(College of Public Administration, Nanjing University of Finance and Economics, Jiangsu Nanjing)
Abstract: This paper based on the CHIP2007 provided by the survey of income of residents in China, drawing lessons from and reforming the wage determination equation, estimating the net effect of training return rate by traditional regression method and quasi-experimental research of PSM. It is indicated that（1）The traditional regression method reveals that the rates of return on training, education and university education are 17%, 6.6% and 8.3%; （2）The model of Logit and average marginal effect show that the probability of urban labor force participating in training can be increased by 2% and 0.9% for every one year's increase in education and work experience, while unmarried and self-perceived workers in a healthy state negatively affect the training participation rate; （3）The PSM reveals that the range of return on training and university are [10.4%, 12.9%] and [33.8%, 36%], while the annual return on university education is about 8.45%~9%; （4）Although the economic value of on-the-job training is lower than the overall rate of return on University education, it is higher than the investment in education, Ignoring self-selection bias will overestimate the economic value of training.
Keywords: On-The-Job Training; Rate of Return; Human Capital; Educational Human Capital; Return on University Education