Rudolph Penner, a senior fellow at the Urban Institute, is an expert in federal budgets and Social Security. In government, he has been director of the Congressional Budget Office from 1983 to 1987, assistant director for economic policy at the Office of Management and Budget, deputy assistant secretary for economic affairs at the Department of Housing and Urban Development, and senior staff economist at the Council of Economic Advisors. His most recent policy brief, with C. Eugene Steuerle, is A Radical Proposal for Escaping the Budget Vise.

Five Questions Archives

Dozens of American proposals for individual accounts are circulating and about 20 countries now have some sort of individual account as a central part of their Social Security reforms. But individual accounts have very different characteristics and operate very differently in different countries. Most U.S. proposals are not very radical since they would leave much of the traditional Social Security system intact. Recent proposals, mainly from Republicans, would allow individuals to voluntarily divert some of their current payroll taxes into individual accounts. In return, these individuals would forfeit some of their traditional benefits.

Diverting payroll taxes deprives the Social Security system of resources and the AARP and other advocates of the traditional system oppose it. Some proposals reduce this effect by raising the payroll tax base. Other taxes could be raised as well, but at a huge cost to the reform's popularity. Rerouting payroll taxes also raises the budget deficit until the benefit reductions kick in.

Many design issues must be resolved when creating individual accounts. The president's proposed design has some nailed down. Administrative costs would be kept low by running the accounts like the Federal Employees Thrift Savings Plan. Investment choices would be limited to a few broad bond and stock funds—a way to limit risk. Anyone reluctant to make investment choices would be put into a life cycle fund by default. Such a fund allows for taking significant amounts of risk when the investor is young, but reduces risk as a person approaches retirement.

The president cures part of the long-run solvency problem facing Social Security by altering the indexing of initial benefits. Currently, initial benefits are indexed to wages. The president would index initial benefits to prices for roughly the top half of the earnings distribution. Prices typically rise less fast than wages.

The president's overall plan has two major disadvantages. First, like many Republican plans, it greatly increases the short-term budget deficit because it diverts part of the payroll tax to the individual accounts and does not raise any taxes. Second, individuals have to pay back the diverted amounts at a real rate of interest of 3 percent. Individuals would have to take significant risks to earn as much as 3 percent.

A recent Republican plan receiving a lot of publicity would finance individual accounts out of the current Social Security surplus. The accounts would be small and the program would not last long because the surplus won't last long. And the plan does nothing for Social Security solvency. Clearly, the plan is not meant to be taken literally, but is a stalking horse for individual accounts.

2. Can individual savings accounts work in Social Security reform?

Individual accounts can complement Social Security reform by providing individuals with the opportunity, if they're willing to take a bit of risk, to replace some or all of the traditional benefits they lose under proposals to make the system more solvent.

It's no accident that individual accounts have played an important role in 20 countries with significantly reformed systems. Chile pioneered that route and replaced almost their whole system with individual accounts. No one is proposing that here. Most American proposals retain much of the traditional system. But the Chilean experience has gone quite well and given a number of other countries the confidence to build individual accounts into reform.

The reforms are still young in most countries, so it's hard to say how well they'll be doing 30 or 40 years from now. The reform that I like best took place in Sweden, where they radically transformed their traditional system and added individual accounts with contributions of 2½ percent of wages. The Swedish plan's most attractive feature is that it adjusts benefits automatically to increases in life expectancy and to other economic and demographic surprises.

3. What are the potential risks to workers from the individual accounts?

Individuals can take as much risk as they like, but must expect a lower rate of return if they choose lower-risk investments. Investors shouldn't exaggerate the risks associated with equity investments. Assuming historical rates of return, someone contributing over a lifetime to a broad-based equity fund has a very high probability of doing extremely well even if they foolishly remain heavily invested in equities until the eve of their retirement and must sell immediately after a market slump like that following the recent stock market bubble. Such people might complain because they got less than they thought they would a few years earlier. But they can't complain about their average rate of return over a lifetime compared to that in a relatively riskless security, such as an inflation-indexed government bond.

It is unfortunate that opponents of Social Security reform so often attack the concept of individual accounts when their real objection is to proposals that divert payroll taxes out of the Social Security trust fund. You can have individual accounts that don't have this effect. In fact, that's what President Clinton recommended.

Considerable risks already exist in the traditional Social Security system—especially the concern that benefit growth will have to be cut to improve the system's solvency.

Even if there weren't solvency problems, you never know for sure what level of benefits you're going to get from Social Security. The system is indexed to wages. So what you get depends on how wages increase over your working life. The wage increases in the few years before your benefits are computed at age 60 are especially important and they are hard to forecast.

4. What are the benefits of individual saving accounts?

I think an appealing characteristic of the President's proposal is that the individualized accounts are run by the government with administrative costs considerably lower than those in your traditional 401 (k) or IRA. And that's very important because large administrative costs can significantly dent one's ultimate retirement income.

Without individual accounts, Social Security reform is all about pain -- the pain of higher taxes or reduced benefit growth. Individual accounts add something positive to reform. They can replace some or all of the benefit reductions necessary to achieve solvency.

Properly designed individual accounts can increase national savings. The problem with our Social Security system—which is pay-as-you-go-is that there are no real assets underlying it. The return you get as an individual is largely determined by demographics—how many people are paying in versus how many people are receiving benefits. If the system invests in real capital assets, the rate of return still reflects demographics but depends more on the rate of return to real capital.

The President's proposal doesn't increase national saving initially because investments in individual accounts are financed by increasing the budget deficit. This is a major disadvantage of the President's plan that can be corrected by financing the individual accounts with tax increases or by having the individual put money in individual accounts while continuing to pay today's payroll tax.

Individual accounts also earn some political popularity because they can be left to one's heirs.

5. What might secure Social Security benefits for future generations?

I suppose the first question is whether you really want to for everyone, especially for the affluent. Many wealthy couples are getting much more than a welfare family does.

If the nation wants to repair the current system, there will be some pain—either higher taxes or a lower-than-promised rate of benefit growth. Benefit growth can be reduced or payroll taxes raised progressively. The President favors the former.

But any attempt to reform the system progressively will mean swallowing important inequities. Social Security simply isn't very effective at redistributing income. Benefits are based largely on an individual's earnings. Effective redistribution must look at earnings plus investment income and at the family rather than the individual. Many individuals have low lifetime earnings, but have affluent spouses. They look poor to the Social Security system and would receive a windfall from any progressive reform.

It is theoretically possible to design a reform that restores solvency while having the return to individual accounts make up for lost traditional benefits for almost all participants in almost all age cohorts. But the "almost all" is an important caveat. This happy result can only be achieved by individuals willing to take somewhat higher risks.