private- and market-oriented sectors. Privatisation is defined by comparing the size of privatisation in different prefectures. Inspired by Coleman’s micro-macro linkage, the study aims to build a bridge between structural–level characteristics, such as privatisation, and individual-level factors, such as individuals’ education, occupation, and the household registration system. The study first

87

explores the macro relationship between privatisation and income inequality at prefectural level. Then it moves to the individual level and explores how individual income is connected to different stratifying

mechanisms

–

education,

occupation

and

the

household registration system. Finally, income inequality is aggregated from individual income, using coefficients estimated from multilevel analysis. The article confirms that income inequality is significantly associated with the degree of privatisation in a prefecture. Further, educational attainment, occupational class and the household registration system have a different impact on individual

income

in

prefectures

with

different

levels

of

privatisation. The multilevel design was appropriate in this study, since it is able to integrate both structural and individual level factors when explaining the relation between privatisation and inequality. The revised manuscript of this paper has been sent to the journal Social Science Research.

Article 2: Income Inequality and Health The second article aims to examine whether income inequality in China is related to individuals’ physical health outcomes. It also tests whether the income inequality-health hypothesis can be confirmed in the context of China by examining whether a higher degree of inequality has a negative influence on population health. The study also attempts to distinguish the effect on health of individual income at micro level from income inequality at macro level. Panel data structure is used to control for county-level heterogeneity. By doing this, time-invariant factors on the countylevel are reduced in the analysis process.

Further, the study

stands out from previous studies because it defines individual health using physical measures, instead of, for example, selfreported health or the mortality rate. The measures used are: blood pressure, waist-hip ratio, upper-arm muscle circumference

88

Chapter Four

and obesity. By using physical measures, the study attempts to give a more objective assessment of health, with an emphasis on individuals. The study found that the results differ from measurement to measurement, and between genders. We found that income inequality does not have a significant impact on individuals’ risks of having health problems. For men, income is correlated with higher probabilities of having abnormal waist-hip ratio and being obese. For women, higher income is correlated with a lower probability of being overweight. The article is published in Social Science & Medicine.

Article 3: Older Adults’ Mental Health Despite increased individual income in the years after the economic reforms in China, the level of life satisfaction declined. Due to lower fertility and mortality rates, the population is aging rapidly. As the proportion of elderly people grows, challenges arise with respect to the wellbeing of older adults in China. As well-being is sometimes used as a measure for mental health, the third paper can be viewed as a further exploration of the relation between income inequality and health. In this paper, I ask whether a lower degree of income inequality contributes to higher subjective wellbeing for older adults in China. Using county-level fixed-effects models, the study finds support for the ‘Eastern paradox’ in China. Based on county-level fixedeffects estimations, the analyses show that, with the exception of the rural poor, income inequality is negatively associated with subjective wellbeing, net of individual income. Individual income is positively correlated with subjective wellbeing, but only for urban citizens with incomes above the median. The paper has been sent to Social Indicators Research, and is in the review process.

Chapter Five

Conclusion: Emerging Challenges ‘That is why our policy will not lead to polarization, to a situation where the rich get richer while the poor get poorer. To be frank, we shall not permit the emergence of a new bourgeoisie.’ -Deng Xiaoping, 1986. 13

This dissertation has looked at income inequality in China. The first article investigates how privatisation may be a contributing factor to rising income inequality, in connection with the stratifying mechanisms of education, occupation and the hukou system. Furthermore, in the second article the study examines how income inequality may have an impact on individual wellbeing, defined as individual physical health. The third examines the relation between income equality and subjective wellbeing. In this dissertation, I have found that a higher degree of privatisation is correlated with a higher level of inequality. Moreover, a higher degree of inequality is not correlated with higher probabilities of having physical health risks, defined by abnormal values on blood pressure, waist-hip-ratio, upper arm muscle circumferences and being overweight. However, inequality

13

See Literature Editorial Board of the CPC Central Committee 1993a.

90

Chapter Five

has a negative impact on individuals’ subjective wellbeing, when we focus on the disadvantaged social group of the elderly. The Chinese government has recognised the problems of rising inequality, and it has gradually put more emphasis on justice, equity and inequality. However, in reality, economic growth is still the main priority both for central and local governments. Many policies leading to more marketisation, and concrete privatisation measures are being implemented. In western China, privatisation and market opening was introduced at a later stage than in the eastern regions. In the development program ‘Strategy for opening up the West’, initiated in the beginning of the 2000s, privatisation was used as a measure to boost the regional economy, but also had the goal of raising living standards for local people and reducing inequality. However, this study indicates

that

there

is

higher

income

inequality

when

privatisation is increased. Further, a prefecture with greater inequality is correlated with lower average individual income level. Both of these points contradict central goals of the policy, and in addition, of the outcomes rising inequality for Chinese people, especially worse health and lower subjective wellbeing go against the aim of improving living standards. This dissertation can bring some new insights within the field of inequality research and represent a contribution to China studies. The effects of privatisation on inequality in Western China have not been examined before, nor have any studies focused on the relation between wellbeing and income inequality among older adults in China. Moreover, previous studies have mainly measured health using indicators such as population mortality or self-reported health. In the second article, health is measured by individual physical health. Furthermore, although topics of inequality, health and wellbeing have been taken up in the international context, there have been fewer studies in the Chinese context.

91

Theory-led analytical frameworks Introducing the famous directed acyclic graphs (DAGs) into sociology, Morgan and Winship (2007) emphasised the importance of

theories

when

choosing

and

adopting

methodological

approaches. They warned of the danger of offering causal inference in the absence of explicit theories. In a context of causality, the understanding of direct and indirect effects, confounding or colliding variables are all to a large degree dependent on our knowledge of reality. Therefore, a contextual understanding and the use of theoretical frameworks are essential for providing a causal structure. In this dissertation, I have attempted to combine empirical analysis with a contextual understanding of China. I have also emphasised the use of theory-led analytical frameworks, such as the Coleman diagram, as well as using a methodological approach that takes regional heterogeneity into account. The background account gives space to flesh out contextual complexities, while the analytical framework is able to provide an appropriate balance between theory and method, hypothesis testing and empirical analysis. When researching inequality and its links to other factors, one danger is looking away from how micro and macro levels are connected, mechanisms,

specifically which

how

consists

inequality-generating of

individual

behaviour

social and

interaction at the micro level, are related to the aggregated structural level (c.f. Weber and Leuridan 2008). The mode of ‘mechanism-based analysis’ for causal inquiry in observational social science involves introducing intervening variables between the causal variable and the outcome variable in an effort to provide a mechanistic explanation of the relation. In the articles included in this thesis, some important intervening variables frequently appeared, such as educational attainment, occupational class and household registration status.

92

Chapter Five

These factors may serve as mediators between individual attributes and structural outcomes, such as individuals’ indicated labour

market

outcomes,

socioeconomic

position,

and

the

structural level outcome of inequality. However, in a complex process of social transition, the relations between these factors are often intertwined with other important variables, and may vary between different groups. Education and occupation In the first paper I found that individuals with higher education levels are better rewarded in a more market-oriented and privatised society. This is due to the need for higher-skilled labour. However, even though market transformation gives higher returns to education, neither education nor income necessarily predicts better wellbeing. Therefore, a higher degree of market economic activity does not necessarily is related to better wellbeing for subjects in all social strata. For example, the relation between education and wellbeing varies when we consider individuals’ income level and their household registration status. For richer rural older adults, higher education is correlated with better subjective wellbeing, but for people with an urban household registration, and for poorer rural people, education is not a major determinant of wellbeing. The impact of individual income on wellbeing also varies between social groups: For richer urban older adults, higher income is correlated with better wellbeing, but income does not predict wellbeing for poorer urban people, or people with rural household registration. Further, the impact of education on health depends on gender and physical measures. For women, higher education means higher probabilities of having normal scores of waist-hip ratio, and lower probabilities of being overweight. But for men, higher education increases the risk of being overweight. It is unclear why

93

there are such differences. It might be due to different lifestyles for more highly educated men and women, or it may be that men and women experience different kinds of pressure in the labour market. The relation between occupation and health also varies between men and women, and between the different health measures.

Compared

to

farmers,

males

within

all

other

occupations have higher probabilities of having abnormal blood pressure and being overweight. The same pattern can be found when it comes to WHR, the only difference being that the semi/non-skilled workers are not significantly different from farmers when considering this health measure. For women, service class and non-manual workers have significantly reduced probabilities of having abnormal blood pressures. Household registration The impact of household registration varies greatly when we consider its relation to individual income, income inequality, wellbeing and health. In paper 1, I found that hukou does have a significant association with individual income, at least in the context of western China. However, the impact of hukou on income does not vary between prefectures with different degrees of privatisation. Moreover, household registration plays an important role for men regarding their health, but the correlation is not statistically significant for women. Compared to rural men, men with an urban household registration have higher probabilities of having abnormal values for blood pressure, WHR and overweightness, and lower probabilities of being undernourished. Furthermore, the relation between income inequality, income and wellbeing varies between urban and rural citizens, and between income groups. The negative impact of income inequality on wellbeing is strongest among richer urban older adults, weaker

94

Chapter Five

for poorer urban older adults and weakest among richer rural people; it disappears for poorer rural older adults. When considering the effect of individual income on wellbeing, increased income is correlated with better subjective wellbeing, but only for richer urban older adults. Epidemiologic transition In researching the relation between income inequality and health in China, three theoretical approaches have been central in the health-inequality

debate.

