China’s Pension Problem
Demographic Transition & Rural-to-Urban Migration
Bachelor Thesis by Livia Remeijers
2015
Student Number: 10217002
Supervisor: dhr. prof. dr. S.J.G. Sweder van Wijnbergen Field: Economics
Abstract
The aim of this thesis is to examine the effect of China’s demographic transition and rural-to-urban migration on old-age dependency ratios and system solvency of China’s urban and rural pension systems. Dependency ratios published by the National Bureau of Statistics of China are based on household registration, hence giving a distorted picture of reality. This thesis realizes a simulation that corrects for non-registered migrants residing in urban areas in the calculation of urban and rural dependency ratios. As a result rural dependency ratios are significantly higher than urban dependency ratios. In order to maintain system solvency of the PAYG system, rural contribution rates should drastically increase, putting excessive pressure on the rural pension scheme. Due to large rural-to urban migration of the young, there is a growing surplus in urban pension revenues and a growing deficit in rural pension revenues. The challenge to maintain system sustainability becomes greater as China’s ageing society increases and rural-to-urban migration
continues. A possible solution is to unify urban and rural pension schemes, hence avoiding high contribution rates, but increasing pressure on China’s political system.
Acknowledgement
Foremost, I would like to express my sincere gratitude to my supervisor dhr. prof. dr. S.J.G. Sweder van Wijnbergen for his enthusiasm for my topic, for his time and his interesting and useful comments throughout my bachelor thesis process.
My sincere thanks also goes to my professors form Fudan University in Shanghai who shaped my interest in issues regarding the development of China’s economy. Besides, I thank Fudan University for giving me the opportunity to take part in their exchange program 2014.
Last but not least, I would like to thank my close friend Valerie Dusseldorp with whom I could always talk about my topic and for giving me useful tips, as well as my father Leon Remeijers who enthusiastically read my first draft of my bachelor thesis.
Table of Contents
Abstract ... 2
Chapter 1: Introduction ... 5
Chapter 2: Demographic Background ... 6
2.1 Ageing trend ... 6
2.2 Urbanization trend ... 8
Chapter 3: Pension history ... 9
3.1 Urban pension scheme ... 9
3.2 Rural family based support... 10
Chapter 4: Current pension reforms ... 12
4.1 Urban pension scheme ... 12
4.2 Rural pension scheme ... 13
Chapter 5: Central issues of urban and rural pension schemes ... 15
5.1 Coverage ... 15
5.2 Mismanagement ... 16
5.3 Confidence problem... 16
5.4 Rural-to-urban migration obstacle ... 16
5.5 Inequality between urban and rural pension schemes ... 17
Chapter 6: Empirical evidence on China’s pension problem... 18
6.1 Hypothesis ... 18
6.3 Methods & Results ... 19
6.4 Discussion ... 22
Chapter 7: Simulation ... 24
7.1 Method & Results ... 25
7.2 PAYG system solvency ... 28
Chapter 8: Conclusion ... 30
Chapter 9: Appendix ... 32
Chapter 1: Introduction
China finds itself in the midst of a demographic transition. Fertility rates below
replacement level have led to a shrinking share of the working-age population, while the proportion of elderly has significantly increased over time (Dorfman et al., 2013).
Additionally, since China turned into a ‘socialist market economy’, large groups of young migrants have moved from the countryside to the cities (Peng, 2011). As a consequence age pyramids are uneven across regions. China’s pension system is divided in an urban and a less developed rural pension scheme that operate independently. The uneven and changing urban and rural age pyramids affect pension financing of both schemes. Hence China’s demographic transition and rural-to-urban migration seem to challenge urban and rural pension systems.
There is extensive academic literature about China’s rising pension problem. However, research on how China’s urban and rural pension systems are affected by changing rural and urban age pyramids is missing. Therefore my thesis examines the following research question: What is the effect of China’s demographic transition and rural-to-urban migration on old-age dependency ratios and system solvency of China’s urban and rural pension systems? The old-age dependency ratio is defined as the ratio of elderly, those aged 65 and above, to the working-age population, those aged between 15 and 64, and will be referred to as the dependency ratio in the rest of the thesis. China is the most populated country in the world and has the world’s largest elderly population, indicating the importance of my research (Kinsella & Wan, 2009). Furthermore, in order for China to reach sustainable development of both urban and rural regions, it is of great importance that the pension system is sustainable and that rural-urban economic
disparities diminish.
The remaining of my thesis is structured as follow: chapter two deliberates China’s demographic background regarding its ageing and urbanization trends. Chapter three draws an image of China’s pension history, while current pension reforms are discussed in chapter four. Chapter five elaborates the central issues of urban and rural pension schemes. Followed by chapter six which analyzes whether published dependency ratios reflect the phenomenon of China’s ageing society as well as the shift of
rural-to-urban migration. Chapter seven demonstrates a simulation in order to correct dependency ratios for non-registered migrants. The last part draws the conclusion.
Chapter 2: Demographic Background
China is in the middle of a demographic transition. This chapter will elaborate the development of China’s ageing trend and urbanization trend in order to understand the country’s demographic situation.
2.1 Ageing trend
China’s aging trend is not entirely unique since such a trend is similarly observed in other countries such as Japan, New Zealand, the European Union, the United States of
America, and Canada (Hu & Yang, 2012). The ageing trend is also a global trend as a consequence of increasing urbanization and industrialization. It is a trend that occurs due to what is called ‘demographic transition’, which is the process of declining fertility rates and increasing life expectancy resulting in an older age population structure (Chen & Liu, 2009). Even though we see a similar ageing trend in other countries, some characteristics of China’s growing grey population are unique. The time space in which China has reached an ageing society is exceptional. Whereas developed countries such as the United States and the United Kingdom took over 100 years of demographic transition, China took only 40 years to reach an ageing society (Cai, Giles, O’Keefe, & Wang, 2012). Another exceptional characteristic is the role played by the Chinese government in order to manage population growth. In 1979 the Chinese government implemented the one child policy aiming to reduce birth rates and hence slow the population growth rate (Hesketh, Lu, & Xing, 2005). Due to the introduction of the policy the total fertility rate, defined as the average number of children born per woman, has drastically decreased from 2.7 in 1980 to 1.7 in 2013, see figure 1 (National Bureau of Statistics, 2014).
