Shihe Fu: Learning in Chinese Cities: Do Rural Migrants Benefit from Agglomeration Economies?



On October 17, 2016, Prof. Shihe Fu, a Fulbright Scholar in the Center for Real Estate at MIT and professor of Urban and Real Estate Economics at the Southwestern University of Finance and Economics in China, gave a lecture as part of the fall 2016 China Talk Series. He spoke about the benefits rural migrants gain from agglomeration economies in China. 

Prof. Fu began by defining the essential terms for understanding agglomeration economies in China. He first defined a city from the perspective of urban economics as a concentration of people and firms in a limited geographic area. People and firms generally want to be close together in a city due to the benefits of concentrating resources. This phenomenon is called agglomeration. Indeed, the benefit of concentrating workers and firms is one of the reasons why cities exist.

There are two types of agglomeration economies: localization and urbanization economies. In localization economies, benefits accrue from the concentration of the same industry in the same area (e.g. Silicon Valley). In urbanization economies, benefits accrue from the diversity of other industries and people located in the same place (e.g. New York City). This is especially true for small, specialized firms, which need input from a variety of other firms.

Localization economies are based on sharing expensive inputs like highways and airports, pooling resources such as workers and firms, and gaining knowledge from information spillovers. Industries that are collocated tend to grow quickly in the long run due to these affects, which are called Marshallian externalities.

Urbanization economies are much the same as localization economies. However, theorists originally speculated that economic performance for urbanization economies was based solely on the size of a city. Jane Jacobs, however, shifted the argument from urban size to urban diversity with her book The Death and Life of Great American Cities.

The focus of Prof. Fu’s talk was on labor market agglomeration economies. In large agglomeration economies, workers enjoy a wage premium, which is the difference in wages earned by workers in large versus small cities. This difference exists in part due to the higher worker productivity found in agglomeration economies. These labor market agglomeration economies are founded on labor market pooling (reducing search friction and increasing labor mobility) as well as knowledge spillovers (human capital externalities) including poaching, peer effects, formal communication, and informal communication between individuals and firms.

The spread of ideas and innovation in such an economy is a good example of a Marshallian externality. Prof. Fu quoted Lucas (1988) to illustrate the point that social interaction in cities creates positive spillover affects in the workforce: “Most of what we know we learn from other people…without a distinction between student and teacher.”

Prof. Fu explained that the existing empirical evidence for labor market agglomeration economies focuses on developed countries, city size, and urban workers. This leaves the question of whether skilled people move to cities or if cities make people more skilled. Moreover, the existence and extent of labor market agglomeration economies in China has not been studied. Prof. Fu therefore posed several research questions: whether Marshallian externalities exist in Chinese cities; if they do, how large are they, and whether rural migrants benefit from labor market agglomeration. Prof. Fu argued that understanding labor market agglomerations in China is important given the massive rural-urban migration of low-skilled workers, regulations on urban growth such as the Hokou system, global competition in the Chinese market, the massive growth of Chinese cities, and the need to make the urban labor force more skilled.

To answer these research questions, Prof. Fu created a model that incorporates wages, industries, cities, and individual attributes such as gender. The variables in the model were college share (number of workers with college degree out of all industry workers) and a variable of agglomeration (total employment in an industry in a city). The results of the model showed that as education increased, wages increased, which mirrors the findings in studies on Marshallian externalities in other countries. Rural migrants, however, benefit less from Marshallian externalities than urban migrants or local urban residents.

The main findings of Prof. Fu’s study are as follows:

  • Marshallian externalities exist in the urban labor market in China
  • Rural migrants benefit from Marshallian externalities, but benefit much less than do urban workers or local workers with an urban Hokou
  • There is double discrimination against rural migrants based on Hokou and migration status

Based on these findings, Prof. Fu asked why rural migrants benefit so much less from labor market agglomerations than other sectors of the population. Because industry and skill were controlled for in the model, Prof. Fu argues that this group’s low benefit from agglomeration is due primarily to its small socialization network or, in other words, discrimination based on status, which excludes rural migrants from many of the benefits of the agglomeration economy.

Other studies also suggest that the lack of benefits is due to double discrimination against rural migrants, who cannot find jobs and, when they do, get paid less. Prof. Fu showed that rural migrants live far away from the city center, whereas urban migrants live closer to downtown with greater access to jobs and beneficial networks.

Prof. Fu concluded his lecture by asking how Chinese cities can attract skilled people. He argued that they can do so by making cities safe, clean, accessible, livable, and most importantly, by making cities open, tolerant, and less discriminatory. This final point will ensure that the benefits of agglomeration can be enjoyed by all.

Following the talk, Albert Saiz, associate professor of urban economics and real estate and director of the Center for Real Estate at MIT, responded to Prof. Fu’s lecture. He emphasized the novelty of Prof. Fu’s study and findings, commenting that wage differences are the key to understanding differences in productivity. This is because in economic reasoning, wages (W) are assumed to equal marginal product of labor (MPL). However, Prof. Saiz pointed out that we don’t know for a fact if MPL = W in China because of the unique structural characteristics of the Chinese economy. For MPL to equal W, the economy must reach equilibrium. Due to China’s intense growth, restrictions on mobility, and the prevalence of State Owned Enterprises, it is unclear whether the economy in any given city has reached equilibrium. Moreover, equilibrium is affected by institutional factors such as unions and legal regulations, which add another layer of uncertainty to the Chinese case.

Prof. Saiz pointed out that this question could provide an interesting new direction to Prof. Fu’s research, which could look at differences between cities in China and also compare them to other areas of the world with a mixed portfolio of spatially desperate cities, such as in Europe.  

Prof. Saiz also applied the same logic of institutional factors to question the magnitude of the wage premium between large and small cities. He concluded his response to the talk by agreeing with Prof. Fu’s overall assessment and stress on the need to reduce discrimination to attract more skilled labor to Chinese cities. Prof. Saiz added that the urban-rural divide is not a zero-sum game, but rather one in which increased capacity can benefit all.

- Max Budovitch, first-year DUSP student and STL Fellow

Cities facilitate learning and human capital accumulation. In a dense, local labor market, workers can benefit from knowledge spillovers and therefore enhance their productivity. This is supported by many empirical studies from developed countries. Less is known in cities in developing countries. Using micro data from the 2004 manufacturing census and 2005 population census in China, we find that overall workers benefit from labor market pooling and knowledge spillovers in Chinese cities, but rural migrants benefit much less than do local urban residents. This is not because rural migrants are low skilled or work in informal sectors. This may be because they lack social network and suffer “double discrimination” for being “rural” and being “migrant.” Our findings suggest that social interactions in cities provide a channel of learning alternative to formal schooling. Our findings also have policy implications on how Chinese cities can become “skilled” during the rapid urbanization process coupled with global competition.

Shihe Fu received his Ph.D. in economics from Boston College in 2005. He is a professor at Southwestern University of Finance and Economics in China and currently a Fulbright visiting scholar at the MIT Center for Real Estate. His research focuses on housing, labor, and environmental issues in cities. His papers appear in Journal of Public EconomicsJournal of Labor EconomicsJournal of Urban EconomicsJournal of Regional Science, andUrban Studies.

Recommended Reading List:

Cities and skills, by E. Glaeser and D. Maré, Journal of Labor Economics 19(2001), pp. 316-342.

Communication externalities in cities, by S. Charlot and G. Duranton, Journal of Urban Economics 56 (2004), pp.581–613.

Do rural migrants benefit from urban labor market agglomeration economies? Evidence from Chinese cities, by G. Yang, L. Li, and S. Fu, 2016, Working Paper. Available upon request: