Gary Findlay, Executive Director of the Missouri State Employees’ Retirement System confronts critics of current pension accounting standards not with substantive arguments, but through a series of ad hominem attacks and non sequiturs (see Academic and Think Tank Hypocrisy In the Public Retirement Arena). As one of the producers of what Mr. Findlay terms “think tank diatribes” in favor of market valuation of public pension liabilities, a response seems in order.
Public sector pensions adhere to a principal known as “interperiod equity,” under which, in the Governmental Accounting Standards Board’s terms, “taxpayers of today pay for the services that they receive and the burden of payment for services today is not shifted to taxpayers of the future.” Accounting methods should reveal the degree to which a plan has met this standard.
But current methods clearly fail that test. When guaranteed benefit liabilities are discounted using the interest rate expected for a risky portfolio of assets, the resulting liability figure ignores the very real possibility that the assts will fall short of the value needed to meet plan obligations and the taxpayer will be obliged to make up the difference. Unless plan sponsors implausibly believe they can achieve their expected rates of return with certainty, a significant contingent liability is being omitted. Market valuation techniques, by contrast, capture the costs of taxpayer guarantees of accrued benefits.
Mr. Findlay does not confront these arguments. Rather, he makes a number of arguments that either are simple personal attacks or are essentially unrelated to the pension accounting debate.
First, he says, “Given their vehement opposition to putting retirement assets at risk, one might expect the academics to avoid risk in the investment of their own retirement savings.” This reflects a simple misunderstanding of market valuation, which focuses on how to value liabilities, not how to invest assets. Moreover, if an individual invest his assets in equities, he accepts both the return and the risk. When a public sector pension shifts to riskier investments, this risk imposes a contingent liability on future generations. Yet public pension accounting nowhere acknowledges this cost.
Second, Mr. Findlay questions the funding sources for critics of public pension accounting. I personally have received no direct funding for pension work from anyone with a financial interest in the accounting debate. And for what it is worth, most of the vested interests – financial advisors, hedge funds, private equity, public employee unions, and so forth – are on the other side. But, unlike Mr. Findlay, the critics of pension accounting have stuck with substantive arguments and attempted to resolve disagreement through debate rather than innuendo.
Finally, Mr. Findlay says that, “To support their theory, the academics seem fond of using the first decade of this century which included the bursting of the tech-bubble at the outset and, more recently, the global credit crisis.” Mr. Findlay counters by citing average market returns over longer periods. I am not aware of critics relying on poor market returns in recent years to support their theory, since the very essence of market valuation is that the discount rate applied to a liability should be based upon the risk characteristics of the liability, not of any assets set aside to fund that liability. In other words, recent market returns – good, bad or indifferent – are not relevant to the market valuation debate.
The bottom line is that when a benefit liability must be paid with 100 percent certainty, a plan should be considered “fully funded” only when it is capable of paying that liability with 100 percent certainty. Market valuation satisfies this criterion. Current accounting standards do not, encouraging pensions to promise too much, fund too little, and take too much investment risk.
Andrew G. Biggs is a resident scholar at the American Enterprise Institute, and thinks he may well be one of the “think tank hypocrites” referred to in Mr. Findlay’s article.