The

income

inequality–health

hypothesis emphasises the importance of inequality in itself. The absolute income–health hypothesis focuses on individual income. The epidemiologic transition theory looks at changes in healthmortality patterns, and claims that the major causes of mortality/health problems are not necessarily determined by income or income inequality. When analysing several waves of the CHNS data set and defining health using physical measures, I found no statistically significant connection between inequality and health outcomes in China. Therefore, the research does not support the income inequality–health hypothesis. The relation between individual income and wellbeing was not significant for all health indicators. When income increases, it also increases the probability of having abnormal values for men’s waist-hip ratio and overweightness. But for other health measures and for women, individual income does not play an important role. These conclusions may lead to an increased awareness of the need to not only keep an eye on factors like income inequality or income in itself, but also to have a more nuanced discussion that includes the ongoing epidemiologic transition in China. There is a need to place more emphasis on developing China’s healthcare structures, as well as general discussions concerning welfare support. This faces challenges due to China’s demographic

95

and epidemiologic transitions. The population is aging rapidly. The increasingly elderly population, one-child families, dualemployed parenthood and large-scale rural-to-urban migration led to a more fragile family base. At the same time, China has experienced an epidemiologic transition in a much shorter time than many other countries – there has been a shift in leading mortality causes and behavioural changes such as smoking and lack of physical activity. With 177 million adults already diagnosed with hypertension, 303 million smoking adults and the emergence of non-communicable diseases and chronic diseases, China’s healthcare reforms and healthcare institutions face particular challenges (Meng and Tang, 2013; Yang et al., 2008). The challenges include the integration of primary care services with secondary and tertiary healthcare, prevention strategies and treatment for chronic diseases, and public policy support. Further, such policies are integrated with welfare policies for employment, income maintenance, housing, education, etc. Reducing non-communicable disease is one of the key interventions for a reduction of poverty and social and health inequalities (WHO, 2002). Therefore, it is important that welfare institutions can also adapt to new challenges caused by the transition.

Healthcare and welfare support Welfare provision, which is an important factor for income distribution, is closely related to income inequality. However, the three articles did not have a major focus on institutional welfare factors. The healthcare system in China is directly linked to people’s physical and mental health, and the marketisation of healthcare provision has become a major contributor to increased income inequality. Market-oriented competition creates effects such as a greater share of heavy patients in public hospitals, because private health providers only treat patients with less

96

Chapter Five

severe diseases (Cooper, 2010). Additionally, more profit-oriented health services are often associated with higher costs and a lower quality of medical services (Eggleston et al., 2006; Silverman et al., 1999, 1999; Tiemann et al., 2012). Furthermore, welfare support is also a type of social investment. Welfare policies may remove barriers to work (such as illness) and increase economic participation, which in turn contribute positively to development (Midgley, 1999). As a type of welfare support, public healthcare can be seen as a social resource for individuals, supporting socioeconomic, psychological and physical wellbeing (Fritzell and Lundberg, 2007; Lundberg, 2009). Therefore, collective welfare arrangements are expected to be more important for the health of persons in lower social positions and with less social resources (Lundberg, 2009). This further affects the structure of inequality in society. Income inequality is closely related to the labour market structure. Generous benefits are associated with employment and economic growth (Esser, 2005; Garfinkel et al., 2010; van der Wel, 2011). Social policy in health care is expected to be important in terms of securing distribution of resources to people with bad health, as well as reducing the negative social consequences of bad health such as employment (Lundberg, 2009; Lundberg et al., 2008). Therefore, the government provision of welfare resources is particularly important for labour market participants among disadvantaged social groups (van der Wel et al., 2011). However, since the introduction of the economic reform in China, there has been a significant reduction in healthcare use (Liu et al., 2007; Mao, 2007; Wagstaff and Yu, 2007). High medical costs and low coverage of insurance coverage stands out as the most important factors (Hong et al., 2006; Liu et al., 2007; Wagstaff and Lindelow, 2008; Wang and Luo, 2005). Welfare provision can to some extent cancel out the market influence and outcomes of unequal distribution. When this is cut

97

or privatised, the income gap is just larger, and this has a direct impact on peoples’ subjective wellbeing. At the same time, insurance

coverage

often

heavily

depends

on

individuals’

household registration status, employment situation, work sector, migration background and class origin (cf. Whyte 2010). Regional factors such as urban-rural diversity and regional disparities also have important influence on insurance coverage and care receiving (Liu et al., 2007; Vedom and Cao, 2011). Although this dissertation found that household registration has a varying impact (and may not have a direct effect) on people’s wellbeing and physical health, it may still have a great impact on people’s entitlement to healthcare, depending on local differences such as welfare policies. The right to and utilisation of healthcare systems has an important influence on people’s physical and mental health. Further research may focus more on concrete indicators for healthcare policies and services to find more precise and detailed correlations.

Privatising public hospitals Article 1 explored processes of privatisation in western China. The effect of privatisation on income inequality may occur via different stratifying mechanisms, and the underlying relation between privatisation and income inequality is transferable from the case of western China to other regions, and from defining privatisation as the proportion people working in private sectors, to the arrangement and praxis which transfer public properties to private ownership. The ideal of privatisation is based on efficiency, profit, individualism, market regulation and the withdrawal of state provision. On one hand, a market with a higher degree of privatisation rewards higher educated people and skilled-workers with high wages, on the other hand, the lower educated, unskilled and

rural

household

registration

holders

become

more

98

Chapter Five

disadvantaged. When state provision, in the form of welfare, is privatised, individuals are left on their own in the stratified labour market. Income inequality rises, social gaps become larger, and it becomes more difficult to achieve ‘common prosperity’. Public hospitals in China are the most important healthcare providers, and deliver more than 90% of the country’s health services (Yip et al. 2012). One problem with Chinese healthcare is the lack of primary healthcare services, and the expansion and overuse of secondary and tertiary services. It has led to increased total costs and greater inequality. In 2010, the Ministry of Health started a pilot project to reform the public hospitals. The pilot project was started in 17 cities. The central goals of the reform are to increase equity, affordability and accessibility (MoH 2010). No practical effort was made to reintroduce

primary

healthcare.

Instead,

the

important

components were to allow for private investment in healthcare, as well as to encourage and support the private ownership and operation of health services. As recently as 2012, the 12th Five-Year-Plan for Health was announced, emphasising non-public health service providers and an extension of the pilot project to reform public hospitals in other counties and cities (MoF 2014; State Council 2012, 2014a; Wang 2014). At least one city in each province was chosen to participate in the public hospital reform during 2014, and over half of the counties will be included in reforming their county-level public hospitals, according to the plans (State Council 2014b). The main goals have been the same across the cities, but local goals and practices also vary. It seems that most of the governments chose to adapt the market mechanisms and private actors as healthcare solutions. Only four of these cities mentioned goals of providing access to affordable basic healthcare through public and non-profit hospitals. Some other cities, to varying degrees, shifted their strategies towards market competition and

99

private ownership of public hospitals (Yip et al. 2012). In Hubei Province, the number of new approved private health care institutes increased from 32 in 2012 to 130 in 2013. In Beijing, in just the first two quarters in 2013, 163 new private hospitals and clinics were approved (Chinanews 2014). In 2011 and 2012, the China Resources Pharmaceutical Group bought 5 public hospitals and planned to extend to 30 hospitals in five years (Caixin 2014). If the public hospital reform in China reaches its goal of equal and affordable access to health services, it would lead to efficient and quick treatment of patients, independent of their background. This would contribute to higher probabilities of patients being healthy sooner and returning to work. The trend is reflected in higher labour market participation in regions where public hospital reforms took place. Moreover, combined with higher work participation and lower medical costs, it gives reason to believe that a successful healthcare system increases economic equality in general. However, to date, most entrants in China have been motivated by profit, including the private hospital chains, pharmaceutical sector, medical equipment producers, and real estate developers (Yip and Hsiao 2014). The opening of the market for private health providers may also reinforce the use of high-tech medical tests and expensive medicines in order to signify high quality of services to patients (Yip and Hsiao 2014). This may indicate an even higher barrier for the poor and for people with poor health, and these developments may be followed by greater income inequality.

The active social volcano? Income inequality has increased in China. But what do the Chinese people think about inequality? In a survey on people’s perceptions of fairness and inequality, Whyte and colleagues found that although income inequality has

100 Chapter Five

increased dramatically in China, Chinese people are more concerned about state provision of welfare distribution and government efforts to reduce inequality; they are less concerned about equity and social mobility (Whyte, 2010b; Whyte and Im, 2014). Compared to the 2004 survey, in 2009 Whyte and Im (2014) found that, the Chinese people had increased acceptance of current inequalities, opportunities and social justice and were more optimistic. However, they were more critical of inequality caused by government redistribution. This relates to previous discussions in this chapter about welfare provision and the role of the state. Privatising welfare services and the state sector has consequences for distribution and income inequality. The market transition theory argues that the rising position of entrepreneurship and increased influence of economic capital may reduce inequality in the beginning, since the ‘playing field’ is levelled out due to reduced returns to political capital and prestige. Along with market expansion, inequality may increase in a later phase of market transition, when economic capital enhances the redistributional inequality. Chinese people are critical of unequal state redistribution and transmission of wealth, and expressed a strong desire for state welfare responsibility, a government levelling of social services, job security, reduce structurally caused income inequality and for the government to play a role of distributing wealth and income among people (Whyte and Im, 2014). Although living standards have improved, and income levels have increased, Chinese people’s subjective wellbeing has declined. I found in the third article that higher income inequality leads to lower wellbeing for older adults. Older adults, who have experienced great changes in their lifetimes, have also lived through a time of massive societal changes, in which China changed from being a relatively egalitarian society before the 1980s, to a market economy with high income inequality. Unlike

101

younger generations, which have grown up viewing inequality as legitimate, older adults may have an attitude to wellbeing influenced by the contrast between an egalitarian past and unequal present. While economic development may have led to higher incomes and a better standard of living, increased income inequality may be an indicator for what older adults’ might miss compared to the past: the previously relatively egalitarian society, an ideology that they might find meaningful, and the previously generous welfare support. Another interpretation is to look at income inequality in terms of relative position. Worsened subjective wellbeing may be caused by individuals comparing themselves to people who are in more advantaged situations. It is stressful living among wealthier people. Stress may also translate into mental health problems for people living in more unequal places. Moreover,

regional

characteristics

matter.

Stratifying

mechanisms such as education, occupation and settlement patterns differ with regional characteristics. In places with a higher level of privatisation, there is a higher return to education, occupation and household registrations status. One factor is that wealthier people may have greater resources to invest in education, because they have higher occupational positions, and are able to pay for social and medical services. In regions where wealthy residents can buy their way out of welfare services, less social cohesion and investment in public health and education may affect physical and mental health outcomes.