Figure 1. Total Fertility Rate in China 1980-2013. The fertility rate decreased from 2.7 in
1980 to 1.7 in 2013. Data obtained from China Statistical Yearbook 2014, National
Bureau of Statistics.
While the fertility rate decreased, the life expectancy rate increased due to improved hygiene and health services. Rapid declining fertility rates together with rising life expectancy result in a higher proportion of the elderly relative to the others and thus to a growing grey population (Hesketh et al., 2005). According to the World Bank those aged 65 and above made up only 5% of the total Chinese population in 1980. In 2013 it was 9.7% and Hesketh et al. (2005) expect it to increase to over 15% by 2025, see table 1.
Table 1.
Age composition of China’s total population in 1990, 2000 and 2013 (in %)
1990 2000 2013
Aged 0-14 27.7 22.9 16.4
Aged 15-64 66.7 70.1 73.9
Aged 65 and above 5.6 7.0 9.7
Source. China Statistical Yearbook 2014 National Bureau of Statistics
0 0,5 1 1,5 2 2,5 3 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013
Total Fertility Rate
2.2 Urbanization trend
Before the economic reforms in the early 1980s China’s urbanization was rather low. According to the China Statistical Yearbook 2014 only 11.18% of the population lived in urban areas in 1950. The proportion of urban residents increased to 19.39% by 1980, which is still relatively low. However, the urbanization process has sped up over recent decades. In 2013 more than half, 53.73% of the population lived in urban regions
(National Bureau of Statistics, 2014). The urbanization process is to a large extent driven by China’s exceptional internal migration. Millions of young rural workers move to cities where better job opportunities and higher incomes are available (Peng, 2011). To
emphasize the contribution of rural-to-urban migration to China’s urbanization process we need only look at the example of migrants living in Shanghai. In 2010 the Shanghai population counted for 23 million people of whom only 14 million hold the Shanghai household registration status. This means the other 9 million are categorized as migrants, or ‘temporary population’ (Zhang, 2008).
Assuming that present urbanization trends continue to hold, Peng (2011) argues that urbanization will continue to increase at an exceptional rate. The report “Preparing for China’s Urban Billion” says that China’s urban population will extend by 39% between 2010 and 2025 and reach the 1 urban billion by 2030 (Woetzel et al., 2009). Woetzel et al. (2009) believe that internal migration will continue to be the major driver of China’s urbanization process. According to their research migration will count for 70% of population growth in urban areas between 2005 and 2025. Furthermore, they predict future migrants and existing migrants will count for 40% of the total population living in cities by the year 2025.
Due to large rural-to-urban migration the proportion of elderly is higher in rural compared to urban areas. A consequence is that the ageing process in urban areas has slowed down due to the arrival of young workers. On the contrary, this kind of internal migration has led to a serious ageing challenge in rural regions (Peng, 2011). Hence, we observe a double ageing trend. China as a whole faces an ageing trend and on top of that there is a concentration of a rising old population in rural areas as a consequence of large rural-to-urban migration of the younger ones, leaving the elderly behind. Before studying the effect of changing urban and rural age pyramids on China’s pension system, the
pension history and urban and rural pension schemes will be discussed in the following chapters.
Chapter 3: Pension history
This chapter discusses China’s urban and rural pension history. The Chinese government focused solely on the development of the urban pension scheme, while support for the rural elderly has been family based.
3.1 Urban pension scheme
The Chinese old-aged security system originates from the 1950s after the establishment of the Communist Party (Dunaway & Arora, 2007). The old urban pension system was part of the so called ‘iron rice bowl’, a system that ensured lifetime employment in State Owned Enterprises (SOEs) mostly situated in urban regions (Trinh, 2006). SOEs were obliged to provide their former employees with generous pension benefits of 80%. A Pay-As-You-Go (PAYG) system, in which current contribution revenues finance current pension benefits, was used. State enterprises provided the pensions directly to their retired employees and contribution by employees were not needed (Shen & Williamson, 2010).
However, the social security system as defined above became unsustainable when SOE reforms commenced in the late 1980s. While private companies started to gain large market competitiveness the inefficient state enterprises with a large surplus of employees faced financial difficulties. It turned out that the promised pension benefits were a too big financial burden on the SOEs competitiveness resulting in huge losses (Trinh, 2006). As a consequence many employees got laid off, hence losing all benefits connected to the ‘iron rice bowl’. While the number of people employed in SOEs drastically declined, the amount of people employed in private companies significantly increased. Trinh (2006) argues that up to the late 1990s all people working at private companies as well as self-employed people did not take part in any pension system. Thus, the coverage of the old age urban pension system declined together with the decline in SOEs employees. This
emphasizes the need for urban pension reform aiming to cover all Chinese urban residents.
3.2 Rural family based support
In the 2010 population census provided by the National Bureau of Statistics one observes that family support is the main source of support for Chinese elderly aged 60 and above living in rural areas (National Bureau of Statistics, 2010). In 2010, 47.7% of the rural elderly aged 60 and older depended on family support, whereas this was 33.5% for urban elderly. In the same year 41.2% of rural elders depended on labor income. This was only 14.4% for urban elders. The previous suggests that rural seniors depend for a longer period on income, thus continue working till a later age than urban seniors. For old-aged registered in urban areas the main source of elder support is a pension; 46.3.3% of urban seniors received a pension in 2010, whereas this was only 4.6% for rural seniors, see table 2. The latter emphasizes higher pension coverage in urban compared to rural regions, implying that the rural pension scheme is largely underdeveloped.
Table 2.
