Introduction
Social Security, the largest of all federal programs, is likely to undergo many legislative changes early in this century. Governments around the world, forced to deal with the aging of their populations and a corresponding growth in the percentage of adults depending on public transfer programs, must reconsider the nature of their social contracts. In the United States, Social Security expenditures will begin to exceed payroll tax revenues by increasing amounts as each wave of the baby-boom generation ceases to pay Social Security taxes and begins receiving Social Security benefits.
The need for reform has provided policymakers with an opportunity to rethink some of the choices that determined Social Security's original design. Yet the discussion often centers on how the system can accumulate additional savings—either by investing a portion of Social Security taxes in individual accounts or by building up government-held trust fund assets—without considering the more fundamental question of how Social Security benefits might be restructured.
Social Security was modeled on the single-earner, married-couple family. Although more common in the late 1930s, this type of household is not the norm today. Indeed, since Social Security's inception, sweeping social and demographic shifts have radically altered the structure of the American family. Nonetheless, Social Security law still largely reflects the realities of the 1930s. As a result, the program could soon spend hundreds of billions of additional dollars per year without providing a significant number of retirees and their spouses and survivors with a basic level of support.
Social Security, while never designed purely as an antipoverty program, was conceived largely to prevent poverty among the elderly, and it has proved instrumental in reducing poverty in this segment of the population. According to recent demographic and income projections, however, poverty rates among the elderly are likely to remain significant in the next couple of decades, even though real Social Security benefits are scheduled to rise substantially. Social Security already pays out more than enough per person to keep everyone out of poverty, even if some individuals have no other source of income. Yet Social Security falls far short of eliminating poverty and will continue to fall short, despite a planned increase of hundreds of billions of dollars in annual real spending over the next couple of decades. The failure to achieve greater poverty reduction in a system of this type generally stems from the structure of the benefit formulas, which over time fail to concentrate a larger share of resources among those with relatively greater need. The gap between Social Security's structure and the needs of today's families partly explains why the system does not target benefits optimally.
Three factors are contributing to the disconnect between Social Security and America's families. Most significantly, the share of low-income never-married and divorced mothers among retirees has risen. Many of these economically vulnerable women are not eligible for the Social Security income protections provided to widows and long-time spouses.
Second, the number of dual-earner married couples has increased. Roughly speaking, spousal and survivors benefits are add-on features that are not paid for with additional contributions or a reduction in the worker's benefit. A spouse essentially receives either his or her own earner's benefit or, if the amount is higher, a spousal or survivor benefit. This design creates a disadvantage for some couples today—married couples who have the same total household earnings do not always receive the same level of benefits. In general, households where each spouse contributes about the same earnings receive fewer benefits than households where one spouse contributes a larger share of earnings. The decline in the benefits of a two-earner couple relative to those of a one-earner couple with the same total earnings is greater the more evenly divided the couple's earnings are.
Today, most married couples have two earners, and the percentage of spouses with fairly equal earnings is increasing rapidly. Compared with previous decades, more women are earning higher education, entering the labor force, working more hours, and receiving wages comparable with those of men. Consequently, an increasing number of couples participate in a program that provides them with fewer benefits than it provides households with the same total earnings, yet where one spouse brings in all, or much of, the earnings. This unequal treatment goes against Social Security's goal of "horizontal equity," or providing roughly comparable benefits to workers who make the same contributions over their lifetimes.
Another factor that has contributed to the apparent disconnect between Social Security's family benefits and today's families is the work and savings disincentives built into spousal and survivor benefits. In many instances, workers are entitled to spousal or survivor benefits that exceed, or nearly equal, their own potential worker benefits. Hence, they receive little or no return on the payroll tax contributions they make to the Social Security system over their lifetimes. In contrast, the decision of a high-earning spouse to accumulate additional earnings (e.g., through a second job or a salary increase) usually adds to total benefits.
This problem is less about income adequacy—survivors and disability protections are still available—than about how the lack of return on significant payroll contributions might affect work decisions. With demand for different sources of labor likely to be high as baby boomers age and leave the workforce, now is a good time to minimize work disincentives. In fact, the aging of the population provides a strong argument for removing barriers to work. Households that devote extra hours to work typically earn more over their lifetimes and have more resources to support themselves in retirement. Their income tax contributions also reduce the strain on other taxpayers to support other government programs, such as education and defense.
