Wisconsin’s Smoking Gun

smokinAre public sector workers overcompensated? A month ago, I’d have said “probably”. Today I think we’ve found the smoking gun.

Here’s what I knew a month ago: Public-sector quit rates are roughly one-third of their private-sector counterparts. The obvious explanation is that public-sector jobs are generally too cushy to walk away from. It seems to me that it would be hard to account for that factor of three in any other way, though you can see some reasonable attempts in the comments here. (To be clear: I think that some of the factors in these comments can reasonably account for part of the difference in quit rates. I find it implausible that those factors are collectively substantial enough to account for a factor of three.)

A month ago, that was the best evidence on the table. Today, thanks to the protestors in Wisconsin, we’ve got something like proof positive.

Here’s why: If you cut the pay of an overpaid worker, he’ll generally scream bloody murder. After all, overpaid workers like to stay overpaid. But if you cut the pay of a non-overpaid worker, you haven’t really damaged him. He just quietly leaves and gets a job elsewhere. After all, the ability to find a comparable job elsewhere is pretty much the definition of not being overpaid.

Now how are the Wisconsin public workers reacting to projected pay and/or benefit cuts? As if the rug’s been pulled out from under them, that’s how. Every time a worker says “These cuts will cause me severe pain”, that worker is saying, in effect, “I can’t get anyone else to pay me at the level I’m accustomed to”, or, in briefer words, “I am overpaid!”.

So yes, they’re overpaid. And the louder they get, the surer you can be.


74 Responses to “Wisconsin’s Smoking Gun”

  1. 1 1 MattW

    Are there any public sector jobs that are specialized enough that no private sector alternative exists? If so, does the argument still stand that non-overpaid will simply leave and get a job elsewhere?

    My sister is a social worker with the state she lives in, and I can’t think of a private sector alternative. But even if there is one, are there any public sector jobs that have no comparable private job? Or does that even matter?

  2. 2 2 Nathan

    An argument premised on the perfect efficiency of labour markets would be more compelling at a time when unemployment was not 9.8% despite a declining participation rate.

    You really can’t think of any other explanation for why a large number of workers, many of whom have very specialized skills, may not be keen on switching jobs right now?

  3. 3 3 Bennett Haselton

    Also, even with full employment, the fact that public sector workers are paid more than they could make in the private sector, does not necessarily mean that their public sector salaries are wasteful — couldn’t it also be that their services create positive externalities, as is probably the case for teachers? (i.e. when a person gets educated, some benefits accrue to them, and other benefits accrue to the society around them.) Then presumably it’s economically efficient to pay teachers a higher salary — commensurate with the value of the education to the student, plus the value of the positive externalities — which is higher than what they would get paid in the free market, which would only be the value to the student (or to their parents).

  4. 4 4 Harold

    What’s the quit rate for economics professors?

  5. 5 5 Thomas Purzycki

    @Bennett: On the other hand, theoretically, the government hires teachers precisely because they are trying to provide a public good that the market doesn’t provide enough of. The fact that many teachers have jobs in the first is because the government is already paying more money than the market would.

  6. 6 6 Coupon Clipper

    I know this wasn’t your main point, but I always assumed that the reason their quit rate was lower was because so much of their compensation was deferred comp, i.e. a pension. It’s expensive to walk away from a job like that, but that doesn’t necessarily mean they’re overpaid.

  7. 7 7 DBeno

    The politicians who are determining the fate of the public workers in Wisconsin are elected and are therefore more susceptible to public opinion. Large scale, media attracting, protests can be effective here in ways that are not available to private sector workers. It’s not surprising that Wisconsin public workers are visibly more upset than non-unionized private workers in trying to save their jobs\benefits. They can reasonably expect that their complaints (if they can raise them to the point of a broad, sympathetic awareness among voters) may be more effective than they would be for others.

    For a variety of reasons, most likely their protests and complaints will not be successful. In any case, it certainly looks like it takes a lot of work to take over a state capital building. Nevertheless, its a potentially effective recourse for these workers. So, while it’s true this is a job-security advantage and therefore a type of benefit, it’s no more than one more (seemingly minor) factor in calculating if these workers are overpaid overall.

  8. 8 8 Harold

    If the Govt. is virtually a monopoly employer, then quit rates are not directly comparable with private sector ones. Say, teaching, for example. If quitting for higher pay can only be done by leaving the profession, then this is not the same as people switching employers to do the same job. The lower quit rate may indicate that they really, really like their profession compared to the alternatives. However, we do not have the comparable data for how much people love their *profession* in the private sector, only their jobs.

    The problem with a broad brush comparison is that different quit rates are appropriate for different sectors. There is a cost to turnover, but this varies a lot. A waiter can switch jobs easily, and the employer can replace him easily. There is not too much cost involved, so a high turnover rate is appropriate. If a skilled production worker quits, he cannot easily be replaced, and there will be significant cost in training his replacement. In this case, a low turnover is appropriate. Looking at the statistics, ( http://www.bls.gov/news.release/jolts.t04.htm ) it is clear that manufacturing has a quit rate much lower than retail or leisure, and only slightly higher than public sector, exactly as we may expect if employers are trying to optimise turnover. I don’t believe that quit rates alone are sufficient to say that an entire group is overpaid.

    I also think it quite likely that some public employees are overpaid, but I don’t think this is quite the smoking gun you say it is.

  9. 9 9 Steve Landsburg

    Bennett: I do not understand your externality argument. Given a pool of identical workers, some of whom are producing $100 a week in the private sector and others of whom are producing $150 a week in the public sector, you’d want to move the marginal worker from the private sector to the public sector (i.e. you’d want the state to hire him), and this should continue until the two productivities equalize. So in an efficient allocation, the output of the public school teacher (inclusive of externalities) should be the same as it would be in a private sector job.

  10. 10 10 math_geek

    I don’t find this argument at all compelling.

    As Coupon Clipper said, deferred compensation is a pretty big part of public sector pay, to the point where people really do quit in large numbers once they fill out those benefits. The correct response of a potential employee to an underpaid government sector is to not take the government job in the first place, not to quit and take another job.

