Counting Persons in Poverty on the Current Population Survey

08/01/1998

This paper explains differences between the counts of persons in poverty as published by the Census and the counts one gets when doing a simple tabulation of the public use CPS file.

The March 1997 CPS (Table 1) shows 36.5 million persons in poverty in 1996 according to the Bureau of the Census. This represents 13.7% of a total population of 266.22 million persons.1 Analysts who tabulate the CPS to arrive at persons in families whose income is less than the poverty threshold without adjustment, however, will get neither the same number of persons in poverty nor the same total population. Rather, depending upon how they tabulate the file, they will get a larger total population and as many as 11% more persons in poverty. This article explains the two reasons for the differences: the treatment of children under 15 who are unrelated to anyone else in the household in which they live and the treatment of families containing subfamilies.

As Table 1 shows, the analyst who tabulates the file in a relatively straightforward manner will get a total population of 266.79 million persons. This is the same as the Census total for nonpoverty purposes, but 570,000 more than the Census total population for poverty purposes.2 This 570,000 person disparity is the population of children under 15 who live in households, but are unrelated to anyone else in those households.

By definition, these children have no income since persons under 15 are not asked about their income. Yet, they tend to live in families which are not poor. Census resolves the problem of these children by omitting them from the poverty group and from the population when any tabulation involves a poverty count.

HHS has always included them in its count of persons in poverty and in the total population on the grounds that about 40% of them are foster children placed with the family by the local social service agency. Thus while the family is not poor, these children were poor when they were placed with the family and probably will again be poor after they return to their birth parents. The number of persons in poverty in HHS tabulations is 37.1 million, 570,000 more than the Census number. In HHS tabulations, there are 13.9% in poverty, compared to the 13.7% published by Census.

Family Structure and Poverty

A second difference between the poverty population as shown by Census and that which one gets when tabulating the file without adjustment is the treatment of families containing subfamilies. A subfamily exists when three or more related persons live together and some of them are married or more than one generation apart. For example, a married couple whose 16-year-old daughter and her 6-month-old child live with them would be a primary family with four members. In addition the younger mother and her child would constitute a related subfamily. In our example all the family's income is earned by the married couple.

If one counts the subfamily as part of the primary family (as Census does), the entire family is likely to have income higher than the poverty level and no one in the family would be counted as being in poverty. If one counts the subfamily as having its own income, our example younger subfamily will be in poverty since its income is zero.

The Census file shows as the family income of each primary family the total income including that of the subfamily. The family income of the subfamily includes only that of the members of the subfamily. Thus some relatively small number of primary families would be in poverty if their family income excluded the subfamily members' income. But they appear not to be in poverty because subfamily members incomes' are included. In addition, the poverty threshold on the file pertains only to the family size of the (larger) primary families and is thus too high for subfamilies. An analysis based on simply dividing income by poverty level will show about 4 million (11%) more persons in poverty than Census shows. This results because the income of subfamilies tends to be low, because the poverty level is set for families including the subfamily members, and because unrelated children in households are included. By this definition, 15.3% of the population is in poverty, compared to 13.7%.

Children Under Age 18

Of more current policy interest are estimates of children under age 18 in poverty. In March 1997, according to Census tabulations of the CPS which do not involve poverty (Table 2), the population included 71.22 million children.3 Again, 570,000 of them were unrelated children under age 15 in households. Therefore in its poverty calculations, Census indicates that there were only 70.65 million children.4

The number of children in poverty is 14.46 million if these unrelated children are left out.5 It is 15.03 million if they are included.6 Using subfamily income to determine poverty would increase the number of children in poverty to 16.83 million.7 An unadjusted tabulation of the CPS file will produce 2.37 million (16%) more children in poverty than Census shows in its poverty document. By this definition, 24% of children are in poverty, compared to 20%.

Notes

  1. Lamison-White, Leatha, U.S. Bureau of the Census, Current Publication Reports, Series P60-198, Poverty in the United States, 1996, U.S. Government Printing Office, Washington, DC, 1997, Pages vii and 2.

  2. Census Health Insurance Table A from http://www.census.gov/hhes/hlthins, September 29, 1997.

  3. Census Health Insurance Table B from the Census web site, September 29, 1997.

  4. P60-198, Page 2.

  5. P60-198, Page vii.

  6. ASPE tabulation.

  7. ASPE tabulation.


CONTACT PERSON: M. Eugene Moyer, Office of Health Policy.

