Economics of Rural–Urban Migration
Abstract and Keywords
In the past century, many developing countries have experienced rapid economic development, which is usually associated with a process of structural transformation and urbanization. Rural–urban migration, shifting the labor force from less productive agricultural sectors to more productive industrial sectors in cities, plays an important role in the growth process and thus has drawn economists’ attention. For instance, it is recognized that one of the important sources of China’s growth miracle is rural–urban migration.
At the early stage of economic development, an economy usually relies on labor-intensive industries for growth. Rural–urban migrants thus provide the necessary labor force to urban production. Since they are more productive in industrial sectors than in agricultural sectors, aggregate output increases and economic growth accelerates. In addition, abundant migrants affect the rates of return to capital by changing the capital–labor ratio. They also change the skill composition of the urban labor force and hence the relative wage of skilled to unskilled workers. Therefore, rural–urban migration has wide impacts on growth and income distribution of the macroeconomy.
What are the forces that drive rural–urban migration? It is well understood that cities attract rural migrants because of better job opportunities, better career prospects, and higher wages. Moreover, enjoying better social benefits such as better medical care in cities is another pull factor that initiates rural–urban migration. Finally, agricultural land scarcity in the countryside plays an important role on the push side for moving labor to cities.
The aforementioned driving forces of rural–urban migration are work-based. However, rural–urban migration could be education-based, which is rarely discussed in the literature. In the past decade, it has been proposed that cities are the places for accumulating human capital in work. It is also well established that most of the high-quality education institutions (including universities and specialized schools for art and music) are located in urban areas. A youth may first move to the city to attend college and then stay there for work after graduation. From this point of view, work-based migration does not paint the whole picture of rural–urban migration. In this article, we propose a balanced view that both the work-based and education-based channels are important to rural–urban migration. The migration story could be misleading if any of them is ignored.
It is well recognized that the economic development of a country proceeds in stages. From the macroeconomic perspective, these stages are known as structural changes or transformation, which are characterized by resources reallocation from less productive to more productive sectors (for a macro-oriented survey on the topic, see Herrendorf, Rogerson, & Valentinyi, 2014). One of the most important structural changes in developing countries is the shift of labor input from the agricultural sector into the manufacturing and/or services sectors. Since the agricultural sector is mainly located in the rural area of a country, whereas the manufacturing and services sectors are concentrated in cities, rural–urban migration is regarded as the main momentum of economic development in the literature.1 As pointed out by Bosworth and Collins (2008), an important sustaining factor for future economic growth in China and India is their large share of the labor force in agriculture, which can support continuing rural–urban migration. This review focuses on the recent development of the topic.
As summarized in the survey by Williamson (1988) on urbanization and growth, traditional explanations of internal migration and urbanization are classified into two categories—rural push and urban pull. The rural push hypothesis is that unusually rapid population growth lowers rural labor productivity and pushes landless labor into the cities, whereas the urban pull hypothesis focuses on economic forces pulling migrants into the cities. The rural push hypothesis reflects the idea of surplus labor in Lewis (1954) and leads to the prediction that rural–urban migration raises unemployment in cities in the Todaro (1969) model. One unique feature of these dual-sector models is that arbitrage in terms of labor mobility is restricted so that rural migrants are paid less than the marginal product of labor in the formal urban sector. The modeling of migration decision in Todaro (1969) is the same as that in Sjaastad (1962). Sjaastad takes migration as an investment in human capital. However, it is regarded as the human capital approach under the disequilibrium perspective of the migration literature. It is “disequilibrium” because migration is driven by the rural–urban wage gap which is not labor market-clearing. The only difference is that it is expected income that matters in Todaro, highlighting the fact that rural–urban migration is an investment in job search in light of the rural–urban wage gap. The expected income is a product of urban wage and the probability of employment, which is identified as a random job lottery in Harris and Todaro (1970).
While the canonical Lewis–Todaro literature of internal migration emphasizes the imperfection of labor markets and the role of the rural–urban wage gap (Greenwood, 1997; Lucas, 1997), there is another strand of the literature that focuses on country-specific factors such as financial constraints, risk, and transaction costs. The latter literature explains rural–urban migration in terms of location-specific factors instead of the rural–urban wage gap (Stark, 1991). Internal migration is an optimal outcome of the resource-allocation problem in a specific environment that varies across countries and over time. Although urban job risk is important, as emphasized in the Todaro and Harris–Todaro migration models, rural risks are equally important when we discuss internal migration in less developed countries. As a result, migration models of family strategies that link migration with factors such as risk behavior of households, fertility, and education are potentially promising.
Although the excellent survey conducted by Lucas (1997) was completed in the late 1990s, there have been few important additions to the work on rural–urban migration. Foster and Rosenzweig (2007) provide an explanation based primarily on the limitation of data availability. As pointed out in their review:
There is growing evidence that internal migration in developing countries is increasing and this pattern has raised new concerns about consequences for both sending and receiving areas of this process. This growth is particularly evident in two of the world's largest and fastest growing economies, India and China largely as a consequence of reductions in economic and residential restrictions. (p. 3058)
For India, and sometimes Bangladesh and Pakistan, Foster and Rosenzweig (2007) summarize their findings from data sets within an integrated framework so as to provide the requisite information of sectoral migration and economic development. However, Foster and Rosenzweig highlight several important limitations of the existing literature. First, the setup is mainly static and deterministic. Second, they should pay attention to the opportunity costs of migrants in terms of what they have to give up in rural areas for their migration decisions. Third, the rural–urban wage gap, which is the main factor in the Lewis–Todaro analysis, is not the only determinant of rural–urban migration. Finally, migration decisions of individuals and households should be differentiated. As we will discuss later in this article, recent studies on rural–urban migration are moving in these directions.
Recent works on rural–urban migration have echoed the suggestions of Foster and Rosenzweig (2007). For instance, by constructing a data set on inequality and migration based on the 170 Demographic and Health Surveys (DHS) for 65 countries, Young (2013) finds that the rural–urban gap in living standard is able to account for 40% of the mean inequality and much of the cross-country variation in inequality levels. In addition, given the detailed characteristics of migrants in the DHS data, their skill levels can be identified from their educational attainments in the regions. As a result, Young proposes that residence probabilities are good proxies for the sectoral skill intensity of production. He finds that once the residence probabilities of workers with different education are controlled so that the skill intensity is the same for the regions, the rural–urban gap disappears. Young then concludes that rural–urban migration is an outcome of individual choices based on one’s education and skill, an insight taken from Lagakos and Waugh (2011).
Another interesting example of highlighting factors beyond the rural–urban wage gap that affect migration is Cattaneo and Peri (2016). Following the literature on climate and migration (e.g., Feng et al., 2010; Munshi, 2003), Cattaneo and Peri (2016) examine the effects of global warming on rural–urban migration. According to the Roy–Borjas model (Borjas, 1987; Roy, 1951), global warming affects agricultural productivity so that in response, resource reallocation takes place across sectors. By examining the climate and migration data of 115 countries from 1960 to 2000, from Dell et al. (2012) and Ozden et al. (2011) respectively, it is found that migration patterns in response to climate changes depend crucially on income categories of the countries. For poor countries whose per capita income is below US$1,500, temperature increases do not raise rural–urban migration. On the contrary, global warming causes an increase in rural–urban migration for middle-income countries. This highlights the fact that rural–urban migration is not a uniform global phenomenon and its patterns depend on the specific characteristics of the countries.
