Penner, Rudolph G. - Urban Institute

THE URBAN INSTITUTE
2100 M STREET, N.W.
WASHINGTON D.C. 20037
RUDOLPH G. PENNER
Senior Fellow
Arjay and Frances Miller Chair
Direct Dial: (202) 261-5212
Fax: (202) 728-0232

COMMENTS ON SOCIAL SECURITY REFORM FOR THE WHITE HOUSE CONFERENCE ON SOCIAL SECURITY [1]

It is urgent that we reform Social Security and Medicare. Long-term projections by the General Accounting Office and the Congressional Budget Office suggest the possibility of a declining economy if literally nothing is done. Although Social Security poses a lesser economic problem than Medicare, it may be easier to reform, both technically and politically.
Social Security may be the most popular government program ever invented. It has greatly reduced poverty among the elderly and it has paid a very high rate of return on the payroll tax payments of past retirees.
Unfortunately, the glory days of Social Security are over. As a pay-as- you-go system, it can only pay a high rate of return to the extent that each successive cohort of workers pay more into the system than previous cohorts. In the past, each successive cohort did pay considerably more, because the size of the labor force was growing, real wages were increasing, and the average payroll tax rate was continually increased. Until recently, the most important contributor to the high rate of return was a continually increasing tax burden, as each successive generation of retirees enjoyed benefits financed by taxes on workers that were much higher than the retirees had faced during their own working lives.
It is implausible to think that we can continually increase future payroll tax burdens and there will be little growth in the labor force in the period 2010 to 2030. For both reasons, rates of return to future retirees will be very low, and without reform, the system is likely to lose political support in the very long run.
It is necessary to move from a pay-as-you-go to a funded retirement system to avoid this outcome. I believe that the safest way to accomplish this goal is through a system of mandated individual accounts that would complement a slimmed-down Social Security system.
Some worry that individual accounts will involve transaction costs that are too high and thus will greatly reduce the rates of return available to individuals. Even under the worst assumptions, the rates of return on individual accounts are likely to be higher for most than the rates provided by traditional Social Security. But if high transaction costs are a concern, the National Commission on Retirement Policy (NCRP), of which I was a part, has shown that transaction costs can be reduced to insignificant levels by having the information and collection process administered by the Social Security Administration and by limiting the number of investment options for the individual and the number of trades allowed each year.
Others worry that individual accounts impose too high a risk on investors. Data on IRAs and 401ks suggest that investors are quite modest in their risk taking, especially as they near retirement. If, however, risk is a major concern, plans by Senator Gramm and Martin Feldstein show that much of the risk can be left with the government by having traditional Social Security benefits make up for a part of any earnings disappointments from individual accounts. Alternatively, a minimum benefit can be guaranteed that is higher than the current minimum from Social Security. Although I prefer not to have the government left with a large, implied, contingent liability, I can see moving in this direction as a compromise. In any case, an approach that shares risk is much superior to plans that would have the trust fund invest in equities, in which case, the government bears the entire risk of market fluctuations.
There are many available options for reducing the future growth of Social Security benefits. One can do it through explicit reforms, such as increasing normal and early retirement ages, or one can, as in the Gramm/Feldstein proposal, reduce the Social Security benefit by an amount linked to the investor's success with his or her individual account. The latter guarantees that no one has to do worse than under the current Social Security system. That may give the plan a major political advantage, but it misses the opportunity to make the structure of Social Security more equitable and more conducive to encouraging later retirement.
One of the most prominent competitors with individual accounts is the notion that the trust fund should invest in equities. I find this approach highly troubling. It is scary to think of the political temptations posed by the government owning trillions of dollars of corporate equity. Proponents say that they will avoid this danger by organizing the management of this fund like the Federal Reserve System. That is not reassuring, since the Federal Reserve has been accused of bowing to political pressure in the past, most notably when Chairman Arthur Bums was rightly or wrongly accused of pumping up the money supply on behalf of Richard Nixon's re-election.
Proponents of equity investments by the trust fund also suggest that government can more easily spread risk among the generations. If so, this is also a characteristic of the Gramm/Feldstein approach which leaves much of the risk with the government. But there is absolutely no reason to believe that government would spread risk in an equitable manner. More likely, the strongest political incentive would be to distribute the benefits of positive market surprises immediately by raising benefits or cutting taxes while delaying the pain of negative surprises until future generations.
If we move to individual accounts, it is important to ask how they will be financed. Will the mandate be imposed on top of the current payroll tax system or will payroll or other taxes be cut to offset the pain imposed by mandates? The latter approach could be characterized as using the surplus to save Social Security. I would like to keep a significant budget surplus to supplement the deplorably low level of current private saving. However, this is likely to be an unrealistic goal given the long list of demands for tax cuts and spending increases lurking in the background. Using the surplus to facilitate individual accounts represents a much better use of the money than virtually all the other items on the list.

1 The views expressed in this piece are those of the author and do not necessarily represent the views of the trustees and employees of the Urban Institute.