Labor institution environment and labor agglomeration based on provincial panel data analysis 1

: With the shift in regional economic patterns in China, Chinese labor forces are showing a trend toward concentration. Many scholars are analyzing this trend of labor agglomeration from the perspective economies of scale and transaction costs. This study explains the phenomenon of labor agglomeration from the perspective of the labor system environment. First, it considers that the labor system environment includes the social contract environment and labor contract environment. Second, it attempts to show that a good labor system environment can promote labor agglomeration through theoretical analysis, while technological advancement can promote the agglomeration of professional and technical talents. Third, this study uses provincial panel data to conduct an empirical analysis. Finally, the paper presents some policy suggestions for maintaining a good labor system environment, including improving the level of public service of the floating population and building harmonious labor relations.


Research background
Since the reform and opening up of the Chinese economy, major changes have occurred in China's regional economic pattern. The eastern region's economy has taken the lead in developing international industrial transfer and integration into the international production and trade system. It has also induced the central and western labor force to transfer to the east, creating the trend of labor force gathering from the central and western regions to the eastern region and transferring from rural to urban areas. Recently, some changes have occurred in the patterns of economic agglomeration and labor force agglomeration in China. First, the rural surplus labor force is exhausted, and the trend of labor force concentration from rural to urban areas is gradually weakening. From the perspective of the dual economic structure theory, the differences between urban and rural economic development have led to the migration of the floating population from rural areas to cities. Cai Fang (2011a) believes that, overall, the surplus rural labor force has ended and the Lewis turning point of labor transfer has arrived.
According to the survey data of the floating population, the trend of labor mobility has changed, and the proportion of labor force employment nearby has gradually increased.  3 . This also leads to the one-way flow and agglomeration trend in the labor force from the central and western regions to the eastern region changing to a two-way flow and agglomeration trend. It is, therefore, meaningful to study the motivation of labor force agglomeration in different regions.
The explanation of the new economic geography theory for economic agglomeration lies in the assumption of increasing returns to scale and monopoly competition (incomplete competition) (Krugman, 1991). Rapid economic development can easily result in the phenomenon of economic agglomeration. The expansion of the regional economic scale results in stronger economic competitiveness, stronger attraction for capital, labor, and other production factors, improvement of total factor productivity, and widening of the gap with other regional economies rather than a balanced development among regions. The increasing returns to scale can be explained as a forward market effect, backward market effect, and technology spillover effect.
Economic agglomeration certainly leads to labor agglomeration. Undeniably, the labor system environment plays a significant role in population agglomeration. A good institutional environment can reduce labor market transaction costs, promote the matching efficiency of labor supply and demand, and ensure the optimal allocation of labor force.
The labor system environment includes the labor contract environment between workers and enterprises as well as the social contract environment between workers and local governments. The labor contract environment is mainly reflected in labor contracts, wages and benefits, the settlement of labor disputes, and other factors; the social contract environment is mainly reflected in public social service and social welfare. The main economic motivations of labor flow and agglomeration can be considered from two aspects: labor contract environment and social contract environment. A good labor contract environment can meet the expectations of workers has made relevant supporting measures and implemented subsidies and subsidies in housing and employment to attract college students to stay in Hubei and work in Wuhan. By 2019, the target has been exceeded. "According to the Organization Department of the Wuhan municipal Party committee, since the implementation of the" million college students' employment and Entrepreneurship Project "in Wuhan in 2017, a total of 1.095 million college students studying in Han have been added, and the original plan has been completed two years in advance." https://hb.qq.com/a/20200107/022121.htm. In 2018, Xi'an also proposed the "five-year retention of millions of talents" plan, providing certain subsidies in terms of household registration, housing, employment, and entrepreneurship.
concerning wages and benefits, working conditions, and labor relations. A good social contract environment enables the labor force to enjoy better social welfare and public services. Williamson (1981)  reinforcing. In the low transaction costs context, skilled labor is still concentrated, while unskilled labor will flow again. The transaction cost of a labor contract can be divided into prior cost and after cost. A good labor system environment can reduce both these transaction costs, improve matching efficiency, and promote labor agglomeration.