The fact that effects of rural–urban migration are country-specific explains why the majority of the latest literature on rural–urban migration are country studies. Kaplan and Schulhofer-Wohl (2017) account for the decline of interstate migration in the United States during the period 1991–2011. Kaplan and Schulhofer-Wohl (2017) develop a two-location, two-occupation model where agents with occupation-specific skills choose locations and conduct job searches in order to maximize their expected discounted lifetime utility. Utility depends on income and amenities, where the latter are non-labor market features of locations. Thus, location is an experience good, so that one has to stay in a location to learn the amenities there. By calibrating the model, it is found that half of the decline in gross migration between 1991 and 2011 was caused by the fact that jobs across the nation have become more similar in the sense that returns to occupation-specific skills are falling over time. The remaining decrease of the gross migration rate of the period can be accounted for by the increase in information in the macroeconomy about location amenities so that the experience nature of the location matters less.
Chernina, Dower, and Markevich (2014) investigate the effect of property right and land transferability on internal migration, with a focus on the finance of the migration costs. They take on the Stolypin reform of land titling in 1906 and examine its effects on the migration of peasants from Europe to the Russian Empire from 1901 to 1914. The reform allowed the ownership of land to be transferred from the repartition commune to individual peasants. By constructing a panel of provincial data and performing a difference-in-differences analysis, Chernina, Dower, and Markevich conclude that an increase in land liquidity caused by the Stolypin reform led to an increase in internal migration. Finally, to understand why the long-lasting huge rural–urban wage gap (above 25% adjusted for cost-of-living differences) does not lead to large rural–urban migration flows of male workers in India, Munshi and Rosenzweig (2016) focus on the cultural and institutional background of the rural caste. Specifically, they interpret the close relations of people within the caste as an informal insurance network. As a result, the migration of a villager implies that he/she will lose the insurance of the caste. Although group migration of the caste (Munshi, 2011) and temporary migration (Morten, 2016) can serve as a solution to the information problem, Munshi and Rosenzweig argue that the big wage gap and the low permanent rural–urban migration is the outcome of spatial misallocation generated by the informal insurance of the castes. Interestingly, counterfactuals show that if formal insurance can be allowed so that risk-sharing for households is improved by 50%, the rate of rural–urban migration will be doubled.
Since the mechanism of rural–urban migration at work for an economy is specific to the institutional background of a country, it is conventional to discipline the underlying model to the economy of particular interest. As Foster and Rosenzweig (2007) emphasize, the two most important countries on which the literature has focused are China and India. Given that they have already reviewed the literature for India, the representative work on rural–urban migration and economic growth of China will be reviewed in this article. Our plan is as follows. First, we first provide some institutional background about China in terms of the data from various sources. We then lay down the basic framework of our discussion of the literature. Specifically, we base it on the driving forces of rural–urban migration to separate the literature into work-based and education-based categories. As the existing literature on migration and growth in China is huge and there already exist some excellent surveys for the first decade of the 21st century, we mainly focus on work done since 2008.
Since the 1990s, there has been rapid growth in rural–urban migration in China, with the number of rural migrants increasing by more than 100 million to 145 million (Meng, 2012).2 Since it is well understood that rural–urban migration in China is closely related to the hukou system, we begin with a brief description of it.
The Hukou System
China implemented the unique hukou regulation system in 1958 in order to solve the serious problem of rural labor migration to cities in the early 1950s. The power of the hukou system in regulating migration came not only from the system itself, but also from its integration with social and economic controls on, for example, food rations, job distribution, education, and social benefits. Therefore, it was difficult to stay outside the place of one’s hukou registration before China’s economic reform.
A citizen’s hukou was classified in two parts: hukou suozaidi (the place of hukou registration) and hukou leibie (the type or status of hukou registration). Hukou suozaidi was a person’s presumed regular residence. Everyone was required to register in one and only one place of residence. Hukou leibie referred to “agricultural” and “non-agricultural” hukou. Hukou registration place and status mattered because it determined, for example, access to good jobs, grain rations, education for children, health benefits, and the opportunity to purchase a house. A citizen with “non-agricultural” hukou status would lose the right of land rental and the right of inheriting the land that his/her parents rented.
Changing from agriculture to non-agriculture was commonly known as nongzhuanfei. The regular channels of nongzhuanfei included recruitment by a state-owned enterprise (zhaogong), enrollment in an institution of higher education (zhaosheng), and promotion to a senior administrative job (zhaogan). A formal rural–urban migration involved both a change in hukou registration place and in entitlement status. To complete the official migration process, a citizen had to satisfy both the requirements and obtain a quota, which was controlled by the central government.
Beginning in the early 1980s, the growing market-oriented economy demanded more cheap labor. It became easier for people to survive outside their hukou registration place. The pull factor attracted migrants from rural to urban areas. The increase in mobility also forced the government to relax its hukou policies, introducing, for example, a new type of urban hukou with “self-supplied food grain,” temporary residence certificates, and the “blue-stamp” urban hukou. By 2005, some provinces had begun or would soon begin to implement a unified urban–rural household registration system, removing the distinctions between agricultural and non-agricultural hukou status. The ultimate aim in the future is to abolish the regulations of migration in the hukou system.
We now provide a discussion of two overview articles on migration and growth in China that start with data description.3 First, Chan (2013) explains China’s migration statistics, clarifies misinterpretations, and resolves problems in the key data. The chapter examines the trends and geography of migration, paying special attention to the role played by the hukou (household registration) system. After identifying different categories of migrants under the hukou system, Chan concludes that, in the previous 25 years, the majority of migrants were predominantly non-hukou migrants from the rural areas. Starting from the early 1980s, the hukou migration rate declined slightly, but overall remained quite stable relative to the size of the Chinese population. This reflects the strong government intervention in hukou migration. In addition, the non-hukou migrant population has been growing since the early 1980s, especially in the early 1990s; however, the data shows a slow increase in the number of migrants in the second half of the 1990s. Since the beginning of the 21st century, there has been a migrant labor shortage. For the geography of rural–urban migration, the 1995 survey reports a total of 33.23 million migrants crossing county-level units in the preceding five-year period. Seventy-two percent of the inter-county migration was within provinces, and the remaining 28% was inter-provincial migration. Since then, however, inter-provincial migration has also become more prevalent, accounting for only 32% of all inter-county moves in 1990–1995 and about 47% in 1995–2000. The trend is likely to have continued in the previous decade.
