The Chinese Social Benefit System in Transition

Reforms and Impacts on Income Inequality


Address for correspondence: Qin Gao, Fordham University Graduate School of Social Service, 113 West 60th Street, New York, NY 10023. Voice: +1-212-636-6638; fax: +1-212-636-7876.


Economic reforms since the late 1970s have made China one of the world's fastest growing economies but have also led to greater unmet social welfare needs and growing income inequality. This article describes social policy trends in urban and rural China and reviews recent empirical evidence on the transitions and impacts of its social benefit system. The evidence reveals that although urban social benefits experienced significant cutbacks, they are still much more generous than the minimal social benefits for rural residents and migrants. This extremely inequitable situation calls for a larger, progressive social benefit system in China.


Political and economic reforms during the past quarter century have made China one of the world's fastest growing economies. However, some inherent drawbacks of market economies, including unemployment, inequality, and potential social unrest, have emerged during China's reform era. The enlarging inequality gaps, both within and between the urban and rural areas, the emergence of urban unemployment, the migration from rural to urban areas, and the rising aspirations for equal opportunities and social justice have all led to greater unmet social welfare needs and an increasing urgency to address them.

How has the Chinese social benefit system changed since the economic reforms? How has it impacted income inequality? To provide an updated, comprehensive picture, this article describes social policy trends in urban and rural China since the economic reforms and reviews recent empirical evidence on the transitions and impacts of the social benefit system. Alongside many fundamental structural distinctions between urban and rural China, the social benefit provision mechanisms and their impacts differ substantially across the two areas. Therefore, this article examines the urban and rural social benefit systems separately and then discusses their implications for future reform directions in the national context.

In reviewing recent empirical evidence, this article focuses on the literature that has used survey data from the China Household Income Project (CHIP) (unless otherwise noted). CHIP is a national, cross-sectional study collectively designed by a team of Chinese and Western scholars and conducted by the Institute of Economics at the Chinese Academy of Social Sciences. Thus far, three waves of data have been collected, in 1988, 1995, and 2002. Samples of the CHIP study were drawn from larger National Bureau of Statistics samples using a multistage, stratified probability sampling method. With sample provinces from eastern, central, and western regions of China, the CHIP study is nationally representative and arguably the best publicly available data source on Chinese household income and expenditures.1,2 The sample sizes for CHIP urban surveys were 9009, 6931, and 6835 households for the three years, or 31,827, 21,694, and 20,632 individuals, respectively. The sample sizes for CHIP rural surveys were 10,258, 7998, and 9200 households, or 51,352, 34,739, and 37,968 individuals, respectively.

Retrenchment in Urban Social Benefits

Before the economic reforms were launched in the late 1970s, the urban social benefit system was comprehensive in coverage and generous in provision. Under the “full employment” policy in urban areas, almost all working-age urban residents were employed in state-owned or collective enterprises and received various social benefits through their work units.3–6 Since the economic reforms, a series of social policy reforms have been carried out in urban areas, mainly to facilitate market restructuring and to stimulate economic growth and efficiency.7–9

The reforms of the urban social benefit system (or welfare reforms) have been a process of redefining the welfare responsibilities between work units, their employees, and the government. First, the financing of social benefits has shifted from enterprises to general taxes to ease the excess welfare provision burden borne by work units. Housing has become privatized. Pensions and health benefits have shifted to a wider risk-pooling scheme across enterprises to ensure participation of those bearing higher or lower costs. Second, there has been an increasing emphasis on the responsibilities of individuals to finance social benefits, either indirectly through taxation or directly through contributions to specific programs, such as pensions and health. Third, the government has taken a bigger and more direct role in social benefit provision. Two new public assistance programs, unemployment insurance and the Minimum Living Standard Assistance (MLSA), have been established and enforced since the early-to-mid 1990s to provide a safety net for the new urban poor.8,10–12

Empirical evidence confirms the significant retrenchment in urban social benefits since the economic reforms. Analyses using CHIP data show that the average share of social benefit income in total household income for all urban families shrank from 44% in 1988 to 27% in 1995 and then to 25% in 2002, as shown in the left panel of Table 1.13 More specifically, housing benefits and food assistance diminished, while cash transfers and health benefits increased. Housing had been privatized largely through subsidized purchasing over the years, yielding a sharp decrease in its share in total household income (from 19% in 1988 to 10% in 1995 and to 2% in 2002). Food assistance plummeted from 11% of total household income to almost nothing in 1995 and 2002.

