Testimony Before Senate Aging Committee Regarding Progressive Personal Accounts

Witness: Peter Ferrara
Testimony Date 06/15/2004
Body of Congress: United States Senate
Committee: Senate Aging Committee
Subcommittee:
At the Request of:
Bill:

Testimony Subject:
Progressive Personal Accounts for Social Security

The Progressive Personal Account reform plan would allow lower income workers to contribute a larger portion of their Social Security taxes to personal accounts than higher income workers. The reform plan provides for a large personal account overall enabling workers on average to contribute 6.4 percentage points of the current 12.4% Social Security payroll tax to the accounts.

The reform plan roughly mirrors the progressivity of the current Social Security system, with workers across the board enjoying relatively equal net gains for equivalent investment portfolios. Moreover, the plan would enhance progressivity in the following ways:

• It would sharply increase future retirement benefits for low and moderate income workers.

• It would pay far higher, market based rates of return for workers across the board than Social Security now offers. As a result, the plan would sharply improve both benefit adequacy and equity.

• Under the reform plan, low and moderate income workers would gain much greater accumulations of personal wealth than under Social Security.

• These much larger accumulations of personal wealth by low and moderate income workers would greatly broaden wealth ownership in our nation, and sharply reduce the concentration of wealth.

• The reform plan addresses the problem of those with lower life expectancies, who tend to be those with lower incomes, suffering lower returns and lifetime benefits as a result.

Indeed, the Progressive Personal Retirement Account plan is more progressive on these grounds than either the Diamond-Orszag plan (which does nothing to increase the personal wealth of working people), or the much smaller accounts proposed in Option 2 of the 2001 reform commission.



Components of the Reform Plan

Last summer, I authored a study published by the Institute for Policy Innovation which proposed a progressive personal account option for Social Security. Peter Ferrara, A Progressive Proposal for Social Security Personal Accounts, Institute for Policy Innovation, Policy Report 176, June 2003. The main components of the reform plan are as follows:

• Out of the 12.4% Social Security payroll tax, workers would be free to choose to shift to personally owned, individual accounts, 10 percentage points on the first $10,000 in wages each year, and 5 percentage points on all wages above that, to the maximum Social Security taxable income. This creates a progressive structure with an average account contribution among all workers of 6.4 percentage points

• Benefits payable from the tax free accounts would substitute for a portion of Social Security benefits based on the degree to which workers exercised the account option over their careers. Workers exercising the personal accounts would receive Recognition Bonds guaranteeing them the payment of Social Security retirement benefits based on the past taxes they have already paid into the program. Workers would then also receive in addition the benefits payable through the personal accounts.

• Workers choose investments by picking a fund managed by a major private investment firm, from a list officially approved for this purpose and regulated for safety and soundness, similarly to the operation of the Federal Employee Thrift Retirement System.

• The accounts are backed up by a safety net guaranteeing that workers would receive at least as much as Social Security promises under current law.

• Apart from this personal account option, there would be no change in currently promised Social Security benefits of any sort, for today’s seniors, or anyone in the future. Anyone who chooses to stay in Social Security would receive the benefits promised under current law. Survivors and disability benefits would continue as under the current system unchanged.

Critically, the unique structure of this reform plan maintains both a defined benefit and a defined contribution benefit structure for workers in the same system. Guaranteeing that all workers would receive at least the benefits promised under current law maintains the current defined benefit structure for all workers. Yet, the defined contribution benefits of the personal accounts provide a high probability that all workers across the board would get much higher benefits from the accounts.

Moreover, the reform plan maintains insurance protection for workers while adding savings and investment. The plan requires each worker at retirement to use the funds in his or her account to purchase an annuity providing at least the benefits promised by Social Security. That annuity provides insurance protection against living too long. Moreover, the survivors and disability insurance of the current system is continued without change.


