The Economist on “the muddle-headed world of American public pension accounting”

As state governments across the nation struggle to address a public pension underfunding crisis they can no longer deny, The Economist is the latest major news outlet to turn its gaze on the ongoing debacle. In the current issue, the magazine’s “Buttonwood” column draws a sketch of U.S. public pension accounting that is not only dysfunctional, but that runs against plain common sense.

American public-sector schemes discount their liabilities by the expected return on their assets. The riskier the asset mix, the higher the assumed return—and the lower the bill appears to be.

This is an odd way of thinking. Suppose a car company borrowed $10 billion in the form of a 20-year bond to build a manufacturing plant and planned to pay off the debt with the profits from running the plant. The car company will assume a higher return on capital than its financing cost (otherwise it should not build the plant). But it still has to recognise the $10 billion bond liability on its balance-sheet. It cannot say it owes only $2 billion because it expects a very high return.

The reason is clear. If the plant fails to earn a high return, the firm will still be liable to repay the bond. Similarly, if pension schemes fail to earn a high return on their assets, they still have to pay benefits. Final-salary pensions are a debt-like liability.

The Buttonwood columnist (currently Philip Coggan) notes recent changes to the nation’s largest public pension plan, the California Public Employee Retirement System (CalPERS), that would require greater employer contributions. But such changes will be ineffective in the long run unless they were to be accompanied by major reforms that address some of the structural factors that have made public pension shortfalls severe and chronic: payouts based on final-year pay, negotiation of benefits through collective bargaining, benefit increases through binding arbitration, politicized pension fund boards, and flawed accounting standards.

Last year, The Economist itself noted the problem of politicized public pension boards in the U.S., in a comparison with their Canadian counterparts, which “attract people with backgrounds in business and finance to sit on their boards, unlike American public pension funds, which are stuffed with politicians, cronies and union hacks.” That hasn’t changed over the past year.

The main problem with accounting standards centers around the discount rates used by most public pension plans, which are based mainly on the expected rate of return on investments. Many base them on an unrealistically high rate of 7 to 8 percent. While many pension fund may achieve such returns on average over a period of years, they typically do not achieve such returns year on year, while liabilities in the form of payout commitments to retirees continue grow without interruption. The result is a growing funding gap.

While the Government Accounting Standards Board (GASB) last year issued new rules for public pensions, Coggan notes, “the revised rules still throw up absurdities. In a paper for the Financial Analysts Journal, Robert Novy-Marx of the University of Rochester argues that by destroying assets invested in cash a scheme can reduce its deficit by increasing the expected return on remaining assets. ‘A plan can sometimes improve its funding status by literally burning money,’ he remarks.” In his paper, Novy-Marx explains with the following example:

Consider two pension plans, plans “A” and “B.” Plan A has a single member, a 35 year old worker with five years of service who plans to retire in thirty years with a projected salary of $105,000. The plan holds $10,000 of stocks that have an expected return of 10%. Plan B also has a single member, identical in all ways to that in plan A, and holds the exact same stocks, but additionally owns $10,000 dollars worth of T-bills providing a risk-free yield of 4%. Common sense demands that plan B is better funded than plan A by exactly $10,000. Under GASB’s methodology, however, Plan A appears better funded than plan B.

In short, a pension fund can improve its funding profile by holding higher-yielding but riskier assets. As Andrew Biggs of the American Enterprise Institute noted last year, “the new regulations tell pensions that boosting investment risk automatically makes them better funded, before a dime of higher returns have been realized. Since riskier investments have higher expected returns, shifting to a riskier portfolio allows public pensions to use a higher discount rate, instantly improving their funding status.”

In a study also cited by Buttonwood (and Biggs), Aleksandar Andonov and Rob Bauer of Maastricht University and Martijn Cremers of Yale University point out that new GASB rules would incentivize overly optimistic revenue projections for pension funds at risk of falling into the underfunded category.

Specifically, GASB (2011) proposes severing the link between liability discount rates and expected rates of returns but only for funds that can be classified as underfunded, i.e. where the plan assets are not “projected to be sufficient to pay benefits and the net position projected to remain after each benefit payment can be invested long-term.” … However, for funds where plan assets are projected to be sufficient, the wrong incentives would remain in place as the assumed rate of returns of the assets could still be used to discount the liabilities. Moreover, these projections (of whether or not assets are sufficient to pay the benefits) can seemingly still be based on liability discount rates that are linked to expected asset returns. As a result, the new GASB proposals would create even stronger incentives to camouflage liabilities and engage in reckless risk-taking for funds that are close to being underfunded (and that may indeed be underfunded if liabilities would be discounted at – currently – lower high-quality municipal yields) and that rationally want to avoid being classified as underfunded.

A more accurate picture of pension funding would be provided by projecting revenues based on low-risk investments, such as Treasury bonds, as AEI’s Biggs and the Mercatus Center’s Eileen Norcross have proposed.

However, such policy changes can be undone, especially considering the defined benefit/diffuse costs scenario that public pensions pose. As Andonov, Bauer, and Cremers describe it, “Current stakeholders, including boards, members and their representatives as well as politicians and taxpayers, have a direct incentive to underestimate the current value of the existing liabilities and transfer this risk to future generations. In this era of general underfunding, this will allow current members to receive higher benefits without boards and politicians having to make tougher choices now.”

Assembling a pro-reform coalition to counter the status quo, while not impossible, would be incredibly difficult. Even more difficult would be holding such a coalition together and ensuring that reform last. Government employee unions and the politicians they support, on the other hand, aren’t going anywhere, and likely would reverse any reforms at their earliest opportunity. Therefore, public pension reform, if it is to last, should include alternatives to the defined benefit model, whereby payments are guaranteed regardless of a pension plan’s level of funding.