The Political Economy of Government Sector Unionism

Charles Baird, California State University

Download PDF Version

Words matter. The subject is government-sector unionism, not public-sector unionism. Milton Friedman often reminded us that the word "public" includes all of us. Government is merely a subset of the public. Thus he spoke of government schools, not public schools. People who are employed by governments are government employees, not public employees. Much less are they public servants. In what follows I will use the acronym GEU for government-employee unions. The acronym sounds like their substance.

I use "political economy" instead of "economics" in my title, because politics and economics are inextricably interwoven in government-sector unionism. As Justice Lewis Powell wrote in Abood v. Detroit Board of Education [431 US 209 (1977)]:

The ultimate objective of a union in the public sector, like that of a political party, is to influence public decision-making in accordance with the views and perceived interests of its membership…. In these respects, the public sector union is indistinguishable from the traditional political party in this country….

Nor is there any basis here for distinguishing 'collective bargaining activities' from 'political activities' so far as the interests protected by the First Amendment are concerned. Collective bargaining in the public sector is 'political' in any meaningful sense of the word (at 256-257).

The Emergence of Government-Sector Unionism

Franklin D. Roosevelt, a dedicated crony of private-sector unionism, believed that government-sector unionism was “unthinkable and intolerable” ( In 1955 George Meany, the first president of the modern AFL-CIO, opined that “It is impossible to bargain collectively with the government” ( This sentiment against GSU was almost universally shared, but it could not withstand the realities of electoral politics.

As Fred Siegel recently wrote in The Wall Street Journal (, the first government-sector union was created in New York City in 1958 at the behest of Mayor Robert Wagner -- the son of Senator Robert Wagner, the principal author of the 1935 National Labor Relations Act (NLRA). Mayor Wagner and union boss Jerry Wurf agreed that as many city workers as possible should be assembled into unions and become dues payers. Wagner saw that a well-organized and well-funded union could be a formidable force in future elections by providing him with disciplined boots on the ground as well as other in-kind and pecuniary support. Wurf saw that he would get a special place at the table around which public policy is formed.

Their plan bore fruit in Wagner’s 1961 re-election effort. President Kennedy noticed the role government-sector unions played in Wagner’s 1961 victory. In January 1962, with an eye to his expected 1964 re-election campaign, Kennedy signed Executive Order 10988 which imposed GEUs on many groups of federal workers. Wisconsin became the first state to authorize GEUs in 1959. After Kennedy's executive order, government-sector unionism spread to as many as thirty states.

Union Membership Rates

Consistent data on union density (the percent of workers who are union members) in both the private and government sectors are available since 1983. That, and earlier data are available at Figure I shows union densities from 1983 through 2010. The top line shows union densities in federal, state and local government combined. That figure has been fairly steady in the rage 35-40 percent. In 1994 it reached a peak of 38.7 percent. In 2010 it was 36.2 percent. The bottom line shows that private-sector union densities have steadily declined since 1983. In 2010 it was 6.9 percent. Peak private-sector density was approximately 35%, and that was back in the mid 1950s.

Figure II disaggregates government-sector density into its three components – federal (bottom), state (middle) and local (top). Most GEUs are in local government. These consist mainly of government-school teachers, police, and firefighters. Federal density does not include postal unions. (Postal union density in 2010 was 68.6 percent. No wonder the post office loses billions of dollars each year.)

Figure III shows the percent of all union members (private and government) who are in GEUs. That percent has been steadily increasing. In 2009, for the first time ever, that number exceeded 50 percent (51.5 percent). In 2010 it was 52 percent. Although government employees are only 17% of the total number of employees in the U. S., American unionism is now dominated by GEUs. This is, I assert, the major reason why the union movement in the U. S. has moved so far left. Consider the politics of these successive presidents of the AFL-CIO. George Meany (1955-1979) and Lane Kirkland (1979-1995) had very conservative political views compared to John Sweeny (1995-2009) and Richard Trumka, the incumbent AFL-CIO president. Although Meany and Kirkland were diligent to get as much as they could for their private-sector members, they did not advocate big government. Today Sweeney and Trumka are champions of big government because government employees pay most union dues.