Primary sources of support for China's elderly aged 60 and above in 2010 (in %)
Urban Rural
Labor Income 14.4 41.2
Pensions 46.3 4.6
Minimum Living Allowance 3.3 4.5
Property Income 0.6 0.2
Family Support 33.5 47.7
Others 1.9 1.8
Source. China Statistical Yearbook 2010 National Bureau of Statistics, author’s calculation
However, there are multiple reasons for the little attention paid by the Chinese government to the development of a rural pension scheme. One of the reasons is the strong tradition known as ‘filial piety’ that finds its origins in Confucianism (551-479 BC). It is a long-standing moral virtue that goes back thousands of years and obliges
children to support their aged parents (Fan, 2006; Giles, Wang, & Zhao, 2010). Besides, in the time that Mao ruled collectivism was key policy. When rural elder were not able to work anymore they were taken care for by their family or other collectives. They were guaranteed, “food, clothing, housing, medical care, and burial expenses”, which is also known as the social assistance program (Shen & Williamson, 2010, p.240). This good old working tradition ensured no necessity for implementing a new elderly safety net for rural residents (Mitchell, 2013). However, things started to change as demographics changed over time together with the occurrence of social economic reforms.
While for years it was common that aged parents lived with their adult children more young workers move to large cities, not being able to live and take care of their aged parents anymore. Furthermore, another concern for rural elder support is the reduced family size as a result of the one child policy. As people get older due to
improved health service and family size decreases, pressure on adult children to support their family drastically grows. This problem is known as the ‘four-two-one’ problem, which means that one child needs to take care of two parents and four grandparents. This is a large financial burden on working adult children (Pozen, 2013). It has been widely discussed that growing rural-to-urban migration together with falling family size might result in erosion of the traditional elderly support system in rural China (Zhang & Tang, 2008). Therefore, it has been of major importance that the Chinese government
established a new rural pension scheme, not only in the name of social equality, but also to enhance economic development of rural China.
Chapter 4: Current pension reforms
The following sections focus on China’s current pension reforms. The three-pillar urban pension scheme and the new rural pension scheme established in 2009 will be analyzed.
4.1 Urban pension scheme
The current Old Age Insurance urban pension scheme was established in 1997 and is a multi-pillar system financed by the state, employees and employers (Yuan & Feng, 2005). The system consists of three pillars and is designed to cover employees of all sorts of companies. The first pillar is a compulsory social pool based on a PAYG system solely financed by employers. The employer contributes 20% of the employee’s monthly wage to the pool. Seniors are entitled to a pension after contribution of minimum 15 years and when reaching the retirement age, which is 55 and 60 for females and males respectively. The second pillar is the mandatory individual account jointly financed by employees and employers. The employee starts of at a contribution of 4% and increases up to 8% over time, while the employer adds 7% of the wage to the individual account (Trinh, 2006). The pension will be calculated as a percentage of the average salary in the corresponding city in the year before retirement together with the accumulated capital in the individual account (Wu, 2013). In case the retiree has served the Old Age Insurance pension scheme for 35 years he/she receives a pension equal to the replacement rate of 58.5% (20% first pillar plus 38.5% second pillar) of the average salary in the respective city on a monthly basis. The third pillar consists of a voluntary supplement private savings account
managed by individual firms or commercial insurance companies for people who desire a higher pension payment upon retirement, see table 3 (Shao & Xu, 2001; Yuan & Feng,
Table 3.
Three pillar-urban pension scheme from 1997 – present
Pillar Type Contribution
Pillar I Compulsory Basic Pension
(DB, PAYG)
Employer: 20% of employee’s wage
Pillar II Compulsory Individual
Account (DC)
Employer: 7% of employee’s wage
Employee: 4-8% of wage Pillar III Voluntary Private Savings
Account
Individually
Note. DB: Defined Benefit; DC: Defined Contribution Source. Trinh (2006)
4.2 Rural pension scheme
Up to 2009, the Chinese Ministry of Civil Affairs conducted many experimental pension pilots in China’s more developed rural regions. However, none of these pension pilots have succeeded in large coverage (Li, 2007). Reasons for consistent failure have been the solely individual contributions, very low pension benefits and weak management of the pension scheme by local governments. Another reason and maybe one of the most important was the general lack of confidence in the sustainability of the provided rural pension plans (Dorfman et al., 2013). Rural workers rather saved their money than putting it in personal accounts not being entirely sure that local governments took good care of their savings. As a consequence rural pension coverage has remained extremely low at less than 10% (Cai et al., 2012).
Eventually, in 2009 the State Council started to establish the new rural pension system with the aim of full coverage of all counties by the year 2020 (Dorfman et al.,
2013). This is the first time that the government decided to introduce a nationwide rural
pension scheme try-out and directly contributes to the rural pension scheme. The new rural system target is to guarantee all rural residents with a basic pension. The scheme design consists of two components, see table 4. The first component is the basic pension,
based on a PAYG system, subsidized by the central government. This basic pension is equal to the minimum benefit of 55 RMB/month per pensioner, which is close to China’s poverty level (Shen & Williamson, 2010). The second component is the individual account financed through individual contributions and a partial match by local
governments depending on the level of individual contribution. The higher the individual contribution the greater the partial subsidy made by the local government (Cai et al., 2012). The latter is to increase the incentive to participate in the new rural pension scheme and to make higher individual contributions.
The new rural pension system remains voluntary. A retiree has the right to receive the pension provided he/she has reached the retirement age (55 for women and 60 for men) and has taken part in the pension system for at least 15 years (Salditt, Whiteford &
Adema, 2008). Rural pension benefits are calculated as in the urban pension scheme, by
dividing the total accumulated capital on the individual account upon retirement by 139, assuming life expectancy to be 11 years and 7 months after retirement (Dorfman et al., 2013).Seniors older than 60 at the time the new pension system is introduced receive the basic pension benefit on condition that their adult children are participating in the new scheme. This is called ‘family binding’ and is another strong incentive for rural residents to participate in the new rural pension scheme. The management of the new scheme is currently at county level, but aims to reach provincial levels as soon as possible, hence centralizing management of the individual accounts (Cai et al., 2012). Although, China’s pension system is further developed than ever before, there remain some issues
concerning urban and rural pension schemes. The next chapter will discuss the central issues.
Table 4.
New rural pension scheme since 2009
Component Type Contribution
I, Basic Pension DB, PAYG 55RMB/month funded by
central government
II, Individual Account DC Individually and partial
match by local government
Note. DB: Defined Benefit, DC: Defined Contribution Source. Cai et al. (2012)
Chapter 5: Central issues of urban and rural pension schemes
Several issues concerning urban and rural pension systems will be discussed. First, the concern about pension coverage will be deliberated, as well as the problem of
mismanagement of individual accounts and the general lack of confidence in the pension schemes. Followed by the issues that hamper rural-to-urban migration and the problem of inequality between urban and rural pension systems.
