Union wage premiums by gender and race: Evidence from PSID 1980–1992

We find that the overall union wage premium is relatively stable (ranging from 22.3 to 28.4 percent), but there seems to be a convergence of union wage premiums across different demographic groups between 1980 and 1992. Nonwhite men (whose premium ranges from 23.5 to 36.2 percent) show the largest gain, followed by white women (17.1 to 30.5 percent), white men (19 to 26.4 percent), and nonwhite women (10 to 20 percent). One explanation for this convergence of union wage premiums might be the “equalization hypothesis” associated with unions. This converging trend could have important implications for the future of unions. If union membership can explain a portion of the gender/racial wage gap, and if women/nonwhites can obtain, through union membership significant wage premia, increased female/nonwhite union participation in highly unionized sectors that offer high union wage gains could, in time, greatly decrease the gender/racial wage differential. This study was supported in part by National Science Foundation funds [OSR-9350540]. An earlier version of this paper was presented at the Southern Economic Association conference in New Orleans in 1995. We thank Emilia Lulcheva and Michael Lauze for their able research assistance and William Warren for his valuable editorial comments on an earlier draft of this paper. The usual caveat applies.     

Some empirical observations of relevance to the analysis of union wage determination

Six important empirical characteristics of the union sector need to be incorporated into future research on wage determination. These are 1) the extent of unionization, 2) the statistical correlates of unionization, 3) divergent trends in union and nonunion earnings, 4) union/nonunion wage differentials, 5) the determinants of union and nonunion wage change, and 6) wage imitation. Examination of these characteristics suggests the following about union wage determination. Union wages have advanced relative to nonunion since the mid 1950s, despite relative shrinkage of the union sector. Union wage changes show less sensitivity to business-cycle pressures than nonunion. Limited spheres of wage imitation surround certain major union negotiations. These observations can be fitted into recent analyses of wage determination under long-term employer employee relationships, and have relevance for anti-inflation policy. Research for this paper was undertaken while the author was a Senior Fellow at the Brookings Institution and was supported by a grant from the Alfred P. Sloan Foundation. Views expressed should not be attributed to the Brookings Institution, its staff or trustees.     

Right-to-work and union compensation structure

There are many reasons to expect that right-to-work legislation should affect unionism, independently of whether or not such legislation reflects the sentiments of the electorate. The strongest reason is that employees protected by right-to-work legislation can quit a union without quitting their job. This should make collective job actions more difficult and prompt local union leaders to strive more for consensus among members. If so, unions in right-to-work states should negotiate less pay for seniority than do unions in non-right-to-work states. PSID wage data generally confirm this prediction. The authors thank James Bennett, Art Blakemore, Dan Heldman, Barry Hirsch, Stuart Low, and an anonymous referee for comments on an earlier draft.     

Union wage differentials for covered members and nonmembers in Great Britain

I present the first longitudinal estimates of covered union member and covered nonmember wage differentials in Great Britain. Crosssectional estimates show that covered union members receive a premium of about 10 percent over other workers, but covered nonmembers have no significant wage differential. Longitudinal estimates that accounted for fixed effects, selectivity, and measurement error indicated that OLS estimates are downward biased. The “true” union wage differential is likely about 30–35 percent. There is a large negative selection effect to union membership. No robust estimate could be found for covered nonmembers. This paper has benefited from conversations with Martyn Andrews, Wiji Arulampalam, Alison Booth, Tim Hatton, John Hutton, George Jakubson, Costas Meghir, and Andrew Oswald. Alison Booth kindly made extensive comments on an earlier draft, which greatly improved the exposition. John Budd, Hank Farber, and Larry Katz directed me towards related work in the U.S. I am also grateful to seminar participants at the University of York University of Essex, and the CEPR Workshop on Labour Market Issues. Thanks to the ESRC Data Archive for supplying the Family Expenditure Survey (FES) data. All data, computer programs, and results alluded to in the text are available on request. The British Household Panel Study (BHPS) and FES data are available from the ESRC Data Archive at the University of Essex. The Panel Study of Manufacturing Establishments (PSME) data are available from the author on request.     

