Now that Wisconsin Gov. Scott Walker has signed a bill eliminating collective bargaining for most public-sector workers in that state, the political spotlight is shifting back to the issue that first provoked the conflict: the cost to taxpayers of public pensions.
Rep. Devin Nunes (R-CA) provided a new focal point for the discussion when he introduced the Public Employee Pension Transparency Act. This measure would require state and local governments to report certain information about their pension plans to the U.S. Treasury once a year, including how they expect to eliminate any current unfunded liability in their plans.
They would calculate that liability in two ways: using the actuarial assumption they currently use, which is usually discounted based on the expected long-term rate of return on plan investments; and again using a “risk-free” discount rate using U.S. Treasury bond yields. Any state or local government that failed to submit these reports would lose their right to issue tax-exempt bonds, potentially damaging the market for state and municipal debt.
The Nunes bill, which is supported by some leading House Republicans including Budget Committee chair Paul Ryan of Wisconsin, was also endorsed last month by Moody’s Investor Services. It would “provide new incentives to state and local governments to take action to ensure public-employee pension plans’ long-term viability,” the rating agency said in a statement.
Both supporters and critics of the Nunes bill agree it would have a significant impact on investors in public-sector bonds and on public pension funds themselves as investors. But they disagree sharply on what that impact would be. Proponents say using a “risk-free” rate of return as the discount rate for pension obligations yields a truer – and quote sobering – picture of their funded status. It would also bring much-needed attention to the investment policies of public-sector pension funds, which critics say have tended to chase returns in recent years in order to make up for funding shortfalls.
Opponents say a discount rate based on Treasury bonds is inappropriate for public pensions, distorting the picture of their funded status by tying employer contributions to volatile, unrelated shifts in interest rates. This would make a manageable set of problems look much less soluble, presenting state and lawmakers with two equally problematic alternatives: either to contribute much larger amounts to the pension plans, possibly resulting in overfunding; or to phase out the plans altogether. Critics of the Nunes bill also defend public funds’ investment strategies, which they say have not been to chase returns but rather to seek long-term balance by creating more diversified portfolios.
Public pensions currently have some $2.8 trillion in assets, which by their own calculations gives them about $700 billion less than they need to cover liabilities over the next three decades. The current discussion about the discount rate began with a paper published in Fall 2009 by two academics, Joshua D. Rauh of Northwestern University’s Kellogg School of Finance and Robert Novy-Marx of Booth School of Business at University of Chicago. They argued that the current discount rate used by public plan sponsors was much too conservative, and that in reality the plans’ unfunded liability was much larger – as high as $3.23 trillion.
Deciding who’s theory to rely on could make a huge difference for taxpayers. State and local governments currently devote an average 3.8 percent of their annual operating budgets to pension funding, according to a study the Center for Budget and Policy Priorities. That includes a small number – New Jersey, Pennsylvania, and Illinois most prominently – that had let extremely large shortfalls develop. To close a $700 billion liability gap over 30 years, public employees would have to contribute 5 percent to pensions, which the CBPP estimates would not be unduly burdensome once the economy has recovered. If the gap was more than four times that large, according to a study by the Center for Retirement Research at Boston College, the requirement could rise to 9%.
Who’s right? In an argument that formed the basis of thinking in the Nunes bill, Rauh and Novy-Marx contended that only a risk-free discount rate – one based on Treasury bond yields – is appropriate to public pensions, because their liabilities consist of guaranteed payments to retired workers. Currently, most public plans use a discount rate based on an assumed long-term investment return of about 8% on the plan’s assets. That is not appropriate, Rauh and Novy-Marx argued, since those returns come from risky investments that do not match the variability and timing of the plans’ liabilities.
From the point of view of taxpayers, public pension funds actually consist of two things, argues Andrew Biggs, a resident scholar at the American Enterprise Institute: the assets presently in the plan, and an “implicit put option” that represents the taxpayer guarantee that retirees will receive their benefits. Such a guarantee would be quite expensive to buy from, say, a private insurer, if it were available at all. Applying a risk-free discount rate is necessary to price that guarantee into the value of plan liabilities and give taxpayers a true picture of the promises they have made.
Contrast that, Biggs says, with the current practice, under which it is OK for state governments to call a pension plan fully funded when its projections say the investment portfolio will earn, say, 8 percent over 30 years. In reality, chances are only 50/50 that the investments will make that 8% mark, rather than under- or overperforming.
The expected-rate-of-return measure has been used by pension actuaries, and widely accepted, for many years, however, and Biggs, for one, acknowledges that the discount rate debate is “kind of an actuaries-vs.-economists argument.” Keith Brainard, research director at the National Association of State Retirement Administrators, says the argument for the risk-free rate is “perfectly in accordance with economic theory, but that’s what it is: an interesting economic argument.”
Treasury bond yields, which would be the basis of a risk-free discount rate, are extremely sensitive to interest rate shifts, Brainard notes. Using them to calculate required pension funding would result in “wild gyrations” from year to year. When interest rates were low, contributions could be disproportionately large; when interest rates were high, public employers might get an undeserved holiday from making contributions. In any event, state and local governments would have difficulty creating their budgets every year, because fluctuations in their pension obligations could be extreme.
Biggs suggests one way to moderate that problem might be to allow public employers that adopt the risk-free discount rate to “smooth” their pension contributions over a period of years, as they do presently with the rate-of-return calculation. Effectively, that would average out some of the peaks and valleys, making budgeting easier.
