Housing Expenditures of Social Security Beneficiaries, 2005–2018

by
Social Security Bulletin, Vol. 81 No. 3, 2021

This article uses data from the public-use files of the Census Bureau's American Community Survey for selected years 2005–2018 to examine the annual housing expenditures of households that include at least one person who received income from Social Security, regardless of age. In all years, the median percentage of income spent on housing was higher in households that included at least one Social Security beneficiary than in households with no beneficiaries. In households with at least one Social Security beneficiary, the median share of income spent on housing varied by tenure. The median shares for renter households were 31.7 percent in 2005 and 32.5 percent in 2018. The median shares for homeowner households with a mortgage were 27.3 percent in 2005 and 25.1 percent in 2018. The median shares for homeowner households without a mortgage were 13.9 percent in 2005 and 12.4 percent in 2018.


Patrick Purcell is with the Office of Research, Evaluation, and Statistics, Office of Retirement and Disability Policy, Social Security Administration.

Acknowledgments: Thanks to Paul Davies, John Murphy, and Brad Trenkamp for helpful comments and suggestions.

The findings and conclusions presented in the Bulletin are those of the author and do not necessarily represent the views of the Social Security Administration.

Introduction

Selected Abbreviations
ACS American Community Survey
OASDI Old-Age, Survivors, and Disability Insurance

Housing is the largest single expenditure category for U.S. households. On average, in 2018, expenditures for housing were equal to 25.5 percent of household income among all U.S. households; among those with householders aged 65 or older, housing expenditures were equal to 32.8 percent of household income (Bureau of Labor Statistics 2020).1 Although spending for health care may receive more attention in the media, housing expenditures typically are more than twice the amount of out-of-pocket spending for health care among households headed by individuals aged 65 or older (Johnson 2015).2 Given the significance of housing expenditures in household budgets, trends in housing expenditures among Social Security beneficiaries are of potential interest to policymakers and the public.

Using data from the Consumer Expenditure Survey (CES),3 the Social Security Administration (SSA) has documented the expenditure patterns of Americans aged 55 or older in the Expenditures of the Aged Chartbook, most recently with data for 2015 (SSA 2018a). This article focuses on the housing expenditures of households in which at least one person received income from Social Security, regardless of age. The data are from the public-use files of the Census Bureau's American Community Survey (ACS). The ACS collects a wide range of demographic and economic data from a representative sample of more than 3.2 million U.S. households.4 With ACS data, research can examine topics such as housing tenure and expenditures and income sources and amounts, with detail by householder age, sex, marital status, and other demographic characteristics (Census Bureau 2014).5 This analysis covers selected years from 2005 to 2018. Those years were chosen because full implementation of the ACS began in 2005 and data for 2018 were the most recent available when this research began.

Housing Tenure, Household Income, and Percentage of Income Spent on Housing in 2018

In 2018, an estimated 121.5 million households resided in the United States, distributed among three categories of housing tenure: renters, homeowners with mortgages, and homeowners without mortgages. Chart 1 shows that renter households numbered 43.7 million (36.0 percent) and homeowners numbered 77.8 million (64.0 percent). Among the homeowner households, 48.2 million (61.9 percent) had mortgages on their homes and 29.6 million (38.1 percent) did not; respectively, those groups accounted for 39.6 percent and 24.4 percent of all U.S. households.6

Chart 1. Number of households, by tenure and presence of O A S D I beneficiaries, 2018. Stacked bar chart with tabular version below.
Show as table
Table equivalent for Chart 1. Number of households, by tenure and presence of OASDI beneficiaries, 2018
Presence of OASDI beneficiaries Rent Own with mortgage Own without mortgage All tenures
All households 43,712 48,171 29,614 121,496
No beneficiaries 34,473 36,002 12,706 83,180
One or more beneficiaries 9,239 12,169 16,908 38,316
SOURCE: Author's calculations based on ACS data.

Of the 121.5 million U.S. households in 2018, 38.3 million (31.5 percent) included at least one person of any age who received Social Security (Old-Age, Survivors, and Disability Insurance, or OASDI) benefits (Table 1).7 The 9.2 million renter households that included at least one beneficiary constituted 24.1 percent of all beneficiary households. Among households with mortgages, 12.2 million included one or more beneficiaries, constituting 31.8 percent of all beneficiary households. Homeowner households without mortgages in which at least one person received Social Security benefits numbered 16.9 million and constituted 44.1 percent of beneficiary households. Households with at least one Social Security beneficiary constituted 21.1 percent of renter households, 25.3 percent of households with mortgages, and 57.1 percent of homeowner households without mortgages.

