“The reason that Government Employee Union density is higher and more stable than private-sector density is twofold: GEUs can make a better case for government employees to unionize than private-sector unions can; and it is in the interest of government employers to promote unionization.”

Every Labor Day most print and electronic media anoint labor union leaders as spokesmen for all American workers. Yet in 2008 7.6 percent of private-sector workers and 36.8 percent of government workers were union members. Overall, the figure was 12.4 percent. So, at best, unions speak for a minority of workers. Moreover, they do not even speak for all of their member because only some are members by choice.

Labor unions get more respect than they deserve. They are nothing other than labor cartels. Like all cartels, their success depends on the extent to which they can cut off their trading partners—employers, workers, and the customers of employers—from alternatives. Notwithstanding that the National Labor Relations Act (NLRA) helps private-sector unions capture their victims, over time those unions lose market share because of the process of creative escape.

The Changing Face of Unionism

Figure 1. Union Densities, 1983-2008

Figure 1. Union Densities, 1983-2008


Union density is the percent of workers who are union members. Prior to the passage of the NLRA in 1935, density was less than seven percent. Because of special privileges granted to unions by the NLRA, by the mid-1950s density reached nearly 35 percent. For reasons discussed below, it has been declining ever since. Figure 1 depicts union densities, as reported by the Bureau of Labor Statistics (BLS), from 1983 to 2008. The top line shows that density in the government sector has been stable in the 35- to 40- percent range. The bottom line depicts a steady decline of private sector unionism, and the middle line shows the corresponding decline in overall unionism.

Figure 2. Government Union Works / Total Workers, 1983-2008

Figure 2. Government Union Works / Total Workers, 1983-2008


Prior to the 1960s, government unionism was almost nonexistent. As more and more state governments permitted the unionization of state and local workers in the 1960s and 1970s, government unionism exploded. By the early 1980s it started to level off, peaking at 38.7 percent in 1994. The percent of unionized workers who are government workers has been growing since the mid-1970s. In 2008, it was 48.6 percent. Figure 2 portrays that ratio from 1983 to 2008. Soon it will exceed 50 percent. The present recession has increased private-sector unemployment to over ten percent, and feckless attempts by government to ameliorate the recession have greatly expanded government employment. I predict that BLS data for 2009 will show government unions crossing the 50 percent threshold this year. The union movement is gradually becoming little more than a contest between government employees and taxpayers.

Unions Are Labor Cartels

In any market there are buyers and sellers. Given the number of competing buyers, the bargaining power of any one seller depends on the number of other sellers from whom buyers may buy. If there are no rivalrous sellers the one seller has high bargaining power, and if there are many sellers he has little bargaining power. A cartel is a group of hitherto rivalrous sellers of some product or service who join together to eliminate their rivalry.1 A cartel tries to fix price and other terms of trade thereby eliminating alternatives available to buyers. The infamous OPEC oil cartel is a good example. It seeks to eliminate rivalry among sellers of crude oil and set a monopoly price.

In the labor market, workers are sellers and employers are buyers. A labor union is a group of workers who join together to eliminate rivalry in the sale of their labor services. Its goal is to take wages and other terms of employment out of competition. It seeks to set a monopoly price for the labor services it sells so that employers must pay the monopoly price or go without those services.

Any cartel has to guard against “cheating” by any of its members. If individual members offer alternatives to the cartel price, the cartel loses bargaining power. Similarly a cartel must try to exclude any rivalrous outsiders from entering its market as sellers, or it must at least bring any newcomers in as cartel members. Ordinarily cartels find it very difficult to maintain discipline over their members and fend off competition from interlopers. However, in the United States, government has intervened to bolster labor cartels. For example, strikes and threats of strikes, often accompanied by violence, are the most common means for unions to discipline members and exclude rivals. The U.S. Supreme Court has ruled that union violence committed in the course of a “labor dispute” may not be prosecuted in federal courts.2 Extraordinary union violence and government’s disgraceful failure to provide any protection against it are well documented.3

Labor’s Alleged Bargaining Power Disadvantage

Title I, Section 1 of NLRA, as amended in 1947, asserts that unions are necessary to redress an inherent “inequality of bargaining power between employees… and employers.” This argument is still the main justification politicians use to support labor unions. The formation of labor cartels would, other things equal, increase the bargaining power of workers relative to employers. However there is another side to bargaining power. Other things equal, the bargaining power of labor increases as the number of rivalrous employers seeking to hire workers increases. If workers have more alternative employments from which to choose, their bargaining power increases without unions.

From Labor Unions, by Morgan O. Reynolds, in the Concise Encyclopedia of Economics:

Despite considerable rhetoric to the contrary, unions have blocked the economic advance of blacks, women, and other minorities. That is because another of their functions, once they have raised wages above competitive levels, is to ration the jobs that remain. The union can discriminate on the basis of blood relationships or skin color rather than auctioning off (openly selling) the valuable jobs to the highest-bidding applicants. Because craft unions such as the carpenters’ and railway unions have had more monopoly control over wage rates and hiring practices than industrial unions such as the auto and steel workers have had, craft unions have had more opportunities to exclude minority workers.