Meng (2013) studies the dynamic process of rural–urban migration based on the data of 2008–2012 from the Rural–Urban Migration in China (RUMiC). It characterizes the demographic composition of the migrants and the resulting labor market outcomes. The RUMiC is a longitudinal survey that follows migrants over time. It starts with a random sample of 5,000 migrant households from 15 cities in 2008. Because of the significant attrition rate each year, starting from 2009 the RUMiC city migrant survey has two sub-samples: the old sample traces part of the previous year sample and the new sample draws a new random sample. The former presents a dynamic picture of migrants’ life and work, whereas the latter gives a representative picture of migrants in general. Meng first explores the relation between rural–urban migration and China’s unskilled labor supply. This is closely related to the hukou system which restricts labor mobility between the rural and urban sectors so that the potential labor supply comes from the rural hukou population.
According to the age and gender distribution of the migrant and non-migrant labor force (individuals aged 16 to 65 who are employed or unemployed) in urban and rural areas from the RUMiC 2008 samples of Rural Household, Urban Household, and City Migrant surveys, Meng summarizes the following facts. First, the majority of migrant workers in cities are aged between 16 and 30, and they are much younger than their urban counterparts. Second, migrants only account for a small portion of the total rural hukou labor force. Third, more than half of the young rural hukou workers (aged 16 to 30) have migrated to cities outside of their home county, and the majority of the rural labor force between 16 and 64 years of age have not migrated. Fourth, since 2006, the proportion of newly arrived migrant workers who are aged between 16 and 25 years has increased year by year, from 57% in 2006 to 78% in 2010, whereas the proportion of those aged 36 and above has reduced year by year. Fifth, the average migrant wage and salary workers worked 60 hours per week in 2008, whereas the average was 58 hours in 2012. Sixth, the increase in monthly and hourly earnings for migrant wage-salary workers between 2008 and 2012 is 12% and 14% per annum, respectively. Seventh, migrants on average have 8.8 years of schooling, whereas their urban hukou counterparts have 10.3 years and the rural hukou non-migrated workforce has only 6.9 years. Finally, using urban household survey data, it is found that the rate of return to years of schooling is 7.2%, whereas the highest estimate to years of schooling for migrants is 4%. There is a large disparity between the quantity and quality of education between rural and urban areas.
What have we learned from these reviews? It is consistent with the conventional belief that rural–urban migration is mainly work-based. Rural workers migrate to cities in order to have better job opportunity and remuneration. The rural–urban wage gap is the driving force as emphasized in the Lewis–Todaro literature. However, there is a subtle implication from the data. First, the migrant population is getting younger over time. Second, migrants are less educated than their urban hukou counterparts but are better educated than their rural non-migrant counterparts. Rates of return are higher for education in cities than in rural areas. Obviously, there is a difference or gap between urban and rural education. The idea of the rural–urban schooling gap is not novel. For instance, from the latest household surveys of 70 countries prepared for the World Bank (e.g., country censuses, Living Standards Measurement Surveys, and Demographic and Health Surveys), Orazem and King (2007) provide a detailed characterization of the rural–urban schooling gaps. From the year 2000 or later, the urban–rural gap exceeds 10% in half the sampling countries, and by ages 15–17 the schooling gap exceeds 10% in three-quarters of the countries. They also point out that returns to schooling are higher in urban than in rural markets. But what is important and interesting is that the main motivation of rural–urban migration can be education-based; that is, agents are moving from rural areas to cities to obtain better education or training in order to achieve better work opportunities in life.
In the existing literature on internal migration, education-based models are not often seen. Glomm (1992) develops a model that combines the Todaro model of rural–urban migration and the neoclassical growth model to understand the phenomenon of continuous movement of workers from the rural sector to the urban sector as an equilibrium outcome. The economy consists of two locations or sectors where the rural (urban) location only produces agricultural (manufacturing) goods. Technologies depend on agents’ region-specific expertise and the total amount of assessed expertise in the region. It is assumed that the urban sector has a higher population density than the rural sector so that cities have an advantage in the externality of assessed expertise in production. Expertise in each region is accumulated through a simple linear learning-by-producing technology so that learning is a joint production as in Rosen (1972). Preferences are defined over the consumption of both goods, and the income elasticity of the agricultural good is assumed to be less than that of the manufacturing good and below unity. In such a setup, output growth is higher in the urban sector than in the rural one, whereas rural income falls relative to the urban counterpart. As a result, rural–urban migration is an outcome of the growing income differentials.
Another important piece of work related to education-based motivation is Lucas (2004), who constructs a two-sector model where the urban sector pays a higher wage due to its high-skilled jobs which are not available to rural workers. Unskilled rural agents migrate to cities to accumulate human capital and invest in skills. Households’ preferences are isoelastic and they supply one unit of time endowment to either work or invest in human capital. Rural production uses labor and land; it has diminishing returns because land is fixed in quantity. The urban production technology is linear and requires only skilled labor input. Human capital investment can only take place in cities and labor is free to move between sectors. Rural–urban migration takes place once and for all with the growth rate of production of the urban (rural) sector changes from zero (one) to a constant positive level (zero). If we allow for an externality in human capital accumulation as in Eaton and Eckstein (1997) so that it depends on the gap between one’s own level of human capital and the level of human capital of the leader, then different urban people will have different human capital levels. The equilibrium outcome is that only the leader produces and all others accumulate human capital in the urban sector. Numerical computations show that the pace of urbanization is gradual. Finally, Liao, Wang, Wang, and Yip (2017) build a dynamic intergenerational model where migration decisions are based on education. With human capital accumulation and intergenerational mobility, they examine the effects of both work- and education-based rural–urban migration on the economic development of China for the period 1980–2007.
Before closing the section, we present an overview of internal migration in China based on census data.4 We classify migrants into three categories based on their migration reasons: work-based, education-based, and others.5 Figure 1 provides the shares of work-based and education-based migration for the five years 1985, 2000, 2005, 2010, and 2015.
As shown in Figure 1, work-based migration always accounts for more than 40% of the total migration, whereas education-based migration accounts for more than 10% (except in 2005). This suggests that the role of education-based migration cannot be ignored in China’s rural–urban migration. Figure 2 further provides the details of the geography of migration at the provincial level.
Except in 1985, work-based migration is divided equally into the intra-provincial and inter-provincial categories. Migrants are equally likely to work within and outside their original provinces. However, the location choices of education-based migration are biased toward intra-provinces. The phenomenon that education-based migration is mainly intra-provincial is primarily because the total education costs are much lower within than outside the provinces. Finally, the trend of education-based migration has been rising significantly since 2005.6 This could be due to the expansion of college education since 1996. This also explains why research work on education and migration in China has been on the rise in the last decade.
