Trust Fund Truths

November 8, 1999

by Andrew Biggs

Andrew Biggs is a Social Security Analyst for the Cato Institute's Project on Social Security Privatization.

Confusion abounds regarding the Social Security Trust Fund. Congressional Republicans have initiated an ad campaign claiming that Democrats would "raid" the Social Security Trust Fund to fund general government expenditures. Liberals such as the Center on Budget and Policy Priorities claim that spending surplus payroll taxes "would not result in any raid on the Social Security trust fund or have an adverse effect on the Social Security benefits that future beneficiaries will receive." Who, if anyone, is right?

To start, let's get the Trust Fund straight, once and for all. As the President's own budget acknowledges, the IOUs in the Trust Fund - special issue government bonds, really - "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 ... 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."

So the Trust Fund cannot - repeat cannot - delay the need for tax hikes, benefit cuts or an increased retirement age. We will have to find additional resources to repay the Fund's bonds just as surely as we would need those resources to pay benefits directly. Given this, does it make "no difference whether the surplus is used to finance other programs or to pay down debt," as the Washington Post recently declared? Hardly.

The difference is between the obligation to pay benefits and the ability to pay them. Like borrowing from a credit card, the government's use of surplus payroll taxes creates an obligation to repay. And like a credit card statement, the balance of the Trust Fund reflects this obligation, currently at $653 billion and slated to rise to over $2.3 billion. Between 2014 and 2034, that's $115 billion to repay every year (in today's dollars). And as anyone with a credit card balance knows, what really matters is where the money is going to come from to pay it.

The real issue is not the balance of the Trust Fund, but how that balance can be paid off and how current budget surpluses might be used to do it. The liberal plan for the money - spend it - does next to nothing in this regard.

By contrast, using current surpluses to repay public debt would reduce future interest payments, freeing up resources that might be used for Social Security. But "might" is the operative word. Social Security would have to compete with other programs for these interest savings. Who's to say how future Congresses might use this money? In any event, savings would be relatively small since only low interest public debt would be retired.

By far the best way to use today's budget surpluses to cover tomorrow's Social Security benefits is to establish personal retirement accounts for workers. These accounts, invested in higher-rate stocks and corporate bonds, would produce greater savings than retiring public debt. Unlike the government bonds in the Trust Fund, these would be real assets that would reduce the need to raise taxes, increase the retirement age, or cut benefits. And since the accounts would be held by workers, they could not be siphoned off to pay for other programs, as today's surplus taxes are and as tomorrow's interest savings could be.

In short, the obligation to pay future Social Security benefits is like the obligation to pay your credit card bill: worthless unless you've got the money to back it up. There are various ways to save today's surplus to back up tomorrow's obligations to Social Security. Retiring public debt is one way, establishing personal accounts is even better. But spending the money on today's political priorities - whether we call it "raiding the Trust Fund" or issuing IOUs - most certainly is not. "It's your money, your choice, your future."

2001 Index | 2000 Index | 1999 Index | 1998 Index