If employers could form (buyer) cartels to suppress wages, workers would have a bargaining power disadvantage that the formation of countervailing worker cartels could ameliorate. But there is very little evidence that employers have ever successfully conspired to suppress wages. According to Morgan Reynolds, over the 19th century, when unions were almost nonexistent in the U. S., worker-initiated job switching—quitting jobs to seek and obtain other jobs—increased steadily and substantially. The high rate of new business formation and the concomitant increase of employment alternatives open to workers account for this record. Moreover, large firms consistently paid higher wages than their smaller counterparts.4

The Key Principle of Coercion in the NLRA

Section 9(a) of the NLRA compels an employer to recognize a union as the monopoly bargaining agent for all non-managerial employees when a majority of them vote for union representation. A majority vote “certifies” a union as the exclusive representative of all workers, including those who voted against the union and those who didn’t vote. Individual employees are forbidden to represent themselves. Unions justify this coercion of dissenting workers on the grounds that it is workplace democracy. They argue that since it is legitimate to have majority rule in government elections, it must also be legitimate to have majority rule in workplace elections. But this analogy is inapt.

Unions are not governments. Democracy is the mandatory submission of a numerical minority to the will of a numerical majority. It doesn’t allow individuals to make decisions for themselves. Overriding individual preferences is appropriate in the few cases where different individual outcomes cannot peacefully coexist—e.g., rules and budgets for national defense, police and the courts. Such decisions are appropriately made through government. However, buying and selling labor services is a private matter. Different outcomes can coexist peacefully. When a worker decides to accept or reject the terms of a job offer, another worker can make a different decision. A job offer made and accepted is a matter of mutual, voluntary consent between an employer and an employee. Others can decide for themselves among available alternatives. Each can go his own way in peace.

The analogy is also incomplete. Winners of political elections must stand for reelection on a regularly scheduled basis. Not unions. Once a union is certified, it is presumed to have majority support among workers indefinitely. New workers never get a chance to vote. As new workers come in and older workers leave, the eventual result, as with the United Auto Workers, is that none of the unionized workers ever cast a ballot in favor of the union.

Exclusive representation precludes employers from dealing directly with employees about wages and other terms of employment. Employers are forbidden to reward individual workers for meritorious performance without union permission. Unions are loath to grant permission because they want workers to think that they, rather than individual productivity, are the source of wage gains. Thus, highly capable workers often want nothing to do with unions.5 Exclusive representation also prohibits individual workers from speaking directly with employers about any job-related issues without union permission. Individual workers have no voice. Only certified unions may speak.

Government Employee Unions (GEUs)

The reason that GEU density is higher and more stable than private-sector density is twofold: GEUs can make a better case for government employees to unionize than private-sector unions can; and it is in the interest of government employers to promote unionization. Any union-imposed wage increase that exceeds increases in worker productivity will increase the cost of labor per unit of output. In the private sector, union-impaired employers find it difficult to pass these cost increases on to customers. If they try to do so, they will lose market share to their rivals that do not have such cost increases and will have to cut production and employment. Government agencies typically have no competitors. They can pass cost increases on to taxpayers who cannot legally refuse to pay. Thus, GEUs can assure their members of higher wages than comparable workers receive in the private sector. In 2005, the Employee Benefit Research Institute, based in Washington DC, estimated that, on average, unionized government employees received 46 percent more in total compensation than comparable private-sector workers.6

In government employment it is in no one’s interest to seek to minimize costs and improve performance. GEUs and the government agency heads with whom they bargain sit on the same side of the bargaining table. The agency heads want bigger and bigger budgets and staffs to increase their status and influence, and the GEUs want more and more money for their existing members and more and more members to pay union dues. Both sides seek more and more taxpayer money. Since government agencies typically face no competition, they do not have to fear losing customers because of inefficient performance. For example, the U.S. Postal Service has a government-enforced monopoly on first-class mail. In 2008 postal-union density was 61.9 percent.7 Postal worker compensation is, on average, 34 percent higher than that paid to comparable private-sector workers,8 and we are familiar with the resulting quality of service.

The Process of Creative Escape

Unions depend on capture. They try to capture employers by cutting them off from alternative sources of labor; they try to capture workers by eliminating union-free employment alternatives; and they try to capture customers by eliminating union-free producers. Successful capture generates monopoly gains for unions. However, since most captives seek to escape, those gains are temporary. Entrepreneurship is alertness to, and grasping of, opportunities to gain by doing things differently.9 Everyone is a potential entrepreneur. Some are less alert to opportunities than others, but almost all are alert enough to mimic success. The decline of private-sector unionism is a gradual process of entrepreneurship-based creative escape.