    Second, this is totally ignoring any transaction costs of finding another job. People can’t continuously keep searching for jobs in real life because switching jobs quickly sends a poor signal to future employers. Taking a job, working well, and not getting a raise after the first year is frustrating for many because they are caught between sending a bad signal to potential employers by switching jobs or accepting a salary that is lower than what they are worth.

    Third, there is no guarantee that some people aren’t planning on leaving their jobs in Wisconsin. All that could tell you is that some government workers are overpaid. Not much different from the public sector actually.

    Fourth, it is of course, the case that the protests are not at all about benefit cuts, which every union has claimed repeatedly that they are willing to accept. It’s the loss of collective bargaining privileges that has all the people screaming bloody murder. Note that even unions that are exempt from the change in the law are screaming bloody murder. Sometimes, people actually have principles that are worth screaming for, and for many, the ability to negotiate collectively with your employer is one of them.

  11. 11 11 Michael

    Steve, this is somewhat tangentially related to your point (screaming bloody murder about benefits being taken away), but more in the area of performance bonuses.

    Psychologically, people hate losing something they have much more than they like gaining something they don’t have. Or in other words, people will work harder not to lose something than they will to gain something extra.

    In addition, from my own observations, other people tend to be less jealous if someone is lucky not to lose something (say, not lose their house in a flood) than they do if they gain something out of luck (say, win the lottery).

    So anyway, to make the observation concrete, let’s consider these two possibilities with regard to teachers:

    1. Give all teachers $5000 a month. For those who meet a certain standard, give them a bonus of $1000 a month.
    2. Give all teachers $6000 a month. For those who don’t meet a certain standard, make them pay the district $1000 a month.

    Even though both are equivalent (and glossing over the difficulty of deciding what is a suitable standard), I would say that people would work harder in case 2. It’s their money, after all, and they are at risk of losing it. In addition, I think that if there are teachers Anna and Bob, Bob would be more jealous if Anna gets a bonus that he doesn’t get than he would be if Anna doesn’t have to pay a “fine” (or whatever) that he does. At least, I think that would be the case.

    Of course, whether or not their collective bargaining unit would allow lack-of-performance fines when the hate performance bonuses is another question.

    (You may have made a similar point in an earlier post, but I can’t remember it.)

  12. 12 12 Ken B

    Mthe_geek writes
    “Second, this is totally ignoring any transaction costs of finding another job.” But this isn’t true because private sector workers face the same costs.

    His point 1 is even worse. Think about it. Accruing such benefits is part of the pay. If m_g’s numbers are right then the quit rate is even lower while acctruing benefits, strangthening Steve’s argument.
    Point 3 is obscure to me. Wisconsin public workers are not more overpaid than the average overpaid? A six foot five man is not tall by NBA standards, so he’s not tall.

    Point 4 does not address the point at issue: are they overpaid. (It’s also not quite accurate; only some collective bargaining power is being curtailed, bargaining over wages is not.)

  13. 13 13 kzndr

    As I understand it, the public sector unions in Wisconsin have accepted the cuts to their compensation proposed by the governor. What they’re crying bloody murder about is the attempt to get rid of collective bargaining.

  14. 14 14 Cos

    “Now how are the Wisconsin public workers reacting to projected pay and/or benefit cuts? As if the rug’s been pulled out from under them, that’s how.”

    False. Here’s how they’re actually reacting:

    Okay, we’ll take all of the pay and benefit cuts you proposed. But we won’t give up collective bargaining rights.

    Yes, you seem to have missed it, but the unions in Wisconsin offered early on to accept *all* of the pay and benefit cuts the Governor proposed. He refused to back off on eliminating their collective bargaining rights, and that’s what they’re protesting.

    No smoking gun here.

  15. 15 15 Will A


    “Now how are the Wisconsin public workers reacting to projected pay and/or benefit cuts?”

    Many state are negotiating with labor unions to reduce costs. These state capitols aren’t being overrun.

    The California Service Employee International Union workers agreed pretty quickly to concessions when faced with layoffs. California has also furloughed its state worker which has resulted in a 15% cut in pay. This happened without workers storming the Capital.

    Is your point that California public employees are underpaid since they haven’t screamed bloody murder.

    It might be possible however that Wisconsin Public Employees are open to pay/benefit cuts and what they are really reacting to is the loss of collective bargaining rights.

    It might be possible that people react stronger to a perceived loss of a right than they do to a loss of income.

  16. 16 16 Will A

    To be clear, I think your argument that looking at turnover is a good indication of compensation level. I’m just not convinced that screaming bloody murder is an indication of being over paid.

    Using turnover rate on public employees wouldn’t show anything unsurprising:
    - Teachers in poor areas are under paid.
    - Soldiers are under paid.
    - Police and Fire Fighters are over paid.
    - Teachers in wealthy school district are over paid.

  17. 17 17 Ken B

    One of the themes of Steve’s writings (as I take it) is that informal arguments can often be made precise and quantitative; this allows them to be assessed more accurately. “Most people would be happy to have that deal” can be translated into a quantitative measure in several ways; one is the quit rate. We see a large and persisten quit rate difference, so we now know this argument is a strong one.
    Steve is doing a simialr thing now with “overpaid”. Overpaid is not a value judgment. It can be more precisely defined and measured. If you cut my pay or working conditions by a significant amount (to overcome transaction costs) and I don’t move on because I cannot find anything better (even after the cut) then I was overpaid. And I’ll scream instead. And we see that very strongly.

  18. 18 18 math_geek

    @Ken B

    That’s true about private workers facing the same costs, point 2 doesn’t hold up nicely with regards to quit rates, but my thinking was more along the lines that people get really upset about a cut in benefits or salary because even if they find they are now underpaid, they still have to eat the transaction costs in order to rectify that.

    Your refutation of my point 1 indicates a failure to understand point 1. If a benefit schedule is such that 10 years of continued employment earn a cash bonus then people who have worked 9 years are quite unlikely to quit, since that’s a one year to accrue the benefit as opposed to 10. Vesting benefits specifically stop people from quitting their job more effectively than they do at attracting new employees. That may be desirable, but it’s going to result in a lower quit rate regardless of whether or not the workers are overpaid, as the people who are underpaid don’t leave. That says nothing about attracting new employees (or the quality of said employees).