Future studies This dissertation has attempted to contribute to the existing literature of income inequality in China. By looking at the market transformation process with a focus on privatisation, the thesis has

emphasised

the

important

intermediate

stratifying

102 Chapter Five

mechanisms

of

education,

occupation

and

the

household

registration system in distributing income. The papers are based on statistical analysis, but also aimed at capturing contextual complexities and regional and local differences. The relation of income inequality to health and wellbeing has been the main topic. By doing so, the study has not only had a structural level focus, but has also taken individuals into account, both in terms of the objective physical measures and in terms of subjective evaluations and perception. Based on the work of this thesis, several interesting areas for future studies can be identified. First, more research is needed on the welfare state system, how it is affected by income inequality and how it may influence income inequality. Such studies would provide a general framework for the national and local context, as well as highlighting the institutional mechanisms of wealth and income distribution. Second, we need to move a step further to identify potential causal relationships between income inequality and other factors. Moreover, other dimensions of inequality, and not just income inequality, can be studied more systematically. Third, more nuanced and detailed analyses are needed to study different social groups in China. We need a more comprehensive understanding of the society, by noting that groups of people may play different roles, or be affected in different ways by processes of market transition. One example in this thesis is my focus on urban and rural residents with urban and rural hukou. The huge numbers of migrants have not received much attention. These people are certainly highly diverse – from the rich upper class, which is migrating between large cities, to a new ‘precariat’ of rural-to-urban migrants. They may also be influenced differently by income inequality. In the context of a society facing rapid change and a mobile population, such research is as relevant as ever.

References Abbott, A. (2001), Time Matters: On Theory and Method, 1 edition., University of Chicago Press, Chicago. Alesina, A., Di Tella, R. and MacCulloch, R. (2004), “Inequality and happiness: Are Europeans and Americans different?”, Journal of Public Economics, Vol. 88 No. 9-10, pp. 2009– 2042. Allison, P.D. (2009), Fixed Effects Regression Models, SAGE Publications. Almås, I., Johnsen, Å.A. and Kotsadam, A. (2014), “Poverty in China Seen from Outer Space”, Memorandum, Vol. 2014/11. Angrist, J., D. and Pischke, J.-S. (2009), Mostly Harmless Econometrics: An Empiricist’s Companion, 1 edition., Princeton University Press, Princeton. Appleton, S. and Song, L. (2008), “Life Satisfaction in Urban China: Components and Determinants”, World Development, Vol. 36 No. 11, pp. 2325–2340. Araral, E. (2009), “Privatization and regulation of public services: A framework for institutional analysis”, Policy and Society, Vol. 27 No. 3, pp. 175–180. Baltatescu, S. (2007), Towards a Sociological Theory of Subjective Well-Being, SSRN Scholarly Paper No. ID 2196346, Social Science Research Network, Rochester, NY, available at: http://papers.ssrn.com/abstract=2196346 (accessed 16 September 2015). Bandelj, N. and Mahutga, M.C. (2010), “How Socio-Economic Change Shapes Income Inequality in Post-Socialist Europe”, Social Forces, Vol. 88 No. 5, pp. 2133–2161. Becker, G.S. (2009), Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education, University of Chicago Press.

104

Bian, Y. and Logan, J.R. (1996), “Market Transition and the Persistence of Power: The Changing Stratification System in Urban China”, American Sociological Review, Vol. 61 No. 5, pp. 739–758. Birdsall, N. and Nellis, J. (2003), “Winners and Losers: Assessing the Distributional Impact of Privatization”, World Development, Vol. 31 No. 10, pp. 1617–1633. Blecher, M. (2005), “Inequality and Capitalism in China”, presented at the the American Political Science Association Task Force Conference on Inequality and Difference in the Third World, unpublished report prepared for the American Political Science Association Task Force Conference on Inequality and Difference in Developing Societies, available at: http://www.apsanet.org/content_13916.cfm (accessed 10 October 2015). Bloom, G. and Gu, X. (1997), “Introduction to Health Sector Reform in China”, IDS Bulletin, Vol. 28 No. 1. Blumenthal, D. and Hsiao, W. (2005), “Privatization and Its Discontents — The Evolving Chinese Health Care System”, New England Journal of Medicine, Vol. 353 No. 11, pp. 1165–1170. Bose, M., Oliván, B. and Laferrère, B. (2009), “Stress and obesity: the role of the hypothalamic–pituitary–adrenal axis in metabolic disease”, Current Opinion in Endocrinology, Diabetes, and Obesity, Vol. 16 No. 5, pp. 340–346. Bradburn, N.M. and Caplovitz, D. (1965), Reports on Happiness; a Pilot Study of Behavior Related to Mental Health, Aldine, Chicago. Brockmann, H., Delhey, J., Welzel, C. and Yuan, H. (2008), “The China Puzzle: Falling Happiness in a Rising Economy”, Journal of Happiness Studies, Vol. 10 No. 4, pp. 387–405. Byrne, D.G. and Espnes, G.A. (2008), “Occupational stress and cardiovascular disease”, Stress and Health, Vol. 24 No. 3, pp. 231–238. Cameron, A.C., Gelbach, J.B. and Miller, D.L. (2006), Robust Inference with Multi-Way Clustering, Working Paper No. 327, National Bureau of Economic Research, available at: http://www.nber.org/papers/t0327 (accessed 10 February 2016). Caporale, G.M., Georgellis, Y., Tsitsianis, N. and Yin, Y.P. (2009), “Income and happiness across Europe: Do reference values matter?”, Journal of Economic Psychology, Vol. 30 No. 1, pp. 42–51.

105

Carroll, D., Phillips, A.C., Der, G., Hunt, K. and Benzeval, M. (2011), “Blood pressure reactions to acute mental stress and future blood pressure status: data from the 12-year follow-up of the West of Scotland Study”, Psychosomatic Medicine, Vol. 73 No. 9, pp. 737–742. CCCPC, (Central Committee of the Communist Party of China, State Council). (2000), “CPC’s suggestion for the tenth fiveyear plans of national economic and social development Stratify (中共中央关于制定国民经济和社会发展第十个五年计 划 的 建 议 )”, News of the Communist Party of China, 11 October, available at: http://cpc.people.com.cn/GB/64162/71380/71382/71386/4837 946.html (accessed 21 June 2013). CCCPC Party Literature Research Office (Ed.). (1986), Selected Works of Chen Yun, Vol. 3 (陈云文选第三卷), Vol. 3, Central Party Literature Press ( 中 央 文 献 出 版 社 ） , Beijing, available at: http://www.reformdata.org/content/19860902/2031.html. CGMH. (2011), Chinese Guidelines for the Management of Hypertension (CGMH), 3rd Edition (中国高血压防止指南第 三版 ), MoH Centers for Disease Control and Prevention, China Hypertension Association, National Center for Cardiovascular Diseases (卫生部疾病控制局，高血压联盟， 国家心血管病中心), Beijing. Chan, C.K., Ngok, K.L. and Phillips, D. (2008), Social Policy in China: Development and Well-Being, Policy Press, Bristol, UK. Chang, S., He, W. and Chen, C. (2006), “The growth characteristics of children under 5 in the past 15 years”, Journal of Hygiene Research (卫生研究), Vol. 35 No. 6, pp. 768–771. Chang, S., He, W., Jia, F. and Chen, C. (2007), “Analysis on the changes of nutritional status in China anemia status of children under 5 in China”, Wei Sheng Yan Jiu = Journal of Hygiene Research, Vol. 36 No. 2, pp. 210–212. Chan, J.M., Rimm, E.B., Colditz, G.A., Stampfer, M.J. and Willett, W.C. (1994), “Obesity, fat distribution, and weight gain as risk factors for clinical diabetes in men”, Diabetes Care, Vol. 17 No. 9, pp. 961–969. Chen, G.J. (2010a), “Income Inequality in China: U-form Transformation and Uncertainty (中国收入不平等:U 型变化 与不确定的未来)”, Jiangsu Social Sciences, No. 5.

106

Chen, G.J. (2010b), “Market or Non-Market: Empirical Analysis of Reasons for Income Inequality in China (市场抑或非市场:中 国收入不平等成因实证分析)”, Sociological Studies, No. 6. Chen, J., Zhao, X., Zhang, X., Yin, S., Piao, J., Huo, J., Yu, B., et al. (2005), “Studies on the effectiveness of NaFeEDTAfortified soy sauce in controlling iron deficiency: a population-based intervention trial”, Food and Nutrition Bulletin, Vol. 26 No. 2, pp. 177–186; discussion 187–189. Chen, W.-Q., Wong, T.W. and Yu, I.T.-S. (2008), “Association of Occupational Stress and Social Support with Healthrelated Behaviors among Chinese Offshore Oil Workers”, Journal of Occupational Health, Vol. 50 No. 3, pp. 262–269. Chen, Z. (Ed.). (2012), Healthy China 2020: Strategic Research Report ( 健 康 中 国 2020 战 略 研 究 报 告 ), People’s medical publishing house, Beijing. Chiasson, J.-L., Gomis, R., Hanefeld, M., Josse, R.G., Karasik, A., Laakso, M. and Group, T.S.-N.T.R. (1998), “The STOPNIDDM Trial: An international study on the efficacy of an α-glucosidase inhibitor to prevent type 2 diabetes in a population with impaired glucose tolerance: rationale, design, and preliminary screening data”, Diabetes Care, Vol. 21 No. 10, pp. 1720–1725. Chow, C.K., Lock, K., Teo, K., Subramanian, S.V., McKee, M. and Yusuf, S. (2009), “Environmental and societal influences acting on cardiovascular risk factors and disease at a population level: a review”, International Journal of Epidemiology, Vol. 38 No. 6, pp. 1580–1594. Chow, G.C. (2007), China’s Economic Transformation, Wiley. Clark, A.E. and Senik, C. (2011), Will GDP Growth Increase Happiness in Developing Countries?, Working Paper No. IZA Discussion Paper No. 5595, Institute for the Study of Labor (IZA). Clays, E., De Bacquer, D., Leynen, F., Kornitzer, M., Kittel, F. and De Backer, G. (2007), “Job stress and depression symptoms in middle-aged workers-prospective results from the Belstress study”, Scandinavian Journal of Work, Environment & Health, Vol. 33 No. 4, pp. 252–259. Coan, R.W. (1977), Hero, Artist, Sage, or Saint?, 1St Edition edition., Columbia University Press. Coburn, D. (2004), “Beyond the income inequality hypothesis: class, neo-liberalism, and health inequalities”, Social Science & Medicine, Vol. 58 No. 1, pp. 41–56.

