# Income of the Population 55 or Older, 2008

### There are statistics for aged units and the family income of persons. What is the difference? Which statistics should I use?

The wider variety of tables incorporated in this edition gives more options to users needing information on the income of the aged. The questions asked about the income of the elderly often fall into one of two categories: what income do the elderly provide for themselves and those they live with, and what income is available as a resource for the elderly.

Statistics for aged units treat each marital unit (married couple or nonmarried individual) as one unit. A nonmarried individual has only their income and demographic attributes. A married couple's age is defined as the age of the husband—unless he is under age 55 and the wife is 55 or older, in which case it is the age of the wife. All other demographic characteristics, including the statistical weight, are that of the husband. Income for the married couple is the sum of both spouses' income; if either spouse has income from a specific source the married couple is considered to be a recipient unit. Tables on aged unit income are designed to be flexible enough to answer both types of questions, but with a couple of qualifications. First, unlike tables in which each person counts as a unit, aged unit tables treat each married couple as 1 unit and each nonmarried person as 1 unit. Interpreting aged unit tables like a person table will emphasize the economic well-being of nonmarried persons relative to married persons. Second, the aged unit tables exclude the income of other family members, which may not provide a complete picture of the resources available to the unit.

Statistics for the family income of persons are also based on the demographic attributes (age, sex, race, Hispanic origin) of each person. Total income from all family members (related through blood, marriage, or adoption) is treated as another attribute of the person. If any person in the family has income from a specific source the aged person is considered to be in a recipient family. These tables are designed to answer questions on the resources available to an aged person.

The table below illustrating the differences across units for total money income is taken from Tables 3.A1 and 3.B1. Median income is higher for aged units than it is for persons because aged unit income includes spousal income. Family income of persons 65 or older is higher still because it includes income from all family members, not just a spouse. The number (thousands) is the same for persons and family income of persons because only the attribute of interest (person versus family income) has changed; both are concerned with the same population—persons 65 or older. The number of aged units is lower because a married couple is counted as 1 unit, not 2 persons.

Persons 65
or older (unpublished)
Aged units 65
or older
Family income
of persons
65 or older
Median income (dollars) 17,657 24,857 34,568
Number (thousands) 37,788 28,921 37,788

### I have seen other statistics on the income of the elderly that are different from these. What would cause these differences?

Statistics may differ across publications because different data are used. Income of the Population 55 or Older uses the March Supplement to the Current Population Survey, which is conducted by the U.S. Census Bureau. Some publications producing statistics on the income of the elderly may use other surveys like the Survey of Income and Program Participation or the Health and Retirement Survey. Still other publications, such as Social Security's Annual Statistical Supplement, use administrative data.

Different publications also may not employ the same procedures for similar statistics. The units (persons, aged units, families, households, and so forth) may differ. One methodology may exclude income or units that another does not.

The statistics below on the average amount of money the elderly received from Social Security illustrate these points. The first column shows the number of beneficiary aged units 65 or older and median Social Security income from Table 5.A1. The second column shows the number of beneficiaries and average benefit from Table 5.A16 of the Annual Statistical Supplement, 2009. The number of beneficiaries differs because two different sources of data and two different units were used. The Annual Statistical Supplement statistics are based on administrative records from Social Security's Master Beneficiary Record drawn for December 2008. The administrative records include institutionalized beneficiaries (for example, in nursing homes and hospitals) and do not rely on accurate self-reporting by the beneficiary to a survey. The benefit amounts differ not only because the data are different, but also because the Income of the Population statistic is the median of the annual amount received and the Annual Statistical Supplement statistic is the mean of the monthly benefit in December 2008.

Income of the Population Table 5.A1 Supplement Table 5.A16
Beneficiary aged units 65 or older (thousands) 25,210 . . .
Beneficiaries 65 or older (thousands) . . . 35,745
Median Social Security income (dollars) 14,966 . . .
Average monthly benefit . . . 1,124.40
Average monthly benefit (annualized) . . . 13,492.80

### Why are there only 4 quintile limits?

The quintile limits indicate the boundaries between the quintiles. The bottom quintile has income below the lowest quintile limit with no lower boundary. Likewise, the top quintile has income of at least the highest quintile limit with no upper boundary. The middle quintiles are bounded on the top and bottom by the quintile limits.

