The Impact of Social Security Reform on Women in Three Countries

Policy Reports | Social Security | Work & Wages

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

Appendix I: Methodology

Analysis of how women fare relative to men in the new and old social security systems is difficult for a number of reasons. First, the new systems have not been in effect long enough to be mature. That is, current retirees are subject to a mixture of old and new system benefits and we cannot be sure how someone fully under the new system will fare in the future. Moreover, we don't know the future rate of wage growth and rate of return on investments, which determine retirement accumulations. At the same time, longitudinal data from the past are not available. Thus we could not use actual employment histories of current retirees and workers to estimate their benefits under the new systems.

Representative Men and Average, Full-Career and 10-Year Women. We solved these problems by constructing synthetic men and women - using cross-sectional data on current behavior of people at different ages, educational levels and marital status to proxy the lifetime employment, wage and contribution histories of "typical" persons in each category. We then simulated how the average man and woman in each educational category would fare under the rules of the old and new systems, given these histories. Five educational levels are presented, ranging from incomplete primary to several years of postsecondary. We use education as a proxy for "potential permanent income."

This methodology assumes that age-specific labor force participation and wage behavior will remain constant through time, except for secular wage growth, for each schooling level. In reality, aggregate female labor force participation rates have been rising and will probably continue to rise. Female labor force attachment is strongly positively correlated with education levels, which have been rising dramatically. Changing social norms may lead to additional increases in female employment probabilities within each educational category. Moreover, the work incentives and disincentives in the new pension systems may alter work habits.

These potential changes in female labor force participation rates were not taken into account directly. However, in addition to the "average" woman in each educational group, we also calculated pensions for "10-year women" who work only 10 years, prior to childbearing and for "full career women" who have the same labor force participation and retirement age as men. Full career women give us an indication of the impact of increasing female labor force participation rates. The absence of longitudinal data meant that we could not vary wages as a function of experience, so the lifetime earnings and pensions of full career women are probably understated.

Our married men and women are assumed to be single until the median age of marriage in each country, and married thereafter. They marry within their educational class, and the average husband is three years older than the wife. We use the same wage and work profiles for single women, because of the small sample size for singles in the older age-higher education cells. Since single women probably have greater labor force attachment than married women, our simulations for full career women give us a rough approximation of their lifetime earnings and benefits. Thus, altogether we model five categories of men by education grouping and 15 categories of women - five education groups and three levels of labor force attachment - as well as distinctions by marital status for both genders. In constructing our synthetic men and women, we use national data sets for urban areas.18

Assumptions for Simulations. We use these employment histories to simulate the accumulations, annuities and public benefits that different groups of men and women can expect under the new and old systems. Accumulations and annuities in individual accounts are very sensitive to rates of return on investments and rates of wage growth. In our baseline simulations, we assume a "moderate growth" scenario in which economy-wide real wage growth is 2 percent per year and the real rate of return is 5 percent prior to retirement. The return during the payout stage is assumed to be 4 percent, given the likelihood that many will choose a lower-risk or fixed-rate annuity.19 Sensitivity analyses assuming a 3 percent real rate of return during the accumulation stage, 2 percent during annuitization and a 0 rate of wage growth were also carried out. The gender results in this "slow growth" case are very similar to the baseline, except that the relative role of the public benefit increases dramatically, especially in Chile, and therefore gender ratios are compressed. In this paper we focus on the baseline case, which produces less favorable gender results from the new systems.20

Differences in Investment Returns. Some evidence from the United States and other countries indicates that women may choose less risky portfolios with lower expected rates of return than men, in which case their accumulations and annuities would end up lower, leading to a lower gender ratio. However, in Latin America limited financial markets and regulations have given workers little portfolio choice. They can choose an asset manager, but all managers offer very similar portfolios. Thus, gender differences in portfolio risk and return are minimized. This may change, as Chile in 2002 started allowing differentiated portfolios and other countries may follow suit. For the United States, we simply note that any observed gender differentials are reduced once earnings differentials are controlled and may disappear once women acquire more financial experience. Moreover, the differential return would be much smaller if measured in risk-adjusted terms.21

Annuitization at Retirement. Although both gradual withdrawals and annuities are permitted at the payout stage, to impute a stable annual flow for purposes of analysis we assume that these accumulations are fully annuitized upon retirement. For simplicity in calculating the value of the annuity, we assume that these average people all have a certain lifetime, which corresponds to the national expected life spans. Life expectancies are differentiated by country and gender. Based on analyses of annuity markets in Chile and other countries, we assume a "money's worth ratio" of 100 percent, using a low-risk discount rate.22 Men and women are assumed to retire at the retirement age that is specified in each country - lower for women than for men in Chile and Argentina. In fact, in Chile most men retire from the system 5 to 10 years earlier than the "normal" age, which narrows the gender gap in pensions relative to that presented in this paper.

