Setting The Record Straight On Personal Accounts

Volume IV, Number 4 April 30, 1998

Proposals to transition in whole or in part from the current Social Security system to a funded system of personally owned accounts are winning growing support among economists, legislators, and the American public. This is hardly surprising given the huge potential benefits: Not just higher national savings, and hence ultimately higher wages for workers, but higher returns on contributions, and hence higher benefits for retirees.

Yes, the personal account approach raises legitimate design issues--especially, how to pay for the transition to a funded system. What this policy discussion does not need, however, is the kind of misinformation dispensed by Brookings economists Henry Aaron and Robert Reischauer in a recent essay in the Washington Post. Surprisingly, Aaron and Reischauer barely touch on the transition issue. Instead, using reckless assertions, they draw an utterly distorted picture of personal accounts that does nothing to help Americans think clearly about the future of Social Security.

* Distortion One: Without any change in its "basic structure," Social Security "can provide workers with higher returns than could any privatized alternative." The implication is that Social Security is a good deal--which is patently untrue.

Average-earning workers born in 1945 and retiring in 2010 will earn a real return on their FICA contributions of between 1.2 percent (if they are single males) and 2.5 percent (if they are in a two-earner marriage). Even this assumes that scheduled taxes are sufficient to pay future benefits. But they are not. Under current law, benefits would have to be cut to match revenues after Social Security goes bankrupt in 2032. As a result, rates of return for workers retiring in 2030 would fall by nearly an additional percentage point--meaning that for single males they would fall all the way to zero.

What about personal accounts? If workers invested entirely in equities, they could earn a return of roughly 7 percent--assuming the market matches its long-term historical average. If they invested in equities when young and shifted into bonds as retirement nears, they could earn a return of 4.5 percent with little market risk.

Social Security's dismal rate of return is acknowledged by practically every expert. The result of the program's pay-as-you-go structure, it condemns younger Americans to huge losses in potential lifetime wealth. Consider a worker whose lifetime real salary is $30,000 and who invests 10 percent of that throughout his or her career, roughly the contribution rate for Social Security retirement benefits. At a 7.0 percent rate of return, he or she would accumulate a real $857,000 by age 65. At a 4.5 percent rate, that amount would be $417,000. But at a zero percent rate, it would be only $135,000.

        Social Security (OASI) Real Rates of Return for
           Average-Earning Workers by Retirement Year

                        Single  Single  Two-Earner
                         Male   Female    Couple
                1975      5.0     6.5       7.2
                1995      1.8     2.9       3.5
                2010      1.2     2.1       2.5
                2030*     1.0     1.9       2.3
                2030**    0.1     1.0       1.4

        * Assumes current-law benefits are paid in full.
        ** Assumes benefits are cut to match revenues starting in 2032.

        Source: Eugene Steuerle, Urban Institute; and authors' calculations.
* Distortion Two: Aaron and Reischauer can't refute such numbers. Instead, they play bait and switch, which leads them to distortion two: Social Security, to the extent it is allowed to invest in private assets, could earn the same rate of return as personal accounts. In fact, they say, because administrative costs would be lower, the net rate of return would actually be higher.

Their logic is disingenuous. For one thing, there's the question of scale. Many personal account plans would ultimately fund most of Social Security. Aaron and Reischauer, however, would invest no more than a tiny fraction of total worker contributions in private assets--which is why they can assure us they would leave the "basic structure" of Social Security intact.

More important, the government investment approach (whatever the scale) doesn't address the central concern behind calls for personal ownership--namely, that voters don't trust government to save their tax contributions. Defenders of the status quo are starting to grasp that Social Security's pay-as-you-go structure is indefensible, and so want to climb on board the funding band wagon. Unfortunately, the historical track record shows that Congress can always find some way around any legislative vow that Social Security will engage in genuine savings. The public knows this, which is why most Americans under fifty would prefer the security of personal ownership to a politician's promise.

* Distortion Three: Social Security "provides a secure and predictable financial guarantee"--while in a system of personal accounts, retirement benefits would "ride the financial market roller coaster." Neither half of this proposition is true.

Remember, the current system provides a mere statutory right to benefits that Congress can (and indeed must, as Aaron and Reischauer themselves point out) cut at some future date. Only personal accounts offer workers ownership of constitutionally protected property. Nor is it true that personal account assets would have to ride the roller coaster of the stock market. Many plans would require workers to shift from equities into fixed-income assets as they grow older. Simulations show that such portfolio rules could easily protect workers nearing retirement from just about any market decline--even a crash on par with 1929. An economic collapse large enough to short-circuit these safeguards would probably destroy a public system as well.

* Distortion Four: Whatever the market does on average, personal accounts would put workers who make unwise investments at risk of catastrophic losses. Once again, most plans directly address this concern. Some would limit investment choices to just a few index funds. In no plan would workers be allowed to invest their assets in just one or two companies--or in anything as risky as Uncle Fred's hardware store.

* Distortion Five: Administrative fees would "chomp away at the returns" to personal accounts. Of course, without regulatory guidelines, workers might squander any amount of money on pricey brokers. But Aaron's and Reischauer's estimate of average annual fees (1 to 2 percent of assets) is way above what is now charged by an efficient no-load index fund. The expense ratio of the Vanguard 500 Portfolio, for instance, is only 0.2 percent. Many plans, moreover, would limit allowable fees. And most also provide for one or more default investment options administered by a government-sponsored board. These options would probably cost even less than the Vanguard indexes--while offering those workers who either can't or don't want to manage their own accounts a sure way to match the market.

* Distortion Six: The high cost of marketing and administering private annuities--as much as 20 percent of account balances--makes the personal account approach prohibitively wasteful. Wrong again. Most economists expect that the private annuity market would become vastly more efficient as the pool of annuitants expands under a personal account plan. Government, in any case, could sell a minimum annuity to everyone--at a single pooled rate with negligible waste. Or else, as in Chile, each retiree could be given the choice between purchasing an annuity and making voluntary annual withdrawals, with the maximum annual withdrawal determined by his or her remaining life expectancy.

* Distortion Seven: Social Security provides insurance to many nonaged persons, such as young widows and children, that personal accounts wouldn't. But most plans would leave these benefits entirely in place. Since account balances would be bequeathable, the personal account approach would actually improve the protection available to these young survivors and dependents. In any case, the benefits in question only account for about 5 percent of total OASI outlays.

* Distortion Eight: Personal accounts leave no room for assistance to low-earners, who would have to be helped through a separate program with a "demeaning means test." Again, Aaron and Reischauer ignore what's being proposed. Some plans would provide a government savings match for low-earners. Others would "top up" the accounts of workers with insufficient assets to buy a minimum annuity. The effect of these provisions would be to create a funded system that duplicates the progressive tilt of the current system.

Aaron and Reischauer do a great disservice to the national discussion over Social Security. They make claims for the current system that aren't true--and they grossly distort how personal accounts would actually work.

They rightly point out that most Americans are extremely dependent on Social Security. But they draw the wrong conclusion. They suggest that this dependence means we cannot make fundamental changes to the current system. To the contrary, it means that we must make fundamental changes. Giving people personal ownership over assets invested in the real economy would not only help the economy, it would greatly increase the income and security of tomorrow's retirees.


. Last updated: 1 May 1998