Forced Association

Unions represent their members, but, because of exclusive representation (monopoly bargaining) they also represent workers who choose not to be union members. The data at report the percent of workers who are union members (densities), and they also report the percent of workers who are covered by union contracts. I use the percent covered minus density as one measure of the extent to which unionism trespasses against workers' freedom of association. Figure IV depicts this measure of forced association in the government sector. The top line is the percent covered. The bottom line is density. The area between them (in blue) is forced association. In 2010, coverage was 40 percent, density was 36.2 percent, and forced association was 3.8 percent.

A Public Choice Analysis of GEUs

Public choice is the economics of politics. It is an attempt to understand what happens in the political world by assuming that all political actors are self-interested and pursue those interests through exchange with each other. In other words, public choice is politics as exchange. As James Buchanan, a Nobel laureate and one of the founders of public choice analysis, put it: public choice is "politics without romance" ( replaces the assumption that political actors pursue the "public interest" with the self-interest assumption that underlies economic analysis of private-sector markets.

Inside-government political actors include politicians, bureaucrats (those who are appointed to head agencies and those who are hired to actually operate the various agencies), and organized interest groups. Unorganized voters are also self-interested political actors. They try to do the best they can for themselves in the political marketplace just as they do in the private marketplace. However, the inside-government political actors often succeed at acquiring gains for themselves at the expense of unorganized voters.

Concentrated Benefits and Diffused Costs

During the Scott Walker v. Wisconsin GEUs battle earlier this year, we witnessed the spectacle of government employees chanting "raise my taxes." This illustrates one of the key principles of public choice analysis – the significance of concentrated benefits and diffused costs. Other thing equal, no one, including government employees, likes having to pay higher taxes. But when everyone pays higher taxes the extra tax receipts are disproportionately spent in ways that benefit government-sector workers. Their wages go up, or other terms of employment are improved, or the budgets of their agencies expand so their prominence and power increase. Perhaps all of the above. A private-sector employee pays the higher tax and gets little or nothing back. A government-sector employee pays the higher tax, but gets back much more. It is benefits focused on a few paid for by costs spread out over every taxpayer. Those who get the focused benefits work hard to expand them by supporting politicians who will keep the money flowing. Since the costs are widely dispersed, the per-person cost is usually not high enough to cause the unorganized taxpayers to resist. (The revolt against GEUs in Wisconsin and several other states suggests that in some venues the per person cost is getting high enough to create significant resistance.)

Gordon Tullock, another founder of public choice analysis, illustrates the concentrated benefits and diffused costs phenomenon with I call "Tullock's Favorite Law." Imagine a law is passed that imposes a one-dollar tax on every American (including Tullock) the proceeds of which are then given to Tullock. All the benefits are focused on Tullock, and the costs are widely spread over all taxpayers. Tullock is delighted and a mere one-dollar tax is not enough to get the taxpayers to revolt.

The Iron Triangle Model

Of course Tullock could not possibly get politicians to impose such a law. He is not an organized interest group. He is just one taxpayer. To see how the votes of politicians are typically determined we must consider the Iron Triangle Model. Figure V illustrates an iron triangle and the exchanges that occur among the three inside-government political actors. There is a separate iron triangle for each special interest group. Elected politicians are at the top corner, bureaucrats are at the bottom left corner, and an organized special interest group is at the bottom right corner. The arrows represent the exchanges that take place.

First, the exchange that takes place between the organized interest group and politicians: Politicians want to be elected and reelected to political office. They want to acquire and maintain political power. This means that they must seek pecuniary and in-kind electoral support. A special-interest group seeks protections from competition and other privileges that only politicians can give. For example, the United Auto Workers (UAW) would like politicians to enact laws making it easier for the UAW to force employees of the so-called "transplant" auto makers (foreign-owned firms making cars in America) to pay union dues. A deal is struck. The UAW will donate money to the electoral campaigns of compliant politicians. It will also pay many of its members to "donate" their time to phone banks, precinct organizing, and get-out -the-vote drives. In return, the politicians promise legislatively to "level the playing field" in favor of the UAW.

Second, the exchange between politicians and bureaucrats: Bureaucrats seek larger budgets for their agency programs. Larger budgets give them more power and influence over those they regulate. They also seek a broader regulatory scope so they can wield power over more and more people. Bureaucrats are ordinary people. They value prestige and power. Politicians want the bureaucrats to keep the crony organized interest group happy. A deal is struck. Politicians give the bureaucrats what they want in exchange for the bureaucrats servicing the special interest group.

