The Differential Returns To Education In Urban China And Its Causes

The rate of return to education as a measure of the return to investment in education, has become an important practice of human capital theory in the field of applied economics. According to neoclassical economics, if in the perfect competition market, workers will eliminate intercity or interlocal wage differentials by migration or job hopping, homogeneous workers will have the same marginal labor productivity, and the return to education of each worker will be equal, which means that in the perfect competition market, the rate of return to education is invariant across cities.However, in practice, even though China’s urban labor market has been transferred from the planned economy to the market economy, and wages get close to the value of labor factors like education gradually because of wage reform and labor mobility, our urban labor markets are still at the immature stage, various types of local labor markets develop unequally, and some segmentation factors in labor markets like household restriction and monopoly industries still hinder the free movement of labor factors, which makes wage differentials expand among cities. On existing condition, the rates of return to education will be the same across different cities? If not, what kind of differences will be existed? How much do regional differentials in pay depend on different levels of education? Those are main issues that need to be discussed in the paper.In recent years, academe which studies the return to education in urban china mainly focuses on analysis of the trend of time series and marginal changes, and rarely refers to regional differentials in the rates of return to education. Even if there exists this kind of research, it didn’t answer the question like whether the rates of return to education across cities are the same at the interprovincial or national levels, even no answer of the question whether the rates of return to education are invariant across cities and how much difference there exists to be at the structural level covered by the whole body.Therefore, this paper examines whether the rates of return to education in Chinese cities are the same with the scope of urban areas and the angle of empirical study. According to urban distribution, we divide cities into three types:municipalities, provincial capitals and other cities. After controlling the elements which would affect wages like sex, ability, industry, occupation, we use CGSS(2006) data to analyze the differences in the rates of return to education across three types of cities, and the results show that the average rate of return to education in municipalities is 9.9%,8.4% in provincial capitals and 7.4% in other cities, which can reflect the overall differences in rates of return to education across cities. Since the average rates of return to education cover the impact on individual wages from different levels of education, we divide education variable into three levels: "college and above", "senior high school" and "junior high school and below". At the "junior high school and below" level, the rates of return to education in municipalities, provincial capitals and other cities are 4.1%,2.9% and 4.3%; at the "senior high school" level, the rates in three types of cities are 7.3%,8.6% and 7.4%; at the "college and above" level, the rates are respectively 14.1%,11.6% and 11.5%, which can reflect the structural differences in rates of return to education across cities. Meanwhile, this paper uses chow test to test the group regression method and uses quantile regression to test the heterogeneity problem.The main conclusions of this paper are:firstly, there are not only the overall differences but also the structural differences in rates of return to education across three types of cities in china. The existence of structural differences does not negate the overall differences, and there are detailed structural differences behind the overall differences.Secondly, the overall differences show that the rate of return to education in municipalities is the highest, in provincial capitals is in the middle, and in other cities is the lowest. As to these differences, this paper gives four explanations:1) the urban difference of industrial productivity. As it is an obvious phenomenon of "metropolis wage premium" in china which is manifested in municipalities in a productivity advantage of the high-end tertiary industry and the provincial capitals in a comparative advantage of the modern manufacturing sector, the wage level effect caused by these productivity advantages makes the differences of the average rates of return to education at three types of cities.2) the urban difference of human capital accumulation. Facing the information dissemination, technical exchange and knowledge sharing caused by various industrial aggregations, the difference is shown to consist of both way and speed of workers obtaining human capital across cities, and the highly skilled people may have a comparative advantage of human capital accumulation and labor productivity.3) the urban difference of labor mobility. When the urban labor market is larger, the cost of search is lower and the efficiency of search is higher, then labor mobility is more adequate and job turnover is more productive with greater job-to-job wage growth. Therefore, job turnover is the most frequent thing and the effect of job-to-job wage growth is the largest in municipalities, then it is in provincial capitals, and the lowest is in other cities.4) the urban difference of cost of living. Cost of urban living may affect the calculated rate of return to education by affecting workers’ requirement to earnings. Because of the high cost of living like housing price and traffic cost, residents in big cities have to ask for higher wages and firms have to pay higher wages, which may be one of the reasons that the average rate of return to education in municipalities is the highest, in provincial capitals is in the middle, and in other cities is the lowest.Thirdly, the structural differences of rates of return to education across cities are larger, and the reasons are more complex. This paper mainly analyzes the differences of rates of return to education at the "college and above", "senior high school" and "junior high school and below" levels across three types of cities, and the conclusions are as follow:1) the rates of return to "junior high school and below" are low with little difference across cities. This is because the positive externalities tend to be strong at the primary education level with the characteristic of return to society bigger than return to person, and the workers at this level cannot get the wage growth effect of "metropolis wage premium", so the rates of return to "junior high school and below" are low across cities; in addition, it is shown that the mobility of workers at this education level is more adequate across cities.2) the rates of return to "senior high school" increase obviously and the highest is in provincial capitals. It is because the education at this level has some characteristic of specificity and the interiorization of return to education makes the rates increase. The reasons of the highest rates at this level in provincial capitals could be following:the productivity advantage of modern manufacturing sector in provincial capitals and the phenomenon of "skill shortage" at the modern manufacturing sector.3) the rates of return to "college and above" increase substantially which is consistent with the point that the rates of return to education in our country has a marginal increase with the improvement of education level. The highest rate of return to "college and above" is in municipalities, and the main reasons are that the elements like agglomeration of high-end tertiary industry, household restriction, school level restriction and high urban living cost make the rate of return to this level in municipalities is the highest, and the difference of the other two types of cities is not obvious.Fourthly, the differences in rates of return to the same education level are a manifestation of low economic efficiency. If workers with the same education level can move to cities with high salary, the labor market can get a Pareto improvement to improve efficiency of overall economy.Finally, according to above research findings related to the practical issues, some policy implications are as follows:1) attaching importance to primary education continuously; 2) optimizing the urban industrial development; 3) supporting vocational education and training strongly; 4) eliminating labor mobility barriers; 5) paying attention to controlling the cost of living in big cities.