Table 1.  Shares of social benefits in total household income (%)
  1. Source: Gao and Riskin.13

Total social benefits442725111
 Cash transfers 91115011
   Pensions 61014000
 Health 4 5 7000
 Housing1910 2000
 Food assistance11 1 0000

On the other hand, the shares of pensions, health benefits, and public assistance income all increased during the same period. Pension income increased significantly (from 6% in 1988 to 10% in 1995 and to 14% in 2002), reflecting the dramatic aging of the urban population during this period rather than a more generous pension system per se. Health benefits also increased from 4% in 1988 to 5% in 1995 and to 7% in 2002. Given that the welfare reforms actually had cut the levels and coverage of health benefits, such an increase most likely reflected the dramatic rise in healthcare costs and increased health consciousness among the public.8 In addition, the value of public assistance saw a sharp increase (from 1 yuan per capita per year in 1988 to 46 yuan per capita per year in 2002, adjusted for the consumer price index); however, its share in total household income remained almost zero and thus played a minimal role in lifting total household income.13

Who received these urban social benefits? Until recently, urban social benefits mainly went to the more privileged socio-demographic groups.14 Employees of public institutions and state-owned and collective enterprises were the main beneficiaries of social benefits before and during the early stages of the welfare reforms. However, economic restructuring reduced the share of such enterprises in the overall economy, while the welfare reforms reduced social benefit provision through the work units. As a result, the social benefits enjoyed by this privileged group have been steadily decreasing. Housing, health, and food benefits all favored those at the top of the market income distribution, especially in the early stages of the welfare reforms. Less educated household heads and larger household size were both associated with lower amounts of social benefits received. Residents in the central and western regions consistently received fewer social benefits of all types than those in the eastern region.

However, some recent demographic and policy trends have helped shift certain social benefits toward the more disadvantaged, mainly the elderly, the unemployed, and those with health problems.8,14 First, the presence of an elderly person (60 years or older) or retiree in a household was increasingly linked to larger amounts of pension income and health subsidies. As mentioned above, this is more of an indicator of the aging population in urban China than of an actual increase in the generosity of the urban pension system.

Second, to accommodate the market economy and to avoid serious social unrest from those who were left behind by the economic reforms, the government has established two new public assistance programs. One is unemployment insurance, which provides a temporary living subsidy for laid-off workers for a limited time. The other program is the MLSA, which provides cash subsidies to families with household incomes below the minimum local living standards. Empirical evidence confirms that both programs heavily target the bottom of the income distribution and help improve the economic conditions of these poorest families in urban China.14 However, recent research has found that the MLSA is far from sufficient to lift families out of poverty, despite the fact that it does help to narrow poverty gaps and relieve severe poverty.15,16 The Ministry of Civil Affairs, which oversees the MLSA program, found that in three sampled cities (Shenyang, Xining, and Bengbu) in 2000, only less than 30% of the urban poor were MLSA recipients.17 Based on survey data from five big cities in 2001 and 2005 (Shanghai, Wuhan, Shenyang, Fuzhou, and Xian), the impact of unemployment insurance on poverty reduction appeared to be even smaller than that of MLSA.18

Urban Social Benefits and Income Inequality

Urban social benefits were distributed more and more unequally over time, both absolutely and relative to market income and overall income inequality. Gao and Riskin13 used the concentration ratio to measure how unequally social benefits were distributed. The concentration ratio is a measure of the inequality of the distribution of a particular income source. It is measured in the same way as the Gini coefficient, except that it considers the distribution of an income source over all income recipients rather than only the recipients of one particular source. Like the Gini coefficient, the higher the value of the concentration ratio, the more unequal a distribution is. As shown in the left panel of Table 2, the concentration ratio of total social benefits in urban areas was consistently higher than that of market income and the overall Gini coefficient, indicating its more unequal distribution. This trend was increasingly predominant over time. More specifically, compared to the overall Gini coefficients, the concentration ratios of total social benefits were higher in all years: 0.25 in 1988 (Gini = 0.23), 0.41 in 1995 (Gini = 0.34), and 0.46 in 2002 (Gini = 0.35).

Table 2.  Inequality in social benefits, market income, and total household income
  1. Source: Gao and Riskin.13

Concentration ratio of income source
 Social benefits0.250.410.460.370.440.71
   Cash transfers0.330.320.330.350.260.72
   Food assistance0.120.270.400.56NANA
 Market income0.200.310.310.350.400.36
Gini coefficient of total household income0.230.340.350.360.420.37

Among individual benefits, pensions were distributed more equally over time, while health and food assistance were distributed more unequally. More specifically, the concentration ratio of pensions was much higher than the overall Gini coefficient in 1988 (0.42 as compared to 0.23), but it decreased to a level slightly lower than the overall Gini in 1995 and 2002. Health benefits saw a striking increase in inequality, with a concentration ratio of 0.83 in 2002 as compared to 0.45 in 1995 and 0.19 in 1988. Food assistance, despite its decreasing share in total household income, also became more unequally distributed over time.13

Housing benefits had the most unusual trends in their distribution: their concentration ratio was 0.30 in 1988, jumped to 0.51 in 1995, and then decreased back to 0.31 in 2002. This transition, however, is a vivid reflection of the housing privatization process in urban China. In 1988, very few urban residents owned their homes, and the majority still lived in free or heavily subsidized public housing. In the same year, the government started a nationwide housing reform, increasing rent charges and the sale of public housing to current occupants. The reform progressed, and by 1995, a larger group of privileged urban residents were given priority to purchase housing from their work units at heavily subsidized prices (which was not considered a housing benefit), while a large proportion still lived in public housing (which was counted as a housing benefit). Therefore, the distribution of housing benefits in 1995 became much more unequal. The government began to build generally affordable and functional housing in 1998 and introduced the publicly accumulated housing fund nationwide in 1999. These reforms, as well as the spread of subsidized purchase opportunities among the urban population, greatly increased urban housing ownership while sharply reducing inequality in urban housing benefits in 2002.8,13

According to several widely used inequality measures, the income inequality reduction impact of total urban social benefits appeared to decrease between 1988 and 2002.14 For example, Table 3 shows that, as a direct result of social benefits, the overall Gini coefficient was reduced by 18% in 1988 and by 14% in 2002. The economic distance between the poor and the rich (as measured by the difference between the ratios of the 10th percentile and the 90th percentile to the medium, respectively) declined by 13% in 1988 and 12% in 2002. Furthermore, this decrease was calculated to be even more significant when giving additional weight to the bottom of the income distribution, as shown by the results using the Atkinson indices. Specifically, as more weight was given to the lower end of the income distribution (that is, as the value of e changed from 0.5 to 1 and to 2), the reduction in income inequality due to social benefits increased in each respective year, and again the role of social benefits in alleviating income inequality was stronger in 1988 (i.e., larger percentage changes) than in 2002.

Table 3.  Reduction in income inequality due to total social benefits in Urban China (%)
  1. Sources: Gao14; and author's calculation from CHIP data.

Gini coefficient1814
Economic distance ( = p90/p50 − p10/p50)1312
Atkinson index
 e = 0.53831
 e = 14239
 e = 26258

Although urban social benefits as a whole helped to reduce income inequality, they were insufficient to close the rising income gap, which was driven by growing market income inequality. The consequential gaps in total household income widened from 1988 to 2002. Certainly, the shrinking size of urban social benefits and their modest effects on income inequality reduction partially accounted for this trend. However, the rising income inequality was mainly due to the increasing share of market income in total household income and the growing inequality in market income itself.14 As shown in Table 4, total social benefits contributed to 48% of the overall income inequality in 1988, and this contribution decreased to 33% in both 1995 and 2002. On the contrary, the contribution of market income to overall income inequality continued to rise from 48% in 1988 to 66% in 1995 and 70% in 2002.

Table 4.  Contribution of income sources to overall income inequality (%)a
  1. Sources: Gao and Riskin.13aThe contribution of a particular income source to overall income inequality is calculated as the share of the income source in total household income multiplied by that source's concentration ratio and then divided by the Gini Coefficient of the total household income. The summed contribution of social benefits and market income was larger than 100% in urban China in 2002 because a negative income component—taxes and fees—reduced overall income inequality by 4%.

Social benefits4833330.60.91.2
Market income486670979996

The Marginal Rural Social Benefit System

In contrast to the comprehensive and generous social benefits in urban China before the welfare reforms, the rural social benefits had always been marginal in coverage and minimal in provision.4,5 The government has done little to improve this situation. Only a very small privileged group of rural residents has had access to pensions due to their prior employment in state-owned or collective enterprises. Housing benefits have never been available to the rural residents.

The prereform, community-based rural health financing and provision system, called the Rural Cooperative Medical System (RCMS), provided basic health services and wide coverage to rural residents. It was jointly funded by family contributions, a portion of the collective welfare fund, and sometimes the government. The system collapsed after the rural economic reforms were launched in 1978, leaving many rural residents unable to afford the dramatically increasing healthcare expenses.19–21 To remedy the negative consequences stemming from the absence of the RCMS, the government has conducted a series of trials since the mid-1990s of a rural healthcare system financed by a combination of household contributions and subsidies from local governments, but a national system has yet to be developed.22

There have been three public assistance programs in rural China. The “Five Guarantees” program was developed as the main institutional arrangement for the rural elderly, disabled, and minors who had no family support or income sources to receive five types of goods for basic needs: food, clothing, medical care, housing, and burial expenses. This system, however, has been ineffective in providing sufficient support for the target population in many rural areas. Two other public assistance programs aiming to protect the vulnerable from natural disasters and human misfortunes—the natural disaster relief system and the collective welfare fund—were in place before the reforms and have been maintained ever since.4,5,23

