FPA Journal - = When Should Women Claim Social Security Benefits?

Executive = Summary

This paper summarizes = the=20 incentives facing older women when deciding when to = claim=20 their Social Security benefits. Nearly 60 percent of = women=20 opt for actuarially reduced benefits at age 62=97a = greater=20 percentage than men. Yet women are expected to live = longer=20 than men. Longer life expectancy would generally = suggest=20 delaying Social Security benefits. But the analysis = shows=20 that single women and married women face very = different=20 choices.
Married women are = entitled to=20 three types of benefits: (1) benefits based on their = own=20 earning records, (2) spouse's benefits based on = their=20 husband's earning records, or (3) survivor's benefit = equal=20 to 100 percent of their husband's benefits. These = benefits=20 are reduced if claimed earlier than the full = retirement=20 age.
From the authors' = study, two=20 key points emerge. First, in the case of survivors'=20 benefits, the husband usually can maximize the = benefits of=20 the couple or his surviving wife by delaying his=20 claim.
Second, the wife is = usually=20 better off claiming her own Social Security benefits = as=20 early as possible, though this can change depending = what=20 percentage her own benefits are of her husband's=20 benefits.
For most married = women, the=20 Social Security benefit structure actually = encourages them=20 to grab their benefits as soon as possible. While = this may=20 maximize the wife's Social Security "wealth," it = also=20 encourages them to withdraw from the labor force, = creating a=20 loss of earnings and 401(k) savings, and extending = the=20 period over which they need to support themselves in = retirement.

Alicia H. Munnell is = the Peter=20 F. Drucker Professor of Management Sciences in Boston=20 College's Carroll School of Management and director of = the=20 Center for Retirement Research at Boston College in = Chestnut=20 Hill, Massachusetts.

Mauricio Soto is a = senior=20 research associate at the Center for Retirement = Research at=20 Boston College in Chestnut Hill,=20 Massachusetts.

If individuals continue = to=20 withdraw completely from the labor force in their = early=20 sixties, a large and growing number will be hard = pressed to=20 maintain an adequate standard of living throughout = retirement.=20 Economic and demographic pressures are gradually = eroding key=20 sources of retirement income; at the same time = increases in=20 life expectancy mean that people can expect to live = for 20=20 years, on average, after they stop working. But = averages do=20 not tell the whole story. Nearly one-third of women = and almost=20 one-fifth of men will live into their=20 nineties.

Women's low wages, interrupted = work=20 histories, and role as caregivers make them especially = vulnerable in old age. A solution to the income = security=20 challenge, with a potentially enormous payoff, is for = women to=20 work longer. One important benefit of continued = employment=97in=20 addition to increasing current income, allowing = additional=20 401(k) contributions, and postponing the drawdown of=20 savings=97is avoiding the actuarial reduction, or = enjoying the=20 actuarial increase, in Social Security benefits. The = higher=20 benefits are particularly important for women who, on = average,=20 have much longer life expectancy. Unfortunately, = women, even=20 more than men, tend to claim Social Security benefits = as soon=20 as they become available. The question is,=20 why?

This paper summarizes the incentives = facing=20 older women when claiming their Social Security = benefits. The=20 analysis shows that single women and married women = face very=20 different choices. For most married women, the Social = Security=20 benefit structure actually encourages them to grab = their=20 benefits as soon as possible. These incentives = reinforce the=20 tendency for wives, who are usually younger, to retire = when=20 their husbands do. Early claiming may maximize the = wife's=20 Social Security "wealth," but it also encourages them = to=20 withdraw from the labor force, creating a loss of = earnings and=20 401(k) savings, and extending the period over which = they need=20 to support themselves in retirement.

A = note of=20 caution is warranted here. The retirement decision = entails=20 much more than maximizing the present discounted value = of=20 Social Security benefits=97the presence of health = insurance, the=20 value of leisure, and other sources of wealth and = income=20 should be taken into account. But the decision of when = to=20 claim benefits (whether individuals continue working = or not)=20 is one of the major financial decisions of older = households.=20 This paper indicates that there is an optimal claiming = strategy for married couples. For most households, = wives=20 should claim early and husbands should delay=20 claiming.