The arrow at the bottom of the triangle represents the flow of services from the bureaucrats to the organized interest. Three current examples: The bureaucrats at the National Labor Relations Board (NLRB) are helping the unions that donate to and deploy manpower to the campaigns of union-friendly politicians by shortening the time that employers have to make their case against unionization during union representation elections. The same NLRB is trying to prevent the Boeing company, which is unionized in the state of Washington, from opening a union-free branch in South Carolina, a right-to-work state. And the same NLRB recently ruled that if a union can get an employer to agree to unionization without a representation election, the affected employees may not challenge that outcome by holding a representation election.

Rational Abstention and Rational Ignorance

While the in-government political actors dance around their respective iron triangles, unorganized voters are excluded, ignored and exploited. They get to pay for the game, but their interests are ignored. How can this be? Two other public choice principles provide the answer: rational abstention and rational ignorance.

Many people lament the fact that voter turnout is typically low. In most elections more than one-half of eligible electors abstain from voting. Why? Consider the costs and benefits of voting as they are perceived by any individual voter.

The only time that any one voter's vote affects an electoral outcome is if the votes of all the other voters produce a tie. From each individual elector's perspective, in elections with thousands or more electors who are unknown to each other, the ex ante probability that his one vote will break a tie is miniscule – insignificantly different from zero. If each voter considers the benefit of casting a vote is to affect the outcome, that benefit will be very small indeed. The costs of casting a vote include registering to vote and actually going to the polling place to vote. Those costs are significantly greater than zero. Hence it is rational for each potential voter to decide to abstain from voting. Why do some people who are not organized as a special interest vote? When asked most say that it is their civic duty, or that they experience peer pressure to vote, or that it makes them feel good about themselves.

You can count on people who are organized to pursue some special interest to vote. They will vote early, eagerly (and often, if they can get away with it). They will go door-to-door to round up people who will vote the right way. They will try to discourage and obstruct people who would vote the wrong way. They are organized to vote and work to obtain focused benefits while their per-person cost of those benefits is small. It is rational for people organized as a special interest to vote.

Many people also lament that most people who vote are uninformed about candidates and issues. Why is this true? To cast an informed vote a voter must spend a great deal of time and effort gathering information about candidates and issues. For what? To cast an informed vote that will not affect the outcome? The voter's costs of casting an informed vote greatly exceed the benefit of doing so, so it rational for each voter to choose not to become informed about candidates and issues. In other words, it is rational for each voter to choose to be ignorant about candidates and issues. Unorganized voters rationally choose not to pay attention to the political process. They determine their vote by their feelings and perceptions at the moment of voting. They might as well decide how to vote by flipping a coin.

In contrast, when an individual is considering whether or not to buy a car it is rational for her to evaluate the car very carefully before the decision is made. Her one vote "yes" or "no" will determine the outcome. The benefit of making the right choice clearly exceeds the cost of acquiring the information necessary to make the right choice.

A result of rational abstention and rational ignorance is that organized interests trump unorganized taxpayers almost every time a new government program, or expansion of an existing government program, is proposed. People organized into a special-interest group, who expect to receive concentrated benefits from government spending , work hard to increase spending. Unorganized taxpayers rationally choose not to use time and money to oppose the increased spending.

Meat Cleaver or Scalpel?

If anyone proposes that all government spending be cut by X percent, the inside-government political actors always reply that such an action would amount to using a meat cleaver when a scalpel would be more appropriate. The scalpel approach is to consider each program one at a time. But to do so is to play into the hands of the individual special interests. It gives the special interest in each iron triangle the advantage. The only way for taxpayers to win out over the various iron triangles is to tie them all together in one big package deal. When total spending and total debt are the issue, unorganized voters are likely to pay more attention to the process than when the pros and cons of any single program are in dispute. When it comes to cutting government spending, the meat cleaver does the job better than the scalpel. Scott Walker was able successfully to take on Wisconsin GEUs because he attacked them in the context of out-of-control total spending.

The Iron Dagger

Notice that in the case of the GEUs, the two corners at the base of the iron triangle converge into one. The bureaucrats – those hired to actually carry out government activities – and the organized interest group – the GEUs that represent such government employees – are one in the same. Figure VI shows that the iron triangle becomes a two-way arrow between politicians and the GEUs. We can think of the two-way arrow as the handle of a dagger aimed at unorganized taxpayers who must pay the resulting bills.

Campaign Contributions by GEUs

The wages and salaries paid to government employees come from all taxpayers. Some of those wages and salaries are taken by GEU officials as union dues. Those GEU officials then spend large portions of those dues to elect big-government politicians. Therefore, the iron dagger is the means by which all taxpayers, even those who favor small government, are forced to make campaign contribution to big-government politicians.