The influence of legal status on the labor market impact of immigration

This article explores US labor market changes that would take place as a result of an amnesty that would regularize the status of undocumented workers without changing the total size of the alien workforce. The theoretical analysis suggests that the influence of legal status on market wage rates and on minimum wage enforcement is weak and that to the extent that there is an effect, it depends on particular institutional arrangements. Although data are not adequate for a definite measurement of these effects, those data that are available support this conclusion. It appears that the presence of undocumented as opposed to resident aliens can weaken union organizing efforts.     

Reconsidering union wage effects: Surveying new evidence on an old topic

I examine evidence on private sector union wage gaps in the United States. The consensus opinion among labor economists of an average union premium of roughly 15 percent is called into question. Two forms of measurement error bias downward standard wage gap estimates. Match bias results from Census earnings imputation procedures that do not include union status as a match criterion. Downward bias is roughly equal to the proportion of workers with imputed earnings, currently about 30 percent. Misclassification of union status causes additional attenuation in union gap measures. This bias has worsened as private sector density has declined, since an increasing proportion of workers designated as union are instead nonunion workers. Corrections for misclassification and match bias lead to estimated union gaps substantially higher than standard estimates, but with less of a downward trend since the mid 1980s. Private sector union gaps corrected for these biases are estimated from the CPS for 1973–2001. The uncorrected estimate for 2001 is .13 log points. Correction for match bias increases the gap to .18 log points; further correction for misclassification bias, based on an assumed 2 percent error rate, increases the gap to .24. Reexamination of the skill-upgrading hypothesis leads to the conclusion that higher union gap estimates are plausible. The conventional wisdom of a 15 percent union wage premium warrants reexamination.     

Union and nonunion wages at the firm level: A combined institutional and econometric analysis

This paper presents an analysis of union-nonunion wage differences at the intrafirm level. At this level of analysis, modeling of the union effect can be informed by a historical overview of the organizational context of unionization, which is not as feasible at other levels of analysis. The findings suggest that a simultaneous equation (2-stage) model, in which union status is treated as an endogenous variable, is not appropriate in this case. Moreover, OLS estimations of the union effect on wages from a one-stage model in which union status is exogenous is found to be efficient and unbiased. The study suggests that combining institutional details of the organizational context with econometric analysis facilitates a better understanding of not only the magnitude of wage differences but also of how the differences may be interpreted in the context of industrial relations.     

Whatever happened to the monopoly theory of labor unions?

This paper argues that unions act in accord with the conventional cartel or monopoly model. The basic premise is that it is useful to ask what a “union maximizes” because if more wealth is available, union decision-makers have an incentive to capture it for themselves or their membership. In the formal model, unions negotiate wage rates which maximize the monetary surplus above the supply price of labor, providing an endogenous answer to the questions of how union employment and wages are simultaneously determined. Comparative static analysis yields empirical predictions about the behavior of union employment, wage rates, and union-nonunion wage differentials. I would like to acknowledge helpful comments by Richard Anderson, Ray Battalio, Hugh Macaulay, Michael Ormiston and Akira Takayama on earlier drafts of this paper. The usual caveat applies.     

Do union members receive compensating differentials? The case of employment security

This study presents evidence in the ongoing examination of whether or not union wage effects represent, in part, premiums to unpleasant aspects of unionized work. Three major empirical results follow: first, approximately one-sixth of the union differential can be attributed to the greater employment risk of union members; second, nearly one-half of the union return, and over one-third of the non-union return, to union density can simultaneously be attributed to employment risk; and, third, union members, on balance, receive larger premiums for employment risk. Such results extend previous work which argues that unions help reveal preferences about workplace public goods. In addition, they support those who contend that union density proxies other relevant omitted variables. The author thanks Randall Crane for reading an earlier draft, Mohanty Madhu for research assistance, and an anonymous referee for helpful suggestions.     

Unionism and voter turnout

Conventional wisdom suggests that union members and their families are more politically active, and more likely to vote, than nonmembers. This study presents, to our knowledge, the first systematic empirical examination of that conventional wisdom. Results suggest that union members are more likely than nonmembers to vote in a general election, and that union campaign efforts increase voter turnout generally. There is no evidence, however, that union family members are more likely to vote than nonmembers, or that union status affects an individual’s likelihood of voting in a primary election. The authors wish to thank Robert Perloff, Donna Sockell, and an anonymous referee for helpful comments on an earlier draft of this paper. This study was partially supported by a Columbia University Graduate School of Business Faculty Research Fellowship and a Faculty Research Grant from the University of Pittsburgh Graduate School of Business.     

