Hit Job: Daniel DiSalvo on Public Sector Unions

May 04, 2015

Dean Baker
The Huffington Post, May 3, 2015

View article at original source.

Daniel DiSalvo doesn’t like public sector unions. That is the main takeaway from Government Against Itself (Oxford University Press), DiSalvo’s new book on public sector unions. In the course of reading the book, readers are likely to conclude that he is not especially fond of unions or workers in general. He also doesn’t like Social Security and Medicare. He even manages to get in a drive-by directed at Senator Elizabeth Warren.

But the main villain of the book is clearly public sector unions. DiSalvo paints a dark conspiracy where public sector unions push for ever higher pay and benefits, work rules that allow for endless loafing on the job, and disciplinary policies that prevent even the most incompetent from being fired. High costs and low productivity strain public budgets, but the political power of public sector unions prevents effective steps to counter their abuses. Since the unions are such large donors to political campaigns, politicians can’t stand up to them. It’s a moving story; the data just don’t quite fit the picture.

Before getting into the details, I will follow DiSalvo’s example and give some personal biography. My mother worked for 30 years in the Illinois Department of Public Aid. Much of this time she was a member of AFSCME, however she did move up to a supervisor position, thereby becoming management. She is now living on her pension from the state. When I was a grad student in Michigan, as a teaching assistant I was a member of the Michigan Federation of Teachers. The Center for Economic and Policy Research, where I am co-director, gets about 10 percent of its funding from unions, roughly half of which is from public sector unions. Now, let’s get back to the story.


DiSalvo’s begins his book with horror stories about three relatively young retirees from public sector jobs, all of whom enjoy comfortable pensions. He then tells us that the best estimate of the unfunded liability of the public pension system is $3.2 trillion (page 3), which he adds is approximately 21 percent of GDP. (Fortunately, his estimate of the shortfall had fallen to $2.7 trillion and 17 percent of GDP by page 155.) These numbers undoubtedly sounds pretty scary, but of course these liabilities refer to a 30-year pension planning period, not a single year. Over that time period, DiSalvo’s $3.2 trillion figure is equal to a bit less than 0.8 percent of the discounted value of GDP.  

But wait, it gets less scary. DiSalvo’s calculations use a rate of discount that is far below the expected and historic return on pension fund assets. If future liabilities are discounted using the expected return on pension fund assets, the cumulative shortfall is roughly $1 trillion, according to The Pew Charitable Trusts. This would come to roughly 0.25 percent of GDP over the 30-year planning period. That is not a trivial sum, but probably not the sort of thing that need keep us awake at night. By comparison, the increase in military spending associated with the wars in Iraq and Afghanistan was roughly 1.6 percent of GDP at its peak.

Furthermore, the tales of lavish pensions for early retirees are the exceptions, as can be easily discovered. The average benefit for a retired New York (non-uniform) state public employee in 2014 was $21,300. Non-uniform Detroit city workers got average benefits of $19,200 before the bankruptcy agreement. As a result of the bankruptcy, they took a 4.5 percent cut in pensions, and gave up their cost of living adjustment (COLA). The loss of the COLA will lead to cuts of close to 50 percent after 25 years. Retired public employees in Illinois get an average pension of $32,200. Sound generous? These workers don’t get Social Security so this will be the bulk of the retirement income for most of these workers.

There are cases where workers are able to retire early due to 25-years- or 30-years-and-out clauses and these provide the poster children for DiSalvo. Most public sector employees don’t start their work at age 20. The vast majority of non-uniform public sector employees don’t start collecting their benefits until they are over age 60. Police and fire fighters generally are able to retire with relatively generous pensions in their fifties. But these are dangerous and often physically demanding jobs where we probably would not want people working into their sixties.    

There are certainly state and local governments with pensions that are seriously underfunded, but it is hard to link this back to unions, as DiSalvo tries. Alicia Munnell, one of the country’s leading retirement experts, found no statistically significant link between pension funding and the power of public sector unions. A quick look at the data shows that heavily unionized states can be found at both the top and bottom on the funding ratio spectrum. Illinois and New Jersey have seriously underfunded pensions, but New York and Washington rate near the top with 87 percent and 95 percent funding, respectively. By contrast, South Carolina and Virginia, both states without public employee unions, rank near the bottom with funding ratios of 65 percent.


