Commentary | Unions and Labor Standards

How Unions Can Help Restore the Middle Class

March 10, 2009

How Unions Can Help Restore the Middle Class

The following testimony on the importance of unions to restoring a solid middle class was given to the Senate Committee on Health, Education, Labor and Pensions on Tuesday, March 10, by Dr. Paula B. Voos, a professor in the School of Management and Labor Relations at Rutgers, and an EPI research associate. 

Download print-friendly PDF version

Thank you for the invitation to speak today. I am pleased to have the opportunity to consider the role unions can play in rebuilding the American middle class, a matter of utmost importance not only for ending the current economic downturn, but also for our nation in the longer term. As an economist, I have been studying the role unions play in our economy for some time and in 1993-94, I had the opportunity to serve on the Dunlop Commission in its consideration of how labor law should be modernized to serve the “Future of Worker Management Relations in the United States.”

There is now a substantial body of research evidence on the economic impact of U.S. unions. Unions typically:

– Raise the wages of the employees they represent;

– Increase the fringe benefits of those same employees, usually by a greater extent than they increase wages;

– Reduce income inequality within the represented firm, by reducing differentials between low-paid and high-paid employees, men and women, various racial/ethnic groups, younger and older employees, and so forth;

– Increase pay of nonunion workers in occupations and industries with substantial union presence as nonunion employers move closer to union standards;

– Reduce income inequality in the wider society by reducing inequality not only within and between represented firms, but also across entire industries as nonunion employers increase compensation to discourage unionization, all of which strengthens the middle class (Card, Lemieux, and Riddell, 2007).

– Reduce employee turnover by lessening the number of quits (voluntary separations); and

– Thus increase the retention of skilled employees, enhancing human capital and productivity in both the firm and the economy as a whole;

(See Freeman and Medoff, 1984; Bennett and Kaufman, 2007).


– Because they suffer less turnover, unionized employers have greater incentives for employee training and for high-skill, high commitment human resource policies, rather than low-skill, high-turnover or other “low road” approaches to human resources. Reduced turnover avoids costs to employers but also lessens society’s costs associated with unemployment, such as Food Stamps, uncompensated care and other social programs.

– Union-represented employees have been found to be more productive, on average. This is probably both due to the fact they have more work experience and due to greater employer investments in them and in physical capital (see Doucouliagos and Laroche 2003 for an overview of seventy-three statistically independent studies);

– The nature of the labor-management relationship is crucial in this regard: good union-management relationships are ones that foster high workforce productivity, but workplaces characterized by labor strife and worker resentment — whether union or non-union — do not (Belman, 1992).

– Union employees typically cannot be disciplined or discharged without a reason, termed “just cause.” This assurance of fair treatment is one reason union employees have greater “voice” than non-union employees and typically are more willing to make suggestions or speak up to improve business operations.

The most important reason to improve the ability of employees to organize into unions is that such membership is a fundamental right in democratic societies, related to freedom of association and the right of all human beings to band together to improve their lives. For that reason alone, I would urge you to pass legislation to make real in the U.S. once again the promise of the National Labor Relations Act. Section 1 of that Act puts federal law behind “the practice and procedure of collective bargaining and … the exercise by workers of the full freedom of association, self-organization, and designation of representatives of their own choosing.” (NLRA Sec. 1).

Nonetheless, some may be concerned with the economic consequences of increased unionization at this moment in time. They should be assured that the economic consequences would be positive. There are two main reasons:

– First, greater union membership would help the United States recover from the current economic downturn and help prevent future economic crises.

– And second, greater union membership would help the United States make the transition to competing internationally on the basis of high productivity, high quality, and innovation, rather than on the basis of low wage labor or long hours – a race to the bottom that we can never win against nations like China.

Let me explain.

The Economic Crisis and the Middle Class

The growth of income inequality in the U.S. and the related decline of the middle class are critical factors in the current economic crisis: the collapse in the housing market, the crisis of inadequate capital in the nation’s banking institutions, the decline in the stock market, the free-fall in consumer spending, declining employment and other aspects of the recession that are worsening daily.

In the early part of this decade, stagnating incomes for the bottom 80 percent of American families led many people to go into excessive debt to meet ordinary needs such as adequate housing — particularly in parts of the country like California in which housing prices and rents had soared. Many took on inappropriate subprime mortgages because low “teaser rates” made them able to afford monthly payments. All this was common in an era in which wages and salaries were failing to rise even though productivity was rising steadily and profits were good.