Table 1. Number of U.S. households, and percentage distribution by householder characteristics and housing tenure: By presence of OASDI beneficiaries, selected years 2005–2018
Characteristic 2005 2010 2015 2018
All households: Number (in thousands) 114,682 114,542 118,178 121,496
  Households with one or more beneficiaries
Number (in thousands) 30,104 32,463 36,253 38,316
Percentage distribution by householder—
Age 100.0 100.0 100.0 100.0
18–61 20.4 20.4 19.0 17.6
62–69 26.9 29.2 30.6 29.4
70 or older 52.7 50.4 50.4 53.0
Marital status 100.0 100.0 100.0 100.0
Married couple 47.7 47.7 47.8 47.9
Unmarried male 14.9 15.8 16.7 17.1
Unmarried female 37.4 36.5 35.5 35.0
Race/ethnicity 100.0 100.0 100.0 100.0
Single race, non-Hispanic
White 80.0 79.1 77.3 76.1
Black 10.1 10.1 10.6 10.9
American Indian/Alaska Native 0.6 0.5 0.5 0.6
Asian/Pacific Islander 2.2 2.4 2.9 3.1
Two or more races, non-Hispanic 0.9 1.1 1.2 1.3
Hispanic origin, any race(s) 6.3 6.8 7.5 8.1
Percentage distribution by housing tenure 100.0 100.0 100.0 100.0
Rent 22.5 23.6 24.6 24.1
Own with mortgage 29.0 31.4 31.8 31.8
Own without mortgage 48.6 45.1 43.5 44.1
  Households with no beneficiaries
Number (in thousands) 84,578 82,079 81,926 83,180
Percentage distribution by householder—
Age 100.0 100.0 100.0 100.0
18–61 95.7 94.1 92.6 91.3
62–69 3.2 4.7 5.9 7.1
70 or older 1.1 1.3 1.5 1.7
Marital status 100.0 100.0 100.0 100.0
Married couple 53.8 51.8 51.0 51.1
Unmarried male 20.0 20.5 21.3 21.6
Unmarried female 26.2 27.6 27.7 27.3
Race/ethnicity 100.0 100.0 100.0 100.0
Single race, non-Hispanic
White 68.8 67.2 64.4 62.7
Black 12.3 12.4 12.7 12.7
American Indian/Alaska Native 0.7 0.6 0.6 0.6
Asian/Pacific Islander 4.4 4.7 5.4 5.8
Two or more races, non-Hispanic 1.2 1.6 1.9 2.2
Hispanic origin, any race(s) 12.5 13.5 15.1 15.9
Percentage distribution by housing tenure 100.0 100.0 100.0 100.0
Rent 37.0 38.9 42.3 41.4
Own with mortgage 51.4 48.9 43.5 43.3
Own without mortgage 11.6 12.1 14.2 15.3
SOURCE: Author's calculations based on ACS data.
NOTE: Rounded components of percentage distributions do not necessarily sum to 100.0.

The high percentage of beneficiary households among homeowners without mortgages indicates that the proportion of residents who have attained age 62—the earliest eligibility age for retired-worker benefits—is higher in these households than in the other tenure groups. Using data from the 2018 ACS, I calculate that the median age of homeowners without a mortgage was 66, and the median ages of homeowners with a mortgage and of renters were 53 and 44, respectively. The householder was aged 62 or older in 82.4 percent of households in which one or more people received income from Social Security in 2018. By contrast, the householder was 62 or older in only 8.8 percent of households with no Social Security beneficiaries.

Household income differed substantially across housing tenure groups in 2018. It was lower among renter households than among homeowners with or without mortgages; and among homeowners, it was lower for those without mortgages than for those with mortgages (Chart 2). One reason homeowners without mortgages had lower income than homeowners with mortgages is that the nonmortgage-holding householders were older on average, and thus were more likely to have retired from full-time employment. As noted above, in 2018 the median age of homeowners with mortgages was 53 while the median age of homeowners without mortgages was 66. Using 2018 ACS data, I calculate that 80 percent of 53-year-old householders were employed and only 17 percent were not in the labor force (3 percent were unemployed; not shown). By contrast, only 38 percent of 66-year-old householders were employed and 61 percent were not in the labor force (1 percent were unemployed).

Chart 2. Household income at the 25th, 50th (median), and 75th percentiles, by housing tenure and presence of O A S D I beneficiaries, 2018. Box plot with tabular version below.
Show as table
Table equivalent for Chart 2. Household income at the 25th, 50th (median), and 75th percentiles, by housing tenure and presence of OASDI beneficiaries, 2018
Percentile Rent Own with mortgage Own without mortgage
One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries
75th 48,000 76,600 109,700 153,400 81,100 120,000
Median 24,400 45,000 66,000 100,000 46,200 68,000
25th 13,700 22,800 37,200 62,450 25,230 34,000
SOURCE: Author's calculations based on ACS data.

Across all three tenure groups, households that included at least one person receiving Social Security benefits had lower income at the 25th, 50th, and 75th percentiles than did households in which there were no Social Security beneficiaries. This was due in part to the presence, on average, of fewer workers in households with Social Security beneficiaries than in nonbeneficiary households. Using ACS data for 2018, I calculate that households with no Social Security beneficiaries included an average of 1.6 people who reported having earned income during the preceding 12 months (not shown). Households with at least one Social Security beneficiary averaged only 0.6 residents who reported having earned income during the preceding 12 months.

ACS data on housing expenditures include amounts spent on shelter and utilities. Shelter expense categories include rent, mortgage interest and fees, property taxes, and homeowners' and renters' insurance. Utility expenses include water, electricity, gas, and heating oil.8 The proportion of household income spent on housing in 2018 differed substantially across tenure groups (Chart 3). Renters spent a greater share of income on housing than did homeowners, and homeowners with mortgages spent a greater proportion of income on housing than did those without mortgages. Within each tenure category, households with one or more Social Security beneficiaries spent a larger proportion of household income on housing than did households with no Social Security beneficiaries. This result aligns with the data exhibited in Chart 2, which show that in all three categories of household tenure, households with one or more Social Security beneficiaries generally had lower incomes than households in which no one received income from Social Security.

Chart 3. Percentage of household income spent on housing at the 25th, 50th (median), and 75th percentiles, by housing tenure and presence of O A S D I beneficiaries, 2018. Box plot with tabular version below.
Show as table
Table equivalent for Chart 3. Percentage of household income spent on housing at the 25th, 50th (median), and 75th percentiles, by housing tenure and presence of OASDI beneficiaries, 2018
Percentile Rent Own with mortgage Own without mortgage
One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries
75th 56.9 48.5 41.2 28.9 22.0 17.4
Median 32.5 28.0 25.1 19.6 12.4 9.0
25th 20.0 17.8 16.2 13.9 7.3 5.3
SOURCE: Author's calculations based on ACS data.

