How big is the public pension funding gap?
California is facing a whopping $26.6 billion deficit, but budget discussions this week focused on a question that might sound small: whether the state’s public retirement system should cut its assumed future return on portfolio investments from 7.75 percent to 7.5 percent.
The California Public Employees’ Retirement System (CALPERS) decided to hold steady at 7.75 percent — and the impact was significant. It meant the pension fund won’t have to dip into general revenues for an additional $200 million or more.
The CALPERS decision underscores an ongoing debate about the size of unfunded public sector pension liabilities as battles rage in state capitols between elected officials and employee unions: What is the proper assumption for return on portfolio investments over time?
The debate isn’t academic; the return assumptions are critical in determining the size of unfunded public sector pension liabilities.
Most state and municipal pension funds assume a long-term rate of return around eight percent, reflecting a portfolio invested in equities, bonds and alternative assets such as hedge funds. That number reflects the approach preferred by actuaries. It’s supported by actual investment history, and it’s endorsed by the Governmental Accounting Standards Board (GASB), the independent accounting and financial standards reporting group for state and local governments.
Economists, meanwhile, prefer a more conservative approach adhering to standard financial theory: cash flows should be discounted at a rate that reflects their risk. Measuring pension funds, that requires use of a “riskless rate of return” assumption reflecting what could be earned on Treasuries or corporate bonds, around four to five percent.
Joshua Rauh, an associate professor at the Kellogg School of Management at Northwestern University, says the riskless rate method pegs aggregate unfunded liabilities of state and local governments north of $3 trillion — a sum larger than the overall $2.6 trillion of debt on state and local government balance sheets. The GASB-endorsed eight percent assumption indicates a smaller $1.3 trillion aggregate funding gap, he calculates.
“You can write out a simple equation on how to keep a pension fund in balance,” says Karen Harris, vice president of capital markets research at Callan Associates, an investment consulting firm that has researched the riskless rate-of-return issue. “The benefits paid plus expenses of running the plan must equal contributions plus investments. You can only pay for the program through contributions or investment returns, or a combination of the two. So the rate of return assumption is important in understanding how you are funding benefits.”
The argument for eight perecent is based on actual historical rates of return — and the fact that investment time horizons for pension funds are very long. Harris says that for a worker who starts a job at age 20 and retires at 60, a pension fund won’t need dollars to pay out benefits for 40 years, and the total investment horizon is 55 years, when life expectancy is included.
“Next, an actuary asks what she can earn over that time horizon,” she adds. “And what she’ll say is, ‘I can earn something.’” Part of that return will come from buying the longest government and corporate bonds available, holding them to maturity and reinvesting the proceeds; the difference must come from returns on equities where risk is involved on everything but dividends.
Pension portfolios very likely will yield eight percent or better over time. But if returns fall short, public sector pensions must turn to taxpayers to make up the gap, Rauh points out. “If they fail to make that eight percent, taxpayers are on hook,” he says. “Governments need to come clean about what they actually owe.”
The two rate of return assumptions produce staggering differences in funded ratio projections. Rauh has projected unfunded liabilities using the two methods as of June 2009; the chart below shows how the states with the largest unfunded liabilities compare (see the yellow-highlighted column, which compares the plan sponsors’ stated unfunded liability against liability using a riskless return assumption).
The riskless rate debate isn’t just between actuaries and economists — it’s political. In Washington, Republican legislators worry that states will seek federal pension bailouts, and have been pushing for legislation requiring more conservative accounting in order to get the full extent of the problem into the daylight. Progressives argue that the riskless rate approach overstates the underfunding issue and creates other policy and political challenges.
“It’s too conservative, says Elizabeth McNichol, a senior fellow at the Center on Budget Planning and Priorities who specializes in state fiscal issues. “There’s a danger in using too low a rate — if your contribution is bigger than needed you have to cut funding for other things in the present, and may end up with an over funded pension in the future. When that happens, it’s too easy for politicians to over-promise by expanding benefit promises in the future.”
Solutions? Rauh favors federal legislation that would require state and local governments to meet all future pension commitments using a riskless calculation — or lose their tax-exempt status on bond issues. “They can choose to fund their defined benefit pensions, or adopt less expensive defined contribution plans,” he says. “If we do that kind of fundamental reform you could have states issue very large amounts of new debt with very long maturities to restructure the legacy liability and spread it out over a long time.”
Harris endorses the riskless rate method for disclosure and measurement purposes, but thinks plans should be permitted to make contributions based on expected long-term rates of return.
“The alternative is to use the riskless rate, and only if they earn a greater rate of return, can they then reduce their future contributions,” she says. “Financial economics would say start contributing today based on the riskless rate and only cut the contributions back when they realize the higher investment returns.”

Comments RSS










Defined contribution plans are not necessarily less expensive. In Massachusetts, teachers hired in recent years contribute 11% of their paycheck to the pension system. If the pension system meets its 8.25% investment target, then the state ends up contributing little or nothing.
Defined contribution plans offer the employer cost-certainty by shifting the investment risk onto the employee. On the other hand, there would be political pressure for employers to match employee contributions at a 3% to 6% rate — money they do not presently contribute to the pension system. Moreover, they might need to start participating in Social Security (from which they are presently exempt). That’s another 6.2% contribution.
That’s potentially much more expensive than simply guaranteeing the pension returns on employees’ contributions.
State’s have failed to adequately fund their pension obligations even with the current assumptions. Lowering return assumptions won’t create more money, just a sharper political dialogue about how money gets spent. (As if that dialogue isn’t sharp enough already.) Long term equity returns are at the heart of this debate. If equities fail to return in the next 100 years what they have in the past 100 years (9-11%)then the current actuarial assumptions will be wrong and pensions will fail. In any event, requiring State’s to solve current underfunding problems is the first step. Lowering assumptions now is like raising interest rates on a home owner who faces foreclosure. It just makes matters worse.