Meanwhile, at the top of the income distribution, there was an explosion of speculation as the wealthy put their money into multiple homes, hedge funds, securities, and new financial instruments, like bonds securitized by mortgages. This other aspect of inequality of incomes in the U.S. — excessive compensation for CEOs, Wall Street executives, hedge fund managers, and other wealthy individuals — contributed to the bubble that inevitably burst, precipitating the current recession.

In fact, fifty-nine percent of all the income growth since 1989 accrued to the upper one percent of households and about thirty-six percent accrued to the upper tenth of that upper one percent (Mishel et al., 2008). The shift of income to the upper one percent since 1979 (their income share rising from 10% to 22.9%) represents an additional one trillion dollars of income for that group. This type of unbalanced income growth has greatly contributed to our current economic misfortunes.

Increased union organization would tend to shift the income distribution in favor of the middle class, enhancing the purchasing power of this key group of the nation’s consumers and allowing them to once again afford to buy automobiles, homes with 30-year fixed rate mortgages, and all the other goods and services important to American life. Unionization of low-wage service workers similarly would increase purchasing power and help revive the economy. Putting more dollars into the pockets of working families stimulates the American economy – both in the short term and in the long run – because they spend such a high proportion of those dollars here.

It is no accident that the prosperity and consumer boom of the 1950s – a period of unprecedented middle class expansion, broad business growth, increased home ownership, rising consumer spending, and the shared expectatio
n that a college education was within the reach of everyone and that the lives of our children would be better than our own – followed the greatest sustained expansion of unionization in American history.

The notion that greater unionization is harmful to an economic recovery is misguided. Unions, as institutions, and the members that form them are economically rational and do not pursue demands that force firms out of business. There are several studies that show that firms that become unionized (see the review of studies in DiNardo and Lee, 2004) are no more likely to fail than are firms that remain nonunion. If anything, unions are more important in a recession. As was stated in a statement signed by forty prominent economists and released on February 25th, “The current recession will further weaken the ability of workers to bargain individually. More than ever, workers will need to act together.”

Economic recovery and future economic stability depend on a middle class once again having sufficient purchasing power to sustain the economy; we must not rebuild another bubble economy. Greater unionization can contribute to that goal because wages and benefits for ordinary workers will rise and income inequality in the economy as a whole will be reduced. In short, unions help foster the broad middle class that is essential to our nation’s economic strength.

The Long-run Impact on American Competitiveness

A crucial question is whether in an increasingly global economy, U.S. economic competitiveness would be hurt by an increase in union representation. Contrary to the conventional wisdom, there is little reason to fear in this regard.

First, most parts of the world, including all of the high-end economies with which we compete, have much higher levels of unionization than we do. Those high-end economies also pay higher benefits to their blue-collar workers. Of the 20 richest countries tracked by the U.S. Bureau of Labor Statistics, the United States ranks 17th in hourly pay for production workers in manufacturing. This group of trading partners accounts for almost half of total U.S. trade flows (Bivens, 2009). The key difference in competitiveness is not unionization; it is that we burden our businesses, especially our largest corporations, with the high cost of health insurance, whose cost is spread across society in other high-end economies, and the disadvantage of an overvalued currency. In fact, high rates of unionization are associated with smaller trade deficits, a good measure of international competitiveness (Bivens, 2009).

Second, low labor costs are never going to be a reliable basis for U.S. competitiveness in a global economy – rather, the U.S. needs to compete on the basis of innovation, high value-added, high quality, and high productivity. Unionization tends to promote the shift to these latter bases of competition by foreclosing the low-wage alternative.

Unions increase productivity through a variety of channels. They reduce turnover and, hence, firm-specific skills are retained. One benefit is that turnover costs are lowered for employers. Moreover, the lower turnover makes it economically rational for employers to provide more training to union-represented employees, increasing employee skills and productivity further. In addition, since unions increase compensation, firms are incentivized to invest in new technology (which tends to be labor -saving), increasing productivity. Unionized employers also tend to shift to higher value-added goods and services in their product mix. And in sectors in which there are union-supported apprenticeship programs, employers can take advantage of this source of highly-skilled labor.

Research on this topic indicates that there is substantial variation in the “union productivity effect.” The effect is much larger where there is a good relationship between labor and management, whereas in high-conflict situations, there is little likelihood that unions enhance productivity (Belman 1992). Strikes, of course, are particularly deleterious. Hence it is important that public policy not only makes it possible for workers to organize should they so desire, but also that the federal government provides a path to unionization that reduces conflict and gets the labor management relationship off to a good start.