However, all of these programs were supplemental and only played a marginal role in supporting rural families. Consequentially, social benefits for rural families remained at a minimal level, accounting for only 1% of total household income over time (as shown in the right panel of Table 1), as opposed to more than 25% for urban social benefits.13 Because of this small size, rural social benefits as a whole played a very insignificant role in shaping overall income inequality. Table 4 indicates that overall rural income inequality was predominantly determined by market income, which contributed over 96% to overall inequality, while the contribution of social benefits was only around 1%. However, the impact of rural social benefits did appear to be increasing over time, contributing to 0.6% (in 1988), 0.9% (in 1995), and 1.2% (in 2002) of overall income inequality.13

Despite its very small size relative to total household income, rural social benefits were distributed increasingly unequally over time, both in absolute terms and relative to market income and total household income. As shown in the right panel of Table 2, the concentration ratio of total social benefits increased from 0.37 in 1988 to 0.44 in 1995 and then jumped to 0.71 in 2002, all of which were higher than the concentration ratio of market income and the Gini coefficient for total household income. The 0.71 concentration ratio in 2002 was strikingly high, which was driven by the increased inequality in pensions and health benefits. This reflects that the majority of the rural social benefits—though minimal on average—went to the more privileged groups.13

Minimal Social Benefits for the Migrants

Since the early 1980s an increasing number of migrants who hold rural household registration status (hukou) have moved to work and live in the cities and towns. This migration wave greatly expanded during the 1990s. The number of migrants jumped from 18 million in 1989 to 70 million in 1993 and to 150 million in 2004.24,25 Migrants now make up 11% of the national population and more than 20% of urban residents.

Migrants receive virtually no social benefits. They are usually of working age with earning capabilities and thus do not qualify for rural benefits. Meanwhile, they are not entitled to any urban social benefits due to the lack of registered local city resident status. In 2002, less than 5% of migrants received any pensions, unemployment insurance, or health benefits.13 The net subsidies (total subsidies less taxes and fees paid) received by migrant families were negative (about 1% of total household income), indicating that taxes and fees paid by migrants exceeded the sum of any support they received from the government.1

Conclusion and Discussion

Based on a set of recent literature using the CHIP data, the Chinese social benefit system appears to be a tale of two sides. On the one hand, the urban social benefit system suffered significant cuts from its comprehensive coverage and generous provision before the welfare reforms, although on average social benefits still made up one-quarter of urban families' total household income in 2002. On the other hand, social benefits remained minimal for both rural residents and migrants from rural to urban areas, making up 1% or less of their total household incomes. In both urban and rural China, social benefits, particularly pensions and health care, were distributed increasingly unequally over time. Urban social benefits contributed to reducing overall income inequality, but this reduction decreased over time. Rural social benefits played only a marginal role in reducing overall income inequality because of their small share in total household income. Essentially, market income served as the driving force for overall income inequality in both urban and rural China.

Some important policy implications emerge from these empirical findings. First and foremost, the difficulties faced by rural residents and migrants should be of serious concern to China's policymakers. The virtual nonexistence of social benefits for these two groups, which make up about 70% of the total population, indicates a lack of respect for their equal rights. Since 2005, the government has started several important policy initiatives to address the extremely inequitable situation. These initiatives include eliminating the formerly heavy taxes and fees levied on peasants, restoring and subsidizing the RCMS, and expanding the MLSA program to rural areas. Most of these policies have yet to be fully carried out, and their outcomes need to be evaluated in the future. At the same time, the lack of social benefits for the 150 million migrants remains mostly ignored.

Second, as an explicit redistributive mechanism, social benefits should play a larger, progressive role in order to improve the economic conditions and overall well-being of those in need, to help reduce the income inequality driven by market income, and to promote equal opportunities and social justice for all. Special attention should be paid to the very vulnerable, who are often left behind by both market competition and the social benefit system. Demographic trends, especially population aging and migration, should be taken into consideration in further developing the social benefit system. In the long term, a more progressive social benefit system will be able to improve both societal cohesion and efficiency and thus benefit the nation as a whole.


The author is grateful for the support from the Chiang Ching-Kuo Foundation for International Scholarly Exchange and the Horowitz Foundation for Social Policy. The CHIP surveys that provide data for this research were funded by the Asian Development Bank, the Ford Foundation, and the Swedish International Development Agency. The author also thanks Fuhua Zhai for helpful comments and Rocky Citro for excellent editing assistance.

Conflicts of Interest

The author declares no conflicts of interest.