Public Sector Union Delema

America’s Public Sector Union Dilemma

By Lee E. Ohanian Saturday, November 26, 2011

Filed under: Economic Policy, Public Square

There is much less competition in the public sector than the private sector and that has made all the difference.

Since the Great Recession began in 2008, there has been a growing criticism of public sector unions, reflecting taxpayer concerns about union compensation and unfunded pension liabilities. These concerns have led to proposals to change public sector union policy in very significant ways. Earlier this month, voters in Ohio defeated by a wide margin a law that would have restricted union powers, although polls showed broad support for portions of the law that would have reduced union benefits. In Wisconsin, a state with a long-standing pro-union stance, Governor Scott Walker advanced policy in February that would cut pay and substantially curtail collective bargaining rights of many public sector union workers. In Florida, State Senator John Thrasher introduced legislation that would prevent governments from collecting union dues from union worker state paychecks. And it is not just Ohio, Wisconsin, and Florida that are attempting to change the landscape of public unions. Cash-strapped governments in many states are considering ways to reduce the costs associated with public unions.

It is important to determine why public unionization rates are so much higher than in the private sector, and whether public union employees are excessively raising costs to taxpayers. Public sector workers may be paid significantly more than private sector workers and their pensions and job security are often higher than in the private sector. Factoring in the lower likelihood of dismissal and layoffs in the public sector, public sector compensation may be 10 percent higher than market rates.
I calculate that bringing public sector wages closer in line with private sector wages by reducing them by 5 percent can reduce state fiscal deficits considerably. For California, which is among the most fiscally strapped states in the nation, reducing state worker wages by 5 percent would reduce the state deficit by about 15 percent. Moreover, some public sector workers, such as California prison guards, are paid far in excess of competitive levels, reflecting a strong union and effective lobbying that has fostered rapid compensation growth. Other unions, such as teacher unions, do not drive up compensation nearly as much, but instead have substantial negative impact by protecting poor teachers, which in turn reduces the quality of public education and reduces human capital.


The Economic Implications of Unions

It is clear that governments must use a systematic approach to public sector compensation if the public is to avoid overpaying government employees.

A union is a form of monopoly, or cartel. It is a single seller of labor services to a business. This means that unions have the ability to raise compensation for its members above the level that would prevail in a competitive marketplace, as well as to define work rules for its members that reduce efficiency. There has been considerable research on the effects of collective bargaining on wages, and consensus estimates are that unions raise wages by about 10 to 15 percent above the rate that would prevail in their absence.¹
There is comparatively less research on the impact of work rules on economic activity, but the available data suggests that union work rules, particularly in industries that face little competition, can substantially reduce efficiency and output. James A. Schmitz of the University of Chicago estimates that during periods of very limited competition, union work rules in the iron ore industry reduced output per worker by about 50 percent.²
By raising wages and adopting inefficient work rules, unions increase business costs and prices, which in turn reduce employment and output. From this perspective, unions seem misplaced in a modern economy in which there is considerable competition in the labor market and in other markets, and in which there is general recognition among economists and policymakers that increasing competition benefits society by improving the allocation of scarce resources, increasing efficiency, and maximizing output.
Unions are largely a carryover from many years ago, when there was much less competition in the economy for workers and unions were considered an important economic force that was necessary to protect worker safety and health. But both labor market conditions and worker health and safety conditions are much different today. Most workers now live in locations with many employers competing for their services. This competition for workers means that wages coincide with worker productivity. And workplace health and safety is now largely covered by federal and state laws. As a result of these changes, the socially useful role of unions has declined considerably over time.
Divergent Trends in Private and Public Unionization Rates
These changes in the importance of unions to workers over time are also reflected in changes in unionization rates. One of the most striking trends in labor markets over the last century is the very rapid increase, followed by the subsequent substantial decline, in unionization rates in the private sector. Figure 1 shows the share of unionized employment from 1929 to the present. The data show a very significant increase in private sector unionization during the 1930s as union membership jumped from about 12 percent in 1929 to about 35 percent during World War II.