With the system's fiscal balance in jeopardy, Social Security is likely to be reformed in the years ahead, with no one measure likely to achieve all the goals set out by reformers. To ensure meaningful reform at each stage, the Social Security debate must focus not only on creating more savings, but also on how well the system's structure serves the changing profile of the American family. In addition, benefits experts must better inform policymakers about the history of Social Security and reform options affecting the structure of family benefits.
This volume tracks the evolution of family benefits in Social Security and shows how changes in the retired population have affected the nature of these benefits as well as the system's ability to serve the elderly. It examines the current structure, adequacy, and efficiency of spousal and survivors benefits and evaluates specific reform proposals that could improve the living standards of the neediest Social Security beneficiaries. A look at social security systems in other countries, the potential effects of individual account reform options on women's benefits, and other related topics add important perspectives to the Social Security debate.
The History of Family Benefits
In "Family Benefits in Social Security: A Historical Commentary," Edward Berkowitz explains that the evolution of family benefits within Social Security can only be understood as part and parcel of a historical series of enactments pertaining to overall benefits. The Social Security Act of 1935 did not provide benefits for anyone other than workers contributing taxes. But in 1939, right before the government mailed out the first beneficiaries' checks, Congress extended benefits to spouses and survivors of Social Security contributors. In the program's early years, the government collected significantly more payroll taxes than it paid out in benefits. Family benefits were a politically attractive way to reduce this surplus.
Social Security's family benefits resemble pure transfers in that recipients are not required to make additional contributions. That is, unlike private pensions and life insurance plans, Social Security does not reduce a worker's benefit to fund spousal and survivors benefits. The program's formula also assigns larger benefits to higher earners' spouses and survivors rather than allocating benefits evenly among all spouses.
Little attention was paid to this particular design feature in the first few decades of Social Security. Early on, the legislative focus was how to raise spending and cover more workers, with less emphasis placed on ensuring that each dollar was equitably spent or on efficiently pursuing income adequacy goals for families. As time went on, the structure of family benefits took a backseat to legislative debates over provisions that held much larger financial consequences, such as expanding overall benefits (for example, instituting Disability Insurance benefits). Since the 1930s, changes to the structure of Social Security family benefits have been modest. For instance, certain divorced people eventually received coverage, and Congress set the eligibility age for spousal and survivor benefits to correspond to those for retired workers.
Modifying the basic structure of Social Security's family benefits will require significant legislative change. Political concerns have made it difficult to remove inequities, especially because reform typically means lowering one person's benefit while raising another person's. Social Security law uses language built around the idea of entitlement, making support for any reduction in promised benefits or obligations unpopular, even if such measures are necessary to restore equity.
The Changing Retired Population
While the structure of Social Security and family benefits have largely stayed the same, Americans' work, marriage, and education patterns have undergone dramatic changes. As Karen Smith reports in "The Status of the Retired Population, Now and in the Future," recent societal trends are reshaping the profile of the retiree population and the distribution of benefits.
In 1950, less than 40 percent of women aged 20 to 64 were in the labor force; by 2000, more than 70 percent of women held paying jobs. From the 1950s to the 1990s, marriage rates declined, and in recent decades nonmarital cohabitation has increased markedly. From the 1960s to the 1980s, the divorce rate more than doubled, though it has since leveled off. Also dramatic was the rise in the percentage of births to unwed mothers—from 5 percent of all births in 1960 to 33 percent in 1999.
These patterns affect all beneficiaries, but they have the greatest implications for the poor, who marry less and divorce more than the population as a whole. Social Security provides no spousal or survivors benefits to individuals married for fewer than 10 years, regardless of both spouses' earnings contributions to the household. In 1990, more than 63 percent of divorces occurred in marriages lasting fewer than 10 years, which suggests that Social Security does not provide shared benefits in the majority of marriages ending in divorce.1 Thus, many divorced women do not benefit from the antipoverty protection built into Social Security's spousal and survivor benefits. Black and Hispanic women, in particular, appear vulnerable, because their divorce rates, like those of the poor, are higher than average.
In the past, rising levels of Social Security benefits (and the maintenance of benefits' purchasing power through cost-of-living adjustments) led to a consistent decline in the poverty rate among the elderly. In the future, however, the rise in divorced and never-married individuals (particularly women) who do not qualify for spousal and survivor benefits could prevent the retiree poverty rate from declining significantly. A decrease in earnings gains among workers with the least education also clouds some future retirees' income prospects.