### Why isn't Social Security differentiated by type of benefit?

Social Security is not broken out by type of benefit because a person may receive more than one type of Social Security benefit (such as retired-worker benefit and spouse benefit). Aged units and families may have even more combinations of benefits (such as one spouse receiving retired-worker benefits and the other receiving disability benefits). For this reason, it also should not be assumed that beneficiary units 65 or older or persons 65 or older in beneficiary families are receiving retirement benefits, even though disability benefits are converted to retired-worker benefits at full retirement age.

### Why don't you differentiate between income from defined benefit (DB) pensions and income from defined contribution (DC) plans like IRAs and 401(k) accounts?

We do not publish statistics differentiating between DB and DC pensions because a significant portion of payments from DC plans are not collected in the Current Population Survey. The Census Bureau only includes "regular payments" from retirement, survivor, and disability income in its definition of total money income. Many people do not choose to annuitize their pension accounts and instead make withdrawals from their pension accounts on their own. These withdrawals are not part of total money income, and data are not collected on withdrawals from pension accounts in the March Supplement to the Current Population Survey.1

### Do the statistics on receipt of asset income and the receipt of pension income in Section 2 indicate whether an elderly person has assets or a pension account?

No. The March Supplement of the Current Population Survey does not ask about asset and pension holdings, and not all asset and pension income is included in the Census Bureau's definition of total money income. Two notable exclusions are withdrawals from defined contribution pension accounts and capital gains or losses. Only "regular payments" from retirement, survivor, and disability income are included as pension income. Many people do not choose to annuitize their pension accounts and instead receive lump sums or make withdrawals from their pension accounts on their own. These withdrawals are not included as part of pension or total money income. Using receipt of income from assets or pensions will underestimate asset or pension holdings.

### What is the difference between the relative importance of an income source in Sections 8 and 9 and a source's share of aggregate income in Section 10?

The most important difference between the two concepts is that the relative importance of an income source is based on aged units/family income of persons, while a share of aggregate income is an average over dollars. This means that the relative importance of Social Security for an aged unit is calculated by dividing each aged unit's Social Security income by its total income. Based on the ratio of Social Security income to total income, aged units are then placed in the appropriate row of the table (0 percent of income from Social Security, 1 percent to 19 percent of income from Social Security, and so forth). Only aged units with positive total income and nonnegative earnings and asset income are included. The mean at the bottom of each panel is an average of the ratio of Social Security income to total income. The measures of relative importance of income sources in Sections 8 and 9 are designed to examine the resources available to an elderly person or aged unit.

On the other hand, aggregate income shares are designed to answer the question of what income sources the elderly as a whole are providing to their living units. Income provided by the nonelderly is excluded (except in the case of aged units because spouses may be nonelderly). The Social Security share of aggregate income for persons 65 or older is the total number of dollars of Social Security received by persons 65 or older divided by the total number of dollars received by the elderly from all sources of income. Only the total number of dollars matters for aggregate income shares; neither the number of persons or aged units nor the distribution of money is taken into account.

### Do statistics on the relative importance of income sources accurately reflect the resources available to the elderly?

Not in all instances. First, the survey on which these data are based does not include some potentially important resources as income, including lump-sum pension payments and capital gains. In addition, these statistics do not take into consideration noncash benefits that supplement money income (housing and energy subsidies or food stamps) or the amount of savings available to supplement monthly income. This could overstate the relative importance of earnings or Social Security and understate the relative importance of pensions and assets as resources.

Second, comparisons of the survey data used in this publication, the Current Population Survey (CPS), with other surveys indicate that certain sources of income are increasingly underreported, particularly asset income and pension income. One survey designed to capture small and/or infrequent amounts of income is the Survey of Income and Program Participation (SIPP). Researchers at the Census Bureau and SSA have used these surveys to estimate how misreported income leads to over- or underestimation of total income, poverty, and the relative importance of income sources. Estimates of the proportions of the elderly receiving all of their income from a single source ("100 percent reliance") are particularly affected by underreports of receipt of income from other sources because the receipt of just \$1 from a second income source changes whether an aged unit receives all of their income from a single source. The table below from Fisher (2007) illustrates the differences in receipt of income by source between the CPS and the SIPP. These differences in receipt, particularly of asset income and pension income, can cause the estimated relative importance of Social Security to vary greatly. In 1996, 17.9 percent of elderly aged units in the CPS reported receiving all of their income from Social Security; however, only 8.4 percent of elderly aged units in the SIPP reported receiving all of their income from Social Security.