    Point 3 is that we can account for all the people who are protesting, but not for all the people that are not protesting. It’s pretty clear that the majority of the public workers may be annoyed, but they are not screaming bloody murder. Prof. Landsburg takes one subset of public workers and attributes their qualities to the entire class. Some athletes are overpaid… Barry Zito, Jason Werth, Albert Haynesworth, Jamarcus Russel come to mind. That must mean that all athletes are overpaid. In any field of course, some people are overpaid and others are underpaid, as identifying exact productivity levels is very costly. A tenure system increases the cost of removing the overpaid workers. The best teachers in Wisconsin might very well be teachers in Minnesota in 2 years. The worst teachers might be the ones screaming the loudest.

    Point 4 does address the issue of whether screaming bloody murder implies that they are overpaid. The implicit assumption is that only people who are overpaid scream, therefore screaming implies that one is overpaid. However if people scream because they believe that collective bargaining is a moral obligation of employers, then it is no longer the case that screaming implies overpaid.

    As for your side point, limiting collective bargaining to wages is stupid. If additional sick days are worth more to workers than they cost employers then negotiation will result in additional sick days. Shutting off negotiation on non-wages results in inefficiency and increasing tax payer costs compared to full bargaining rights. The very fact that this is what is on the table implies that Gov. Walker cares about something other than controlling costs and efficient government.

    Finally, as to the overpaid underpaid spectrum. I find the debate to be very silly, as it always assumes that teachers are all more or less identical. It is certainly the case that we could hire teachers at lower wages, but they would be less effective teachers as more talented people would be off doing other things. If you paid teachers more, you would attract more people wanting to become teachers, and would have a more talented teacher pool. What is optimal here is a harder question, and a very interesting one.

  19. 19 19 Loweeel

    math_geek, if teachers are not more-or-less identical, all the more reason to eliminate collective bargaining, which really only saves transaction costs of individual bargaining in cases where the labor itself is commoditized rather than individualized.

  20. 20 20 math_geek


    You would have to scrap our entire civil-service system first in order to get truly individualized bargaining. And that civil-service system was put in place so as to fix the political rewards system that preceded it. Of course, you can institute policies designed to reward to teachers that appear to be the most effective, and collective bargaining can still give insight as to what the “best” number of sick days is.

  21. 21 21 Will A

    Ken B:

    For sure I hope you don’t take this as a strawman question/point. I don’t mean it as such.

    I’m really am with you that looking at turnover can help to quantify if someone is over compensated.

    However, CISCO has a pretty low turnover. Does this mean that CISCO overpays its employees? It would seem that the answer is obviously yes.

    CISCO however seems to be a relatively successful company and overpaying their employees seems to be a thought out strategy.

    The way I read math_geek’s point is that we kind of need to balance a person’s compensation level with the level of skill we want from the position.

    If the Wisconsin law passes, it would be interesting to track the the percentage of teachers graduating from Wisconsin’s top schools that take jobs in Wisconsin versus other states that have better collective bargaining rights.

    This wouldn’t necessarily point to compensation, but it might point to competitiveness in a state’s ability to attract teachers based on a state’s collective bargaining rights.

    My personal (unfounded) hope is that the Wisconsin law passes and graduating teachers from top WI schools move to California.

  22. 22 22 Ken B

    Math_geek: “Finally, as to the overpaid underpaid spectrum. I find the debate to be very silly, as it always assumes that teachers are all more or less identical. It is certainly the case that we could hire teachers at lower wages, but they would be less effective teachers as more talented people would be off doing other things.”

    Steve’s argument is that the evidence suggests we could hire the SAME teachers at lower wages. Hence they are overpaid. That is what overpaid means.

    Math_geek: “Shutting off negotiation on non-wages results in inefficiency and increasing tax payer costs compared to full bargaining rights.” This would be true in a true market. It is not the same with public sector unions as the other negotiating party has different incentives than a private employer would have. This is a well known phenomenon.

    As for point 4, yes your point is logically well taken. It could be that all the protestors are yelling for some other reason. They could be objecting to Walker’s necktie for instance. They say it is principle, but that principle corresponds exactly to leverage. To me it looks like all the stuff about “uinons gave us the weekend” is just cover. So my initial criticism of point 4 was wrong: the point is logically coherent, it is just not factually correct.

  23. 23 23 math_geek

    @Will A

    Costco was really panned by Wall Street a couple years back for having benefits and wages way above the then industry standard. Costco defended itself by saying it had among the lowest levels of employee turnover and employee theft, and that the savings from those benefits well outweighed the increased wages. They’re now the 3rd largest retailer in the US.

  24. 24 24 Will A


    It would be interesting to know if there is a correlation between profitable companies and low turnover.

    If companies that have lower turnover are more profitable on average and if – as many conservatives are saying – we need to run government like a successful business, then we would want to come up with a compensation system that has a low turnover of government employees.

  25. 25 25 math_geek

    @Will A

    I don’t know about any statistical study that says that low turnover is good. But common sense should dictate that it is. Not all jobs are created equal, and switching jobs has built in transaction costs for both sides. New employees need to be trained, and I’ve heard stories from businesspeople saying it takes 2 years for an employee to become fully productive. If training is a cost, than low turnover is beneficial.

    As for wanting to come up with a compensation system that has a low turnover, well, that’s exactly what we have in our federal government. Personally, I think that we probably overdo it, and we could benefit from a shift in vested benefits to wage benefits and certainly we should make it easier to fire unsatisfactory employees, but I’m totally aware I can’t argue that point very well.

  26. 26 26 Ken B

    Not just costco. Henry Ford started the $5 workday to retain trained employees.

    In a competitive market some employers will pay more for less turnover: wages & benefits are the not whole cost of labor. It will happen more don’t you think when there are many employers cometing for workers. Does anyone doubt that?

    Well it won’t in some cases. If some workers are sheltered from the market and overpaid then they will not benefit from more competiton for their services. The boss’s nephew may be better off at his uncle’s company no matter what.

    And teacher union workers may be better off no matter what. Will A and math_geek deny this. But they also object to school choice, which would create more competing employers for teachers. So they don’t see more competing employers creating better choices for teachers. Like I don’t see more employers creating better choices for that nephew.