Migration is an important force that causes demographic change. The “demographic dividend” (Asian Development Bank, 2011) describes how demographic change can have a profound and positive impact on economic growth: When a country experiences a decline in mortality and fertility rates, its share of the working-age population rises and hence per capita income growth increases. For China, estimates of the contribution of the demographic dividend to income growth for the three decades after the “reform and opening up” lie in the range of one-sixth to one-quarter (Bloom et al., 2010; Wei & Hao, 2010). Special attention is paid to the changing composition of employment, owing to the large-scale movement of surplus labor from rural to urban areas along the line of the Lewis–Todaro literature (Gong, Kong, Li, & Meng, 2008; Minami & Ma, 2010). Using national-level data, Bloom et al. (2010) estimate the impact of the sector change from rural–urban migration, which accounted for 16.0% of China’s per capita income growth over the period 1980–2000. Golley and Wei (2015) indicate that sectoral change accounted for 18.7% of per capita GDP growth over the period 1980–2010 based on “mechanical” growth accounting exercises. Following Kelley and Schmidt (2005), they use regressions instead to show that sector change has had a significant negative impact on both the industrial and service sector productivity growth, so that the standard “mechanical” growth decomposition analyses may have overstated its benefits.
China may have experienced history’s largest flow of internal migration in the world since the barriers on migration from rural to urban areas were gradually removed in the late 1970s and early 1980s. Rural people have been allowed to move freely between cities and their homes, but most of them are only circular migrants because the hukou system denies them permanent urban residency rights and associated social benefits. The number of permanent migrants remains rather stable, whereas temporary migrants have continued to move into cities since the 1980s (Chan, 2001; Chan & Buckingham, 2008). Although there is a growing body of literature (such as De Brauw & Rozelle, 2008; Hare, 1999; Lu & Song, 2006; Zhu, 2002) on China’s internal migration in recent years, most migration studies in China exclusively focus on temporary migration of China’s rural labor. Zhao (2007) provides a detailed literature survey of China’s temporary migration studies. The possible reasons for the lack of studies on the permanent migration decision made by rural migrants in China may include: (1) the number of permanent migrants from rural areas with the change in hukou status is few due to the strict constraints imposed by the hukou system; and (2) it is difficult to obtain survey data concerning permanent migrants. Hu, Xu, and Chen (2011) differentiate between temporary and permanent migration to study the permanent migration choice made by rural migrants using the China General Social Survey data. The China General Social Survey allows them to overcome the data problem because it is a unique, nationally representative data set that contains plenty of information on characteristics of rural migrants, their family background, and their hometown. Hu, Xu, and Chen (2011) show that estimating the permanent migration decision with the simple probit model can result in selection bias. Using the two-stage method proposed by Van de Ven and Van Praag (1981) to avoid selection bias, they find that more educated and more experienced migrants tend to be permanent urban residents, while those people with more children and more land at home are more likely to migrate circularly rather than permanently.
It is well understood that the hukou system plays an important role in controlling and restricting rural–urban migration in China. Since the start of the Chinese economic marketization in the late 1970s and early 1980s, hukou system reforms have been carried out in stages. Song (2014) presents a literature review on the evolution of the hukou system and its reforms. Pi and Zhang (2016) investigate how hukou system reforms affect the skilled–unskilled wage inequality through a general-equilibrium approach. Following Harris and Todaro (1970), they develop a multi-sector model and find that the effects of hukou system reforms on the skilled–unskilled wage inequality depend on the sectoral factor intensities. Specifically, no matter whether there is an urban informal sector, the wage inequality will be reduced if the urban skilled sector is more capital intensive than the urban unskilled sector when the government increases the strength of hukou system reforms. Using the Chinese General Social Survey data from 2010–2013, Pi and Zhang (2016) provide empirical evidence for their findings.
However, most of the existing studies explore the rural–urban work-based migration in China by adopting reduced form approaches or partial-equilibrium frameworks.7 Without a dynamic general-equilibrium setup, it is difficult to examine the overall impacts of rural–urban work-based migration on the Chinese economy, especially the influences of institutional factors on migration decisions and the interaction between institutional and macroeconomic factors. Being aware of this, Liao, Wang, Wang, and Yip (2016) construct a dynamic general-equilibrium model to study how institutional factors affect work-based rural to urban migration decisions in China.
Three institutional factors are considered in Liao, Wang, Wang, and Yip (2016). The first one is the hukou system and its continual reform. Since the 1980s, the growing market-oriented economy has demanded a large amount of cheap labor for urban production. It thus became easier for people to be privately employed in cities. Higher wages and better social benefits in cities were the pull factors that initiated rural–urban work-based migration and resulted in significant temporary migration.
The second institutional factor that affects rural–urban migration decision is the one-child policy. The principle of the one-child policy restricted each couple to have only one child. One-child families could obtain benefits and subsidies, including child allowance, priority access to schools, employment, health care, and housing. In rural areas, one-child families could even enjoy tax credits and receive more land. However, continual penalties on above-quota births would be applied if the one-child policy was violated. In addition to penalties, parents working in public sectors could be forced to quit their jobs. The one-child policy also interacted with the hukou system and had impacts on migration decisions. Permanent migrants had to follow the fertility policies of their destination cities with a transition period. For example, if the wife had not been pregnant when a couple obtained Beijing hukou, the certificate they received in the rural area for having a second child would be withdrawn. In contrast, the fertility of temporary rural–urban migrants was restricted by their original places. In reality, family planning officials had difficulty tracking temporary migrants. Thus, rural–urban couple migrants would take the stricter fertility regulation in cities into consideration when they were making permanent migration decisions.
The third institutional factor considered in Liao, Wang, Wang, and Yip (2016) is the land tenure arrangement system. Under the household responsibility system that was introduced in the early 1980s, each rural household had to produce a quota of goods and was compensated by the government. Above-quota production could be sold in the free market at unregulated prices. However, the land rights for rural households were incomplete. The land was officially owned by village collectives and distributed to households according to household size. Rural households only had the right to use the land for production activities with fixed-term contracts. After several years, village collectives had the right to reallocate land according to changes of household size. Besides, it was unclear whether households had the right to rent land or transfer their land contract. If land could not be rented or transferred, the lower productivity of land became an opportunity cost of rural–urban migration. Therefore, a risk of land expropriation in the next land reallocation and idle land became the pull factors on rural–urban migration decisions. To strengthen land tenure security, land reforms were implemented in 1998 and 2002. However, the land transfer market is not mature. Contracts for land transfer remain informal and are usually made with relatives. These are still costs for rural–urban migrants.
To study how the above three institutional factors interact with rural–urban migration, Liao, Wang, Wang, and Yip (2016) construct a general-equilibrium framework with endogenous fertility and migration decisions. The setup is an infinite-horizon model of one-period lived heterogeneous agents. Agents are heterogeneous in their preference toward quantity of children and their ability level. The framework includes the main elements of the three institutional factors. First, rural agents are attached to land but only have the rights of land use to highlight the pull role of the land tenure arrangement system. Land will be withdrawn after a period of time of their migration. In other words, rural–urban migrants do not collect land return. Only those who actually work on the farm obtain land returns. Second, above-quota penalties between rural and urban areas are differentiated to be consistent with the fact that rural fertility policies are more lenient. Rural agents take the fertility policies of the destination into account when making migration decisions. Third, regulations on rural–urban mobility are considered to capture the main spirit of China’s hukou system.
