CEPR - Center for Economic and Policy Research


En Español

Em Português

Other Languages

Home Publications Blogs Beat the Press More Public Pension Scare Stories at the Post

More Public Pension Scare Stories at the Post

Thursday, 05 April 2012 14:08

The Washington Post is always willing to accommodate those who want to make a big issue out of budget deficits. In that spirit it ran a column today by Robert Pozen and Theresa Hamacher warning readers about "public-pension pitfalls."

The piece begins by decrying the fact that almost 80 percent of state and local government employees are still covered by traditional defined benefit pensions even as these pensions are rapidly disappearing from the private sector. This may seem a bizarre complaint to most people.

After all, few workers have been able to accumulate enough in 401(k)s to guarantee themselves any sort of security in retirement. In 2009, the financial wealth for the median household between the ages of 55-64 was only around $50,000, including all 401(k) assets.

Most public sector workers will have some pension income to support them in addition to just being dependent on Social Security. This might be considered a source of security that we would like to see brought back for private sector workers rather than eliminated for public sector workers. Of course this is the Washington Post.

It is also important to remember that close to a third of state and local employees are not covered by Social Security so their public pension will be their only regular source of retirement income. Somehow, Pozen and Hamacher forgot to mention this fact in their piece.

Next we are told that the unfunded liabilities of these plans are $600 billion. This is supposed to sound very scary, since $600 billion is a big number. To make sense of big numbers we need a context.

The planning period for a pension fund is typically 30 years. Over the next 30 years, GDP is projected to be over $400 trillion in today's dollars. This means that the unfunded liability is equal to about 0.15 percent of projected GDP over this period. To make another comparison, relative to the size of the economy it is equal to a bit more than 3 percent of what we are currently spending on the military. Are you scared yet?

Next Pozen and Hamacher complain about the rates of return being assumed by these pensions in assessing their liabilities. They tell readers that the pension funds mostly assume 8 percent nominal rates of return but:

"Over the past 10 years, the Standard & Poor’s 500-stock index has achieved only a 1.9 percent annualized return."

This one should be scary but only because Pozen is identified as a senior lecturer at Harvard Business School and he may be teaching this stuff to young people. What Pozen apparently missed is that we had a huge stock bubble in the late 90s that burst beginning in 2000. It was precisely because stocks were hugely over-valued in the late 90s (price to earnings ratios topped 30 to 1) that we had very low returns.

Some of us noticed this at the time and tried to warn that pension funds were being overly optimistic in assuming 8 percent returns when price to earnings ratios were so high. Unfortunately, Mr. Pozen was not among those issuing such warnings. Instead he sat on President Bush's Social Security commision, which advocated privatizing Social Security based on the high returns it assumed to be available in the stock market.

However the flip side of this picture is that now that price to earnings ratios have returned to more normal levels, it is safe to assume that stocks will provide historic rates of return. This means that pension funds are being entirely reasonable in assuming 8 percent returns going forward precisely because the return on Standard & Poor's 500-stock index was so bad over the last decade. In fact, it is almost impossible to construct a scenario in which pension fund returns will be substantially worse than what the pension funds are now assuming.

Pozen should know this, just as he should have been able to recognize that the stock market was hugely over-valued when he wanted workers to put their Social Security money in the market. There is a consistency in Pozen's seemingly contradictory positions. It is the interests that are being served by his arguments. 



I'll respond quickly to a couple of comments below. First, the 8 percent nominal return for stocks is not based on historic performance. It is a calculation that assumes that firms pay out 60 percent of profits to shareholders (either as dividends or share buybacks), that the price to earnings ratio stays constant and that profits grow at the same rate as the economy. (We adjust for being somewhat above trend now.) If readers check the link to the calculator they will see that you need to make some unusual assumptions to get something much lower than an 8 percent nominal return.

One comment took issue with my comparison to 30 years of GDP. My calculation used a 3 percent real discount rate, it was not simply summing GDP over a 30 year period.

Comments (14)Add Comment
Is an 8% stock-market return reasonable?
written by matthew, April 05, 2012 3:38
I agree with the gist of Dean's argument, but I have a question about his statement toward the end that it is "entirely reasonable" to assume an 8% return on stocks going forward. Why is this so reasonable? If I'm reading him right, Dean considers any return less than 8% "almost impossible."

