What Do Unions Do?

A Twenty-year Perspective

James T. Bennett, Bruce E. Kaufman

• Publisher: Routledge
• Year: 2017
• eBook ISBN: 9781351299466
• Topic: Business
• Subtopic: Business General
• Index: Business

One of the best-known and most-quoted books ever written on labor unions is What Do Unions Do? by Richard Freeman and James Medoff. Published in 1984, the book proved to be a landmark because it provided the most comprehensive and statistically sophisticated empirical portrait of the economic and socio-political effects of unions, and a provocative conclusion that unions are on balance beneficial for the economy and society.The present volume represents a twentieth-anniversary retrospective and evaluation of What Do Unions Do? The objectives are threefold: to evaluate and critique the theory, evidence, and conclusions of Freeman and Medoff; to provide a comprehensive update of the theoretical and empirical literature on unions since the publication of their book; and to offer a balanced assessment and critique of the effects of unions on the economy and society. Toward this end, internationally recognized representatives of labor and management cover the gamut of subjects related to unions.Topics covered include the economic theory of unions; the history of economic thought on unions; the effect of unions on wages, benefits, capital investment, productivity, income inequality, dispute resolution, and job satisfaction; the performance of unions in an international perspective; the reasons for the decline of unions; and the future of unions. The volume concludes with a chapter by Richard Freeman in which he assesses the arguments and evidence presented in the other chapters and presents his evaluation of how What Do Unions Do? stands up in the light of twenty years of additional experience and research. This highly readable volume is a state-of-the-art survey by internationally recognized experts on the effects and future of labor unions. It will be the benchmark for years to come.

1

What Do Unions Do? A Twenty-Year Perspective

James T. Bennett and Bruce E. Kaufman

I. Introduction

Labor unions are at least as old as industrial economies and have been the subject of debate and controversy for just as long. All sides agree that the objective of unions is to advance the interests of their worker members, and toward this end they exert pressure on employers and governments for improved terms and conditions of employment. Controversy quickly flares, however, once the discussion moves beyond this point. Two issues, in particular, occupy center stage. The first is one of fact, and concerns the effects of unions. Central questions include: how do unions affect wages?, firm performance?, labor market efficiency?, workers’ welfare?, and the political process? The second issue involves evaluation and judgment: Are unions good or bad for workers, firms, and the economy? Are the methods unions use to achieve their objectives consistent or inconsistent with widely shared ethical and legal principles? And, would social welfare be advanced by encouraging or discouraging additional unionism?

Twenty years ago this debate was taken to a new level with the publication of What Do Unions Do? by Richard Freeman and James Medoff. In the first paragraph of their book, they observe, “For over 200 years, since the days of Adam Smith, economists and other social scientists, labor unionists, and businessmen and women have debated the social effects of unionism. Despite the long debate, however, no agreed-upon answer has emerged to the question: What do unions do?”

In the remainder of the book they attempt to fill in the gaps and holes in our knowledge of “what unions do.” Their comprehensive approach brought to the subject new theory, an impressive collection of data sets, and sophisticated statistical tools. Not only did Freeman and Medoff (F&M) cover oft-considered subjects, such as the union effect on wages, but also a variety of new or seldom-treated topics, such as how unions affect employee benefits, turnover, productivity, and firm profitability. Going further, they also gave attention to a variety of noneconomic dimensions not often addressed by economists, such as democracy in unions and the union role in dispute resolution and work force governance.

II. Theory

F&M contend that economists and the general public hold two quite different conceptions of unions, one is negative and the other is positive. The negative view they call the “monopoly face” of unions, the positive they call the “collective voice/institutional response” face.