Union wages, rents, and skills in health care labor markets

We examine the effect of unions on the earnings of health care workers, with emphasis on the measurement and sources of union wage premiums. Using data constructed from the 1973 though 1994 Current Population Surveys, standard union premium estimates are found to be substantially lower among workers in health care than in other sectors of the economy, and to be smaller among higher skill than among lower skill occupational groups. Longitudinal analysis of workers switching union status, which controls for worker-specific skills, indicates a small impact of unions on earnings within both high and low skilled health care occupations. Evidence is found for small, but significant, union threat effects in health care labor markets. It has been argued that recent legal changes in bargaining unit determination should enhance union organizing and bargaining power. Although we cannot rule this out, such effects are not readily apparent in our data. The authors appreciate the assistance of David Macpherson, who helped develop the CPS data files used in the paper.     

The effect of the extent of unionism on union and nonunion wages

This paper provides estimates, derived from micro wage equations, of the effects of unionism on the wages for both union and nonunion labor. These equations control not only for union status, but also include measures of the extent of unionism in product and labor markets. The results suggest,inter alia, that an increase in the extent of unionization in an industry has substantial positive effects on the wages of nonunion as well as union workers. Increases in the extent of union coverage within an occupation, however, have little or no effect on nonunion wages.     

WHY DO WORKERS JOIN UNIONS? THE IMPORTANCE OF RENT SEEKING

Union status models ignore the fact that rent-seeking prospective members have an incentive to bid up entry costs so that higher union wage gains make union jobs more costly to obtain. The standard presumption that higher union wages cause firms to substitute toward higher quality workers is shown to be incorrect under most plausible assumptions; the observed positive correlation between wage gains and the propensity to join a union underestimates the size of the true supply response. The union/nonunion wage differential reveals more about the social cost of unions than the gain to an individual worker from union membership.     

The Ross-Dunlop debate and union wage concessions: A median voter analysis

The Ross-Dunlop debate concerns the extent to which unions take into account the trade-off between wages and employment in formulating their wage demands. This paper develops a median voter model of union behavior that offers a new approach to resolving the Ross-Dunlop debate. The model shows that when the binding constraint on the median union member in the seniority distribution is the threat of layoff, the union will behave as a “Dunlop-type” union; when the binding constraint is the cost of striking, the union will behave as a “Ross-type” union. The model is then applied to the related issue of union wage concessions. Two questions are examined: Under what conditions will a union agree to wage concessions? How large a cut in wages will be accepted?     

Recent trends in the union wage premium

Estimates of union-nonunion wage differentials are updated by examining a wide variety of subgroups in the general population. Variations in union wage premiums are tracked over a 15-year period, allowing inter-temporal comparisons that are not usually available. One important finding is that union wage premiums have been generally rising since 1972–1973. It also appears that females and nonwhites no longer gain the huge payoffs from unionism that evidence has suggested. The impact of race and education on union wage premiums is weighed, and there is an examination of how levels of education, association with blue-collar or white-collar occupations, and employment in specific industries affect union premiums. The author thanks Professor Jack Skeels for invaluable comments and Northern Illinois University for financial support.     

Countercyclical union wage premium? Evidence for the 1980s

Empirical results based on pooled male data from the Panel Survey of Income Dynamics indicate an overall union wage premium of about 11.92 percent for the 1980s. In response to fluctuations in local labor market conditions, proxied by the local unemployment rate, a much more flexible wage-setting process is found in the nonunion sector relative to the union sector. The long-term effect of unemployment on nonunion real wages suggests an approximate 0.6 percent decline for every one percentage point increase in unemployment, a statistically significant reduction, but the long-term effect of unemployment on real wages of union members is negligible. The union wage premium ranges between 11.6 to 12.3 percent for the sample years. Even though union wages are insensitive to short-run fluctuations in local labor market conditions, and are somewhat countercyclical in nature, widespread union wage concessions which occurred during the 1980s may now be exerting a downward pressure on union wages. We acknowledge financial support of National Science Foundation [OSR-9350540], the Ada Howe Kent Research Fund, and The Fogelman Academic Research Excellence Fund. We thank Barbara Ganley for valuable editorial comments and Noga Peled for her able research assistance. The usual caveat applies.     

