‘What a Piece of Junk’: Responding to the Economic Policy Institute on Public Employee Pay

In the original”Star Wars” trilogy (not those crappy prequels), Luke Skywalker uttered these words when he first saw Han Solo’s ship, the Millennium Falcon: “What a piece of junk.” I had the same reaction to a new policy brief from the Economic Policy Institute (EPI), which attacks my work with Jason Richwine on public sector pay, in particular a recent study for the Ohio Business Roundtable. (I should note that these comments are my own and that Jason is far more polite than I am).

In previous work for EPI, Rutgers University’s Jeffrey Keefe concluded that Ohio state and local government workers receive slightly lower total pay and benefits than similar workers in the private sector. We find, by contrast, that the combined value of public sector salaries, benefits, and job security exceeds private sector levels by roughly 43 percent.

Keefe questions practically every part of our Ohio paper and, trust me, we have answers to almost all of his points. If you read EPI’s paper and are tempted to believe its claims, post a comment and I’ll address it.

But pensions are where the real action is. Once you accept our view of pensions, you could buy pretty much everything else EPI says and still conclude that public employees are overpaid. EPI thinks pay studies should focus on what employers contribute toward pensions while we look at the benefits employees actually receive.

Put simply, EPI believes that government pensions can generate a given dollar of future retirement benefits at roughly one-third the cost of a private sector employer. And it’s true that, for each dollar of guaranteed future benefits, governments actually do contribute about one-third as much as private pensions. These lower contributions are based on aggressive accounting rules that let public plans “discount” their future benefit liabilities using high interest rates of around 8 percent, versus about 5.5 percent for private defined benefit (DB) plans and, implicitly, around 4 percent for 401(k)-type pensions. Based on low employer contributions, EPI concludes that public sector pensions aren’t actually all that generous. We counter that if you look at the benefits employees actually receive, most public employees’ total compensation package is well above private sector levels.

In effect, EPI’s argument rests on you believing that “Public employees are overpaid, but we aren’t overpaying them.” That is, government possesses some magic by which it can pay far higher pensions at far lower costs than the private sector. This claim is doubly wrong.

1)    The vast, vast majority of professional economists don’t believe that government possesses such magic. As Keefe himself admits, most economists argue that accounting rules that let public plans contribute so much less than private pensions are simply wrong. Don’t take my word for it, though: Nobel Prize-winning economists, the Federal Reserve, and the Congressional Budget Office all say the same thing. If public pensions followed economically-sound accounting rules, their contribution rates would rise and it would be obvious that public employees receive higher total compensation.

2)    Moreover, even if government can magically generate pension benefits at one third the cost, that does not imply that employees should be the beneficiaries of that little miracle. According to the theory of “equalizing differences”—which the Handbook of Labor Economics callsthe fundamental (long-run) market equilibrium construct in labor economics”—government just as well could pay lower wages and use the savings to reduce taxes or increase other government programs.

For a public/private pay comparison, what we want to know is whether that offset has taken place. By measuring the pensions people will actually receive, along with their salaries and other benefits, we can accurately compare total compensation packages between the public and private sectors. And, based on these actual benefits, it is unequivocal that public employees in Ohio, and in most other states for that matter, receive higher total pay than private sector workers.

Other public sector pay studies—such as from the (hardly conservative-leaning) Center for State and Local Government Excellence, which EPI cited in its highly-misleading blog post on our paper—note that, “the public sector contribution under-states public sector compensation,” for exactly the reasons we describe. (We have other issues with the CSLGE study, but on this point they’re correct).

So to accept EPI’s arguments regarding pensions and overall public sector compensation, you have to reject the views of both the vast majority of financial economists and the vast majority of labor economists. Nice going, EPI.