5.1 Coverage
There exist a serious concern about pension coverage. According to the China Statistical Yearbook 2014 urban pension coverage was 29.6% in 2006 and increased up to 44.1% by 2013 (National Bureau of Statistics, 2014). Hence, more than half of China’s urban employees are not covered. The people who are not covered are mostly long-term unemployed, self-employed, disabled and illegal migrants. For rural areas it is estimated that less than 10% of the residents are covered (Wu, 2013). Reasons might be the
differences in the economic development level among rural regions. For example, the coverage in more developed rural areas such as Beijing’s suburbs is much larger than in poorer rural regions. Simply because residents of less developed rural areas cannot afford the demanded contributions (Hu & Yang, 2012). Hence, the concern whether ‘the poorest of the poor’ will be reached with the new rural pension scheme remains present.
5.2 Mismanagement
Experience has shown that there is a high risk of mismanagement of the individual accounts. In urban regions it widely happens that personal accounts become notional, meaning that they are ‘empty’, because the capital is used for other means, such as current basic pension payments (Frazier, 2004; Pozen, 2013). Besides, the goals of local governments might not be in line with those of the central government. While, the central government aims at balancing social insurance expenditure between urban and rural areas, local governments might not have the incentive to diminish regional inequality (Wu, 2013).
5.3 Confidence problem
The new rural pension system is voluntary and thus rural residents will only participate in the plan if they have enough confidence in the pension scheme. The occurrence of
‘empty’ individual accounts is a discouragement for employees to participate in the rural pension scheme. Employees become reluctant to placing their money on an individual account knowing that their saved money might be used for other means (Salditt et al.,
2008). In order to gain trust from the working population local governments should meet
their promises and operate in a transparent way (Shen & Williamson, 2010).
5.4 Rural-to-urban migration obstacle
There is a growing concern that current pension schemes hamper rural-to-urban
migration. It is uncertain whether rural residents moving to urban areas are able to bring their individual accounts supplemented by local governments (Dorfman et al., 2013). It is more difficult for local governments to manage the personal accounts of residents who moved away from their home counties. Theoretically rural ‘hukou’ residents living in the urban area who not participate in the urban pension scheme are allowed to take part in the new rural pension system. However, in practice local governments might be reluctant to match individual accounts with a partial contribution knowing that participants live beyond their home county (Cai et al., 2012). Hence, it is possible that a large group of migrants living in cities do not receive at all or a very small pension upon retirement.
However, numbers to support the previous are unfortunately not accessible. This problem needs more attention by emphasizing portability rights of the individual accounts such that coverage and overall labor mobility can be improved.
5.5 Inequality between urban and rural pension schemes
Of all issues related to urban and rural pension schemes, the urban-rural inequality problem is a rising concern. For decennia, the Chinese government solely focused on the development of the urban pension scheme, leaving rural elderly disadvantaged (Shen & Williamson, 2010). Besides, it is widely acknowledged that there exist large income disparities between urban and rural areas (Wu, 2013). Rural incomes are much lower than urban incomes, due to lower labor productivity. In table 5 we observe that income
disparity between urban and rural regions has increased over time. In 1978 urban wages were 2.57 times higher than rural wages, while the urban-rural income ratio had risen to 3.03 by 2013.
Table 5.
Urban and rural per capita net income
Aggregate Data
1978 1990 2000 2012 2013
Per Capita Net Income Urban Residents (RMB)
343 1510 6280 24565 26955
Per Capita Net Income Rural Residents (RMB)
134 686 2253 7917 8896
Urban-Rural Income Ratio 2.57 2.20 2.79 3.10 3.03
Sources. China Statistical Yearbook 2014 National Bureau of Statistics, author’s calculation
Since, basic pensions are calculated as a percentage of local average wages, a rural retiree received an average basic pension three times lower than that of an urban retiree in 2013. Furthermore, Lu (2010) claims that 90% of government expenditure goes to urban residents while only 10% is left for rural residents. However, two thirds of China’s
Depending on the state and including individual account differences, rural retirees receive nearly a 10 to 20 times lower total pension compared to urban retirees (Wu, 2013). Moreover, with a growing ageing society and increasing rural-to-urban migration one might expect the dependency ratios to increase further, resulting in even larger demographic disparities between urban and rural regions (Cai et al., 2012). The latter can be illustrated by looking at the development of dependency ratios (those aged 65 and above as a proportion of those aged between 15 and 64). Dependency ratios are
calculated in order to show changes in the population age structure, which in turn have consequences for pension financing (Mason & Kinugasa, 2008). In the next chapter we will look at whether these demographic disparities are reflected in regional dependency ratios.
Chapter 6: Empirical evidence on China’s pension problem
“The dependency ratio is used to describe the relationship between the accumulation of pension funds and demographic changes” (Hu & Yang, 2012, p.203). A change in dependency ratios affects the accumulation of capital available for pension payments. In order to maintain financial stability of the PAYG-system, demographic changes will affect pension benefits or require modified contribution rates (Disney, 2000; Leers,
Meijdam & Verbon, 2004). This chapter focuses on the development of China’s national
dependency ratio as well as the development of rural and urban dependency ratios. Solely the effects for a PAYG pension system will be analyzed.
6.1 Hypothesis
The hypothesis is formulated in three parts.
I: China’s national dependency ratio increases over time.
II: Dependency ratios are larger in rural states than in urban states. III: Dependency ratios increase faster in rural states than in urban states.
6.3 Methods & Results
For all three hypotheses I use data provided by the National Bureau of Statistics of China (NBS). Every year the NBS publishes a China Statistical Yearbook with statistics. For my analysis I will look at population data. More specifically I will look at national and state specific dependency ratios as well as urbanization rates for various years. National dependency ratios from China’s Statistical Yearbook 2014 are plotted in figure 2 for a time interval of 1982-2013. One might observe that the national dependency ratio was at 8.0% in 1982 and rose to 13.1% by 2013, an increase of 5.1%. In other words, in 1982 there were nearly 12 workers to one aged person, while in 2013 there were only 7
workers to one elderly person. The suggested rise in national dependency ratios in recent decades by the National Bureau of Statistics reflects the first hypothesis. Hence,
hypothesis 1 is justified.