Early Claiming of = Social=20 Security Benefits

The existence of Social = Security's=20 earliest eligibility age (EEA) means that most = Americans have=20 an important choice to make once they turn 62: claim = reduced=20 Social Security benefits right away or delay until = some later=20 date and receive a larger benefit.=B9 The adjustments = in=20 benefits do not depend on the gender of the = beneficiary. The=20 reductions are approximately actuarially equivalent = for the=20 person with average life expectancy of the population. = Monthly=20 benefits are lowered by an amount that offsets the = longer=20 period for which they will be = received.=B2

The=20 precise amount of the reduction for early retirement = depends=20 on when benefits are claimed relative to the full = retirement=20 age (FRA). The fact that the FRA is rising from 65 to = 67=20 complicates the story. With an FRA of 65, a person who = claims=20 benefits at age 62 receives monthly benefits 20 = percent lower=20 than the full amount. The scheduled increase in the = FRA from=20 age 65 to 67 raises the actuarial reduction for = claiming=20 benefits at age 62 from 20 percent to 30 percent (see = Table=20 1).=B3 At the same time, the delayed retirement = credit, which=20 has been increasing, now provides an actuarially fair=20 adjustment for delaying claiming benefits beyond the = full=20 retirement age.⁴ Thus, on average, workers = are=20 expected to receive the same lifetime benefits = regardless of=20 when they claim benefits between ages 62 and 70. In = the=20 following discussion, age 66 will be used as the full=20 retirement age=97the benchmark for full = benefits.

3D"Table

In 2003, 59 = percent of women=20 and 53 percent of men opted to claim actuarially = reduced=20 benefits at age 62 (see Table 2).⁵ The fact = that a=20 greater proportion of women than men claim benefits at = this=20 age is somewhat surprising, since the women are = expected to=20 live longer than the men and the actuarial adjustments = are not=20 gender-specific.⁶ A longer life expectancy = means=20 that people will receive benefits for an extended = period of=20 time, which makes the increase in monthly benefits = from=20 postponing receipt more valuable (see Figure 1). That = is,=20 individuals with average life expectancy of, say, 81 = years,=20 will receive the same lifetime benefits if they claim=20 actuarially reduced benefits at 62 or full benefits = four years=20 later at 66. But women at age 62 have a life = expectancy that=20 is about three years longer than that for men. = Therefore, more=20 women than men are likely to live beyond the = "break-even age"=20 and would gain from postponing retirement past age 62. = In=20 other words, women can make up for forgoing = actuarially=20 reduced benefits at ages 62=9665 by claiming the full = benefit at=20 age 66 and enjoying it over a longer life = span.

3D"Table

3D"Figure

Using the = scheduled actuarial=20 reductions and delayed credits for individuals born = between=20 1943 and 1958 (the cohort for whom 66 will be the full = retirement age) and a 3 percent real discount rate = suggests a=20 break-even age of 81. The 1948 mortality tables = project a life=20 expectancy at 62 equal to 80.0 for men and 82.6 for = women.=20 With a 3 percent discount rate, men maximize expected=20 discounted present value of their own Social Security = benefits=20 by claiming at age 62; women reach the maximum by = claiming at=20 age 68, although the gain from delaying retirement is = less=20 than 2 percent of the present discounted value of = Social=20 Security benefits.

Thus, one would expect = a=20 smaller percentage of women than men to claim benefits = at age=20 62=97just the opposite of the pattern shown in Table = 2. Clearly=20 other factors are at play.

The Claiming Behavior of = Single=20 Women

One way to sort out the = puzzle is=20 to look at the retirement patterns of men and women by = marital=20 status. This information comes from the Health and = Retirement=20 Study (HRS), a nationally representative survey of = older=20 Americans.⁷ Table 3 reports the = distribution of=20 claiming ages for workers who claimed Social Security = benefits=20 between 1992 and 2002.

