Whither Social Security? Transition it, then Retire It

by William A. Niskanen

William A. Niskanen is chairman of the Cato Institute and a former economic advisor to President Reagan.

Social Security, the largest and most popular federal program, is now 63 years old and should soon be retired. For Social Security is an intergenerational Ponzi scheme, a tax on those now working to pay for those now retired, and the bills are now coming due.

The first Social Security tax was 2 percent on earnings up to $3,000 a year. The current Social Security tax in 12.4 percent on earnings up to $72,600 a year -- of which 10.7 percentage points is for retirement income and 1.7 percentage points is for disability insurance. In addition, workers pay a 2.9 percent tax on all earnings to finance Part A of Medicare.

The first Social Security recipient was a hardy lady named Ida Mae Fuller who paid a total of $44 in payroll tax and, by living to the age of 100, received a total of $21,000 in benefits; for most of those who have since retired, Social Security has also been a bargain. For workers born in the past forty years, however, the average real (inflation-adjusted) rate of return on their social security taxes will be about 1.4 percent, lower than the current real yield on Treasury bills. And for different reasons, blacks, the second worker in two-worker households, and high earning workers will receive an even lower return.

And that is if Social Security pays all of the benefits that it has promised. The current payroll tax, however, will only finance about 75 percent of the promised benefits by the year 2030. The primary reason for this ominous portent is that the number of workers per beneficiary is expected to decline from the current ratio of 3.3 to a ratio less than 2.0 by that year with the retirement of the baby boomers.

One way or another, a business-as-usual approach to Social Security will require some combination of a reduction in benefits or other federal spending, an increase in the payroll tax or some other federal tax, or an even larger federal debt burden on our grandchildren. All because federal politicians chose to finance Social Security on a pay-as-you-go basis more than 60 years ago. The manager of a private or state pension fund would go to jail for the same behavior.

Tinkering with Social Security is no longer enough. Reducing future benefits or increasing the payroll tax would lead to a negative real return to today's young workers. The only type of pension program that is demographically stable in the long run is a prefunded plan in which each generation pays for their own retirement. This point is now generally recognized, even in that island of unreality called Washington. The challenge is to design a transition plan from the current pay-as-you-go system to a prefunded system while maintaining the benefits of those who are already retired or nearing retirement. And that is now the primary focus of the Social Security debate.

This debate, fortunately, has been pulled forward more than any of us anticipated, largely by President Clinton's commitment to submit his own proposal on Social Security this winter. President Clinton, to his credit, was also correct to urge Congress not to commit the pending federal budget surplus to other ends until this issue is sorted out. Financing the transition to a stable prefunded private retirement program is surely more important than the mishmash of spending increases and tax cuts recently considered by Congress. The Social Security plan that Clinton outlined in his recent State of the Union Address, however, is unfortunately not a serious proposal:

1. About 50 percent of the projected unified budget surplus over the next 15 years would be committed to reducing the federal debt owed to the public. This may have better effects on the economy than many other proposed uses of the surplus but would have no direct effect on Social Security. By sleight-of-hand, Clinton would increase the federal debt owed to the Social Security Trust Fund by twice this amount, claiming that this would extend the solvency of this mythical trust fund to the year 2055, but this left-handed bookkeeping would have no substantive effect on anything. The issue is not double counting but phony accounting; two times nothing is still nothing.
   
   
2. About 12 percent of the projected surplus would finance a small new complicated IRA. The federal government would put a small fixed amount into all these Universal Savings Accounts and would then augment individual deposits to these accounts on an income-tested basis. This measure would also have no effect on Social Security.
   
   
3. Another 12 percent of the projected surplus would be invested in private securities to be held by the Social Security Trust Fund. Clinton proposed a system designed to protect these investments from political pressure, but the record of similar systems is not encouraging. Many state and local pension funds and some national provident funds have been manipulated for political objectives, reducing the rate of return to these funds. From early in the Clinton administration, the Department of Labor has tried to pressure private pension fund to invest in politically targeted investments. Although this component of the Clinton plan has been the focus of most early criticism, this is the only component that would have any effect, albeit minuscule, on the future funding of Social Security.

This plan, apparently assembled by some wordsmith with a stapler, is not a serious plan, and there is no reason for Congress to give it serious attention. My guess is that this plan, like the 1993 health care proposal, will never reach a floor vote.

In that case, where should Congress go from here? The worst response to Clinton's failure to submit a serious proposal would be to act as if Social Security is not a serious issue; this would trigger a frenzy of down-payment spending and junk tax cuts that would fritter away the budget surplus. A better alternative would be to do nothing; approve a tight budget, go home, and let the surplus reduce the federal debt until there is a sufficient consensus to resolve the Social Security issue. The best alternative is to recognize that the projected surplus is a rare opportunity to resolve the major long-term fiscal issues. And the most important fiscal challenge is to transform Social Security from an unsustainable pay-as-you-go government pension into a sustainable system of prefunded private retirement accounts.

The Cato Institute, fortunately, has been studying Social Security for over 20 years, a program that was regarded as the third rail of American politics for most of that period. The general provisions of the several transition plans that make the most sense to us are the following:

1. The benefits of those who are currently retired would be fully guaranteed. Anything less would be morally wrong and politically stupid.
   
   
2. All current workers would have the choice to stay in Social Security or to move to a new system of private retirement accounts. For those who stay in Social Security, the age for full benefits could be slowly increased, maybe by 2 months a year, up to age 70.
   
   
3. Those who choose the private retirement account would receive a recognition bond equal to the cumulative value, including interest, on the Social Security taxes they have already paid. In addition, they would put some part of their payroll tax in one or more government-authorized portfolios of private stocks and bonds.
   
   The sum of the recognition bonds and the accumulating private portfolios would make most workers better off, primarily because of the substantially higher return on private equities. Over the past 70 years, for example, the average real return on U.S. corporate equities was 7.7 percent, and there was no 30 year period in which the real return was less than 4 percent. Since the average real return on Social Security is now only 1.4 percent, most workers would be better off even if only half of the Social Security tax were diverted to private accounts. Moreover, for those who die prematurely, the recognition bonds and accumulated private portfolios would be part of the person's estate, compared to the complete loss of these accumulated payments under Social Security.
   
   
4. All new workers would be in the private retirement system.
   
   
5. The transition cost would be financed by some combination of the following:
   1. the projected surplus on the unified budget,
   2. the part of the Social Security tax that is retained by the government,
   3. the higher tax revenues from the increased private investment and earnings, and
   4. some increase in the federal debt that would be repaid when the benefits payments to the remaining Social Security recipients decline.
   
   
6. The federal government would retain five important roles:
   1. set the payroll tax rate and the share that may be diverted to private accounts,
   2. maintain the benefits of those who are currently retired or choose to stay in the Social Security system,
   3. maintain a safety net financed from general revenues for those who, for whatever reason, do not accumulate a sufficient private account for a minimally adequate retirement annuity,
   4. authorize the private funds into which some part of the payroll tax may be diverted. At such time that a person has accumulated assets of a specified level, any additional retirement saving may be put into any investment, and
   5. regulate the rate at which an individual can draw down the principle of his or her retirement account or require that individual to buy an annuity of a specified amount.

That's about it. The primary political issues involve the share of the payroll tax that would be diverted to private accounts, whether the age for full retirement should be gradually increased, the level of the safety net, and how much the government should regulate the authorized private portfolios. Most of these problems, of course, would be larger if Congress fritters away the projected budget surplus on less important matters. We have a brief window of opportunity to sort out this issue, and the sooner the better.

This article originally appeared in The Colorado Springs Gazette, October 16, 1998.