Among beneficiary households, renters with the median ratio of housing expenditures to household income spent nearly one-third of their income (32.5 percent) on housing, while the median ratio among homeowners with a mortgage was about one-quarter (25.1 percent). The median ratio of housing expenditures to household income for a nonmortgage-holding homeowner household with at least one Social Security beneficiary was only 12.4 percent in 2018. One-fourth of renter households with at least one Social Security beneficiary spent 57 percent or more of household income on housing in 2018. One-fourth of beneficiary households with mortgages spent 41 percent or more of income on housing. One-fourth of homeowning beneficiary households without mortgages spent 22 percent or more of income on housing expenditures for items such as property taxes, homeowners' insurance, and utilities.

In summary:

Number of Households by Tenure and Receipt of Social Security Benefits, 2005–2018

From 2005 to 2018, the number of households in the United States rose from 114.7 million to 121.5 million (Table 1), a 5.9 percent increase. The number of households in which no one received Social Security benefits fell from 84.6 million to 83.2 million, a decrease of 1.7 percent. In that period, the number of households in which one or more people received income from Social Security rose by 27.3 percent, from 30.1 million to 38.3 million.

The number of households with one or more Social Security beneficiaries rose mainly because waves of individuals born during the 1946–1964 baby boom became eligible for and claimed retired-worker benefits. From December 2005 to December 2018, the number of Social Security beneficiaries rose by 30.0 percent, from 48.4 million to 62.9 million. The number of Old-Age and Survivors Insurance beneficiaries increased by 31.4 percent, from 40.1 million to 52.7 million—substantially more than the increase in the number of Disability Insurance beneficiaries, which rose 22.9 percent, from 8.3 million to 10.2 million (SSA 2019, Table 5.A4).

During the period from 2005 to 2018, the number of renter households in the United States rose from 38.0 million to 43.7 million (Chart 4), a 14.9 percent increase. The number of renter households with no Social Security beneficiaries rose by 10.2 percent, from 31.3 million to 34.5 million, while the number of beneficiary renter households rose from 6.8 million to 9.2 million (36.7 percent). Households with one or more Social Security beneficiaries constituted 17.8 percent of renter households in 2005 and 21.1 percent of them in 2018. Conversely, renter households constituted 22.5 percent of the 30.1 million Social Security households in 2005 and 24.1 percent of the 38.3 million Social Security households in 2018 (Table 1).

Chart 4. Number of renter households, by presence of O A S D I beneficiaries, selected years 2005-2018. Stacked bar chart with tabular version below.
Show as table
Table equivalent for Chart 4. Number of renter households, by presence of OASDI beneficiaries, selected years 2005–2018
Presence of OASDI beneficiaries 2005 2010 2015 2018
All households 38,039 39,605 43,551 43,712
No beneficiaries 31,280 31,953 34,616 34,473
One or more beneficiaries 6,759 7,652 8,935 9,239
SOURCE: Author's calculations based on ACS data.

From 2005 to 2018, the number of homeowner households with mortgages declined by 7.7 percent, from 52.2 million to 48.2 million (Chart 5). The number of nonbeneficiary households with mortgages declined by 17.2 percent, from 43.5 million to 36.0 million, but the number of beneficiary households with mortgages rose from 8.7 million to 12.2 million (39.5 percent). Beneficiary households accounted for 16.7 percent of all homeowner households with mortgages in 2005 and 25.3 percent of them in 2018. Homeowner households with mortgages constituted 29.0 percent of the 30.1 million Social Security households in 2005 and 31.8 percent of the 38.3 million Social Security households in 2018 (Table 1).

Chart 5. Number of mortgage-holding homeowner households, by presence of O A S D I beneficiaries, selected years 2005-2018. Stacked bar chart with tabular version below.
Show as table
Table equivalent for Chart 5. Number of mortgage-holding homeowner households, by presence of OASDI beneficiaries, selected years 2005–2018
Presence of OASDI beneficiaries 2005 2010 2015 2018
All households 52,202 50,342 47,219 48,171
No beneficiaries 43,479 40,161 35,675 36,002
One or more beneficiaries 8,723 10,181 11,544 12,169
SOURCE: Author's calculations based on ACS data.

The increase in the number of Social Security beneficiary households with mortgages during this period coincided with an increase in the proportion of all older homeowners with mortgages. Collins, Hembre, and Urban (2018) examined data from the Census Bureau and reported that from 2000 through 2015, the number of households headed by individuals aged 65 or older who held mortgage debt rose by 3.6 million, “increasing older American mortgage usage by thirty-nine percent.” Haurin, Loibl, and Moulton (2019), using data from the Health and Retirement Study, found that from 2004 to 2014, the proportion of households headed by persons aged 62–71 who had mortgage debt rose from 40 percent to 46 percent. Among households headed by persons aged 72 or older, the proportion with mortgage debt rose from 17 percent to 22 percent. The Joint Center for Housing Studies of Harvard University (2019) used data from the Federal Reserve Board's Survey of Consumer Finances to find that, from 1989 to 2016, the proportion of households headed by individuals aged 65–79 who had mortgages or home equity loans increased from 24 percent to 46 percent. Over the same period, the proportion of households headed by individuals aged 80 or older who had mortgage debt rose from 3 percent to 26 percent.

The increase in retirement-age households with housing debt occurred in a period during which interest rates fell to historically low levels. Some older homeowners may have chosen to refinance their mortgages at lower interest rates and thereby extended the term of their loans into their retirement years. The availability of home equity loans and lines of credit also might have provided incentives for older homeowners to borrow against the equity in their homes. Other homeowners may have been unable to pay off their mortgage debts before retiring because of other financial obligations. For example, some older homeowners may have helped their adult children repay student loans or purchase first homes, delaying repayment of their own mortgage debt to do so.