    I do though see more employers who do the job more efficiently putting the uncle out of business and costing the nephew his job. So I can see him and his partisans actually objecting to more competition.

  27. 27 27 Ken B

    @Will A: I know nothing about CIsco so I don’t know if they overpay. Not being an owner of Cisco I don’t much care and I certainly make no objection. I hope they buy out my employer though.

    As a taxpayer I think I may object if the government overpays, either directly or with deferred payments.

  28. 28 28 Will A

    Ken B:

    I don’t object to school choice. I object to a private school (private company) being required to accept any student that applies.

    In my opinion, private schools should have the right to accept/reject who they choose.

    I’m also against any community that attempts (explicitly or implicitly) to prevent anyone/group from moving into a their community because that community has a public school desired by said person/group.

    Right now, Americans are free to choose to move to any community that has a school that they feel meets the needs of their child. They are also free to choose any private school that will accept their child for admission.

    And as it relates to this blog, I think that American Teachers should have the ability choose to pursue a job in any school in any state to which they would like to apply.

  29. 29 29 math_geek

    I like how “school choice” boosters always assume that once the voucher system gets put in place, there won’t be massive pressure to regulate the entire private school industry.

    If you take the governments money, you take their regulations too. I’m sure Sidwell Friends would love to be informed they no longer have the right to discriminate against learning disabled children as per the Americans with Disabilities Act.

  30. 30 30 Will A


    Ken B didn’t mention anything about a voucher system (on this post at least). He just said that he was in favor of school choice.

    He could have been talking about:
    - Open enrollment to allow any child to go to any public school they want.
    - Charter Schools
    - Magnet Schools
    - The right for a parent to do homeschooling.
    - The ability for people to live in any school district they choose.

  31. 31 31 neil wilson

    Can someone explain what I am missing?

    You and I are equally qualified. We both should get $50,000 a year.

    You and your employer decide that you should get $40,000 in cash and $10,000 in retirement benefits.

    My employer and I decide that I should get $48,000 in cash and $2,000 in retirement benefits.

    Now, both of our employers decide that they can’t afford to pay us $50,000 anymore.

    They decide to cut our pay by $3,000.

    Which one of us is more likely to quit?

    I would guess that I am more likely to quit because it won’t cost me very much in lost retirement benefits. Maybe neither of us has vested yet. I would lose 20% of what you would lose. So I would be far more likely to quit. Maybe we have a defined benefit plan where the longer you stay and the older you are the more valuable the contribution.

    Now, in reality, my guess is that the public employee is getting paid more than the private employee precisely because of retirement benefits.

    Corporate accounting used to not do a good job of calculating the value of future benefits. That made it easier for a private employee to give someone $1,000 in retirement benefits than $1,000 in salary. It is still true today with stock options. You can give an employee stock options without having to expense it.

    Governments are interested in balancing a budget. They aren’t interested in making logical choices. Why should Indiana be able to do a sale leaseback on the tollroad and have that help their budget? Sure, it helps their cashflow but they had a $1billion asset and now they have $800 million in cash. It seems to me they lost $200 million but the budget looks far better.

    In any event, government can get better employees at a lower cost by promising a big pension that won’t be paid for years. It used to be that government didn’t even have to calculate the unfunded liability. It is getting better but it is still easier for a government to cook the books than a private business.

    So, getting back to Steve’s main point. Why don’t government employees quit? Because they made a long term deal to defer some of their income and if they leave now, they will lose a lot of promised benefits.

    On a side issue, a lot of bankers took huge paycuts when the market tanked. Why didn’t they leave if they weren’t overpaid? Well, they were, and still are, overpaid.

  32. 32 32 math_geek

    @Will A. I suppose that is true, but catchphrase interpretation stands I think.

    Well just for fun…

    A – We can do this and not have it be an absolute mess? How does it even work?
    B-E, I’m pretty inclined to support all those things, although Charter schools haven’t shown very much empirical improvement over the normal public school system.

  33. 33 33 Will A


    Also as it relates to this post and school choice, I think of charter school teachers as non-union independent contractors.

    They seem to trade union benefits for the ability to have more freedom in how they teach.

    It’s possible that giving teachers more freedom (choice) in how they taught and not requiring them to teach to tests would also might reduce the screams of bloody murder.

  34. 34 34 Will A

    math_geek (last one I promise):

    Robert B. Reich has a different take on vouchers. I checked out his “Aftershock” book from the library after listening to an interview of him on my favorite NPR station.

    If you are are looking for a progressive take, he recommends giving kids vouchers for public schools inversely proportionally to their parents income level. E.g. poor kids get $ 12,000, rich kids get $ 6,000.

    He argues that this would either:
    - Help poor public schools.
    - Encourage rich public schools to open slots for poor kids to get access to their vouchers.

  35. 35 35 Josh

    The issue is collective bargaining per se. Why should individuals not allowed to collude when collusion is implicitly allowed to occur in the business world? ..is it illegal for wal-mart to say “we sell for less”? No, but it’s a form of collusion.

  36. 36 36 bart.mitchell

    Interesting way to look at it. Lets apply the same logic to other professions.

    How many CEO’s quit their jobs? Almost none, they have to be forced out of those jobs. Conclusion? CEO’s are over paid.

    How many University Economics professors quit? Very few, they too are obviously over paid.

    We’ve identified at least two more groups that need their wages and benefits cut. Who else is next?

  37. 37 37 scott

    First, you base your entire point on the idea that the workers are picketing a pay cut. THEY OFFERED TO TAKE A PAY CUT AND WERE IGNORED. So, that means the pay cut (i think it was about 5%) was clearly much, much less than the gov had in mind. thats why he wants to gut the unions, so he can DRASTICALLY cut their pay.

    Second, you assume that being overpaid is the ONLY reason someone would want to stay at a job. again, simply wrong. there are many other factors to employee loyalty. If they are overpaid or not, that is not something im going to argue. Clearly they are since, again, they OFFERED to take the pay cut.

    Third, you assume that all public works hold skills that are applicable to the private sector. Some are, but not all, a good amount of public workers cant take the same skills in the private sector. They would have to shift to a job with little or no experience.