Review of relevant research literature
There is a phenomenon of urban-rural segmentation in the labor market in developing countries. Economists generally believe that a reasonable flow of labor in different regions plays a key role in economic development. Cai Fang (2011b) believes that since the reform and opening up, much of the surplus rural labor force has entered the city, contributing enormous labor resources and demographic dividends to economic development, bringing about sustainable economic prosperity. From the human capital theory perspective, Schultz (1970) postulates that labor mobility is an important means of promoting the accumulation of human capital, and breaking the institutional barriers hindering labor mobility can promote economic growth. The classic model of labor mobility includes the Lewis model (1954), which considers that economic factors lead to labor flow from rural to urban areas. When the marginal output of labor in urban and rural areas is equal, the surplus rural labor force will cease to flow to the city, and the Lewis turning point will be established. Harris and Todaro (1970) proposed a two-sector model for labor mobility. The model considers that the expected income gap between urban and rural areas leads to the labor flow from rural to urban areas, which occurs not only before the Lewis turning point, but also when there is no surplus labor force in rural areas. With industrial agglomeration and labor concentration, technological advancement is more likely to occur in the city, increasing the income gap between urban and rural areas to a certain extent, with a significant impact on labor mobility. Generally, advancement in agricultural technology will lead to a decline in the demand for agricultural labor and promote the transfer of labor force from the agricultural to the industrial sector; urban technological advancement will lead to the rebalancing of labor force, which may have an agglomeration or dispersion effect (Peng A good institutional environment can reduce transaction costs, promote the development of contract-intensive and labor-intensive industries, and promote the agglomeration of high-quality labor. As an embodiment of social contract relationships, high-level public service may, to a certain extent, promote the labor agreement between workers and employers. Through empirical analysis of data, scholars have verified the role of public services in the promotion of population agglomeration (Tiebout, 1956;Day, 1992;Dahlberg et al.2012). In addition, some domestic scholars use survey data to find that more public services for the floating population will contribute to stable work in cities (Xia Yiran, 2015;Hou Huili, 2016). Some scholars have also studied the impact of the social security system on the cross-border flow of labor. Doris (1998) believes that if a country's social security system is not comprehensive, it will lead to the migration of talent and capital from that country to a country with a perfect social security system. Hassler  From the a forementioned literature review, there are two mechanisms concerning the impact of the institutional environment on labor agglomeration. The first is that the institutional environment can reduce transaction costs, increase contract-intensive and technology-intensive industrial investment, and attract high-quality talent. The second aspect is the theory of new economic geography, which holds that agglomeration can achieve economies of scale, generate higher returns, and promote labor agglomeration.
This study analyzes the internal impact of the labor system environment on labor agglomeration from a transaction costs perspective.

Static analysis
According to the new economic geography theory, monopoly competition is the cause of the developmental imbalances among regions. Theoretically, the economies of scale and technological advantages of economic agglomeration must be maintained by continuous gathering of high-quality talent. A region can maintain a monopoly advantage (economic agglomeration advantage) from the perspective of cost advantage.
Assuming that the cost of a typical initiative is related to the population size of the region ( ) , ( ) can be considered an inverted U-shaped curve. With population agglomeration, there is a positive agglomeration effect, that is, an increase in economic efficiency and a decrease in costs. With the increase in regional population agglomeration, there is a negative agglomeration effect. Economic efficiency starts to decrease and costs begin to increase. The benefit spectrum of economic agglomeration (labor agglomeration) is the opportunity cost (minimum cost of other cities 2 , 3 ), represented in Figure 1 by ̅ ; the portion lower than the opportunity cost (shadow area) can be viewed as the benefit of agglomeration economy. If the institutional environment is added to the model, the search and ex-post contract costs can be reduced by a good institutional environment. The cost curve will therefore be reduced to ′ ( ), which will expand the scope and benefits of the agglomeration economy. Figure 1 Impacts of institutional environment on labor agglomeration.

Dynamic analysis
According to the traditional Cobb Douglas production function, the economic output function of a region is assumed to be where is the regional technical level, is the level of capital investment, is the size of the labor force in the region, is the level of human capital in the region, and is the proportion of labor matching in the region. The equilibrium growth rate can be obtained as follows: From formula (1), the balanced growth rate depends not only on the capital growth rate, technological advancement rate, and comprehensive human capital growth rate, but also on the matching rate, labor growth rate, and human capital growth rate. 1) Technological advancement is the main driving force behind economic growth. In addition to promoting the advancement of overall factor productivity, technological advancement can generate more jobs, encourage integration with the labor market, and contribute to economic growth.
2) A good labor system environment can promote the integration of human and technological resources and foster economic development.