From 2005 to 2018, the number of homeowner households without mortgages rose by 21.2 percent, from 24.4 million to 29.6 million (Chart 6). The number of nonbeneficiary, nonmortgage-holding homeowner households rose by 29.4 percent, from 9.8 million to 12.7 million, while the number of nonmortgage-holding beneficiary households rose from 14.6 million to 16.9 million (15.6 percent). Households that included one or more Social Security beneficiaries constituted 59.8 percent of all homeowner households without mortgages in 2005 and 57.1 percent of them in 2018. Nonmortgage-holding homeowner households accounted for 48.6 percent of the 30.1 million Social Security households in 2005 and 44.1 percent of the 38.3 million Social Security households in 2018 (Table 1).

Chart 6. Number of nonmortgage-holding homeowner households, by presence of O A S D I beneficiaries, selected years 2005-2018. Stacked bar chart with tabular version below.
Show as table
Table equivalent for Chart 6. Number of nonmortgage-holding homeowner households, by presence of OASDI beneficiaries, selected years 2005–2018
Presence of OASDI beneficiaries 2005 2010 2015 2018
All households 24,442 24,596 27,409 29,614
No beneficiaries 9,820 9,966 11,635 12,706
One or more beneficiaries 14,622 14,630 15,774 16,908
SOURCE: Author's calculations based on ACS data.

From 2005 to 2018, the number of households with one or more Social Security beneficiaries increased by 8.2 million. Renter households accounted for 30.2 percent of that increase, as their numbers rose by 2.5 million. The number of beneficiary households with mortgages rose by 3.4 million, constituting 42.0 percent of the increase, and the number of nonmortgage-holding homeowner beneficiary households rose by 2.3 million, representing 27.8 percent of the increase. As noted earlier, the number of Old-Age and Survivors Insurance beneficiaries rose by 12.6 million from 2005 to 2018, accounting for 86.9 percent of the increase in the number of all beneficiaries during this period. Although retirees are more likely to have paid off their mortgages than younger people are, homeowner households without mortgages accounted for only about 28 percent of the increase in beneficiary households. The increase in the proportion of older households with mortgages appears to account for the relatively small contribution of nonmortgage-holding homeowner households toward the overall increase in beneficiary households.

In summary:

Household Income by Tenure and Receipt of Social Security Benefits, 2005–2018

In both 2005 and 2018, the median income of homeowner households was higher than that of renter households, and among homeowner households, it was higher among mortgage-holders than nonmortgage-holders (Table 2). In both years and all three tenure categories, median income was higher in households with no Social Security beneficiaries than in households with one or more beneficiaries.

Table 2. Median annual income of U.S. households, by householder characteristics, housing tenure, and presence of OASDI beneficiaries: Selected years 2005–2018 (in 2018 dollars)
Characteristic 2005 2010 2015 2018
All households 58,208 57,348 58,587 61,000
  Households with one or more beneficiaries
All 39,757 42,032 44,819 45,900
By householder—
Age
18–61 50,146 48,596 49,794 53,700
62–69 49,118 51,821 52,866 51,400
70 or older 33,174 35,584 39,210 41,100
Marital status
Married couple 58,247 62,070 66,215 67,300
Unmarried male 30,088 30,402 31,042 31,750
Unmarried female 24,430 26,141 27,863 28,300
Race/ethnicity
Single race, non-Hispanic
White 41,403 43,656 47,039 48,000
Black 28,031 30,517 31,783 33,100
American Indian/Alaska Native 30,088 31,323 30,883 33,800
Asian/Pacific Islander 58,118 60,458 62,507 64,200
Two or more races, non-Hispanic 36,260 38,290 41,001 42,900
Hispanic origin, any race(s) 34,331 37,081 38,352 39,400
By housing tenure
Rent 21,730 23,147 24,367 24,400
Own with mortgage 61,461 63,106 65,156 66,000
Own without mortgage 39,217 41,457 45,344 46,200
  Households with no beneficiaries
All 65,576 63,341 65,898 70,000
By householder—
Age
18–61 66,862 64,143 66,745 70,000
62–69 64,740 69,094 71,195 72,200
70 or older 18,516 20,728 19,070 22,100
Marital status
Married couple 90,006 89,995 95,350 99,800
Unmarried male 50,146 46,178 49,794 51,600
Unmarried female 38,574 38,002 39,200 41,000
Race/ethnicity
Single race, non-Hispanic
White 74,577 72,549 76,280 80,000
Black 42,432 40,996 41,318 44,000
American Indian/Alaska Native 46,289 42,723 43,437 46,500
Asian/Pacific Islander 77,148 80,610 84,756 92,000
Two or more races, non-Hispanic 55,290 55,276 60,812 64,000
Hispanic origin, any race(s) 48,346 46,788 48,734 53,000
By housing tenure
Rent 38,574 38,002 42,378 45,000
Own with mortgage 90,251 90,180 95,986 100,000
Own without mortgage 61,976 60,918 65,686 68,000
SOURCE: Author's calculations based on ACS data.

From 2005 to 2018, the real median income of renter households with one or more Social Security beneficiaries rose 12.3 percent, from $21,730 to $24,400 (Chart 7). The real median income of nonbeneficiary renter households rose from $38,574 to $45,000 (16.7 percent). At the 25th percentile, the real income of beneficiary renter households rose from $12,871 to $13,700 (6.4 percent), while the real income of nonbeneficiary renter households rose 16.7 percent, from $19,544 to $22,800. At the 75th percentile, the real income of beneficiary renter households rose from $40,760 to $48,000 (17.8 percent), while the real income of nonbeneficiary renter households rose 16.8 percent, from $65,576 to $76,600.

