Social Security Reform without Illusion: The Five Percent Solution

America’s entitlement programs for senior citizens are on an unsustainable course. Unless changes are made soon, we face the prospect of exorbitant tax rates or severe benefit cuts. Fortunately, there is a solution. Retirement benefits can be secured without raising payroll taxes by giving people the opportunity and incentive to save for their own retirement.

The Need for Reform. Although the federal payroll tax currently pays for almost all Social Security and Medicare benefits, the shortfall will grow rapidly during the baby boomer retirement years. Eventually, retirement benefits paid to the elderly will consume the entire federal budget, crowding out every other spending program. For example:

• This year, for the first time in more than 20 years, the combined deficit in Social Security and Medicare will require a net transfer from the Treasury equal to almost 4 percent of federal income tax revenues.
• That figure will double in the next five years and double again in the five years after that.
• Ten years from now we will need one-in-seven income tax dollars, in addition to the payroll tax, to pay retirement benefits.
• By 2020, elderly entitlements will consume one-in-four income tax dollars and by 2030 they will consume one of every two.
• By mid-century, when today’s college students retire, we will need three-fourths of all federal income taxes to pay their retirement benefits.

To avoid this unpleasant and unsustainable future, we must move quickly to a funded system, under which each generation pays its own way. The transition to a new system will not be easy. But each year we delay increases the cost of making it. What follows are the main features of our reform proposal.

Creating Personal Retirement Accounts. All workers who have not yet reached retirement age will be able to set aside part of their earnings in a personal retirement account (PRA). Specifically, PRA deposits will equal 10 percent of the first $7,650 in annual earnings, 3 percent of earnings between $7,650 and $55,000, and 1 percent of earnings above $55,000. Thus for a worker who earns $7,650 per year or less, the PRA deposit will equal 10 percent of wages. For an average-income worker, the PRA deposit will equal about 5 percent of wages. (The PRA contribution rates are specifically designed to replicate the progressivity of the current system.) Funds in these accounts will be invested in assets, and as the balances grow over time, they will replace the government’s promises to pay benefits. The youngest workers will completely pay their own way as the private accumulations provide a retirement income equal, on the average, to what the current system promises.

Funding the Accounts. Workers will be able to divert a portion of their payroll taxes into a PRA, with lower-income workers able to divert more than higher-income workers. In return, workers must make their own additional contribution of 1.25 percent of wages, to be matched by their employer. Roughly speaking, for each dollar an average-income worker invests in a PRA, three additional dollars will be invested by an employer and funds that otherwise would have been paid to the government. For each dollar invested by the lowest-income workers, seven dollars will be invested by an employer and diverted payroll tax dollars. The PRAs of the highest-income workers will be almost totally funded by the 1.25 percent contributions of employees and their employers, with little or no payroll tax diversion.

After eight years, the worker’s contribution (matched by the employer) will gradually rise to 1.75 percent and the percent of payroll taxes diverted will fall. At that point, for each dollar an average-income worker contributes to his PRA, about two dollars will be contributed by his employer or the government. For each dollar a low-income worker contributes, almost five dollars will be contributed by someone else. The contribution rates will remain at these levels until 2038 when the current payroll tax rates are no longer needed to sustain the program.

Easing the Transition. To make the transition easier, for the first five years of the program employees and their employers could divert monies currently sent to defined contribution retirement plans — such as 401(k)s — to meet their 1.25 percent PRA contribution requirement. Additionally, small businesses could also be allowed a year’s delay before matching their employees’ contributions.

Choosing to Participate. A worker’s participation in the reformed system will be voluntary. However, since there will be no increase in the Social Security payroll tax, those who choose to remain in the current system will have to accept lower benefits in future years as payroll tax revenues fail to keep pace with Social Security’s promises. By contrast, those who participate in the reformed system can expect benefits that will equal currently promised benefits, on the average.

Investing Prudently. Workers will not buy and sell individual stocks and bonds with their PRA funds. Instead, they will invest in approved, diversified funds that reflect the performance of the market as a whole, including stock index funds, bond funds and government securities funds. The management of these funds will be subject to strict accounting and financial standards. Funds will be approved by an independent governing board responsible for establishing safety and soundness criteria.

Administration. To avoid creating additional burdens for employers, all the administration and paperwork could be done internally at the Social Security Administration. Employers would send employee and employer contributions to the government, just as they do under current law. However, firms already administering defined contribution plans could make deposits directly to PRAs on behalf of their employees, just as they do now with their 401(k) plans.

Securing a Retirement Income. Over time, the Social Security benefits paid by the government to retired workers who participate in the PRA system will be gradually reduced by a predetermined formula. In general, increased PRA account balances will offset these reductions. During retirement, individuals will receive two monthly checks — one from the Social Security Administration (as under the current system) and one based on their private accumulation.