107

Cohen, S., Line, S., Manuck, S.B., Rabin, B.S., Heise, E.R. and Kaplan, J.R. (1997), “Chronic social stress, social status, and susceptibility to upper respiratory infections in nonhuman primates”, Psychosomatic Medicine, Vol. 59 No. 3, pp. 213–221. Coleman, J.S. (1986), “Social Theory, Social Research, and a Theory of Action”, American Journal of Sociology, Vol. 91 No. 6, pp. 1309–1335. Coleman, J.S. (1994), Foundations of Social Theory, Harvard University Press. Collins, S.M., Karasek, R.A. and Costas, K. (2005), “Job strain and autonomic indices of cardiovascular disease risk”, American Journal of Industrial Medicine, Vol. 48 No. 3, pp. 182–193. Collyer, F.M. (2003), “Theorising Privatisation: Policy, Network Analysis, and Class”, Electronic Journal of Sociology, available at: http://www.sociology.org/content/vol7.3/01_collyer.html (accessed 10 October 2014). Cook, I.G. and Dummer, T.J.B. (2004), “Changing health in China: re-evaluating the epidemiological transition model”, Health Policy, Vol. 67 No. 3, pp. 329–343. Cooper, Z. (2010), “Does hospital competition improve efficiency? : an analysis of the recent market-based reforms to the English NHS”. Cosselman, K.E., Navas-Acien, A. and Kaufman, J.D. (2015), “Environmental factors in cardiovascular disease”, Nature Reviews Cardiology, Vol. 12 No. 11, pp. 627–642. Cox, D.R. and Wermuth, N. (1996), Multivariate Dependencies: Models, Analysis and Interpretation, CRC Press. D’Arpizio, C., Levato, F., Zito, D. and de Montgolfier, J. (2014), Luxury Goods Worldwide Market Study Fall-Winter 2014: The Rise of the Borderless Consumer, Bain & Company, Paris. Deaton, A. (2003), “Health, Inequality, and Economic Development”, Journal of Economic Literature, Vol. 41 No. 1, pp. 113–158. Deaton, A. (2007), Income, Aging, Health and Wellbeing Around the World: Evidence from the Gallup World Poll, Working Paper No. 13317, National Bureau of Economic Research, available at: http://www.nber.org/papers/w13317 (accessed 20 September 2015).

108

Deaton, A. (2008), “Income, Health, and Well-Being around the World: Evidence from the Gallup World Poll”, Journal of Economic Perspectives, Vol. 22 No. 2, pp. 53–72. Diener, E. (1984), “Subjective Well-Being”, Psychological Bulletin, Vol. 95 No. 3, pp. 542–575. Diener, E. and Diener, C. (1995), “The wealth of nations revisited: Income and quality of life”, Social Indicators Research, Vol. 36 No. 3, pp. 275–286. Diener, E., Lucas, R.E. and Oishi, S. (2002), “Subjective WellBeing: The Science of Happiness and Life Satisfaction”, in Lopez, S.J. and Snyder, C.R. (Eds.), Handbook of Positive Psychology, Oxford University Press, Oxford and New York, pp. 63–73. Diener, E. and Seligman, M.E.P. (2004), “Beyond money: toward an economy of well-being”, Vol. 5, Psychological Science in the Public Interest. Diener, E., Suh, E.M., Lucas, R.E. and Smith, H.L. (1999), “Subjective Well-Being: Three Decades of Progress”, Psychological Bulletin, Vol. 125 No. 2, pp. 276–302. Di Tella, R. and MacCulloch, R. (2008), Happiness Adaptation to Income beyond “Basic Needs”, Working Paper No. 14539, National Bureau of Economic Research. Di Tella, R., MacCulloch, R.J. and Oswald, A.J. (2003), “The Macroeconomics of Happiness”, Review of Economics and Statistics, Vol. 85 No. 4, pp. 809–827. Djukanovic, V., Mach, E.P., UNICEF and Organization, W.H. (1975), Alternative Approaches to Meeting Basic Health Needs in Developing Countries: A Joint UNICEF/WHO Study, World Health Organization. Dollar, D. (2007), Poverty, Inequality, and Social Disparities during China’s Economic Reform, No. WPS4253, The World Bank, pp. 1–28. DPES, (Department of Population and Employment Statistics of National Bureau of Statistics of China) (Ed.). (2005), China population statistics yearbook, China Statistics Press. DRC (Ed.). (2005), An Evaluation of and Recommendations on the Reforms of the Health System in China: Executive Summary, Project team of DRC. 中国国务院经济技朮社会发 展研究中心课题组, Beijing. Easterlin, R.A. (1974), “Does economic growth improve the human lot? Some empirical evidence”, in David, P.A. and Reder, M.W. (Eds.), Nations and Households in Economic Growth :

109

Essays in Honor of Moses Abramovitz, Academic Press, pp. 89–125. Easterlin, R.A. (1995), “Will raising the incomes of all increase the happiness of all?”, Journal of Economic Behavior & Organization, Vol. 27 No. 1, pp. 35–47. Easterlin, R.A. (2001), “Income and Happiness: Towards a Unified Theory”, The Economic Journal, Vol. 111 No. 473, pp. 465– 484. Easterlin, R.A., Morgan, R., Switek, M. and Wang, F. (2012), “China’s life satisfaction, 1990–2010”, Proceedings of the National Academy of Sciences of the United States of America, Vol. 109 No. 25, pp. 9775–9780. Easterlin, R.A. and Sawangfa, O. (2010), “Chapter 7. Happiness and Economic Growth: Does the Cross Section Predict Time Trends? Evidence from Developing countries”, in Diener, E., Kahneman, D. and Helliwell, J. (Eds.), International Differences in Well-Being, Oxford University Press, pp. 166–216. Eggleston, K., Shen, Y.-C., Lau, J., Schmid, C.H. and Chan, J. (2006), Hospital Ownership and Quality of Care: What Explains the Different Results?, Working Paper No. 12241, National Bureau of Economic Research. Elster, J. (1989), Nuts and Bolts for the Social Sciences, Cambridge University Press. Eriksson, J., Lindström, J., Valle, T., Aunola, S., Hämäläinen, H., Ilanne-Parikka, P., Keinänen-Kiukaanniemi, S., et al. (1999), “Prevention of Type II diabetes in subjects with impaired glucose tolerance: the Diabetes Prevention Study (DPS) in Finland. Study design and 1-year interim report on the feasibility of the lifestyle intervention programme”, Diabetologia, Vol. 42 No. 7, pp. 793–801. Esser, I. (2005), Why Work? : Comparative Studies on Welfare Regimes and Individuals’ Work Orientations, dissertation, Stockholm. Fan, C.C. and Sun, M. (2008), “Regional Inequality in China, 1978-2006”, Eurasian Geography and Economics, Vol. 49 No. 1, pp. 1–18. Fang, H. and Rizzo, J.A. (2011), “Does inequality in China affect health differently in high- versus low-income households?”, Applied Economics, Vol. 44 No. 9, pp. 1081–1090. FAO. (2010), The State of Food Insecurity in the World Addressing Food Insecurity in Protracted Crises, Food and Agriculture Organization of the United Nations, Rome.

110

Fararo, T.J. and Butts, C.T. (1999), “Advances in generative structuralism: Structured agency and multilevel dynamics”, The Journal of Mathematical Sociology, Vol. 24 No. 1, pp. 1–65. Feng, N. and Xiao, N. (2007), Population Aging in China as Reflected by the Results of the 2005 Population Sample Survey, Department of Population and Employment Statistics, National Bureau of Statistics of China. Firebaugh, G. and Schroeder, M.B. (2009), “Does your neighbor’s income affect your happiness?”, American Journal of Sociology, Vol. 115 No. 3, pp. 805–831. Fischer, C.S. (2007), “What wealth-happiness paradox? A short note on the American case”, Journal of Happiness Studies, Vol. 9 No. 2, pp. 219–226. Flannery, R. (2015), “China Now Has a Record 400 Billionaires And Billionaire Families; Greater China 500+”, Forbes, 21 April, available at: http://www.forbes.com/sites/russellflannery/2015/04/20/chin a-now-has-a-record-400-billionaires-and-billionairefamilies-greater-china-500/ (accessed 8 September 2015). Frey, B.S. (2008), Happiness: A Revolution in Economics, MIT Press, Cambridge, Mass. Frey, B.S. and Stutzer, A. (2002), “What Can Economists Learn from Happiness Research?”, Journal of Economic Literature, Vol. 40 No. 2, pp. 402–435. Fritzell, J. and Lundberg, O. (2007), Health Inequalities and Welfare Resources: Continuity and Change in Sweden, Policy Press. Gaetano, A.M. and Jacka, T. (2004), On the Move: Women and Rural-to-Urban Migration in Contemporary China, Columbia University Press. Gallup-Healthways. (2015), State of Global Well-Being: 2014 Country Well-Being Rankings, Gallup-Healthway. Gan, J., Guo, Y. and Xu, C. (2008), “What Makes Privatization Work? The Case of China”, presented at the HKU conference, Hong Kong University of Science and Technology. Gao, Q. (2008), “The Chinese Social Benefit System in Transition”, Annals of the New York Academy of Sciences, Vol. 1136 No. 1, pp. 342–347. Garfinkel, I., Rainwater, L. and Smeeding, T. (2010), Wealth and Welfare States: Is America a Laggard or Leader?, Oxford University Press, Oxford ; New York.

