National Longitudinal Surveys: Discussion Paper. Employer Learning and the211 Signaling Value of Education

摘要

If profit maximizing firms have limited information about the general211u001eproductivity of new workers, they may choose to use easily observable 211u001echaracteristics such as years of education to 'statistically discriminate' among 211u001eworkers. The pure credential value of education will depend on how quickly firms 211u001elearn. To obtain information on employer learning, we work with a wage equation 211u001ethat contains both the interaction between experience and a hard to observe 211u001evariable that is positively related to productivity and the interaction between 211u001eexperience and a variable that firms can easily observe, such as years of 211u001eeducation. The time path of the coefficient on the unobservable productivity 211u001evariable provides information about the rate at which employers learn about 211u001eworker productivity. Using data from the NLSY we obtain preliminary estimates of 211u001ethe rate at which employers learn about worker quality and use these, along with 211u001esome strong auxiliary assumptions, to explore the empirical relevance of the 211u001eeducation screening hypothesis. We show that even if employers learn relatively 211u001eslowly about the productivity of new workers, the portion of the return to 211u001eeducation that could reflect signaling of ability is limited.