Tuesday, October 05, 2004

That Mythical, Magical Social Security Trust Fund

You know the trust fund that's supposed to keep Social Security solvent until 2042, even though benefit payments will begin to exceed receipts in 2018? First, it's phony, as explained by Olivia S. Mitchell and Thomas R. Saving in a July 31, 2001, Washington Post op-ed piece:
...When Social Security ran annual surpluses in the past, it enabled other parts of government to spend more. The trust fund measures how much the government has borrowed from Social Security over the years, just as your credit card balance indicates how much you have borrowed. The only way to get the money to pay off your credit balance is to earn more, spend less or take out a loan. Likewise, the only way for the government to redeem trust fund IOUs is to raise taxes, cut spending or borrow.

Politicians and independent analysts have recognized the past inability of government to "lock-box" the trust fund. In 1989, for example, the assistant comptroller of the General Accounting Office said, "We shouldn't kid the American people into thinking extra savings is going on." It wasn't until 1999 that the non-Social Security portion of the government budget finally reached balance. Only then was the Social Security surplus actually used to retire government debt.

We are surprised that this perspective on the trust fund is controversial. The commission's interim report quotes credible sources -- the Congressional Budget Office, the General Accounting Office and the Congressional Research Service -- supporting the view that the trust fund is an asset to Social Security but a liability to the rest of the government. The Clinton administration's fiscal year 2000 budget indicated a similar perspective:

"These [trust fund] balances are available to finance future benefit payments and other Trust Fund expenditures -- but only in a bookkeeping sense. . . . They do not consist of real economic assets that can be drawn down in the future to fund benefits. Instead, they are claims on the Treasury that, when redeemed, will have to be financed by raising taxes, borrowing from the public, or reducing benefits or other expenditures. The existence of large Trust Fund balances, therefore, does not, by itself, have any impact on the Government's ability to pay benefits."

Does this mean the government will not make good on those trust fund claims? Of course not, and the commission never suggested this. In fact, one principle guiding the commission's work is that current and near-retirees must have their benefits preserved. Still, the nation has only three ways to redeem trust fund bonds: raising taxes, cutting spending or increasing government borrowing. If there is some alternative source of funds, no one has yet suggested it. And the annual amounts to be repaid are large and rising over time: $93 billion in 2020, $194 billion in 2025, and $271 billion in 2030 (in today's dollars).

Some have suggested we could solve our Social Security problems simply by legislating a higher interest rate on trust fund bonds. This has superficial appeal, since it would be a simple bookkeeping matter for the Treasury. It would make the trust fund look bigger and extend the date when the trust fund is exhausted -- on paper. But no increase in the amount we owe ourselves creates new saving. The same people will be retiring, expecting the same benefits and posing the same need for revenues irrespective of the size of the trust fund.

One way or another, promises of Social Security benefits made under the current system must be financed by taxpayers. These costs are slated to grow from 10.5 percent of taxable wages today to 13.3 percent in 2016, 17.8 percent in 2038 and 19.3 percent in 2075. That is why reform now is crucial, when time is still on our side. With each passing year of non-action, the day of reckoning draws nearer, leaving us with fewer options.

Olivia S. Mitchell, a Democrat, is a professor of insurance and risk management at the University of Pennsylvania's Wharton School. Thomas R. Saving, a Republican, is a professor of economics at Texas A&M and a member of the Social Security Board of Trustees.
So much for the mythical trust fund. Now, about those magical interest rates. According to the Social Security Administration, bonds issued to the trust funds in 2003 had an interest rate of 3.5 percent, compared with an interest rate of 5.25 percent for bonds issued in 2002. That's not too bad, considering that the average rate on 20-year Treasury bonds (the Treasury's longest maturity) was 4.96 percent in 2003 and 5.43 percent in 2002. But it's not nearly as good as, say, AAA and BAA rated corporate bonds and conventional mortgages, which had these average yields: AAA 2003, 5.66 percent; AAA 2002, 6.49 percent; BAA 2003, 6.76 percent; BAA 2002, 7.80 percent; conventional mortgage 2003, 6.54 percent; conventional mortgage 2002, 5.82 percent. In other words, high quality corporate bonds and conventional mortgages carry significantly higher interest rates than government bonds (because of the perception that corporate bonds and conventional mortgages are riskier than government debt).

These facts point to a way to ease the transition to private Social Security accounts. First, convert the mythical trust fund to a real one by creating an independent agency to invest the trust fund in high-quality corporate bonds and conventional mortgages. The agency would gradually sell off the trust fund's portfolio of highly marketable government bonds and replace them with high-quality corporate bonds and conventional mortgages. In the end, the trust fund would comprise real assets, and those real assets would earn real income, at rates higher than those magically credited to the mythical trust fund upon which the future of Social Security now rests uneasily.

The additional revenue earned by the new trust fund wouldn't wipe out the pending deficit in Social Security, nor would it prevent the eventual depletion of the trust fund. But the depletion of the trust fund could be held off long enough to enable a graceful transition to private social security accounts. The slower that transition, the less painful would be the temporary -- but necessary -- combination of tax increases and/or benefit reductions.

In the end, we'll be better off because each worker will be investing for his or her own retirement -- unlike the present system, which places an increasingly heavy burden on future generations. But we need more time to get there, and "privatizing" the trust fund is a good way to buy that time.