Figure VII is a chart included in a Heritage Foundation paper titled "Opportunity, Parity, Choice: A Labor Agenda for the 112th Congress" by James Sherk ( It illustrates how important GEUs are in the buy-a-politician game. The American Federation of State County and Municipal Employees, The Service Employees International Union, and the National Education Association are all GEUs. The chart only shows monetary donations. The just-as-important in-kind contributions (especially dedicated boots on the ground ) of GEUs are unmatched by any other organized interest groups. All taxpayers are, willy-nilly, dragooned into supporting the GEUs' favorite politicians.

The Power of Monopoly

Perhaps the main reason that government-sector unionism is growing relative to private-sector unionism is that the latter is constrained by competition while the former is not. They have to confront intense competition from union-free private sector firms. If a private-sector union is, through collective bargaining, successful in obtaining wage (or benefit) hikes that increase a firm's unit labor costs, that firm will not be able to pass the cost increase forward to its customers. Its customers have far too many alternative sellers from whom to buy.

Back when American car buyers had only three possible suppliers – GM, Chrysler, and Ford – all which were unionized by the UAW, the union could successfully obtain cost-increasing compensation hikes for its members. GM, Chrysler and Ford couldn't care less. Those costs increases could easily be passed forward to car buyers. This is no longer true. Although the UAW still controls GM, Chrysler and Ford, American car buyers have easy access to cars produced in foreign countries. Moreover, today there are many foreign-owned car companies which produce cars in the U. S. Almost all of them are union-free.

This means that when a union-impaired employer sits down at bargaining table with a union the two parties have conflicting ends. The employer, playing with her own money, will try to avoid any collective bargaining contract that will decrease her ability to compete with union-free rivals. The union will seek as high a compensation package as possible, but it is unlikely to get one that is significantly higher than what comparable union-free workers get. This decreases the attractiveness of unionization to union-free workers. In March 2009, Rasmussen released a poll indicating that only nine percent of union-free workers would prefer to be union members (

In contrast, government agencies which employ workers do not face competition. In most cases such agencies – e.g. government schools, fire departments, police departments, prison guards, and government transit authorities -- are monopoly, or near-monopoly providers of whatever goods or services they produce. Union-obtained compensation hikes are routinely passed forward on the "customers" these agencies "serve." Those customers, of course, are taxpayers; and taxpayers cannot legally refuse to pay the tax hikes that emerge from government-sector collective bargaining. Governors and legislatures sometimes try to cut government spending by abrogating union contracts. When they do they are often overruled by union-friendly judges (

Two recent studies illustrate the power of monopoly in government-sector collective bargaining. In "Public Sector Unions and the Rising Costs of Employee Compensation," published in the Winter 2010 issue of The Cato Journal (, Chris Edwards finds that, on average, state & local government compensation exceeds comparable private-sector compensation by 45 percent. In "Inflated Federal Pay: How Americans Are Overtaxed to Overpay the Civil Service," published by the Heritage Foundation on July 7, 2010 (, James Sherk finds a 30-40 percent federal compensation premium over comparable private-sector compensation.

Another indication of the power of monopoly in the government sector is illustrated by Figure VIII: The difficulty of getting rid of free-riding employees in government employment are well known. It is more difficult than turning the tides.

Taxation Without Representation

In 1761, James Otis, Jr., a politician and activist in colonial Massachusetts, famously proclaimed that "Taxation without representation is tyranny" ( This rapidly became a rallying cry for American revolutionaries opposing the British imposition of taxes – e.g. the tea tax – on Americans in the thirteen colonies. It was the driving force behind the Boston Tea Party of December 16, 1773. The idea that taxation without representation is a violation of fundamental democratic principles became and remains a basic principle of American taxation. Except when it comes to the taxes that emerge from government-sector collective bargaining.

According to American democratic process values (see Robert S. Summers,, the imposition of taxes is supposed to be decided by elected legislators and elected executives deliberating in the open with taxpayers having both access and voice. Collective bargaining in the private sector is carried on behind closed doors, in strict secrecy, until an agreement is reached and made public. It is an "unfair labor practice" (a crime) for either party – the employer or the union – to reveal anything that goes on during bargaining while it is taking place. In their statutes that created government-sector unionism, politicians applied these private-sector rules to government-sector collective bargaining. They did so without even thinking about the implications.