The model is calibrated to the data from China during 1980–2007 as the benchmark economy. Then, quantitative analyses are performed to investigate the interactions between migration decisions and the three institutional policies. Several interesting results related to the interactions are observed in the paper. First of all, the results suggest that the abolishment of the one-child policy in urban areas would disproportionately encourage high-skilled migrants, especially those who prefer having more children. Thus, urban fertility would strongly rebound and rural fertility would drop. Rural–urban migration makes rural production more effective, thereby narrowing the rural–urban income gap and lowering the incentive for low-skilled migration. Second, the higher frequency of land reallocation would result in higher income of rural workers that actually work on the farm because they could obtain more land. This policy would narrow the rural–urban income gap, thus lowering the incentive for rural–urban migration, especially for high-skilled migrants. Since those who prefer to have more children would like to stay in rural areas, rural fertility would increase slightly. Third, two types of hukou reforms are explored: relaxing the hukou regulation on high-skilled workers and shortening the waiting time to obtain urban hukou. The paper shows that the relaxation of the hukou regulation on high-skilled workers would mainly encourage high-skilled migrants, while shortening the waiting time to obtain urban hukou would only encourage low-skilled workers. This is because the tax burden on low-skilled workers is lower, but high- and low-skilled workers actually enjoy the same urban benefits after they obtain urban hukou.
The Higher Education System and Its Reform
To obtain admission to universities or colleges, students in China have to take the National Entrance Examination (gaokao). The system of gaokao was established at the beginning of the 1950s. It was abolished for several years during the Cultural Revolution and was then restored in 1977.
As discussed, zhaosheng, entering college through gaokao, was one of the official channels for rural students to obtain urban hukou. Because of the scarcity of education resources, the acceptance rates were very low, especially in the 1980s. Students who passed gaokao enjoyed lots of benefits. First of all, college tuition was almost free before 1990. Second, a college freshman could choose to move his hukou registration place to his school. His hukou status would become non-agricultural as well. Third, during the period before 1994, a college graduate would be distributed to a stable government job. Thus, a college graduate simply moved his hukou registration to the working unit after graduation and still kept the non-agricultural status. In the periods when the policy of government job assignment was abolished, jobs were not guaranteed. A graduate could move his hukou registration to the working unit, temporarily put it to the collective joint household of personal exchange center, or move it back to his hometown (while still keeping the non-agricultural status). Thus, under China’s hukou system, entering college implied upward mobility in the society.
China has introduced a series of educational reforms since 1996, focusing on the expansion of college education and increases in college tuition. The expansion policy has advanced China’s higher education by a much faster pace than primary and secondary education and has been a strong source for China’s R&D growth (Yao, 2014). The expansion policy also provided broader access for rural students to enter college. For example, Gou (2006) shows that, during 1989–2005, the admission rate for rural students increased from 18.7% to 62.9%, while that for urban students increased from 30.4% to 68.4%. The percentage of freshmen originally from rural areas was even higher than it was from cities in 2005, 53.0% versus 47.0%, respectively. However, because of the rise in college tuition, lower-income families now shoulder a much higher burden for children’s college education than higher-income families (Li, 2007). Thus, the rural–urban disparity in access to top universities has been widened. Besides, since 2003, universities in Hong Kong have been allowed to enroll new students from Mainland China, and attending university in Hong Kong is becoming a popular choice. All of these factors affect education-based migration decisions.
In terms of education-related rural–urban migration, some studies have focused on the well-documented rural–urban schooling gap that motivates rural agents to migrate to get a better education in the city. Following the existing literature on the rural–urban schooling gap from the perspectives of educational attainment, enrollment, and education funding (Li & Wu, 2011; Tsang & Ding, 2005), Zhao, Ye, Li, and Xue (2017) study this gap according to the dimension of cognitive ability. Using the China Education Panel Survey (CEPS) 2013/2014 data, they confirm that the rural–urban schooling gap is significant in China. They find that the cognitive ability test scores of urban students are approximately 1.41 points (17%) higher than those of rural students, on average. It is suggested that the government can provide parenting skills training and can improve the quality of rural schools in order to narrow the rural–urban gap in cognitive ability.
Furthermore, the school district combination program encouraged by the central government since the late 1990s has led to an unexpectedly large scale of school closings in rural areas (Ding & Zheng, 2015). Using the China Health and Nutrition Survey and the prefecture-level and micro data from the 2005 mini census, Liu and Xing (2016) investigate how the relatively disadvantaged educational condition in rural China has affected the migration behavior of rural residents. They find that the decline in the number of primary schools in villages and rural areas increased the migration probability of the residents there. They conclude that the closing of rural primary schools has contributed to the increased number of rural to urban migrants.
These results suggest that rural–urban migration is related to the rural–urban gap in education opportunities. Thus, the education investment motive of rural–urban migration plays an important role in China’s urbanization and economic development. Zhang (2017) examines the role of the hukou system in migrant children’s educational attainment in cities. It explores whether the new education–poverty trap that results from the urban–rural disparity in education quality and institutional barriers of the hukou system can be broken by rural–urban migration. Applying logistic regressions and survival analysis to the longitudinal survey RUMiC, this paper finds that migrant children either drop out of urban schools early or are forced to return to rural schools with low quality. Thus, the hukou registration system reinforces the formation of the new poverty trap.
Others have analyzed the relation between human capital accumulation and rural–urban migration, as proposed in the seminal paper by Lucas (2004). Lall, Selod, and Shalizi (2006) contains a comprehensive survey of the empirical literature. They find that rural–urban income gaps, education, gender, and family composition are the most important factors that affect the decision of migrants. Numerous papers on rural–urban migration in China offer corroborating evidence, such as Zhao (1999a, 1999b), Hare (1999), De Brauw, Huang, Rozelle, Zhang, and Zhang (2002), and Giles and Mu (2007). Using migrant data from the 2005 One-Percent Population Survey of China from the National Bureau of Statistics, Xing and Zhang (2017) estimate semi-parametrically the expected earnings of rural–urban migrants in each possible destination city and their preference for city population size. They find that rural–urban migrants are willing to give up a substantial amount of income in order to live and work in larger cities. The first possible explanation suggested by Xing and Zhang (2017) is that larger cities help migrants to accumulate human capital faster than smaller ones. This is consistent with the theoretical prediction of Lucas (2004).
Qin, Wang, and Zhuang (2016) focus on the intergenerational aspect of human capital accumulation. They develop a three-period overlapping-generations model to analyze the impact of intergenerational transmission of human capital on income mobility in China. By estimating a simultaneous equations model using the 1989–2009 China Health and Nutrition Survey data, they find that human capital, measured by health and education, is directly transmitted from one generation to the next, reflecting the parent-induced inequality of development opportunities among offspring in China. The estimated intergenerational income elasticity is 0.481, which ranks higher than most of the developed and some of the developing countries in the literature, indicating a low level of intergenerational income mobility. Liu (2008) examines the determinants of rural–urban migration in the presence of human capital externalities in the rural sector. By focusing on human capital externalities in the rural sector, we are able to gain new insights into the general role of human capital in influencing the migration decision and to derive policy recommendations that help curtail rural–urban migration and hence alleviate urban unemployment pressure. Human capital at the aggregate level can act as an inhibitor to migration. If human capital can raise, through its external effect, the payoffs from rural activities, an increase in the overall level of educational attainment of the rural population would make staying in the rural sector a more attractive choice (at least for some potential migrants) relative to migrating to the city. This would be particularly true if there were plenty of local off-farm employment opportunities, in which human capital externalities are more important than in on-farm productions.