    So, in the end. You are being paid by the GOP/Tea Party to open your mouth and type out blatantly inaccurate info on the situation. You are being paid to create propaganda. I hope you sleep well at night, you are the perfect example of what is wrong with the media today.

  38. 38 38 Peter John

    People who are passionate about their jobs then to stay.

  39. 39 39 Andre

    What does this have to with what is going on in Wisconsin? They agreed to the pay cut. They just don’t want collective bargaining to become illegal. It seems to me that ought to be a basic human right.

  40. 40 40 Babinich

    “If you are are looking for a progressive take, he recommends giving kids vouchers for public schools inversely proportionally to their parents income level. E.g. poor kids get $ 12,000, rich kids get $ 6,000.”

    Why are the “poor kids” leaving their school? Threats of violence, student under performance, teacher under performance, poor infrastructure, inadequate teaching supplies/materials?


  41. 41 41 Harold

    There are two issues here: do quit rates indicate if someone is overpaid, and do the protests indicate the same?

    Last first: the Wisconsin workers are apparently protesting about other matters of principle, so this does not necessarily indicate they are overpaid. You could argue that the principle they are demonstrating about is the right to collectively bargain so they can remain overpaid. However, we do not know the motives of the individuals, and even if collective bargaining did result in overpayment, the people concerned may not believe this. So the volume of their protests does not necesarrily correlate with their level of overpayment.

    So back to the quit rate issue. As I mentioned earlier, quit rates vary significantly in different sectors: e.g. Government 0.5, manufacturing 0.7, construction 1.3 and leisure and hospitality 2.9. So does manufacturing overpay, and as mentioned by others, does Cisco overpay? No, the difference is probably better explained by different costs of turnover and the benefits of employee retention. Steve’s point in the original post was that Krugman was wrong to correct for education, whilst ignoring other factors such as work hours. Quit rates would take into account all these factors. However, the different quit rates in different industries shows that this is not the case: we still need to correct the quit rate data for costs of turnover. A reasonable proxy for this might be education level – it costs more to replace more educated workers as they are more likely to have specialised knowledge and experience. Oh dear, this brings us right back to using Krugmans education level correction factor.

    Steve does acknowledge that a direct comparison is not applicable, and that some difference would be expected, just not so large a difference as is actually seen. If I assume for now that manufacturing does not overpay (due to Govt. subsidies or similar), then the difference between 0.7 and 1.6 (the overall average private sector quit rate rate) can be explained by these turnover cost factors. So can we explain the difference between 0.5 (public sector) and 0.7 (manufacturing)? I don’t know, but it is plausible from the arguments given above by others. More detailed examination of the different quit rates in sub-sectors of private and public employment would be required.

  42. 42 42 skip

    “But if you cut the pay of a non-overpaid worker, you haven’t really damaged him. He just quietly leaves and gets a job elsewhere”

    I assume you have some evidence for the mind-bogglingly contentious empirical claim that properly paid employees don’t care about wage cuts.

  43. 43 43 Mick

    It’s collective bargaining stupid. The protesting is not about pay nor benefits. They already made concessions regarding those areas.

  44. 44 44 John McCullough

    You’ve missed the point entirely. The public sector unions have agreed to make concessions on their wages and benefits, but are fighting to keep their collective bargaining rights.


  45. 45 45 teachermant

    I must admist that you pose an intruiguing argument. There are some merits to your thought process. In some industries, there are overpaid workers. However, I do not agree with the context in which you place your argument. Public service workers are not blue collar workers. In these industries, people have spent thousands of dollars in higher education in order to perform their jobs. In light of this fact, just compensation is due. I am an educator in the State of Connecticut. Teachers are required as well as other public service workers to be highly qualified. In order to do this, you need to continue in the process of education. I have a huge problem with all of the responsibilities that we have incurred and as a result, the monetary compensation has not increased! I all I want to say is that teachers earn their money. Every penny!

  46. 46 46 Ken B

    @Will A: Yes, thank you. There are more ways to get school choice than just vouchers. They can even be combined in various ways. And we have 50 states to try things out in. Personally I like the voucher idea plus having public charter schools. But I have no prior committment to the *mechanism* as long as the goal — competition, accountability, and therefore better results — follow.

    And math_geek may be right that some “anti-discrimination” laws impede progress, and I am glad he sees the value in repealing them! In any case if you need to change law X to get school choice then I will consider if we really need law X. It’s like gay marriage: you really need to repeal some sodomy laws for it to work. Should that be an impediment?

  47. 47 47 Harold

    There is one fundamental problem with school choice – the recipients of the education do not get to choose. It is the parents that do. If the parents do not value education (perhaps because they do not have it themnselves), they may put a low value on it, to the detriment of their children. School choice assumes a complete alignment with the welfare of the children and the parents.

    Is it not at least in principle possible that it is necessary to overrule the choice of some, so that their children can become educated enough to make valid choices?

  48. 48 48 Chicago Methods


    That isn’t the argument at all. Steve’s argument is that how loudly a person complains about a loss of wages, or a loss of their job, is positively correlated to how overpaid they are. While I too question what seems like simplistic logic, at the very least we should be able to get the argument correct.

    On a side note, wouldn’t this deal more with economic rent or am I way off base? If this does deal with economic rent, that might be a place to start.

  49. 49 49 Jim from Wisconsin

    I am a state employee in Wisconsin, working just down the street from the protests at the Capitol. I am directly affected by the bill Governor Walker has proposed. I am not a member of the union but many of my coworkers are. Your argument is based on very faulty logic. It’s also based on, what I presume, the reports from national media that unfortunately many have misrepresented facts.

    The issue, and the reason for the protest, has NEVER been about the money. We knew the platform Walker ran on, and we fully expected cuts when he won the election in November. What was not expected was the end of collective bargaining. That reason, and that reason alone, is why the protests are happening.

    Now the national debate that has taken place has focused on our pay and what public sector workers earn. What they have very rarely, if at all, have mentioned are:

    1. We have not received any raises in 4 years.

    2. We have also had mandatory furloughs in the last 2 years, resulting in a 3% decrease in pay.

    Now I have a long career that has been both in public and private sectors. While I enjoy a comfortable salary in the public sector thanks to my years of experience and education, it’s not nearly as much as I made working in the private sector.