At the time they retire, individuals will use their accumulated PRA balances to purchase annuities. If the sum of their annuity check and Social Security check equals at least 150 percent of the poverty level, any surplus PRA funds may be used for other purposes — including certain tax-free health care expenses. As an alternative to purchasing an annuity, retirees may be given the opportunity to leave their account with a pension fund manager and withdraw an amount set each year by law, as is currently done in Chile.

Reducing Risk. This proposal has two explicit guarantees to PRA participants:
(1) Everyone at or near retirement will receive all promised Social Security benefits; and (2) Everyone else with at least 35 years of full-time work will have a retirement income equal to at least 150 percent of the poverty level. If any qualifying worker’s total benefit falls below 150 percent of the level of poverty, the federal government will supplement that worker’s benefit up to the 150 percent level.

Taxes During Retirement. Like deposits to Roth IRAs, individual deposits and payroll taxes diverted to PRAs will be made with after (income) tax dollars; thus withdrawals of these funds will be tax free. The portion resulting from employer contributions will be taxed as ordinary income at the time of withdrawal.

Accommodating Modern Family Life. To accommodate the changing nature of marriage and family life, all PRA contributions will be treated as community property. That is, PRA deposits will be divided fifty-fifty between a husband’s and wife’s accounts, regardless of who earns the wages.

Paying for Long-Term Care. Retired workers who have accumulated more than the amount required for the minimum annuity can use additional PRA balances to purchase long-term care insurance and to pay for long-term care directly during retirement. Tax-free withdrawals will be allowed for certain health care expenses, including long-term care expenses for debilitating end-of-life diseases such as Alzheimer’s disease, and for nursing home or stay-at-home care.

Paying for Reform. Unlike other reform proposals advanced in recent years, this proposal is fully funded. Deposits to PRA accounts are not funded by government borrowing. They are funded by expected Social Security payroll tax surpluses, the government’s promise to redeem the Social Security Trust Fund and new contributions to be made by employees and their employers.

Consequences of Reform. After about three decades, the reformed Social Security system will finance itself. At this point, workers’ Social Security payroll taxes and contributions will be more than sufficient to pay benefits and make contributions to PRA accounts. As a result, government can reduce the Social Security payroll tax, and over the next three decades, the combined contribution rate could be cut in half.

[page]

Countries around the globe are reforming their government-sponsored pension programs. Many of the reforms involve shifting away from pay-as-you-go financing in favor of partial or fully funded programs. Often, these reforms involve the creation of some form of individual accounts. The motivation is the same almost everywhere: To secure adequate retirement incomes without incurring mounting taxpayer burdens in future years.

“Workers should be given incentives to save more for their retirement.”

If we make real economic investments today, the income from those investments will pay for some or all of Social Security’s future scheduled benefits. If these investments are individualized, people will own their future retirement funds. This will reduce the likelihood of political interference in investment choices and eliminate the government’s ability to spend surplus payroll taxes on other programs. But there is no free lunch. To guarantee their retirement benefits, current generations must bear higher costs than would otherwise be required.

A transition to a funded system has several benefits:

• First, we will avoid the high taxes the current system promises to impose on future workers.
• Second, we will ensure that currently scheduled Social Security benefits for future retirees will actually be paid on average.
• Third, when workers own their individual retirement accounts, they have a property right lacking in the current program.
• Finally, the increase in savings resulting from growing personal account balances will expand the nation’s stock of capital, leading to more plant and equipment, and higher wages for future workers.

Because today’s pay-as-you-go Social Security system allows individuals to avoid saving for their own retirement, the nation’s current stock of capital is lower than it might otherwise be. Reversing the process will provide an economic gain for future generations, although at the cost of lower consumption for the generations that increase their saving.

[page]

“Without reform, Social Security and Medicare will consume all income tax revenues by 2070”

The 2004 Medicare and Social Security Trustees Reports show that programs for the elderly are on an unsustainable course. Expenditures exceed anticipated revenues, and the funding gap is projected to grow through time.

One way to assess the problem is to calculate the present value of the difference between expenses and revenues. This year, for the first time, the Trustees reported calculations for all the elderly entitlement programs, and the numbers are startling. Measured in current dollars:

• Over the next 75 years, scheduled benefits exceed dedicated revenues by $33 trillion.
• Looking indefinitely into the future, the present value of the additional revenues required by Social Security and Medicare totals almost $74 trillion.

What does it mean to have a $74 trillion revenue shortfall? It means that in order to pay benefits to current and future generations without using general revenues or cutting benefits, we need $74 trillion on hand right now, invested at the government’s borrowing rate. Because we don’t have $74 trillion invested today, next year the liability will be even larger. The year after that it will be larger still.

“Currently, Social Security surpluses are spent on other programs”

Some have asserted that an immediate solution to the problem is unnecessary, because the Trust Funds are flush with surpluses that can pay benefits well into the future. But the Social Security Trust Funds are not flush with assets that can pay benefits like those in a conventional pension fund. They are more like IOUs the government has written to itself. Surplus funds are not invested in financial assets like stocks and bonds. Instead, the surpluses are credited to the Trust Fund but are spent on other government programs. Every asset of the Trust Funds is a liability of the Treasury. Summing over both parts of government, the assets and liabilities net out to zero. In order to pay benefits in future years, the government will have to tax, borrow or cut spending on other programs. [See Figure I.]