In fact, this was part of the reasoning behind the National Labor Relations Act when it was passed in 1935. Section 1 of that Act, quoted earlier, speaks about the need to protect commerce “from injury, impairment, or interruption… by encouraging practices fundamental to the friendly adjustment of industrial disputes arising out of differences as to wages, hours, or other working conditions, and by restoring equality of bargaining power between employees and employers.” The idea in 1935 was that if employers were legally required to recognize and bargain with their employees’ chosen representatives, recognition strikes would be unnecessary and contentious disputes over wage and working conditions would be channeled into the collective bargaining process, to the benefit of all.

Unfortunately, because of a series of changes in the interpretation of the law over time, employers are now able to insist that before collective bargaining can commence, employees must prove their support for their chosen bargaining representative through an election process that is so conflict-laden that it fails to fulfill the purpose of getting collective bargaining relationships off to a constructive beginning. The waiting period prior to an NLRB representation election creates a period of counterproductive labor-management strife that increases workplace tension and undoubtedly hurts workplace productivity. Even when employees win the right to representation through an election, they are often unable to negotiate a first contract. This occurs because the strike is the dispute resolution procedure when the parties are unable to agree on a contract. American workers often don’t want to strike, and yet they often cannot get a first contract without a successful strike. The entire representation election process is still extremely conflict-laden and is ripe for reform. The proposed Employee Free Choice Act is one option that shows particular promise to lessen labor-management conflict during the unionization stage.

In short, we can be competitive while allowing American employees to exercise their rights to form a union. To do so, we need a way for workers who want union representation to organize in a less conflict-laden way and to initiate a constructive labor-management relationship.

What About Small Business?

Another issue is whether small business would be particularly disadvantaged if employees who wanted union representation had an easier way of organizing than the current NLRB process. There are several reasons to doubt that would be the case.

For one reason, small employers often have a different employment atmosphere than that which exists in large bureaucratic organizations; it may well be that employees in small firms have little demand for union representation. Interestingly, rates of union representation in small employers are currently lower than those in larger organizations in the United States, even though unions are in fact more likely to win representation elections in small than in large units. This means many small business owners should not be overly concerned about possible changes in the law governing union organizing.

At the same time, individuals who work in small business should have the same rights to freedom of association and union representation as anyone else. So if the employees of a small employer do form a union, what then?

Actually, there can be substantial benefits to small business from union representation. When an industry is characterized by many small employers, each firm can benefit from area-wide unionization that standardizes compensation across competing firms, stabilizing the industry. The unio
n provides a pool of well-trained labor that becomes attached to the industry. Moreover, the union often serves important functions in training and benefit-provision for the entire set of employer signatories to a union contract. Furthermore, unions recognize the need to preserve and enhance the competitiveness of unionized employers. Unions organizing small businesses in the service sector often defer negotiated wage increases until the majority of competing employers are also unionized, and give newly unionized firms several years to catch up to union contract levels. Contrary to popular opinion, unions, like businesses, also act in an economic rationally manner.

In short, while I doubt that a new process of union formation would cause an explosion of union representation in small firms, if some small business sectors were to be organized because their employees are frustrated with current conditions and seek change, that could provide positive benefits for small businesses and their employees.


For all these reasons, I urge you to enact changes in our nation’s labor law that would make it easier for workers to organize, should they so desire, to obtain an initial agreement, and to build a successful working relationship with their employer, free of unnecessary labor-management conflict.

The restoration of a strong middle class is indispensable to the restoration of the American economy. Unions are an essential part of rebuilding that middle class.


Bivens, Josh. 2009. Squandering the Blue-Collar Advantage, EPI Briefing Paper #229, Economic Policy Institute, Washington, DC. February 12, 2009.

Belman, Dale. 1992. “Unions, the Quality of Labor Relations and Firm Performance,” in Larry Mishel and Paula B. Voos, eds., Unions and Economic Competitiveness. New York: M.E. Sharpe.

Bennett, James T. and Bruce E. Kaufman, eds. 2007. What Do Unions Do? A Twenty-Year Perspective. New Brunswick, NJ, Transaction Publishers.

Card, David, Thomas Lemieux, and W. Craig Riddell, “Unions and Wage Inequality,” in Bennett and Kaufman, op. cit., 2007.

“DiNardo, John and David Lee. 2004 “Economic Impacts of New Unionization On Private Sector Employers: 1984-2001”, Quarterly Journal of Economics, 119 (November).

Doucouliagos, Chris (Hristos) and Patric Laroche. 2003. What Do Unions Do to Productivity? A Meta-Analysis.” Industrial Relations, 42 (October): 650-91.

Freeman, Richard B. and James L. Medoff. 1984. What Do Unions Do? New York, Basic Books.

Mishel et al. The State of Working America 2008/09, Economic Policy Institute Series, Cornell University Press, 2008.

Download print-friendly PDF version