But isn't this 8% figure based on the past performance of the U.S. stock market? And the past performance of the U.S. stock market took place within the context of very usual and very favorable circumstances. These circumstances -- particularly World War II and its aftermath -- saw the emergence of the U.S. as the preeminent global superpower. These conditions are not likely to obtain in the future. So why should we believe so confidently that the U.S. stock market's future performance will be so stellar?

On a side note, does this 8% return take into account both inflation and fees?

Any clarifications are appreciated.
option out of social security
written by tew, April 05, 2012 4:28
The reason "a third of state and local employees are not covered by Social Security" is because they don't want to contribute to social security. This allows them to evade the embedded income distribution feature of social security. Anyone familiar with the basics of how social security benefits are calculated knows that the curve of benefits vs. lifetime contributions is strongly concave - some of the lifetime contributions of high earners are shifted to benefits for the lower earners. (This is not a disputable fact and it easy information to vet.)

In exchange for opting out of social security most of these workers pay less of their salary towards their pensions than they would contribute to social security. In return for this lower contribution, they receive pensions that are usually substantially larger than social security benefits and which start at an earlier age.
misleading aggregation
written by tew, April 05, 2012 4:33
The use of a 30 year undiscounted GDP aggregate against a current pension deficit is misleading. (My guess is that the author is deliberately misleading the readers.)

If the current pension deficit were fixed and would be resolved in equal proportion to the next 30 years' GDP, then this comparison would be less misleading. However, we should note that government employment tends to grow with GDP. Thus, those larger future GDPs will be associated with larger public sector employee bases which increase the demands on the pension funds. So if the root causes of the pension deficit are not solved, future deficits will grow along with employment.
written by skeptonomist, April 05, 2012 5:40
For some scaling, Social Security outlays for the one year 2011, which are unfunded (paid out of current revenue), were almost $750B. I would guess that for state and local retirement programs to be exempted from SS, they would have to be similar, that is mostly not funded - is that true? If not the state and local retirees could not have been getting benefits equivalent to SS in the early days. Or has there been some kind of gradual conversion from unfunded to funded?
Facts about SS and some state and local employees pension plans
written by Pauleytel, April 05, 2012 8:41
@ tew - the benefits paid by some state and local public pension programs replace SS benefits partially or in their entirety for their member retirees - e.g. retired teachers under California's CALSTERS forfeit ALL claim to prior SS credit earned through other jobs; CALPERS's state and local public employee members lose about half of non-CALPERS SS credits. Public employees' defined benefit contributions vary by agency and/or agreements relative to their employer's, which also do not pay into the SS trust. However, their benefits upon retirement, are taxed as ordinary income while the SS trust is largely or entirely relieved of supporting them. The general economy benefits from these retirees spending at likely greater levels than pensionless retirees or those who have saved the meager rate Dean notes above: "financial wealth for the median household between the ages of 55-64 was only around $50,000, including all 401(k) assets." Incidentally, such SS-exempt people receive none of the current payroll tax cut and so are not party to that tax relief.
Forced to gamble on the market
written by Innocent Victim, April 05, 2012 9:08
As a senior with lifetime savings that earn me no interest, I have no other way of deriving income from them but to gamble on the stock market. Barack Obama, Tim Geithner and Ben Bernanke keep interest rates artificially low in order to assist the fraudulent banks that are responsible for our current difficulties. This policy does not put money into circulation, because the banks prefer putting their cash into questionable financial assets rather than lend it. They get my deposits for nothing, and invest them.
more equal than others
written by tew, April 05, 2012 10:22

Thank you for clarifying what I was saying w.r.t. public sector employee pensions and social security participation. However, not all who collect public pensions forfeit their entire claim to social security benefits if they'd worked enough to reach the threshold for qualifying.

I have to say that this one was a howler though: "The general economy benefits from these retirees spending at likely greater levels than pensionless retirees or those who have saved the meager rate" Oh man, that's good stuff. That's the same justification the more crass defenders of wealth use! "Hey, it's goooooood that they have so much money, since when they spend it the economy benefits!"