In addition to shaping spousal and survivor benefits, societal changes have important implications for worker benefits, especially among women. Sheila Zedlewski and Rumki Saha, in "Social Security and Single Mothers: Options for ‘Making Work Pay' into Retirement," examine how work trends will affect single mothers' retirement options. During much of the mid- to late-20th century, the rise in single parenthood was accompanied by an increase in the proportion of women who received welfare benefits and did not work. Following welfare reform in the mid-1990s, however, the employment rate of single heads of household rose rapidly. Within a short time, more single mothers began working, and as a group, they started working more hours than in the past.
Although this increased work effort means that more former welfare mothers will qualify for Social Security benefits, many women in this segment of the population will remain vulnerable to poverty. In particular, the authors' estimates show that single mothers who have spent more than 10 years on welfare will likely qualify for Social Security benefits equaling only 30 to 60 percent of the poverty level. These women (as well as a significant share of single mothers receiving welfare for less than 10 years) will need to turn to Supplemental Security Income (SSI), a social assistance program available to the elderly. But SSI denies benefits to anyone who has significant unearned income or assets and thus discourages workers from saving for retirement. In addition, benefits under SSI, while indexed to inflation, fail to keep pace with average wages and income levels.
Single mothers who never marry, or only marry for a short time, also do not qualify for the additional poverty protections Social Security provides spouses and survivors. As a consequence, these low-income working single mothers often pay more in taxes but receive fewer benefits than do other spouses and survivors.
A Comparison with Other Developed Countries
Policymakers considering how to implement reforms that best serve the retired population will benefit from understanding some of the features of systems in place in other countries. In "Social Security and the Treatment of Families: How Does the United States Compare with Other Developed Countries?" Lawrence Thompson and Adam Carasso look at the U.S. Social Security system and the pension systems of 15 other industrialized countries. The authors also consider how countries calculate benefits for different kinds of households and compare their benefit coverage with that of the United States.
Their analysis shows that the countries use a variety of approaches to balance the two primary goals of retirement programs—individual equity and social adequacy. Many use one public program to address social adequacy concerns and another program to allay individual equity concerns. The United States is unusual in that it relies on a single public program that uses a progressive benefit formula to balance both goals. On the whole, the U.S. system for individual workers (excluding family benefits) tends to be less generous than the other countries' systems, particularly for low and average earners. Supplemental benefits for spouses and survivors, however, tend to be higher in the United States, and thus play an important role in ensuring a minimum level of income protection for qualifying individuals. Although spousal supplements are fairly common in most of the other national systems' adequacy components, they are often structured as an additional flat benefit and, in some cases, are linked to income.
Benefits for surviving spouses tend to be related to earnings in countries that tie worker benefits to earnings (i.e., benefits rise in tandem with the worker's earnings). Still, as a percentage of a worker benefit, a survivor benefit is higher in the United States than in most other countries. As with spousal benefits, the U.S. system supplies relatively generous survivor benefits to offset the impact of its fairly modest worker benefits.
In most countries, survivors benefits reduce the surviving spouse's potential return on additional Social Security tax or contribution payments. The reduction is greatest in the United States as well as in countries with flat benefits, where survivors who work may get exactly the same benefit as those who do not. The more common approach in countries with an earnings-related component is to allow some combining of credits by spouses.
The countries also treat divorce differently. Several countries give each partner equal Social Security credit for earnings accrued during their marriage. Other countries, including the United States, extend spousal benefits only in divorces between couples that were married for a specified number of years. Countries that split the credits between spouses either use a standard formula or rely on court supervised divorce proceedings to determine the benefit. Where there is no flat benefit floor, splitting by itself can create serious adequacy problems, because many divorced spouses, especially women, have low earnings records.
A few countries have pension systems that deal with divorce through pure supplemental benefits (a divorced worker does not relinquish any benefits or make additional contributions to pay for a former spouse's benefits). These plans are subject to many of the same criticisms as the U.S. system. In fact, the U.S. program offers divorced spouses more adequate benefits while the worker is still alive than do pension systems in the small number of other countries using this approach.
Relative to the other countries studied, the U.S. public pension system is the least supportive of child rearing. Thompson and Carasso find that the United States is the only country that does not offer some direct form of compensation to people who temporarily leave the labor force to care for young children.