10 thoughts on “‘What a Piece of Junk’: Responding to the Economic Policy Institute on Public Employee Pay

  1. Andy….You are obviously not an actuary or a financial person, because your pension argument could not be more inacccurate. Everyone knows that public and private plans use different actuarial conventions, but that’s where it ends. The differences are a result of differing business models, and have nothing to do with chicanery as you suggest. The private sector recognizes gains immediately while the public sector smooths returns. The private sector also uses slightly lower return assumptions among other differences. The end result however is simply this: Private sector plans have more volitile contribution rates while public sector plans have less volitile contribution rates. In the long run both trend toward equilibrium, which is full funding. During the late 80s and throughout the 90s, most public plans were over funded. If it is your contention that the public sector is understating salaries because of pension under funding, then it must surly be your contention that the public sector was overstating its salaries during the 90s because it was putting more toward pensions than was neccessary to fund benefits? Any actuary or financial person will easly see the holes in your argument, but some unsespecting folks will surly be duped by your rhetoric, so congrats on perpetrating myth!

  2. Robert — I don’t accuse public pensions of not following their actuarial or accounting rules; I argue, like the many others I cite, that those rules are incorrect. There’s now a large literature on this and I haven’t seen any response from public pension actuaries that I find particularly convincing. It’s not enough to say that they have different business models; you have to say how and why that matters. And, in my second point, I showed that even if you accept that public pension accounting rules are correct and that they can produce higher benefits at lower costs than the private sector, you don’t and shouldn’t accept that the benefits of government’s ability to do so should flow to public employees rather than taxpayers, beneficiaries of government programs that compete with employees for funding, etc.

    As for your example from the 80s and 90s, public sector plans were not underfunded if measured on a market value basis. Pensions were still guaranteeing participants an 8% average return when the return that plans could guarantee themselves on their investments was much lower. In effect, they’re giving participants an unpriced, but very valuable, put option.

  3. I’ve been following this discussion for some time, and while understanding (and supporting) the value and benefit of a WIDE range of opinions, on subjects such as this where fundamental mathematics (and quantifiable economics) plays a big role …. at what point should an academic’s opinions (as well as perhaps his position in academia) be cast aside BY THE INSTITUTION THAT EMPLOYS HIM when his conclusions are so utterly illogical and unsupportable ?

    Rutgers … why is Jeffrey Keefe still on your faculty ?

  4. Robert, basically what you are saying is that public pensions invest in a completely a different set of assets than private pensions plans. Hogwash, the only asset that is different is the taxpayer guarantee and ability to put losses on the taxpayer. That is what makes them so immoral.

    • Oliver — You’ve basically got it right. The government offers a guaranteed benefit funded with a (relatively small) investment in risky assets. If the assets fall short then the taxpayer has to honor the guarantee, as we’ve seen, through increased taxes or reduced spending on other government programs. So it’s clear that it’s the taxpayer, not “the government” that is bearing the risk, which is why economists almost universally believe that we should price these risks as taxpayers themselves do. In effect, the taxpayer is forced to offer an implicit “put option” for public pensions, but the cost of this guarantee — which is really easy to calculate — isn’t revealed anywhere.

      Put another way, a pension that assumes an 8% rate of return also offers a GUARANTEED 8% return on contributions to participants — and that’s on both their personal contribution and the employer contribution made on their behalf. The result is actual dollar benefits in retirement that are a LOT higher than a typical private sector employee with a 401(k) would receive. Our studies capture the actual benefits public employees will receive, while Keefe’s simply don’t. It’s really as simple as that.

  5. Andrew, I understand what you understand … via decades of (very technical) work in this field.

    A suggestion …. Concentrate more on Public Vs Private sector BENEFIT comparisons rather than the funding cost, the latter being too fungible and difficult for the masses to understand (and filter out the myriad of distortions and outright lies from supporters of the status quo).

    Express this first as the typical monthly annuity …. easiest for ALL parties to understand.

    Then bring in the additional Pension elements that make Public Sector Plans MORE expensive than just a simple ratio of Public Sector monthly income to Private Sector monthly income …. e.g., the earlier retirement ages, the inclusion of COLAs, the MUCH more liberal definition of “pensionable compensation” in Public Sector Plans, etc.

    Next explain that the BEST (and fairest) Public Sector vs Private Sector pension “cost” comparison is what an immediate annuity issuer would charge (as of the date of retirement) to assume responsibility for the promised pension….. noting that this “cost” factors in the particulars mentioned in my last paragraph.

    And lets not forget the VERY high OPEB cost of subsidized retiree healthcare, often with a “value” (for family coverage) approaching $500K for Public Sector employees who retire in the early 50s.

