CIAO DATE: 03/02


Critical Review

Critical Review

Winter–Spring 1998 (Vol.12 Nos.1–2)

The State and Labor in Modern America

By Lowell Gallaway and Richard Vedder *

Abstract

One of the distinctive developments of the postwar era in the United States has been the relative decline in the economic significance of labor unions. Melvyn Dubofsky offers the hypothesis that this has resulted from a shift in public policy that represents a return to pre-New Deal notions of the proper relationship between government and unions. But a much more likely explanation lies in the changing nature of American life. Increases in education and advances in transportation and communications have weakened greatly the traditional worker-versus-capitalist paradigm that is at the core of Dubofsky’s thinking. As this has happened, labor unions have increasingly become relics of a previous age.

Melvyn Dubofsky’s The State and Labor in Modern America (Chapel Hill: University of North Carolina Press, 1994) is very much in the left-wing mainstream of labor histories written over the past half century. It is highly readable, and his central premise is one that must, at least in part, be accepted: The rise and fall of organized labor organizations in America is closely associated with governmental policies and behavior. When government favors and promotes unions, they grow. When government shows hostility (or, we would argue, even neutrality) toward them, they tend to decline. Dubofsky is absolutely correct in proposing that the heyday of American trade unions (1930 to perhaps 1960) was, in general, a period when the federal government successfully used its coercive abilities to enhance the power and size of the organized American labor movement.

 

Workers vs. Capitalists

Dubofsky continues a tradition of labor historiography begun by John R. Commons (1918 and 1935) three generations ago. 1 In this tradition, there are two classes of people: “capitalists” or “employers” on the one hand, and “workers” or “laborers” on the other, with unions serving as the agents of the latter. Not until the next-to-last page of text (237) does Dubofsky acknowledge that most Americans identify themselves “neither with capital nor with the working class, preferring to see themselves as a protean ‘middle class.’” Much to Dubofsky’s chagrin, America, as Tocqueville observed more than a century and a half ago, is not a class-oriented society. Accordingly, even at the peak of union power, there were far more “workers” who labored outside unions than inside. To consider the AFL-CIO to be synonymous with working Americans, as Dubofsky does continuously throughout this book, misrepresents reality.

A good example of Dubofsky’s misuse of the word “worker” comes early (4): “By the 1890s...the strike had become the workers’ chosen weapon in their ongoing struggle against capital.” In reality, most “workers” in the Gay Nineties were not “struggling with capital.” Even late in the decade of Grover Cleveland and William McKinley, union membership was well under 3 percent of total employment, and under 5 percent of the number of nonagricultural workers (U. S. Department of Commerce 1975, 178). In 1894, the year with the most widespread strike activity in the nineteenth century, substantially less than 5 percent of the working population engaged in a work stoppage (Perlman 1950, 136). It is hard to believe that if workers had been able to vote by secret ballot on collective bargaining, more than a small percentage actually would have chosen union representation.

Nor does Dubofsky come to grips with changes in American life that have made organized labor a collection of increasingly irrelevant, anachronistic organizations. 2 American individualism, structural changes in the U.S. economy (e. g., the trends toward service employment and greater female labor-force participation), and increased global competition have led to a precipitous decline in the proportion of the labor force in unions, something Dubofsky laments but says relatively little about—possibly because it is not consistent with his view that since unions are obviously beneficial for their members, any membership decline must be due to governmentally enforced or inspired acts designed to constrain labor.

Further, it does not seem to have occurred to Dubofsky that when something becomes more expensive, people may want to buy less of it. Thus, he overlooks the hypothesis that organized labor may have been too successful, pushing wages up to levels that encouraged the substitution of capital for labor and the migration of businesses to cheaper, nonunion locales, both within and outside the United States. Similarly, economist Charles Baird (1996) has argued that when the legislative environment becomes extremely pro-union, at some point the result is to reduce the proportion of workers in unions. Baird suggests that below some threshold, the gains to unions from a more favorable legal environment serve to enhance membership, but as membership grows, a point comes at which employers, responding to the higher costs associated with a unionized labor force, begin to fight the unions more fiercely in representation elections.

Dubofsky appears to accept dubious economic arguments for labor unions, based in part on the theory of monopsony. According to this theory, when a single powerful employer of labor services negotiates with a multitude of competing workers, it is likely that workers will be compensated less than would be the case in a purely competitive market. Under monopsony, government promotion of unions may be justified, enabling labor monopolies (unions) to slug it out with monopsonists (employers) in local labor markets.