Figure 2. National dependency ratio 1982-2013. China’s national dependency ratio rose
from 8% in 1982 to 13.1% in 2013. Data obtained from the China Statistical Yearbook
2014 National Bureau of Statistics.
7,5 8,5 9,5 10,5 11,5 12,5 13,5 1982 1991 1994 1997 2000 2003 2006 2009 2012 Dependency ratio (%) Year
To test hypothesis 2 I plotted provincial urbanization percentages together with the corresponding state specific dependency ratios. I have done this for all Chinese
provinces for a time interval of 9 years from 2005 till 2013. For simplicity I assume that a province is ‘urban’ when the urbanization rate of the respective province is 50% and above. Provinces with an urbanization rate below 50% are considered to be ‘rural’. The results for 2005 are shown in figure 3 and the graphs for 2006-2013 are presented in the appendix. In the year 2005 the provinces Jiangsu, Jilin, Heilongjiang, Zhejiang, Liaoning, Guangdong, Tianjin, Beijing, and Shanghai are assumed to be urban states. Hence, the other 22 provinces are considered to be rural states. In order for the second hypothesis to be valid all 9 urban states should have lower dependency ratios than the other 22 rural states, yielding a decreasing line. However, the 2005 graph shows nearly a horizontal line. The latter indicates that rural and urban provinces yield similar dependency ratios, thus the plotted dependency ratios for 2005 do not reflect the second hypothesis. The other 8 years presented in the appendix yield similar graphs. Hence, the second hypothesis is not reflected in the data; therefore hypothesis 2 is not justified.
Figure 3. Dependency ratio & provincial urbanization (urbanization in rising order) 2005.
In order to test the third hypothesis I have calculated the average urbanization percentage per state between 2004 and 2013. The lowest urban percentages are found in Tibet, Guizhou and Yunnan – these states are considered to be an example of rural states. Highest urban percentages are found in Shanghai, Beijing and Tianjin – these states are considered to be an example of urban states. However, Tibet and Shanghai are particular states in terms of age pyramids. Of all Chinese provinces dependency ratios are lowest in Tibet. A possible explanation for this might be the large influx of young Chinese Han migrants into Tibet stimulated by the Chinese government (Wong, 2010). Shanghai has the highest dependency ratios of all Chinese states. This result may be explained by Shanghai’s extreme low fertility rates. One of the explanations might be that Shanghai women choose more often for their career over a family. Besides, costs to raise children as well as accommodation costs for a larger family are high. Hence, couples living in Shanghai more often choose not to have children or to have one child, even though some couples are allowed to have more than one child (J.P., 2013). In order to avoid distorted results, I decided to exclude Shanghai and Tibet from my data analysis. I have plotted the dependency ratios of Guizhou province – being an example for rural states together with Beijing province – as an example for urban states for a time period of ten years (2004-2013). The resulting graph is presented in figure 4. The plotted dependency ratios seem to reflect the third hypothesis. Although, there is not a dramatic shift, rural dependency ratios do seem to increase faster than urban dependency ratios. Hence, hypothesis 3 is justified.
Figure 4. Development of urban & rural dependency ratios 2004-2013. Data obtained
from the China Statistical Yearbook 2014 National Bureau of Statistics.
6.4 Discussion
The suggested rise in the national dependency ratio in recent decades by the National Bureau of Statistics is in line with what is widely agreed on in the available academic literature. The reason for such increase in the national dependency ratio is China’s demographic transition towards an ageing society; low fertility rates together with increasing longevity (Chen & Liu, 2009). The increase of China’s national dependency ratio is expected to be largest between 2010 and 2035. This can be explained by the fact that the baby boomers from the 1950s are all about to retire. Hence, the proportion of elderly to the working population will sharply increase (Peng, 2011). Simulation results generated by Wang, Xu, Wang, & Zhai (2004) suggest that the national dependency ratio is expected to further increase to 25% in 2030 and will be at 39% by 2050. Considering a PAYG-system, a rise in the national dependency ratio will negatively affect the
accumulation of pension finances. As a result, pension benefits will decrease or the financial burden on the working population will increase as a result of higher contribution
by Wang et al. and a replacement rate of 0.6 the contribution rate should be 15% in 2030 and 23.4% in 2050 in order to maintain financial sustainability of the PAYG system. Hence, China’s ageing problem asserts a serious challenge on the pension system.
The published dependency ratios by the NBS do not reflect the second hypothesis. State specific dependency ratios do very, however they do not necessarily yield higher ratios for rural states compared to urban states. This suggests that while China’s national dependency ratio increases over time (validity of hypothesis 1), there are no significant demographic differences between urban and rural regions. This result gives the
impression that China’s pension problem is solely driven by its growing ageing society and not by rural-to-urban migration of the young ones. However, the previous is different from what is expected and widely agreed on in the academic literature. Due to large rural-to-urban migration of the young ones, rural dependency ratios are expected to be larger than urban ratios. The following statement supports the previous: “Not only will old-age dependency ratios rise rapidly, but this trend is likely to be much more pronounced in rural areas (consistent with anecdotal concerns about the “hollowing out” of villages because of migration)” (Cai et al., 2012, p.11). Hence, in reality rural-to-urban migration results in a larger pension challenge in rural regions than in urban regions.
Furthermore, dependency ratios provided by the NBS do seem to reflect the third hypothesis. Rural dependency ratios increase faster than urban dependency ratios, however they do not show a dramatic shift between 2004-2013. The obtained results are in agreement with those attained by Wang stating, “Aging is becoming far more
pronounced in rural than in urban areas, due to the demographic transition and continued movement of young adults into cities” (The World Bank, 2012). Similar, Dorfman et al. (2013) emphasize that the gap between rural and urban dependency ratios will widen in future decades. To sum up, according to dependency ratios provided by the NBS,
hypotheses one and three are justified, while the second hypothesis is proven invalid. The following part will elaborate the most evident explanation for why my results do not reflect hypothesis 2, while the academic literature widely supports this.