• This year, for the first time in many years, the federal government will have to draw on general revenues to cover the excess of spending over revenues in the combined elderly entitlement programs; the funding deficit this year is equal to about 3.6 percent of federal income taxes.
• In less than five years, the share of income taxes needed will double, and five years beyond that it will double again.
• By 2020, the federal government will need more than one-in-four federal income tax dollars to pay benefits to the elderly, in addition to payroll taxes and other dedicated revenues.¹
• By 2030 (toward the end of the baby boomer retirement years) we will need more than half of all federal income tax revenues to pay for the deficits of Social Security and Medicare.
• By 2040, we will need two-thirds of federal income taxes; by 2050, three-fourths.
• And, by 2070, the elderly will need all federal income taxes (in addition to all payroll taxes), leaving nothing to pay for any other federal programs.

Clearly, we cannot sustain a pay-as-you-go system, under which promises made to today’s workers must be paid by future generations. Instead, we must move quickly to a funded system, under which each generation pays its own way. Perhaps the most compelling case for Social Security reform is the state of the Medicare program. Medicare’s unfunded liability is seven times greater than Social Security’s. A reformed Social Security system will ultimately lessen the tax burden, paving the way for much-needed Medicare reform.

[page]

The reform plan presented here is based on a time-honored principle: There is no such thing as a free lunch. While the benefits of reform are substantial, they cannot be realized without sacrifice. Realistically, today’s generation of workers face a double burden: They must secure the retirement benefits of their parents and simultaneously begin funding their own retirement benefits.

Who must sacrifice and how much? This reform plan makes explicit certain budget requirements that already exist implicitly: (1) expected Social Security surpluses over the next decade must be reserved exclusively for Social Security, and (2) the promises represented by the current Social Security trust fund must be redeemed.

“Initially, employees and employers would each contribute 1.25 percent of wages to fund the accounts.”

The projected surpluses and the redemption of the trust fund will provide part of the finances necessary to fund personal retirement accounts. The workers who expect to benefit must shoulder the rest of the burden of reform. In order to fully realize promised benefits, employees must be willing initially to set aside and invest an additional 1.25 percent of payroll (matched by their employer) rising gradually to 1.75 percent after eight years. Over time, private accumulations of assets will replace government promises. In fact, the youngest workers will be able to fully fund their own retirement without the need to impose any new taxes on future generations.

This plan stands in stark contrast to reform plans that require unrealistic and unspecified spending cuts, unrealistic and unspecified tax increases, and/or large amounts of federal borrowing. A reform plan that promises to pay scheduled benefits without new revenues requires significant reductions in other federal spending:

• As noted above, paying Medicare and Social Security benefits with no reform will require one-in-four income tax dollars by 2020 in addition to payroll tax collections and premium payments.
• A personal retirement account reform plan that does not include additional contributions from employees or their employers would require almost one-in-two income tax dollars by 2020!

The reform plan proposed here makes the costs of prepayment explicit, so that costs can be compared to the benefits.

[page]

“In return, workers are able to deposit payroll taxes that otherwise would have been paid to the government.”

Given the current federal budget environment, we sought to develop a Social Security reform plan that incorporates a personal investment component without requiring more government revenue than is already promised to Social Security through payroll taxes and the bonds in the Trust Fund. Our plan creates personal retirement accounts funded partly from existing payroll taxes and partly from a small additional investment by participating workers and their employers. [See the sidebar, “Proposal Outline.”]

Personal Retirement Accounts (PRAs). Workers who have not yet reached the retirement age will set aside part of their earnings in a personal retirement account (PRA). Specifically, PRA deposits will equal 10 percent of the first $7,650 in annual earnings, 3 percent of earnings between $7,650 and $55,000, and 1 percent of earnings above $55,000.² Thus for a worker who earns $7,650 per year or less, the PRA deposit will equal 10 percent of wages. For a worker who earns $27,000, the PRA deposit will equal about 5 percent of wages.

“For each dollar an averageincome worker invests, three dollars is invested by someone else.”

Contributions . In order to fund their PRA accounts, all workers will be able to divert a portion of their payroll taxes. Also, workers must initially make their own additional contribution of 1.25 percent of wages, to be matched by their employer. In general, lower-income workers will be able to invest more of their payroll taxes than higher-income workers. [See Table I.]

Roughly speaking, for each dollar an average-income worker invests in their PRA, three additional dollars will be invested by an employer and funds that otherwise would have been paid to the government. For each dollar invested by the lowest-income workers, seven dollars will be invested by an employer and by diverted payroll tax dollars. The PRAs of the highest-income workers will be almost totally funded by the employee and employer 1.25 percent contributions, with little or no payroll tax diversion.

“Employer and employee contributions to personal accounts would rise to 1.75 percent of wages after eight years.”

