Labor Supply under Real Wage Uncertainty: A New Look at the Intertemporal Substitution Hypothesis

The idea that timing may play crucial role in labor supply decisions has existed in several forms since Mincer's [12] seminal work on the labor force participation decisions of married women. Mincer suggested at that time that married women's labor supply decisions may be quite responsive to transitory variations in earnings. Lucas and Rapping [11] generalized the notion of timing by suggesting that for any individual, consumption of leisure and other goods in the current period is readily substitutable with consumption in other periods. In this framework the laborer is viewed as a capitalist and the to transfer one's [labor] supply from one market [period] to another . . . is recognized as an investment decision [11, 729]. This concept of intertemporal substitution has been controversial, as it has proven quite difficult to substantiate empirically. While the estimates obtained by Lucas and Rapping were favorable, the recent and much more elaborate tests provided by Altonji [1] and Clark and Summers [5] have failed to support the intertemporal substitution hypothesis. These studies have used models in which individuals base their current labor supply decisions on the magnitudes of current and expected variables. If current decisions can be shown to be responsive to expected levels of relevant variables then this indicates that individuals do in fact choose when to and when not to supply labor over the course of the business cycle, and the notion of intertemporal substitution is supported. The approach then, has been to estimate the net present value of the expected levels of wages and prices at each point in time and to base tests of the intertemporal substitution hypothesis on the performance of these expectations variables. Implicitly, these models assume that people live in an uncertain world and that they base their current decisions on what they judge to be the likely future values of important economic variables. A key element which has been missing from previous studies is consideration of how the level of uncertainty associated with peoples' expectations affects decisions made at each point in time. This paper attempts to fill that gap by focusing upon the effects of real wage uncertainty on labor supply decisions.