111

Garnaut, R., Song, L., Tenev, S. and Yao, Y. (2005), China’s Ownership Transformation: Process, Outcomes, Prospects, World Bank, Washington, DC. Gaylin, D.S. and Kates, J. (1997), “Refocusing the lens: epidemiologic transition theory, mortality differentials, and the AIDS pandemic”, Social Science & Medicine (1982), Vol. 44 No. 5, pp. 609–621. Gong, L., Xu, Y., Zhang, W., Liu, W. and Wang, X. (2006), “Chinese Guideline for Early Vascular Disease Detection (First report) (中国血管病变早期检测技术应用指南(第一次报 告))”, Medical Journal of Chinese Peoples Health (中国民康 医学), Vol. 9. Goodman, D.S.G. (2004), “The Campain to ‘Open Up the West’”, in Goodman, D.S.G. (Ed.), China’s Campaign to “Open Up the West”: National, Provincial and Local Perspectives, Cambridge University Press. Gravelle, H. (1998), “How much of the relation between population mortality and unequal distribution of income is a statistical artefact?”, BMJ, Vol. 316 No. 7128, pp. 382– 385. Groenewold, N., Chen, A. and Lee, G. (2008), “China’s Regions and Regional Disparities, 1953 to 2005”, Linkages between China’s Regions: Measurement and Policy, Edward Elgar Publishing Limited, UK. Gu, D., Reynolds, K., Duan, X., Xin, X., Chen, J., Wu, X., Mo, J., et al. (2003), “Prevalence of diabetes and impaired fasting glucose in the Chinese adult population: International Collaborative Study of Cardiovascular Disease in Asia (InterASIA)”, Diabetologia, Vol. 46 No. 9, pp. 1190–1198. Gustafsson, B.A., Shi, L. and Sicular, T. (2008), Inequality and Public Policy in China, Cambridge University Press. Gustafsson, B.A., Shi, L., Sicular, T. and Ximing, Y. (2008), “Income Inequality and Spatial Differences in China, 1988, 1995, and 2002”, in Gustafsson, B.A., Shi, L. and Sicular, T. (Eds.), Inequality and Public Policy in China, Cambridge University Press. Haas, J.D. and Brownlie, T. (2001), “Iron deficiency and reduced work capacity: a critical review of the research to determine a causal relationship”, The Journal of Nutrition, Vol. 131 No. 2S-2, pp. 676–690. Hedström, P. (2005), Dissecting the Social: On the Principles of Analytical Sociology, Cambridge University Press.

112

Hellevik, O. (2007), “Linear versus logistic regression when the dependent variable is a dichotomy”, Quality & Quantity, Vol. 43 No. 1, pp. 59–74. Hirschman, A. and Rothschild, M. (1973), “The Changing Tolerance for Income Inequality in the Course of Economic Development”, The Quarterly Journal of Economics, Vol. 87 No. 4, pp. 544–66. Holbig, H. (2004), “The Emergence of the Campaign to Open Up the West: Ideological Formation, Central Decision-making and the Role of the Provinces”, The China Quarterly, Vol. 178, pp. 335–357. Holland, P.W. (1986), “Statistics and Causal Inference”, Journal of the American Statistical Association, Vol. 81 No. 396, pp. 945–960. Holland, P.W. and Rubin, D.B. (1983), “On Lord’s Paradox”, in Wainer, H. and Messick, S. (Eds.), Principals of Modern Psychological Measurement: A Festschrift for Frederic M. Lord, Routledge. Hong, Y., Li, X., Stanton, B., Lin, D., Fang, X., Rong, M. and Wang, J. (2006), “Too Costly To Be Ill: Health Care Access and Health Seeking Behaviors among Rural-to-urban Migrants in China”, World Health & Population, Vol. 8 No. 2, pp. 22–34. Horrace, W.C. and Oaxaca, R.L. (2006), “Results on the bias and inconsistency of ordinary least squares for the linear probability model”, Economics Letters, Vol. 90 No. 3, pp. 321–327. Horton, R. (2015), “Offline: China—the triumph (and danger) of transition”, The Lancet, Vol. 386 No. 10006, p. 1808. Hu, R., Shen, C. and Zou, H. (2011), Health Care System Reform in China: Issues, Challenges and Options, CEMA Working Paper No. 517, China Economics and Management Academy, Central University of Finance and Economics, available at: https://ideas.repec.org/p/cuf/wpaper/517.html (accessed 5 October 2014). Hurun Report. (2013), The Chinese Millionaire Wealth Report 2013, The Group M Knowledge, Hurun Wealth Report, available at: http://up.hurun.net/Humaz/201312/20131218145315550.pdf. Kahneman, D. and Deaton, A. (2010), “High income improves evaluation of life but not emotional well-being”, Proceedings of the National Academy of Sciences, Vol. 107 No. 38, pp. 16489–16493.

113

Kahneman, D. and Krueger, A.B. (2006), “Developments in the Measurement of Subjective Well-Being”, The Journal of Economic Perspectives, Vol. 20 No. 1, pp. 3–24. Knight, J. and Gunatilaka, R. (2010), “Great Expectations? The Subjective Well-being of Rural–Urban Migrants in China”, World Development, Vol. 38 No. 1, pp. 113–124. Knight, J. and Gunatilaka, R. (2011), “Does Economic Growth Raise Happiness in China?”, Oxford Development Studies, Vol. 39 No. 1, pp. 1–24. Knight, J. and Gunatilaka, R. (2014), “Subjective Well-being and Social Evaluation: A Case Study of China”, in Clark, A. and Senik, C. (Eds.), Happiness and Economic Growth: Lessons from Developing Countries, Oxford University Press, United Kindom. Knight, J. and Song, L. (1995), “Towards a Labour Market in China”, Oxford Review of Economic Policy, Vol. 11 No. 4, pp. 97–117. Knight, J. and Song, L. (1999), The Rural-Urban Divide: Economic Disparities and Interactions in China, Oxford University Press. Kondo, N., Sembajwe, G., Kawachi, I., van Dam, R.M., Subramanian, S.V. and Yamagata, Z. (2009), “Income inequality, mortality, and self rated health: meta-analysis of multilevel studies”, BMJ, Vol. 339, available at:http://doi.org/10.1136/bmj.b4471. Layard, R. (2003), Happiness: Has Social Science a Clue?, Lionel Robbins Memorial Lectures 2002/3, London School of Economics, available at: http://www.lse.ac.uk/publicEvents/events/2003/20030106t1 439z001.aspx (accessed 20 September 2015). Le Grand, J. and Robinson. (1984), “Privatization and the Welfare State: in introduction”, in Le Grand, J. and Robinson, R. (Eds.), Privatization and the Welfare State, Allen & Unwin, London, UK, pp. 1–18. Lei, X., Yin, N. and Zhao, Y. (2010), SES Health Gradients During the Epidemiological Transition: The Case of China, SSRN Scholarly Paper No. ID 1599012, Social Science Research Network, Rochester, NY, available at: http://papers.ssrn.com/abstract=1599012 (accessed 15 March 2016). Lemieux, T. (2008), “The changing nature of wage inequality”, Journal of Population Economics, Vol. 21 No. 1, pp. 21–48.

114

Leueng, J.C.B. (2005), “Social Welfare in China”, in Walker, A. and Wong, C.-K. (Eds.), East Asian Welfare Regimes in Transition, From Confucianism to Globalisation, The Policy Press, Bristol. Liang, R. (2013), “Fighting child malnutrition in rural China”, UNISEF East Asia and Pacific, available at: http://www.unicef.org/eapro/media_20532.html (accessed 9 September 2015). Li, G., Hu, Y. and Pan, X. (1996), “Prevalence and incidence of NIDDM in Daqing City”, Chinese Medical Journal, Vol. 109 No. 8, pp. 599–602. Li, H. (2003), “Economic transition and returns to education in China”, Economics of Education Review, Vol. 22 No. 3, pp. 317–328. Lindenberg, S.M., Coleman, J.S. and Nowak, S. (1986), Approaches to Social Theory: Based on the W.I. Thomas and Florian Znaniecki Memorial Conference on Social Theory, Russell Sage Foundation. Li, S. and Luo, C. (2007), “Income Differences and Social Justice (收入差距与社会公平)”, in Chi, F. (Ed.), 2007 Evaluation Report on China’s Reform, Foreign Languages Press. Li, S. and Luo, C. (2010), “Reestimating the Income Gap between Urban and Rural Households in China”, in Whyte, M.K. (Ed.), One Country, Two Societies: Rural Urban Inequality in Contemporary China, Harvard University Press. Literature Editorial Board of the CPC Central Committee (Ed.). (1993a), “Unity Depends on Ideals and Discipline (一靠理想， 二靠纪律才能团结起来)”, Selected Works of Deng Xiaoping, Vol. 3 (邓小平文选第三卷), Vol. 3, People’s Press (人民出版 社）, Beijing, pp. 110–111. Literature Editorial Board of the CPC Central Committee (Ed.). (1993b), “Interviewed by Mike Wallace (邓小平答美国记者迈 克·华莱士问)”, Selected Works of Deng Xiaoping, Vol. 3 (邓小 平文选第三卷), Vol. 3, People’s Press (人民出版社）, Beijing, pp. 110–111. Liu, J., Ma, H., He, Y.L., Xie, B., Xu, Y.F., Tang, H.Y., Li, M., et al. (2011), “Mental health system in China: history, recent service reform and future challenges”, World Psychiatry, Vol. 10 No. 3, pp. 210–216. Liu, M., Zhang, Q., Lu, M., Kwon, C.-S. and Quan, H. (2007), “Rural and urban disparity in health services utilization in China”, Medical Care, Vol. 45 No. 8, pp. 767–774.

115

Liu, X. and Cao, H. (1992), “China’s Cooperative Medical System: its historical transformations and the trend of development”, Journal of Public Health Policy, Vol. 13 No. 4, pp. 501–511. Liu, X. and Yi, Y. (2004), The Health Sector in China Policy and Institutional Review, No. 34525, The World Bank, pp. 1–99. Liu, Y. (2002), “Reforming China’s urban health insurance system”, Health Policy, Vol. 60 No. 2, pp. 133–150. Liu, Y. (2004), “China’s public health-care system: facing the challenges.”, Bulletin of the World Health Organization, Vol. 82 No. 7, pp. 532–538. Liu, Y., Rao, K. and Fei, J. (1998), “Economic transition and health transition: comparing China and Russia”, Health Policy, Vol. 44 No. 2, pp. 103–122. Liu, Y., Rao, K. and Hsiao, W.C. (2003), “Medical expenditure and rural impoverishment in China”, Journal of Health, Population, and Nutrition, Vol. 21 No. 3, pp. 216–222. Lou, V.W.Q. and Gui, S. (2011), “Chapter 14. The psychological well-being of older people”, in Chen, S. and Powell Jason L. (Eds.), Aging in Perspective and the Case of China, Nova Science Publishers, Inc., pp. 213–229. Lozoff, B., Jimenez, E. and Smith, J.B. (2006), “Double burden of iron deficiency in infancy and low socioeconomic status: a longitudinal analysis of cognitive test scores to age 19 years”, Archives of Pediatrics & Adolescent Medicine, Vol. 160 No. 11, pp. 1108–1113. Lu, D. and Neilson, W. (2004), “Introduction: West China Development — Issues and Challenges”, China’s West Region Development, World Scientific, pp. 1–14. Lundberg, O. (2009), “How do welfare policies contribute to the reduction of health inequalities?”, Eurohealth, Vol. 15 No. 3, pp. 24–27. Lundberg, O., Yngwe, M.Å., Stjärne, M.K., Björk, L. and Fritzell, J. (2008), “The Nordic Experience: Welfare States and Public Health”, Health Equity Studies, No. No. 12. Lynch, J., Smith, G.D., Harper, S., Hillemeier, M., Ross, N., Kaplan, G.A. and Wolfson, M. (2004), “Is Income Inequality a Determinant of Population Health? Part 1. A Systematic Review”, Milbank Quarterly, Vol. 82 No. 1, pp. 5–99. Lynch, J.W., Smith, G.D., Kaplan, G.A. and House, J.S. (2000), “Income inequality and mortality: importance to health of individual income, psychosocial environment, or material conditions”, BMJ, Vol. 320 No. 7243, pp. 1200–1204.