Chart 7. Renter households: Annual household real income at the 25th, 50th (median), and 75th percentiles, by presence of O A S D I beneficiaries, selected years 2005-2018. Box plot with tabular version below.
Show as table
Table equivalent for Chart 7. Renter households: Annual household real income at the 25th, 50th (median), and 75th percentiles, by presence of OASDI beneficiaries, selected years 2005–2018
Percentile 2005 2010 2015 2018
One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries
75th 40,760 65,576 42,839 66,791 46,616 73,102 48,000 76,600
Median 21,730 38,574 23,147 38,002 24,367 42,378 24,400 45,000
25th 12,871 19,544 13,819 18,655 14,091 21,189 13,700 22,800
SOURCE: Author's calculations based on ACS data.

From 2005 to 2018, the real median income of beneficiary mortgage-holding households rose 7.4 percent, from $61,461 to $66,000 (Chart 8). The real median income of nonbeneficiary mortgage-holding households rose 10.8 percent, from $90,251 to $100,000. At the 25th percentile, the real income of beneficiary mortgage-holding households rose from $34,588 to $37,200 (7.6 percent), while that of mortgage-holding nonbeneficiary households rose 7.9 percent, from $57,861 to $62,450. At the 75th percentile, the real income of beneficiary mortgage-holding households rose from $102,093 to $109,700 (7.5 percent), and that of mortgage-holding nonbeneficiary households rose 13.6 percent, from $135,009 to $153,400.

Chart 8. Mortgage-holding homeowner households: Annual household real income at the 25th, 50th (median), and 75th percentiles, by presence of O A S D I beneficiaries, selected years 2005-2018. Box plot with tabular version below.
Show as table
Table equivalent for Chart 8. Mortgage-holding homeowner households: Annual household real income at the 25th, 50th (median), and 75th percentiles, by presence of OASDI beneficiaries, selected years 2005–2018
Percentile 2005 2010 2015 2018
One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries
75th 102,093 135,009 102,605 137,037 107,428 147,263 109,700 153,400
Median 61,461 90,251 63,106 90,180 65,156 95,986 66,000 100,000
25th 34,588 57,861 36,332 56,427 37,081 59,594 37,200 62,450
SOURCE: Author's calculations based on ACS data.

From 2005 to 2018, the real median income of beneficiary homeowner households without mortgages rose 17.8 percent, from $39,217 to $46,200 (Chart 9). The real median income of nonbeneficiary homeowner households without mortgages rose from $61,976 to $68,000 (9.7 percent). At the 25th percentile, the real income of beneficiary nonmortgage-holding homeowner households rose from $22,167 to $25,230 (13.8 percent). The real income of nonbeneficiary homeowner households without mortgages rose 5.8 percent, from $32,145 to $34,000. At the 75th percentile, the real income of beneficiary nonmortgage-holding homeowner households rose from $67,505 to $81,100 (20.1 percent). The real income of nonbeneficiary nonmortgage-holding homeowner households rose 12.6 percent, from $106,593 to $120,000.

Chart 9. Nonmortgage-holding homeowner households: Annual household real income at the 25th, 50th (median), and 75th percentiles, by presence of O A S D I beneficiaries, selected years 2005-2018. Box plot with tabular version below.
Show as table
Table equivalent for Chart 9. Nonmortgage-holding homeowner households: Annual household real income at the 25th, 50th (median), and 75th percentiles, by presence of OASDI beneficiaries, selected years 2005–2018
Percentile 2005 2010 2015 2018
One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries
75th 67,505 106,593 71,398 107,799 78,297 116,539 81,100 120,000
Median 39,217 61,976 41,457 60,918 45,344 65,686 46,200 68,000
25th 22,167 32,145 23,722 30,033 25,427 32,207 25,230 34,000
SOURCE: Author's calculations based on ACS data.

In summary:

Housing Expenditures by Tenure and Receipt of Social Security Benefits, 2005–2018

In both 2005 and 2018, renter households in which one or more people received Social Security benefits paid a larger percentage of their income on housing expenses than renter households with no beneficiaries (Chart 10). The median proportion of household income spent on housing remained relatively stable among beneficiary renter households, at 31.7 percent in 2005 and 32.5 percent in 2018. The median proportion of income spent on housing by nonbeneficiary renter households remained essentially the same, at 27.8 percent in 2005 and 28.0 percent in 2018.9

Chart 10. Renter households: Percentage of household income spent on housing at the 25th, 50th (median), and 75th percentiles, by presence of O A S D I beneficiaries, selected years 2005-2018. Box plot with tabular version below.
Show as table
Table equivalent for Chart 10. Renter households: Percentage of household income spent on housing at the 25th, 50th (median), and 75th percentiles, by presence of OASDI beneficiaries, selected years 2005–2018
Percentile 2005 2010 2015 2018
One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries
75th 55.4 48.4 55.5 53.6 55.5 50.1 56.9 48.5
Median 31.7 27.8 32.5 29.8 32.5 28.6 32.5 28.0
25th 19.8 17.4 20.4 18.7 19.8 18.0 20.0 17.8
SOURCE: Author's calculations based on ACS data.

At the 25th percentile of the expenditure-ratio distribution, renter households with at least one Social Security beneficiary spent approximately 20 percent of household income on housing in both 2005 and 2018, and nonbeneficiary renter households spent 17.4 percent of income on housing in 2005 and 17.8 percent of income on housing in 2018. At the upper end of the expenditure-ratio distribution (the 75th percentile), housing expenses consumed about half—or more—of household income. For beneficiary renter households at the 75th percentile, housing expenses consumed 55.4 percent of income in 2005 and 56.9 percent of income in 2018. Nonbeneficiary renter households spent 48.4 percent of income on housing in 2005 and 48.5 percent of income on housing in 2018.