Selection of statistical indicators of labor market environment
The supply and demand of the labor market are determined not only by labor In the empirical analysis, we analyze the causes of labor agglomeration from three aspects: first, the availability of employment in the local labor market before employment, that is, whether the labor demand is strong. If the labor market demand is strong, it will lead to more labor agglomeration. Second, employment is a level of security for workers' rights and interests, which may arise from the signing of labor contracts and labor disputes, and the level of social welfare. If workers have access to better social welfare, including medical and education services, and better public services, labor agglomeration will be more conducive.

1) Regression results of fixed effect model
Since panel data regression is selected for regression analysis and 31 provinces and autonomous regions in mainland China are selected for regression analysis, the fixed effect model is selected when selecting the regression method. At the same time, the Hausman test finds that the fixed effect model is more suitable for this regression model. Table 2: Impact of labor institutional environment on labor agglomeration (FE model)

Variable Explained variable (total employment)
Model Ⅰ Model Ⅱ Model Ⅲ Model Ⅳ coefficient P-value coefficient P-value coefficient P-value coefficient P-value  Table 2 for details.
2) Robustness test GDP, fixed asset investment, and patent authorization are all related to the overall employment-the endogenous problem of explanatory variables. The empirical analysis department adopts the first-order lag term of these variables as the instrumental variable (IV) of these explanatory variables. Through regression analysis, it was found that the instrumental variable method does not affect the overall regression conclusion.
See Table 3 for details. Model Ⅰ Model Ⅱ Model Ⅲ Model Ⅳ coefficient P-value coefficient P-value coefficient P-value coefficient P-value

3) Categorical regression analysis
This study analyzes the effect of the labor system environment on talent concentration. In the Statistical Yearbook, the number of employees with a college degree or above in some years in Tibet is zero. Therefore, in this regression analysis, the data of the Tibet region are excluded, and the data of the remaining 30 provinces and cities are analyzed. From the analysis results, the institutional environment also has a significant role in promoting the concentration of professional talents; the closing rate is positively correlated with the number of employees with college degrees or higher, and the number of cases filed per 10,000 people is negatively correlated with the number of employees with college degrees or higher. This is consistent with the preset conclusion of the study. In this regression, it is also found that the regional R&D level has a significant positive effect on the agglomeration of professional talents. In the regression, the coefficient of patent licensing is positive, indicating that the index is positively correlated with the explained variables, and it also verifies hypothesis 2 in the theoretical model. At the same time, it also verifies the positive promotional effect of public financial expenditure on talent agglomeration (Lnss coefficient is positive and significant), indicating that a good social contract relationship is conducive to talent agglomeration. In this study, the robustness of the regression is verified and passes the robustness test. This study also conducts an empirical analysis on the number of employments requiring a degree or higher, and the conclusion is consistent with the results in Table 4 below.

Conclusion and policy implications
The spatial agglomeration of the population (labor force) in China continues to be a trend of economic and social development, and the characteristics and paths of this agglomeration are also changing. The surplus rural labor force is almost exhausted, and the labor force has shifted from rural to urban concentration to the transferring of labor between cities; the reason for labor migration has shifted from direct economic motivation to comprehensive factors-direct economic motivation and indirect economic motivation. The institutional environment factors related to labor employment play a significant role in labor agglomeration. Based on the empirical analysis of provincial panel data from 2001 to 2017, this study draws the following conclusions: first, establishing a good social contract relationship with workers and providing good public services will promote labor agglomeration; second, handling labor disputes properly, building harmonious labor relations, and improving the labor system environment will promote labor agglomeration; and third, promoting the ability of science and technology to innovate in a region will promote labor agglomeration, promoting the clustering of professional and technical personnel.
The key policy implications of this study include the following: first, the local arrangements should be made more convenient pertaining to children's enrollment, accommodation, and transportation arrangements, to ensure the stability of enterprise employment and encourage the growth of migrant labor agglomeration and regional economic and social development. Third, a sound labor dispute settlement system must be developed and a harmonious labor relationship created. We should encourage all government departments to effectively collaborate and cooperate, set up an early warning system for labor disputes, settle labor disputes promptly, and minimize the number of labor disputes. We should strengthen the labor dispute mediation system, actively facilitate the reconciliation of all sides of labor disputes, and improve the effectiveness of labor dispute resolution. We should also improve the mediation system for labor disputes.