A most evident explanation for why the data does not reflect the second hypothesis is that the NBS might have calculated the dependency ratios based on
based on where people are registered instead of where they actually live. The household registration system, also referred to as the ‘hukou’ system, dates from the 1950s (Chan & Zhang, 1999). The system works as follow: every person is registered at his/her place of birth. In general there are two types of ‘hukou’ residents: urban and rural. Furthermore, public and social services are strictly linked to someone’s place of registration. Hence, rural ‘hukou’ workers living in urban suburbs are not eligible to enjoy the same social benefits as urban residents (Wong, Fu, Li, & Song, 2007). Although, the disadvantages of the ‘hukou’ system have been a huge obstacle to labor mobility, rural-to-urban migration has significantly increased in recent decades. In the past thirty years about thirty percent of China’s total population migrated from rural to urban regions (Peng, 2011). Most of these migrants are young rural workers searching for better job opportunities in the cities. If this large number of young rural migrants would be added to the urban working-age population, dependency ratios will look different. Urban dependency ratios will be smaller and rural dependency ratios larger, reflecting the second hypothesis. The following chapter presents a simulation in which I will analyze the effect of large rural-to-urban migration on rural and urban age pyramids followed by the influence on the sustainability of a PAYG system.
Chapter 7: Simulation
As presumed in the previous chapter, the NBS has calculated the dependency ratios based on household registration, not taking into account non-registered migrants residing in urban areas. However the number of migrants moving from the countryside to the cities has drastically increased in recent decades, altering urban and rural age pyramids (Peng, 2011). Therefore, by means of a simulation, I will correct for non-registered migrants in the calculation of urban and rural dependency ratios. Subsequently, I will analyze the effect of changing dependency ratios on the sustainability of a PAYG system. By doing so I venture an attempt to give a better picture of reality and so of the actual scale of China’s pension problem.
7.1 Method & Results
According to annual national sample surveys of population, about thirty years ago nearly 80% of China’s total population lived in rural regions. Nowadays, a bit less than half of China’s total population remains in rural areas (National Bureau of Statistics, 2014). Hence, during the last three decades about 30 % of China’s total population moved from rural to urban areas. The latter is equal to 37.5% of China’s rural population that moved to the cities. Consistent with the academic literature, I assume that all migrants originate from the rural working-age group (those aged between 15 and 64).
To realize the simulation, regional population numbers published by the NBS from the year 2013 are used. In order to get urban and rural population numbers I have taken the age composition of population by region and sorted the provinces on
urbanization in rising order. For simplicity I assume that states with an urbanization rate lower than 50% are rural states. Provinces with an urbanization rate of 50% and above are assumed to be urban states. Subsequently, I have summed up total population
numbers, working-age population numbers (Nw), and old-aged population numbers (Np) for the rural states as well as for urban states.
First, I calculate urban and rural dependency ratios for the year 2013 based on household registration. Table 6 presents the urban and rural population numbers as well as the corresponding dependency ratios, which are 12.73% and 13.57% respectively. One might notice that there is not a large difference between urban and rural dependency ratios. Subsequently, I move 37.5% of China’s rural population to urban areas, hence correcting for non-registered migrants residing in urban regions. The 37.5% of China’s rural population will be taken from the rural working-age group and added to the urban working-age group. As a consequence the urban working-age population increases by 40.0%, while the rural working-age population declines by 52.7% (see table 7). The new urban and rural population numbers are presented in table 6. The new urban and rural dependency ratios yield 9.10% and 28.69% respectively (see table 6). Hence, including non-registered migrants residing in urban regions result in a larger urban-rural
dependency ratio disparity, showing a dependency ratio gap of nearly 20% (28.69% - 9.10%). The latter illustrates significant demographic differences between urban and rural
regions, showing a much higher proportion of elderly remaining in rural areas than in urban areas, hence reflecting hypothesis 2 from chapter 6.
To give a clear overview of the simulation results, table 7 presents the percentage change in urban and rural working-age population numbers as well as the percentage change in urban and rural dependency ratios. The next section will analyze the effects of the changed urban and rural dependency ratios on system solvency of a PAYG pension system.
Table 6.
Old (based on household registration) and new (including non-registered migrants) age compositions 2013 (1,000 persons)
Old New
Total Nw Np Np/Nw (%) Total Nw Np Np/Nw (%)
Urban 617,146 469,943 59,843 12.7 805,129 657,926 59,843 9.1
Rural 501,289 356,731 48,416 13.6 313,306 168,748 48,416 28.7
Note. . a Total stands for the total number of people (x1,000 persons) living in rural or urban areas. b Nw is the working-age population, those aged between 15 and 64. c Np is the old-age population, those aged 65 and above. d Np/Nw is the dependency ratio in percentage. 37.5% of 501,289 is 187,983 and moves from rural to urban. 187,983 is subtracted from the rural working-age population group and added to the urban working-age population group.
Sources. China Statistical Yearbook 2014 National Bureau of Statistics, author’s calculation
Table 7.
Percentage change in working-age population and dependency ratios (based on table 6)
∆Nw (%) ∆Np/Nw (%)
Urban 40.0 -28.3
Rural -52.7 111.0
Note. a ∆Nw is the percentage change in the working-age population, those aged between 15 and 64. b ∆Np/Nw is the percentage change in the dependency ratio.
The 40.0% increase in the urban working-age population is obtained as follows (187,983/469,943) (table 6) and the 52.7% decline in the rural working-age population is calculated as follows (187,983/356,731) (table 6)
7.2 PAYG system solvency
Considering the PAYG revenue and expenses formula, the dependency ratio has influence on the contribution rate and the replacement rate. The PAYG formula is as follows:
× × = × × (1)
= × (2)
is the number of workers. is the contribution rate, which workers need to pay in order to participate in the pension scheme as a percentage of the local average wage .
is the number of pensioners who will receive a pension benefit equal to the
replacement rate multiplied by the local average wage . For system solvency pension revenues should be equal to pension expenses.