The transition under the reform is financed by 4 factors:

1. Short term Social Security surpluses projected until 2018.

2. Reducing the rate of growth of Federal spending by 1 percentage point per year for 8 years, and devoting those savings to the transition. The proposal would consequently involve a Federal spending limitation measure providing for this reasonable and moderate spending restraint. The proposal, therefore, provides a vehicle for beginning to get runaway Federal spending under control. The spending savings for those years are maintained until all short term debt issued to fund the transition is paid off in full.

3. The revenue feedback from increased saving and investment in the accounts due to taxation of increased investment returns at the corporate level, as developed by Harvard Prof. Martin Feldstein and former Sen. Phil Gramm for his legislative proposal;

4. To the extent needed, the sale of surplus Social Security trust fund bonds. This just involves paying Social Security back for all the surpluses it has lent to the Federal government in the past for other government spending. Under the current system, those bonds are just going to be redeemed for cash from the Federal government anyway after 2018, until the trust fund is exhausted in 2042.

The Official Score of the Reform Plan by the Chief Actuary of Social Security

This proposal has already been scored by the Chief Actuary of Social Security. 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 That official score shows:

• The large personal accounts in the plan are sufficient to completely eliminate Social Security deficits over time, without any benefit cuts or tax increases. That is because so much of Social Security’s benefit obligations are ultimately shifted to the accounts. As the Chief Actuary stated, under the reform plan, “the Social Security program would be expected to be solvent and to meet its benefit obligations throughout the long-range period 2003 through 2077 and beyond.” Id., p. 1. Indeed, the eventual surpluses from the personal accounts are large enough to eliminate the long term deficits of the disability insurance program as well, even though the reform plan does not otherwise provide for any changes in that program.

• The accounts achieve this not only with no benefit cuts or tax increases in Social Security. Over time, in fact, the accounts would provide substantially higher benefits, as well as tax cuts. The official score shows that by the end of the 75 year projection period, instead of increasing the payroll tax to over 20% as would be needed to pay promised benefits under the current system, the tax would be reduced to 3.5%, enough to pay for all of the continuing disability and survivors benefits.

• Moreover, as discussed further below, at standard, long term market investment returns, the accounts would produce substantially more in benefits for working people across the board than Social Security now promises, let alone what it can pay. This is the only reform proposal that achieves that result. 

• The reform also achieves the largest reduction in government debt in world history, by eliminating the unfunded liability of Social Security, almost three times the current reported national debt.

• The reform would also greatly increase and broaden the ownership of wealth and capital through the accounts. All workers would participate in our nation’s economy as both capitalists and laborers. Under the Chief Actuary’s score, workers would accumulate $7 trillion in today’s dollars in their accounts by 2020. Wealth ownership throughout the nation would become much more equal, and the concentration of wealth would be greatly reduced. This is also discussed further below.

• With the above transition financing, Social Security achieves permanent and growing surplus by 2029 under the Chief Actuary’s score. Before that time an average of about $52 billion in surplus Social Security trust funds bonds are sold each year for 24 years, for a total of $1.25 trillion, all in today’s dollars. 

• Even with the sale of the surplus trust fund bonds, the trust fund never falls below $1.38 trillion in today’s dollars, or 145% of one year’s expenditures, with the official standard of solvency being 100%. After 2029, the trust fund grows permanently, reaching 12.5 times one year’s expenditures by the end of the projection period, or about $6.3 trillion in today’s dollars, far too much.

• Within 15 years after 2029, the reform produces sufficient surpluses to pay off all the bonds sold to the public during the early years of the reform. So this surplus completely eliminates the Federal debt sold to the public in the earlier years of the reform, leaving the net impact of the reform on debt sold to the public at zero. Indeed, as mentioned above, the reform goes on to completely eliminate Social Security’s current unfunded liabiltiy of $10.5 trillion, close to three times the reported national debt.

