Reforming the Structure for Public Sector Pay

Last Saturday, the New York Times published a sad story of another Californian town heading for bankruptcy under the weight of unaffordable pensions and salaries. The figures should astonish even the politically jaded. In a town where the median household income is less than $32,000 the average pay and benefits for a member of the police department exceeded $177,000.

Rich public employee compensation schemes create public tragedies. First, poor people suffer the injustice of overpaying people richer than themselves. Second, when the bubble bursts, a city can no longer afford needed services, leading to higher crime rates and worse educational results. Those with most to lose live in high crime areas and low performing school districts. So the poor take a double hit.

Often the reaction is to blame” greedy” public employees. But this sentiment is wholly inappropriate. These people were just operating rationally within a structure of bad laws. The real question is what can be done to prevent similar tragedies in the future. Here are three suggestions:

First, reduce the legal privileges of public employees, like any right to strike, to require government to engage in collective bargaining, and to collect union dues by coercion. As Max Schanzenbach and I have discussed, public sector unions are not like those in the private sector. In the private sector, negotiations over wages are genuinely two sided with management vigorously representing the interests of shareholders. By contrast, in the public sector the real party in interest—the citizens of the city or state—are generally not well represented at the bargaining table. The elected officials cannot be counted on to negotiate effectively because they want the campaign contributions and political muscle that unions can bring. The recurring dilemma of democratic politics is that politicians often benefit by giving away benefits to such concentrated groups at the expense of the rest of us. Privileges for public sector unions exacerbate this fundamental problem rather than diminish it.

Second, make the salaries, pensions, and health benefits of public sector employees transparent so that citizens can compare them with their own. By law citizens should receive annual summaries of compensation and hours worked by the civil servants who work for them. While most people do not pay much attention to politics, many are likely to agitate when they discover that government workers get more benefits for similar or less work than they do.

Third, privatize public services where possible and prudent. Private companies have every incentive to hold down pay to the market wage. To be sure, privatization risks sweetheart deals with companies rather than with unions. But public contracting laws can temper this problem. Moreover, while private companies can influence politicians with campaign contributions, unions wield two weapons—both resources and an organization that can deliver its members’ votes on election day.

John O. McGinnis

John O. McGinnis is the George C. Dix Professor in Constitutional Law at Northwestern University. His book Accelerating Democracy was published by Princeton University Press in 2012. McGinnis is also the coauthor with Mike Rappaport of Originalism and the Good Constitution published by Harvard University Press in 2013 . He is a graduate of Harvard College, Balliol College, Oxford, and Harvard Law School. He has published in leading law reviews, including the Harvard, Chicago, and Stanford Law Reviews and the Yale Law Journal, and in journals of opinion, including National Affairs and National Review.

About the Author

Recent Popular Posts

Related Posts


  1. Kevin R. Hardwick says


    The devil is always in the details, and the NYT story that provides your opening hook is a bit vague on the details. According to the story, the town in question devotes 7 million of its 10 million total to police department salaries. The story also specifies that there are 39 police officers in the city force. So simple math does indeed produce the eye-catching number of 177,000 in pay for each police officer.

    The remainder of the story, however, implies that the real problem is not the salaries that police draw, but rather the very expensive pensions offered. The focus of the story, in other words, is not on bloated salaries, but rather on bloated pension benefits. Here, however, the story provides us with almost no data. What are the typical pension benefits provided to police officers? Are state of California pensions out of line with those of other states? Why are pensions for police officers more generous that those for, say, public school teachers, or city administrators? The story does not say–but without this information, we cannot adequately assess the implications of the story.

    Finally, the story is solely focused on the salaries and benefits of police officers, and not public officials more broadly. To generalize on it, as you do in your blog post, is to expand the scope of the analysis in a fashion that may not be warranted by the underlying data. Since the story does not tell us much at all about the salaries and pension benefits of other public employees, it is hard to tell whether there is a meaningful problem there that merits the kinds of solutions that you propose.

    So the problem with your blog post is a disconnect between the opening “hook,” which you introduce to suggest that there is a real, broad problem that requires solution, and the broadness of your following analysis. The news story, when read closely and carefully, implies that there is a specific problem with police pensions in California, and perhaps elsewhere too. The story implies that the financial problem that the town in question confronts (and other towns in California too) is a problem created by the police pension obligations that the town has chosen to assume. The solutions you suggest in your blog post might or might not adequately resolve the issue–for us to know that, we would need to know a lot more about the trade-offs involved. For example, what is the norm for police pensions in other towns in the state, and why is the norm what it is? What happens to the quality of policing when a town chooses not to offer normative pension benefits? (The NYT story suggests that some people–likely those with a stake in keeping generous benefits–argue that police quality declines when pension benefits are cut. They argue in essence that you get what you pay for–which a priori sounds plausible. But is it really true?)

    In summary then, I find your three proposed solutions to be plausible enough, and to make a fair bit of sense. Implementing them strikes me as theoretically sound, and from the perspective of theory, to be conducive to good public policy. But your post also implies that there is an actual, empirical issue here–that the problems are not just theoretical or hypothetical, but actual and real. And here the evidence does not support the generalization you have made. The actual problem seems to have something to do with the scale of police pensions–and we just don’t have the evidence available to us, from the NYT story, to know one way or the other whether or not the theoretical solutions you suggest would in fact resolve that.

