I just returned from a great conference co-sponsored by Vanderbilt, U. of Missouri, and my department at the University of Arkansas on teacher pensions. One of the major issues discussed at the conference was the financial sustainability of those pension systems. And at the heart of that discussion was a debate over the appropriate discount rate to apply to pension liabilities. Essentially, the debate was over what we should assume to be the return on pension investments in the future.
This may seem like an arcane and dry issue, but let me tell you that it is incredibly important. If you don’t care about it now, you will care if those pensions fall short of their assumed rate of return on investments over a long period of time. If teacher pension plans run out of money, taxpayers are on the hook to make up the difference to the tune of tens of billions of dollars. I know that in the era of trillion dollar bailouts tens of billions don’t seem like a big deal, but after a few of these trillion dollar bailouts there may not be much left for the teacher pensions if they go kablooey (that’s the technical term).
Public pensions are generally considered well-funded when they have assets that are roughly 80% of liabilities. Don’t ask me why it is considered OK to know that you are short 20% of what you owe, but most folks who work on these issues just don’t think it’s realistic to have 100%. Besides, when plans get close to 100% funding of their liabilities they tend to increase the generosity of their benefits to bring that ratio down.
About 40% of the major teacher pension plans failed to meet the 80% standard for being adequately funded as of 2007 — before the current market meltdown. But the 60% that did meet this standard did so assuming that they would return 8% on their investments going forward. Is 8% the right rate to assume?
Remember that teacher pensions are promising to pay teachers a certain amount of money 20, 30, or 40 years from now. They also expect to receive a certain amount in contributions from the teachers, from their employers (the state or school districts), and from investment returns on those funds. Whether you have enough money to make the promised payments to teachers is extremely sensitive to the assumed rate of return on investments.
Some folks, often public finance economists, argue that assuming an 8% return is irresponsible. They say that the market tells us what rate we should use and it is the risk-free rate of long-term US treasury bonds, currently earning a little less than 4%. To get a significantly higher return one has to take-on significantly more risk, with the distinct possibility that one will earn far less than 8% and even less than 4%.
If one assumes a 4% return rather than an 8% return, a teacher pension that would have been 70% funded assuming 8% would drop to 44% funded assuming a 4% return. Just how under-funded teacher pensions are hinges heavily on whether we assume a 4% or 8% return.
Other folks, often pension plan actuaries, argue that the 8% assumption is reasonable. They point to the historic rate of return on pension assets to support 8% as a reasonable figure. They also say that risk is different for the government since it is a perpetual entity. They can endure losses for a long time and eventually make up for them in a way that a private organization cannot.
I find it hard to support the 8% assumption. Historic rates of return are hardly reassuring. We may have received an average return of 8% over the last century, but who knows whether the next century will resemble the last one? And who knows if we might be like Japan, where stock indices are still less than half of the peak they obtained three decades ago. There is a good reason why investments ads say past returns are no guarantee of future returns.
And being a perpetual entity provides no protection if future rates turn out to be less than 8%. Being perpetual only means that one can endure a very long period of under-performance. That assumes that eventually the mean return will revert to being 8%. But why should that be? What if the mean return over the next infinite period is only 4%? No matter how long we wait, we would never get 8%.
If it were really true that the government could be assured an 8% return while private entities can only be assured a 4% return, then it would make no sense to have a private financial industry. The government could borrow at 4% to buy up the entire private sector and guarantee everyone an 8% perpetual return. People should give all of their money to the government to invest so that they could be assured the 8% return with no risk. If they invest it privately they can only be assured a 4% return with no risk.
But defenders of using an 8% discount rate respond saying that if you assume a 4% rate and end up making 8%, the plans will be grossly over-funded. That would essentially involve the transfer of wealth from this generation to a future generation, which would be grossly unfair.
Of course, the only way that the pensions could get more than 4% would be if they took on additional risk by investing in equities, real estate, hedge funds, etc… If you lent me money at 4% and I took it to Vegas and put it all on black, I might also come back with a lot more than the 4% I would owe you. I just can’t be guaranteed to come back with extra money. Similarly, the teacher pension plans cannot be guaranteed the 8% return and should not assume extra risk in the attempt to get it. If pensions switch to a 4% discount rate they should probably be restricted in their investments to mostly risk-free investments, like government bonds.
Switching to a 4% discount rate is going to be painful, especially with already under-funded teacher pension plans and with the recent market meltdown. But if we don’t do it, eventually we may well face a future financial crisis brought on by pensions rather than by housing. Let’s spot the bubbles before they burst.
In addition to reading the papers listed at the conference web site, you may also want to check out this new piece by my colleagues Bob Costrell and Mike Podgursky in the current issue of Education Next.
Posted by matthewladner 
(Guest Post by Matthew Ladner)
Jay P. Greene's Blog 