If DiSalvo’s attack on public employee unions and pensions falls short, he doesn’t do much better with his wage story since it isn’t clear what story he is trying to tell. When he lays out his main themes at the start of the book, he makes a point of telling us that public sector unions don’t do much to advance equity.

“While the benefits that public sector unions win for their members in terms of wages, health care, and pension benefits although substantial, tend to be smaller than those won by unions in the private sector, government unions don’t do much for workers with high school diplomas and limited skills because they don’t represent many of them (page 32).”

The problem with this picture is that virtually all the research on the relative compensation of public sector and private sector workers finds that less-educated workers in the public sector may enjoy a premium relative to their counterparts in the private sector. This is less obviously true for workers with college degrees and workers with advanced degrees are generally found to face a pay penalty compared with their counterparts in the private sector.

DiSalvo notes this fact in his longer discussion of wages in chapter 7, and in fact explicitly decries the compression of wages in the public sector. He argues that the compression of wages makes it more difficult for the public sector to get and retain high quality workers. That is a debatable point, but it goes directly opposite his earlier dismissal of public sector unions as a force promoting equity. In fact, there is little disagreement that workers with just high school degrees get better pay and benefits in the public sector on average than they would in the private sector.

Looking at the question of pay comparability more generally, DiSalvo has a largely fair discussion of the literature on the topic. Most of the research finds that in straight pay, public sector workers face a modest penalty, but this is mostly or entirely offset by better benefits. We can beat this one up in various ways, but you can’t make the story of comparability go away even if you can get the public sector pay being a bit higher or lower depending on the assumptions and the specific comparisons.

DiSalvo does follow the direction of some research in arguing for a substantial premium based on greater job security and the value of a defined benefit pension. The case on job security is less compelling that DiSalvo implies. As he notes repeatedly, public sector employees are much more likely to be college grads than the population as whole. This means his comparisons of turnover or unemployment rates between public sector workers and the workforce as a whole are inappropriate. It is likely the case that there is less turnover and layoffs even when adjusting for education, experience and other factors, but the difference between the private and public sector would be considerably smaller.

Also, as DiSalvo notes, private sector employees are likely to enjoy a premium in the form of more desirable workplaces. Public sector offices are usually functional and nothing more. Compare the Google campus to your favorite public sector office building. Where would you rather work?

But to really make a case that public sector workers are better paid than private sector workers it is necessary to assign a considerable value to the security offered by the defined benefit pensions that most public sector workers still enjoy and relatively few private sector workers now have. DiSalvo cites work by Andrew Biggs, an economist at the American Enterprise Institute, which values a dollar placed in a defined benefit (DB) pension plan at more than three times the value of a dollar placed in a defined contribution plan.

This calculation leads to a substantial compensation premium for public sector workers, but this is not due to what the government is paying, it is due to Biggs’ calculation of the value of the benefit to workers. If the government paid the exact same amount for its workers’ pensions, but put the money into a 401(k), by Biggs calculation the compensation premium for public sector employees would disappear.

Biggs is undoubtedly right that the guaranteed benefit has considerable benefit to workers, even if there are reasons for questioning his exact calculation. But one of the ironies of this debate is that most of those arguing for reducing the power of public sector unions are also big proponents of defined contribution pensions. If they had their way, they would quickly eliminate the premium Biggs’ finds, not by saving taxpayers money, but by providing less security for public sector workers.

Historically DB pensions were thought to be a way of allocating risk to the party that can better bear it. It matters little to a state or local government if the stock market is down in a given year, as long as they correctly planned for its average return. On the other hand, it matters hugely to an individual worker if the market is down at the point at which they retire. It can mean a lower standard of living throughout their retirement years. By offering a DB pension, governments assume this risk. This has relatively little cost to the government, but provides a substantial benefit to workers. It would be unfortunate if political pressures forced an end to DB pensions in the public sector instead of finding a way to offer some equivalent security to workers in the private sector.  

Productivity and Efficiency

One of DiSalvo’s main themes is that public sector unions impose work rules and protect seniority in ways that prevent governments from providing quality service. In particular, he focuses on teacher unions, telling readers:

“Take the clearest example of inefficiency: the inability of school administrators to fire incompetent teachers. Indeed, protecting such ineffective workers has the effect of weakening the overall public workforce. Stanford University economist Eric Hanushek found that if middling instructors could somehow replace the bottom 5 to 8 percent of American teachers, the United States would be catapulted to near the top of international math and science rankings (page 35).”