They introduce the monopoly face with the statement (p. 6), “Most, if not all, unions have monopoly power, which they can use to raise wages above competitive levels. Assuming that the competitive system works perfectly, these wage increases have harmful economic effects, reducing the national output and distorting the distribution of income.” F&M do not analytically elaborate the monopoly face of unions, but use the concept as a convenient metaphor for the general idea that unions — like monopolistic firms — possess market power and use it to raise the price of labor above the market level. The artificially high price of labor, in conjunction with restrictive work rules and strikes, leads to a number of undesirable outcomes: resource misallocation, inefficiency in production, a bias toward cost-push inflation, and lower capital investment and productivity growth. Unions thus appear in their monopoly guise as a special interest group that uses economic and political “muscle” to gain more income for their members (and leaders) at the expense of the welfare of the larger community.

In constructing the voice/response face of unions, F&M draw on ideas from the public choice area of economics and Hirschman’s influential book, Exit, Voice, and Loyalty (1970). They contend workers have two ways of dealing with workplace problems: the “exit” option of quitting and the “voice” option of speaking up to management. As F&M note, in a perfectly competitive labor market worker exit and entry among firms produces a situation in which no individual can be made better off without making someone else worse off. Firms are also operating on their efficiency frontiers, so no organizational slack exists. If, however, markets have significant imperfections, such as limited/asymmetric information or mobility costs, the exit option of turnover suffers from two drawbacks: Turnover imposes positive cost on workers and firms and fails to achieve the optimal configuration of working conditions and production methods. Firms may also not minimize costs. Efficiency, therefore, may be promoted by relying on voice as a way to solve workplace problems.

But, argue F&M, voice is also likely to be undersupplied relative to the social optimum if it only takes the form of individual communication between workers and managers due to public good (free-rider) problems and workers’ fear of being fired. Effective voice, therefore, has to be collective voice, such as a labor union, since only then will workers have the incentive to express their true preferences and the assur-ance that they will not suffer for it. Typically, F&M conceptualize union voice in the relatively narrow sense of a form of communication and information flow and assume it is largely separable from the exercise of union bargaining power. In places (e.g., p. 11), however, the concept is broadened to subsume a political process of rule making, workplace governance, and rights enforcement and couched as a limitation on the employer’s power.

According to F&M, the traditional view of unions is largely negative because the assumption of “perfect markets” leaves no room for union voice to play a beneficial role. Illustratively, they state (p. 12), “The greater the imperfection of markets, and the further the real-world management is from a computer programmed by the Invisible Hand, the greater are the possibilities for management’s response to unions to improve the operation of the economy.” In the presence of market imperfections, however, union voice can promote efficiency in a number of ways, they claim, such as by reducing turnover cost, improving working conditions and production methods, enhancing specific on-the-job training, improving decision making, and by fostering a greater sense of cooperation and trust. Union voice, suggest F&M, can also promote efficiency by inducing/helping management to improve coordination, reduce slack, and implement human resource policies more fairly and professionally.

Another consequence of collective voice, note F&M, is that it leads to a fundamentally different labor contract between the firm and employees. In a regime of individual bargaining, the employer’s attention is on keeping/hiring the “marginal” worker — the one just on the margin of going to another firm, while it may ignore or take for granted the preferences of the inframarginal workers who are locked into their jobs by seniority, fringe benefits, or other such factors. Since the marginal worker is more likely to be young, single, and mobile, while the inframarginal workers are more likely to be older, have families, and tied to the community, the preferences of the two groups regarding the mix of wages, benefits, and working conditions may be substantially different. With individual bargaining, the employment package will be structured to appeal to the marginal worker; with collective bargaining the union through a median voter process will articulate the preferences of the “average” employee. Again, the outcome under collective bargaining, by moving the labor contract toward what is desired by a majority of the work force, may promote greater efficiency and employee satisfaction in the labor market.

The bulk of F&M’s arguments focus on the efficiency effects of unionism. At points, however, they broaden the analysis to include the effect of unions on other social outcomes, such as income inequality, democracy in the workplace, and representation of workers in the nation’s political process. F&M claim that the conventional (monopoly) view of unions also leads to a negative assessment of the contribution of unions in these areas. When unions raise wages, for example, they (allegedly) widen income inequality among workers, while it is claimed that unions themselves are often nondemocratic or corrupt and distort the political process through sizable lobbying and campaign contributions. Acknowledging that all of these criticisms have an element of truth, F&M also claim that a voice/response theory of unions leads to a much more positive picture and one more consistent with the overall performance of unions. The democratic nature of unions, for example, leads to wage-leveling policies that reduce income inequality. Unions also protect employees’ rights in the workplace and counterbalance corporate power in the political arena.