3D"Table

For single women, = the pattern=20 is as predicted. A significantly smaller percentage of = single=20 women claim benefits earlier than either married or = single=20 men. That makes sense. Women, on average, live longer = than the=20 break-even age, which means that they will enjoy the = higher=20 benefits they gain from postponing retirement for = enough=20 additional years to more than make up for forgoing = benefits=20 from ages 62=9664. Thus, it is perfectly sensible that = single=20 women are less likely than men to claim benefits = early.
But=20 what about married women? They are much more likely = than men=20 to claim benefits early. Are these women acting = irrationally=20 or are they responding to the incentives in the=20 system?

The Claiming = Behavior of=20 Married Women

Benefits = available for=20 married women. Married women are entitled to = one of=20 three types of benefits: (1) a benefit based on their = own=20 earnings record; (2) a spouse's benefit equal to 50 = percent of=20 their husband's primary insurance amount = (PIA)=97regardless of=20 the age at which the husband claims benefits=97if that = benefit=20 exceeds their own benefit; or (3) a survivor's benefit = equal=20 to 100 percent of their husband's actual benefit if = that=20 exceeds their own benefit (see sidebar, "Supplementary = Benefits Available to Married Women"). The percentages = are=20 reported at the full retirement age, and are lower if = the=20 benefits are claimed early. For example, a woman = claiming the=20 spouse's benefit at age 62 will receive only 35 = percent of her=20 husband's primary insurance amount rather than the 50 = percent=20 payable at age 66.

These three different = types of=20 benefits are shown in Figure 2 for a woman whose = unreduced=20 Social Security benefit equals 40 percent of her = husband's.=20 This example assumes that she and her husband are the = same=20 age, her husband will claim benefits at 65, she will = live to=20 90, and he will die at 80. The wife can claim benefits = from=20 her own earnings history starting at age 62, although = this=20 will cause a permanent reduction of her own benefits. = Once her=20 husband retires, she will be eligible for the 50 = percent of=20 her husband's benefit, which is reduced to reflect her = early=20 claim of benefits. Once her husband dies, she will be = entitled=20 to a survivor's benefit, which is mainly the = continuation of=20 her deceased husband's benefit.

3D"Figure

An equivalent = chart for the=20 household would just increase the section labeled = "spousal=20 benefit" to 140 percent. The illustration is about = women's=20 benefits to show that the survivor's benefits are an = important=20 consideration, often overlooked, for the decision of = when to=20 claim Social Security benefits.

Two = interesting=20 insights emerge from the stylized profile presented in = Figure=20 2. First, from the wife's perspective, the decision as = to when=20 to claim does not affect the survivor benefit. The = survivor=20 benefit is solely determined by her husband's earnings = history=20 and the actuarial reductions or increase to reflect = his early=20 or delayed claiming. That is, from her perspective, = the=20 survivor benefit is a fixed amount to be received = after her=20 husband's death. To maximize that fixed annual amount, = she=20 would like him to delay claiming. If the husband is = trying to=20 maximize the benefits received by the couple, delay is = also=20 his choice. Ironically, in this example, when the = husband is=20 deciding when to claim benefits, it is his wife's life = expectancy, rather than his own, that is the = determining=20 consideration.

Second, the decision the = wife does=20 have control over is when to claim the benefits she = receives=20 until the death of her husband. Her goal would be to = maximize=20 the present value of these benefits over her lifetime. = Again,=20 ironically, in this calculation her life expectancy = becomes=20 irrelevant; the relevant life expectancy is that of = her=20 husband. In this way, her choice mirrors that facing a = single=20 man. Because these benefits are expected to be = received for a=20 period shorter than the life expectancy of the average = person,=20 she has an incentive to claim as soon as=20 possible.