Assessing Current U.S. Law and Some Alternatives
Research on Social Security's family benefits has been fairly limited and, as in the political arena, the results have often been overshadowed by concerns about workers' primary benefits. Several significant exceptions are important to note, however. In the 1980s, researchers evaluated proposals that would share earnings or benefits between married partners, for example, by granting each spouse half the credit for the taxable earnings of the other (see Burkhauser and Holden [1982]; Congressional Budget Office [1986]; and Zedlewski [1984]). Ross and Upp (1993) provide an overview of the earnings sharing debate and speculate about the factors that led to its ultimate failure. Other family-related proposals surfacing over the years include child care dropout years and homemaker credits, but analyses of the equity and efficiency of these proposals have been fairly rare. (The studies of Iams and Sandell [1994; 1998] and Sandell and Iams [1997] are important exceptions.)
In "Social Security Benefits for Spouses and Survivors: Options for Change," Melissa Favreault, Frank Sammartino, and Eugene Steuerle build on the research that is available, suggesting a variety of ways to rigorously quantify and compare the effects of different reform proposals. The chapter's emphasis on specific tools and evaluation methods gives policymakers an alternative to the anecdotal evidence often trumpeted in particular reform agendas. Unlike many of the earlier studies cited, the chapter also considers how different proposals could be combined to offset the potentially negative effects of any single proposal.
The authors start with four proposals that merely increase benefits—either by raising survivors benefits, introducing child care credits, creating a new minimum benefit, or reducing the length of the marriage requirement for spousal and survivors benefits. They then evaluate three expenditure-neutral packages that combine different aspects of the benefit-increase proposals with benefit cuts for some retirees and assess the performance of the specific proposals (many others could have been chosen) against four efficiency and fairness measures, arguing that these tests should serve as benchmarks for all reforms:
- Income Adequacy. Combating poverty among the elderly is one of Social Security's explicit goals. The authors' specific measure of adequacy is the percentage of Social Security benefits that go to individuals with the lowest lifetime earnings.
- Equal Treatment of Equals. According to the principle of horizontal equity, Social Security law should treat individuals in similar circumstances the same. The authors quantify the reform proposals' performance against this measure by comparing benefit levels among couples with the same lifetime earnings but different relative shares of earnings, as well as among divorced individuals with marriages lasting for different intervals.
- Efficiency. In a system concerned with efficiency, additional taxes paid would bear a strong relationship to additional benefits earned. For each reform option, the authors examine whether spouses and low-earning individuals receive at least the same marginal return as higher earners on each additional tax dollar contributed to Social Security.
- Support for Single Heads of Household. Single parents typically have the least income protection in retirement and the worst rates of return among workers with equal earnings levels. The authors assess the reforms' effects on benefits for single heads of household who work and raise children relative to the level for spouses who do not work or raise children. In a sense, this test encompasses several principles, such as income adequacy and equal treatment of equals. Thus, if a provision does not meet the single-head-of-household test, the authors suggest it is not well designed.
Reform Alternatives and Income Adequacy
Congress added spousal and survivors benefits in part to help resolve income-adequacy problems among women. The benefits replace family income lost because of retirement, death, or disability of a worker (and reduce poverty in these workers' families). By the income-adequacy measure, existing spousal and survivors benefits are not well targeted. While the benefits are important for maintaining the income adequacy of many families, they give greater benefits to individuals married to richer people, less support to individuals married to poorer people, and no additional benefits to single heads of household.
Some observers contend that Social Security's spousal and survivor benefits are not meant to support marriage but rather to adjust for a family's child care responsibilities. In fact, the correlation between an individual's potential benefits and the amount of child care he or she provides is weak. For example, spouses and survivors who do not raise children are entitled to the same benefits as those who do.2
Examining some alternatives, Favreault, Sammartino, and Steuerle find that proposals that increase survivors benefits by themselves (without lowering spousal benefits at the same time) are also poorly targeted and do not increase the share of benefits to those with the lowest lifetime earnings. In contrast, certain minimum benefit proposals—the minimum benefit can be based upon years of work or have no work requirement—are better targeted and cost-effective in meeting an income adequacy standard among individuals with low lifetime earnings.
Reform proposals that provide limited earnings credit for years spent providing child care also tend to target benefits more precisely to individuals in need, although the effectiveness of these types of proposals varies over time, depending upon work patterns, particularly among women. In particular, the rapid growth in the number of working mothers means that even if child rearing were considered in reforming Social Security law, there is less connection between lack of access to workers' benefits and child rearing today.