Like renter households, mortgage-holding households with one or more people receiving Social Security benefits paid a higher percentage of income on housing expenses than nonbeneficiary households did in both 2005 and 2018 (Chart 11). Unlike for renter households, however, the proportion of household income spent on housing by mortgage-holders declined during this period: Among beneficiary households, the median proportion declined from 27.3 percent in 2005 to 25.1 percent in 2018; among nonbeneficiary households, it declined from 23.1 percent in 2005 to 19.6 percent in 2018.10

Chart 11. Mortgage-holding homeowner households: Percentage of household income spent on housing at the 25th, 50th (median), and 75th percentiles, by presence of O A S D I beneficiaries, selected years 2005-2018. Box plot with tabular version below.
Show as table
Table equivalent for Chart 11. Mortgage-holding homeowner households: Percentage of household income spent on housing at the 25th, 50th (median), and 75th percentiles, by presence of OASDI beneficiaries, selected years 2005–2018
Percentile 2005 2010 2015 2018
One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries
75th 44.6 33.9 45.7 36.1 41.9 30.3 41.2 28.9
Median 27.3 23.1 28.2 24.0 25.4 20.4 25.1 19.6
25th 17.7 16.2 18.3 16.7 16.6 14.5 16.2 13.9
SOURCE: Author's calculations based on ACS data.

Among mortgage-holding beneficiary households, one-fourth spent 17.7 percent or less of household income on housing in 2005. By 2018, the 25th percentile expenditure-to-income ratio had declined by 1.5 percentage points to 16.2 percent. Among nonbeneficiary households, the 25th percentile expenditure-to-income ratio declined from 16.2 percent in 2005 to 13.9 percent in 2018. One-fourth of Social Security beneficiary households with mortgages spent at least 44.6 percent of income on housing in 2005. By 2018, the 75th percentile of that ratio had declined by 3.4 percentage points to 41.2 percent of household income. Among nonbeneficiary households, the 75th percentile expenditure-to income ratio declined from 33.9 percent in 2005 to 28.9 percent in 2018.

Homeowner households without mortgages spent a smaller proportion of income on housing than did either renters or homeowners with mortgages in both 2005 and 2018. As with mortgage-holding households, the proportion of income spent on housing by nonmortgage-holding homeowner households declined from 2005 to 2018, but the decline was proportionally smaller. The median proportion of income spent on housing expenses by nonmortgage-holding homeowner households with one or more Social Security beneficiaries declined from 13.9 percent in 2005 to 12.4 percent in 2018 (Chart 12). Among nonbeneficiary households, the median proportion was 9.4 percent in 2005 and 9.0 percent in 2018.11

Chart 12. Nonmortgage-holding homeowner households: Percentage of household income spent on housing at the 25th, 50th (median), and 75th percentiles, by presence of O A S D I beneficiaries, selected years 2005-2018. Box plot with tabular version below.
Show as table
Table equivalent for Chart 12. Nonmortgage-holding homeowner households: Percentage of household income spent on housing at the 25th, 50th (median), and 75th percentiles, by presence of OASDI beneficiaries, selected years 2005–2018
Percentile 2005 2010 2015 2018
One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries One or more beneficiaries No beneficiaries
75th 24.1 17.5 23.6 19.3 22.0 17.8 22.0 17.4
Median 13.9 9.4 13.7 10.0 12.5 9.2 12.4 9.0
25th 8.3 5.6 8.2 5.8 7.4 5.4 7.3 5.3
SOURCE: Author's calculations based on ACS data.

Among nonmortgage-holding homeowner households with at least one Social Security beneficiary, one-fourth spent 8.3 percent or less of household income on housing in 2005. By 2018, the 25th percentile expenditure-to-income ratio had declined to 7.3 percent. Among nonbeneficiary households, the 25th percentile proportion was 5.6 percent in 2005 and 5.3 percent in 2018. Among Social Security beneficiary nonmortgage-holding homeowner households, one-fourth spent at least 24.1 percent of income on housing in 2005. By 2018, the 75th percentile expenditure-to-income ratio had declined to 22.0 percent. Among nonbeneficiary households, the 75th percentile expenditure-to-income ratio was 17.5 percent in 2005 and 17.4 percent in 2018.

In summary:

Conclusion

This article documents trends in housing tenure, household income, and the proportion of income spent on housing expenses in selected years from 2005 to 2018, as reported in the ACS. During that period, the number of households in which at least one person received income from Social Security increased from 30.1 million to 38.3 million (27.3 percent), and the proportion of U.S. households with one or more Social Security beneficiaries rose from 26.2 percent to 31.5 percent. In 2018, 24.1 percent of the 38.3 million households in which at least one person received Social Security benefits were renter households, 31.8 percent were homeowners with mortgages, and 44.1 percent were homeowners without mortgages.

In both 2005 and 2018, the median income of homeowner households was higher than that of renter households, and the median income of households with mortgages was higher than that of homeowner households without mortgages. From 2005 to 2018, the real median income of renter households in which at least one person received Social Security benefits increased from $21,730 to $24,400 (12.3 percent). The real median income of beneficiary households with mortgages increased from $61,461 to $66,000 (7.4 percent), and the real median income of nonmortgage-holding homeowner households with beneficiaries increased from $39,217 to $46,200 (17.8 percent).

Expenditures for housing consume a substantial proportion of household income, especially among renter households and mortgage-holding homeowner households. In renter households with one or more Social Security beneficiaries, the median share of income spent on housing was 31.7 percent in 2005 and 32.5 percent in 2018. One-fourth of renter beneficiary households spent 56.9 percent or more of income on housing in 2018. For mortgage-holding households with one or more Social Security beneficiaries, the median share of income spent on housing was 27.3 percent in 2005 and 25.1 percent in 2018. In 2018, one-fourth of mortgage-holding beneficiary households spent at least 41.2 percent of income on housing. The median proportion of income spent on housing by nonmortgage-holding homeowner households with one or more Social Security beneficiaries declined from 13.9 percent in 2005 to 12.4 percent in 2018; however, one-fourth of nonmortgage-holding beneficiary households spent at least 22.0 percent of income on housing in 2018.

