The purpose of this paper is to provide empirical support for the theory of effort in a multi-sector model developed in Leamer( 1996). That theory is built on the familiar idea that a firm can contract with workers regarding both the wage level and also the working conditions. Those features of the labor contract that enhance productivity but are disliked by workers are called “effort” and the labor market thus offers a set of wage-effort contracts with higher wages offsetting higher effort. If “effort” does not affect capital depreciation, the high effort-high wage jobs occur in the capital-intensive sectors where the capital cost savings from high effort are greatest.

Among the implications of this theory are: Communities inhabited by industrious workers who are willing to exert high effort for high wages have high returns to capital. A minimum wage does not cause unemployment. It forces effort in the low effort-low wage contracts up enough to support the higher wage. These and many other aspects of the model of endogenous effort are discussed in Leamer(1997b). In this paper we focus on the following two implications:

• The capital savings from effort are greatest in the capital-intensive sector which is where the high-wage high-effort contracts occur.

• Price declines in the labor-intensive goods twist the wage-effort offer curve, lowering the compensation for low-effort work but increasing the marginal return for hard work. Thus increased competition with the emerging third world adversely affects the low-effort workers who find themselves having to work harder to maintain their living standards. High-effort workers may be made better off by the increased marginal compensation for effort in high-wage high-effort jobs in the capital-intensive sectors.

We find empirical support for both of these hypotheses. We show that there is a surprisingly clear relationship between effort, wages and capital intensity, with higher capital intensity of the sector associated with both higher wages and higher effort. We look for and we find a twisting of the wage-effort offer curve in the 1970s, which we (somewhat casually) associate with increased globalization.

The obvious and possibly insurmountable problem that we face is how to measure effort, the product of hours of operation multiplied by the “intensity” of use. We have no measurement of job intensity, which is not merely of the speed of operations but also the attentiveness and willingness to take risks and any other intangibles that raise productivity without increasing capital costs. Absent any obvious measure of job intensity, we take a first step in the direction of least resistance and use production workers’ annual hours as our indicator of effort. We are thereby acting as if intensity is not so negatively correlated with hours that there is no relationship between hours and effort.