To analyze the effect of the 52.7% shift of rural young workers to urban areas on the sustainability of a PAYG system it is worth looking at the change in urban and rural age pyramids and thus at the change in dependency ratios. As shown in table 7, the 52.7% influx of young rural workers in urban regions increases the urban working-age population by 40.0%. As a consequence the urban dependency ratio decreases by 28.3% ((9.1-12.7)/12.7) see table 8. As a result, there is a surplus in pension revenues in urban areas; meaning revenues exceed pension expenses. Contrarily, the departure of young workers in rural areas has resulted in a 52.7% decline of the rural working-age
population. As a consequence the rural dependency ratio increases by 111.0% ((28.7-13.6)/13.6) see table 8. Hence, pension expenses exceed the revenues, meaning that there is a deficit in pension revenues in rural areas. In order to maintain system solvency, these changes in dependency ratios require modified contribution rates or modified
replacement rates, hence resulting in a different pension benefit. The contribution rate is defined as in equation (2) the replacement rate multiplied by the dependency ratio
Table 8.
Required contribution and replacement rates for PAYG system solvency
∆Np/Nw (%) ∆ (%) ∆ (%)
Urban -28.3 -28.3 40.0
Rural 111.0 111.0 -52.7
Note. a ∆Np/Nw is the percentage change in the dependency ratio. b ∆ γ is the required percentage change in
the contribution rate for system solvency while holding the replacement rate τ constant (using equation (2)).
c
∆ τ is the required percentage change in the replacement rate for system solvency while holding the contribution rate constant (using equation (1)).
Sources. Table 7, author’s calculation
Table 8 shows the required change in urban and rural contribution rates for system solvency while holding the replacement rate constant. The same table also shows the required change in urban and rural replacement rates while holding the contribution rate constant. For urban areas, a decrease in the dependency ratio of 28.3% means that more workers can take care of each pensioner, hence diminishing the burden on employees and employers. As a result contribution rates can be lowered by 28.3% (holding the
replacement rate constant) such that higher coverage of urban employees can be
achieved. Technically another option in order to maintain system solvency would be to increase the replacement rate by 40.0% (holding the contribution rate constant).
However, this seems unlikely since current replacement rates are already high compared to developed countries (Dorfman et al., 2013).
For rural areas, the departure of young workers results in an increase in the dependency ratio of 111.0%. This implies that each worker needs to take care of a much larger number of pensioners, hence dramatically increasing the burden on the rural working-age population. To maintain financial stability of the PAYG system, the
contribution rate should increase by 111.0% (holding the replacement rate constant). As a consequence the burden on employees and employers will drastically increase, with ‘take-home-pay’ becoming much smaller and labor costs higher, resulting in less employment (James, 2002). Technically another option in order to maintain system solvency would be to lower the replacement rate by 52.7% (holding the contribution rate
constant). However, rural pension benefits are already low, with most retirees only receiving the basic pension of 55 RMB per month, which is nearly equal to China’s poverty level (Shen & Williamson, 2010). Hence, lowering the rural replacement rate is not a feasible possibility.
To sum up, urban residents benefit from the inflow of young rural ‘hukou’ residents resulting in a growing pension surplus, while rural workers are largely disadvantaged with an increasing pension deficit. The previous raises concerns about growing urban and rural disparities, with increasing pension pressures, illustrating China’s growing pension problem.
Chapter 8: Conclusion
China’s demographic transition is in crisis, which has consequences for the sustainability of China’s pension system. The relationship between demographic changes and the accumulation of pension funds is represented by the dependency ratio. The effect of demographic transition and rural-to-urban migration on dependency ratios has been examined. Dependency ratios provided by the NBS support the expectation that China’s national dependency ratio increases over time. A reason is China’s declining working-age population and growing grey population. However, NBS dependency ratios do not reflect the hypothesis of higher dependency ratios in rural areas compared to urban areas. The most evident explanation is the omission of rural ‘hukou’ migrants in the calculation of urban and rural dependency ratios. The simulation has demonstrated that by including non-registered migrants in the calculation of dependency ratios, the urban-rural gap becomes more apparent. As a result rural dependency ratios increase significantly and urban dependency ratios decline, illustrating greater urban-rural demographic disparities. Subsequently, the effect of changing dependency ratios on system solvency of a PAYG system is analyzed. The large increase in the rural dependency ratio has a negative effect on the accumulation of pension funds. To maintain system solvency, higher
contribution rates or lower replacement rates are required. This is problematic since the working-age population and the elderly population will be disadvantaged. The drop in urban dependency ratios seems to be beneficial for urban employees and enterprises since
contribution rates can be lowered, hence resulting in larger coverage. However, it is important to bear in mind that urban residents only benefit from the influx of young workers if the rural ‘hukou’ migrants participate in the urban pension scheme. In practice most migrants are not taking part in any pension scheme. The challenge to maintain system sustainability becomes greater as China’s ageing society increases and rural-to-urban migration continues.
It is of major importance that the Chinese government adjusts its pension system geared towards demographic changes, taking into consideration the process of internal migration. Solutions for China’s pension problem is an on going debate in the academic literature, however several studies suggest a national pension system covering all Chinese residents. The idea is that a unified pension scheme can lift rural pension deficits with the pension surplus of urban regions. An advantage of a unified pension system is the
avoidance of extremely large contribution rates in rural areas. However, unifying urban and rural pension schemes will exercise pressure on the political system, since urban residents will have to subsidy the pension deficit of rural areas. As a result, urban residents might not want to participate in the pension system anymore, giving rise to political instability. Therefore, it would be interesting to study to what extent a unified pension system would be feasible considering China’s growing ageing society and increasing rural-to-urban migration.
Research limitation: to study the validity of hypothesis two and three in chapter six a time space of eight and nine years respectively is used. However, with a small sample size of only nine or ten years, caution must be applied, as the findings might not be up for generalization. Moreover, a short time space cannot really capture changes. Hence, one should be cautious with drawing a general conclusion. Further research should try to include data of a larger time space.
Chapter 9: Appendix
Figure 5. Dependency ratio & provincial urbanization (urbanization in rising order) 2006.
Data obtained from China Statistical Yearbook 2007 National Bureau of Statistics.
Figure 6. Dependency ratio & provincial urbanization (urbanization in rising order) 2007.
Figure 7. Dependency ratio & provincial urbanization (urbanization in rising order) 2008.
Data obtained from China Statistical Yearbook 2009 National Bureau of Statistics.
Figure 9. Dependency ratio & provincial urbanization (urbanization in rising order) 2010.
