The Cost of Personal Retirement Accounts

Some analysts of Social Security reform are advancing the notion that the cost of a Social Security reform plan based on personal retirement accounts is shown by the amount of general revenues that would be needed to finance the transition to the accounts. This notion is sorely mistaken. Among other problems, this yardstick does not count increases in Social Security payroll taxes or cuts in future Social Security benefits as costs.

The Transition to Personal Accounts

All personal account reform plans involve a transition-financing burden. That is because Social Security operates primarily on a pay-as-you-go basis, where today’s taxes are not saved and invested to finance the future retirement benefits of today’s workers. Rather, those taxes are mostly immediately paid out to finance the benefits of retirees. The benefits of tomorrow’s retirees are then to be paid out of the taxes of tomorrow’s workers.

Under a personal account reform plan, some of the money that would be paid into the system today to finance current benefits would instead be saved and invested in personal accounts for future benefits. Therefore some additional funds must be added to the system from somewhere in order to continue to pay current benefits in full, until the personal accounts begin taking over benefit payment responsibilities in the future. That is the transition-financing burden.

Financing the Transition

This transition can be financed from many possible sources, some internal to Social Security itself, such as increasing payroll taxes or cutting future promised benefits, and some from outside the program, by providing general revenues to the program derived from a myriad of possible sources.

Assuming that the amount of general revenues needed for a personal account reform plan is an accurate and complete measure of the costs of such a reform plan is fallacious. That measure excludes some items that are true costs, or financial burdens, and includes other items that are not true costs.

For example, financing the transition by increasing payroll taxes does not involve a general revenue transfer to Social Security. Therefore, it would not be counted as a cost if the measure of costs is the amount of general revenues required by a reform plan. If a personal account plan were financed entirely by removing the cap on taxable income and increasing the payroll tax rate, then under the general revenue measure of costs, such a plan would have no costs.

Similarly, financing the transition by reducing future promised Social Security benefits also does not involve a general revenue transfer to Social Security. So, again, any such reduction in benefits would not be counted as a cost if the measure of costs is the amount of general revenues devoted to Social Security during the transition. But of course, any such reduction in benefits would be a cost to those losing the benefits.

On the other hand, suppose a personal account reform plan increased economic growth and income, resulting in increased general revenues to the government. Suppose those increased general revenues were devoted to Social Security to help finance the transition. This would then be counted as part of the costs of the reform plan if the yardstick were simply the amount of general revenues devoted to Social Security.

But these general revenues were generated by the reform plan itself. They would not exist without the reform. Consequently, they cannot logically be considered part of the net cost of the reform plan. Quite to the contrary, these additional revenues are a benefit of the reform plan, used to offset, and hence reduce, the net transition-financing burden.

Another possible transition-financing source would be to reduce the growth of non-Social Security federal spending, and devote the savings to Social Security. The Cato Institute has touted this as a source of transition financing for many years, citing at various times potential savings from eliminating unecessary spending found in corporate welfare, agriculture subsidies, the defense budget, and many other programs.¹

This funding, of course, would involve additional general revenue transfers to Social Security, and so would be counted as a cost if such general revenue transfers were the simple yardstick. But counting reductions in non-Social Security spending as a cost while not counting reductions in Social Security spending as a cost is transparently erroneous.

Moreover, to the extent the reductions in non-Social Security spending involve wasteful or counterproductive federal spending, those reductions would not represent a cost. Again, quite to the contrary, these reductions would, in fact, be another benefit of the reform plan, used to offset, and hence reduce, the net transition-financing burden.

The True Costs of Personal Accounts

As discussed at the outset, any personal account reform plan involves a transition-financing burden, as some of the funds that are used to pay current benefits under the present system are saved and invested in the personal accounts instead. So additional funds for Social Security must come from somewhere to ensure the continued payment of promised benefits, until the personal accounts start taking over benefit payment responsibilities.

For many years, Jose Pinera, William Shipman, and myself, as well as others, have argued that this whole transition-financing burden is actually not a cost in the true sense of the word as a sacrifice or expenditure of resources. What the transition really involves is paying for the increased savings and investment of the fully funded personal account system in shifting from the non-savings, simply redistributionist, pay-as-you-go system of the current Social Security framework. The money devoted to the transition in a personal account system is not lost or spent. It effectively is saved through the personal accounts for future use.

In other words, what is involved in this transition from a pay-as-you-go to a fully funded system is advance funding, not net additional or new funding. Indeed, the whole reason for that advance funding is that over the long run it generates higher returns that effectively reduce the cost of the system.