In reality, however, the = choices of the=20 household are not as straightforward as this stylized = example.=20 First, the year of death is unknown for both husband = and wife,=20 so a married couple's choice of when to claim benefits = depends=20 on the probabilities (P) assigned to four outcomes: = (1)=20 P(Husband Alive, Wife Alive), (2) P(Husband Alive, = Wife Dead),=20 (3) P(Husband Dead, Wife Alive), and (4) P(Husband = Dead, Wife=20 Dead). These are properly modeled in the following = section=20 (Table 4, p. 62). Second, the figure shows the = undiscounted=20 real benefits. Third, in the stylized example, the = survivor's=20 benefits are determined solely by when her husband = claims his=20 own benefits. This is also generally true, with the = notable=20 exception being when the husband dies before the wife = reaches=20 the full retirement age, the point at which the = reductions=20 shown in the sidebar would apply.

3D"Table

Maximizing the = Expected=20 Present Discounted Value of Future Benefits

If the main goal of = married=20 couples is to maximize the expected present discounted = value=20 of Social Security benefits, the joint decision as to = when to=20 claim benefits depends on their relative primary = insurance=20 amounts and the age difference between the = two.⁸=20 Table 4 shows the combination of ages at which married = men and=20 women should claim benefits to maximize the expected = present=20 discounted value of Social Security=20 benefits.

Each of these claim/age = combinations=20 maximizes the present discounted value of the expected = benefits that the household receives after the husband = reaches=20 age 62. Consistent with the actuarial adjustment of = Social=20 Security, the discounted rate is 3 percent real, and = for=20 mortality the calculations use the Social Security=20 Administration's 1948-cohort tables. The actuarial = reductions=20 and delayed credits are those for the cohort = 1943=961954 for the=20 main earner and spouse's benefit and 1945=961956 for = survivors.=20 The percent of households was estimated using the HRS. = The=20 remainder (9.7 percent) of households are those where = the=20 female PIA is higher than the male PIA. Both members = of the=20 household are assumed to be alive with a probability = of 1=20 until the husband reaches age 62. There are four = possible=20 outcomes: (1) P(Husband Alive, Wife Alive), (2) = P(Husband=20 Alive, Wife Dead), (3) P(Husband Dead, Wife Alive), = and (4)=20 P(Husband Dead, Wife Dead). For outcomes one and two, = the=20 calculations follow the rules of the program. For = outcome=20 three, the survivor's benefit reduction, given the = husband's=20 benefit, is calculated as the weighted average of the=20 reductions at each age combination before the wife = reaches the=20 full retirement age, given the probability that the = husband=20 will die before that point. This expected reduction is = then=20 multiplied by the actuarial reduction applied to the = husband's=20 PIA because of his age of claiming to obtain the total = survivor's (wife's) reduction relative to the = husband's=20 PIA.

For ease of exposition, these = calculations=20 assume that the primary insurance amount is unaffected = by the=20 age in which benefits are claimed. In reality, if = individuals=20 delay retirement, additional earnings would enter the = PIA=20 formula and might increase the PIA. This means, for = example,=20 that the wife should take into consideration the = expected=20 effect of an extra year of earnings on the ratio of = PIA. Then,=20 the decision of the optimal age to claim benefit = should be=20 based in the expected PIA ratio.

The = results=20 suggest that married men should in most cases delay = claiming=20 Social Security benefits until age 69.⁹ The = intuition is that the delay of claiming by the husband = increases the value of the survivor's benefits more = than it=20 reduces the man's own benefit because of his lower = life=20 expectancy.

In terms of the wife's = decision, the=20 optimal claiming age depends on her earnings relative = to her=20 husband's. If the wife's earnings produce a benefit = equal to=20 40 percent or more of her husband's, she should claim = benefits=20 as early as possible. The intuition is that once the = woman=20 reaches some moderate level of benefits, she maximizes = the=20 lifetime value of this component by claiming early = because she=20 receives these benefits only over the relatively short = expected lifetime of her husband. That is, her time = frame for=20 claiming her own benefits and her spouse's benefit is = the life=20 expectancy of her husband. Once he dies, she claims = the=20 survivor benefit, which is maximized by his claiming = at age=20 69. About half of all married women fall into this=20 category.