From both binomial and multinomial logit regression analyses of the Chinese Household Income Project 1995, Liu (2008) finds that in rural areas human capital externalities have a discouraging effect on rural–urban migration. That is, everything else being the same, a rural resident from a county rich in human capital is less likely to migrate to the city than his/her counterpart from another county poor in human capital endowment. One important policy implication of their findings is that China’s official policy of “leaving the farmland without leaving the village,” which expands education opportunities in rural areas, can be a viable strategy to curtail rural–urban migration and hence help alleviate urban unemployment pressure.
Using Chinese provincial data (1980–2005), Ha, Yi, and Zhang (2016) examine the effects of permanent and temporary emigration on human capital formation and economic growth in the Chinese source regions. As suggested by the literature of brain drain (Beine et al., 2001; Jagdish & Hamada, 1974), they find that both permanent and temporary emigration have a detrimental effect on the economic growth of source regions. This implies that the brain drain channel of emigration dominates the brain gain one, while the latter suggests that the emigration of skilled laborers may provide an incentive for those left behind to invest in human capital. Using school enrollment rates for human capital formation, Ha, Yi, and Zhang (2016) found that the two types of emigration have different effects on human capital accumulation in source regions. In a within-group fixed effects model, the estimates suggested that permanent emigration has a positive effect on human capital formation in the source regions, whereas the effect of temporary emigration is mixed. Specifically, they found that the magnitude of temporary emigration only has a significantly positive effect on middle school enrollment but does not significantly affect high school enrollment.
From the education-based perspective of rural–urban migration, we first note that most of the high-quality education institutions (including universities and specialized schools for art and music) are located in urban areas. Besides, Lucas (2004) proposes that cities are the places for accumulating human capital in work. Therefore, it is possible that students may first move to the city to attend college and then stay there for work after graduation. As a result, work-based and education-based migration together provide the whole picture of rural–urban migration. Then, the next questions are: How important is the role played by education-based migration in the process of economic development, and how significant is education-based migration compared with work-based migration?
Liao, Wang, Wang, and Yip (2017) take the lead in addressing these questions by constructing a dynamic general-equilibrium framework with endogenous education migration decisions. Since parents’ support is the most important source for financing a college education, it is natural to employ a two-period overlapping generations model with altruistic parents making education-migration decisions for children. Human capital accumulation and intergenerational mobility are also considered in the model to allow for the existence of work-based migration. Their theoretical analysis suggests that rural parent s will be willing to send their children to attend colleges in cities if the expected return of college education is higher than that of other options, including work-based migration. The model is then calibrated to the data from China from 1980–2007. Under China’s hukou regulation system, zhaosheng is a unique channel that mitigates rural–urban migration barriers. Therefore, zhaosheng provides a good example of education-based migration for quantitative analysis.
Table 1. Decomposition of zhaosheng
Percentage Explained by Interaction With:
Output Per Capita
Urban–rural TFP ratio
Human capital accumulation
Distortion on skilled workers
Based on the calibrated results, Liao, Wang, Wang, and Yip (2017) further perform numerical experiments to investigate the influences of education- and work-based migration on development. The result suggests that, during 1980–2007, the contribution of education-based migration to per capita output is quite comparable to that of work-based migration (4.9% vs. 4.4% of changes in per capita output). Without examining the education channel, the picture of rural–urban migration in China would be very misleading. The paper further finds that, although the quantity of education-based migrants is much fewer than that of work-based migrants in the survey of population census in 2000, the contribution of education-based migration in the second sub-sample period (1995–2007) still cannot be ignored, and is even larger than that in the first sub-sample period. To better understand it, a difference-in-difference quantitative analysis is performed to examine the underlying driving forces with which education-based migration is interacting. As reported in Table 1, although education-based migration interacts negatively with work-based migration and intergenerational mobility, the dominant forces are all positive: the interactions between education-based migration and the improvement in urban total factor productivity (relative to rural total factor productivity), human capital accumulation, and less distortion on skilled workers. The latter two interactions even account for approximately two-thirds of the effects of education-based migration on per capita output and skill premium. As a result, although the quantity of education-based migrants is less significant compared with that of the work-based channel in the second sub-sample period, the interactions with the underlying driving forces enhance the role of education-based migration in the development of China.
It is widely recognized that one of the engines of development is rural–urban migration. In the middle of the 20th century, Lewis (1954) and Todaro (1969) already pointed out that the movement of surplus labor from the rural agricultural sector to the urban industrial sector was the driver of urbanization and growth for developing countries. Following the Lewis–Todaro tradition, the majority of research studies have focused on the work-based motive of migration, which is determined by the rural–urban wage gap. But in the 1990s, researchers started to understand that the difference in earning capacity summarized by the rural–urban wage gap is not the only factor that dictates migration decisions. For instance, urban amenities are also an important consideration for migrants (de Laat, 2005). In this article, we emphasize that the gap in education opportunity and quality is crucial in making the migration decision from rural to urban. Both work-based and education-based motives must be considered in order to have a complete understanding of rural–urban migration.
In this article, we have focused on rural–urban migration in China from both the work-based and education-based perspective. This is because the rural–urban migration flows are huge in China, increasing from 27 million in 1985 to 1.7 billion in 2010. The literature on rural–urban migration in China is also expanding rapidly. We have seen that the hukou system has played an extremely influential role in the process of rural–urban migration. Its reform, together with other policy changes such as those on the education system, have had important impacts on the process of internal migration in China. Although the work-based proportion is larger in the overall internal migration than the education-based, the latter proportion is rising significantly whereas the former has been stable since 2003. In addition, as pointed out by Liao, Wang, Wang, and Yip (2017), numerical experiments show that both types of migration have similar contributions to per capita output for the period 1980–2007. It is also clear that there have been more and more studies on the relation between education, human capital accumulation, and growth in the past decade. We believe education-based migration is a promising research avenue for studying rural–urban migration in China.
Liao thanks the research assistant service from Hui-wen Chang and the financial support from the Ministry of Science and Technology of Taiwan, project No. MOST 103-2410-H-001-016-MY2. Needless to say, the usual disclaimer applies.