The worker’s contribution (matched by the employer) will gradually rise to 1.75 percent after eight years and payroll tax diversions will fall. [See Table II.] At this point, employer and government contributions fall to about two dollars for every dollar the average-income worker invests, and about five dollars for every dollar the low-income worker invests. The contribution rates will remain at these levels until 2038 when the current payroll tax rate is no longer needed to sustain the system.

A few examples for the first year of the program illustrate how the contribution rates would work:

• Example 1: A Low-Wage Worker. A worker who earns $7,000 per year contributes $700 = (10 percent x $7,000). The realized contribution rate for this worker is 10 percent. Of that amount, the worker contributes $87.50 (1.25 percent x $7,000) and the employer contributes another $87.50 for a total of $175. The rest, $525 (7.5 percent) is deposited by the government from diverted payroll taxes.
• Example 2: A Medium-Wage Worker. A worker who earns $35,000 per year has earnings between the first and second thresholds and contributes $1,585.50 (10 percent x $7,650 + 3 percent x ($35,000 – $7,650). The realized contribution rate for this worker is 4.53 percent. The worker and employer each contribute 1.25 percent and the government contributes 2.03 percent.
• Example 3: A Higher-Wage Worker. The individual who earns $60,000, above the second threshold of $55,000, contributes $2,235.50 (10 percent x $7,650 + 3 percent x ($55,000 – $7,650) + 1 percent x ($60,000 – $55,000)). This contribution results in a realized contribution rate of 3.73 percent. The worker and employer each contribute 1.25 percent and the government contributes 1.23 percent.

In future years the two income thresholds will rise by the growth in the Social Security Wage index. With this contribution rate structure, the average contribution is 5.14 percent.³

Easing the Transition. To make the transition easier, for the first five years of the program employees and their employers could divert monies currently sent to defined contribution retirement plans — such as 401(k)s — to meet their 1.25 percent PRA contribution requirement. Additionally, small businesses could be allowed a year’s delay before matching their employee’s contributions.

Funding the Personal Accounts. There are three current dedicated revenue sources for Social Security: payroll taxes, taxes on benefits and the Trust Fund. While the Trust Fund does not provide additional revenues to the Treasury, it does represent a commitment by the Treasury to provide resources to Social Security. In our reform, we require the Treasury to honor its commitment and aid in the transition to a retirement system with individually-owned retirement accounts. This year, Old Age and Survivors Insurance (OASI) revenues will exceed spending by an amount equal to 1.56 percent of payroll. The surpluses are expected to grow to 2.01 percent by 2008 and will continue until 2017. Thus, for the next 4 to 5 years, the government can contribute to PRAs without significant draws on the Trust Fund.

“Total deposits to low income workers’ accounts will equal about 10 percent of wages.”

Investments. Workers will not be able to buy and sell individual stocks and bonds with their PRA funds. Instead, they will be able to invest in approved diversified funds that reflect the economy as a whole. They also will have investment options, including stock index funds, bond funds and government securities funds. The management of these funds will be subject to strict accounting and financial standards.

Administration. To avoid creating additional burdens for employers, all the administration and paperwork could be done internally at the Social Security Administration. Employers would send employee and employer contributions to the government, just as they do under current law. The employees would make their investment selections with the government — not with employers, who bear no additional administrative burden. However, firms already administering defined contribution plans would be allowed to make deposits directly to PRAs on behalf of their employees, just as they do now with their 401(k) plans.

“Administrative costs can be minimized.”

It is reasonable to expect firms managing personal retirement account assets to receive compensation for their services; after all, there will be some 148 million accounts to manage. By limiting options and structuring the accounts carefully, administrative fees could be reduced.⁴

• The 1994-1996 Advisory Council on Social Security assumed administrative costs of 0.105 percentage points for the Individual Accounts option.⁵
• The President’s Commission to Strengthen Social Security assumed administrative costs of 0.3 percent.⁶

Retirement Benefits. Over time, the Social Security benefits paid by the government to retired workers who participate in the PRA system will be gradually reduced by a predetermined formula. In general, these reductions will be offset by increases in PRA account balances. During retirement, individuals will receive two monthly checks — one from the Social Security Administration (as under the current system) and one based on their private accumulation.

Annuitization. At the time of retirement, individuals will use their PRA funds to purchase annuities. If the sum of the annuity check and the Social Security check equals at least 150 percent of the poverty level, any surplus PRA funds may be used for other purposes — including certain tax-free health care expenses.

For the purpose of the simulations reported here, we assumed that all workers annuitize 100 percent of their retirement-age PRA accumulation. The unisex life table for each birth cohort was used to calculate the annuity amounts. This implies that no benefits are awarded to the annuitant’s survivors once the annuity has been purchased, but because each spouse has an account, the surviving spouse continues to receive annuity payments. However, annuitants could also choose an array of options including those with survivor’s benefits. (In Chile, joint annuities are required.)

“Workers will purchase annuities with their account funds at retirement.”