The union wage effect in Canadian manufacturing industries

Data for Canadian manufacturing industries, at the two-digit level, are used to examine the component elements of the union wage effect. The results show that absence of compulsory union membership for all employees in the bargaining unit served by a union does not significantly impair the ability of the union to negotiate wage gains. That is, our results imply that there is little reason for unions to devote much effort to negotiating the stronger forms of union security — union or closed shops. A second implication of our results is that significant bargaining advantages may accrue to unions with an international (U.S.) link, relative to Canadian national unions.     

The threat of unionism and wage-coverage effects

This study finds that the union/nonunion wage differential decreases with local labor-market coverage. In general, SMSA coverage has a negligible impact on union wages, and nonunion wages increase significantly with coverage. This is consistent with strong threat effects operating at the local labor-market level. As with most other wage-coverage studies, however, union wages increase more quickly with industry coverage than do nonunion wages. These results support the argument that distinctly different economic processes underlie local labormarket and industry-coverage effects. Economies in the provision of union services imply that union threat effects will be most salient at the local labor-market level. Industry wage-coverage relationships are dominated by the positive effect of product-market coverage on union bargaining power. Estimated coverage effects vary by major industry groupings and are sensitive to changes in the specification of the wage equations. The author acknowledges the helpful comments of an anonymous referee and the competent research assistance of Harold Leong.     

The twenty-percent majority: Pro-union bias in prevailing rate determinations

Each individual wage rate set by Davis-Bacon or by any similar state or local prevailing wage determination petrifies the outcome of competing views of how construction work should be staffed and paid on public works projects. Although presented with great precision (to tenths of a cent for both wages and fringe benefits), the level of wages themselves are of surprisingly little consequence: Those set at union levels soon rise, being automatically updated to new contracts and conditions; the rest fast become obsolete in any rising market, because surveys to update them are rare. But in addition to setting wage levels, determinations also delineate which jobs get to have rates set for them, and perhaps most critically, whether those delineated are identified as union or notunion. Whatever pattern is found may remain in effect for years or even decades, influencing which journeymen and laborers own what job tasks and who may perform what. Also, if a particular job happens to be set as union, it may bring with it dozens or even hundreds of related special job categories, grades of sub-groups, fine distinctions of fringe benefits, and complex divisions of geographic applicability based on local union jurisdictional areas. This study uses determinations recently made in Pennsylvania as an example to examine the mechanics of the wage-setting process. I find that, in addition to the endemic problems one might expect associated with a complex and partly judgmental process, every step of finding and setting prevailing rates includes overwhelming deference on the part of government towards union views and methods. It shows why unions representing less than 20 percent of the private construction work force consistently set the parameters controlling most of public construction. It ends with some suggestions on how better surveys and determinations could be made.     

Is Collective Bargaining Pareto Efficient? A Survey of the Literature

It would be difficult, even today, to argue that labour unions are not important economic institutions, and it is this importance that makes their consequences for efficiency so substantial. Interest in the economic analysis of unions was revived in the early 1980s, in large part by a paper by Ian McDonald and Robert Solow, which formalized ideas first expressed in the context of labour markets 35 years earlier by Wassily Leontief. The standard textbook model of the labour union treats the union as a conventional monopoly seller of labour, selecting the wage while the firm chooses the level of employment; McDonald & Solow, however, drew from Leontief’s work to suggest an alternative in which the firm and union negotiate to a Pareto efficient contract. Further theoretical work followed, and a still-growing empirical literature began to develop; a wide variety of empirical procedures and tests have been attempted, with a diverse and contradictory range of findings. Given the importance of the question of union contract efficiency, an up-to-date survey of the literature may be useful in synthesizing past results and pointing the way to future research, and it is this role which the current paper will attempt to fill.