This dramatic increase reflects the passage of a number of important pieces of pro-union legislation, including the National Labor Relations Act in 1935, which not only made it easier to organize workers into a union, but also increased union bargaining power, which in turn raised union wages and thus increased the attractiveness of unions for workers.

The cost of California’s prison system is about $44,000 per inmate, compared to a national average of $28,000.

But union representation in the private sector began to decline, at first slowly in the 1960s, and then accelerating in the 1970s. Private sector unionization rates have declined from about 37 percent in 1952 to only about 6 percent today. Declining private sector unionization reflects a number of factors, including that the economy is much more competitive than it was 60 years ago, and that many of today’s workers prefer to negotiate their own opportunities rather than relinquish their individual bargaining rights to collective bargaining.
It is also important to recognize that declining unionization is not the result of the country’s declining industrial base, as is often suggested (see for example Bluestone, 1990, and Rowthorn and Ramaswamy, 1997). In particular, declining unionization characterizes most of the private sector economy, including industry. As Barry T. Hirsch shows, unionization rates in manufacturing and construction, two of the most heavily unionized sectors, fell from about 40 percent in the early 1970s to less than 15 percent in 2006.³
Increased competition is considered by many economists to be a major factor in understanding lower private sector unionization.^4,5 In a competitive industry, competition for workers drives wages up to the level of worker productivity, and competition in product markets drives output prices down to the level that is consistent with the market return on capital. Thus, union attempts to raise compensation or implement inefficient work rules in a highly competitive market would result in firms becoming unprofitable.


In contrast, if a firm or industry is protected from competition, then profits will be higher compared to profits under competition. Excess profits are called economic rents by economists, and unions can obtain a share of these rents for their members by raising wages and/or changing work rules. Thus, the economic rents that result from too little competition are divided between labor and capital, and the relative division between these two parties depends on their respective bargaining positions.

Not surprisingly, much union organization focused on highly concentrated industries in which there was little competition, such as autos and steel.⁶ Unions are likely to be more successful in raising wages in highly concentrated industries because it is feasible to unionize the entire industry, such as the United Auto Workers in organizing General Motors, Ford, and Chrysler in the 1930s.

Consensus estimates are that unions raise wages by about 10 to 15 percent above the rate that would prevail in their absence.

But the auto industry, as well as much of the manufacturing sector, has not only been impacted by foreign competition (auto imports are now about 17 percent of GDP compared to about 5 percent of GDP in 1970), but by competition among the U.S. states. In 1947, the Taft Hartley Act substantially changed the National Labor Relations Act by giving states the right to outlaw the union shop, in which workers must join a union. Today, 23 states have passed these “right to work” laws, including most of the South and many of the Midwest and the Mountain West states. The share of employment in these “right to work” states has increased from about 24 percent of employment in 1955 to about 38 percent today. Thomas J. Holmes has studied how “right to work” laws impact the location of industry and finds that these laws are quantitatively important determinants of where manufacturers choose to locate.⁷ Using detailed county-level data, Holmes examines business activity at state borders in which a “right to work” state borders a non-“right to work” state. He finds that the manufacturing share of employment increases by one-third in the right-to-work state at the border compared to the other state, and concludes that this higher share of manufacturing is significantly due to differences in these unionization policies.

The impact of competition among the states is clearly illustrated by the location decisions of foreign companies like Toyota and Honda. When these companies decided to produce autos in the United States, they chose to locate in right-to-work states, including Kentucky, Texas, Mississippi, and Alabama. Foreign companies now produce about 50 percent of autos manufactured in the United States, and this production share will likely increase in coming years. Auto workers in non-union plants show little interest in United Auto Workers overtures to become organized. One worker at a southern U.S. auto plant stated "We have good communication with management here.
Why would you need a union? The only time a union shows up is to collect dues or at election time".⁸ Another worker of 17 years stated "The UAW has to have a reason to come inside the company. Nissan doesn't give them one. I don't need someone talking to the boss for me”.⁹