Reform Alternatives and Equal Treatment of Equals
The proposals examined have different effects on Social Security's ability to meet the goal of equal treatment of equals. For example, consider a low-wage two-earner couple and a single-earner couple with the same total lifetime earnings. The authors show that some benefit-increase proposals, such as granting higher survivors benefits, would reduce the gap between the two couples' lifetime Social Security benefits. The effect of minimum benefit proposals would vary: under one scenario, the two couples would receive the same total benefits for the same taxes; under another, the two-earner couple who works more years would receive higher benefits than the one-earner couple. The results under the minimum benefit alternatives are not the same for couples at all income levels, making it difficult to generalize the results.
Reform Alternatives and Efficiency/Incentives to Work
A work-related minimum benefit increases lower-income spouses' returns on additional Social Security contributions and thus might add a significant work incentive to the system's structure. Introducing an enhanced, add-on survivors benefit would not add any incentive. And a child credit might discourage work, because the return on additional years worked would likely be zero if an individual could get the same, or a higher number of, Social Security credits without working.3
Reform Alternatives and the Single-Heads-of-Household Test
Single heads of household usually fare better under a system that provides minimum benefits. Under such arrangements, single heads of household can qualify for at least some of the additional transfers that only spouses and survivors gain under current law. If a minimum-benefit proposal requires many years of work, but a head of household works for just a few years, SSI might remain the more valuable retirement option. SSI benefits, however, are means-tested and do not increase with growth in economic productivity over time (increasing only with inflation), whereas the minimum benefits examined here would grow over time along with wages (and thus overall productivity). Introducing additional survivors benefits actually hurts single heads of household relative to others in the current system.
Establishing child credits for nonworking spouses creates mixed results, but the credits do the least for single heads of household who already work and thus receive no additional credit. These individuals often get more out of their own earnings contribution than out of an imputed minimum child care supplement. Zedlewski and Saha (this volume) find that minimum benefits with work requirements do the most for individuals who received welfare for a significant number of years.
Women's Benefits and Individual Accounts
One type of reform proposal receiving a great deal of recent attention goes beyond restructuring Social Security's defined benefits and recommends placing a portion of tax dollars in individual accounts whose assets would be invested privately—usually in a limited set of market options. These proposals, recommended by elected officials from both major political parties, raise important questions, such as how the accounts would be financed and administered, what factors determine the size and distribution of transaction costs, and how the strengths and weaknesses of individual accounts compare with those of more collective diversification approaches.4
Rudolph Penner and Elizabeth Cove focus on the potential impact of individual accounts on women's benefit levels given a reasonable set of economic and demographic assumptions. The authors look specifically at a "carve-out" proposal that diverts a portion of payroll tax payments into individual accounts.5 The analysis assumes that traditional benefits are cut proportionately by an amount sufficient both to eliminate the current deficit in the program by 2040 and to finance the diversion of payroll taxes into individual accounts. This type of proposal—a fairly draconian (and, in fact, politically unlikely) scenario in that it finances the accounts entirely through benefit cuts—serves as a useful benchmark for considering whether even the most drastic proposals would have adverse effects on women's retirement incomes. The authors also compare the carve-out option with proposals that "add on" to current-law payroll taxes to finance individual accounts.
The analysis assumes various rates of return on individual accounts and compares the total retirement income under the carve-out and add-on options to a reduced-benefit option, which scales benefits back sufficiently to eliminate the existing deficit by 2040 and does not contain individual accounts. For the carve-out proposal, a particularly interesting question is whether the individual account provides sufficient retirement income to make up for the further cut in traditional benefits necessary to make up for the trust fund's loss in payroll tax revenues.
Under a carve-out proposal that assumes a 5.5 percent net return on individual account investments (after transaction costs), married couples in which the wife works fewer than 25 years do very well compared with the reduced-benefits proposal. Among these couples, annual income would be about one-fourth higher, and nearly all the couples would be better off.
Under a 4 percent rate of return, the majority of couples still do better under the individual account carve-out than under the reduced-benefit option. Even in the bottom quintile of lifetime earnings, about three quarters of couples do better under the reformed system. The reform option, however, looks less favorable as the assumed rate of return is scaled back. The carve-out assuming a 2.7 percent return produces lower benefits than does the reduced-benefit option for more than 70 percent of couples. Still, among those who do worse under the carve-out option, the average benefit is only slightly lower.