The next category is where the = wife's=20 earnings produce unreduced Social Security benefits = equal to=20 between 30 and 40 percent of her husband's benefits. = This=20 group accounts for 10 percent of all married women. = Again,=20 optimizing behavior requires that the husband defer = claiming=20 for at least a year in order to maximize the = survivor's=20 benefit. In terms of the wife's decision, if the age=20 difference is significant, she should elect to take = benefits=20 at the earliest possible age. Because she is so much = younger=20 than her husband, the span over which she can expect = to=20 receive her actuarially reduced small benefit based on = her own=20 earnings and her actuarially reduced spouse's benefit = is very=20 short. In other words, for this decision, she is like = a man=20 with a very short life expectancy. At the other = extreme, when=20 the wife and husband are the same age, the wife should = wait=20 until the full retirement age (here assumed to be 66). = The=20 intuition is that the couple will spend most of the = retirement=20 period together, and the goal is to maximize the = couple's=20 benefit=97the survivor's benefits become less = important. By=20 waiting, the wife will get 100 percent of her benefit = and,=20 more importantly, when her husband retires the = household will=20 receive nearly 160 percent of his PIA=9750 percent = spousal=20 benefit plus 108 percent for his delayed retirement = credit=20 from claiming at 67.

The final category = is where=20 the wife's primary insurance amount is less than a = third of=20 the husband's. In this case, the only meaningful = benefit is=20 the spouse's benefit. To claim the spouse's benefit, = her=20 husband must retire. If the age difference is = significant, the=20 husband should defer until age 68 so that the wife = will have=20 the largest benefit for her expected extensive period = as a=20 survivor. The wife again should claim as early as = possible to=20 maximize the benefits over her husband's relatively = short life=20 expectancy. At the other extreme, if the husband and = wife are=20 the same age, they will spend most of their retirement = together. In this case, the survivor's benefit becomes = less=20 important, so the husband should claim at the full = retirement=20 age, and the wife should again claim as soon as=20 possible.

In short, the foregoing = analysis=20 demonstrates that, in most cases, the structure of = Social=20 Security benefits makes it advantageous for married = women to=20 claim benefits as early as possible. When they have = meaningful=20 earnings, they can claim at 62. When they have = virtually no=20 earnings, they need to wait until their husband = retires to=20 claim. The question, as always, is whether people = really=20 undertake these detailed actuarial calculations before = they=20 make decisions. Therefore, it is worth considering an=20 alternative or reinforcing factor that may lead to = women=20 retiring early=97namely, joint = decision-making.

Joint=20 Decision-Making

Reinforcing the = incentives in the=20 Social Security program is the fact that husbands and = wives=20 appear to coordinate their retirement. That is, = married women=20 facing the prospect of retirement are likely to = coordinate=20 their withdrawal from the labor force with that of = their=20 husband. To the extent that their husbands are = generally=20 older, women may face pressure to retire earlier than = they=20 would otherwise. Interestingly, Social Security's = earliest=20 eligibility age (EEA) of 62 was established in 1956 = for women,=20 primarily as a way for husbands and wives, who were = presumed=20 to be two to three years younger than their husbands, = to claim=20 benefits at the same time.¹⁰

A = growing=20 number of studies have examined how husbands and wives = coordinate their retirement decisions.¹¹ = These=20 studies show that spouses tend to retire at the same = time,=20 generally because they want to spend time together. = That is,=20 when husbands and wives view the trade-off between the = costs=20 of forgone leisure and the benefits of increased = income from=20 paid employment, they place greater weight on the = leisure when=20 they can spend it with their spouses.¹² = Economists=20 have been unable to find any support for alternative=20 hypotheses, such as husbands and wives have the same = taste for=20 work and leisure or that they face the same financial=20 incentives.¹³

The fact that = couples=20 generally retire together reinforces the incentives in = the=20 Social Security program for married women to withdraw = from the=20 labor force at young ages.

