ABSTRACT
The existence ofwage differentials acrossworkers in different demographic groups has been documented in many empirical studies. Three types of differences across workers and the reasons behind them have received a great deal of attention and are the focus of this chapter. First, estimates of wage differentials associated with age or experience are used to examine implications of human capital models of wage growth. Second, estimates of wage differentials associated with sex or race are used to test for wage discrimination. Third, estimates of wage differentials associated with marriage have been interpreted as reflecting productivity effects. Additional areas of inquiry include wage differentials by union status, education, and industry. The problemwith the traditional approach of estimating wage regressions to test
theories of wage determination is that, without independent measures of worker productivity, it is difficult to determine whether wage differentials associated with worker characteristics reflect productivity differentials or some other factor, such as discrimination. For example, with data only on wages and worker characteristics over the life cycle, it is difficult to distinguish human capital models of wage growth (such as Ben-Porath, 1967; Becker, 1975; Mincer, 1974) from incentivecompatible models of wage growth (Lazear, 1979) or forced-saving models of life-cycle wage profiles (Loewenstein and Sicherman, 1991; Frank and Hutchens, 1993). Typical wage regression results report positive coefficients on age, conditional on a variety of covariates, but these positive coefficients neither imply that older workers are more productive than younger ones, nor that wages rise faster than productivity. Similarly, without direct measures of the relative productivity of workers, discrimination by sex, race, or marital status cannot be established based on significant coefficients on sex, race, or marital status dummy variables in standard wage regressions, since the usual individual-level wage regression controls may not fully capture productivity differences (e.g. Becker, 1985). The major contribution of this chapter is to use a unique new data set that
combines data on individual workers with data on their employers to estimate relative marginal products for various groups of workers, which we then compare
with relative wages. This employer-employee data set, the Worker Establishment Characteristics Database (WECD), matches long-form respondents to the 1990 Decennial Census of Population to data on their employers from the Longitudinal Research Database (LRD). These data are a major improvement over previously available data sources because they combine detailed demographic information on workers in a sample of plants with information on plant-level inputs and outputs.1
We use these data to estimate production functions in which workers with different demographic characteristics have potentially different marginal products, thereby obtaining estimates of these relative marginal productivities. In addition, we explore numerous issues regarding the estimation of these production functions in an attempt to obtain reliable estimates of these productivity differentials. For the most part, we find that our estimates of marginal productivity appear relatively robust and reasonable, although, not surprisingly, they do change somewhat as we vary our specification and sample. Because we have information on plant labor costs, we also specify and esti-
mate plant-level earnings equations. Theseplant-level earnings equations represent the aggregation of individual-level earnings equations over workers employed in a plant, and hence are the plant-level counterparts to the individual-level wage regressions that motivate this research. By simultaneously estimating the production functions and earnings equations at the plant level, we can compare the relative marginal products and relative wages of workers distinguished by various demographic characteristics.2,3 Thus, the data and empirical framework we develop supply the independent productivity measures needed to draw more decisive conclusions on numerous topics regarding the determination of wages, including race and sex discrimination in wages, the causes of rising wages over the life cycle, and the returns to marriage.