Analysis of Retirement Benefits Under the Reform Plan

In the IPI study, I reviewed how the reform plan would affect retirement benefits for three hypothetical family/worker combinations. Ferrara, A Progressive Proposal for Social Security Personal Accounts, pp. 13-14. This is not a comprehensive analysis of how workers would fare under the reform plan. But the compelling results strongly indicate the great gains that would result for workers across the board.

First, take the case of an average income worker age 40 in 2003 earning $35,000 that year. He entered the work force at age 23 earning $17,677 per year then, and earns only the average annual salary increase each year. Suppose he was able to exercise the personal account option in the proposal from the start of his career. The payment into his account for the year, at age 40, would be 10% of the first $10,000 in wages, and 5% of everything above that, for a total of $2,250.

Assume he invests each year in a diversified portfolio of half stocks and half bonds and earns standard market investment returns. The long term real rate of return on corporate bonds is around 3.5% and for corporate stocks 7% to 7.5% and more. Ibbotson Associates, Stocks, Bonds, Bills and Inflation 2003 Yearbook, Market Results for 1926-2002 (Chicago, 2003); Peter Ferrara and Michael Tanner, A New Deal for Social Security, (Washington, DC: Cato Institute, 1998), pp. 72-73. The comprehensive work by William Shipman on an administrative framework for personal accounts indicates that the administrative costs for such a system would be less than 25 basis points. William G. Shipman, Administrative Challenges Facing A Market-Based Social Security System, CarriageOaks Partners, 1999; Ferrara and Tanner, pp. 88-89. Consequently, we may assume that a portfolio of half stocks and half bonds would yield a net annual real return of about 5% over a worker’s career, at standard market investment returns.

With such returns, the worker would reach retirement with a total accumulated trust fund of $334,095, in today’s dollars after adjusting for inflation. That fund would be enough to pay an annual annuity about 70% more than what Social Security promises but cannot pay, $2,653 per month compared to $1,567. With a higher percentage invested in stocks, the account would pay even more. If the account were invested two thirds in stocks and one third in bonds, at standard market investment returns the worker would reach retirement with an account of $417,815. That fund would pay $3,317 per month, just over twice what Social Security promises but cannot pay. With the account invested entirely in stocks and earning standard market investment returns, the worker would retire with a fund of $576,761, paying him $5,186 per month, again all in today’s dollars. This would be well over three times what Social Security promises, but cannot pay.

Now take the example of an average income two earner couple, each 40 years old in 2003. The husband earned an income of $40,000 that year, and the wife earned an income of $30,000, which is consistent with U.S. Census Bureau data regarding the average income of two earner married couples. They again each entered the work force at age 23, with the husband earning $20,202 that year and the wife earning $15,152. They also each earn only the average salary increase each year. Suppose again as well that they exercised the personal account option described above from the start of their careers.

At standard market investment returns, with a diversified portfolio invested half in bonds and half in stocks, the couple would reach retirement with a total fund of $668,178, again in today’s dollars after adjusting for inflation. That fund would be able to pay them about 60% more than what Social Security promises but cannot pay, $4,987 per month compared to $3,133. Remember this results from the contributions to the accounts of only 10% of the first $10,000 in wages and 5% of everything above that, compared with about 10 percentage points of the Social Security tax going to Social Security retirement benefits.

Moreover, with a higher percentage invested in stocks, again the account would pay more. With the account invested two-thirds in stocks and one third in bonds, the couple would reach retirement with $835,506. That would pay monthly benefits of $6,236 in today’s dollars, again about twice what Social Security promises but cannot pay. With the account invested entirely in stocks, then at standard returns the couple would reach retirement with a fund of over $1 million ($1,153,188) in today’s dollars. That fund would finance a monthly benefit of $9,840, again more than three times what Social Security promises but cannot pay.

