Percent of Women Kept out of Poverty by Marriage, Other Family, and Social Security, by Age'

1
(No Transcript)
2
Population Aging and the Availability of Workers
to Support Retirees
3
Individual Aging and the Availability of Men to
Marry Women
4
The Social Security Administrations Biennial
Report on Income and Poverty of Older Americans

• Based on CPS data from March, 1998
• HRS 1998 is approximately one-third smaller than
  CPS in the 55 and over age group
• HRS has high-quality wealth data, and
  significantly better data on asset income (48
  have none in CPS, only 34 in HRS)

5
The Evolution of the HRS Sample Design
6
(No Transcript)
7
Percent of Women Kept out of Poverty by Marriage,
Other Family, and Social Security by Age.
8
Percent of Women Kept out of Poverty by Social
Security by Marital Status and Age.
9
Widow Poverty Rates by Age and Duration of
Widowhood
10
Poverty Rates of Widows, by Age at Widowhood and
Age(simulated profiles based on cross-section
regression)
11
Median Net Worth , by Age of Woman and Marital
Status
12
Net Worth of Widows, by Age at Widowhood and Age
(simulated profiles based on cross-section
regression)
13
Social Security Benefits by Duration of
Widowhood,Holding Age Constant

• Real Social Security benefits do not decline over
  time for an individual
• A negative correlation with duration must be due
  to lower benefit amounts when claimed
• Lower couple lifetime earnings, younger age at
  claim, earlier year of claim

14
Supplemental Security Income Benefits by Duration
of Widowhood, Holding Age Constant

• SSI is a means-tested program
• Increasing benefits with duration of widowhood
  indicates declining economic status

15
Findings from Cross-Sectional Analysis of HRS 1998

• The economic status of widows at any age is
  strongly negatively related to the length of time
  since husbands death
• This appears to result from a combination of
  younger women entering widowhood in worse
  economic status than older women, and a trend
  toward worsening status over time.
• Other interpretations of the same evidence are
  possible longitudinal analysis is needed.

16
Questions for Longitudinal Analysis Comparing
HRS and AHEAD

• Is the income-mortality relationship stronger at
  younger ages?
• Is there a greater economic loss associated with
  husbands death at younger ages?
• Is there deterioration in the economic status of
  widows, and is it greater at younger ages?
• How do living arrangements and family
  relationships respond?

17
Relative Risk of Husbands Death, by Family
Income as a Percent of Poverty
HRS 1992 to 1998, risk relative to HRS Married
Men at 300-600 of poverty AHEAD 1993 to 1998,
risk relative to AHEAD Married Men at 300-600 of
poverty
18
Relative Risk of Husbands Death, by Percentile
Ranking of Family Income Within Each Cohort
HRS 1992 to 1998, risk relative to HRS Married
Men at 50-75th percentile of married HRS incomes
AHEAD 1993 to 1998, risk relative to AHEAD
Married Men at 50-75th percentile of married
AHEAD incomes
19
Table 4. Transition Rates into Poverty by
Marital Status at Beginning and End of Period
HRS and AHEAD Compared With RHS (percent).
20
What Part of Retirement Income Security is Lost
at Widowhood, or Declines Afterward?

• Savings?
• Social Security?
• Pensions?

21
Changes in (mean) Net Worth by Marital Status
from 1992 to 1998 HRS Women (51-61 in 1992)

• New widows were similar to other married women in
  1992
• New widows lost both absolutely and relatively
• Continuing widows remained poorer than new.

22
Changes in (median) Net Worth by Marital Status
from 1992 to 1998 HRS Women (51-61 in 1992)

• Median wealth of new widows was lower than that
  of other married women in 1992
• New widows lost both absolutely and relatively
• Continuing widows remained poorer than new and
  gained little
• Only married women increased median wealth

23
Changes in (mean) Net Worth by Marital Status
from 1993 to 1998 AHEAD Women (70 in 1993)

• New widows had less initial wealth than other
  married couples, but more than widows
• All groups experienced increases in wealth, but
  new widows lost relative to other groups

24
Changes in (median) Net Worth by Marital Status
from 1993 to 1998 AHEAD Women (70 in 1993)

• New widows had less initial wealth than other
  married couples, but more than widows
• Continuing widows did not increase in median net
  worth
• New widows gained at about the same percentage
  rate as married women

25
Out-of-Pocket Medical Expenses Associated with
Husbands Death HRS vs AHEAD

• Out-of-pocket medical expenses were substantially
  higher on average at younger ages, not covered by
  Medicare
• Not shown here, medical expenses were as likely
  to be high at low incomes as at high

26
Changes in Social Security Benefits by Marital
Status from 1992 to 1998 HRS Women (51-61 in
1992)

• This cohort experienced rapid increases in
  numbers eligible for benefits.
• Relative to other groups, new widows gained less.

27
Changes in Social Security Benefits by Marital
Status from 1993 to 1998 AHEAD Women (70 in
1993)

• New widows and other married couples had similar
  Social Security benefits in 1993
• New widows lost about 35 of their couple
  benefits, only slightly more than the loss of
  spousal benefits
• Continuing widows had lower benefits in 1998

28
Changes in Pension and Annuity Income by Marital
Status from 1992 to 1998 HRS Women (51-61 in
1992)

• New widows had similar pension income as other
  married women in 1992.
• New widows lost pension income, while married
  couples and widows gained.
• By 1998, new widows were similar to continuing
  widows.

29
Changes in Pension and Annuity Income by Marital
Status from 1993 to 1998 AHEAD Women (70 in
1993)

• New widows had lower pension income in 1993 than
  other married couples.
• New widows lost both absolutely and relatively
• Continuing widows gained, but remained lower than
  new widows in 1998

30
Changes in Income of Other Family by Marital
Status from 1992 to 1998 HRS Women (51-61 in
1992)

• New widows and other married couples had lower
  income from other family members than widows in
  1992.
• New widows gained other family income, but
  remained lower than continuing widows

31
Changes in Income of Other Family by Marital
Status from 1993 to 1998 AHEAD Women (70 in
1993)

• New widows and other married couples had similar
  low levels of other-person income in 1993
• New widows increased other family income
• Continuing widows had much higher income from
  other family members in both years

32
Answers from Longitudinal Analysis Comparing HRS
and AHEAD

• Is the income-mortality relationship stronger at
  younger ages? Yes, a little, but
  differences are greater at high incomes than low
• Is there a greater economic loss associated with
  husbands death at younger ages?
  
  Yes, losses in wealth and
  pensions appear to be greater. Medical expenses
  associated with death contribute to wealth loss.
• Is there deterioration in the economic status of
  widows, and is it greater at younger ages?
• Perhaps, but the period 1992-98 is short,
  and unusually prosperous. Widows did not benefit
  nearly as much as married couples from the boom.

33
CONCLUSIONS

• The HRS is a valuable source for studying
  cross-section patterns and longitudinal dynamics
  of poverty and widowhood
• Poverty is worse among women widowed relatively
  young
• Loss of a husband in middle age deprives the
  woman of the wealth accumulation made by couples
  just before retirement, of some pension rights
  and accruals, and in some cases creates
  substantial uninsured medical expenses
• Changes to Social Securitys widow benefit
  formulas need to be carefully targeted to assist
  the groups at highest risk of poverty