We have a retirement crisis. Pew Charitable Trusts estimates that public pensions are underfunded by $1.38 trillion. Standard & Poor's estimates that S&P 500 companies are $355 billion short of pension obligations. "Of the 500 companies, 338 have defined-benefit pension plans, and only 18 are fully funded," writes The New York Times.
Part of the reason pensions find themselves severely underfunded is because investment returns haven't come close to previous assumptions. After the booming 1990s, many assumed high returns were a steady bet. Whoops. Add with interest rates near 0%, it's exceedingly difficult to generate the kind of returns pensions once assumed.
A common criticism these days is that pensions still aren't living up to reality, and continue to use return assumptions they'll never be able to reach.
Last month, I asked Joseph Dear, chief investment officer of CalPERS, the nation's largest pension fund, what he made of that criticism. His response might surprise you (transcript follows):
Morgan Housel: There are a lot of people who are skeptical of both private and public pension funds' returns assumptions -- that they're way too bullish and that that is dramatically understating their liability gap. Do you think most ... pension funds are reasonable with their return outlooks?
Joseph Dear: Well, CalPERS' assumed rate of return is 7.5%, and we do the development of that assumption in a very public and transparent way. There are some people who think that's unrealistically high and thus we're taking on too much risk; there are others who think that our liabilities are understated. If you're running a pension fund, you have a pretty simple kind of situation. If you want low contribution rates, then you have to have high investment-return expectation, and ultimately you have to achieve that. If you want to reduce the risk in your portfolio, you have to increase the contribution rates.
Now let's put that in practical terms. In the CalPERS system, investment pays for about two-thirds of the pension dollar that's paid out to our beneficiaries. So $0.67 of every dollar that a CalPERS pensioner receives comes from investment, and so one-third comes from contribution rates, and the contribution rates are from the governmental employer and typically also from the worker.
So when you reduce the return expectation from investment, one dollar, and you've got to raise two bucks on the contribution side to make that up, so there is a tendency to say, let's be reasonably optimistic about investment outcome and choose a higher number because it does reduce the cost. That's fine if it works. If it doesn't work, the thing about public pension funds is it's not the pension recipient who bears the risk of inadequate investment return; it's the governmental employer, and thus the taxpayer, who's behind that.
So your policy decision is do you want a higher-risk portfolio, which is going to earn a bigger return and thus keep contribution rates low so the governmental budgets can be spent on direct service delivery? Or do you want a lower-risk portfolio and higher contribution rates and trotting that out. And that's played out all over California at the state level, local government level and around the U.S. Look, if our target was 4%, my job would be a lot easier, right? So it's not like I'm out there saying, Make it higher, make it higher. I just think people should understand the nature of that trade-off.
Now there's another important part of that question which is the liability side, which is a whole other sphere. We determine our liabilities by applying a discount rate which is typically equal to or very close to our assumed rate of return. Now there are critics who say that that's unreasonably high, that we should discount against a governmental borrowing rate or against some other lower rate. And when you reduce the discount rate on your liabilities, you greatly increase the size of those liabilities, so some people think that the liabilities of public pension plans are really understated, because in the United States, most funds use the rate of return as the discount rate.
I think the issue there is you have to decide what are you trying to understand? Are you trying to understand and apply the discount rate so that you have a view of the economic value of the liabilities? Then yeah, you want to use something like the yield curve and that would increase CalPERS liabilities and reduce our funded status considerably. It's also important if you're trying to manage the risk in the portfolio and understand that as the difference between the duration of the liabilities and the duration of the assets and that mismatch being a major source of risk.
But if you're trying to set contribution rates, then it may be appropriate to assume a higher rate of return because what you're then incorporating is the equity risk premium. The difference between what you earn from riskier assets like equities against less risky assets, like fixed income. And I'm not sure it's necessarily a good idea or a good bargain to assume there is no return from equity. There better be or there's something really broken about our economy and our financial system if riskier assets don't provide higher returns. I mean something's badly mispriced.
Now if you get it wrong, that is you underestimate your liabilities, don't earn enough, don't contribute enough, then you have very, very painful consequence in the future of having to make direct cash payments to provide for benefits. So the risk has shifted to the future and those taxpayers could have a large burden from not having saved enough going forward.
The other mistake is you're saving too much today, so today's taxpayers are having to shoulder too big a burden and workers and government employers and then you end up in the future with too much money. Now too much money may not sound like a big problem, right? But in the pension world too much money means somebody's going to think of a way to increase benefits, right? So I think it's really good to keep an eye on the assets and liabilities, to use different discount rates and show what the differences are and to have a robust public discussion about what a pension fund is assuming in terms of its investment and let people question it, let our board members question it, and then look at the liabilities and then see how things change.
In a super-low rate environment like we have today, liabilities are definitely bigger. Are interest rates going to stay? Is the 10-year Treasury going to stay at 1.6% indefinitely? I doubt it. So it's going to go up and the interest rates will go up and even those who want to do the yield curve will see liabilities coming down. That's a long answer, but it's a really important question.