Periodic Withdrawals. The Chilean system of privately-owned individual accounts illustrates a successful way to handle personal retirement account payouts. Retirees in Chile can choose to purchase an annuity or make programmed withdrawals from their personal account. Workers choosing annuities receive an inflation-protected guaranteed income for life, but forgo the right to leave a bequest to heirs. Workers choosing programmed withdrawals leave the account with a pension fund manager and withdraw an amount each year set by law. Retirees can leave a bequest to their heirs, but run the risk of exhausting the account before they die. Regardless of the option chosen, the government provides a minimum benefit guarantee to all workers who have contributed to the system for at least 20 years.⁷

Government-Funded Benefits During the Transition. Workers close to retirement (those between 51 and 64 years of age, inclusive) will have their scheduled taxpayer-funded Social Security benefits reduced by 1 percentage point for each year between their current age and the age of 64.⁸ For example, 60-year-old workers will receive 96 percent of currently scheduled benefits from Social Security plus the annuities from their PRAs. Funds invested in the PRA earning a real 3 percent rate of return (the rate of return that is assumed to be earned on the Social Security Trust Fund bonds) will produce an annuity equal to 5 percent of scheduled benefits. Thus, the total benefit would equal 101 percent of scheduled benefits. Similarly, 55-year-old workers will receive 91 percent of currently scheduled benefits from Social Security, plus annuities equal to 9 percent of scheduled benefits, with their PRAs again earning a real return of 3 percent — a total of 103 percent of their anticipated benefits.

For workers 20 to 50 years of age, benefits will continue to accrue according to the current law schedule; but at the time of retirement, these workers will receive a preset proportion of these scheduled benefits.⁹ For example, upon reaching full retirement age, workers 50 years old at the time of reform will receive 85.7 percent of the benefit they would have received under the current Social Security system. Workers 30 years of age will receive 28.6 percent of the benefit they would have received under the current system. New workers, 20-year-olds in 2004, will be entirely in the new personal account system.

The combination of an individual’s personal account annuity and their entitlement to a share of scheduled benefits will, on the average, equal total currently scheduled benefits.¹⁰ With these contribution rates and a 5.4 percent rate of return on PRA accumulations and a 3 percent annuity return, workers will fully replace their scheduled benefits.¹¹

Minimum Retirement Income. Any participant with at least 35 years of full-time participation in Social Security (cumulatively, both before and after the reform) will be guaranteed a retirement income of 150 percent of the poverty level for persons 65 years of age and over, adjusted for inflation.¹² If a participant’s personal account annuity plus his or her share of scheduled benefits falls short of this amount, the government will supplement the individual’s monthly income to reach the 150 percent level. Married couples who have full work histories will be guaranteed 150 percent of the poverty level for two-person households above the age of 65.¹³

“Annuities from personal accounts will eventually replace Social Security benefits.”

Another approach to minimum benefits can be found in the reformed Swedish pension program, which includes a private account, a notional account and the guaranteed benefit amount. The private account represents the portion of the program that is prepaid in that a small share of payroll taxes are invested in the market and accumulate for retirement. The notional account generates a formula based benefit, but the payroll taxes are used to fund current retiree benefits. The base guaranteed pension is funded by tax revenues and is awarded regardless of one’s work history. The guarantee amount is means tested, based on the size of the flows from the notional and private accounts. Minimum benefit payments in a system like the Swedish program would be paid through contemporaneous general taxes rather than through payroll taxes. They would replace Supplemental Security Income (SSI) and could be paid to all elderly. In such a two-part system, PRA annuities would be intertwined with these minimum benefits. Since PRAs are not funding the totality of one’s retirement pension, the required PRA contribution rate would be smaller. The sidebar further addresses the topic of guaranteeing currently scheduled benefits. [See the “Guarantees” sidebar.]

[page]

Because this proposal ultimately replaces currently scheduled defined benefits with defined contributions, we must address the issue of the current progressive benefit formula.

“The reformed program replicates Social Security’s progressive benefit structure.”

If the goal of reform is to replace scheduled benefits, different contribution rates by income levels are necessary, given Social Security’s redistributive benefit formula and differing lifetime earnings. Targeting scheduled benefits requires redistribution during the accumulation phase or at retirement. That is, we can have different rates of contribution allowing low-income workers to deposit more taxpayer dollars in their PRAs than high-income workers (as is proposed here) or we can have different payout rates with low-income retirees getting a higher benefit relative to preretirement wages than high-income workers (as occurs in the current system).

Figure II illustrates the relative outcomes using the differential contribution rates that we recommend. Our proposal basically replicates the outcomes of the current defined benefit formula for workers at different incomes who are 55 and 40 years of age. For workers 25 years of age in 2004, lower earners will be better off relative to the current system when compared to higher-income earners.

[page]

“Initially, the reformed program will cost more than the current program.”