Asian Development Bank (2011). Asian Development Outlook 2011 Update: Preparing for Demographic Transition. Philippines: Asian Development Bank.Find this resource:
Beine, M., Docquier, F., & Rapoport, H. (2001). Brain drain and economic growth: Theory and evidence. Journal of Development Economics, 64, 275–289.Find this resource:
Bloom, D. E., Canning, D., Hu, L., Liu, Y., Mahal, A., & Yip, W. (2010). The contribution of population health and demographic change to economic growth in China and India. Journal of Comparative Economics, 38, 17–33.Find this resource:
Borjas, G. J. (1987). Self-selection and the earnings of immigrants. The American Economic Review, 77(4), 531–553.Find this resource:
Bosworth, B., & Collins, S. M. (2008). Accounting for growth: Comparing China and India. The Journal of Economic Perspectives, 22(1), 45–66.Find this resource:
Cattaneo, C., & Peri, G. (2016). The migration response to increasing temperatures. Journal of Development Economics, 122, 127–146.Find this resource:
Chan, K. W. (2001). Recent migration in China: Patterns, trends, and policies. Asian Perspectives, 25, 127−155.Find this resource:
Chan, K. W. (2013). China: Internal migration. In I. Ness & P. Bellwood (Eds.), The Encyclopedia of Global Human Migration (pp. 1–15). Hoboken, NJ: Wiley-Blackwell.Find this resource:
Chan, K. W., & Buckingham, W. (2008). Is China abolishing the hukou system? The China Quarterly, 195, 582−606.Find this resource:
Chernina, E., Dower, P. C., & Markevich, A. (2014). Property rights, land liquidity, and internal migration. Journal of Development Economics, 110, 191–215.Find this resource:
De Brauw, A., Huang, J., Rozelle, S., Zhang, L., & Zhang, Y. (2002). The evolution of China’s rural labor markets during the reforms. Journal of Comparative Economics, 30, 329−353.Find this resource:
De Brauw, A., & Rozelle, S. (2008). Migration and household investment in rural China. China Economic Review, 19, 320–335.Find this resource:
de Laat, J. (2005). Moral Hazard and Costly Monitoring: The Case of Split Migrants in Kenya, Job Market Paper. Brown University.Find this resource:
Dell, M., Jones, B. F., & Olken, B. A. (2012). Temperature shocks and economic growth: Evidence from the last half century. American Economic Journal: Macroeconomics, 4(3), 66–95.Find this resource:
Ding, D., & Zheng, F. (2015). Dismantling teaching points: Integrate education resources or reduce investment in education based on the analysis of 1996–2009 provincial panel data. China Economic Quarterly, 14(2), 603–622 (in Chinese).Find this resource:
Eaton, J., & Eckstein, Z. (1997). Cities and growth: Theory and evidence from France and Japan. Regional Science and Urban Economics, 27, 443–474.Find this resource:
Feng, S., Krueger, A. B., & Oppenheimer, M. (2010). Linkages among climate change, crop yields and Mexico–US cross-border migration. Proceedings of the National Academy of Sciences of the United States of America, 107(32), 14257–14262.Find this resource:
Foster, A. D., & Rosenzweig, M. R. (2007). Economic development and the decline of agricultural employment. In T. Paul Schultz & John A. Strauss (Eds.), Handbook of Development Economics (Vol. 4, pp. 3051–3083). Amsterdam, North-Holland: Elsevier.Find this resource:
Giles, J., & Mu, R. (2007). Elder parent health and the migration decision of adult children: Evidence from rural China. Demography, 44(2), 265–288.Find this resource:
Glomm, G. (1992). A model of growth and migration. Canadian Journal of Economics, 25(4), 901–922.Find this resource:
Golley, J., & Wei, Z. (2015). Population dynamics and economic growth in China. China Economic Review, 35, 15–32.Find this resource:
Gong, X., Kong, S. T., Li, S., & Meng, Z. (2008). Rural–urban migrants: A driving force for growth. In R. Garnaut, L. Song, & W. T. Woo (Eds.), China’s Dilemma (pp. 110–152). Canberra: Asia Pacific Press.Find this resource:
Gou, R. (2006). Examining equality in higher education from the perspective of rural–urban access to higher education. Research in Educational Development, 5, 29–31 (in Chinese).Find this resource:
Greenwood, M. J. (1997). Internal migration in developed countries. In M. R. Rosenzweig & O. Stark (Eds.), Handbook of Population and Family Economics (Vol. 1B, pp. 647–720). Amsterdam, North-Holland: Elsevier.Find this resource:
Ha, W., Yi, J., & Zhang, J. (2016). Brain drain, brain gain, and economic growth in China. China Economic Review, 38, 322–337.Find this resource:
Hare, D. (1999) “Push” versus “pull” factors in migration outflows and returns: Determinants of migration status and spell duration among China’s rural population. The Journal of Development Studies, 35(3), 45–72.Find this resource:
Harris, J. R., & Todaro, M. P. (1970). Migration, unemployment and development: A two-sector analysis. The American Economic Review, 60(1), 126–142.Find this resource:
Henderson, J. V. (2005). Urbanization and growth. In P. Aghion & S. N. Durlauf (Eds.), Handbook of Economic Growth (Vol. 1B, pp. 1543–1591). Amsterdam: Elsevier.Find this resource:
Herrendorf, B., Rogerson, R., & Valentinyi, A. (2014). Growth and structural transformation. In P. Aghion & S. N. Durlauf (Eds.), Handbook of Economic Growth (Vol. 2, pp. 855–941). Amsterdam, North-Holland: Elsevier.Find this resource:
Hu, F., Xu, Z., & Chen, Y. (2011). Circular migration, or permanent stay? Evidence from China’s rural–urban migration. China Economic Review, 22, 64–74.Find this resource:
Jagdish, B., & Hamada, K. (1974). The brain drain, international integration of markets for professionals and unemployment. Journal of Development Economics, 1(1), 19–42.Find this resource:
Kaplan, G., & Schulhofer-Wohl, S. (2017). Understanding the long-run decline in interstate migration. International Economic Review, 58(1), 57–94.Find this resource:
Kelley, A., & Schmidt, R. (2005). Evolution of recent economic-demographic modelling: A synthesis. Journal of Population Economics, 18, 275–300.Find this resource:
Lagakos, D., & Waugh, M. E. (2011). Selection, agriculture, and cross-country productivity differences. The American Economic Review, 103(2), 948–980.Find this resource:
Lall, S. V., Selod, H., & Shalizi, Z. (2006). Rural–urban migration in developing countries: A survey of theoretical predictions and empirical findings. World Bank Policy Research Working Paper 3915.Find this resource:
Lewis, W. A. (1954). Economic development with unlimited supplies of labour. The Manchester School, 22(2), 139–191.Find this resource:
Li, H., & Wu, B. (2011). China’s educational inequality: Evidence from College Entrance Exams and Admissions. In the conference China economics summer institute 2011 (July 1–3).Find this resource:
Li, W. (2007). Family background, financial constraints and higher education attendance in China. Economics of Education Review, 26, 725–735.Find this resource:
Liao, P.-J., Wang, P., Wang, Y.-C., & Yip, C. K. (2017). Educational choice, rural–urban migration and economic development. NBER Working paper No. 23939.Find this resource:
