The Impact of Social Security Reform on Women in Three Countries

Studies | Social Security | Women In The Economy

No. 264
Tuesday, November 04, 2003
by Estelle James, Alejandra Cox Edwards & Rebeca Wong

Implications for Social Security Reform in the United States

Individual account systems can improve relative outcomes for women and have done so in Latin America. But the relative gains to women are not inevitable; detailed design features matter. We can draw lessons from the three countries' experiences so as to improve the gender outcomes of a new multipillar system in the United States.

Social Safety Net. The purpose of the individual account is to require people to save for their own old age, while the safety net aims to redistribute income to those with a low capacity to earn and save. The progressivity of the current Social Security system in the United States is limited by lower expected lifetimes for low earners, a ceiling on taxable earnings for high earners, the absence of a minimum benefit and the presence of idiosyncratic redistributions. Reform provides an opportunity to strengthen the safety net and target redistributions more carefully. As in Latin America, low-earning women likely would benefit - if we do it right.

"Working women in the United States would benefit from personal accounts, low-earning women would benefit from minimum guarantee, and married women would benefit from joint annuities."

Annuitization. Annuitization, which provides a guaranteed income for life, is especially important for women in view of their greater longevity. This requirement - at least up to the poverty line - could be built into the individual account system. In Latin America lump sum withdrawals are allowed only for the amount left over after the retiree has purchased a pension that exceeds a specified floor.

Indexation of Annuities. Inflation insurance is especially important to women for similar reasons; otherwise the purchasing power of their pensions will fall substantially as they age. In the old Latin American systems, pension benefits were not price-indexed. This situation is changing in the new systems. During the accumulation stage account balances grow with the rate of interest, which is generally greater than the inflation rate. During the payout stage, annuities are price-indexed in Chile, and Mexico plans to begin price-indexing. This is facilitated in Chile by the prevalence of indexed bonds and other financial instruments. It will be more difficult and costly in Mexico because of the paucity of indexed instruments. The existence of long-term price-indexed bonds makes indexed annuities feasible and advisable in the United States.

Joint Pensions. Women often have low years of market work as part of an informal family contract in which the husband supports the wife in exchange for the time she spends caring for the family. In Latin America this contract is enforced by requiring that husbands purchase survivors' insurance before retirement and joint pensions upon retirement. This is an important requirement to build into any individual account system. An added bonus of joint annuities is that they defuse the contentious unisex issue. When purchasing individual annuities, unisex mortality tables produce higher payouts for women and lower payouts for men as compared with gender-specific tables, but in the case of joint annuities the choice of mortality tables makes very little difference. Since joint annuity pricing takes into account the combined lifetimes of husband and wife, similar monthly and lifetime payouts are produced whether unisex or gender-specific mortality tables are used.

"American women should be allowed to keep annuities based on their own earnings in addition to a joint annuity from their husband's personal account."

Own-Annuities and Joint Annuities - either or both? Financing spousal and widows' benefits from the common pool raises tax rates and penalizes working women and single men and women.16 In the United States and many other countries, working women must choose between their own benefit or the widow's benefit - they can't get both. Thus, women who work in the market for much of their lives pay substantial taxes with no incremental benefit. In contrast, in the new Latin American systems the widow keeps her own annuity as well as the joint annuity, which is financed by the husband. Market work by the wife is rewarded rather than penalized, yet this does not impose an additional burden on the public treasury.

Accounts and Annuities in Divorce or Cohabitation. At present the divorce rate is low in Latin America and divorce is prohibited in Chile (although marriages can be "nullified"). Thus, the systems we are analyzing do not contain detailed mechanisms for handling assets in or payouts from individual accounts in case of divorce. Legal protections regarding retirement accumulations and annuities would obviously be needed in the U.S. for divorced women and women cohabiting without formal marriage. These might include, for example, treatment of pension assets acquired during the marriage as community property in case of divorce and for legally recognized partnerships.17

Investor Education. Some evidence from the United States and other countries indicates that women may choose lower-risk portfolios with lower expected rates of return than men, in which case their accumulations and annuities would be lower, leading to a lower gender ratio. In Latin America little portfolio choice has been available until recently, so this issue has not arisen, but that is now changing in Chile, which will have to face the possibility of systematic differences in investment choices by gender and socioeconomic groups. The MPG, which sets a floor on pensions, will help to counteract excessive risk aversion among low earners. Investor education is essential so that women do not dissipate the gains from the new system by making overly conservative investment choices.

"Currently, Social Security provides a higher rate of return to nonworking married women than to single women or women who work."

Public Benefits for Working versus Nonworking Women. Women can have low lifetime earnings because they receive low wage rates or because they work few years. Should both these groups be considered equal targets for redistribution through the public pillar? Chile does not distinguish between these two causes in determining eligibility for its MPG, but Mexico does in allocating its SQ. In the current U.S. system, a higher rate of return is provided to those with low lifetime earnings, providing they have worked at least 10 years. The U.S. formula does not distinguish between people who have worked a little at high wage rates versus those who have worked a lot but earned low rates. Studies have shown that often the low earnings are due to low years of work rather than low wage rates.18 Furthermore, married women get a spousal benefit even if they have never contributed but single women must work to secure a benefit. Given two households with the same family income, the one with a wife who stayed at home gets a larger total benefit and pays lower payroll taxes than the one where both husband and wife worked in the labor market. We need to think through whether we are satisfied with this formula.

Indexation of the Public Benefit. Our simulations for Chile showed that if the MPG is price-indexed it becomes almost irrelevant in the future as wages and own-annuities grow in the face of a constant MPG. It will then be ineffective as a gender-equalizing instrument. This is part of the reason why very few pensioners receive the MPG in our simulations. In reality, the MPG has been increased on an ad hoc basis with wages over the past 20 years and is currently received by many women (as well as men). But a wage-indexed MPG is costlier and produces larger work disincentives than a price-indexed or nonindexed MPG. Similar issues arise in the debate over social security reform in the United States. One of the three proposals of the President's Commission to Strengthen Social Security would price-index the public benefit, instead of wage-indexing it for new cohorts, as is done now. A public benefit indexed to inflation would eventually become smaller relative to the average pension and the average wage and would therefore afford less protection to future cohorts of women and other low earners. An alternative proposal by the Commission would index the public benefit half to wages and half to prices, in order to maintain its relevance through time. But this implies a higher cost. Policy makers and citizens will have to evaluate this trade-off.

How we design our individual account system is up to us. The Latin American experience shows that a careful design will improve the position of women while also making our system more financially sustainable.

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.

Read Article as PDF