Appendix A

Table A-1. Renter households: Median percentage of household income spent on housing costs, by householder characteristics and presence of OASDI beneficiaries, selected years 2005–2018
Characteristic 2005 2010 2015 2018
All renter households 28.5 30.3 29.4 29.0
  Households with one or more beneficiaries
All 31.7 32.5 32.5 32.5
By householder—
Age
18–61 33.2 34.2 33.2 32.3
62–69 29.6 30.5 31.4 31.8
70 or older 31.9 32.7 32.6 33.2
Marital status
Married couple 26.0 26.5 26.2 26.0
Unmarried male 31.7 32.8 33.0 33.1
Unmarried female 34.7 35.7 35.8 36.2
Race/ethnicity
Single race, non-Hispanic
White 30.7 31.9 31.4 31.7
Black 34.8 35.2 35.5 35.5
American Indian/Alaska Native 30.5 32.0 27.8 28.2
Asian/Pacific Islander 29.5 30.8 30.7 30.1
Two or more races, non-Hispanic 32.8 34.5 34.4 32.9
Hispanic origin, any race(s) 34.3 33.0 34.0 33.7
  Households with no beneficiaries
All 27.8 29.8 28.6 28.0
By householder—
Age
18–61 27.6 29.7 28.4 27.8
62–69 29.2 29.6 29.2 29.0
70 or older 40.0 42.0 48.3 52.1
Marital status
Married couple 23.2 25.0 23.8 23.3
Unmarried male 25.8 28.0 27.0 26.7
Unmarried female 33.7 35.7 34.5 33.7
Race/ethnicity
Single race, non-Hispanic
White 25.2 27.2 25.7 25.3
Black 32.4 34.9 33.5 32.4
American Indian/Alaska Native 25.1 27.5 27.9 24.9
Asian/Pacific Islander 28.1 27.5 27.2 25.9
Two or more races, non-Hispanic 29.7 31.8 29.8 29.1
Hispanic origin, any race(s) 31.4 33.6 32.3 31.5
SOURCE: Author's calculations based on ACS data.
Table A-2. Mortgage-holding homeowner households: Median percentage of household income spent on housing costs, by householder characteristics and presence of OASDI beneficiaries, selected years 2005–2018
Characteristic 2005 2010 2015 2018
All mortgage-holding homeowner households 23.7 24.7 21.4 20.6
  Households with one or more beneficiaries
All 27.3 28.2 25.4 25.1
By householder—
Age
18–61 25.5 26.4 22.8 21.4
62–69 26.2 27.1 24.4 24.4
70 or older 31.0 31.6 29.0 28.3
Marital status
Married couple 24.1 25.0 22.3 22.0
Unmarried male 32.2 33.3 30.7 30.1
Unmarried female 36.2 36.5 33.9 32.8
Race/ethnicity
Single race, non-Hispanic
White 26.3 27.3 24.7 24.5
Black 33.2 32.5 29.6 27.7
American Indian/Alaska Native 27.3 28.0 27.2 25.9
Asian/Pacific Islander 27.6 29.4 26.3 26.2
Two or more races, non-Hispanic 29.4 30.9 26.8 27.3
Hispanic origin, any race(s) 31.1 31.2 28.1 27.5
  Households with no beneficiaries
All 23.1 24.0 20.4 19.6
By householder—
Age
18–61 23.1 24.0 20.3 19.5
62–69 23.0 23.7 21.3 20.8
70 or older 39.7 43.2 40.6 36.8
Marital status
Married couple 21.4 22.1 18.9 18.1
Unmarried male 26.1 27.4 23.2 22.3
Unmarried female 29.4 30.4 26.4 25.3
Race/ethnicity
Single race, non-Hispanic
White 22.0 22.8 19.5 18.6
Black 26.4 28.1 23.5 22.2
American Indian/Alaska Native 22.4 23.3 19.9 20.0
Asian/Pacific Islander 27.4 28.1 23.3 22.9
Two or more races, non-Hispanic 26.1 27.5 21.6 21.3
Hispanic origin, any race(s) 29.6 30.0 24.6 23.4
SOURCE: Author's calculations based on ACS data.
Table A-3. Nonmortgage-holding homeowner households: Median percentage of household income spent on housing costs, by householder characteristics and presence of OASDI beneficiaries, selected years 2005–2018
Characteristic 2005 2010 2015 2018
All nonmortgage-holding homeowner households 12.0 12.2 11.1 10.9
  Households with one or more beneficiaries
All 13.9 13.7 12.5 12.4
By householder—
Age
18–61 11.6 12.2 11.2 10.8
62–69 11.7 11.6 10.8 11.1
70 or older 15.2 14.9 13.6 13.3
Marital status
Married couple 11.0 10.9 10.0 10.0
Unmarried male 15.0 15.2 14.2 14.2
Unmarried female 19.8 19.1 17.6 17.4
Race/ethnicity
Single race, non-Hispanic
White 13.7 13.5 12.4 12.3
Black 17.5 16.5 14.5 13.9
American Indian/Alaska Native 12.4 11.6 10.8 10.1
Asian/Pacific Islander 11.6 11.9 11.6 11.8
Two or more races, non-Hispanic 14.4 14.2 12.6 12.8
Hispanic origin, any race(s) 15.0 14.2 12.9 13.1
  Households with no beneficiaries
All 9.4 10.0 9.2 9.0
By householder—
Age
18–61 9.1 9.7 8.9 8.7
62–69 9.8 9.8 9.0 9.2
70 or older 21.0 21.9 21.3 20.2
Marital status
Married couple 7.6 7.9 7.4 7.3
Unmarried male 11.3 12.5 11.5 11.3
Unmarried female 15.1 15.7 14.4 14.2
Race/ethnicity
Single race, non-Hispanic
White 8.9 9.5 8.8 8.6
Black 12.2 13.2 11.5 11.0
American Indian/Alaska Native 9.0 9.9 8.8 8.5
Asian/Pacific Islander 8.8 9.9 9.3 9.1
Two or more races, non-Hispanic 11.1 12.3 10.0 10.2
Hispanic origin, any race(s) 11.7 12.1 10.6 10.4
SOURCE: Author's calculations based on ACS data.