Conclusion

Social=20 Security benefits are actuarially equivalent for an = individual=20 with average mortality=97in theory, individuals should = be=20 indifferent about when to claim benefits. The average=20 individual's life expectancy is somewhere between the = life=20 expectancy of men and the life expectancy of women. = Since=20 actuarial adjustments are not gender-specific, women = (who are=20 expected to live longer than the "average individual") = can=20 increase the present discounted value of benefits by = delaying=20 claiming. Men, on the other hand (since they live = shorter=20 lives than the "average individual") should claim as = early as=20 possible. But the main insight from this paper is that = there=20 is also an optimal claiming pattern for married = couples. And=20 it is not completely intuitive: to maximize the = present=20 discounted value of Social Security benefits for the = married=20 couples, males should generally delay claiming and = women=20 should generally claim as early as = possible.

To=20 the extent that the availability of Social Security = benefits=20 causes women to withdraw from the labor force earlier = than=20 they would otherwise, they face a less secure = retirement. The=20 inadequacy of retirement income will become an = increasingly=20 serious problem in the future as replacement rates = under=20 Social Security decline and retirees are forced to = rely=20 increasingly on accumulations in defined-contribution = plans,=20 such as 401(k) plans as opposed to traditional = defined-benefit=20 plans.

Endnotes

Under current law, = the=20 procedure to calculate initial benefits involves = four steps.=20 First, a worker's previous earnings are restated in = terms of=20 today's wages by indexing past earnings to wage = growth.=20 Second, earnings for the highest 35 years are then = averaged=20 and divided by 12 to calculate average indexed = monthly=20 earnings. Third, the Social Security benefit formula = is=20 applied to average indexed monthly earnings to yield = the=20 primary insurance amount (PIA)=97the benefit payable = at the=20 full retirement age. Finally, benefits are adjusted = to=20 produce permanently lower benefits for those who = claim=20 before the full retirement age, and higher benefits = for=20 those who delay retirement.
Benefits are reduced = by 5/9 of=20 1 percent for each month they are received prior to = the full=20 retirement age up to 36 months, and 5/12 of 1 = percent for=20 each month thereafter. This is equivalent to a 6.67 = percent=20 reduction for the first three years before the full=20 retirement age and 5 percent thereafter. Remarkably, = these=20 reductions have remained close to actuarially fair = despite=20 the changes in mortality and interest rates (Jivan=20 2004).
The full retirement = age is=20 scheduled to increase from age 65 to age 67 by 2022. = The=20 increase began with individuals born in 1938, for = whom the=20 full retirement age is 65 plus 2 months, and = increases 2=20 months a year until it reaches age 66. Then, after a = 12-year=20 hiatus, the full retirement age again increases by 2 = months=20 a year until it reaches age 67 for individuals born = in 1960=20 or later.
This paper uses the = term=20 actuarially fair "in the sense that the = total value=20 of the reduced monthly benefits that an average = worker could=20 expect to receive between age 62 and death is = similar to the=20 total value of the full monthly benefits that the = worker=20 could expect to receive over that time by waiting = until he=20 or she was eligible for full benefits." See = Congressional=20 Budget Office (2001).
These numbers = represent the age=20 at which different cohorts claim Social Security = benefits.=20 They should be interpreted as an approximation to = the=20 percent of individuals of a given cohort claiming = Social=20 Security benefits at a particular age.
Consistent with this = idea,=20 Delavande, Perry, and Willis (2006) show that = people's=20 beliefs about their personal survival chances have = an effect=20 on Social Security benefit claiming.
