Sleep must be considered subject to choice and affected by the same economic variables that affect other uses of time. Using aggregated data for 12 countries, a cross-section of microeconomic data, and a panel of households, we demonstrate that increases in time spent in the labor market reduce sleep time. Each additional hour of market work reduces sleep by roughly 10 minutes (and waking nonmarket time by 50 minutes). The total time available for work and leisure is thus itself subject to choice. Interestingly too, otherwise identical women sleep significantly less than men (even though the average Woman sleeps slightly more). We develop a theory of the demand for sleep that differs from standard models by its assumption that sleep affects wages through its impact on labor-market productivity. Estimates of a system of demand equations demonstrate that higher wage rates reduce sleep time among men, an effect that is entirely offset by their positive effect on waking nonmarket time. Among Women the wage effect on waking nonmarket time is negative and small, but the effect on sleep is negative and quite large. These results, and the model they are based on, allow a more s subtle interpretation of standard results in the labor supply literature.