Another way to see this is through the concept of the unfunded liability of Social Security, currently officially estimated at about $11 trillion. What the transition really involves is the payoff of that unfunded liability, by reforming pay-as-you-go Social Security into a fully funded personal account system of savings and investment with no unfunded liabilities. As William Shipman has argued quite correctly for the Cato Institute, the advance funding of a personal account system will over the long run reduce the financial burden of that unfunded liability, not add to it.²

The Costs Versus the Benefits

This is not to say that the transition-financing burden of personal account reform plans is not properly considered in evaluating such plans. Since the transition involves an increased short-term burden in order to get increased future gains, of course that short-term burden must be considered.

But the full net economic value of that burden must be included, considering the transition financing from all sources, net of additional resources for the transition produced by the reform itself. If the transition is financed in part by increasing payroll taxes or cutting future promised Social Security benefits, that must be considered as part of the financing burden as well. To the extent, however, that the transition is financed by additional resources generated by the reform plan itself, those resources are not part of the net transition-financing burden of the reform plan. To the extent the reform can be financed by such resources, the reform is self-financing and does not involve an additional financial burden. As discussed above, this total net transition-financing burden is not measured by the amount of general revenue transfers to Social Security involved in any reform plan.

Moreover, this transition-financing burden must then be weighed against the full long-term benefits of the reform plan. Bigger account plans naturally involve bigger net transition-financing burdens. But they also involve much greater long-term benefits that must be weighed in the balance. These benefits include the following:

• If the accounts are large enough, they would eliminate the long-term deficits of Social Security without raising taxes or cutting future promised benefit levels. That is because the big accounts shift so much responsibility for payment of retirement benefits from the old pay-as-you-go Social Security framework to the personal accounts that the deficits are ultimately eliminated through this process. This point is established through the official scores by the Chief Actuary of Social Security of both the Progressive Personal Account proposal³ issued by the Institute for Policy Innovation last summer, and the legislation introduced by Rep. Paul Ryan (R-WI) and Sen. John Sununu (R-NH).⁴
• In the process of eliminating these deficits and achieving permanent solvency for Social Security, the unfunded liability of Social Security, currently estimated at $11 trillion, is eliminated as well.
• The large personal accounts generate substantial wealth and assets directly and personally owned by each worker in his or her personal account. Under the Ryan-Sununu bill, the Chief Actuary’s official score projects that workers would accumulate $7 trillion in these accounts in today’s dollars after just 15 years. This would greatly broaden the ownership of wealth in our society, and sharply reduce the concentration of wealth.
• At standard, long-term, market investment returns, workers paying into personal accounts would receive much higher retirement benefits. For the Ryan-Sununu bill, for example, the net gain is estimated at two-thirds to 100 percent more.
• Once the transition to personal accounts is completed, the payroll tax can be reduced. The official score of the Ryan-Sununu bill projects that the Social Security payroll tax rate can ultimately be reduced from the current 12.4 percent to about 4 percent, with another 6.4 percent going into the personal accounts owned directly and individually by each worker (which consequently is not a tax). This compares to a total payroll tax rate of close to 20 percent that would ultimately be necessary to pay all currently promised benefits under the current Social Security framework. This long-term tax reduction, as well as the higher benefits from the personal accounts and elimination of the long-term Social Security deficits, reflects the net gain from advance funding.
• The personal accounts would greatly increase economic growth through increased saving and investment, and through the extensive tax relief involved in the accounts. The proportion of payroll taxes workers can shift to the accounts is effectively an immediate payroll tax cut, as the money remains directly and personally owned by each worker. The accounts also involve a new immediate tax-free investment opportunity. Then there is the long-term payroll tax reduction. All of this adds up to new jobs, higher wages, and more overall national income.

Using the amount of general revenue transfers to Social Security to help finance the transition as the yardstick by which to rank reform plans would not take into account any of these benefits. Consequently, that yardstick is biased against personal accounts, and large personal accounts in particular, as they would produce even larger benefits that are not counted. Moreover, counting only general revenue transfers to help finance a personal account transition as a cost of a reform plan also is biased against personal accounts, and again large personal accounts in particular which are only feasible with such general revenue transfers.

Endnotes

1. See, e.g., Peter Ferrara and Michael Tanner, A New Deal for Social Security , Washington, D.C.: Cato Institute, 1998, pp. 187–194.
2. William Shipman, “Facts and Fantasies About Transition Costs,” Cato Institute, SSP. No. 13, October 13, 1998.
3. “Estimated Financial Effects of the Progressive Personal Account Plan,” December 1, 2003, Office of the Actuary, Social Security Administration; “Additional Estimated Financial Effects of the Progressive Personal Account Plan,” April 6, 2004, Office of the Actuary, Social Security Administration.
4. “Estimated Financial Effects of the ‘Social Security Personal Savings and Prosperity Act of 2004’”, July 19, 2004, Office of the Actuary, Social Security Administration.