Data obtained from China Statistical Yearbook 2011 National Bureau of Statistics.
Figure 10. Dependency ratio & provincial urbanization (urbanization in rising order)
Figure 11. Dependency ratio & provincial urbanization (urbanization in rising order)
2012. Data obtained from China Statistical Yearbook 2013 National Bureau of Statistics.
Figure 12. Dependency ratio & provincial urbanization (urbanization in rising order)
References
Cai, F., Giles, J., O'Keefe, P., & Wang, D. (2012). The elderly and old age support in
Rural China. World Bank Publications.
Chan, K. W., & Zhang, L. (1999). The hukou system and rural-urban migration in China: Processes and changes. The China Quarterly, 160, 818-855.
Chen, F., & Liu, G. (2009). Population aging in China. In International handbook of
population aging (pp. 157-172). Springer Netherlands.
Disney, R. (2000). Declining public pensions in an era of demographic ageing: Will private provision fill the gap?. European Economic Review, 44(4), 957-973. Dorfman, M. C., Holzmann, R., O'Keefe, P., Wang, D., Sin, Y., & Hinz, R.
(2013). China's pension system: A vision. World Bank Publications.
Dunaway, S. V., & Arora, V. B. (2007). Pension reform in China: the need for a new
approach. International Monetary Fund.
Fan, R. (2006). Confucian filial piety and long term care for aged parents. In HEC
Forum (Vol. 18, No. 1, pp. 1-17). Springer Netherlands.
Frazier, M. W. (2004). After pension reform: navigating the “third rail” in China. Studies
in Comparative International Development, 39(2), 45-70.
Giles, J., Wang, D., & Zhao, C. (2010). Can China’s rural elderly count on support from adult children? Implications of rural-to-urban migration. Journal of Population
Ageing, 3(3-4), 183-204.
Hesketh, T., Lu, L., & Xing, Z. W. (2005). The effect of China's one-child family policy after 25 years. New England Journal of Medicine, 353(11), 1171-1176.
Hu, N., & Yang, Y. (2012). The real old-age dependency ratio and the inadequacy of public pension finance in China. Journal of Population Ageing,5(3), 193-209. James, E. (2002). How can China solve its old-age security problem? The interaction between pension, state enterprise and financial market reform.Journal Of
Pensions Economics & Finance, 1(01), 53-75.
J.P. (2013). Ladies from Shanghai. The Economist. Retrieved from
http://www.economist.com/blogs/feastandfamine/2013/03/chinas-one-child policy
Kinsella, K. G., & Wan, H. E. (2009). An aging world: 2008 (p. 37). US Department of Commerce, Economics and Statistics Administration, US Census Bureau. Leers, T., Meijdam, L., & Verbon, H. A. (2004). Ageing, migration and endogenous public pensions. Journal of Public Economics, 88(1), 131-159.
Li, B. (2007). Pension reform in China: who are left out? (pp. 29-48). NOVA Science Publishers.
Lu, J. (2010). Jinrong weiji xia shehui baozhang de zhengce xuanze [Policy Selection of Social Security in Financial Crisis]. Jingjishi [China Economist], (007), 9–11. Mason, A., & Kinugasa, T. (2008). East Asian economic development: two demographic dividends. Journal of Asian Economics, 19(5), 389-399.
Mitchell, S. L. (2013). China's Pension Crisis: System Reforms Start with Relaxation of One-Child Policy. Frontiers L. China, 8, 233.
National Bureau of Statistics. (2010). Population Census 2010. Retrieved June 8, 2015, from http://www.stats.gov.cn/english/Statisticaldata/CensusData/
National Bureau of Statistics. (2014). China Statistical Yearbook 2014. Retrieved June 8, 2015, from http://www.stats.gov.cn/tjsj/ndsj/2014/indexeh.htm.
Peng, X. (2011). China’s demographic history future challenges. Science,333(6042), 581 587.
Pozen, R. C. (2013). Tackling the Chinese pension system. Paulson Institute.
Salditt, F., Whiteford, P., & Adema, W. (2008). Pension reform in China. International
Social Security Review, 61(3), 47-71.
Shao, Y., & Xu, J. (2001). China's urban pension system: Reforms and problems. Cato
J., 21, 395.
Shen, C., & Williamson, J. B. (2010). China's new rural pension scheme: can it be improved?. International Journal of Sociology and Social Policy, 30(5/6), 239 250.
The World Bank. (2012). New Bank Report Looks at China’s Rural Elderly [Press release]. Retrieved from http://www.worldbank.org/en/news/press
release/2012/04/09/new-bank-report-looks-at-china-rural--elderly
Trinh, T. (2006). China’s Pension System: Caught between mounting legacies and unfavourable demographics. Deutsche Bank Research, February 17.
Wang, Y., Xu, D., Wang, Z., & Zhai, F. (2004). Options and impact of China's pension reform: a computable general equilibrium analysis. Journal of Comparative
Economics, 32(1), 105-127.
Willmore, L. (2004). Population ageing and pay-as-you-go pensions. Ageing Horizons, 1, 23-29.
Woetzel, J., Mendonca, L., Devan, J., Negri, S., Hu, Y., Jordan, L., ... & Yu, F. (2009). Preparing for china’s urban billion. McKinsey Global Institute, March.
Wong, E. (2010). China’s money and migrants pour into Tibet. The New York
Times. Retrieved from
http://www.nytimes.com/2010/07/25/world/asia/25tibet.html
Wong, K., Fu, D., Li, C. Y., & Song, H. X. (2007). Rural migrant workers in urban China: living a marginalised life. International Journal of Social Welfare,16(1), 32-40.
Wu, L. (2013). Inequality of pension arrangements among different segments of the labor force in China. Journal of aging & social policy, 25(2), 181-196.
Yuan, Z., & Feng, J. (2005). Rethinking pension reform in China. China and the World
Economy, 13(2), 20-32.
Zhang, L. (2008). Conceptualizing China's urbanization under reforms. Habitat
International, 32(4), 452-470.
Zhang, W. & Tang, D. (2008). The New Rural Social Pension Insurance Programme of Baoji City, HelpAge International-Asia/Pacific, Chiang Mai.
