Pension Bonds a Risky Cop-Out

CalPERS isn’t supposed to be in business to serve up free lunches, but you wouldn’t know it from the governor’s revised budget proposal. As things stand now, Sacramento plans to punt on next year’s payment to the giant state pension fund by floating $2.2 billion in taxable five-year bonds next September. The plan, if adopted, would spare legislators from having to cut more spending or raise taxes. But it comes with hidden costs, glossed over by slick salesmanship from Wall Street. The fear is that issuing “pension obligation bonds” could become a new habit.

Many cities and counties in California, however, have been using these bonds to top up underfunded pension plans since the mid-1990s. The city of Fresno and San Diego, Kern, San Luis Obispo and Contra Costa counties are among the latest issuers. They hope to cut their total pension costs by borrowing to invest in stocks and bonds.

“Wall Street predators zeroed in on the vulnerabilities in pension funding methods,” says Jeremy Gold, a pension actuary. “Just as a virus attacks a body, they exploited weaknesses in accounting and actuarial methods, as well as in state budgeting. Investment bankers have succeeded in making these financially unsound transactions look very attractive in order to extract great profits underwriting pension bonds.”

Start with a practical problem that hit Philadelphia hard. The financially strapped city borrowed $1.3 billion in 1999 to close the gap in its poorly funded pension plan. But now it faces the unpleasant choice of either paying an additional $21.3 million in pension costs this year or watching its unfunded liability escalate that much faster. That’s a big risk for Philadelphia taxpayers, and it showed up within a year or two after investment returns fell below those assumed by the pension plan’s actuary.

Another practical problem involves the perfectly understandable temptation to raise benefits after the proceeds from the sale of pension obligations bonds make a pension plan look healthier. San Diego County, now a serial pension-bond issuer, floated $430 million of these bonds in 1994. The stock market’s astonishingly good performance during the next five years or so made the pension plan look flush. And the county felt comfortable enough to boost benefits. Then, the bubble popped. “We are concerned for several reasons that pension obligation bonds can add to fiscal stress and contribute to rating downgrades,” says Parry Young, a director of state and local government ratings at Standard & Poor’s.

Pension bonds also are expensive. The first ones, issued in the early 1980s, were tax-exempt. So, a state or local government, armed with the benefit of tax-exempt rates, could borrow for less than the U.S. Treasury. Officials put the money from these early pension obligation bonds to work, investing them in U.S. Treasury bonds. It was a risk-free yield pickup. The pros call this a pure “tax arbitrage,” and it was valuable.

These bonds became quite popular until the Internal Revenue Service cracked down and took away the tax exemption. They didn’t come back into fashion until the mid-1990s, when the allure of a booming stock market revived their appeal.

Pension bonds wouldn’t exist without a fundamental mistake in the way actuaries -- sometimes described as number crunchers without the personality of an accountant -- report the value of promised benefits. They err when they assume that the value of a monthly retirement check is somehow related to the performance of the pension plan’s investments. Rest assured that California’s pension promises are rock-solid. Taxpayers, after all, stand behind them. But there’s risk galore in the portfolio.

A bird in the hand is worth two in the bush, to put it in terms of the cliche. With the California Public Employees Retirement System, the actuary assumes that it’s closer to three. It turns out that a risk-free pension promise worth $1 today will buy $3.75 in benefits 30 years from now, based on recent interest rates for the long-term Treasury bond. But that same payment of $3.75 would be worth only about 35 cents today if its risk was commensurate with the much higher rate of return of 8.25% assumed by CalPERS.

So, today’s pension obligation bonds miss the point. The idea is that the government issuing them can reduce its pension contributions if the investments outperform the assumed rate of return set by the actuary of the individual pension plan. But that’s an actuarial arbitrage, not a tax arbitrage, which made the first pension bonds so valuable.

There are winners and losers, of course. A plethora of laws, regulations and accounting standards require officials to systemically understate the true value of future pensions. These low estimates, in turn, determine annual contributions. Our state and local pension systems therefore transfer vast amounts of wealth from people in the future, principally taxpayers, to plan participants and others today.

Pension obligation bonds try to take the process one step further. They work like magic for public officials seeking to avoid difficult budget choices. They lighten the burden on current taxpayers. Beneficiaries of today’s spending programs can continue to enjoy government largess. The investment bankers become richer. And the actuaries look like heroes.

Many traditional defined-benefit plans, which pay a monthly pension based on earnings and length of service to a single employer, offered by private companies also have been struggling lately. Recent estimates suggest that these arrangements are underfunded to the tune of at least $300 billion.

General Motors announced plans last week to issue $13 billion in long-term corporate debt to help close a shortfall of about $20 billion in its U.S. pension plans. Here, a true tax arbitrage is possible, because interest payments on the debt will be tax-deductible.

So GM is on firmer ground. The late Fischer Black of the Massachusetts Institute of Technology and later of Wall Street, one of the greatest financial thinkers of the 20th century, worked out a tax arbitrage theory of corporate pension funding with Irwin Tepper. Black and Tepper explained that taxpaying companies are best off when they borrow to fund their pension promises, taking tax deductions on the interest payments and investing their pension assets in bonds -- the most heavily taxed of all securities. Pension plans, they reasoned, are not places where companies should take stock market risk.

A spokesman for GM reports that the company invests about half of its pension portfolio in stocks, with the remainder going to bonds and other investments like real estate. So, the GM deal deserves two cheers. One for borrowing and another for putting at least some of the money into bonds.

But what might be good for General Motors is almost certainly not going to be good for California.