The monopsonistic model may make sense where there are company towns with one large employer hiring unskilled workers who have few other work options because of high transportation costs. One might argue that those situations were moderately common in the late nineteenth and early twentieth centuries, although our own research suggests that as early as the 1830s, the monopsony power of employers in New England mill towns was modest at best (Vedder, Gallaway, and Klingaman 1978). In any case, monopsony is far rarer today. The proportion of Americans working for distant corporate behemoths has declined sharply. For example, the percentage of workers employed by Fortune 500 corporations today is half what it was a generation ago. The proportion of manual workers has declined from about half of all nonfarm employees around 1930 to about a quarter of them today. In 1900, when labor-management strife was high, only about 10 percent of workers had professional, technical, or managerial jobs; today, 30 percent do—far more than have factory jobs. More education; near-universal automobile transportation; and radio, television, and computer communication have made it much easier for employees to escape from anticompetitive labor markets. In short, workers do not need unions as much as once may have been the case.

In any case, the old capitalist-worker distinction, greatly overdrawn by labor historians such as Dubofsky, is now more inappropriate than ever. Millions of Americans are both workers and capitalists. By the early 1990s, over 11 million workers were in Employee Stock Ownership Plans, and a far greater number were indirect capitalists through their pension-fund investments. Membership in private pension plans grew from 2.7 million in 1935 to over 77 million in 1990.

Moreover, the disparities in wealth that gave rise to resentments and class thinking have declined over time. Compare the richest American early in this century, John D. Rockefeller, with his contemporary counterpart, Bill Gates. At the time the Standard Oil Trust was broken up in 1911, Rockefeller’s net worth was equal to about eight or nine days of output from the American economy. Bill Gates, the richest American today, has a net worth approaching two days of output—proportionally 75 to 78 percent less than Rockefeller’s. The taxes and regulations of the welfare state have had a leveling influence that negates much of the psychological raison d’être of trade unionism.

In addition, today’s billionaires do not run sweat shops to support conspicuous consumption. Microsoft’s employees are prosperous; some two thousand of them are millionaires in their own right, as a result of stock options and stock purchase plans. The late billionaire, Sam Walton, drove his own pickup truck to work, and super-billionaire Warren Buffet lives in a house in Omaha that is less impressive than the stables in some of the Vanderbilt mansions of a century ago. Unlike three generations past, the employees of today’s wealthy capitalists work eight hours a day, five days a week in air-conditioned comfort, with lots of vacation time to enjoy a plethora of material possessions, ranging from automobiles and boats to personal computers. For all these reasons, envy and resentment are not as strong as they were generations ago.

It is no wonder, then, that American “workers” increasingly accept the proposition that their best friend is the American (or, in some cases, Japanese) capitalist. When profits are high, job opportunities tend to abound, and consumption spending by employees is higher than when profits are low. The adversarial zero-sum society is increasingly being replaced with a milieu where employee-worker distinctions and interests are merged or blurred. What is good for General Motors, by and large, is good for America, and most Americans know it, including many members of the United Auto Workers.

 

Workers vs. Unions

Indeed, the evidence is mounting that the labor unions are worse than anachronisms—that their adversarial attitudes have spillover effects that hurt innocent third parties. If the United Auto Workers prices its members out of the Michigan labor market and mandates union rules that constrain productivity, the resulting unemployment hurts retailers and their employees in the Detroit suburbs, who make a living serving the auto workers. We have used econometric techniques to evaluate variations in the economic performance of the 50 American states (Vedder 1996; Vedder and Gallaway 1986). One variable included in our models measures the incidence of labor union membership in a state. Consistently, we have observed that economic growth is negatively correlated with union membership.

Simple, descriptive evidence, while not conclusive, makes the point lucidly. For example, we compared the economic growth experience from 1980 to 1995 of the 10 states with the highest proportion of union membership against that of the 10 states with the lowest proportion. 3 The ten states with low union membership had an average rate of economic growth about 30 percent higher than the ten high union-membership states (29.43 percent versus 22.69 percent). High union membership seems to mean low growth in real per-capita income. Even while they may prompt the replacement of unskilled labor by machines, unions scare away capital resources vital to expanding output and job opportunities for high-wage laborers. Still other evidence suggests that labor unions increase unemployment; other things equal, states with substantial unionization tend to have higher average or “natural” rates of unemployment (Vedder and Gallaway 1996).

Perhaps an even more devastating portrayal of the negative spillover effects of unions is found in statistics on human migration. If people tend to move away from a locale, it is a good sign that the overall quality of life in that area is relatively unappealing; in-migration would seem to indicate that an area has a high appeal to the populace. This is why East Germany built the Berlin Wall. During the early 1990s, more than 1.4 million native-born Americans moved away from the highly unionized states, while migrating in even larger numbers into the states with a low union presence. 4 It is hard to reconcile the traditional labor historian’s view that unions protect oppressed workers from capitalistic exploitation with the fact that in the early 1990s, one thousand Americans were moving into the states with the least union presence every day.