    • Tough Love — Good points, and you hit on how hard it is to catch all of this. The pension normal cost, adjusted to a reasonable discount rate, is a decent approximation of the generosity of the plan. But you’re right that there are any number of variables to monkey with, so there’s no real way to get it perfect. A head-to-head comparison of benefits for a stylized worker does show that public sector pensions are simply far more generous than what the typical private sector employee will receive. And, of course, OPEBS only sweeten the deal.


      • Andrew, an interesting (and enlightening) BENEFIT comparison would go something like this:

        (a) assume a full career Private Sector worker (lucky enough to still have a generous Traditional DB Plan) retires with 50% of final average salary after 30 years of employment at age 60 …. noting that this would be quite a GENEROUS Private Sector Plan…. and receiving an “actuarially unreduced” pension stating at age 60 is both gererous and somewhat unusual in Private Sector Plans.
        (b) Assume that his/her final average salary was $100K so that the annual single-life annuity is $50K, and determine what the “cost” to purchase that benefit from an annuity-wrtier is. My guess … about 13.5 times the annual fixed (non-CLOA adjusted )annuity or $13.5x$50,000 = $675,000.
        (c) now lets calculate the “typical” pension of a Public Servant after a 30 year career… likely retiring at age 55 (not 60) also with a final average salary of $100K.
        (d) The annual (COLA-adjusted) annuity is likely about 75% of pay after 30 years or $75K annually. Hence, had it started at age 60, and using the same 13.5 times factor, the cost would be 13.5x$75,000= $1,012,500.
        (e) But wait, since he/she begins collecting an unrediuced pension at age 55 (not 60, as does the Private Sector worker) the cost is not 13.5 time the annual annuity , but about 15 times. Re-calculating, we now have a “cost” of 15x$75,000=$1,125,000.
        (f) But we’re not done. The inclusion of a 3% annual COLA adjustment (to an other wise identical Plan, but w/o the COLA benefit) increases a pension’s cost (to someone retiring at age 55) by just about 1/3. YES …. the inclusion of COLA is a VERY expensiove benefit. So, we now have a “cost” of $1,125,000 (from (e) above) times 4/3 = $1.5 Million.
        (g) So where are we …. $675,000 pension “cost” for the Private sector worker and $1.5 Million for the Public Sector worker … making the SAME pay at retirement.
        (h) Now add about another $250K-$500K (call it $375K) for heaviiy subsidize retiree healthcare that the Public Sector workly gets (but NOT the Private Sector worker) and we have a Public Sector “Retirement package” worth $1,500,000+$375,000= $1,875,000 vs $675,000 for the Private Sector worker. Note that the Public Sector worker’s retirement package has a value and hence a “cost” 2.78 times greater than an equally compensated Private Sector worker WITH a generous (for Private Sector standards) retirement package.

        Taxpayers should be outraged by this. After all, THEIR contributions (including the investment earnings thereon) pay for 80-90% or the total cost of Public Sector pensions.

  6. Robert: You are wrong on many fronts. Are actuaries God? The state pensions were overfunded? Then why did most states raise taxes and sell Pension Obligation Bonds?
    Your biggest problem with 8% is the historical drawdown. From 1966 – 1980 the stock market
    lost 4 %. So, how will we pay 8% plus a 2% COLA, plus benefits ( NOT FUNDED) for 14 years
    if that repeats which is still within the parameters of the Bell Curve.

    2nd Question: Why didn’t these “financial people” (idiots) forward contract 30 year govt. bonds when rates were 12 – 16% in the early 80′s? or even a few years later at 10%? What Brains?

    3rd Question: Using your Bell Curve Crude Oil exceeded 2.00 from the mean basis Cash Monthly and Cash Weekly in 1998 and 2008. Why didn’t states forward their gas costs? This was a loss of 1.5 or more per gallon from 1998 and 1.0 or more since December of 2008. This was at least a loss of 500 million from 1998 to 100 million plus since 2008. What happened with your theory? Answer the question: LOSER.

  7. Great beat ! I wish to apprentice whilst you amend your website, how could i subscribe for a blog web site? The account helped me a acceptable deal. I had been tiny bit familiar of this your broadcast offered brilliant clear idea

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=""> <strike> <strong>