Liao, P.-J., Wang, P., Wang, Y.-C., & Yip, C. K. (2016). To stay or to migrate? The one-child policy, land tenure arrangement and work-based migration in China. Working paper.Find this resource:
Liu, J., & Xing, C. (2016). Migrate for education: An unintended effect of school district combination in rural China. China Economic Review, 40, 192–206.Find this resource:
Liu, Z. (2008). Human capital externalities and rural–urban migration: Evidence from rural China. China Economic Review, 19, 521–535.Find this resource:
Lu, Z., & Song, S. (2006). Rural–urban migration and wage determination: The case of Tianjin, China. China Economic Review, 17, 337–345.Find this resource:
Lucas, R. E. (1997). Internal migration in developing countries. In M. R. Rosenzweig & O. Stark (Eds.), Handbook of Population and Family Economics (Vol. 1B, pp. 721–798). Amsterdam, North Holland: Elsevier.Find this resource:
Lucas, R. E. (2004). Life earnings and rural–urban migration. Journal of Political Economy, 112(S1), S29–S59.Find this resource:
Meng, X. (2012). Labor market outcomes and reforms in China. The Journal of Economic Perspectives, 26(4), 75–101.Find this resource:
Meng, X. (2013). Rural–urban migration: Trend and policy implications (2008–2012). In R. Garnaut & C. Fang (Eds.), China: A New Model for Growth and Development (pp. 179–197). The Australian National University Press.Find this resource:
Minami, R., & Ma, X. (2010). The turning point of Chinese economy: Compared with Japanese experience. China Economic Journal, 3(2), 163–180.Find this resource:
Morten, M. (2016). Temporary migration and endogenous risk sharing in village India. NBER Working Paper No. 22159.Find this resource:
Munshi, K. (2003). Networks in the modern economy: Mexican migrants in the US labor market. The Quarterly Journal of Economics, 118(2), 549–599.Find this resource:
Munshi, K. (2011). Strength in numbers: Networks as a solution to occupational traps. Review of Economic Studies, 78(3), 1069–1101.Find this resource:
Munshi, K., & Rosenzweig, M. (2016). Networks and misallocation: Insurance, migration, and the rural–urban wage. The American Economic Review, 106(1), 46–98.Find this resource:
Orazem, P., & King, E. M. (2007). Schooling in developing countries: The roles of supply, demand and government policy. In T. P. Schultz & J. A. Strauss (Eds.), Handbook of Development Economics (Vol. 4, pp. 3475–3559). Amsterdam, North-Holland: Elsevier.Find this resource:
Ozden, C., Parsons, C. R., Schiff, M., & Walmsley, T. L. (2011). Where on earth is everybody? The evolution of global bilateral migration 1960–2000. World Bank Economic Review, 25(1), 12–56.Find this resource:
Pi, J., & Zhang, P. (2016). Hukou system reforms and skilled–unskilled wage inequality in China. China Economic Review, 41, 90–103.Find this resource:
Qin, X., Wang, T., & Zhuang, C. C. (2016). Intergenerational transfer of human capital and its impact on income mobility: Evidence from China. China Economic Review, 38, 306–321.Find this resource:
Rosen, S. (1972). Learning by experience as a joint production. The Quarterly Journal of Economics, 86, 366–382.Find this resource:
Roy, A. D. (1951). Some thoughts on the distribution of earnings. Oxford Economic Papers, 3(2), 35–46.Find this resource:
Sjaastad, L. A. (1962). The costs and returns of human migration. Journal of Political Economy, (Suppl. 70), 80–89.Find this resource:
Song, Y. (2014). What should economists know about the current Chinese hukou system? China Economic Review, 29, 200–212.Find this resource:
Stark, O. (1991). The Migration of Labor. Cambridge, MA: Basil Blackwell.Find this resource:
Todaro, M. P. (1969). A model of labor migration and urban unemployment in less developed countries. The American Economic Review, 59, 138–148.Find this resource:
Tsang, M. C., & Ding, Y. (2005). Resource utilization and disparities in compulsory education in China. China Review, 5(1), 1–31.Find this resource:
Van de Ven, W. P. M. M., & Van Praag, B. M. S. (1981). The demand for deductibles in private health insurance: A probit model with sample selection. Journal of Econometrics, 17, 229−252.Find this resource:
Wei, Z., & Hao, R. (2010). Demographic structure and economic growth: Evidence from China. Journal of Comparative Economics, 38(4), 472–491.Find this resource:
Williamson, J. G. (1988). Migration and urbanization. In H. Chenery & T. N. Srinivasan (Eds.), Handbook of Development Economics (Vol. I, pp. 425–465). New York, NY: Elsevier.Find this resource:
Xing, C., & Zhang, J. (2017). The preference for larger cities in China: Evidence from rural–urban migrants. China Economic Review, 43, 72–90.Find this resource:
Yao, Y. (2014). The Chinese growth miracle. In P. Aghion & S. N. Durlauf (Eds.), Handbook of Economic Growth (Vol. 2B, pp. 943–1031). Philadelphia: Elsevier.Find this resource:
Young, A. (2013). Inequality, the urban–rural gap, and migration. The Quarterly Journal of Economics, 128(4), 1727–1785.Find this resource:
Zhang, H. (2017). Opportunity or new poverty trap: Rural–urban education disparity and internal migration in China. China Economic Review, 44, 112–124.Find this resource:
Zhao, G., Ye, J., Li, Z., & Xue, S. (2017). How and why do Chinese urban students outperform their rural counterparts? China Economic Review, 45, 103–123.Find this resource:
Zhao, Y. (1999a). Leaving the countryside: rural-to-urban migration decisions in China. The American Economic Review, 89(2), 281–286.Find this resource:
Zhao, Y. (1999b). Labor migration and returns to rural education in China. American Journal of Agricultural Economics, 79(4), 1278–1287.Find this resource:
Zhao, Z. (2007). Migration, labor market flexibility, and wage determination in China: A review. The Developing Economics, 43, 285−312.Find this resource:
Zhu, N. (2002). The impacts of income gaps on migration decisions in China. China Economic Review, 13, 213–230.Find this resource:
(1.) For a summary of the literature on rural–urban migration, see Williamson (1988), Greenwood (1997), Lucas (1997), and recently Foster and Rosenzweig (2007). For a discussion on rural–urban migration and cities growth, a topic which we skip in our review, see Henderson (2005).
(4.) The data of 1985 is from the 1990 Population Census of China; that of 2000 is from China 2000 Population Census; that of 2005 is from China 1% Population Sample Survey Data 2005; that of 2010 is from China 2010 Population Census; and the data of 2015 is from China 1% Population Sample Survey 2015.
(5.) The first category, work-based migration, includes the migration reasons of “job transfer,” “job assignment,” and “work or business.” The second category, education-based migration, refers to the reason of “study and training.” In the data of 2015, “for children’s education” is also included when we count education-based migration. Other migration reasons are summarized in the third category, others, such as “to relative and friend,” “retired and resigned,” “moved with family,” “marriage,” and “pull down and move.”
(6.) The higher share of education-based migration in 1985 is due to the zhaosheng system and the government job assignment.