Today’s increasingly competitive global and domestic economy indicates that there are important limitations on what unions can plausibly achieve compared to what they were able to achieve in the past. Thus, workers will have little demand for union representation when unions cannot deliver better pay and working conditions than what workers can achieve on their own. From this perspective, the very large decline in private sector unionization is not surprising.
But unionization trends among public sector workers differ considerably. Figure 1 also shows unionization rates for state and local government workers since the early 1980s. But rather than declining, as in the case of the private sector, public sector unionization rates have been relatively high and stable over time, at around 43 percent for local government workers and about 33 percent for state workers. These very different trends reflect large differences in the impact of competition on private versus public sector employees. Specifically, the very large decline in unionization in the private sector has been significantly impacted by increased competition, which has reduced the ability of unions to raise wages or change work rules. But much less competition exists in the public sector, and this means that unions have more opportunities, which makes union membership more attractive.



Figure 2 shows compensation rates for public and private sector workers since 1929, measured in 2008 dollars. Between the end of World War II and 1980, note that private sector and public sector compensation moved in lockstep with each other, as both rose from about $25,000 after World War II to nearly $50,000 in 1980. But after 1980, public sector compensation diverges from private sector compensation significantly, as public sector compensation rises to nearly $70,000 per worker, representing about a 40 percent increase from 1980, but private sector compensation rises to only about $60,000, representing about a 20 percent increase from 1980. The gap tends to become larger during the recessions of the early 1990s and also the Great Recession, reflecting slower compensation growth in the private but not the public sector. Moreover, this rising compensation differential between the public and private sector may be significantly understated, as it does not include pension benefits, which tend to be more generous in the public sector. While other factors are likely involved, it is plausible that the relatively high unionization rates in the public sector can account for some of the acceleration in public sector pay relative to private sector pay over the last 30 years.

Is Public Sector Compensation Too High?

The socially useful role of unions has declined considerably over time.

Some economists have argued that the acceleration of public sector compensation compared to private sector compensation is imposing a significant cost to taxpayers. At first glance, it may seem reasonable that the public sector should provide approximately the same compensation rates as the private sector. However, if one factors in the higher job security and pensions of public sector jobs, public sector wages should actually be lower than those in the private sector.
Regarding job security, Chris Edwards documents that the chance of job loss by layoff is three times higher in the private sector than in the public sector.¹⁰ Since most individuals dislike risk (what economists call risk aversion), the fact that public sector jobs have more job security than private sector jobs raises the benefit of working for the public sector and suggests that the public sector may be able to pay lower compensation rates than the private sector because of this added benefit. This explanation for why the public sector could compete with the private sector even while paying lower compensation follows the same reasoning as why relatively riskless assets, such as U.S. Treasury securities, provide a lower rate of return than risky assets such as stocks and real estate.

I provide an estimate of this differential by conducting an analysis of the difference in private and public compensation, accounting for the additional job security offered by the public sector. To simplify the analysis, I focus on a representative worker, and thus abstract from potential differences in worker characteristics between public and private sector workers that might impact the analysis. I use a standard economic model in which a risk-averse individual faces unemployment risk, which is about three times higher in the private sector than the public sector. In the model, there is a representative worker with a risk aversion coefficient of two, which is close to the consensus estimate in the related literature. At any point in time, the worker in the model will be in one of three possible states: employed in a relatively high paying job, employed in a job paying 25 percent less than the high paying job, or unemployed, in which they receive unemployment benefits. The relatively low paying job is included in the analysis because workers frequently take jobs following a layoff that pay considerably less than their previous job.

I use historical data from the Bureau of Labor Statistics to identify the probability that a worker in the model is involuntarily separated from their job, which is about a 4 percent chance per month, average duration of unemployment, which is about 3.5 months, and the probability of continuing employment, which is about 96 percent.

For the case of the public sector, the probability of involuntary separation is just 1.3 percent, which is one-third as high as the probability in the private sector case. I then calculate the difference in compensation between the public sector (low unemployment case) and the private sector, such that a worker would be indifferent between working in either sector. I find that workers would be willing to work for about 10 percent less compensation in the public sector, given the additional benefit of much higher job security. This estimate is conservative in terms of considering today’s labor market, as average unemployment duration today is much higher than its historical average.