116

Manson, J.E., Rimm, E.B., Stampfer, M.J., Colditz, G.A., Willett, W.C., Krolewski, A.S., Rosner, B., et al. (1991), “Physical activity and incidence of non-insulin-dependent diabetes mellitus in women”, Lancet (London, England), Vol. 338 No. 8770, pp. 774–778. Manzo, G. (2007), “Variables, Mechanisms, and Simulations: Can the Three Methods Be Synthesized?: A Critical Analysis of the Literature”, Revue Française de Sociologie, Vol. 48, pp. 35–71. Mao, Q. (2011), “Can Western Development Program Help to Reduce Income Inequality in China’s Western Region—— An Empirical Study Based on the Difference in Difference Model (西部大开发有助于缩小西部地区的收入不平等吗——基 于双倍差分法的经验研究)”, Finance & Economics, No. 9, pp. 94–103. Mao, Z. (2007), Health System of China: Overview of Challenges and Reforms, United Nations Economics and Social Commission for Asia and the Pacific (UNE State Council AP), Thailand. Ma, R.C.W., Lin, X. and Jia, W. (2014), “Causes of type 2 diabetes in China”, The Lancet Diabetes & Endocrinology, Vol. 2 No. 12, pp. 980–991. Marmot, M.G., Smith, G.D., Stansfeld, S., Patel, C., North, F., Head, J., White, I., et al. (1991), “Health inequalities among British civil servants: the Whitehall II study”, Lancet, Vol. 337 No. 8754, pp. 1387–1393. Marshall, G. and Firth, D. (1999), “Social mobility and personal satisfaction: evidence from ten countries”, The British Journal of Sociology, Vol. 50 No. 1, pp. 28–48. Matthews, K.A., Katholi, C.R., McCreath, H., Whooley, M.A., Williams, D.R., Zhu, S. and Markovitz, J.H. (2004), “Blood Pressure Reactivity to Psychological Stress Predicts Hypertension in the CARDIA Study”, Circulation, Vol. 110 No. 1, pp. 74–78. McNally, C.A. (2004), “Sichuan: Driving Capitalist Development Westward”, The China Quarterly, Vol. 178, pp. 426–447. Meng, Q. (2005), Review of Provider Organization Reforms in China, No. 34528, The World Bank, pp. 1–40. Meng, Q. (2006), Health Care Pricing and Payment Reform in China: The Implications for Health Service Delivery and Cost Containment, Department of Public Health Sciences, Karolinska Institutet, Stockholm, Sweden, Stockholm.

117

Meng, Q. and Tang, S. (2013), “Universal Health Care Coverage in China: Challenges and Opportunities”, Procedia - Social and Behavioral Sciences, Vol. 77, pp. 330–340. Midgley, J. (1999), “Growth, Redistribution, and Welfare: Toward Social Investment”, Social Service Review, Vol. 73 No. 1, pp. 3–21. Miller, G., Luo, R., Zhang, L., Sylvia, S., Shi, Y., Foo, P., Zhao, Q., et al. (2012), “Effectiveness of provider incentives for anaemia reduction in rural China: a cluster randomised trial”, BMJ : British Medical Journal, Vol. 345, available at:http://doi.org/10.1136/bmj.e4809. Miller, K.A., Siscovick, D.S., Sheppard, L., Shepherd, K., Sullivan, J.H., Anderson, G.L. and Kaufman, J.D. (2007), “LongTerm Exposure to Air Pollution and Incidence of Cardiovascular Events in Women”, New England Journal of Medicine, Vol. 356 No. 5, pp. 447–458. MoH. (2012), The Nutrition Development Report of Chinese Children Aged 0–6 (2012), Ministry of Health, Beijing, pp. 1–18. MoH, MoST and NBSC. (2005), The general survey report of nutrition and health in China in 2002 (中国居民营养与健康 状况调查报告 2002), People’s Medical Publishing House, Beijing. Mok, K.H., Wong, Y.C., Walker, R.M. and Zhang, X. (2010), “Embracing the market: Examining the consequences for education, housing, and health in Chinese cities”, in Mok, K.H. and Ku, Y.-W. (Eds.), Social Cohesion in Greater China: Challenges for Social Policy and Governance, World Scientific Publishing, New Jersey, pp. 187–224. Morgan, S.L. and Winship, C. (2007), Counterfactuals and Causal Inference: Methods and Principles for Social Research, 1 edition., Cambridge University Press, New York. Moulton, B.R. (1990), “An Illustration of a Pitfall in Estimating the Effects of Aggregate Variables on Micro Unit”, The Review of Economics and Statistics, Vol. 72 No. 2, pp. 334– 38. Naughton, B. (2007), The Chinese Economy: Transitions And Growth, Mit Press. NBSC. (2013), “Ma Jiantang Answering Reporters’ Questions about National Economic Operation in 2012 (马建堂就 2012 年 国 民 经 济 运 行 情 况 答 记 者 问 )”, National Bureau of Statistics of China, 18 January, available at:

118

http://www.stats.gov.cn/tjdt/gjtjjdt/t20130118_402867315.h tm (accessed 21 June 2013). NBSC. (2015), “National data”, National Bureau of Statistics of available at: China, http://data.stats.gov.cn/easyquery.htm?cn=C01 (accessed 5 August 2015). Nee, V. (1989), “A Theory of Market Transition: From Redistribution to Markets in State Socialism”, American Sociological Review, Vol. 54 No. 5, pp. 663–681. Nee, V. (1991), “Social Inequalities in Reforming State Socialism: Between Redistribution and Markets in China”, American Sociological Review, Vol. 56 No. 3, pp. 267–282. Nee, V. (1992), “Organizational Dynamics of Market Transition: Hybrid Forms, Property Rights, and Mixed Economy in China”, Administrative Science Quarterly, Vol. 37 No. 1, pp. 1–27. Nee, V. (1996), “The Emergence of a Market Society: Changing Mechanisms of Stratification in China”, American Journal of Sociology, Vol. 101 No. 4, pp. 908–949. Nee, V. and Cao, Y. (2005), “Market transition and the firm: institutional change and income inequality in urban China”, Management and Organization Review, Vol. 1 No. 1, pp. 23–56. Nee, V. and Matthews, R. (1996), “Market Transition and Societal Transformation in Reforming State Socialism”, Annual Review of Sociology, Vol. 22, pp. 401–435. NHFPC, (National Health and Family Planning Commission of PRC). (2015a), China Family Development Report 2015, China Population Publishing House. NHFPC, (National Health and Family Planning Commission of PRC). (2015b), “Press conference for releasing the ‘2015 Report on Chinese Nutrition and Chronic Disease’ (国新办 《中国居民营养与慢性病状况报告（2015）》新闻发布会文字 实录 )”, National Health and Family Planning Commision of the PRC, 30 June, available at: http://www.nhfpc.gov.cn/xcs/s3574/201506/6b4c0f873c174ac e9f57f11fd4f6f8d9.shtml (accessed 9 September 2015). Norton, B.J. (1978), “Karl Pearson and statistics: the social origins of scientific innovation”, Social Studies of Science, Vol. 8 No. 1, pp. 3–34.

119

Oberschall, A. (1996), “The Great Transition: China, Hungary, and Sociology Exit Socialism Into the Market”, American Journal of Sociology, Vol. 101 No. 4, pp. 1028–1041. Ohlson, L.O., Larsson, B., Björntorp, P., Eriksson, H., Svärdsudd, K., Welin, L., Tibblin, G., et al. (1988), “Risk factors for type 2 (non-insulin-dependent) diabetes mellitus. Thirteen and one-half years of follow-up of the participants in a study of Swedish men born in 1913”, Diabetologia, Vol. 31 No. 11, pp. 798–805. Olshansky, S.J. and Ault, A.B. (1986), “The Fourth Stage of the Epidemiologic Transition: The Age of Delayed Degenerative Diseases”, The Milbank Quarterly, Vol. 64 No. 3, pp. 355– 391. Omran, A.R. (1971), “The epidemiologic transition. A theory of the epidemiology of population change”, The Milbank Memorial Fund Quarterly, Vol. 49 No. 4, pp. 509–538. Omran, A.R. (2005), “The Epidemiologic Transition: A Theory of the Epidemiology of Population Change”, The Milbank Quarterly, Vol. 83 No. 4, pp. 731–757. O’Toole, T.E., Conklin, D.J. and Bhatnagar, A. (2008), “Environmental risk factors for heart disease”, Reviews on Environmental Health, Vol. 23 No. 3, pp. 167–202. Ouyang, Y. (2013), “China tackles illness-led poverty as financing gap grows”, The Lancet Oncology, Vol. 14 No. 1, p. 19. Pan, X.R., Yang, W.Y., Li, G.W. and Liu, J. (1997), “Prevalence of diabetes and its risk factors in China, 1994. National Diabetes Prevention and Control Cooperative Group”, Diabetes Care, Vol. 20 No. 11, pp. 1664–1669. Park, A., Song, X., Zhang, J. and Zhao, Y. (2004), The Growth of Wage Inequality in Urban China, 1988 to 1999, Department of Economics, University of Michigan, Ann Arbor. Park, A. and Wang, S. (2001), “China’s poverty statistics”, China Economic Review, Vol. 12 No. 4, pp. 384–398. Park, S.-G., Min, K.-B., Chang, S.-J., Kim, H.-C. and Min, J.-Y. (2009), “Job stress and depressive symptoms among Korean employees: the effects of culture on work”, International Archives of Occupational and Environmental Health, Vol. 82 No. 3, pp. 397–405. Pearson, K. (1957), The Grammar of Science, Meridian Books. Pei, X. and Rodriguez, E. (2006), “Provincial income inequality and self-reported health status in China during 1991-7”, Journal of Epidemiology and Community Health, Vol. 60 No. 12, pp. 1065–1069.