Low income workers would receive similar gains under our progressive proposal, even though the Social Security benefit formula is skewed to favor lower income workers. Take the example of a worker age 40 in 2003 who earned only $20,000 that year. Assume he entered the work force at age 23 earning $10,101 that year. He also earns only average wage increases each year, which leaves him in the same position relative to other workers each year as a worker earning $20,000 this year. Assume again that he is able to exercise our proposed personal account option from the start of his career.

At standard market investment returns, with a diversified portfolio half in bonds and half in stocks, he would reach retirement with a total fund of $223,282, in today’s dollars. That fund would be able to pay him 64% more than Social Security promises but cannot pay, $1,773 per month compared to $1,083. With the account invested two thirds in stocks and one third in bonds, the worker would reach retirement with an account of $279,607. That fund would pay him $2,220 per month in today’s dollars, again over twice what Social Security promises but cannot pay. With the account completely invested in stocks and earning standard market returns, the worker would reach retirement with a fund of $386,542 in today’s dollars. That fund would again pay him over three times what Social Security promises but cannot pay, $3476 per month compared to $1013.

Moreover, such a career low income worker would usually not be entering the work force at age 23, after college. Suppose we assume the worker enters the work force at age 19 earning $8,600 that year. With the additional funds from four years of early work at ages 19-22, and a portfolio of half bonds and half stocks earning standard returns, the worker would reach retirement with a trust fund of $271,505 in today’s dollars. That fund would pay the worker 84% more than Social Security promises but cannot pay, $2156 per month compared to $1172.

With the account invested two-thirds in stocks and one third in bonds, the worker would reach retirement with an accumulated fund of $350,498. That would pay him $2,783 per month in today’s dollars, again over twice what Social Security promises but cannot pay. A fund invested entirely in stocks starting at age 19 would accumulate for this lifetime low income worker to one half million by retirement ($500,471), which would pay him close to 4 times what Social Security promises but cannot pay.

We can safely conclude from this analysis that workers across the board will receive substantially higher retirement benefits through the personal accounts than Social Security promises them under current law, let alone what Social Security will actually be able to pay in the future. This is especially so given the broad margins of gain for low and moderate income workers in these examples. The vast majority of family and worker histories would fall within the parameters of these examples, or would involve above average income workers that would clearly gain by broad margins as well.

Why this enormous gulf between what the personal accounts can pay and what Social Security can pay? The personal accounts operate as a fully funded system. The money paid in is saved and invested in new capital investments. These capital investments actually increase production, and the value of this production increase is returned to investors in the form of a rate of return or interest payment on their investments. Over the course of a lifetime, this return would accumulate to large sums, which would then be used to finance benefits in retirement.

But Social Security operates primarily on a pay-as-you-go basis, where the money paid in today is not saved and invested but is mostly immediately paid out to finance current benefits. The future benefits of today’s workers are not to be paid by their savings and investment, but by the future taxes to be paid by future workers. Such a system adds nothing to production. It is a mere redistribution system, transferring funds from one segment of the population to another. This means that workers under such a system lose the full amount of the increased production and associated returns they would get if their money was invested in private, productive assets through a fully funded system. The payroll tax financed redistribution system can pay some effective return as revenues grow over time due to increased wages and population growth, enabling the system to pay more to retirees than just what they paid in. But this effective return, which is still obtained by a tax redistribution from others rather than increased production, will never be anywhere near as great as the full returns produced by capital investment. For further discussion, see Peter J. Ferrara, Social Security Rates of Return for Today’s Young Workers (Wash. D.C.: National Chamber Foundation, 1986), pp. 8-11; Peter J. Ferrara and Michael Tanner, A New Deal for Social Security (Wash.D.C.: Cato Institute, 1998), Chapt. 4

Conclusions Regarding Progressivity

Based on this background, we can conclude as well that the Progressive Personal Retirement Account Plan would provide far more benefits and gains for low and moderate income workers, and ultimately be far more progressive, than either Option 2 offered by the 2001 Social Security reform commission, or by the Diamond Orszag plan. The reasons for this are discussed below.