The dismissal rate of teachers for performance-based reasons, which is about 0.1 percent, is very low compared to dismissal rates in other occupations.

This analysis suggests the possibility that public sector compensation may be significantly higher than competitive levels. Moreover, the fact that public sector workers are only about one-third as likely to voluntarily leave their job as private sector workers is consistent with the conclusion that average public sector compensation rates are in excess of competitive levels, indicating that there are relatively few external employment opportunities that dominate public sector workers’ jobs. The fact that average public compensation is higher than average private sector compensation suggests that public sector worker compensation may be well above competitive levels and indicates that public sector wages could be reduced without significantly impacting public sector employment. For example, I’ve calculated the impact of a 5 percent wage reduction for all public employees in California, a state with one of the most severe fiscal crises in the country. A 5 percent wage cut would reduce state spending by $1.33 billion, which would reduce California’s 2011 state budget deficit by nearly 15 percent.

This finding has implications for how public sector wages should be determined. Specifically, there are no systematic efforts in government to evaluate the level of public sector compensation with reference to the private sector, particularly at the local level where unionization among public workers is the highest. Robert G. Gregory and Jeff Borland analyze wage determination in federal, state, and local governments and conclude that pay scales are significantly determined by wage bargaining, which ranges from national wage agreements to decentralized wage agreements within a local area.¹¹ Federal General Schedule compensation is determined by the executive and legislature on the basis of some evidence on private enterprise pay rates for similar types of work. But at the state and particularly the local level, compensation determination takes place through negotiation between employers and unions representing public sector employees. In this case, compensation outcomes depend upon features of the institutional environment, including the extent to which employers are required to bargain with unions, whether arbitration procedures are available to resolve disputes, and whether unions have the legal right to take strike action as part of wage-setting negotiations. Bargaining between employers and unions is the most common method of wage-setting for local government employees in the United States.

It is clear that governments must use a systematic approach to public sector compensation if the public is to avoid overpaying government employees. This approach is imperative because of the limited scope of competition within government for at least some services and because of the relatively high level of unionization in the public sector.

Governments should first make assessments of the adequacy of public sector compensation with reference to private sector compensation, as in the case of the federal government. But all forms of government should evaluate compensation beyond simple comparisons with the private sector by adding in the relative advantages of public sector employment, including job security, relative differences between vacations and other paid leaves, and relative differences between pension and retirement benefits. This analysis could be an integral component in bargaining with public sector unions and insure that public sector compensation levels do not deviate far from reasonable levels. The following section describes what can happen to compensation levels when there is lack of competition, strong unions, and no comparative reference points for assessing compensation levels.

Some states are pursuing reforms that are in this spirit in dealing with public sector pensions. New York Governor Andrew Cuomo and New Jersey Governor Chris Christie have proposed plans that would reduce public pension costs in their respective states by increasing worker contributions to pensions, changing retirement ages, and eliminating the ability of workers to use overtime in late career years to pad pension payments. Both governors support these proposals by pointing out that current pension rules in their states are uncompetitive compared to other plans.

The Economic Impact of Unions: Excessive Compensation in California’s Prison Guard Union
The California Correctional Peace Officers Association (CCPOA) is the bargaining unit for California’s prison guards. CCPOA has been very effective in raising compensation for them. Figure 3 shows prison guard pay across states, and highlights the high level of California guard pay.

I
n 2006, when California’s last contract with the CCPOA expired, about 900 guards received at least $50,000 in overtime pay, and about 1,600 officers received more than $110,000 in pay. The cost of California’s prison system is about $44,000 per inmate, compared to a national average of $28,000.¹² The CCPOA has been able to engineer remarkable compensation growth for its members not through any demonstrable evidence of substantially higher productivity, which is the typical determinant of compensation growth in a competitive labor market, but rather through its monopoly status as a single seller of labor to the state prison system and through political contributions that selectively support legislation and programs that boost the demand for prison guards.

