FOR IMMEDIATE RELEASE
MONDAY, SEPTEMBER 15, 2003
Robert Bernstein CB03-142
Public Information Office
(301) 763-3030/457-3670 (fax)
(301) 457-1037 (TDD)
email: pio@census.gov
*NOTE TO CORRESPONDENTS*
On Sept. 26, the Census Bureau will release its annual reports on income
and poverty in the United States; and on Sept. 30, it will release its
annual report on health insurance coverage.
With the release of the income and poverty reports, the Census Bureau
will also release a report discussing consumption-based or
expenditure-based poverty measures. This document will provide a context
for further research that may eventually lead to development of an
experimental series of consumption/expenditure measures of poverty. The
Census Bureau would develop such measures in consultation with the Bureau
of Labor Statistics (BLS), the Energy Information Administration (EIA) and
the Office of Management and Budget (OMB); the Bureau of Economic Analysis
(BEA) would provide technical support in this effort.
These experimental measures are not intended to replace the official
poverty measure.1 They are part of a decades-long tradition of Census
Bureau and other statistical agency work on alternative approaches to
measuring poverty.
The Current Population Survey's (CPS) Annual Social and Economic
Supplement is the data source for the official measures of poverty and
income, as well as for the report on health insurance coverage. There are
two modifications to the products for 2002, which the Census Bureau would
like to share with you in advance.
Racial Group Comparisons in the 2003 Current Population Survey:
Income, Poverty and Health Insurance Reports
For the first time in 2003, CPS respondents were allowed to report
themselves in more than one racial group;2 previously they had to choose
one. This complicates year-to-year comparisons. Therefore, there is no
perfect way to compare changes to income, poverty or health insurance by
race between 2002 and the previous years.
To assist, the three new reports will compare 2001 single-race figures
with two different figures for 2002: one comparison will be based on those
who reported only one race, and one will be based on those who reported
one or more races. For example, the poverty report for 2002 will compare
the 2001 poverty figures for blacks with 2002 poverty figures for those
who reported themselves as:
- black and did not report any other race, and
- black only or black in combination with some other race(s).
The income and poverty reports provide year-to-year comparisons for most
racial groups. No data will be provided for American Indians and Alaska
natives, nor for native Hawaiians and other Pacific islanders, because the
sample was not sufficiently large.
Alternative Definitions of Income and Poverty
Traditionally, income and poverty data in Census Bureau reports have
been based on the amount of money that people or households receive during
the calendar year. This income concept is limited and does not provide a
completely satisfactory measure of economic well- being. For example, it
does not include the effect of taxes and, therefore, does not reflect the
effect of tax law changes on economic well-being. Similarly, this
traditional concept excludes the effect of noncash benefits (e.g.,
employer-provided group health insurance, food stamps, school lunches and
housing assistance), which certainly enhance economic well-being.
In the early 1980s, the Census Bureau embarked on a research program to
examine the effects of a broadened range of income measures that subtract
taxes and add the value of noncash benefits. The Census Bureau published
these measures in separate reports beginning in 1988 (P60-164-RD1), and
they were included in the money income and poverty reports beginning in
1995 (P60-188). The 2002 income report will show year-to-year changes between
calendar years 2001 and 2002 based on money income alone (the traditional
income measure), as well as year-to-year changes based on four alternative
definitions of income:
1) The first adds realized capital gains and subtracts taxes (federal
income taxes, state income taxes, payroll taxes).
2) The second includes everything in the first definition and adds
employer-provided health benefits, food stamps, rent subsidies and
school lunches.
3) The third includes everything in the second definition and adds an
estimate of the value of Medicare and Medicaid.
4) The fourth includes everything in the third definition and adds an
estimate of the annuity value of home equity for homeowners.
The poverty report includes several alternative measures of poverty
developed by the Census Bureau. Six are based on recommendations by the
National Academy of Sciences. They account for benefits and taxes on
income, and also use a new set of poverty thresholds based on the
Consumer Expenditures Survey. Also presented are eight additional
measures that use the four alternative income measures mentioned earlier,
with two versions each of poverty thresholds the official thresholds and
the thresholds derived by using an experimental series of inflation
estimates. A final version uses the money income measure used for the
official thresholds, plus the alternative method of adjusting for
inflation over time.
Regarding the consumption-based measure, researchers from the Census
Bureau have worked with the Interagency Working Group on Alternative
Measures of Material Well-Being to produce a supplementary publication.
This report, Supplemental Measures of Material Well-Being:
Expenditures,Consumption, and Poverty 1998 and 2001, will discuss
alternative ways of measuring economic status and poverty, including
measures based on expenditures rather than income (which is the basis for
the current official U.S. poverty measure). This report is another
installment in our series of research reports aimed at improving our
understanding of the low- income population in the United States and how
best to measure it.
1The official definition for poverty is established by the
Office of Management and Budget (OMB). For further information, see
Statistical Policy Directive No. 14.
2OMB establishes the official guidelines for the collection
and classification of data for race (including the option for respondents
to mark more than one race) and Hispanic origin. Race and Hispanic origin
are treated as separate and distinct concepts in accord with OMB's
guidelines. For further information, see Statistical Policy Directive
No. 15.
available free from Adobe.
