What Investment Returns Are Public Pensions Really Assuming?

This was a post by Andrew Biggs at AEI. But the permalink is blank. So here it is:

"Most public sector pension plans assume they’ll receive an 8 percent investment return in future years. There are a few problems with that assumption. First, as I’ve previously argued, even if 8 percent is an accurate figure, pensions shouldn’t be using this number to calculate the value of their liabilities. It’s wrong to apply an interest rate derived from a risky portfolio of assets to a liability that is guaranteed to be paid.

Second, some analysts believe these returns are overestimated. Wilshire Consulting, for instance, argues that most plans will receive only around 6.5 percent average returns going forward. If this turns out to be the case, the typical plans’ costs will rise by almost 80 percent. Pension funding is very sensitive to rates of return.

But there’s a third point that I don’t think has been raised before: most public pensions are actually assuming returns well in excess of 8 percent. This makes it even less likely that they’ll be able to meet their projections. Here’s how it works.

A pension calculates its funding health by calculating the percentage of accrued benefits its assets could pay if those assets generated some assumed rate of return, usually 8 percent. The problem is that these calculations are based on so-called “actuarial assets,” which smooth returns over a 5- or 10-year period in order to mask variability from year to year. According to Wilshire Consulting, as of fiscal year 2010, the actuarial value of pension assets was almost 15 percent above market value.

In effect, pensions are assuming that the market value of their assets will first catch up to the actuarial value and then generate 8 percent from there on out. What does that imply for an overall rate of return? Assuming all this happens over a 30-year period, the true assumed rate of return is around 8.6 percent per year.

Will the pensions get that? Maybe—they’ve had a good year so far in 2011. But 8.6 percent is more than 2 percentage points per year above than Wilshire projects they’ll get; over 30 years, the end value of assets would be almost twice as high under the pensions’ assumptions as under Wilshire’s. This just shows—again—that public pension accounting deserves a closer look."