Declining private sector unionization reflects how many of today’s workers prefer to negotiate their own opportunities.

Some of the CCPOA’s political contributions have probably reduced the effectiveness of California’s prison system, particularly regarding its ability to rehabilitate and reform offenders, and as a consequence may have made crime worse in the state and increased its cost. Today’s California prisons are much less effective than in the past, as 77 percent of crime in California is the result of recidivism (crimes by repeat offenders). California’s recidivism rate is about twice the national average, and this is attributed by many to the lack of rehabilitation and training programs in California prisons. These programs, which at one time were common in California prisons and were acknowledged as a central positive feature of California prison policy, lost funding beginning around 1980 and are now largely absent.

CCPOA has actively lobbied against legislation that would enhance training and rehabilitation in prisons. In 2005, CCPOA spent $11 million on political activities, compared to around $3 million in previous years, as 2005 was Governor Arnold Schwarzenegger’s “year of reform” for California government. Some of CCPOA’s lobbying and contributions helped defeat Schwarzenegger’s 2005 proposal to reduce overcrowded prison populations by as much as 20,000 inmates through a program that would have placed parole violators into rehabilitation programs rather than returning them to prison. Schwarzenegger dropped this proposal following television ads funded by the CCPOA that criticized the governor for proposing to release dangerous criminals. In 1999, the CCPOA opposed a bill that would have funded a pilot program for alternative sentences for nonviolent parole offenders, thus reducing the prison population, and which was vetoed by Governor Gray Davis. The CCPOA has also opposed candidates who advocated private prisons, which tend to have lower operating costs. In 2004, CCPOA strongly supported California’s controversial “ three strikes” law—which requires that three-time offenders face mandatory and extended prison terms, including for petty theft and nonviolent drug crimes—by spending over $1 million to defeat Proposition 66, which would have reduced the number of offenders who would serve life sentences under this law.

In 2010, CCPOA contributed about $1 million to defeat Proposition 5, which would have reduced prison overcrowding by providing treatment rather than prison sentences for nonviolent drug offenders. CCPOA also contributed about $2 million to Jerry Brown’s gubernatorial campaign. Governor Brown recently agreed to bargaining terms with the CCPOA, which had been operating without a contract since 2006, having failed to reach agreement with Schwarzenegger. All told, the CCPOA spent about $7 million on 2010 elections, including supporting 107 political candidates, 104 of whom were elected. California State Senator Juan Vargas, who won by 22 votes in 2010, said “I won by 22 votes and without CCPOA I wouldn’t have been close … They literally won this campaign for me.”¹³

And prior to reaching a contract agreement with Governor Brown, CCPOA President Mike Jimenez remarked, “We’ve had a long-term relationship with Jerry Brown. He’s got really good intuition … on what we need as a profession.” Furthermore, Craig Brown, a CCPOA lobbyist, stated, “We should be able to develop a good contract with this governor and we should have no trouble getting it ratified.”¹⁴

The agreement includes eight weeks of vacation per year and a controversial component that would allow prison guards to cash in unused vacation time at a 100 percent accrual rate, compared to an 80 percent rate that is used for most other state employees. Moreover, the unused vacation time can be used to increase pensions. The accrual of unused vacation days and the significant use of overtime is partially the result of California’s prison system being far above capacity, which likely reflects the lack of reform programs in the system as well as California’s three-strikes law. California State Senator Anthony Cannela, who was supported by the CCPOA in last fall’s election, broke ranks with other Republican lawmakers to cast the deciding vote when the State Senate ratified the contract. Cannela remarked that “I feel that the CCPOA was actually critical to my election.”¹⁵

Unions are likely to be more successful in raising wages in highly concentrated industries because it is feasible to unionize the entire industry.

Overall, the CCPOA spent about $7 million on California’s recent elections. And of the 107 candidates they endorsed, 104 won their election. The CCPOA website boasted, “We won big this year. Played a decisive role in electing the governor. Elected new friends in the legislature. Made a difference for the men and women who walk the toughest beat. We win because we never quit, and that’s what makes us CCPOA.”

