Is it Time to Reform Social Security?

By Edward M. Gramlich. Ann Arbor, MI: University of Michigan Press, 1998. Pp. xi, 103. $24.95.

This brief book will be a useful guide to the present state of the national debate over Social Security. It is a clearly written accessible account of the problems that face the system and the proposed solutions that came out of the President's Advisory Council last year. As the chairman of this Council, Gramlich is a person well situated to write this book. His political views on this issue also place him very much in the center of the debate. He occupies a middle position between those on the right, who support more extreme privatization proposals and people to his left who would like to leave the current system largely intact.

Gramlich's preferred solution is a combination of benefit cuts and small mandated individual accounts, which would be an add-on to the existing program. The benefit cuts would be large for middle income workers but would still preserve the progressive payback structure of the current system. The mandated accounts would be 1.6 percentage points of payroll, less than one-third the size advocated by privatizers. These accounts would be centrally managed through the government, rather than put directly into the hands of the financial industry as preferred by most advocates of privatization. Gramlich's plan, and other similar proposals, have garnered considerable attention in recent months. Many in Washington believe that the Clinton Administration will ultimately end up supporting some version of this proposal.

While this book serves as a solid rendition of the conventional wisdom, it also suffers from all the shortcomings or shortsightedness that characterize the current debate on Social Security. Of course it is not fair to blame Gramlich alone for the narrowness of this debate, but it is at least worth noting some of the key issues that the book sidesteps.

Going in reverse order of importance, Gramlich presents the standard argument on projected trends in entitlement standard as part of the rationale for restructuring Social Security. The basic story is that entitlement spending for the elderly is projected to soar over the next few decades, therefore we should look to contain Social Security spending. This argument is at least misleading, because the real culprit in the story is Medicare spending. Furthermore, the projected increases in Medicare spending are driven primarily by higher per capita medical costs, not demographics. The Health Care Financing Administration projects that large increases in health care costs will occur in both the private and public sector. If health care costs actually do rise at the projected rates it will be disastrous for the economy, regardless of what happens to publicly provided health care. While the projected explosion in health care costs may make a compelling case for comprehensive health care reform, it does not provide much of a basis for cutting back Social Security benefits.

The second issue involves the state of the labor market for older workers. Like many of the restructuring plans currently being floated, Gramlich's plan would accelerate the increase in the normal retirement age, putting it at 67 by 2011 and rising in step with life expectancies to 70 by 2070. The argument is that people are living longer and are healthier, so it is reasonable to expect them to work until later in their lives. While this is a reasonable argument, it ignores the immediate conditions faced by older workers. There is considerable anecdotal evidence and some statistical analysis that indicates that older workers are likely to face considerable difficulties in holding on to their jobs (Aaronson and Sullivan 1998). These older workers are the ones who have been targeted for downsizing because they tend to be at the top of the payscale. In the longer term future, it may be entirely reasonable to expect people to work later into their lives. But even though the projections are made for a 75-year period, the current debate will not actually be setting the nation's retirement policy in 2040-2050, and the issue will be revisited many times by then. However, the debate will be determining the nation's retirement policy for at least the near-term future. It is incredible that discussions of raising the normal retirement age take place in the absence of a solid analysis of the labor market for older workers.

The third issue that has not been seriously examined in this debate is the returns that can be expected on equities over the 75-year planning horizon. The President's Advisory Council worked with the assumption that the real return will be 7% annually based on the historical rate of return on equities. This figure has been widely used in the debate and finds its way into Gramlich's calculations of the income that can be generated by the individual savings accounts in his plan. Projecting the future based on the past is ordinarily a reasonable practice, except when we assume that the future will not be like the past in very fundamental ways. This is the case here. In the past 75 years the economy grew at an average rate of approximately 3.0% a year. Over the next 70 years it is projected to grow at the rate of less than 1.5% annually. The slower growth is the reason there is a projected shortfall in the trust fund. This slower projected growth rate (and the high current price-to-earnings ratios) makes it virtually impossible that a 7.0% real rate of return can be sustained. A 4.0% rate of return would be more in line with the Trustees' growth projections. This gap makes an enormous difference in assessing the merits of privatization proposals.

The last, and most important, unaddressed issue is the implications of a substantial overstatement of inflation by the consumer price index (CPI). Gramlich notes the debate around the Boskin Commission's report claiming that the CPI was overstated by 1.1%. He remains agnostic on the issue but argues persuasively that measuring the inflation rate should be left to BLS, not ad hoc congressional commissions. However, he does not bother to note the implications of the Boskin Commission's claim being true. Specifically, that real wages and income have been rising by 1.1% a year more rapidly than our current data indicate. Because much of the debate on Social Security is framed in terms of generational equity, this would be an enormously important finding. It would mean, for example, that real wages will have risen by 88.5% as of 2030, instead of the 33.2% projected by the Trustees. Of course, it also means that today's beneficiaries were much poorer in the recent past than is indicated by the uncorrected CPI.

It is remarkable that the debate over Social Security has taken place in a context where economists have no idea how wealthy future generations of workers are expected to be, nor how poor current beneficiaries were during their working lives. It is as though the nation were debating a redistributive antipoverty program without knowing the incomes of either the poor people who are expected to benefit or the workers who will pay the taxes. The issues involved in the CPI debate are very complex, but it is virtually impossible to make any serious judgments about generational equity without knowing the true rates of real wage and income growth.

Again, it is not fair to hang all the shortcomings of the Social Security debate on Gramlich's shoulders. On the other hand, as a central figure in this debate, he does bear a portion of the responsibility.

Reference

Aaronson, D., and D. G. Sullivan. 1998. The decline of job security in the 1990s: Displacement, anxiety, and their effect on wage growth. Economic Perspectives, First Quarter, Federal Reserve Bank of Chicago, pp. 17-43.

Dean Baker Economic Policy Institute