Figure III shows two alternative annual costs of funding the OASI program. The status quo cost rate line shows the tax rate necessary to finance scheduled OASI benefits on a pay-as-you-go basis. By 2080 a 16.82 percent tax rate will be needed. The other line, the reformed cost rate, is equal to the sum of the PRA contributions plus the cost of paying the reformed defined benefits as outlined above. As the figure shows, by 2037, the reformed program’s cost falls below the cost of maintaining the program as currently structured.

“Ultimately, the reformed program will cost much less.”

Figure IV shows the relative cost of reform versus pay-as-you go financing, expressed as the difference between the reformed cost rates and the status quo cost rates. Every reform that attempts to prepay benefits imposes costs on current generations that pay-as-you-go financing would have imposed on future generations. The trade-off is clear in this graph. Between 2004 and 2036, the reform is more expensive than continuation of the status quo. However, in all years beyond 2036 the costs are lower. To appreciate the extent of the reduced costs we must imagine the graph continuing into the indefinite future. There are also real economic benefits of reform not captured in this graph. As discussed below, this reform will increase the nation’s capital stock significantly. The increased means of production will result in higher earnings for all future generations. Although we have not accounted for these higher earnings, they must be considered as a significant advantage of prepayment relative to pay-as-you-go financing.¹⁴

“The Trust Fund has a $17 billion balance in 2038, and thereafter no draws on the Trust Fund will be necessary.”

Figure V identifies the evolution of the Trust Fund in 2004 dollars. Between now and 2016, the Trust Fund actually rises as credited interest payments exceed draws. However, by 2038, the first year in which reformed costs are less than reformed revenues, the Trust Fund has a $17 billion balance, and thereafter no further draws on the Trust Fund will be necessary.

[page]

“After 2037, the payroll tax can be reduced. ”

The most important reason to reform Social Security is to avoid significantly higher taxes in future years and painful cuts in benefits. There are also other advantages.

Benefit of Reform: A More Secure Retirement. Just as the current system’s problems compound over time, the benefits of reform also grow. After about three decades, the reformed Social Security system will finance itself, and workers’ payroll taxes and contributions can be reduced. By contrast, without reform, future workers and retirees almost certainly face tax increases that will gradually rise to 50 percent above the current rate, or future retirees will face benefit cuts that gradually rise to a third of benefits.

Benefit of Reform: Taxes During Retirement. Individual deposits and payroll taxes invested in PRAs will be treated like Roth IRAs, in which contributions are made with after (income) tax dollars and withdrawals are tax free.

“Couples will split their contributions to personal accounts. ”

A common assumption behind conventional IRAs and 401(k) accounts is that people will be in a lower tax bracket after they retire. If so, they gain by being able to defer taxes until the time when their tax rate is lowest. However, because of the Social Security benefits tax, many lower- and moderate-income families face higher tax brackets after they retire.¹⁵ As a result, deferring taxes may actually increase their lifetime tax burden. The solution is to allow people to pay taxes during their working years and withdraw funds tax free during their retirement years. According to a recent study on the subject by Boston University Professor Lawrence Kotlikoff, “Every income group would benefit from taking advantage of this form of taxation. But it is especially beneficial to low- and moderate-income families who, if they save on a tax-deferred basis, can expect to face higher tax rates after they retire.”¹⁶ Since the employer contribution to the PRA is made with pretax dollars, the amount attributable to the employer’s contribution will be taxed upon withdrawal, at ordinary income tax rates.

Benefit of Reform: Accommodating Modern Family Life. Given the changing patterns of marriage and divorce it is important that a couple’s retirement savings be shared in some way. Today 50 percent of first marriages and 60 percent of subsequent marriages end in divorce. “Earnings Sharing” is one avenue by which both parties in a marriage share in the assets accumulated during the duration of the marriage. This would require dividing all of a couple’s contributions to personal accounts when they are contributed — crediting half to the husband and half to the wife. If they divorce, each spouse would retain ownership of his or her account.¹⁷

Benefit of Reform: Paying for Long Term Care. Long-term care expenditures are one of the most important retirement policy issues on the horizon for families and state-level policymakers. Many families do little to prepare for the costs of providing long term care for aged relatives. At the same time, long term care accounts for one-third to one-half of total Medicaid expenditures in most states, and Medicaid is one of the fastest growing components of state budgets. Getting long-term care spending under control will go a long way toward restraining state-based Medicaid spending.

Our reform plan will allow retirees to use their PRA balances, above the amount required for the minimum annuity, to purchase long-term care insurance and to pay for long term care directly. Tax free withdrawals also can be used for health care expenses for debilitating, end-of-life diseases.

“Prepaying retirement benefits will reduce taxes and increase incomes for future generations. ”

Benefit of Reform: Greater Economic Growth. One of the primary benefits of prepaying retirement benefits is the increase in the nation’s means of production resulting from higher savings. Investing funds in PRAs in the manner we recommend will create more saving and increase the nation’s capital stock. These higher savings will not be realized if the funds deposited in PRAs come from additional borrowing from the public. Increasing the nation’s means of production requires reducing the government’s debt, both explicit (in the form of government bonds) and implicit (in the form of elderly entitlement promises). Relying on borrowing to fund PRAs will lead to individualizing Social Security but not necessarily to higher savings.