The CCPOA has consistently and systematically taken positions on candidates and bills that directly impact the demand for prison guard services and that in turn drive up prison guard compensation. The remarkable effectiveness of the CCPOA has resulted in some of the highest paid state workers anywhere, with compensation that significantly exceeds the national average.
But public sector unions do not impact the economy just by increasing wages above competitive levels. They also protect union members who don’t perform adequately. And the economic cost of protecting deficient workers can even be higher than simply increasing compensation.

The Economic Impact of Public Sector Unions: Protecting Underperforming Teachers
Unlike the prison guards union, school teacher unions have not driven compensation up dramatically; teacher compensation has increased at roughly the same rate of all workers.
But teacher unions significantly impact the quality of education and the human capital of new workers by protecting teachers who do not perform well. Recent research suggests that improving teacher quality, particularly by replacing the lowest performing teachers, can generate very large productivity and income gains that are essential if the United States is to maintain its competitiveness in the world economy.

There is considerable research on the impact of teacher unions on education quality and teaching outcomes. For example, Caroline Hoxby presents evidence that teacher unions are an important contributing factor for why quality and efficiency of public education has declined since 1960 despite the fact that state and local government spending on K–12 education has increased, reflecting higher teacher salaries and higher per-pupil expenditures.¹⁶ Hoxby identifies the impact of teacher unions through differences in the timing of collective bargaining, with a focus on when legislation is passed that fosters organization of teachers. She concludes that teacher unions are responsible for increasing the resources devoted to public education by using their market power in bargaining with school districts, and that, despite higher spending, unions depress the quality of education by reducing the productivity of teaching.

There is also evidence that the overall quality of teachers has declined over time and that a significant fraction of this decline reflects teacher unions. For example, Caroline Hoxby and Andrew Leigh find that the share of teachers who are among the top aptitude individuals, as measured by SAT scores, has declined over time from about 5 percent of teachers in 1963 to only 1 percent in 2000, and that much of this decline is due to the fact that teacher unions, like most other unions, compress compensation, which means that the spread in compensation between the highest quality and lowest quality teachers is reduced.¹⁷ And it is not only the very top aptitude individuals that are entering teaching at a lower rate. Wage compression benefits lower ability teachers, but reduces compensation of the best teachers, and this decline in compensation at the top end leads to fewer top aptitude individuals pursuing a teaching career.

Much less competition exists in the public sector, and this means that unions have more opportunities, which makes union membership more attractive.

Hoxby and Leigh also note that there has been a large increase in the lowest aptitude individuals—those in the bottom 25 percent of SAT achievement—entering teaching, and that they now make up about 16 percent to around 36 percent of the teacher population.¹⁸ This change in the composition of teachers is negatively impacting teaching outcomes.

There is significant evidence that teacher unions, which bargain for tenure for their members, protect underperforming teachers through the tenure system. For example, the dismissal rate of teachers for performance-based reasons, which is about 0.1 percent, is very low compared to dismissal rates in other occupations. Moreover, the cost associated with dismissing an underperforming teacher is very high, ranging from $100,000 to $200,000 per case. And union contracts also implicitly protect underperforming teachers with seniority-based layoff policies. Last year, Megan Sampson, a public school teacher in Milwaukee, Wisconsin, was laid off because of lack of seniority, even though she received Wisconsin’s outstanding first-year teacher award.

Some teacher unions have largely eliminated competition from the process of setting compensation. Michael Podgursky describes how teacher pay is often set by inflexible salary schedules in which pay depends on years of experience and degrees, but with no salary differences across teaching area, effort, or performance.¹⁹ He notes that these criteria “virtually guarantee shortages by field.” Moreover, some of the criteria used in wage setting, such as taking additional courses, are unrelated to instructional skill and quality. These courses raise salaries in the Los Angeles Unified School District by about $500 million every year without any significant increase in performance.²⁰

Former teacher union insider A.J. Duffy, who for six years was president of one of the largest teacher unions, United Teachers Los Angeles, acknowledges the severe inefficiencies in union teacher contracts. While Duffy served as UTLA president, Los Angeles Mayor Antonio Villaraigosa called the UTLA “one unwavering roadblock to reform.”²¹ Now, Duffy states that the teacher tenure process requires reform. He advocates a longer review period in order to earn tenure, and that teachers continue to establish that they are effective in order to retain tenure. Duffy also states that the teacher dismissal process should be reformed by substantially shortening it, saying, “I would make it 10 days if I could.”