The final component of F&M’s theory is management’s role. The negative monopoly effects of unions, they argue, are real but also potentially more than offset by the positive effects of the voice/response face. The latter, however, are only realized when management takes a constructive approach to collective bargaining, and the parties deal with each other in a spirit of cooperation and mutual gain. The paradox, say F&M, is that while unionism is on net neutral-to-positive for productivity, efficiency, and social outcomes, it nonetheless reduces profits and thus engenders strong employer resistance.

The picture they paint, therefore, is one of missed opportunities — many workers without unions want representation, unions on net are good for the economy and society, and yet management’s resistance to unions — coupled with weak labor laws — has been the central force behind a steady erosion in private sector union density from more than one-third in the 1960s to less than one-tenth four decades later. In contrast, union density in the public sector has remained steady at roughly forty percent for several decades, which they hold up as a model for what private sector density might resemble were it not for management resistance.

III. Empirical Evidence

The two contrasting views of unions give markedly different pictures of unionism’s economic and social repercussions — the monopoly face predicts social loss, the voice/response face social gain. The critical issue, then, is which model better describes the effects of labor unions.

On a conceptual level, Freeman and Medoff state (p. 19) that both models are relevant: “Since, in fact, unions have both a monopoly and a voice/response face, the key question for understanding the impact of private sector unionism in the United States relates to the relative importance of each.” F&M are thus saying that theory alone cannot decide the issue, since both models describe a portion of reality, so the ultimate determination has to come from a weighing and sifting of the empirical evidence. As F&M put it (p. 246), “The central question is not, ‘Who in principle is right?’ but rather, ‘Which face is quantitatively more important in particular economic outcomes?’”

Having proposed the crucial test, F&M devote the remainder of What Do Unions Do? to a chapter-by-chapter empirical examination of the effects of unions on a wide variety of economic and social processes and outcomes. Topics included are: wages, fringe benefits, wage inequality, employee turnover and job tenure, employment adjustment to business cycles, the role of seniority, employee job satisfaction, impact on nonorganized workers, productivity, profits, political power, union governance, and union organizing.

What do they find? F&M provide this summary statement (pp. 19–20),

Although additional study will certainly alter some of the specifics, we believe that the results of our analysis provide a reasonably clear and accurate picture of what unions do — a picture that stands in sharp contrast to the negative view that unions do little more than win monopoly wage gains for their members. Our most far-reaching conclusion is that, in addition, to well-advertised effects on wages, unions alter nearly every other measurable aspect of the operation of workplaces and enterprises…. On balance, unionization appears to improve rather than harm the social and economic system. In terms of the three outcomes in Table 1.1, our analysis shows that unions are associated with greater efficiency in most settings, reduce overall earnings inequality, and contribute to, rather than detract from, economic and political freedom. This is not to deny the negative monopoly effects of unions. They exist. They are undesirable. But … our analysis indicates that, in fact, focusing on them leads to an exceedingly inaccurate representation of what unions do. In the United States in the period we have studied, the voice/response face of unions dominates the monopoly face, though we stress that an accurate portrait must show both faces.

After this statement, F&M list thirteen specific empirical findings from their analysis. Put briefly, they are:

• Unions have a substantial monopoly wage impact, although the exact size of the union-nonunion differential varies considerably over time and across industries and demographic groups. The social cost of union monopoly wage gains is modest — about 0.3 percent of gross domestic product.

• Unions also increase the share of compensation going to employee benefits and shift the composition of the benefits package toward deferred benefits, such as pensions and health insurance.

• Unions reduce the overall level of wage inequality among workers.

• Unionized workers are less likely to quit their jobs and have longer job tenure than similar nonunion workers.