First, the analysis above indicates that the design of the personal account option in the Progressive Personal Retirement Account plan roughly mirrors the progressivity of Social Security. In our examples above, workers seemed to enjoy roughly the same equivalent net gain over Social Security given the same investment portfolios. For example, with a personal account investment portfolio of half stocks and half bonds, the average income worker would gain about 70%, the higher income two average earner family would gain 60%, and the low income worker would gain 64%. With a more realistic assumption that lifetime low income workers end formal education sooner and start work earlier in life, the net gain for this worker was 84%. So workers with investment portfolios invested half in stock and half in bonds seem to gain roughly two thirds from the account option, and perhaps somewhat more for lower income workers.

Similarly, for portfolios invested two thirds in stocks and one third in bonds, workers in these examples seemed to gain across the board about twice as much as Social Security promises but cannot pay. Higher income workers would enjoy about the same margins of net gain from the same investment portfolios.

This conclusion regarding roughly equivalent net gains would seem to apply to the vast majority of workers, if not all workers. Given the complexity of Social Security and widely varying possible income and family combinations, it may be possible to find some cases where workers and their families may not gain as much as the norm. But all workers should gain substantially in any event.

Moreover, there are ways to make the reform plan more progressive. For example, former Senator Bob Kerrey (D-NE) proposed a Kidsave plan, where the government would deposit $1,000 into a personal account at birth for each child. This could be added to the Progressive Retirement Account plan, progressively providing a higher relative net gain for lower income workers than higher income workers.

Secondly, the Progressive Personal Retirement Account plan would sharply increase future retirement benefits for low and moderate income workers. The analysis above indicates that through the personal accounts, at standard long term investment returns, these workers would be able to gain an increase in retirement benefits over what Social Security promises, let alone what it can pay, in the range of two thirds more to twice as much, or even more.

As a result, the large accounts in the Progressive Personal Retirement Account Plan would pay much higher benefits to workers in the future than the Diamond Orszag plan, or Option 2 of the 2001 reform commission, with its much smaller accounts. On this basis, that plan is more progressive than either Diamond-Orszag or the reform commission’s option 2.

Thirdly, the Progressive Personal Retirement Account plan would pay far higher, market based rates of return for workers across the board than Social Security now offers. For most workers today, even if all of Social Security’s promised benefits were somehow paid, the real rate of return paid by the program on the huge taxes paid by workers and their employers over their careers would be 1% to 1.5% or less. For many it would be zero or even negative. Ferrara and Tanner, A New Deal for Social Security, Chapter 4; William Beach and Gareth Davis, Social Security’s Rate of Return, Report of the Heritage Foundation Center for Data Analysis, , no. 98-01, January 15, 1998; Peter J. Ferrara, Social Security Rates of Return for Today’s Young Workers, National Chamber Foundation, Washington, DC: 1986; Peter Ferrara and John Lott, “Social Security’s Rates of Return for Today’s Young Workers”, in Peter Ferrara, ed., Social Security: Prospects for Real Reform, (Washington, DC: Cato Institute, 1983), pp. 13-36. By contrast, as discussed above, the long term real return on stocks is 7% to 7.5% and on corporate bonds 3% to 3.5%. The large accounts in the Progressive Personal Retirement Account plan would consequently sharply raise rates of return for workers across the board. As a result, the plan would sharply improve both benefit adequacy, as discussed above, and equity, providing all workers with higher, market based returns.

Indeed, the Progressive Personal Retirement Account plan would provide workers across the board with higher, more market based returns than either the Diamond-Orszag plan, or the much smaller accounts proposed in Option 2 of the 2001 reform commission. On this basis as well, the Progressive Personal Retirement Account plan is more progressive than either of these two alternatives.