Anyway, I've followed some of Dean's bad pension math in the past. Remember the piece he did showing that because the market was trading at a forward P/E below 15x that very high future returns were certain? He "proved" it by including past bull market cycles that began at single digit multiples and just said that a market P/E of 14x was in the same bucket as one with 8x - same expected future returns! Oh, such great entertainment if it weren't swallowed whole by so many people. (And, yes, I'm saying that the VSPs / WSJ readers aren't the only ones telling themselves fairy tales.)
excellent job!
written by vlasti, April 06, 2012 3:49
very good job!
SS and pensions
written by Jennifer, April 06, 2012 9:28
If public workers get SS it is because they PAID into the system, just like every else, and they PAY into their pension as well. If a worker is dependent on their pension they have the privilege of paying for health insurance as well --they do not get Medicare.
Isn't DOD on a Defined Benefit Plan
written by Marty Heyman, April 07, 2012 7:10
How come we NEVER hear about TERRIBLE PROBLEMS with the Department of Defense benefit plan? It's JUST PEACHY! It doesn't matter what it (or the defined benefit plan for retired legislators and government executives) costs ... we're not worried about that. Just those HOPELESS pensions for the little guys!
Mr. Snark
written by bkrasting, April 07, 2012 9:24
This is a memorable quote from Dean. I will keep it and remind him:

"It is almost impossible to construct a scenario in which pesnion fund returns will be substantially worse than what the pension funds are now assuming."

It takes balls to make statements like this. That, or the person is misinformed.

A more appropriate rate of discount for pension obligations is the the prevailing return on ten-year Treasury Bonds plus 50BP.

That would come to 3%, or less. At that discount rate every private pension fund in the country would be bust.

So who's kidding whom?
The reality of state employee "contracts"
written by annb, April 08, 2012 8:15
@tew: Saying that the reason many state and local employees aren't covered by SS is because we don't want to contribute is ridiculous--our employment contract (even if we are non-union employees and simply employees at will) dictates the terms of our SS contribution. And maybe most workers pay less of their salary towards pensions than SS contributions, but not me (nor the 10,000+ employees in my single branch of government. I pay 8% (mandatory) and have not seen any of the benefits of the payroll tax reduction (which I think is bad public policy anyway with the potential to erode support for SS long term--but that's another debate).

Working for the state is going to cost me, since I will not accrue enough earnings in my SS "replacement" account to replace the reduction in SS payments to me because I am a state employee not covered by SS. I will have to work until full retirement age, and that to get about 90% of what I would have received if I'd been in the SS system.

Pension Returns
written by dean, April 08, 2012 1:45

I made my statement on pension returns on arithmetic -- I suggest you try some. here's a calculator to help guide your thinking http://www.cepr.net/calculators/pecalc.html

Economists and economic policymakers tend to be scared of arithmetic. This is unfortunate. It helps hugely in the understanding of economic policy. Those who learned arithmetic could both the recognize the housing bubble and know that its collapse would be a problem. The same is true for the stock bubble.
Nominal Projected Returns
written by Spinoza500, April 13, 2012 2:59

The article contains two separate points:
1) The excellent point that the US could afford social security if it wishes to do so.

2) A nonsensical forward projection. Dividend yields on US equities are currently 2% . Investors receive dividends not the earnings yield. Let us assume that dividends grow with GDP at 3% (in practice there is a risk of smaller GDP growth and the share of profits in GDP declining from its current high but on the other hand companies may start paying out a higher per centage of earnings by way of dividends e.g Apple ) . You arrive at a projected rate of return of 5%. A reasonable projection for returns (ignoring costs) is therefore 5% real or 7 % nominal (given 2% inflation). Some people might argue that I am being optimistic. See for instance GMO projections for next seven years. Clearly, with a 7 % nominal return on equities and substantial bond holdings, pension funds are unlikely to achieve 8%. Indeed, the projection creates its own dangers as trustees are pushed into allocations into ever higher risk assets (equities, hedge funds, private equity) to make 8% nominal reasonable. A lower target would encourage responsible investing.

Yet all is not lost.
1) Higher inflation may well generate the 8% return although it obviously also impacts the liability number.
2) The taxpayer can pick up the difference. Rather than trying to justify a 8% target, surely the debate should be over the pros and cons of taxpayer subsidy?

Write comment

(Only one link allowed per comment)

This content has been locked. You can no longer post any comments.


Support this blog, donate
Combined Federal Campaign #79613

About Beat the Press

Dean Baker is co-director of the Center for Economic and Policy Research in Washington, D.C. He is the author of several books, his latest being The End of Loser Liberalism: Making Markets Progressive. Read more about Dean.