In looking at results for nonmarried women, the authors find that most fare better under the carve-out option. For widows, the projections are about as favorable as they are for married couples. For divorced women, the story is more mixed, but under an assumed 4 percent return, the majority of divorced women examined still benefit. As with married couples, however, the 2.7 percent return produces less favorable results. In addition, at the 2.7 percent rate, fewer never-married women than divorced women benefit.
Overall, an individual carve-out would likely help boost women's retirement income relative to a reduced-benefits option if account returns matched historical levels. The lack of improvement in the retirement prospects of some lower-income women, however, suggests that a minimum benefit might be needed to help meet distributional goals and that a single mechanism, such as individual accounts, cannot solve all of Social Security's problems. Penner and Cove also review research on whether the investment behavior of women differs radically from that of men and conclude that it does not.
Property Rights under Individual Accounts
Individual account proposals have prompted much investigation into the riskiness of certain investment strategies, but almost nothing has been written about the property rights issues attached to such proposals. Upon distribution of an account, important questions of ownership will inevitably surface. Family events such as divorce and death could lead to the early division and distribution of account assets, considerably reducing available retirement income.
Pamela Perun points out that any property rights system for individual accounts must balance an individual's retirement income needs and his or her family members' support needs. Legally, however, there is no sure or standard way to find that balance. The property rights system attached to individual accounts will depend on how legislators answer a series of questions, including, Who should have rights to the assets? At what point should individuals be permitted to exercise those rights? And should account owners' and beneficiaries' rights be restricted?
As it stands, current law treats property rights inconsistently. Social Security uses federal laws to enforce Social Security property rights the same way in all states. But many other property disputes are resolved by state laws that vary widely in how they divide assets among divorced spouses. Given the complexity of state law and the increase in family mobility, Perun concludes that property rights for individual accounts should be based on a coherent, consistent set of federal laws. She reviews several ways of constructing property rights, drawing on existing models, including the private pension system, community property, earnings sharing, IRAs, and common-law approaches.
Perun also discusses the critical decisions that will shape the system. For example, during marriage, should a spouse have no rights over the account, some rights over the account, or equal rights over the account? Current law inconsistently does all three, depending on whether the asset is an IRA, a private pension, or a financial asset in a community property state. If individual accounts become part of the system, lawmakers will need to resolve the question of spousal access, along with many other property rights issues for spouses, survivors, and divorced individuals.
Timing of the Retirement Decision
As more people live into their eighties and nineties and collect benefits for nearly a third of their lives, the factors that surround the choice to retire could provide important clues to optimizing individuals' work efforts and their contributions to the Social Security system. In "The Family, Social Security, and the Retirement Decision," Melissa Favreault and Richard Johnson examine the factors that contribute to a married individual's decision to retire and consider how Social Security reform might influence retirement timing. They find that recent retirement of a spouse, health and functional impairments, and overall retirement wealth (Social Security entitlement, private pension wealth, and other savings) help determine when an individual will retire.
The high rate of joint retirement apparent among husbands and wives is likely to be less relevant in future decades because of women's increased labor force participation. Greater career investment and more employer incentives may make women less likely to follow their husbands into retirement. Their increased retirement wealth also suggests that they will have greater flexibility in choosing their retirement dates. The importance of a spouse's health, however, underscores the fact that marriage itself provides a form of social insurance. While one member of a couple can increase his or her labor supply in response to a spouse's health emergency, single workers do not have this extra economic protection. Accordingly, expanding the marriage bonuses implicit in the Social Security system through increased spousal and/or survivor benefits would not efficiently target benefits to the most vulnerable among the aged. Options to reduce spousal benefits in tandem with expansions of survivors benefits could, however, increase work incentives and thus delay retirement for recipients of Social Security's spousal and survivors benefits (a group overwhelmingly made up of women).
Establishing Standards for Reform
In reforming Social Security to ensure the system's long-term solvency, elected officials and beneficiaries know that there is no simple solution. The reality is that unless taxes are increased significantly, reform almost inevitably will reduce the rate of growth in future benefits simply to restore fiscal balance. This fact creates a difficult legislative road for politicians. The challenges of restoring fiscal balance, however, also present legislators with an important opportunity to address fundamental concerns with the system and to better meet the retirement needs of an aging population. Lawmakers should strive to improve Social Security so that it adheres to principles of fairness and efficiency. In the difficult process of reform, it would be a mistake to ignore features of Social Security that increase real spending annually by hundreds of billions of dollars but provide little additional antipoverty protection, penalize couples simply because both partners have significant earnings, and give minimal protection to many single and divorced mothers who work and raise children, while providing large transfers to individuals who do neither.