In telling his tale of the interaction between unions and government policy, Dubofsky makes many highly questionable empirical claims. Before the reader even gets to page one, he learns that “it is no coincidence that the precipitate decline of the American labor movement coincided with the presidency of Ronald Reagan” (xiii). The reality is a bit different. Union membership as a proportion of non-agricultural employment had already declined significantly before Reagan assumed office, dropping (for example) from 30.8 percent of employment in 1970 to 25.2 percent in 1980 (U. S. Department of Commerce 1982-83, 401). And the decline continued after the Reagan era. To say that it “coincided” with Reagan’s presidency is a misreading of the evidence.

There is a similar distortion in Dubofsky’s description of the New Deal era. He claims that economic recovery in 1933 resulted in large part from the National Industrial Recovery Act: “The economic recovery precipitated by the NRA’s promise of higher prices brought an increase in union membership” (116). In fact, the evidence is overwhelming that the NRA retarded recovery. Elsewhere, we have estimated that the national unemployment rate fell by 5 points (from about 28 to 23 percent) from March to July 1933—just before the NRA was implemented (Vedder and Gallaway 1997, 77). In the next two years, the rate fell only 2 more points; the recovery largely stalled, only to pick up again after the Supreme Court ruled the NRA unconstitutional in Schecter Poultry v. United States. 5 The high wages required by NRA industry codes priced labor out of the market.

Perhaps less egregiously, Dubofsky blames the 1937-38 downturn exclusively on President Roosevelt’s contractionary fiscal policy (138, 149). In doing so, he ignores Milton Friedman and Anna Schwartz’s argument that this downturn was caused by Federal Reserve monetary policy. Dubofsky also ignores the claim made by business leaders of the era that this recession-within-the-Depression was due to the crisis of business confidence inspired by the New Deal. Least of all does Dubofsky consider the role that may have been played by the double-digit percentage increase in factory wages following the unionization of the mass production industries in 1937. This increase led to a sharp reduction in the quantity of labor demanded. Contemporary observers argued that labor unions were pricing workers out of the market (see Weinstein 1980 and Vedder and Gallaway 1997). The unions’ newfound power resulted from the Wagner Act, which Dubofsky praises lavishly.

The 1937-38 experience appears to show that while unions can manipulate the supply of labor, they cannot repeal the law of demand. During most of 1936, steel industry employment had risen by over 30 percent, because real wages, adjusted for productivity growth, actually fell, making it more lucrative for steel companies to hire workers. From the fourth quarter of 1936 to the second quarter of 1938, however, wages were raised by 22 percent on average—and steel industry employment fell by more than one-half (New York Times 1938, Literary Digest 1937). This wage increase, and the resulting unemployment, reflects the United Steel Workers Union’s success in organizing the industry.

 

Toward a New Labor History

If, as Dubofsky assumes, labor unions protect the “working class” against greedy “capitalist” employers, hostile public policies must be responsible for the recent decline in strike and union activity:

Public policy and practice...resemble more the world of the late nineteenth century than that of the New Deal order, as politics and law once more privilege individualism and define militant collective action as unacceptable, “un-American” and illegal. (231)

In reality, however, recent public policy has been, at most, neutral toward unions, essentially maintaining the legal environment established in the 1930s and 1940s. When Caterpillar workers went on strike in the early 1990s, the government did nothing to aid either the company or the unions. When McDonnell Douglas machinists went on strike in 1996, the state of Missouri gave unemployment insurance benefits to the workers, a strongly pro-union move. Other than during the PATCO strike in 1981, no president for many decades has tried to crush a strike or discourage workers from unionizing. Indeed, President Clinton and Labor Secretary Robert Reich seemed to do everything possible to strengthen unions, even issuing an executive order of dubious legality forbidding government contractors from hiring replacement workers. George Bush, at best, halfheartedly enforced the Beck decision, which would have restricted the use of union dues for political purposes. During the administration of Ronald Reagan, Congress enacted the Worker Adjustment and Retraining Notification Act (1988), giving workers the right to prior notification of plant closings.

All this is merely an extension of the policies of the 1960s and 1970s. During those decades, Democratic presidents sympathetic to labor (Kennedy, Johnson, and Carter) were in power for a majority of the time, and the three Republican presidents (Eisenhower, Nixon, and Ford) were all moderates who did very little to change the legal or political environment in which organized labor operated or to curb the pro-union policies of such agencies as the National Labor Relations Board. Indeed, during the early Carter Administration, strong pro-labor legislation would have been enacted except for a Republican filibuster in the Senate. Yet during these decades, union membership fell significantly as a share of the labor force.

