INTER-INDUSTRY WAGE DIFFERENTIALS: Sector Model 6


After controlling for this bias they find that male wages increase as a function of hours first in an increasing then at a decreasing rate. Less in line with our work are papers on the intertemporal behavior of hours and earnings such as Bernanke(1986) and Abowd and Card( 1987,1989). These papers are concerned again with the supply side. Abowd and Card consider how individuals alter their hours of work over time. Bernanke is interested in how earnings and hours varied in eight industries during the great depression.

The efficiency wage literature has some elements of similarity with our approach, but the differences are substantial and important. Our key variable is the capital-intensity of the task – the greater the capital-intensity the greater is the effort exerted by the worker. Most of the efficiency wage theoiy initiated by Shapiro and Stiglitz( 1984), collected in Akerlof and Yellen(I986) and surveyed in Katz(1986) makes no reference at all to the capital intensity of the operation. These efficiency wage models are all based on the idea that firms can increase profits by raising wages above the market clearing price. These above-market wages reduce monitoring costs since workers are induced to provide high effort by the threat of termination and thus a wage reduction.

The efficiency wage conceptual framework is very different from ours. After controlling for ability, we have workers preferring their own job or indifferent between their wage-effort contract and those contracts available to them in other jobs. The efficiency wage theory, on the other hand, suggests that high wages reflect worker rents needed to coerce workers in the good jobs not to shirk. According to the efficiency wage theory, workers in low-wage jobs prefer and are able to do the high-wage work, but are prevented from bidding for the better jobs in order to make the threat of firing have force in the high-wage contracts. Another important difference is that we have high-effort jobs in capital-intensive sectors and low-effort jobs in labor-intensive sectors. In the efficiency wage literature, worker effort need not vary across sectors. There can be a high-effort low-wage perfectly monitored job and a high-effort high-wage imperfectly monitored job.

Empirically the efficiency wage literature includes one of our two fundamental equations but not the other. We explain wages and hours both as a function of capital intensity. The first equation is part of the efficiency wage tradition, the second is not. Indeed, the efficiency wage literature was partly instigated by observation of the substantial differences in wages across industries. The correlation of the industry premiums with capital intensity was noted at least as far back as Slichter(1950). This has been empirically investigated in recent years in papers by Dickens and Katz( 1987a), Katz and Summers(1989), and Krueger and Summers(1987). payday loans online

The inter-industry wage pattern has been shown to be steady over time and remarkably consistent across different countries. The wage premiums in the capital intensive sectors have been attributed to variety of potential reasons including higher costs of monitoring, more inelastic labor demand and higher cost of worker shirking (close in spirit to our own work). It has been noted that disentangling these various effects can be difficult because of the simultaneity problem; since wages and capital-intensity are considered to be jointly determined.