Economic Meet Politics in Public Pension Accounting
For months I’ve been trying to avoid writing a post about public pension accounting because it’s dull, to say the least . However, the future of our public pension plans probably will be determined by how government accounts for the liability it owes current employees and retirees. I was going to do a four-part series on the various assumptions that go into calculating the assets and liabilities of a pension plan. However, I’m fairly certain that I’d have no readers left by the time I completed the series. So instead, I am going to try to make sense of the biggest pension accounting issue in a single post.
Here’s the problem in a nutshell. According to today’s accounting rules, public pensions have been undervaluing the liability they owe beneficiaries for decades. This miscalculation has permitted state and local governments to invest less than they should have in their pension plans over the years. As a result, the deficit of liabilities over assets has been smaller than it otherwise would be if realistic assumptions were used. The lack of proper estimates and resulting public investment is not a problem unique to pension plans. Government has been scrimping on all sorts of investments, such as bridges, roads, schools, and other long-term assets. Why should pension plans be any different?
For some unknown reason, the Government Accounting Standards Board (GASB) has always allowed public pension plans to use the investment rate at which assets are expected to grow to discount the liabilities. The previous sentence probably gave you a twinge somewhere behind your eyes, so lets take a step back. In order to calculate a pension’s liabilities you have to make all kinds of assumptions about both the existing retirees and current employees in order to arrive at the amount of money the pension plan will have to pay out each year far into the future.
With respect to the existing retirees, the problem is similar to life insurance or annuities. The actuary has to estimate how long the beneficiaries will live and therefore receive their payments. For the current employees, the actuary has to estimate a series of factors in order to project how and when existing employees will start to draw pension benefits and how big those benefits will be. When all those assumptions are completed, the actuary will have a set of cash flows starting in the current year and stretching far into the future. In order to compare those future benefit payments to the pension’s available assets, the actuary has to discount those future payments back to the present.
Proper economics would require those cash flows to be discounted by some sort of high quality interest rate. The idea is that the constitutionally protected pension benefits are a long-term obligation of the government and should therefore be treated like a long-term bond. In fact, corporate pension plans imperfectly follow this method. GASB, as I mentioned, allows public pensions to discount using the expected return for the pension plan, which is a blend of the returns for stocks, alternatives, and bonds. So the typical public pension plan discounts its liabilities using a 7.5% or 8% rate rather than a long-term bond rate which would probably be half as high or less. Discounting by a higher number artificially shrinks the size of the liability and provides a false picture.
GASB and the rating agencies are moving to eliminate the deficiencies in public pension accounting and institute a proper discount rate for the liabilities. Pension trustees and treasurers have uniformly opposed any change in the way pensions assets and liabilities are discounted. I suspect that public officials know full well that the current accounting methods are defective and don’t represent economic reality. However, they’re also aware that proper economics may be bad short-term politics. Bigger pension liabilities and the resulting higher appropriations are probably not palatable to lawmakers. On the other hand, the failure to properly fund pension plans will likely doom many of them in the coming years. It’s like the proper maintenance of a bridge. If you don’t tend to it, it will eventually collapse.
Let’s look at the Teachers’ and State Employees’ Retirement Plan of North Carolina to get a feel for the impact of applying more appropriate accounting principles. According to NC’s actuary, the pension’s assets were worth $58.1 billion and the liabilities were $61.8 for a deficit of $3.7 billion. North Carolina’s funded ratio was said to be 94.0%.
Along comes Moody’s with a study that uses a more appropriate discount rate. Instead of discounting NC’s pension liability at 7.25%, Moody’s used 5.67%. As a result, NC’s deficit doubles from the reported $3.7 billion to $7.5 billion. You can’t blame Treasure Cowell for not liking this result. However, even Moody’s discount rate of 5.67% is too high.
Except for the fact (and it is a hard fact to ignore) that we have a radical and economically ignorant legislature, the $7.5 billion deficit is infinitely manageable. First and foremost, it is not a deficit that needs to be closed right away. Second, as Moody’s points out, the deficit is tiny compared to the size of NC’s economy and our overall tax resources. Third, it simply requires modest appropriations each year to give the pension sufficient assets to meet its obligations.
If you reside in Alaska, Illinois, Connecticut, Kentucky, or Louisiana, the picture is rather ugly. However, you probably knew that already because your state’s pension already had a big deficit under the existing discount method.
Accounting is something that most of us try to avoid thinking about. However, ignoring accounting issues carries a large price. In the case of public pension plans, bad accounting rules have led to large deficits, which will probably sink some plans and make for ugly politics in states like North Carolina. Most people believe that pension plans have generated deficits because the investments haven’t performed as advertised. While it is true that most investment programs have not lived up to their billing, investments are a tiny part of the problem.
 And, the vast majority of plans have been overestimating the rate at which the assets will grow.
 I suspect that when GASB first decreed that the estimate rated of return could be used to discount the liabilities, it didn’t matter as much. Interest rates on bonds were significantly higher than they are today, so the difference between the expected rate of return and interest rates was small. This hasn’t been the case in the last two decades.
 All of this is a bit more complicated, especially if public pension plans allow for cost of living adjustments (COLA), but we’ll keep this simple, since it still makes the point.
 The actuary has to estimate salary increases, turnover, seniority, public employment growth or attrition, and mortality among other factors. Salary increases are a critical element because the future benefit is based on the average salary in final years before retirement. In recent years, the salary increase assumption has tended to be too high as state and local government have limited or eliminated salary increases.
 The assets are massaged a bit by the actuary as well. However, this distortion is much smaller than the improper estimation of pension liabilities.
 Congress has had a habit of intervening from time to time in order to keep the corporate discount rate from being too low, and thereby driving large corporate pension deficits and corporate contributions.
 Adjusted Pension Liability Medians for US States, Moody’s Investor Services, June 27, 2013, page 9.
 Ibid, page 10.