|
In Discussion Paper No. 1124, Research Affiliate Margaret Stevens examines the foundations of the prediction that the costs and returns to investment in specific human capital will be shared between workers and firms, and hence that in the presence of specific human capital there will be a positive relationship between wages and tenure. In fact, the standard model does not predict such a relationship. The paper suggests a more precise definition of sharing, using two models in which sharing does arise, in response to a problem of asymmetric information . In the `whimsical worker' model, workers develop preferences between firms, which affect their responses to relative wages. Firms make wage offers in the second period, and workers choose which to accept. In this model, if information were symmetric, there would be no need for the training firm to share the return to the investment. But if firms are unable to observe workers' preferences, they will choose to offer a wage which is strictly higher than the wage offers of other firms. Thus, sharing arises as the optimal response of firms to information asymmetry.In the `firm-specific shocks' model, firms make wage offers to workers in the same way, but the source of asymmetric information is different: firms are subject to independent shocks which affect their valuation of workers, and the actual valuations are known only to the individual firms at the time when the offers are made. Again, the training firm responds by making a wage offer which is strictly higher in expected value than the offers of other firms, choosing to share the return to specific human capital. These models provide a foundation for the sharing hypothesis, and demonstrate conditions under which specific human capital leads to a positive wage-tenure relationship. They also demonstrate that sharing does not require the assumption of a predetermined wage contract. The Sharing Hypothesis for Specific Human Capital Margaret Stevens Discussion Paper No. 1124, December 1994 (HR) |