Learning in the labor market

This thesis is a collection of three independent essays that study the implication of learning on labor mobility, labor supply, wage distribution, wage dynamics, and allocations of workers under different assumptions about the nature of employer learning. The first essay develops a model of job mobility and wage dispersion under the assumption that the current employers have superior information about their workers over outside firms. The superior information of the workers does not lead to market collapse. Instead, there is a unique mixed strategy equilibrium which leads to a positive amount of turnover and a nondegenerate wage distribution. This model implies that a skill-biased technology change that also favors general skill can lead to increase both in job mobility and wage dispersion. This sheds light, on the joint evolution of job mobility and wage dispersion in the U.S. in the past 30 years. The second essay studies the wage distribution and wage dynamics under matching and symmetric Pareto learning. I develop a model that contains pure learning and pure matching as limiting cases. In addition, the model generates effects that arise from the interaction of learning and matching. In particular, the model generates an earning profile typically obtained in a Mincerian regression.