Dubofsky’s claim (228) that President Reagan’s 1981 firing of striking air traffic controllers was a brutal action reminiscent of nineteenth-century state repression of unions is certainly subject to debate. The controllers were government employees charged with promoting public safety who were expressly forbidden from striking by law, and while it is difficult to envision any president of the 1950s or 1960s doing what Reagan did, by his day Americans had lost much of their sympathy for a movement whose postwar leaders were known mainly for their corruption and their ostentatious conventions in South Florida pleasure spas.

PATCO aside, however, government action against organized labor in the contemporary United States has been distinctly muted. No major laws or judicial decisions have constrained labor. No president has used the armed forces or the courts to constrain union behavior, as was the case a century ago. Yet unions now claim the membership of a smaller proportion of the private-sector nonagricultural labor force than was the case in 1920, more than a decade before the passage of the Wagner Act. It is only a slight exaggeration to suggest that, absent government protection, unions today would be the declining remnants of a fairly vibrant type of early-twentieth century associationism, much like the Odd Fellows or the Rotarians.

 

References

Baird, Charles. 1996. “Union-Friendly Legislation and Union Density: A Laffer Curve Analysis.” Paper presented at Meetings of the Southern Economic Association, Washington, D. C., November 24.

Commons, John R., et al. 1918 and 1935. History of Labor in the United States. New York: Macmillan.

Gallaway, Lowell, and Richard Vedder. 1996. “Labor Laws: Then and Now.” Journal of Labor Research 17: 253–75.

Literary Digest. 1937. December 25: 21.

New York Times. 1938. January 27: 1.

Perlman, Selig. 1950. A History of Trade Unionism in the United States. New York: Augustus Kelley.

U. S. Department of Commerce. 1975. Historical Statistics of the United States, Colonial Times to 1970. Washington, D. C.: U. S. Government Printing Office.

U. S. Department of Commerce. 1982–83. Statistical Abstract of the United States. Washington, D. C.: U. S. Government Printing Office.

Vedder, Richard. 1996. State and Local Taxation and Economic Growth: Lessons for Federal Tax Reform. Washington, D. C.: Joint Economic Committee of Congress.

Vedder, Richard, and Lowell Gallaway. 1986. “Rent-Seeking, Distributional Coalitions, Taxes, Relative Prices and Economic Growth.” Public Choice 51: 93–100.

Vedder, Richard, and Lowell Gallaway. 1996. “Spatial Variations in U.S. Unemployment.” Journal of Labor Research 17: 445–61.

Vedder, Richard, and Lowell Gallaway. 1997. Out of Work: Unemployment and Government in Twentieth-Century America. New York: New York University Press.

Vedder, Richard, Lowell Gallaway, and David Klingaman. 1978. “Discrimination and Exploitation in Antebellum American Cotton Textile Manufacturing.” Research in Economic History 3: 217–62.

Weinstein, Michael. 1980. Recovery and Redistribution Under the NIRA. Amsterdam: North Holland.

 

Endotes

*: Lowell Gallaway and Richard Vedder, the Edwin L. and Ruth E. Kennedy Distinguished Professors of Economics, Ohio University, Athens, OH 45701, telephone (614) 593-2040, telefax (614) 593-0181, are the authors of Out of Work: Government and Unemployment in Twentieth-Century America (Holmes & Meier, 1993).  Back.

Note 1: The citation of the Commons work is complex. It was he who wrote the first two volumes, which were published in 1918. Two additional volumes dealing with the period 1896 to 1932 were published in 1935. The first was written by Don Lescohier and Elizabeth Brandeis and the second by Selig Perlman and Philip Taft.  Back.

Note 2: Throughout this review, we draw heavily on Gallaway and Vedder 1996.  Back.

Note 3: Economic growth was measured by the percentage growth in real personal income per capita. Union membership was measured by taking the average of the percentage of the labor force in labor unions early in the period (1983) and late in the period (1994). The high-union-membership states were New York, Hawaii, Michigan, Washington, New Jersey, Pennsylvania, Alaska, Oregon, Ohio, and Illinois. The low-union-membership states were South Carolina, North Carolina, Mississippi, Texas, Arkansas, Arizona, Florida, Georgia, Virginia, and South Dakota. The data are from the U. S. Department of Labor (union membership) and the U. S. Department of Commerce (income).  Back.

Note 4: Union membership data are for 1994. The high-union states, so defined, differ slightly from those in n3, which averages data for 1983 and 1984. The high-union states are New York, Hawaii, Michigan, Washington, New Jersey, Alaska, Illinois, Minnesota, Pennsylvania, and Indiana; the low-membership states are the two Carolinas, Mississippi, Texas, Florida, Virginia, Arizona, Arkansas, South Dakota, and Idaho. Migration data are from the U. S. Department of Commerce, Bureau of the Census.  Back.

Note 5: 295 U.S. 49; 79 L. Ed. 1570; 55 Sup. Ct. 837.1935.  Back.