The Health and = Retirement Study=20 (HRS) is conducted by the Institute for Social = Research at=20 the University of Michigan. The HRS is a nationally=20 representative data set of about 12,650 individuals = from=20 about 7,600 households. This study began in 1992 by=20 interviewing people ages 51=9661 and their spouses = (regardless=20 of age). The survey has been re-administered every = two=20 years. For a detailed overview of the survey, see = Juster and=20 Suzman (1995).
The analysis here = assumes that=20 households maximize the net present value of Social = Security=20 benefits. They could, instead, maximize Social = Security=20 wealth net of payroll taxes. For example, Diamond = and Gruber=20 (1999) estimate Social Security wealth at different = ages net=20 of any payroll taxes paid after age 55. Assuming a = delayed=20 retirement credit of 5 percent, they find that = married=20 couples have incentives to claim at age 64 and = estimate=20 implied tax rates of around 50 percent for those who = delay=20 retirement after the full retirement age.
The fact that the = optimal=20 retirement age is 69 and not 70 is likely to be the = product=20 of an imperfect actuarial adjustment for delayed = retirement.=20 Our simulations conducted for single women show that = with an=20 8 percent actuarial credit, the optimal age to = retire should=20 be 68, despite their longer life expectancies. See = also=20 Coile et al. (2002).
The Congressional = Budget Office=20 reports the following: "According to Robert J. = Myers, the=20 Chief Actuary of the Social Security Administration = from=20 1947 to 1970, the reduction in the earliest = eligibility age=20 for women in 1956 resulted from a 'domino' effect: a = desire=20 to help widows under 65 and a desire to provide = immediate=20 spousal benefits to wives who were younger than = their=20 husbands resulted in pressure to reduce the = eligibility ages=20 for those types of benefits; then, it seemed unfair = to=20 require female workers to wait until a later age for = benefits than female non-workers. The pressure to = reduce the=20 eligibility age for men in 1961 resulted from the = high=20 unemployment of that period (CBO, 2000)."
Blau (1998), using = the=20 Retirement History Survey, found that among 30 to 40 = percent=20 of married couples the spouses left the labor force = within a=20 year of each other. Hurd (1988), using the Social = Security=20 Administration's New Benefit Survey, estimated that = in=20 one-quarter of couples the husband and wives retired = within=20 one year of each other. Johnson and Favreault = (2001),=20 looking at married couples in the 1998 wave of the = Health=20 and Retirement Study, calculated that between 22 and = 40=20 percent of husbands and wives retired within two = years of=20 each other.
Models estimated by = Gustman and=20 Steinmeier (2000) and Hurd (1988) support the = hypothesis=20 that husbands and wives view their own leisure and = that of=20 their spouse as complementary. Coile (2000), using = the=20 Health and Retirement Study, found that people not = only=20 respond to the financial incentives created by their = own=20 Social Security and employer-provided benefits, but = also to=20 spillover effects from their spouses' incentives. = Coile=20 interpreted these results as an indication that = spouses are=20 eager to coordinate their retirements.
Hurd (1988) and = Gustman and=20 Steinmeier (2000).

References

Blau, David M. 1998. = "Labor Force=20 Dynamics of Older Married Couples." Journal of = Labor=20 Economics 16: 595=96629.

Clements, = Jonathan. 2006.=20 "Squeezing the Most Out of Social Security." Wall = Street=20 Journal July 2, 2006.

Coile, Courtney, = Peter=20 Diamond, Jonathan Gruber, and Alan Josten. 2002. = "Delays in=20 Claiming Social Security Benefits." Journal of = Public=20 Economics 84, 3 (June): 357=96386.

Coile, = Courtney.=20 2003. "Retirement Incentives and Couples' Retirement=20 Decisions." Working Paper No. 2003-04. Chestnut Hill, = MA:=20 Center for Retirement Research at Boston=20 College.

Delavande, Adeline, Michael Perry, = Robert J.=20 Willis. 2006. "Probabilistic Thinking and Early Social = Security Claiming." Working paper. Michigan Retirement = Research Center. University of = Michigan.

Diamond, Peter=20 and Jonathan Gruber. 1999. "Social Security and = Retirement in=20 the United States" in Jonathan Gruber and David A. = Wise eds.,=20 Social Security and Retirement around the = World. The=20 University of Chicago Press.