    All best wishes,

  2. gabe says


    right on all counts!
    however, one should not expect the NY Times to provide sufficient information as that would, perhaps, go against their unique political agenda / party affiliation.
    Yes, the problems with pensions have been documented to be a statewide issue affecting all political subdivisions within the state (NRO has had some decent accounts on this as has Forbes and IBD(?). This is not just a California problem but a national one – however, the fruits of union – Democratic coalition efforts appear to be coming to fruition somewhat sooner in the Golden state.
    Much of it has to do with political deals made with the Democratic controlled state legislature (Goodness, I wish my memory were better so that I could refer you to these essays – although I think our own M Rappaport did a piece – also see Originalism Blog about 2 months back for some discussion) which effectively sold out the citizens of the state by CONTRACTUALLY guaranteeing certain aspects of pensions and promised returns or COLA’s.
    In many instances, the effect upon municipalities is devastating.

    A good while back, I had suggested something similar to what Prof McGinnis advances here. Public employees compensation should be subject to approval of the citizens who are “gunna foot da bill.”

    take care

  3. Kevin R. Hardwick says


    My guess–and its no more than that, given the sloppiness with which the NYT article was written–is that the California city that is budgeting $7 million towards the costs of its 39 person police force is not in fact spending an average of $177,000 per officer. Rather, I’d guess that some significant portion of that $7 million figure is going to defray pension benefits for officers no longer actively on the force. If that is the case, the typical police officer on that town’s force is making less, and likely considerably less, than $177,000. But this is all supposition–the facts simply are not in evidence, so what the real situation is is anybody’g guess.

    Let me give an example here that is closer to home, and for which I have a more or less decent handle on the facts. I live in Harrisonburg, Virginia, population of about 51,000. The average salary for a police officer here is right at $50,000, exclusive of benefits. But the force is smaller than the level of crime merits–the city chooses to have fewer police officers, and instead reaches the requisite hours of policing by paying targeted overtime. Apparently–the city financial officer has the figures to back this up–it makes better sense to pay high overtime, and to have fewer full time police, than it does to pay less overtime and more police. So the average salary is misleading–most police make more than their salary, because many of them work overtime hours. I don’t have the full figures, but when you count overtime and the full range of benefits, it would not surprise me if the cost per officer to the city is more like $90,000.

    Is that a reasonable amount? Well, on the whole I am glad that we have police, and moreover, I want the police that we do have to be high quality. Police, after all, have the very real potential to behave abusively, and speaking as someone subject to the police power, I want to minimize that as much as possible. According to the former police Captain, whom I know moderately well because he attends the same church as I do, the best way to do that is to hire well trained officers. The captain tells me that he preferred to hire men and women who had prior police experience; but absent that, to hire those who had done law enforcement in the military; and absent that, to hire men and women who had finished at least one tour in the armed services. But even when you hire people with those qualifications, you still have to invest a fair bit in training them. If you pay too little, then much of that training goes to waste, because other jurisdictions that pay better will poach your trained officers away from you. So there is something to be said for paying a prophylactic wage, in order to prevent having to maintain an exorbitant training budget.

    To some degree then you can’t treat Harrisonburg in a vacuum. What H’burg finds reasonable and efficient to pay is going to be determined in relationship to the broader market for police officers in the state–and this is true no matter what the median family income of the city is. The city can cut corners to some degree, but in the long run only by being willing to accept shoddier and more less professional police work from its force.

    So what is a reasonable wage for a police officer? In Virginia, starting pay is in the $32-38,000 range, and median pay is in the $47-52,000 range, before you factor in overtime and benefits. Is that a lot, or a little? Is that reasonable, or unreasonable to pay for quality police work?

    The NYT essay implied that the problem in California–or at least in this one town–was *not* pension benefits for *all* public employees, but rather those for law enforcement specifically. So to the extent that this particular essay is the warrant for the suggestion that we face a serious public problem, the problem pertains to a pretty specific group of public employees. But even here, it is hard to gauge ethically or economically just how extensive the problem is. To some degree, it boils down to the question of what is a reasonable to pay for the guy who puts his life on the line to provide public safety for the rest of us? Just speaking personally, I’d venture to say “more than the average public university college professor.”

    All best wishes,

  4. gabe says


    You are certainly correct with respect to the $1777,000 figure. I would be quite surprised if that were in the fact the authorized or standard wage for a police officer.
    The total monetary expenditure surely involves the costs of administration, training and ‘civilian employees and if these factors are not considered the mean and median wage for the “police” tends to be inflated.

    However, and this is a big however, many union / retirement contract provisions allow for the determination of retirement benefits based upon the final 5 years of service. consequently, the unions exert pressure on the various police (and fire, sanitation, etc) departments to assure that the preponderance of overtime is offered to senior employees. This is certainly rational economic behavior on the part of the “soon-to-be-retired” as it inflates their pension base wages considerably. This is true both in my home state of Washington and in California. Boyhood friends who worked as police officers in NYC indicate that this is also the case in NY. (Oops, I forgot, similar provisions apply for correctional officers in both CA and WA.)
    So, yes, these municipal employees do not get $177,000 per annum during the bulk of their career; however many do receive a figure approaching this during their end-term employment.
    And when one considers what our Congressional friends have just done to military pensions – my goodness, it is quite upsetting. Of course, military retirees do not have powerful unions making political contributions to our elected masters – so what does one expect.
    My memory fails me a bit here, but as I recall, there were / are specific provisions enacted into California law (perhaps even constitution) that were quite favorable to unions and pensions which have the effect of guaranteeing unreasonable returns on pension funds and thus assuring a high base for those retirees fortunate enough to have earned them.

    Some rather interesting discussions over at the Originalism Blog, with longer essay at SSRN concerning the legality of breaking these contractual provisions.
    I admit it is beyond my little mind to come up with a solution. However, I think Mr McGinnis’ proposal could be a start. Why should we, who are ultimately tasked with paying for these pensions, not have a say in determining what they shall be.

    Anyway, take care

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>