120

Phillips, M.R., Zhang, J., Shi, Q., Song, Z., Ding, Z., Pang, S., Li, X., et al. (2009), “Prevalence, treatment, and associated disability of mental disorders in four provinces in China during 2001–05: an epidemiological survey”, The Lancet, Vol. 373 No. 9680, pp. 2041–2053. Pickett, K.E. and Wilkinson, R.G. (2015), “Income inequality and health: A causal review”, Social Science & Medicine, Vol. 128, pp. 316–326. Popkin, B.M., Keyou, G., Zhai, F., Guo, X., Ma, H. and Zohoori, N. (1993), “The nutrition transition in China: a cross-sectional analysis”, European Journal of Clinical Nutrition, Vol. 47 No. 5, pp. 333–346. Preston, S.H. (1975), “The Changing Relation between Mortality and level of Economic Development”, Population Studies, Vol. 29 No. 2, pp. 231–248. Rahn, W.M., Kroeger, B. and Kite, C.M. (1996), “A Framework for the Study of Public Mood”, Political Psychology, Vol. 17 No. 1, pp. 29–58. Rodgers, G.B. (1979), “Income and inequality as determinants of mortality: An international cross-section analysis”, Population Studies, Vol. 33 No. 2, pp. 343–351. Roland, G. (2008), Privatization: Successes and Failures, Columbia University Press. Rosenlund, M. (2005), Environmental Factors in Cardiovascular Disease, Institute of Environmental Medicine Karolinska Institutet, Stockholm, Sweden. Rubin, D.B. (1974), “Estimating causal effects of treatments in randomized and nonrandomized studies”, Journal of Educational Psychology, Vol. 66 No. 5, pp. 688–701. Rubin, D.B. (1977), “Assignment to Treatment Group on the Basis of a Covariate”, Journal of Educational Statistics, Vol. 2 No. 1, pp. 1–26. Rubin, D.B. (1978), “Bayesian Inference for Causal Effects: The Role of Randomization”, The Annals of Statistics, Vol. 6 No. 1, pp. 34–58. Savage, M. and Burrows, R. (2007), “The Coming Crisis of Empirical Sociology”, Sociology, Vol. 41 No. 5, pp. 885–899. Schucher, G. and Hebel, J. (2006), The Emergence of a New “Socialist” Market Labour Regime in China, Working Paper No. GIGA WP 39/2006, German Institute of Global and Area Studies, Rochester, NY. Sclar, E.D. (2001), You Don’t Always Get What You Pay For: The Economics of Privatization, Cornell University Press.

121

Shan, J. (2013), “Nearly 120,000 new cases of MDR-TB in China available at: every year”, China Daily, http://www.chinadaily.com.cn/china/201304/19/content_16423914.htm (accessed 10 October 2014). Sidel, V.W. (1982), “Medical care in China: equity vs modernization.”, American Journal of Public Health, Vol. 72 No. 11, pp. 1224–1226. Sidel, V.W. (1993), “New lessons from China: equity and economics in rural health care.”, American Journal of Public Health, Vol. 83 No. 12, pp. 1665–1666. Silverman, E.M., Skinner, J.S. and Fisher, E.S. (1999), “The Association between For-Profit Hospital Ownership and Increased Medicare Spending”, New England Journal of Medicine, Vol. 341 No. 6, pp. 420–426. Sobel, M.E. (2006), “Spatial Concentration and Social Stratification: Does the Clustering of Disadvantage ‘Beget’ Bad Outcomes?”, in Bowles, S., Durlauf, S.N. and Hoff, K. (Eds.), Poverty Traps, Princeton University Press, Princeton and Oxford. Speed, T.P. (1990), “Complexity, calibration and causality in influence diagrams”, in Oliver, R.M. and Smith, J.Q. (Eds.), Influence Diagrams, Belief Nets, and Decision Analysis, Wiley, New York. Stark, D. (1991), “Path Dependence and Privatization Strategies in East Central Europe”, East European Politics & Societies, Vol. 6 No. 1, pp. 17–54. Starr, P. (1988), “The Meaning of Privatization”, Yale Law and Policy Review, No. 6, pp. 6–41. Steptoe, A., Deaton, A. and Stone, A.A. (2015), “Subjective wellbeing, health, and ageing”, The Lancet, Vol. 385 No. 9968, pp. 640–648. Stevenson, B. and Wolfers, J. (2008), Economic Growth and Subjective Well-Being: Reassessing the Easterlin Paradox, Working Paper No. 14282, National Bureau of Economic Research, available at: http://www.nber.org/papers/w14282 (accessed 18 September 2015). Stevenson, B. and Wolfers, J. (2013), Subjective Well-Being and Income: Is There Any Evidence of Satiation?, CESifo Working Paper Series No. 4222, CESifo Group Munich, available at: https://ideas.repec.org/p/ces/ceswps/_4222.html (accessed 20 September 2015). Stoltzfus, R.J. (2001), “Iron-deficiency anemia: reexamining the nature and magnitude of the public health problem.

122

Summary: implications for research and programs”, The Journal of Nutrition, Vol. 131 No. 2, pp. 697–701. Subramanian, S.V. and Kawachi, I. (2004), “Income Inequality and Health: What Have We Learned So Far?”, Epidemiologic Reviews, Vol. 26 No. 1, pp. 78–91. Sun, L. (2008), “Societal Transition: New Issues in the Field of the Sociology of Development”, Modern China, Vol. 34 No. 1, pp. 88–113. Sun, W., Wu, H. and Wang, L. (2011), “Occupational Stress and Its Related Factors among University Teachers in China”, Journal of Occupational Health, Vol. 53 No. 4, pp. 280–286. Sun, Z., Ming, L., Zhu, X. and Lu, J. (2002), “Prevention and control of hepatitis B in China”, Journal of Medical Virology, Vol. 67 No. 3, pp. 447–450. Tang, S., Bloom, G., Feng, X.S., Lucas, H., Gu, X.Y., Segall, M., Singleton, G., et al. (1994), Financing Health Services in China: Adapting to Economic Reform, Institute of Development Studies. Tang, S. and Meng, Q. (2004), “Introduction to the urban health system and review of reform initiatives”, in Bloom, G. and Tang, S. (Eds.), Health Care Transition in Urban China, Ashgate Pub Ltd, pp. 17–38. Tatarkiewicz, W. (1976), Analysis of Happiness, Martinus Nijhoff. The Central People’s Government of the PRC. (2007), “Report from Jiang Zemin at the15th National Congress of the Chinese Communist Party (江泽民在中国共产党第十五次全 国代表大会上的报告)”, The Central People’s Government of the People’s Republic of China, available at: http://www.gov.cn/test/2008- 07/11/content_1042080.htm. The Central People’s Government of the PRC. (2013), “Chinese Center for Disease Control and Prevention: 24% of people above 15 years diagnosed by hypertension (中国疾控中心:我 国 15 岁及以上人群高血压患病率 24%)”, The Central People’s Government of the People’s Republic of China, 9 October, available at: http://www.gov.cn/jrzg/201310/09/content_2502324.htm. The Economist. (2015), “Lifestyles of the rich and infamous”, The Economist, 4 July. The World Bank. (2013), “GINI index”, The World Bank, available at: http://data.worldbank.org/indicator/SI.POV.GINI (accessed 21 June 2013).

123

The World Bank. (2015), “Poverty & Equity”, The World Bank, available at: IBRD-IDA, http://povertydata.worldbank.org/poverty/country/CHN (accessed 8 September 2015). Tian, J. and Wang, X. (2005), “An epidemiological survey of job stress and health in four occupational populations in Fuzhou city of China”, Stress and Health, Vol. 21 No. 2, pp. 107–112. Tiemann, O., Schreyögg, J. and Busse, R. (2012), “Hospital ownership and efficiency: a review of studies with particular focus on Germany”, Health Policy (Amsterdam, Netherlands), Vol. 104 No. 2, pp. 163–171. Torres, S.J. and Nowson, C.A. (2007), “Relationship between stress, eating behavior, and obesity”, Nutrition, Vol. 23 No. 11–12, pp. 887–894. United Nations Statistics Division. (2005), Handbook on Poverty Statistics: Concepts, Methods and Policy Use, United Nations Statistic Division, p. 406. Vedom, J. and Cao, H. (2011), “Health Care Access and Regional Disparities in China”, Risques de Santé En Sociétés, No. 1, pp. 63–78. Veenhoven, R. (1984), Conditions of Happiness, D. Reidel Publishing Company, Dordrecht, Boston, Lancaster. Veenhoven, R. (2008), “Sociological Theories of Subjective Wellbeing”, in Eid, M. and Larsen, R.J. (Eds.), The Science of Subjective Well-Being, Guilford Press, New York, pp. 44–61. Veenhoven, R. (2012), “Happiness: Also Known as ‘Life Satisfaction’ and ‘Subjective Well-Being’”, in Land, K.C., Michalos, A.C. and Sirgy, M.J. (Eds.), Handbook of Social Indicators and Quality of Life Research, Springer Netherlands, pp. 63–77. Verme, P. (2011), Life Satisfaction and Income Inequality, The World Bank. Veselỳ, A. and Smith, M.L. (2008), “Macro-micro linkages and the role of mechanisms in social stratification research”, Sociologickỳ \vcasopis/Czech Sociological Review, No. 03, pp. 491–509. Wagstaff, A. and van Doorslaer, E. (2000), “Income Inequality and Health: What Does the Literature Tell Us?”, Annual Review of Public Health, Vol. 21 No. 1, pp. 543–567. Wagstaff, A. and Lindelow, M. (2008), “Can insurance increase financial risk?: The curious case of health insurance in