Congressional = Budget=20 Office. 1999. "Raising the Earliest Eligibility Age = for Social=20 Security Benefits" January. Available at javascript:Handle= Link('cpe_9332_0','CPNEWWIN:child^toolbar=3D1,location=3D1,directory=3D0,= status=3D1,menubar=3D1,scrollbars=3D1,resizable=3D1@http://www.cbo.gov/ft= pdocs/10xx/doc1058/ssage.pdf');.

Congressional=20 Budget Office. 2001. "Social Security: A Primer" = September.=20 Available at javascript= :HandleLink('cpe_9332_0','CPNEWWIN:child^toolbar=3D1,location=3D1,directo= ry=3D0,status=3D1,menubar=3D1,scrollbars=3D1,resizable=3D1@http://www.cbo= .gov/ftpdocs/32xx/doc3213/EntireReport.pdf');.

Gustman,=20 Alan and Thomas L. Steinmeier. 2000. "Retirement in a = Family=20 Context: A Structural Model for Husbands and Wives."=20 Journal of Labor Economics 18: = 503=9645.

Hurd,=20 Michael D. 1988. "The Joint Retirement Decision of = Husbands=20 and Wives." NBER Working Paper 2803. Cambridge, MA: = National=20 Bureau of Economic Research.

Johnson, Richard = W. and=20 Melissa M. Favreault. 2001. "Retiring Together or = Working=20 Alone: The Impact of Spousal Employment and Disability = on=20 Retirement Decisions." Working Paper No. 2001-01. = Chestnut=20 Hill, MA: Center for Retirement Research at Boston=20 College.

Juster, F. Thomas and Richard Suzman. = 1995.=20 "An Overview of the Health and Retirement Study." = Journal=20 of Human Resources 30 (supplement): = S7=96S56.

U.S.=20 Social Security Administration. 2005. Social = Security=20 Retirement Planner. Available at javascript:HandleLink('cpe_9332_0','CPN= EWWIN:child^toolbar=3D1,location=3D1,directory=3D0,status=3D1,menubar=3D1= ,scrollbars=3D1,resizable=3D1@http://www.ssa.gov/planners');.

= U.S.=20 Social Security Administration. 2005. Annual = Statistical=20 Supplement, 2004. Available at javascript:HandleLink('cpe_9332_0','CPNEWWIN:child^toolbar=3D1,loc= ation=3D1,directory=3D0,status=3D1,menubar=3D1,scrollbars=3D1,resizable=3D= 1@http://www.ssa.gov/policy/docs/statcomps/supplement/2004/supplement04.p= df');.

U.S.=20 Social Security Administration. 2005. The = 2005=20 Annual Report of the Board of Trustees of the Old Age = and=20 Survivors and Disability Insurance Trust Funds. = Available=20 at javascript:HandleLink('cpe= _9332_0','CPNEWWIN:child^toolbar=3D1,location=3D1,directory=3D0,status=3D= 1,menubar=3D1,scrollbars=3D1,resizable=3D1@http://www.ssa.gov/OACT/TR/TR0= 5/tr05.pdf');.

University=20 of Michigan. 1992=962002. Health and Retirement = Study.=20 Available at javascript:HandleLink('cpe_9332_0','= CPNEWWIN:child^toolbar=3D1,location=3D1,directory=3D0,status=3D1,menubar=3D= 1,scrollbars=3D1,resizable=3D1@http://hrsonline.isr.umich.edu');.

Acknowledgment

The=20 authors would like to thank Peter Diamond for helpful = comments=20 and Roberto Medina for explaining Social Security = rules. The=20 authors also thank four anonymous reviewers for useful = comments.

Sidebar

You are here: = FPA Net > FPA Journal = > Past Issues & = Articles > 2007=20 Issues > =20 2007 June Issue - Article = 7=20	Printer=20 friendly view