The problem with this story is that there are no unions to prevent the replacement of bad teachers in large sections of the country. States like Texas, which prohibit collective bargaining in the public sector, have not exactly catapulted to the top of international rankings in spite of the fact that unions are not preventing them from firing their worst performing teachers. There is a literature on the impact of unions on educational outcomes, and it is certainly possible that in some cases they have a negative impact, but the idea that teacher unions are the main factor preventing the United States from having outstanding educational outcomes is absurd on its face.

It is also worth noting that countries that do consistently rank near the top in educational outcomes, such as Finland, have much higher unionization rates among their teachers than the United States. Clearly teacher unions are not inconsistent with high quality education.

DiSalvo’s poster child for unions bringing bad outcomes to schools is the confrontation between the Chicago Teachers Union and Mayor Rahm Emanuel in the fall of 2012. This confrontation led to a strike that was settled on terms relatively favorable to the teachers. While DiSalvo deplores the strike and the demands of the union, one important fact that he leaves out is that the teachers largely had the support of the parents. If Emanuel’s proposals really promised better educational outcomes for students, he failed to convince the parents of this fact.

Unions do ensure that workers cannot be dismissed without cause. This undoubtedly keeps some unqualified workers on the job longer than would otherwise be the case. On the other hand, it also protects qualified workers who might otherwise be dismissed for arbitrary reasons or personal disputes with their bosses. Lower turnover will generally be associated with higher productivity, as will an increased worker voice on the job. Preventing dismissal at will does not have to imply lower productivity.

Public Sector Unions and the Corruption of Democracy

Concerns about excess pay and low productivity are symptoms; the real problem in DiSalvo’s view is that public sector unions undermine democracy. The issue is that the unions push governments for better pay and benefits. They then turn around and support candidates who will give them even better pay and benefits, with their power depending on the gains from their past contracts. DiSalvio sees this as a major corruption of democracy.

There are two obvious problems in this story. The first stems from the sums that provide the basis for DiSalvio’s horror story. He highlights contributions from public sector unions in California, a state where they do have considerable power. His chart (Figure 4.10) shows public sector unions contributed roughly $24 million to candidates for the California legislature in 2012. (Table 4.4, on the previous page, puts the number at $7.6 million.) Should the fact that public sector unions can put such sums behind their favored candidates worry us?

In an era where we have billionaires who are prepared to spend tens of millions or even hundreds of millions to support their favored candidates or their own campaigns, is it a distortion of democracy if unions representing hundreds of thousands of workers spend $24 million (using DiSalvo’s higher number) to support their favored candidates? Since he has chosen to focus on California, in 2010 former Ebay CEO Meg Whitman ran for governor, spending $144 million of her own money. Rather than being concerned about the impact of these billionaires, DiSalvio wants us to be worried about public sector workers spending one sixth of this amount.

The other problem is with DiSalvio’s logic about corruption. Of course there is a clear self-interest in public sector workers supporting the candidates who they expect will give them favorable treatment, but the situation of public sector workers is hardly unique. There are all manner of public contractors who routinely give large contributions to candidates with the expectation of favored treatment. In some cases this may be on broad policy issues, for example a charter school operator may donate money to candidates advocating school “reform” in the hope of boosting business. In other cases the contributions are more straightforward corruption, as when a contractor counts on the candidate they support to steer a major contract in their direction. The major difference between public sector unions and contractors in this story is that the contributions of the unions are fully transparent so that people like DiSalvo can report on them. As a result of recent court rulings, private contractors would have little difficulty hiding whatever sums they choose to contribute in support of their favored candidates. Nonetheless, DiSalvo is only worried about the impact of union contributions in distorting democracy.

This is not the only case where DiSalvo’s concern over principles seems to be asymmetric. Like many conservatives, DiSalvo complains about contracts requiring workers to pay a representation fee to a union that represents their bargaining unit. He uses the phrase “right to work,” implying that this condition in a collective bargaining agreement somehow violates an individual worker’s rights.

Of course this one turns logic on its head. Employers impose all sorts of conditions on work. They set the hours workers must work, which they can change at will. They set dress codes, possibly requiring strict business attire or that workers wear silly uniforms. They can even tell workers how they are supposed to address customers or talk on the phone. In all of these cases, the answer to any worker who objects to the conditions set down by the employer is that the worker has the right to work somewhere else. If a worker doesn’t like the terms set by the employer then they should look for another job.

