monopoly industries. At the same time, location, education, ownership, occupation type and whether having a second career contribute more and more to the income gap. The analysis based on the industry data indicated that, although there is an enlarging trend in wage difference between industries, an obvious convergence and interaction exists in the sub-sectors within the same industry. Using the 3-digit industrial industry data in 1985, 1995 and 2004, and the Spatial Econometrics Method, the research group did practical research on this interactive phenomenon in the similar industry and its possible mechanism. They found that with the market reforming and improvement of labor mobility, wage interaction always exists between similar industry, even the critical factors and their effects on wage in industrial industry changes. In 1985 and 1995, the "wage comparison" between state-owned enterprise and other enterprise of the same industry may play an important role in this wage interaction, but after 2004, indirect interaction between similar industries impacted more.
Based on the peasant survey data in 22 provinces, the research group further studied the impact of social networks on the wage level of migrant workers in urban labor markets, and found only weak evidence indicating that more social networks could help improve the wages level of migrant workers in the urban labor market. The main role of the social network in the urban labor market with high competitive is allocating of work, but does not directly change the labor market's equilibrium price level. It can only affect the migrant workers' occupation type so that to affect their wage level indirectly. Another study of the group shows that the social relations and the local registered residence do not characterize the factors of productivity. In the contrary, it is of more help for workers to enter the high-income industries in the eastern regions where the degree of marketization are higher.