Direct benefit increases to one group, the establishment of individual accounts, or other single-goal provisions might achieve worthwhile objectives, but each measure by itself is not enough to correct all the problems or to achieve all the goals of Social Security. To meet various efficiency and antipoverty objectives, policymakers must consider reform "packages"—combinations of proposals that incorporate the positive effects of certain changes, while mitigating or eliminating their negative effects. If reformers on all sides of the debate are willing to confront tough choices and apply rigorous standards, Social Security's family benefits can be restructured to meet the needs of the modern family.
Notes
1. Of course, some individuals who divorce would not qualify for spousal benefits under current law even if their marriages had remained intact for 10 years or more, because the benefits for which they qualify on the basis of their own earnings exceed the spousal benefits to which they would be entitled. On the other hand, these individuals are still likely to lose out on survivor benefits unless their lifetime earnings exceed those of their former partners. Another complication is that many who divorce remarry and gain entitlement to Social Security spousal and survivor benefits through the subsequent marriage. Regardless, the availability of additional benefits is somewhat arbitrary in its distribution among divorced couples and between those who are or are not divorced.
2. Some might argue that Social Security's practice of averaging earnings over the 35 highest earning years in one's career—thus permitting one to drop out the five lowest earnings years—is a more universal type of subsidy for child rearing. However, the benefit also subsidizes graduate students, early retirees, and many others who do not raise children during their years out of the workforce.
3. Many young people making decisions about whether to work would weight their current income needs more heavily than the marginal changes to pension benefits that they will claim several decades into the future.
4. The already extensive literature on these topics continues to increase. For an overview, including a discussion of the distinction between privatization, prefunding, and diversification, see Geanakoplos, Mitchell, and Zeldes (1998). For a discussion of administrative issues associated with individual accounts, see, for example, Thompson (1999). The analysis does not factor in potential macroeconomic feedback and assumes that earlier transition costs have been completely financed by 2040. In 2040, costs that remain are financed by benefit reductions.
5. The authors assume that the reform requires participants to purchase an annuity at the normal retirement age. They also make the following assumptions: participants pay a unisex rate for the annuity, and, if married, must purchase a joint-survivor annuity. Spouses split accounts upon divorce, and survivors inherit accounts upon workers' deaths if death occurs prior to the normal retirement age.
References
Burkhauser, Richard V., and Karen C. Holden, eds. 1982. A Challenge to Social Security: The Changing Roles of Women and Men in American Society. New York: Academic Press.
Congressional Budget Office. 1986. Earnings Sharing Options for the Social Security System, January. Washington, D.C.: Congressional Budget Office.
Geanakoplos, John, Olivia S. Mitchell, and Stephen P. Zeldes. 1998. "Would a Privatized Social Security System Really Pay a Higher Internal Rate of Return?" In Framing the Social Security Debate, edited by R. Douglas Arnold, Michael J. Graetz, and Alicia H. Munnell (137–156). Washington, D.C.: Brookings Institution.
Iams, Howard M., and Steven H. Sandell. 1994. "Changing Social Security Benefits to Reflect Child-Care Years: A Policy Proposal Whose Time Has Passed?" Social Security Bulletin 57 (4): 10–23.
———. 1998. "Cost-Neutral Policies to Increase Social Security Benefits for Widows: A Simulation for 1992." Social Security Bulletin 61 (1): 34–43.
Ross, Jane L., and Melinda M. Upp. 1993. "Treatment of Women in the U.S. Social Security System, 1970–88." Social Security Bulletin 56 (3): 56–67.
Sandell, Steven H., and Howard M. Iams. 1997. "Reducing Women's Poverty by Shifting Social Security Benefits from Retired Couples to Widows." Journal of Policy Analysis and Management 16 (2): 279–297.
Thompson, Lawrence H. 1999. Administering Individual Accounts in Social Security: The Role of Values and Objectives in Shaping Options. The Retirement Project, Occasional Paper No. 1. Washington, D.C.: The Urban Institute.
Zedlewski, Sheila Rafferty. 1984. "The Distributional Consequences of an Earnings Sharing Proposal." Project Report No. 3344. Washington, D.C.: The Urban Institute.