124

China”, Journal of Health Economics, Vol. 27 No. 4, pp. 990–1005. Wagstaff, A. and Yu, S. (2007), “Do health sector reforms have their intended impacts?: The World Bank’s Health VIII project in Gansu province, China”, Journal of Health Economics, Vol. 26 No. 3, pp. 505–535. Wahl, A.-M. (1998), “Positional power, party politics and social security in Mexico: is urban bias politically negotiable?”, International Journal of Sociology and Social Policy, Vol. 18 No. 2/3/4, pp. 103–156. Wang, F. and Luo, W. (2005), “Assessing spatial and nonspatial factors for healthcare access: towards an integrated approach to defining health professional shortage areas”, Health & Place, Vol. 11 No. 2, pp. 131–146. Wasserstein, R.L. and Lazar, N.A. (2016), “The ASA’s statement on p-values: context, process, and purpose”, The American Statistician, Vol. 0 No. ja, pp. 00–00. Weber, E. and Leuridan, B. (2008), “Counterfactual Causality, Empirical Research and the Role of Theory in the Social Sciences (Review Essay)”, Historical Methods, Vol. 41 No. 4, pp. 197–201. van der Wel, K.A. (2011), “Long-term effects of poor health on employment: the significance of life stage and educational level”, Sociology of Health & Illness, Vol. 33 No. 7, pp. 1096–1111. van der Wel, K.A., Dahl, E. and Thielen, K. (2011), “Social inequalities in ‘sickness’: European welfare states and nonemployment among the chronically ill”, Social Science & Medicine (1982), Vol. 73 No. 11, pp. 1608–1617. WHO. (1999), World Health Report 1999., Geneva. WHO. (2002), Preventing Chronic Diseases: A Vital Investment, World Health Organization, Genevar, Switzerland, available at: http://www.who. int/chp/chronic disease reports/en (accessed 10 October 2015). WHO. (2015), “Cardiovascular diseases”, World Health Organization, Western Pacific Region, WHO Representative Office China, available at: http://www.wpro.who.int/china/mediacentre/factsheets/cvd/ en/. WHO, O. of the W.R. in C. and DRC, S.D.D. of C.S.C.D.R.C. (2005), China: Health, Poverty and Economic Developement, Prepared under the WHO’s Macroeconomics and Health initiative., Beijing.

125

Whyte, M.K. (2010a), One Country, Two Societies: Rural-Urban Inequality in Contemporary China, Harvard University Press. Whyte, M.K. (2010b), Myth of the Social Volcano: Perceptions of Inequality and Distributive Injustice in Contemporary China, Stanford University Press, Stanford, California. Whyte, M.K. (2014), “Soaring Income Gaps: China in Comparative Perspective.”, Daedalus, Vol. 143 No. 2, pp. 39–52. Whyte, M.K. and Im, D.-K. (2014), “Is the social volcano still dormant? Trends in Chinese attitudes toward inequality”, Social Science Research, Vol. 48, pp. 62–76. Wilkinson, R., G. (1992), “Income distribution and life expectancy.”, BMJ : British Medical Journal, Vol. 304 No. 6820, pp. 165–168. Wilkinson, R.G. (1994), “The Epidemiological Transition: From Material Scarcity to Social Disadvantage?”, Daedalus, Vol. 123 No. 4, pp. 61–77. Wilkinson, R.G. (1996), Unhealthy Societies: The Afflictions of Inequality, 1 edition., Routledge, London ; New York. Wilkinson, R., G. (1999), “Putting the picture together: prosperity, redistribution, health and welfare”, Social Determinants of Health, Oxford University Press, Oxford, pp. 256–74. Wilkinson, R.G. (2000), “Inequality and the social environment: a reply to Lynch et al.”, Journal of Epidemiology and Community Health, Vol. 54 No. 6, pp. 411–413. Wilkinson, R.G. and Pickett, K.E. (2006), “Income inequality and population health: A review and explanation of the evidence”, Social Science & Medicine, Vol. 62 No. 7, pp. 1768–1784. Wippler, R. (1978), “The structural individualistic approach in dutch sociology”, The Netherlands Journal of Sociology, Vol. 14, pp. 135–155. Wooldridge, J.M. (2002), Econometric Analysis of Cross Section and Panel Data, MIT Press. World Bank. (2003), World Development Indicators 2003, World Bank, Washington, D.C. Wu, H. (2007), “Changes in Regional Income Disparities and Its Development Trends in China After the Western China Development (西部大开发以来中国地区收入差距的变化及其发 展态势)”, Economy & Technology, No. 12. Wu, H., Chi, T.-S., Chen, L., Wang, L. and Jin, Y.-P. (2010), “Occupational stress among hospital nurses: cross-sectional

126

survey”, Journal of Advanced Nursing, Vol. 66 No. 3, pp. 627–634. Wu, J. (2005), Understanding and Interpreting Chinese Economic Reform, 1 edition., Texere, Mason, Ohio. Wu, S., Zhu, W., Wang, Z., Wang, M. and Lan, Y. (2007), “Relationship between burnout and occupational stress among nurses in China”, Journal of Advanced Nursing, Vol. 59 No. 3, pp. 233–239. Xie, Y. and Zhou, X. (2014), “Income inequality in today’s China”, Proceedings of the National Academy of Sciences of the United States of America, Vol. 111 No. 19, pp. 6928–6933. Xinhua News. (2015), “Survey on Poverty: Facing China’s Poor (扶 贫调查：直面中国贫困角落)”, Xinhuanet (新华网), 22 June, available at: http://news.xinhuanet.com/201506/22/c_1115686370.htm (accessed 29 August 2015). Yang, G., Kong, L., Zhao, W., Wan, X., Zhai, Y., Chen, L.C. and Koplan, J.P. (2008), “Emergence of chronic noncommunicable diseases in China”, Lancet (London, England), Vol. 372 No. 9650, pp. 1697–1705. Yang, G., Wang, Y., Zeng, Y., Gao, G.F., Liang, X., Zhou, M., Wan, X., et al. (2013), “Rapid health transition in China, 1990– 2010: findings from the Global Burden of Disease Study 2010”, The Lancet, Vol. 381 No. 9882, pp. 1987–2015. Yang, W. and Kanavos, P. (2012), “The less healthy urban population: income-related health inequality in China”, BMC Public Health, Vol. 12 No. 1. Yang, W., Lu, J., Weng, J., Jia, W., Ji, L., Xiao, J., Shan, Z., et al. (2010), “Prevalence of diabetes among men and women in China”, The New England Journal of Medicine, Vol. 362 No. 12, pp. 1090–1101. Yao, H., Ren, Z., Xu, Z. and An, S. (Eds.). (2009), Blue Book of the West: Economic Development Report of China’s Western Region ( 中国西部经济发展报告 ), Social Sciences Academic Press. Yip, R.I. (2006), “Present Knowledge in Nutrition”, in Bowman, B.A. and Russell, R.M. (Eds.), Present Knowledge in Nutrition, 9th edition., Intl Life Sciences Inst, Washington, D.C, pp. 311–328. Yip, W. (2010), “Disparities in Health Care and Health Status: The Rural-Urban Gap and Beyond”, in Whyte, M. (Ed.), One Country, Two Societies: Rural-Urban Inequality in

127

Contemporary China, Cambridge MA, Harvard University Press. Yip, W. and Hsiao, W. (2009), “China’s health care reform: A tentative assessment”, China Economic Review, Vol. 20 No. 4, pp. 613–619. Yip, W. and Hsiao, W.C. (2008), “The Chinese Health System At A Crossroads”, Health Affairs, Vol. 27 No. 2, pp. 460–468. Zeng, J. (2013), State-Led Privatization in China: The Politics of Economic Reform, Routledge. Zeng, L., Yan, H., Guo, X., Dang, S. and Xie, H. (2003), “Analysis on malnutrition of children under 3 years old in 40 poor counties in the western areas of China (中国西部 40 县农村 3 岁以下儿童营养状况浅析)”, China Public Health (中国公共 ), Vol. 19 No. 1, available 卫 生 at:http://doi.org/10.3321/j.issn:1001-0580.2003.01.027. Zhang, L. (2002), Strangers in the City: Reconfigurations of Space, Power, and Social Networks Within China’s Floating Population, 1 edition., Stanford University Press, Stanford, Calif. Zhang, L., Yi, H. and Rozelle, S. (2010), “Good and Bad News from China’s New Cooperative Medical Scheme”, IDS Bulletin, Vol. 41 No. 4, pp. 95–106. Zhang, X. and Kanbur, R. (2005), “Spatial inequality in education and health care in China”, China Economic Review, Vol. 16 No. 2, pp. 189–204. Zhang, Y. (2014), “Nation aims to lift 10 million rural Chinese out of poverty this year”, China Daily, ChinaDaily.com.cn (中国 , 14 October, available at: 日 报 网 ), http://www.chinadaily.com.cn/china/201410/14/content_18738444.htm (accessed 8 September 2015). Zhan, L. and He, Y. (2012), “Understanding luxury consumption in China: Consumer perceptions of best-known brands”, Journal of Business Research, Vol. 65 No. 10, pp. 1452– 1460. Zhao, W. (2012), “Economic inequality, status perceptions, and subjective well-being in China’s transitional economy”, Research in Social Stratification and Mobility, Vol. 30 No. 4, pp. 433–450. Zhao, Y. and Si, M. (2011), “Dialogue with sociologist Sun Liping: ‘China needs a social progressive movement’ (对话社会学专 家孙立平: ‘中国需要一场社会进步运动’)”, The Southern Daily (南方日报), 18 April.

128

Zhou, M., Wang, H., Zhu, J., Chen, W., Wang, L., Liu, S., Li, Y., et al. (2016), “Cause-specific mortality for 240 causes in China during 1990–2013: a systematic subnational analysis for the Global Burden of Disease Study 2013”, The Lancet, Vol. 387 No. 10015, pp. 251–272. Zhou, X. (2000), “Economic Transformation and Income Inequality in Urban China: Evidence from Panel Data”, American Journal of Sociology, Vol. 105 No. 4, pp. 1135–1174. Zhou, Y. ’an and Zhao, J. (2004), “Marketization, Economic Structural Changes and Transition in Governmental Policy of Economic Structure (市场化、经济结构变迁和政府经济结 构政策转型——中国经验)”, Management World, No. 5. Zimmer, Z., Kaneda, T. and Spess, L. (2007), “An Examination of Urban Versus Rural Mortality in China Using Community and Individual Data”, The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, Vol. 62 No. 5, pp. S349–S357.

Part II