Figure 3. Manufacturing Employment, 1983-2008

Figure 3. Manufacturing Employment, 1983-2008



Manufacturing has always been a major stronghold of American unionism. Hardly a day goes by without some politician lamenting the decline of American manufacturing and the good, middle-class jobs that go with it. Figure 3 shows the number of private American manufacturing jobs, 1983-2008. While total manufacturing jobs (the top line) has declined, all of that decline has been among union-impaired industries (the bottom line) such as steel, rubber and autos. Union-free manufacturing (the middle line) has held its own.10

The auto industry illustrates the process of creative escape. After World War II there were three dominant American auto firms, all unionized by the United Auto Workers (UAW). Until the late 1970s American car companies faced almost no competition from foreign sellers, and industry-wide collective bargaining with the UAW protected the auto companies from competition among themselves. The UAW and the auto companies shared monopoly gains at the expense of American car buyers. This presented a profit opportunity to foreign auto sellers to enter the American market. UAW contracts made it very difficult for American companies to compete, so they began to lose market share to their foreign rivals. American companies gained less and less from their hitherto cozy relationship with the UAW, so they gradually set up operations in relatively union-free domestic and foreign venues. At the same time, foreign firms began to set up operations in America, especially in southern and border right-to-work states.11 This allowed more and more customers to escape the grasp of the UAW and created union-free employment opportunities for American auto workers. The UAW gradually lost market share. Now, General Motors and Chrysler are subsidiaries of the U.S. government and the UAW. Ford is still privately run, but it remains severely union-impaired.

Unions claim that workers want to join unions but are prevented from doing so by nefarious employer resistance. However, on March 16, 2009, Rasmussen Reports released the results of a poll taken among union-free American workers, who were asked, “Would you like to belong to a labor union where you work?” Only nine percent replied “yes,” while 81 percent said “no.” The rest were undecided.12 It appears that most union-free workers wish to avoid becoming union-impaired. Perhaps they realize that in a world of open, dynamic competition, too much unionization ultimately leads to job losses.

The union response to open, dynamic competition is to try to shut it down. Among other things, they seek tariffs and import restrictions to eliminate foreign competitors; they decry “Benedict Arnold” corporations that export jobs abroad; they seek to make it easier to capture union-free workers by amending the NLRA to eliminate secret-ballot certification elections in favor of face-to-face confrontations between union organizers and recalcitrant workers;13 and they threaten environmental law suits in opposition to development projects unless the projects are constructed with union-only workers.14 Sadly, most members of Congress seem eager to comply.

In Conclusion

This Labor Day, let us honor all labor: voluntary union members, involuntary union members, and union-free workers. Let us celebrate the process of creative escape, which creates expanding economic opportunities for everyone except those who use coercion to live at the expense of others.


Buyers can also form cartels, but seller cartels are relevant here.

U.S. v. Enmons 410 US 396 (1973).

Armand J. Theiblot, et. al. Union Violence: The Record and the Response by Courts, Legislatures, and the NLRB, John M. Olin Institute for Employment Policy and Practice, 1999.

Morgan O. Reynolds, “The Myth of Labor’s Inequality of Bargaining Power,” Journal of Labor Research Vol. 12, No. 2: 167-183.

Daniele Checchi, et. al, “Inequality and Union Membership: The Impact of Relative Earnings Position and Inequality Attitudes,” IZA Discussion Paper No. 2691, March 2007.

“Unions vs. Taxpayers,” by Steve Malanga. The Wall Street Journal, Opinion Journal, May 14, 2009. http://online.wsj.com/article/SB124227027965718333.html

Union Membership and Coverage Database. http://unionstats.com (table for “Federal, Postal, State, & Local” under I. U.S. Historical Tables).

Testimony of Michael L. Wachter Before the Senate Committee on Governmental Affairs, February 4, 2004. http://hsgac.senate.gov/public/_files/020404wachter.pdf. (p.4) PDF file.

Israel Kirzner, Competition and Entrepreneurship, University of Chicago Press, 1973.

Union Membership and Coverage Database. http://unionstats.com (table for “Private Manufacturing” under I. U.S. Historical Tables).

In a right-to-work state union security (forced dues) has been forbidden by the legislature. Section 14(b) of the NLRA gives state governments that power. There are twenty-two such states.

“Just 9% of Non-Union Workers Want to Join Union.” Rasmussen Reports, March 16, 2009. http://www.rasmussenreports.com/public_content/business/jobs_employment/march_2009/just_9_of_non_union_workers_want_to_join_union

The Employee Free Choice Act now under consideration in Congress would do exactly that.

“The Morning Bell,” The Heritage Foundation, June 19, 2009.


*Charles W. Baird is Emeritus Professor of Economics at California State University, East Bay and a past Vice President and member of the Board of Directors of the Mont Pelerin. He was Director of the Smith Center for Private Enterprise Studies at CSUEB from its founding in 1991 to his retirement in 2007. He is a consultant on labor economics to the New Zealand Business Roundtable. His research specialty is in law and economics of labor relations.