Fourthly, under the Progressive Personal Retirement Account plan, low and moderate income workers would gain much greater accumulations of personal wealth than under Social Security. As discussed in our examples above, even a career low income worker would accumulate by retirement close to $300,000 in today’s dollars, and more, in the large personal accounts provided under the reform plan. A lifetime low income worker who starts work at age 19 and invests two thirds in stocks and one third in bonds would accumulate at standard market investment returns about $350,000 in today’s dollars by retirement. The official score of the reform plan by the Chief Actuary of Social Security concluded that just 15 years after the reform plan is adopted, workers would have accumulated $7 trillion in their own personal accounts.

Moreover, under the Progressive Personal Retirement Account plan, low and moderate income workers would gain much greater accumulations of personal wealth than under either the Diamond-Orszag plan (which does nothing to increase the personal wealth of working people), or the much smaller accounts proposed in Option 2 of the 2001 reform commission. The Progressive Personal Retirement Account plan is consequently more progressive than either of these 2 alternatives on these grounds as well.

Fifthly, the much larger accumulations of personal wealth by low and moderate income workers under the Progressive Personal Retirement Account plan would greatly broaden wealth ownership in our nation, and sharply reduce the concentration of wealth. A study by Harvard Professor Martin Feldstein indicates that the large personal accounts in the Progressive Personal Retirement Account plan could reduce the total concentration of wealth by as much as one half. Martin Feldstein, “Social Security and the Distribution of Wealth,” Journal of the American Statistical Association, (December 1976): 90-3.
The Progressive Personal Retirement Account plan is consequently more progressive on these grounds than either the Diamond-Orszag plan (which again does nothing to increase the personal wealth of working people), or the much smaller accounts proposed in Option 2 of the 2001 reform commission.

Sixthly, the Progressive Personal Retirement Account plan addresses the problem of those with lower life expectancies, who tend to be those with lower incomes, suffering lower returns and lifetime benefits as a result. Under the current Social Security system those with lower life expectancies live fewer years in retirement to collect benefits, and so suffer lower returns and lifetime benefits as a result. But with the large personal accounts of the Progressive Personal Retirement Account plan, workers who die before retirement would not lose everything they had paid into the system over the years. They would be able to leave the account funds they had accumulated to their children or other heirs. After retirement, workers would only have to buy an annuity paying what Social Security promises under current law, and could leave all remaining funds again to children or other heirs, as they choose. Moreover, under the plan, groups that suffer lower life expectancies, such as African Americans, could buy their annuities through social organizations such as the NAACP that would reflect the lower life expectancies of the group. These workers would consequently get higher benefits through those annuities, reflecting the fewer years of retirement they would collect benefits on average.

Neither the Diamond-Orszag plan nor Option 2 of the 2001 reform commission would do as much to eliminate this problem. The Progressive Personal Retirement Account plan is consequently more progressive than these alternatives on these grounds as well.

Would the general revenues used to finance the transition to the personal accounts in the Progressive Personal Retirement Account plan undermine these conclusions? About half of these revenues would come from the corporate revenue feedback discussed above. These are new revenues generated by the reform plan itself that would not exist without the plan. These additional corporate revenues would not burden low and moderate income people and so would not counter any of the benefits of the reform plan for those workers, or undermine the above progressive features of the reform plan.

The rest of the general revenues would come from restraining the growth of other Federal spending. This would be a modest restraint equivalent to the budget savings from the long term baseline achieved during the Clinton years. To the extent that results in eliminating wasteful or counterproductive spending, that would be a net gain to society. This consequently would also not burden low and moderate income people or undermine the above progressive features.

Moreover, there is plenty of scope for restraining Federal spending where that would not adversely affect low and moderate income people or the progressivity of the reform. Cutting back on corporate welfare, uneeded military bases, subsidies for large agribusinesses, and other areas would probably benefit low and moderate income people and enhance progressivity. Reductions in government programs resulting from reforms that reduce dependency also should be seen as enhancing rather than reducing progressivity.