Recent research shows that protecting poor teachers is very costly. In particular, research by Eric Hanushek finds that the economic impact of protecting poor teachers is remarkably high.²² He estimates that if the bottom 5 to 8 percent of teachers were replaced with average quality teachers, then U.S. student test scores in math and science would be at the top of international comparisons. This means that the human capital of future U.S. workers would rise, and thus generate higher production and income in the future. Hanushek estimates that the present discounted value of higher future incomes that would result from replacing poorly performing teachers would be about $100 trillion.
Some teacher unions are responding to calls for reform by working with school boards to address some of these issues, including unions in Pittsburgh and in some parts of Florida. And at its 2011 annual meeting, the National Education Association endorsed assessing teachers to ensure that they are accountable for student learning. However, teacher union reforms must be much more broadly and deeply based for public education to be efficient and effective.
The Future for Public Sector Unions

After 1980, public sector compensation diverges from private sector compensation significantly.

Unionization is much higher in the public sector than in the private sector, and there has been no tendency for unionization to decline in the public sector. While private sector unionization has declined from about 37 percent to about 6 percent, unionization in local government has been steady at around 45 percent.
Public sector unions have been able to thrive because of the very limited competition in state and local governments. This has allowed unions to increase wages above competitive levels. In particular, state and local government compensation has increased by about 40 percent since 1980, compared to about a 20 percent increase in the private sector. The average public sector compensation level is now $70,000, compared to an average of $60,000 in the private sector.

Moreover, the much higher rate of job security in the public sector, plus superior public pensions, suggest that state and local government workers would be willing to work for less than private sector pay. My findings indicate that accounting for just the much higher rate of public sector job security suggests that public sector employment could be competitive even with compensation that was about 10 percent lower than the private sector. The fact that average public sector worker compensation is higher than in the private sector, without taking into account pension benefits, suggests that public sector compensation levels may be significantly above competitive levels. I find that reducing the wages of state workers by 5 percent would reduce California’s 2011-2012 state budget deficit by nearly 15 percent.

There may be considerable savings from state and local government reforms that systematically develop competitive compensation analyses. Specifically, government should first benchmark compensation, including pensions, to private sector comparisons. Current efforts regarding pensions that have been proposed by New York Governor Cuomo and New Jersey Governor Christie are useful first steps along these lines. These proposals are aimed at reducing pension benefits to levels that are competitive from previous levels that have been the result of negotiations with powerful unions. Compensation analyses should also take into account other beneficial components of public sector employment, including much greater job security.

At the same time, public sector unions must understand that taxpayers will no longer accept uncompetitive agreements with unions. Rather than continue the union model of yesteryear in which unions focus on maximizing the size of the pie that members can receive, unions should understand that increasing wages for their members requires increasing productivity and reducing costs. Southwest Airlines has long been one of the most successful carriers because both labor and management are focused on achieving higher efficiency and levels of service than their competitors. Unions that can follow these principles will succeed, while unions that cannot will continue to come under fire.
Lee E. Ohanian is Professor of Economics at UCLA, and a Senior Fellow at the Hoover Institution.

FURTHER READING: Michael M. Rosen describes “The Real Problem with Government Employee Unions.” Alan J. Haus writes “Right-To-Work Unionism?” Andrew G. Biggs and Jason Richwine contribute “Public School Teachers Aren't Underpaid” and “Public vs. Private Sector Compensation in Ohio.” Biggs also reports “State Pension Hole Is Much Deeper Than Official Estimates.”

Footnotes