Knowledge that Transforms

The flood of veterans enrolling in college at the end of World War II contributed to widespread rhetoric that the G.I. Bill brought about the "democratization" of American higher education. Whether military service, combined with educational benefits, led World War II veterans to increase their investments in college has received little research attention. Our estimation strategy focuses on between-cohort differences in military service, and we use census data to compare the collegiate attainment of veterans and nonveterans. The net effect of military service and G.I. benefits was substantial gains in the collegiate attainment of World War II veterans.

We study optimal incentives for a two‐stage production process. First, we identify conditions under which the optimal incentive structure for both workers is based on total volume and conditions for when the final stage worker’s incentives are based on relative performance. We show that a bottleneck‐free and balanced line is optimal only when both workers’ contracts are based on volume, while it becomes desirable to limit the final stage worker’s productivity when his compensation is based on relative performance. Thus, we demonstrate that the benefit of removing bottlenecks hinges critically on the structure of the optimal incentive arrangement.

We present a model in which unemployed workers simultaneously sample n potential employers. By varying n, we nest search and Walrasian‐type models of the labor market. We show that low values of n yield typical search equilibria: the wages are dispersed below the marginal productivity of labor. Interestingly, as n exceeds a relatively small threshold, the Walrasian‐type equilibrium emerges with the competitive wage quoted by all firms. For intermediate values of n, the equilibrium is a hybrid of the Walrasian and search equilibria. The model generates wage rigidity and yields novel predictions regarding the comovement of wages, firm turnover, and unemployment.

The negative effect of quits on the willingness of firms to provide on‐the‐job training is well documented in the theoretical literature. Here we explore the strength of this effect by solving a firm’s dynamic optimization problem where there is uncertainty about future productivity and nonzero firing costs. We find that the degree to which quit rates affect hiring depends on the ratio of firing to hiring costs. As this ratio rises, the negative effect of quits becomes less important, eventually reversing itself. We also describe how quit rates affect the firing decision. We highlight some testable implications of our analysis.

A substantial literature has developed to estimate the “true” cyclicality of real wages, that is, composition bias free. Two major issues are addressed in this article: aggregation of heterogeneous workers and potential bias in the measurement of the labor input. A general analysis of the biases is presented, and alternative approaches in the literature are nested in a single framework. Estimates based on an efficiency units concept that avoids the usual aggregation problems are presented. Composition bias underestimates the usual parameters of interest unless both the price and the quantity of the labor input are adjusted appropriately.

While much empirical research exists on labor market consequences of unemployment benefits, there is remarkably little evidence on the forces determining benefits. We present a simple model where workers desire insurance against unemployment risk and benefits increase the unemployment rate. We then conduct one of the first empirical analyses of the determinants of the parameters of the benefit system. Using data for developed countries for 1971–89, controlling for year and country fixed effects and the government's political color, we find evidence that the level of benefits falls when the unemployment rate is high. This is consistent with Wright's tax effect.

This article presents evidence on the relationship between compensation ratios and spans of control within hierarchical organizations. We find that compensation ratios are lower than span of control at any position within the hierarchy, which is consistent with an elasticity of compensation to a number of subordinates lower than one. Managers’ human capital endowments determine a significant part of the salary differences throughout hierarchical levels, as predicted by models of talent allocation in hierarchies. Differences in the size of firms should be attributed more to differences in their number of hierarchical levels than to variations in the span of control.

This article develops a model in which quit rates, and thus the income distribution, depend on employee perceptions of the accuracy of employer assessments of individual productivity because these latter assessments affect wages. When employees believe that these assessments are accurate, income inequality tends to be high. The model can account for the negative correlation across some countries of inequality and the extent to which inequality is deemed to be excessive. It also fits the contrast in U.S. and French experiences concerning the tenure of highly educated workers with high wages relative to the tenure of lower‐paid workers.