Notes

1 The Census Bureau defines the householder as the person (or one of the people) in whose name the housing unit is owned or rented.

2 Out-of-pocket expenditures for health care consist of payments for (1) health insurance premiums and copayments, (2) medical services, (3) prescription and nonprescription drugs, and (4) medical supplies and equipment.

3 The Census Bureau conducts the CES for the Bureau of Labor Statistics (BLS). The CES consists of a diary survey and an interview survey, each with an independent sample. Diary survey respondents record all household expenditures for 2 consecutive weeks. Interview survey respondents are queried four times over the course of 12 months. Approximately 6,000 households complete the diary survey and about 6,000 households respond to the interview survey each quarter. BLS publishes results that integrate data from both surveys.

4 For information on the ACS Public Use Microdata Sample, see https://www.census.gov/programs-surveys/acs/microdata.html.

5 Social Security benefits are a significant source of income in many households. Although the ACS asks whether the respondent received income from Social Security in the previous 12 months, it does not ask the respondent to specify the type of benefit (retired worker, disabled worker, dependent, or survivor).

6 Percentages reported in the narrative are calculated using the more detailed values shown in the charts and tables.

7 The ACS asks separate questions about receipt of OASDI benefits and Supplemental Security Income (SSI) payments. Throughout this article, “Social Security beneficiaries” refers specifically to persons who received OASDI benefits, including those who received both OASDI and SSI. In other words, this analysis omits recipients of SSI payments (unless they also received OASDI benefits). In December 2017, 61.5 million people received OASDI benefits, including 2.7 million who received both OASDI and SSI. Another 5.5 million people received SSI alone (SSA 2018b). Some household survey respondents who received only SSI payments incorrectly reported that they received OASDI benefits. However, because the number of OASDI beneficiaries is much larger than the number of SSI recipients, any such reporting errors would have relatively little effect on the results presented here.

8 Economists treat repayment of mortgage principal as saving because it reduces household debt. Unlike the Consumer Expenditure Survey, the ACS does not collect data on expenditures for household maintenance or repairs.

9 Appendix Table A-1 shows median percentage of household income spent on housing for renters by presence of OASDI beneficiaries and householder demographic characteristics.

10 Appendix Table A-2 shows median percentage of household income spent on housing for mortgage-holding homeowners by presence of OASDI beneficiaries and householder demographic characteristics.

11 Appendix Table A-3 shows median percentage of household income spent on housing for nonmortgage-holding homeowners by presence of OASDI beneficiaries and householder demographic characteristics.

References

Bureau of Labor Statistics. 2020. Consumer Expenditures in 2018. BLS Report No. 1086. Washington, DC: BLS. https://www.bls.gov/opub/reports/consumer-expenditures/2018/pdf/home.pdf.

Census Bureau. 2014. American Community Survey Design and Methodology: Version 2.0, January 30, 2014. Washington, DC: Census Bureau. https://www.census.gov/history/pdf/acsdesign-methodology2014.pdf.

Collins, J. Michael, Erik Hembre, and Carly Urban. 2018. “Exploring the Rise of Mortgage Borrowing Among Older Americans.” CRR Working Paper No. 2018-3. Chestnut Hill, MA: Center for Retirement Research at Boston College. https://crr.bc.edu/working-papers/exploring-the-rise-of-mortgage-borrowing-among-older-americans/.

Haurin, Donald, Cäzilia Loibl, and Stephanie Moulton. 2019. “Debt Stress and Mortgage Borrowing in Older Age: Implications for Economic Security in Retirement.” Working Paper No. WI19-06. Madison, WI: University of Wisconsin-Madison Center for Financial Security, Retirement and Disability Research Center. https://cfsrdrc.wisc.edu/project/wi19-06.

Johnson, Richard W. 2015. “Housing Costs and Financial Challenges for Low-Income Older Adults.” Program on Retirement Policy Brief. Washington, DC: Urban Institute. https://www.urban.org/research/publication/housing-costs-and-financial-challenges-low-income-older-adults.

Joint Center for Housing Studies of Harvard University. 2019. “Housing America's Older Adults, 2019.” Cambridge, MA: JCHS. https://www.jchs.harvard.edu/sites/default/files/Harvard_JCHS_Housing_Americas_Older_Adults_2019.pdf.

[SSA] Social Security Administration. 2018a. Expenditures of the Aged Chartbook, 2015. Publication No. 13-11832. Washington, DC: SSA. https://www.ssa.gov/policy/docs/chartbooks/expenditures_aged/2015/index.html.

———. 2018b. Fast Facts & Figures About Social Security, 2018. Publication No. 13-11785. Washington, DC: SSA. https://www.ssa.gov/policy/docs/chartbooks/fast_facts/2018/index.html.

———. 2019. Annual Statistical Supplement to the Social Security Bulletin, 2019. Publication No. 13-11700. Washington, DC: SSA. https://www.ssa.gov/policy/docs/statcomps/supplement/2019/index.html.