For further information on the reporting of income in surveys and its effect on statistics, please consult articles by Fisher (2007), Koenig (2003), or Roemer (2000).2

Percentage of aged units 65 or older with income from specified source, 1996
Number of aged units (thousands) CPS SIPP
Total 24,553 25,671
Percentage of aged units reporting—
Earnings 20.7 21.4
Retirement benefits 93.1 96.1
Social Security 90.6 94.2
Pensions 41.2 55.7
Public 13.6 20.3
Private 29.9 42.9
Asset income 63.0 73.5
Interest 60.9 71.5
Not interest 24.7 32.0
Public assistance 5.9 9.7
SSI 5.6 7.9
Other public assistance 0.4 2.5
Veterans' benefits 4.6 5.9
Unemployment compensation 0.8 0.8
Workers' compensation 0.6 1.0
SOURCE: Fisher, T. Lynn. 2007. The impact of survey choice on measuring the relative importance of Social Security benefits to the elderly. Social Security Bulletin 67(2): 55–64, Table 2.

### I can't find the information I need. What are some other sources of data?

A good source of data on Social Security benefits and Supplemental Security Income is in Social Security Administration's Annual Statistical Supplement. It and other data publications from Social Security are located on Social Security Administration Office of Retirement and Disability Policy's data Web site, http://www.socialsecurity.gov/policy/data_title.html.

The Census Bureau also has a series of publications based on the Current Population Survey, the most recent of which is called Income, Poverty, and Health Insurance Coverage in the United States: 2008. Several years of these reports can be accessed through the Census Bureau's Income Web site, http://www.census.gov/hhes/www/income/income.html. The Internal Revenue Service's Statistics of Income Division produces data on income that is accessible through its Tax Statistics Web site, http://www.irs.gov/taxstats/.

Data on a variety of topics, including income, wealth, and consumption can be found in the Statistical Abstract of the United States through the Census Bureau at http://www.census.gov/compendia/statab/. The Bureau of Labor Statistics produces a series of reports on consumption from its Consumer Expenditure Survey. These reports and other data on consumption can be found at http://www.bls.gov/cex/. Data on wealth are included in the Survey of Consumer Finances (Federal Reserve Board, http://www.federalreserve.gov/pubs/oss/oss2/scfindex.html), the Panel Study of Income Dynamics (University of Michigan, http://www.psidonline.isr.umich.edu/), and through the Census Bureau's Housing and Household Economic Statistics Division (http://www.census.gov/hhes/www/wealth/wealth.html). Data on health, retirement, and aging are available in the Health and Retirement Study sponsored by the National Institute on Aging (University of Michigan, http://hrsonline.isr.umich.edu/).

### I can't find the answer to my question. Whom do I contact?

If you have questions about how the statistics in this publication were calculated, please e-mail income.pop55@ssa.gov. If you would like to request a copy of this publication, e-mail op.publications@ssa.gov. If you have questions regarding the Current Population Survey, please visit the Census Bureau's CPS Web site http://www.census.gov/cps/. The technical documentation for the March 2009 Supplement is located at http://www.census.gov/apsd/techdoc/cps/cpsmar09.pdf. If you have questions regarding your personal Social Security records, including benefits and earnings history, please call 1-800-772-1213, visit http://www.socialsecurity.gov/onlineservices/ or contact your local Social Security field office.

## Notes

1 A paper by Marc Roemer assessing the quality of income data in the March Current Population Survey and Survey of Income and Program Participation provides further insight into what pensions are and are not counted as money income. See "Assessing the Quality of the March Current Population Survey and the Survey of Income and Program Participation Income Estimates, 1990–1996," unpublished, but available at http://www.census.gov/hhes/www/income/assess1.pdf.

2 Fisher, T. Lynn. 2007. Social Security research: A quartet of articles measuring the economic well-being of the elderly. Social Security Bulletin 67(2): 41–72.
Koenig, Melissa. 2003. An assessment of the Current Population Survey and the Survey of Income and Program Participation using Social Security administrative data. Paper presented at the Federal Committee on Statistical Methodology Research Conference, Washington, DC.
Roemer, Marc. 2000. Assessing the quality of the March Current Population Survey and the Survey of Income and Program Participation income estimates, 1990–1996. http://www.census.gov/hhes/www/income/assess1.pdf.