The wages, salaries and other terms and conditions of government employment are all paid for by taxpayers, but taxpayers are not represented in closed door, secret government-sector bargaining. The secrecy rules deny taxpayers both access and direct voice in the process, and there is no one at the collective bargaining table representing taxpayer interests.

The two parties at the table are bureaucrats representing an employing agency and union leaders representing a GEU. These two parties actually sit on the same side of the bargaining table. They have a common interest – to harvest more taxpayer money for themselves. The bureaucrats want bigger budgets to expand both the size and scope of their power to arrange other people's lives. Extra power gives them more self esteem and public prominence. The union leaders seek additional compensation for their members because that increases the union dues they can harvest. They also seek more and better perks, power and prominence for themselves.

One common perk that government-sector union bosses get is called "official time." The union bosses are government employees. Official time is paid time off from their government jobs to do union business. In 2009 alone, federal employees spent 2,911, 378 paid hours doing union work. Those hours cost American taxpayers $129,100,798 ( Another example is the power given to union bosses to control cuts in government spending. California's 2012 budget includes contingency budget reductions. That is, if tax receipts fall short of projections, which surely will happen, certain spending cuts will automatically take place. The leaders of the California Teachers Association have been granted amnesty. None of the teachers they represent can be laid off. Instead, the school year can be shortened by up to seven days (WSJ Political Diary 7/18/2011).

Binding arbitration is often used to avoid strikes by government employees. In these cases an unelected arbitrator makes the final decision regarding wages and benefits paid to government employees. But those wages and benefits are paid for by taxpayers. Binding arbitration to avoid strikes in the government sector is an egregious example of taxation without representation.

A Fourth Branch of Government

The rule for collective bargaining secrecy is not the only feature of private-sector unionism imposed on the government sector. In fact almost all of the statutes enacted to regulate state & local government collective bargaining simply copied the NLRA which governs private-sector collective bargaining. One provision of the NLRA is mandatory good faith bargaining between employers and unions. When transplanted to the government sector, mandatory good faith bargaining in effect creates a fourth branch of government dedicated to making government as big and as expensive as possible.

Mandatory good faith bargaining means that if a union wants to bargain about an issue the employer must bargain with the union about it (unless it involves some illegal activity). Moreover, the bargaining must be in good faith. It is an "unfair labor practice" for an employer not to bargain in good faith. The only sure defense an employer has against a charge of refusing to bargain in good faith is a record of being willing to compromise with the union. The employer is forbidden simply to say "Here is my offer, take it or leave it."

Ordinary lobbying groups – e.g. the Sierra Club and the Chamber of Commerce – try to get elected and appointed government officials to do their bidding. They often succeed. But any government official is free simply to say "no" and toss the lobbyists out of his office.

Not so with GEUs. In the collective bargaining process the government officials doing the bargaining are agents of the executive and legislative branches. The GEUs do not just try to get such government officials to do their bidding, they command them to do so. Government officials are forbidden to say "no." They must compromise with the GEUs. Nothing regarding wages, salaries and other terms and conditions of employment can be decided without the consent of the GEUs. The GEUs thus have veto power over those matters of public policy. In matters that come under the scope of collective bargaining, the executive and legislative branches of government cannot act without the consent of the GEUs. In such matters GEUs are equal partners with the legislative and executive branches. They are, in effect, a fourth branch of government.

A North Carolina statute declares that any contract between a state governmental unit and a GEU is null and void. In 1974 a teachers union filed suit against the statute in federal court on the grounds that the statute violated the teachers' freedom of association (Winston-Salem v. Phillips 381 F. Supp 644). The decision of the court against the union, which stands to this day, focused on the peculiar nature of GEUs. This excerpt has it exactly right:

The granting of collective bargaining rights to public employees involves important matters fundamental to our democratic form of government…. All citizens have the right to associate in groups in order to advocate their special interests to the government. It is something entirely different to grant one interest special status and access to the decision-making process (at 648).

In Conclusion

The extension of NLRA collective bargaining rules to the government sector was a blunder of enormous magnitude. President Franklin D. Roosevelt had it right. GEUs functioning under NLRA-like rules really are "unthinkable and intolerable." They should be abolished throughout the country. I think Governor Walker in Wisconsin, Governor Kasich in Ohio, and Governor Daniels in Indiana have taken the first steps toward that end.