Present Value of Lifetime Benefits. Although we start by comparing monthly benefits, for the analysis of transfers and systems we shift to a comparison of lifetime benefits, since retirement ages changed as a result of the reforms, expected age of death varies by gender and country, and benefits from the joint annuity start flowing to widows late in old age. We calculate present value of lifetime benefits as of age 65, discounting at the same rate used for annuitization - 4 percent in our baseline case, 2 percent in the slow growth case.

Imputed Taxes and Net Public Benefits. We know each person's contribution to the individual account. However, often we don't know the future cost of the public benefit, its intergenerational burden or its gender incidence, since this is financed, in whole or part, out of general revenues. Our analyses of net redistributions (gross benefits minus taxes) are based on the simplifying assumptions that each cohort covers its own bill and that, within each cohort, the tax burden is distributed proportionally to lifetime earnings as proxied by lifetime annuities from the worker's own retirement accumulation.

The Counterfactual for System Comparisons. Comparing the new versus the old systems introduces an additional set of methodological problems. The old systems were actuarially unbalanced so could not have delivered their promised benefits. What, then, is the counterfactual to the new system? To handle this problem we focus on relative rather than absolute gains and losses to different gender-education-marital groups. Implicitly, this means we are comparing the reform with a counterfactual in which the fiscal imbalance of the old system is corrected in a distributionally neutral way - involving equi-proportional benefit cuts or tax increases for each group, while leaving relative positions unchanged.23 We also compare the old and new gender ratios to see whether these ratios improved or deteriorated due to the reform. [See Appendix Table VI.] In our comparisons of post- and prereform positions across educational groups, we normalize each group's ratio according to the ratio for men in the top educational group. [See Appendix Table VII.] Thus, we are measuring whether a particular group improved its position relative to highly educated men (and relative to each other); we are not measuring absolute gains or losses.

Description of Old Systems. In general, the old systems provided a benefit of the following sort:

B = aYS, where:

B = annual pension benefits, a = incremental benefit per year of work, Y = number of contributory years, S = average salary during last few years of work.

This formula provided a generous benefit for women who worked for only a short time and then withdrew from the labor market, because a was often very high for the first 10 years of work.24 Thus, in all three countries, the first 10 to 20 years of contributions produced a high replacement rate. Women were more likely than men to work for 10-20 years and then leave the formal labor market. Married women got a widow's benefit that was 50 percent of their husband's pension in Chile, 75 percent in Argentina and 90 percent in Mexico. Implicitly, unisex tables were used. Women could retire five years earlier than men with no actuarial penalty in Chile and Argentina. In Mexico a minimum pension protected low earners. All of these provisions favored women.

However, the old systems based their benefits on the last few working years, which favored men. A woman who worked at ages 20 to 30 before childbearing would earn no interest on her contributions and would find her pension based on wages that would appear to be very low compared with prevailing wages when she retired at age 60 to 65. (For example, with 2 percent real wage growth per year the average real wage rate would have doubled over that period.) In addition, using final years' salary as the reference wage especially favored workers with steep age-earnings profiles, who tended to be highly educated men. Furthermore, in Chile and parts of the Argentine system, a woman had to give up the earned pension to get the widow's pension, so women who worked much of their lives in the labor market got little or no incremental benefit from their contributions. Finally, the failure of the old systems to index for inflation was a major disadvantage to all workers, but especially to women. Pensionable salaries were based on past wages that were usually not indexed for inflation. Once a person retired, the initial benefit was usually not indexed for inflation. Yet these countries had such high levels of inflation that often their pensions were worthless. Ad hoc adjustments always lagged the pace of inflation. This especially hurt women who had worked for wages long in the past and received pensions long into the future.

The new systems eliminated these biases, both pro and con women. As discussed above, a woman keeps her own benefit as well as the widow's benefit from the joint annuity. The new public pillars are tilted toward low earners, who are predominantly women. Contributions to the individual accounts that are made in early adulthood add more to present value of lifetime benefits than contributions made in the final years, since the financial instruments in which they are invested generally earn a positive real rate of return over long periods. In Mexico the public benefit is price-indexed, in Chile it has been increasing faster than prices, on par with wages, and in Chile private annuities acquired upon retirement are price-indexed. How women versus men have fared given these complex changes is the empirical issue we have attempted to untangle.

Read Article as PDF