• During recessions unionized firms make more use of temporary lay-offs, rather than wage cuts. On cyclical upswings unionized firms recall relatively more workers, and nonunion firms tend to hire new employees. Unions sometimes agree to wage cuts, but generally only in the face of a substantial threat to employment.

• Unionized workplaces operate under different and more explicit rules than nonunion workplaces. Seniority is more important, dispute resolution more formal, job security protections greater, and management discretion and flexibility restricted.

• The “threat effect” of unions causes nonunion firms, on net, to modestly improve wages and employment conditions of production workers in order to avoid becoming unionized.

• Union workers report greater dissatisfaction with their jobs, particularly with respect to working conditions and relations with supervisors.

• In most cases unionized establishments have higher productivity than nonunion ones.

• Unionized firms earn a lower return on capital and reduced profits. The reduction in profitability is concentrated, however, among firms in concentrated and otherwise highly profitable sectors of the economy (implying union wage gains largely take the form of a redistribution of monopoly rents from firms to workers).

• The political power of unions has contributed to the enactment of numerous pieces of legislation that promote the general social/employment interests of workers. With respect to labor laws that directly impact union power, unions have been able to preserve existing favorable legislation but not expand on it.

• The popular image of unions as boss-run, corrupt institutions is a caricature. Most unions are highly democratic and corruption problems are concentrated in only a few industries.

• The percent of private sector workers belonging to unions has declined sharply since 1950. The decline is due largely to a dramatic increase in the amount and sophistication of legal and illegal company actions designed to forestall union organizing.

IV. National Labor Policy

With these empirical findings in hand, F&M in the last chapter of What Do Unions Do? turn to consideration of the implications for national labor policy, particularly as it bears on union organizing and collective bargaining.

Based on their empirical work, they conclude (p. 247) “in most settings the positive elements of the voice/response face of unions offset or dominate the negative elements of the monopoly face…. In an economy where governments, business, and unions work imperfectly — sometimes for, sometimes against the general welfare — there is a place for unions to improve the well-being not only of their members but of the entire society.” F&M then ask (p. 248), “Should someone who favors, as we do, a thriving market economy, also favor a strong union movement and be concerned with the ongoing decline in private sector unionism?” The answer they give is: “Accord-ing to our research findings, yes.”

This answer leads F&M to the “bottom line” regarding the existing level of unionism. They assert (p. 248), “The paradox of American unionism is that it is at one and the same time a plus on the overall social balance sheet … and a minus on the corporate balance sheet.” Because unionism hurts profits and employer control of the workplace, companies expend great time ...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Acknowledgments
• 1. What Do Unions Do? A Twenty-Year Perspective
• 2. What Unions Do: Insights from Economic Theory
• 3. Historical Insights: The Early Institutionalists on Trade Unionism and Labor Policy
• 4. What Effect Do Unions Have on Wages Now and Would Freeman and Medoff be Surprised?
• 5. Unions and Wage Inequality
• 6. The Effect of Unions on Employee Benefits and Non-Wage Compensation: Monopoly Power, Collective Voice, and Facilitation
• 7. What Do Unions Do for Economic Performance?
• 8. Union Voice
• 9. What Do Unions Do to the Workplace? Union Effects on Management and HRM Policies
• 10. Unionism and Employment Conflict Resolution: Rethinking Collective Voice and Its Consequences
• 11. The Impact of Unions on Job Satisfaction, Organizational Commitment, and Turnover
• 12. De-Unionization and Macro Performance: What Freeman and Medoff Didn’t Do
• 13. Two Faces of Union Voice in the Public Sector
• 14. Unionism Viewed Internationally
• 15. Has Management Strangled U.S. Unions?
• 16. Organized Labor’s Political Scorecard
• 17. What Do Unions Do? Evaluation and Commentary
• 18. What Do Unions Do? A Management Perspective
• 19. What Do Unions Do? A Unionist’s Perspective
• 20. What Do Unions Do? The 2004 M-Brane Stringtwister Edition
• About the Editors and Contributors
• Index