DiSalvio only sees a problem when a condition is being set in a contract negotiated by co-workers. Then he would suddenly have us believe that there is a question of rights for the individual worker at stake. If the co-workers rather than the employer alone play a role in setting the terms of employment then there is suddenly an issue that a worker has the right to work at a particular workplace.

The actual issue here is one of freedom of contract. Can workers sign a contract that requires that everyone who benefits from union representation share in the cost of securing that benefit? DiSalvio argues that they can’t, thereby taking away an important freedom.

The same asymmetry comes up in his discussions on workers’ pensions. In the last pages of the book he attacks the idea of pensions being sacrosanct, even in the context of a bankruptcy like the one in Detroit. A pension is part of a workers’ pay in the form of deferred compensation. The workers in Detroit put in their time and they have good reason to expect that they will receive their pay. As a legal matter, unpaid wages always have top priority in a bankruptcy. In its bankruptcy, Detroit did not review old contracts to try to take back payments from city contractors. It did not review sales of land or other properties to see if it should retroactively charge more money. It is only public sector workers who are expected to give back what they have already earned.

Democracy and Public Sector Unions

In a book that seeks to make bold claims about the role of public sector unions in a democracy, it is striking that DiSalvo makes no use of international comparisons. Countries like Finland, Denmark, and Sweden, have much higher public sector unionization rates than the United States. These countries generally have high quality public services and good schools. They also have strong fiscal positions, with low deficits and low debt to GDP ratios, avoiding one of DiSalvo’s main concerns. Clearly public sector unions have not led these countries on the road to ruin. If there is some inherent conflict between public sector unions and democracy it would be striking if it only seems to show up in the United States. A serious work on the topic of the relationship of public sector unions to democracy would have to explain why they seem to be a problem in some places, as claimed by DiSalvo for the United States, but don’t appear to cause the same problems elsewhere.

Stepping back from this book, the attack on public sector unions in recent years must be understood in its larger political context. Since the election of President Reagan in 1980, the conservative movement has aggressively sought to destroy all the institutional supports for progressive politics in the United States. This drive placed unions directly in the crosshairs.

They pushed a variety of policies designed to weaken private sector unions both directly and indirectly. The direct efforts included a weakening of the National Labor Relations Board and legitimizing the firing of striking workers, reducing the effectiveness of unions’ most important weapon. Indirect policies include the deregulation of heavily unionized industries, such as telecommunications, airlines, and trucking, trade policies designed to put heavily unionized industries in direct competition with low-paid workers in the developing world, and macroeconomic policies that keep the unemployment rate higher than in prior decades.

This effort by the Right has yielded fruit in the form of a sharply diminished role for private sector unions, which now represent just 7.4 percent of the private sector workforce. Having largely succeeded in their attack on private sector unions, the Right is now pointing its guns at public sector unions. While they undoubtedly would like to see public sector workers get lower pay and benefits and have reduced job security, it is at least as important to reduce their ability to support progressive candidates and political movements.

Imagine what elections in even the bluest states would look like if there were no public sector unions to counter the money of billionaires like the Koch brothers? These states would have many fewer Democratic office holders and the ones that were elected would look much more like Republicans, at least on economic issues.

The major political efforts of the last two decades at both the state and national levels would be almost nowhere without the backing of public sector unions. It’s difficult to imagine that any of the state or local minimum wage drives would have succeeded without the support of public sector unions. The same applies to efforts to guarantee benefits like paid family leave or paid sick days. And no one can believe that the Affordable Care Act would have passed without the support it received from the labor movement, including public sector unions.

In all of these cases, public sector unions pushed for policies that benefited workers more generally. For the most part their members already had the gains they sought to secure for other workers. This wasn’t altruism on the part of public sector unions. These are examples of unions acting with foresight, recognizing that it will not be possible to maintain good pay and benefits for public sector workers in a context where private sector workers are poorly paid and without benefits.

This is the main reason why the Right wants to eviscerate public sector unions. It is important for progressives everywhere to understand this fact. There are certainly serious grounds for complaints against public sector unions. There are corrupt officials and time-serving bureaucrats, as is the case with any large institutional structure. But without public sector unions, American democracy would be in a much more dismal state. No progressive should have any doubt on this point.

Support Cepr


If you value CEPR's work, support us by making a financial contribution.

Si valora el trabajo de CEPR, apóyenos haciendo una contribución financiera.

Donate Apóyanos

Keep up with our latest news