This article examines wage determination among union and nonunion truck drivers using the 96 monthly Current Population Surveys for 1983-90. Union density in the previously regulated for-hire sector of the trucking industry fell from about 60% during the regulatory period of the 1970s to about 25% by 1990. Union log wage premiums fell from 0.40 in the 1970s to 0.30 or below in the 1980s. Longitudinal estimates from multiple panels for 1983-84 through 1989-90 suggest far smaller union premiums, supporting the thesis that part of the wage differential following deregulation is a compensating premium for driver quality.

This paper utilizes unique survey data on labor union coverage at the firm level to examine union effects on the profitability of 705 U.S. companies during the 1970s. Market value and earnings are estimated to be about 10 percent-15 percent lower in an average unionized company than in a nonunion company, following extensive control for firm and industry characteristics. Deleterious union effects on firm profitability are sizable throughout the 1972-80 period, but vary considerably across industries. The relatively poor profit performance of unionized companies may help explain the recent decline in U.S. union membership.

About 30% of workers in the Current Population Survey have earnings imputed. Wage gap estimates are biased toward zero when the attribute being studied (e.g., union status) is not a criterion used to match donors to nonrespondents. An expression for “match bias” is derived in which attenuation equals the sum of match error rates. Attenuation can be approximated by the proportion with imputed earnings. Union wage gap estimates with match bias removed are presented for 1973–2001. Estimates for recent years are biased downward 5 percentage points. Bias in gap estimates accompanying other non–match criteria (public sector, industry, etc.) is examined.

During the past several years, a number of studies have appeared in this Review and elsewhere examining the earnings and promotion of women faculty.' A finding common to most of these studies is that the female-male salary differential is smallest at entry level, but widens over time.2 Indeed, George Johnson and Frank Stafford state, after presenting such evidence for Michigan State University: believe this qualitative result would be observed for any university in the United States for which the sample size is sufficiently (1974, p. 899). While the qualitative findings of faculty salary studies have been similar, interpretations have differed. On the one hand, the human capital view attributes the widening sex differential, or flatter female earnings profile, to differences in acquired skill and productivity. This literature emphasizes differences in continuous labor market experience, hours of work, and the relative teaching/research division of labor. On the other hand, the discrimination view attributes the widening differential to increased labor market discrimination with respect to experience. The sex differential is smallest at entry level where universities must compete and pay prevailing salaries to attract incoming faculty members. However, discrimination is more easily exercised in the internal university labor market by male faculty and administrators as job mobility lessens with age (because of fixed costs, tied moves,3 university-specific job training, a shorter benefit span, etc). The strongest evidence supporting the discrimination view is the existence of significant unexplained salary differentials even where detailed data exist on research and teaching performance. The purpose of this note is to examine the salary structure at a large university which was historically the state's university for women, but is now fully coeducational. To our knowledge, no other such study is available. Comparison of results from such a study with those already in the literature can shed light on the interpretation of salary differential studies. We find, somewhat surprisingly, an exception to the finding by Johnson and Stafford and others of a widening femalemale salary differential with experience. Apart from some qualifications discussed below, we find a small differential at entry, but little difference in the reward structure to men and women with respect to experience, ceteris paribus. Alternative interpretations of this evidence are provided. While we cannot clearly test between the human capital and discrimination explanations for salary differentials, our evidence strongly suggests that universities can and do exercise significant discretion in the awarding of salaries, discretion which is presumably made possible by the not-for-profit nature of these institutions. (See Armen Alchian and Reuben Kessel.) Section I briefly describes the data source, while Section II examines specification and then tests for functional form. Section III *University of North Carolina-Greensboro. We thank Terry G. Seaks and Frank Stafford for helpful suggestions. We are equally responsible for the contents of the paper. 'A partial list includes George Johnson and Frank Stafford, with later comments by Steven Farber and by Myra Strober and Alice Quester; Nancy Gordon, Thomas Morton, and Ina Braden; Emily Hoffman; David Katz; Marianne Ferber and Jane Loeb; Ferber and Betty Kordick; Ferber, Loeb, and Helen Lowry; and James Koch and John Chizmar. 2 There is debate over whether the differential begins to narrow late during the working life, as argued by Johnson and Stafford. See Johnson and Stafford (1977) and Ferber, Loeb, and Lowry. 3Viewing mobility within a household context (see Jacob Mincer) the potential for discrimination against married women faculty may be higher than with male or unmarried female faculty.

The Box-Cox transformation is used to compare alternative functional forms of market value equations. Based on evidence from a panel of 480 publicly-traded U.S. manufacturing companies and two additional data sets used previously in the literature, the semilog form of a Tobin’s q equation is found to be strongly preferred to the commonly estimated linear form. We provide illustrations in which inferences can be affected by the choice of functional form. The authors thank Zvi Griliches, Hendrik Houthakker, and two anonymous referees for helpful discussion and suggestions, and Jerry Stevens for providing access to one of the data sets examined in Section III. Remaining errors are ours. A longer working paper version is available on request. 1

This article examines match bias arising from earnings imputation. Wage equation parameters are estimated from mixed samples of workers reporting and not reporting earnings, the latter assigned earnings of donors. Regressions including attributes not used as imputation match criteria (e.g., union) are severely biased. Match bias also arises with attributes used as match criteria but matched imperfectly. Imperfect matching on schooling (age) flattens earnings profiles within education (age) groups and creates jumps across groups. Assuming conditional missing at random, a general analytic expression correcting match bias is derived and compared to alternatives. Reweighting a respondent‐only sample proves an attractive approach.

Wages for black and white workers are substantially lower in occupations with a high density of black employees, following standard controls. Such correlations can exist absent discrimination or as a result of discrimination. In wage level equations, partial correlations fall sharply after controlling for occupational skills. Longitudinal estimates accounting for worker heterogeneity indicate little wage change associated with changes in racial composition. Results support a “quality sorting” rather than discrimination explanation, with racial density serving as an index of unmeasured skills. Discrimination reflected in racial wage gaps occurs within occupations or across occupations in a manner uncorrelated with racial composition.