    So why did I take the cut? I don’t have to worry about coming to work tomorrow after years of loyalty and be told it’s my last day. As someone else pointed out, our take-home salaries are considerably less, and many surveys will back that up. But our contributions to retirement and medical insurance make up for the lack of pay.

    Some will argue that the benefits and stability lead to an apathy on the job. And yes, there are the examples of some workers who contain that. But I can point to several examples in most every private company I’ve worked for or know of people who show up, do the bare minimum, and go home. Eventually those people are weeded out, and for the most part, people in the public sector with that attitude generally are too.

    Futhermore, isn’t the idea in private business that if you want the best and the brightest, you pay them well? Don’t we want our Government programs run effectively and efficiently? Seems to work in the private sector, so why can’t this apply to public sector as well?

  50. 50 50 Steve Landsburg

    Jim from Wisconsin:

    (Responding only to your last point; realizing it’s not your only point):

    Futhermore, isn’t the idea in private business that if you want the best and the brightest, you pay them well? Don’t we want our Government programs run effectively and efficiently? Seems to work in the private sector, so why can’t this apply to public sector as well?

    The problem with this is that every “best and brightest” who is hired by the public sector is unavailable to the private sector, so it’s not at all clear that we WANT the best and brightest in the public sector. To take an extreme case, I don’t want the best Silicon Valley engineers tempted to work as high school teachers; I’d rather have them pushing the limits of technology. From the point of view of economic efficiency, this is the one and only reason why public sector employees ought NOT be overpaid. (It’s also a reason why private sector employees ought not be overpaid, but there’s generally less threat of that happening because of the private-sector profit motive.) It’s the one and only reason not to overpay public employees — but it is a good and sufficient reason.

  51. 51 51 Jim from Wisconsin


    I hardly expect the best engineers from Silicon Valley to become a high school teacher.

    But, what would be wrong with this person working, say, for a State University? They’d have a budget to continue research and develop technology, and more importantly using his or her skills to teach the next generation of technical engineers?

  52. 52 52 Steve Landsburg

    Jim from Wisconsin: Surely you’re aware that time spent teaching is time not spent inventing. Some people’s time is best spent teaching; other people’s time is best spent inventing; other people’s time is best spent in some combination of the two. The trick is to set prices that will draw the right people into the right activities. Private markets do a pretty good job of this (it takes substantial argument to establish this, but that argument can be found in standard economics textbooks); to continue getting it right, it’s important for public employers to pay roughly market rates — not less, and not more.

  53. 53 53 Ken B

    @Harold: What nonsense. The law already requires schooling. How does saying the kid can go to school A or B rather than the kid must go to school A affect that?

    Plus you smuggle in all sorts of ridiculous premises here. We already have some form of school choice: home schooling. If in rare cases parents neglect it the state can and does step in. Why must that change?

    The assertion that “school choice assumes a complete alignment of interests” between parent and children is just false, and obviously so. Also false is the claim “monopoly public schools assumes a complete alignment of interests between children and the school board members”.

    Parents choose lots of stuff for their kids: bedtimes, TV shows, pants, food, city, state, toothpaste. If they neglect or screw up we can intervene.

    A more serious problem is what to do with crappy schools. Google Detroit Public Schools a bit and see. Parents will do a better job than the the DPS has done.

  54. 54 54 Will A


    “Why are the “poor kids” leaving their school? Threats of violence, student under performance, teacher under performance, poor infrastructure, inadequate teaching supplies/materials?”

    For sure turnover of teachers in poor schools is very high for the reasons you mention. If you wanted the same level of teacher in poor and rich schools (in our current system) and if you buy Prof. Landburg’s argument about turnover and compensation, then you would need to either:
    - Increase the compensation for teachers in poor schools.
    - Decrease the compensation for teachers in rich schools.

    As it relates to infrastructure and resources, if we wanted poor and rich schools to have the same level of funding, every time a parent in a wealthy school district donated money to a school, the state would either:
    - give the same amount to poor districts.
    - reduce the amount of state funds to the school that got the donation.

    It is possible however, that the majority of Americans that vote don’t want the same level of teacher or funding for poor schools.

  55. 55 55 Dusty

    1) Why not use the standard, more conventional way of telling if the employees are overpaid. Mathematically? Simply compare the wages, and you’ll see that overall compensation.

    2) Not that it matters, it isn’t about compensation.

    Nah, you’d rather demonize people you disagree with rather than realize that you might have something in common with them.

  56. 56 56 Sarah Hawkins

    Your argument is false on many levels. First, in Wisconsin public sector employees are paid on average LESS than there private sector counterparts-even when factoring in benefits like pensions and vacations days. If Scott Walker were to slash say minimum wages, you are correct, there would not be the mass protests we are seeing now. Poor people are used to shitty treatment by the ruling class and disenfranchised to the point that they wouldn’t demonstrate like you are seeing the middle class do now. The difference is we who are protesting and getting the rug pulled out from under us is that we don’t feel powerless like the lower classes. For years, we were told that the state could not afford to give us raises and thus we settled for increased pension benefits or the like. Now, those promises and CONTRACTS aren’t being honored. You need to be more careful about the conclusions you draw.

  57. 57 57 Steve Landsburg


    1) Why not use the standard, more conventional way of telling if the employees are overpaid. Mathematically? Simply compare the wages, and you’ll see that overall compensation.

    The problem with this procedure is that there are too many unobservables. You want to compare state workers with comparable private sector workers. How do you know who’s comparable? You can count years of education, but that won’t tell you who actually acquired more skills while they were getting educated. You need to adjust for the quality of working conditions. How can you observe all the relevant working conditions?

    Given that it’s essentially impossible to make direct comparisons, it seems to me that the two best measures we’ve got are a) quit rates and b) loudness of protest when wages are cut.

    PS. Where did you get the bit about “demonization”? I said certain people were overpaid. You don’t have to be a demon to be overpaid.

  58. 58 58 Steve Landsburg

    Sarah Hawkins:

    First, in Wisconsin public sector employees are paid on average LESS than there private sector counterparts-even when factoring in benefits like pensions and vacations days.

    You cannot know this without knowing who those private sector counterparts are, and that’s essentially impossible to determine directly (see my response to Dusty just above).

