No time to worry about CalPERS

By Felix Salmon
April 12, 2010
responding forcefully to a rather silly Stanford policy brief which gets very alarmist about California's pension liabilities. There are so many enormous and immediate fiscal problems facing California right now that it seems utterly pointless to put out a paper saying that the state should inject $200 billion into its pension funds -- especially when the logic of the paper is as confused as this:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Well done to CalPERS for responding forcefully to a rather silly Stanford policy brief which gets very alarmist about California’s pension liabilities. There are so many enormous and immediate fiscal problems facing California right now that it seems utterly pointless to put out a paper saying that the state should inject $200 billion into its pension funds — especially when the logic of the paper is as confused as this:

The CalPERS portfolio has had returns averaging 7.91 percent over the last 25 years, with a standard deviation of 11.91 percent. As expected, the high standard deviation means that 68 percent of the time, returns range from –4.0 percent to 19.82 per­ cent. Historically, if CalPERS had simply invested in investment­ grade corporate bonds, the fund could have earned 7.25 percent, only .66 percent less than it has earned with its highly volatile portfolio. This small reduction in earnings would have allowed CalPERS to reduce volatility by a full 7.68 percentage points.

Therefore, in order to avoid future severe underfunded scenarios, we recommend that CalPERS, CalSTRS, and UCRS allocate more of their investment portfolios to fixed income asset classes, thereby reducing risk with a minimal loss of long term investment performance.

I’m not entirely sure where to start on this, but are the Stanford wonks really unaware that the rate of return on fixed-income investments over the past 25 years is largely a function of the fact that interest rates have been declining steadily over that time? And that now they’ve reached zero, they can’t really continue to do so for the next 25 years?

On top of that, the Stanford types seem to think that it makes sense to use a risk-free discount rate to calculate pension-plan liabilities, while even they admit that the assets shouldn’t be invested in a risk-free manner.

I do like the way that the Stanford paper tries to look at the probability of a shortfall of various magnitudes, rather than simply boiling things down to one number. But I’m not convinced by the methodology it uses to arrive at those probabilities, or by the way that the paper ignores all the political realities surrounding pension contributions and payouts.

The fact is that a defined-benefit pension scheme is always going to run the risk that it won’t be able to meet its liabilities as they come due. The California pension plans constitute an attempt to save hundreds of billions of dollars to pay for the pensions of the state’s workers; the attempt might succeed, or it might not.

But right now there are clearly more important and urgent things to do with California’s tax revenues than throw them into a pension pot to support the retirees of the 2040s and beyond. CalPERS might not be perfect, but it’s a lot less dysfunctional than most of the rest of the state government. Let’s get our priorities straight here.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
4 comments so far

Your right Felix, we shouldn’t worry about CalPERS. A pension plan underfunded by $500 billion never hurt anyone.

Posted by RyanM | Report as abusive

Overreact much?

The “Stanford wonks/types” you talk about are students, probably just out of undergrad, who have written this paper as partial requirement for their Master’s degrees in public policy. It’s not like it had any chance of being taken seriously by the Governor’s office anyway.

Best,

Vincent

Posted by VincentMI | Report as abusive

“On top of that, the Stanford types seem to think that it makes sense to use a risk-free discount rate to calculate pension-plan liabilities, while even they admit that the assets shouldn’t be invested in a risk-free manner.”

Why shouldn’t they use a risk-free discount rate to discount liabilities? Don’t confuse assets (which are in present value terms and may grow at greater than the risk-free rate) and liabilities (which are known in future dollar terms, but are certain to be incurred). Absent legislative or judicial action (and I’m not sure what kind of hoops they’d have to jump through to change existing liabilities) they *will be* making the payments in the future, and discounting them at anything but the risk-free rate would understate the obligation.

Posted by Beer_numbers | Report as abusive

How does that disclaimer go? “Past performance is not indicative of future results.” Time after time, we’ve seen analysis that PROVES an investment in XYZ is vastly superior to conventional wisdom (over the past 25 years). That’s usually a sign that XYZ is trading in bubble territory.

I wonder, what was the average return on stocks from 1975 to 2000? What was the average return on real estate from 1980 to 2005?

How long until the bond market collapses 30%?

Posted by TFF | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/