Public pension accounting and low nominal GDP growth
I was reading through an article at Reuters on the problem that public pension plans are having with shortfalls that can only be made up via increased tax revenue, increased investment returns, or decreased benefits. The Reuters article says a recent study shows a shortfall of $1.2 trillion in public pension plans. This is a huge issue that is at the core of debates about state and local government taxes and many predictions of woe for the municipal bond market. I just want to look at this briefly in the context of today’s low nominal GDP growth.
My own macro view here is that public pension shortfalls are only visible as a problem during recession when tax revenue and investment returns are both falling at the same time. If you go back and look at the Meredith Whitney Brouhaha at the end of 2010, what I said at the time is that “my takeaway from the state and local government financial situation is that there will be defaults. The question is when and in what measure. If we see recovery through to the end 2011 and beyond, there isn’t likely to be a crisis in municipal bonds.” The question is whether there eventually will be a crisis.
Here are a few thoughts from the Reuters article:
The 100 funds Milliman studied used a median rate of return for their investments of 8 percent. But the recession slashed into the market, dropping actual median returns to just 3.2 percent for the last five years, according to data from Callan Associates.
The difference has prompted critics to claim that the funds are underreporting their unfunded liabilities, or the gap between what they’ve promised to pay retirees in the future and what they’ll actually have on hand to cover the benefits.
The firm, which has done actuarial work for nearly all of the U.S. states in the past, examined each individual fund in the study, using market valuations instead of smoothed valuations to measure assets and recalibrating liabilities based on Milliman’s own benchmarks of expected long-term returns.
The firm found that the median discount rate should actually be 7.65 percent, rather than the 8 percent median rate the funds used in aggregate.
I don’t buy this math at all. If a diversified pension fund invests in stocks and bonds in a low nominal growth scenario, it needs higher stock returns than it would in a high nominal GDP scenario because low growth connotes low bond yields and, therefore, low returns. To get to 7.65% return, a fund invested 60-40 in stocks over bonds would need well over 10% return on shares if yields stay low in the US as they have in Japan. Is that likely in a low nominal growth environment? I say no.
I think Japan is the place to look for these issues because the demographics there are more stark than they are in the US and the post-bubble malaise has been especially harsh. Bloomberg posted a good article outlining the situation with Japan’s pensions in August that I recommend. I would make two predictions based on that experience. First, if policy makers are unable to get nominal GDP growth up, then US government yields will stay low and returns from those bonds will shrink. Second, if this does occur, then state and local governments will feel much more stress than the central government because they don’t borrow in a currency they can create. If this problem is fixable by relatively small changes, as a friend of mine recently wrote, that makes a compelling case for getting employment and nominal GDP growth up as soon as possible. I think I will leave it there.