  59. 59 59 LucB

    I essentially agree with the main point, I just want to point out a qualification:

    One effect of cuts, which I am not sure at all applies here (I confess having no idea how civil servants are paid in Winsconsin), may apply when the recruitment process is imperfect and/or the career earnings curve over time is steep.

    If for some reason (I don’t claim it’s efficient for most workers) you choose a steep (over time) pay schedule, you are essentially asking your workers to forgo payments now in return for better pay later. Slash the second part when the worker is in the middle of his “firm life cycle” and you’ve damaged his expected earnings even if he is not overpaid before cuts (overpaid as defined on average over the life cycle),
    hence the “if underpaid, you can just leave” argument must be qualified.

    Another version of the argument is competitive entry with a costly specific investment (studying very hard to pass a test, and so on).

  60. 60 60 Ken B

    @Steve: I think you missed a point in your first response to Jim. Another reason to avoid oeverpaying public workers to get “the best and the brightest” is that you won’t get them anyway. A private employer with his own equity on the line has an incentive to get the best, but public functionaries do not. They have their own incentives. These include but are not limited to rent seeking rules imposed by lobby groups, and pressure from the then current public employees.

  61. 61 61 TM

    The defined benefit retirement plans provide a *big* incentive not to switch jobs.

  62. 62 62 Toby

    I didn’t read all the comments, but. . .

    There seems to be a lot of arguing here that teaching is a job without private sector alternatives. But, every time there’s talk of starting a voucher system and allowing people to start private schools, do you see the teachers rushing to embrace the idea?

    It’s the teachers’ unions that throw a fit: Growing up the son of a math teacher (and union president), I’d say they know that things aren’t as cushy outside of state employment, or they’d be all for it.

  63. 63 63 Tom

    I think it comes down to a simple fact of money paid or money taxed.

    Generally, the States and Federal government are the some oflargest single employers in the US, just shy of 10%.

    Revenues were up in the late 90s, Public Sector unions took advantage of these and negotiated fantastic benefit packages, deferred pay as has been mentioned. 100% of that money comes from the tax payers.

    As the economy collapsed, that remaining 90% + had to continue to pay into benefits that had been negotiated during near historic highs. Now the reaper is here to collect and in most states, benefits and salaries must be paid first. So the 90%+ who are earning less, and a larger chunk of that 90% is now unemployed, with wages diminishing. These Public Sector jobs are nearly a protected class, hard to terminate. Unlike my Private Sector job that can be outsourced, even the same equivalent Public Sector position cannot be outsourced.

    So the State governments hands are tied. The Public Sector unions are not subject to bankruptcy, hard to fire, cannot be outsourced, so they have little to no means, to bring sinking ships into balance, without further adding to the tax burden on the remaining 90%+.

    My father flew for United for 30+ years. You know what the did? Canceled his pension, terminated it. He gets 10 cents on the dollar of what he was promised. Do I hate the company? Sure tough licks but 10cents on the dollars, is better than no dollar at all. It’s reality, something the remaining 90%+ of us out here are dealing with.

    If you think of government as company, comparable to the Private sector and make these comparisons, then you have to make the consideration that if you don’t like how a company is treating its employees or the quality of the product, I can boycott that product. I can do that the Public Sector. If I am not happy the way roads are maintained or the quality of our schools or the level of pay that particular positions are paid, I cannot change it. However in the Private Sector, if a company creates an inferior product or overpays the salaries of top bank officials, I can do everything in my power with my dollar, to change that course.

    I don’t know if I think government workers are overpaid, I think they might be and I don’t think they’ve taken much pride in a quality product in a very long time but I can’t change that. Don’t you dare wag a finger at a fireman, police officer, teacher, soldier, agent, TSA official, cubicle dweller, road paver, project planner, city manager or other official who wasn’t elected. You have no idea what they go through, you have no idea how little appreciation they receive, the lack of thanks, lack of Christmas parties and gifts from clients. Oh but wait, I have to pay for your service and we can’t change your benefits and you want more.

    Government will bleed the rest of dry and cry for more. The more I see protests in WI, I think of rearranging deck furniture on the Titanic.

  64. 64 64 Larry B.

    I was sent the link to this post by a friend. I must confess to some confusion over the comments. I’m not sure I buy the argument that the volume of protests (or lack thereof) is somehow an indicator of whether or not people are over/underpaid. However, that is an irrelevant argument. The real point is the debate over the whether or not public sector employees should have the “right” to collective bargaining over pensions/benefits.

    As I understand it, Gov. Walker ran on the proposition that spending was out of control and the steps he would take to put the state back on a solid fiscal foundation. He was elected by a solid majority, and the state Asssembly and Senate both received large Republican majorities for the first time in many years. We can safely assume that this indicates a desire on the part of the electorate for state spending to be reduced to reasonable rates.

    The basic fact is, politicians have been over-promising public sector unions for years in return for their support at election time. The governor is merely doing what he said he would do: namely, rework the bargaining arrangements between the state and the public sector unions. Reasonable enough. The state is facing an untenable fiscal situation; $3.6B in deficit for the current bi-annum. The taxpayers can not afford to continue to subsidize the public sector. Making short term, temporary agreements to reduce costs will do nothing to address the long term structural problems created by collective bargaining. Basic changes to the formula must be instituted to address these issues. Otherwise, the problem will have been simply postponed, not corrected.

    IMO, public sector employees should not be given collective bargaining rights, for the basic fact that there isn’t anyone representing the interests of the private sector taxpayers when these “bargains” are instituted. Anytime you have politicians sitting across the table from the people responsible for their having been elected, and paying for the labor agreements with OPM, you have an open door to corruption.

    If you believe that the public sector should retain collective bargaining rights, let’s start by removing the politicians from the equation completely and institute a panel of private sector (non-union) workers, evenly split between Reps and Dems, responsible for negotiating on behalf of the taxpayer. A fair bargain can only be struck between workers and and those actually paying for their services.

  65. 65 65 bill

    “loudness of protest when wages are cut.”

    Really? You’re a economics professor and “loudness of protest when wages are cut” is really the best way you can come up with judging if someone is overpaid for their job?