[page]

Regardless of one’s position on the pros and cons of prepaying Social Security benefits, all agree that paying future scheduled benefits will require a greater share of the nation’s output. Prepaying retirement benefits — expecting each working generation to provide for some of its own retirement — has been part of the policy discourse since the program’s inception. Since government does not hold the requisite economic assets, prepayment must occur through the vehicle of personal retirement accounts.¹⁸

We suggest that all workers be allowed to invest a portion of their payroll taxes in PRAs, provided they and their employers make an additional contribution of their own. In return, they will avoid the current system’s inevitable path of benefit cuts and tax hikes, and participate in a retirement system where expected benefits are very close to those scheduled under current law.

Future retirees will receive part of their retirement benefits from their PRAs and the rest will be paid as a defined benefit. The defined benefit portion is equal to a percent of the worker’s currently scheduled Social Security benefit. The percent is scaled back over time in a way that recognizes the worker’s years of participation in the program prior to reform and ensures that workers close to retirement can replace their scheduled benefits by investing in conservative assets. Younger workers should achieve a retirement income comparable to the current Social Security benefits by investing in balanced portfolios. In addition, those who work full time for 35 years or more are guaranteed that the combination of their PRA annuity and their defined benefit will yield a retirement income at least 150 percent of poverty.

NOTE: Nothing written here should be construed as necessarily reflecting the views of the National Center for Policy Analysis or as an attempt to aid or hinder the passage of any bill before Congress.

[page]

1. The other dedicated revenues are taxes collected on Social Security benefits.
2. These thresholds will be adjusted with the growth in the average wage index.
3. A worker at the taxable maximum of $87,900 contributes $2,514.50. This contribution results in a realized maximum contribution rate of 2.86 percent. Of this amount, 2.50 percent is contributed by the worker and his or her employer, and the remaining 0.36 percent comes from payroll tax deductions. However, by the eighth year, when the combined employer and employee contribution rises to 3.5 percent of earnings, higher income workers will have higher taxes of, at most, 0.64 percent of earnings for those at the taxable maximum.
4. The 2001 Commission to Strengthen Social Security outlined several key characteristics of administering PRAs. The Commission suggested a two-tier structure in which deposits are initially collected and invested centrally. In this first tier, investment options would include balanced indexed funds and inflation protected bond funds. After a worker’s PRA reaches a stipulated threshold, more investment options would be allowed, provided the investments met safety and soundness rules established by a governing board. The Commission also suggested that asset allocations could only be changed once a year, that preretirement access to PRAs not be allowed, and that the governing board be independent from political pressure. See Chapter 2 of “Strengthening Social Security and Creating Personal Wealth for All Americans,” Report of the President’s Commission, December 2001.
5. Report of the 1994-1996 Advisory Council on Social Security, 1997, Volume I, p. 171.
6. “Strengthening Social Security and Creating Personal Wealth for All Americans,” Report of the President’s Commission, December 2001, p. 97.
7. See Estelle James, “Private Pension Annuities in Chile,” National Center for Policy Analysis, Policy Report No. 271, December 2004.
8. 
   The defined benefit from Social Security for workers 51 to 64 years of age in 2004 will be based on the formula: 
9. 
   Defined benefits from Social Security for workers 20 to 50 years of age in 2004 will be based on this formula:
10. The reform defined benefit formula is guaranteed in that, unlike current Social Security, the right to the benefit would be assigned to each individual. The formula also identifies the rate at which benefits, as scheduled under the current benefit formula, will be phased out. This particular way of reducing the defined benefit portion of the program allows the future behavior of those older than 20 years of age to affect their ultimate benefit, just as it does in the current system. Alternatively, the defined benefit portion of the benefit could be based on accrued benefits, such that benefits are earned solely on past participation in the program. A standard way of calculating current earned benefits is to calculate the disability benefit that a worker would receive based on past earnings and participation; see Stephen C. Gross, “Measuring the Solvency in the Social Security System,” in Prospects for Social Security Reform, Olivia S. Mitchell, Robert J. Meyers, and Howard Young, eds. (Philadelphia: University of Pennsylvania Press, 1999). This benefit is then multiplied by a proportioning factor equal to (age – 22)/40 and the result determines accrued benefits. This accrued benefit can form the basis for issuing recognition bonds to be given to individuals to “recognize” their past participation in the program. These bonds mature when the individual reaches full retirement and have a face value equal to the amount that would generate an annuity equal to accrued benefits. For purposes of our estimates, we use the scaled benefit formula to value the future annual costs of the reformed defined benefit portion of Social Security.
11. The simulations used in this report are limited to the Old-Age and Survivors Insurance (OASI) programs. PRA accumulations are expected to earn a 5.4 percent rate of return. The Commission to Strengthen Social Security assumed a 6.5 percent long-run rate of return on stocks and 3.5 percent on corporate bonds. Thus, a constant 70 percent stock/30 percent bond portfolio with administrative costs of 0.2 percent would produce a 5.4 percent rate of return.
12. New retirees who have worked full-time for 35 years can be identified in several ways. The Social Security Administration records a worker’s earnings and quarters of work credits. A worker is granted credit for a quarter of work if his or her earnings in Social Security covered employment exceed a set threshold. This year the threshold is $900. Thus, a worker earns a quarter of coverage for each $900 in earnings in 2004, for up to 4 quarters of coverage. At the current minimum wage of $5.15, a credit for a quarter of work would be granted for working about 175 hours and a full year of credits would result from 700 hours of work in a given year. Alternatively, adopting the normal definition of full-time / full-year work, equal to 35 hours per week for 50 weeks, will result in a threshold of 1,750 annual hours. Thus, an alternative monetary threshold can be calculated for each year by multiplying the minimum wage by 1,750 hours. This new monetary threshold would then be compared to a retiree’s work history to determine quarters of coverage and whether he or she qualifies for the guarantee. Based on data from the New Beneficiary Survey, 27 percent to 29 percent of new retirees in 1980 and 1981 would have had enough quarters of coverage under the full-time / full-year definition to qualify for the guarantee. If the reform is successful in matching a retiree’s Primary Insurance Amount, on average 2 to 3 percent of new retirees would exercise the guarantee.
13. We suggest a minimum benefit be provided for all those who participate through a normal work-life, defined as 35 years of full-time work. Considering that there are 47 years between 20 years of age and the normal retirement age (soon to be 67), requiring that participants work less than 75 percent of their available years does not seem onerous. Those unable to work will rely on other sources of welfare, like Supplement Security Income, the same sources that take care of them during their normal work years.Scaled back guarantees can be provided for participants with fewer years of active work. However, like the current Social Security program, the new program is not designed to replace contemporaneous transfers to low-income participants.
14. For ease of comparison, we have presented the costs of the reform as a percent of taxable payroll. However, as has been suggested by Kotlikoff in “Fixing Social Security and Medicare for Good,” presented at Improving Social Insurance Programs, University of Maryland, September 2003, a broader based tax, such as a consumption tax, produces smaller economic losses than does a payroll tax. It also distributes the burden of the tax across generations rather than only on workers.
15. Jagadeesh Gokhale and Laurence J. Kotlikoff, “Tax-Favored Savings Accounts: Who Gains? Who Loses?” National Center for Policy Analysis, Policy Report No. 249, January 2002.
16. Ibid.
17. Earnings sharing is not a perfect solution — retired widows who never worked or paid taxes would receive lower benefits. However, that problem will diminish with time; today, few women remain completely outside the labor market and the rate is likely to decline further in the future.
18. The costs of such accounts are similar in magnitude to the long-run costs of prepayment as calculated by the Trustees of the Social Security system. Therefore, PRAs are no more costly than assuring solvency through the Trust Fund mechanism.

