Posted by Jeffrey Brown on Oct 13, 2009
Filed Under (Retirement Policy, U.S. Fiscal Policy)
It is well-known that Illinois has one of the worst track records of funding its public pension plans of any state in the nation. What is less well-known is that the problem is far, far worse than the official statistics would indicate: indeed, the extent of under-funding may be 2.5 times larger than what is typically reported!
To be clear, this is not because anyone in Illinois state government is doing anything unethical – those responsible for calculating the pension liabilities are presumably doing so in accordance with Government Accounting Standards Board (GASB) procedures. Even so, there is near unanimity among economists (and yes, unanimity among economists is rare) that GASB rules themselves are deeply misguided.
This is a complex topic, but I am going to over-simplify for brevity’s sake. As noted in at least two prior posts, the Illinois state constitution provides a rock-solid guarantee that pension benefits will not be reduced or impaired in any way. That makes the benefits that participants have accrued virtually risk-free. It also means that it makes the liability to the State of Illinois risk-free.
Simple (and also advanced) finance theory is unambiguously clear that the appropriate way to discount a risk-free stream of cash flows is to use a risk-free interest rate. While there is not a perfect risk-free asset available for this purpose, municipal bonds and U.S. treasury securities offer reasonable proxies. (I have written on this subject in more depth in a paper with David Wilcox published in the American Economic Review, and I will simply refer readers there for a more in-depth discussion).
Instead of using this clear and rigorously grounded approach, GASB allows public pension plans to discount their liabilities using the expected rate of return on plan assets. There is not a shred of logic to this as an approach for valuing liabilities unless one believes that the liabilities share the exact same risk characteristics as the portfolio in which one is investing (and that is certainly not the case in Illinois. We have riskless liabilities but invest in diversified portfolios of risky assets).
How big of a difference does this make? A relatively new paper by Robert Novy-Marx (University of Chicago) and Joshua Rauh (Northwestern University) estimates the size of the liabilities when calculated using appropriate discount rates. (Read their paper here). What do they find?
In 2008, the four large public pension plans in Illinois had combined assets of $65.7 billion. The combined liabilities of these four plans (calculated under flawed GASB rules) were $151.1 billion, for a shortfall of $85.4 billion.
If one uses a more theoretically appropriate rate on treasury yields, the present value of the liabilities is $284.8 billion, for a shortfall of $219.1 billion! That is more than 2.5 times greater than the official statistics indicate!
What do these numbers really mean? If the state of Illinois wanted to be certain it had enough money set aside today so that it could meet all public pension benefit obligations that have already been accrued, it would need to set aside an additional $219.1 billion. For perspective, that is about 1/3 of Illinois GDP, about 1/3 of state revenues, and about four times the outstanding state debt.
Any amount less than this means that the state would have invest in riskier assets in order to fund the benefits, a strategy that might work … or might make the problem worse.
We’ve all heard to old adage that the first thing to do when you find yourself in a deep hole is to stop digging. In this case, we also need to stop denying just how deep the hole is.