Brainard also calls into question the need to apply a “riskless” standard to pension benefits, even when they are guaranteed by public employers – written into the state constitution, in some places. First, because a discount rate based on expected rate of return has served public employers well for a long time. For the 25-year period ended December 31, NASRA calculates public pension funds returned 8.8 percent on their investments, and for 20 years, 8.7 percent. Over the past 10 years, the figure was 5%, but that reflects the impact of two severe recessions and a very slow economic recovery in between. Some calculations sow a return of about 8 percent holding good for period as long as 80 years.
“It was a crappy decade,” Brainard allows, but “public employers are going concerns. They’re not going to go out of business. So they set their assumptions using very long-term horizons – 20 to 30 years.”
Second, pension liabilities are not the only risks public employers have to be concerned about. “You make tradeoffs,” Brainard says. “They could clamp down on investment risk, putting money into the pension plan until it goes to nothing. But then they run the risk of not having enough money to pay for other spending priorities, or not having enough to pay wages and salaries to qualified workers.”
“We don’t live in a risk-free world,” Brainard says, and probably wouldn’t want to. He analogizes the ideal of risk-free pension funding to a law that everyone drive five miles per hour or a zero-emissions requirement for factories. The health hazards of driving and working or living near an industrial site would drop drastically, but so would economic production.
Requiring public pensions to value their liabilities using a risk-free rate of return isn’t to force them to change their investment policy, Biggs responds, although he questions whether some of their riskier investment are appropriate. Instead, the point he and other critics make is that liability and asset management are separate matters. State and local governments may not be going out of business, but the majority of benefits they owe to current and future retirees will be paid in the next 15 years, Biggs calculates. That represents a significant risk to taxpayers, who should know what they are facing if public pension managers’ investment expectations don’t play out.
The problem, Brainard says, is that mandating a risk-free discount rate “invites selective use by people with agendas.” On the one hand, conservative lawmakers who believe giant pension funds give too much power to public employees and who distrust pension officials who cast themselves as critics of corporate governance would have a new and urgent-sounding argument against the system. “Plan sponsors will just choose to shut them down,” says Brainard. “It’s a scare tactic.”
On the other hand, jacking up pension contributions could result in overfunding – effectively, overcharging one generation of taxpayers and undercharging another. Meanwhile, the need to meet higher liability targets could encourage plans to place assets in excessively risky investments – just the outcome the critics claim they want to avoid. Many state and local governments, the CBPP notes, skipped contributions and made major improvements to employee pensions in the early part of the last decade, when they were flush with cash and expected the high investment performance of the late 1990s to continue – again, the opposite of what advocates of the risk-free rate aim to achieve.
Brainard also objects to suggestions that public pension portfolios are riskier than they should be. While they have upped their commitment to alternative investments like hedge funds in recent years, sometimes to 8% to 10% of the portfolio, that is nowhere near the level that endowments and foundations favor – sometimes as high as 30%. In general, he says, public pensions simply are following modern portfolio theory, which holds that diversifying into a range of assets, some further out the risk curve, actually makes a long-term portfolio less risky overall.
A further problem, Brainard adds, is that the requirement in Nunes’ bill amounts to “an unfunded mandate” on state and local governments. For all of them to make an additional liability calculation using a new discount rate would cost “tens of millions of dollars,” he says – an expense they could not sidestep without losing their status as tax-exempt bond issuers.
State and local governments face real pension funding problems in the near future, even applying the traditional discount rate to liabilities. As a group, they were 100 percent funded until the recession of 2000-01 and then needed seven years to build themselves back to the 85 percent-funded level. The 2008 crisis hit them hard, because on average, 60 percent of public pension income comes from investments. Last year, their collective funding level stood at 77 percent, and this is expected to decline to 73 percent by 2013, according to a study by the U.S. Government Accountability Office.
That is a serious problem, but it does not “amount to a crisis,” the CBPP concludes. Paul Zorn, governmental research director at actuarial consultant Gabriel Roeder Smith, points out that state and local governments have generally restrained benefit increases over the past five years as they have rebuilt their investment portfolios, and are unlikely to reverse course. According to the CBPP, more than 20 states have made changes to reduce pension costs in recent years, despite news reports about some retirees receiving spectacularly generous benefits. That includes raising length-of-service and age requirements for receiving pensions and ratcheting down the formulas use to calculate pensions year by year.
While public employees do have the advantage of legal guarantees for their pensions – a key argument in favor of the risk-free discount rate – in reality they are subject to political pressure during bad economic stretches to give back some of their previous gains. That, says Brainard, suggests that there is less need for Congress to step in and impose a more severe standard for solvency than the traditional one. Public funds in 2008 – the most recent year reported and a major trough for the stock market - took in $118 billion in employer and employee contributions and lost $39.3 billion on their investments while paying out $175 billion, the Census Bureau calculates. Nevertheless, their total assets are sufficient to cover the next 12 to 14 years of benefit payments even in the unlikely event they received no further revenue at all.
Initiatives like the Nunes bill are unlikely to go away, however. One strike public pensions have against them is their sheer size and range.
As investors, they are present virtually everywhere in the economy today. Reporting requirements make their financial status transparent and easy to pick apart, even if some disagree with their methodology. And while their diversified portfolios may may accurately project their long-run average portfolio returns, they also expose the plans, more so than in the past, to wild fluctuations over shorter periods, Biggs points out. That makes situations like the one that took hold in 2008 – “perfect storms” in which investment returns collapse at the same time that tax receipts slump in the face of high unemployment and economic recession – more likely, and more damaging.
One official body considering these issues is the Government Accounting Standards Board, which is now reviewing Standards 25 and 27, governing actuarial calculations for public pensions. The project is expected to be completed by 2013. Until then, the risks associated with public pensions are likely to receive plenty of scrutiny.