[page]

Dr. Andrew J. Rettenmaier is the Executive Associate Director at the Private Enterprise Research Center at Texas A&M University. His primary research areas are labor economics and public policy economics with an emphasis on Medicare and Social Security. Dr. Rettenmaier and the Center’s Director, Thomas R. Saving, have presented their Medicare reform proposal to U.S. Senate Subcommittees and to the National Bipartisan Commission on the Future of Medicare. Their proposal has also been featured in the Wall Street Journal, New England Journal of Medicine, Houston Chronicle and Dallas Morning News. Dr. Rettenmaier is the co-principal investigator on several research grants and also serves as the editor of the Center’s two newsletters, PERCspectives on Policy and PERCspectives. He is coauthor of a book on Medicare, The Economics of Medicare Reform (Kalamazoo, Mich.: W.E. Upjohn Institute for Employment Research, 2000) and an editor of Medicare Reform: Issues and Answers (University of Chicago Press, 1999). Dr. Rettenmaier is a senior fellow with the National Center for Policy Analysis.

Dr. Thomas R. Saving is the Director of the Private Enterprise Research Center at Texas A&M University. A University Distinguished Professor of Economics at Texas A&M, he also holds the Jeff Montgomery Professorship in Economics. Dr. Saving is a trustee of the Social Security and Medicare Trust Funds. His research has covered the areas of antitrust and monetary economics, health economics, the theory of the banking firm and the general theory of the firm and markets. He has served as a referee, or as a member of the editorial board, of several major U.S. economics journals and is currently an editor of Economic Inquiry. Dr. Saving has authored many articles and two influential books on monetary theory. He has been president of both the Western Economics Association and the Southern Economics Association. After receiving his Ph.D. in economics in 1960 from the University of Chicago, Dr. Saving served on the faculties of the University of Washington and Michigan State University, moving to Texas A&M in 1968. Dr. Saving served as chairman of the Department of Economics at Texas A&M from 1985-1991. Dr. Saving is a senior fellow with the National Center for Policy Analysis.