    As a professor of economics, could you explain to me what research backs up this claim?

  66. 66 66 ben

    Oh my god. Really? All this sound and fury and none of you chuckleheads could be bothered to actually look at the numbers?


    There you are, friend.

    If you can’t be bothered to read 13 pages, I can summarize for you. Total compensation (both wages and benefits) for WI’s public employees is dramatically lower than their private sector counterparts.

  67. 67 67 Harold

    @KenB. Sorry to get off topic here, I know this post is not about school choice. Economic theory predicts maximum gains by individuals maximising their own benefits through exercising free, informed choice. You have said that where markets have been tried they have been shown to work. School choice is not such a market, for the reasons I have indicated – the choice is not free or informed for the recipient of the education. An efficient education market will maximise the benefits for the parents – surely this is what economic theory would predict, or have I missed something? I am not promoting school choice (obviously), but neither saying it will not work. I am saying that assuming markets will provide the optimum amount and distribution of education may be wrong, even assuming a perfectly operating market. I was unclear in my earlier post when I said school choice required perfect alignment of interests of parent and child. What I meant was that to apply the theories of markets directly and predict optimum results for the child required such an alignment.

    The introduction of choice *could* result in some children becoming significantly disadvantaged compared to others, and a non-optimum distribution of education.

  68. 68 68 Steve Landsburg


    Oh my god. Really? All this sound and fury and none of you chuckleheads could be bothered to actually look at the numbers?

    I assume you couldn’t be bothered to read the blog post explaining why these numbers tell us nothing: http://www.thebigquestions.com/2010/08/09/i-cant-resist/

  69. 69 69 Corey

    Your hypothesis is specious at best. You’re attempting to look at one factor and posit an outcome using that and only that. It looks to me like you are working backwards – formulating your answer then amassing evidence to back that up instead of vice versa.

    If you really wanted to prove your point, try drilling down to specific jobs and then doing your comparison, not using blanket percentages across the board. For example, compare the quit rates among accountants, teachers, or social workers in the public and private sector. If this data consistently shows your original theory to be correct I would be more inclined to believe you.

    Another factor I think you need to consider is that there are many jobs that are almost entirely specific to the public sector and I believe contain the one immeasurable factor that I think you grossly overlook – people’s passion to serve the public. This can make someone more inclined to keep a job longer and not consider quitting as the feeling of fulfillment they get from helping others can make up for the lower pay rate they receive. Unless you have formulated some equation that factors this in, I don’t believe you can go by the numbers alone.

  70. 70 70 Bob W

    I find it interesting that Wisconsin teachers are paid (per ca pita) with the very highest paid teachers in the country yet our education levels are low. While the states with salaries at about 2/3rds of ours are producing a much better educated group of students all all levels. Could this be because more and more people are becoming teachers because of the pay, benefits and summers off? As apposed to having a passion to be a teacher. The teachers I know personally fall into the first group and tell me their three reason for becoming a teacher are June, July and August.

  71. 71 71 vic

    The notion of an ‘efficiency wage’ is that if the employer pays a premium over transfer earnings (i.e. wages in the next best job) then the worker has a strong incentive to be obedient and adaptable and work his ass off in good faith. However an efficiency wage turns into an inefficiency wage if there are strong property rights in jobs or if workers can collectively determine their own productivity.
    In the first case, the workers inefficiency- i.e. the gap between his present productivity and its potential might go to wage minus the relevant cost of termination. In the second case, workers drive down productivity to extract the employer’s surplus.
    If the Public Sector was subject to the same constraints as a small private firm, this would not be an issue. It may be that in America, the Public Sectors of different States are competing with each other and hence the present crisis.
    However, the problem of ‘inefficiency wages’ remains. The usual argument that the Public Sector tends to be a monopsonistic employer and hence collective bargaining is a necessary and allocatively efficient countervailing force needs to take on board this business of inefficiency wages.
    There is an old adage ‘hard cases make bad law’. No doubt, in the present situation there are ‘hard cases’ which might militate against accepting the argument Prof. Landsburg makes, but that should not cause us to support a bad law- in this case the right of collective bargaining where the employer is not really monopsonistic and where moreover ‘inefficiency wages’ are more likely than the reverse.

  72. 72 72 DK

    So, what are quit rates of tenured professors? I guarantee that the loudest screams EVER would come from academics if tenure is abolished. U Wisconsin academics already scream bloody murder from their only 3% salary reduction in the form of mandatory furlough.

  73. 73 73 Michael

    Ben, I’m not certain you will be reading at this late date, but despite its claims, the Economic Policy Institute is about as non-partisan as the Heritage Foundation, only from a liberal perspective.

    And before you disagree, please take a look at their board of directors, and pay particular attention to their mini-biographies: http://www.epi.org/pages/board/

  74. 74 74 Michael

    Ben, on reading the report, I see one area they deliberately obscured (I say deliberately, because when they were “controlling for education, experience, organizational size, gender, race, ethnicity, citizenship, and disability,” they ignored one of the most obvious factor). In case it isn’t glaringly obvious, when they compared degree to degree, they compared Masters to Masters — it didn’t matter if one person got a Masters in Medieval Literature and another a Master’s in Petroleum Engineering, nor did it matter if one person went to the University of Phoenix and another to MIT. In other words, when they claim to be “controlling for education,” they left out two of the most important factors when determining compensation.

    What makes it even more interesting is they somewhat allude to it in the line, “The compensation differential is greatest for professional employees, lawyers, and doctors.” A person who received a law degree from Harvard may have a higher wage and benefit package than one from the University of Wisconsin, and it certainly seems reasonable that the former might go into public practice rather than become a public defender. That’s not to say he’s worth more as a person, only that he can command a higher salary.

    Until they also control for major field of study and university attended in their “controlling for education” section, the study isn’t worth that much.

  1. 1 Wisconsin’s Smoking Gun » EppsNet: Notes from the Golden Orange
  2. 2 Wisconsin Followup at Steven Landsburg | The Big Questions: Tackling the Problems of Philosophy with Ideas from Mathematics, Economics, and Physics
  3. 3 Wisconsin Followup « Daniel Smith
  4. 4 Wisconsin’s Smoking Gun « Daniel Smith
Comments are currently closed.