Circulated periodically by the Office of the Assistant Secretary for Planning and Evaluation. This paper reflects only the views of its author(s) and does not necessarily represent the position of the Department of Health and Human Services. For further information on poverty tabulations, contact the author at 202-690-7861.


TABLE 1: Poverty Counts(millions of persons)
Census Values:
Total Population for Poverty Purposes   266.22  
+ Unrelated Children under 15 in Households   0.57
Total Population for Non-Poverty Purposes 266.79
 
In Poverty, Census 36.53
+ Unrelated Children under 15 in Households 0.57
+ Family Definition 3.49
In Poverty, simple tabulation from the file 40.59
SOURCE: ASPE and Census Tabulations.
TABLE 2: Child Poverty Counts(millions of persons)
Census Values:
Total Population for Poverty Purposes   70.65  
+ Unrelated Children under 15 in Households   0.57
Total Population for Non-Poverty Purposes 71.22
 
In Poverty, Census 14.46
+ Unrelated Children in Households 0.57
+ Family Definition 1.80
In Poverty, simple tabulation from the file 16.83
SOURCE: ASPE and Census Tabulations.

ASPE RESEARCH NOTES ARTICLES AVAILABLE

Cost of Teenage Childbearing: Current Trends (August 1992)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1992/rn03.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1992/rn03.pdf
Counting Persons in Poverty on the Current Population Survey (August 1998)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1998/rn20.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1998/rn20.pdf
Disability Among Children (January 1995)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1995/rn10.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1995/rn10.pdf
Eldercare: The Impact of Family Caregivers' Employment on Formal and Informal Helper Hours (August 1995)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1995/rn14.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1995/rn14.pdf
Estimating Eligibility for Publicly-Financed Home Care: Not a Simple Task… (April 1992)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1992/rn01.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1992/rn01.pdf
Health Insurance in 1994 from the Current Population Survey: Measurement Difficulties (November 1996)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1996/rn15.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1996/rn15.pdf
Informal Caregiver "Burnout": Predictors and Prevention (April 1993)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1993/rn05.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1993/rn05.pdf
Licensed Board and Care Homes: Preliminary Findings from the 1991 National Health Provider Inventory (May 1993)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1993/rn06.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1993/rn06.pdf
March 1992 Current Population Survey Shows Health Insurance Coverage Up in 1991: Number of Medicaid Recipients Also Rises (February 1993)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1993/rn04.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1993/rn04.pdf
March 1993 Current Population Survey Re-Benchmarked on 1990 Census (March 1995)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1995/rn12.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1995/rn12.pdf
Number of Medicaid Recipients Up: CPS Shows the Number of Uninsured Also Rises (April 1992)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1992/rn02.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1992/rn02.pdf
Population Estimates of Disability and Long-Term Care (February 1995)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1995/rn11.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1995/rn11.pdf
Research and Other Developments of Interest in Employer Group Long-Term Care Insurance ( )
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/rn19.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/rn19.pdf
The Elderly with Disabilities: At Risk for High Health Care Costs (February 1994)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1994/rn08.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1994/rn08.pdf
The Medicaid Personal Care Services Option Part I: Cross-State Variations and Trends Over Time (November 1993)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1993/rn07.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1993/rn07.pdf
The Medicaid Personal Care Services Option Part II: Consumer-Directed Models of Care (December 1994)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1994/rn09.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1994/rn09.pdf
Trends in AFDC and Food Stamp Benfits: 1972-1994 (May 1995)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1995/rn13.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1995/rn13.pdf
Understanding Estimates of Uninsured Children: Putting the Differences in Context (January 1999)
Full HTML Version   http://aspe.hhs.gov/daltcp/reports/1999/rn21.htm
Full PDF Version   http://aspe.hhs.gov/daltcp/reports/1999/rn21.pdf

To obtain a printed copy of this report, send the full report title and your mailing information to:

U.S. Department of Health and Human ServicesOffice of Disability, Aging and Long-Term Care PolicyRoom 424E, H.H. Humphrey Building200 Independence Avenue, S.W.Washington, D.C. 20201FAX:  202-401-7733Email:  webmaster.DALTCP@hhs.gov


RETURN TO:

Office of Disability, Aging and Long-Term Care Policy (DALTCP) Home [http://aspe.hhs.gov/_/office_specific/daltcp.cfm]Assistant Secretary for Planning and Evaluation (ASPE) Home [http://aspe.hhs.gov]U.S. Department of Health and Human Services Home [http://www.hhs.gov]

Preview
Download

"rn20.pdf" (pdf, 1.23Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®