Poverty Measurement
Measuring Poverty
Determining what constitutes a meaningful measure of poverty has confounded analysts for decades. Indeed, most countries do not have an official poverty measure. When analysts in many developed countries measure poverty they often use a ‘relative’ approach, defining poverty as a percentage of median or mean household or family income (adjusted for household or family size). Researchers using Luxembourg Income Study data often define poverty as being below 50% of median (size adjusted) income, while one of the European Union’s new common statistical indicators defines being ‘at risk of poverty’ as having a disposable household income below 60 percent of a country’s median (size adjusted) income.
It wasn’t until President Lyndon Johnson’s War on Poverty in the 1960s that the United States first began to officially measure and count the poor population. The figures that became the official poverty thresholds were developed to count the number of persons in poverty and to provide a statistical yardstick to measure the nation’s progress in reducing poverty.
How the Federal Government Defines Poverty:
The U.S. Census Bureau is the federal agency responsible for preparing
poverty population statistics for the United States.
The official definition of poverty has two components: the poverty thresholds,
and the definition of income that is compared to the thresholds. Families’ and
unrelated individuals’ incomes are compared with a set of poverty
thresholds that vary by family size, by the number of family members who
are children under 18, and (for unrelated individuals and two-person families)
the age of the householder. If a family’s income is below the
appropriate threshold, all members of the family are considered to be in
poverty.
The definition of income that the Census Bureau uses for poverty measurement
is the ‘total money income’ definition, which it has used in
its annual family income reports since the 1940s. This definition
includes pre-tax income from sources such as wages, interest income, and
child support, as well as any government cash benefits such as Social Security,
Supplemental Security Income, public assistance through welfare, or veterans’ benefits. Non
cash benefits, such as housing subsidies, Medicaid, and food stamps are
not included. The income of all family members (over age 15) in a household
is combined, but unrelated individuals, even if sharing a household with
others, are tallied separately.
The Census Bureau does not determine the poverty status of certain individuals,
including people who are incarcerated, homeless individuals who are not
in a shelter, children under age 15 who are not living with a related family
member (such as foster children), and those living in college dormitories,
military barracks, nursing homes or other institutional group quarters.
1 Half of all families have incomes greater than the median; half have incomes less.
History of the Poverty Thresholds
The poverty thresholds with which incomes are compared were developed
in 1963-1964 by Mollie Orshansky of the Social Security Administration. She
based the thresholds on the economy food plan – at that time the
cheapest of four nutritionally adequate food plans being published by the
U.S. Department of Agriculture. A 1955 Agriculture Department food
consumption survey had shown that families of three or more persons spent
about one-third of their after-tax income on food. Accordingly,
the costs of the economy food plan for families of different sizes and
compositions were multiplied by three to calculate the poverty thresholds. (Somewhat
different procedures were followed for one- and two-person units.)
Poverty thresholds are adjusted for inflation annually using the Consumer
Price Index (CPI-U). Some relatively small modifications were made
in the thresholds in 1969 and 1981. The thresholds do not vary by
geographic location; thus, for a family of a given size and composition,
a single threshold applies in all 50 states (and the District of Columbia). The
poverty thresholds for 2005 are shown in the table below.
A simplified version of the thresholds – the poverty guidelines – is
used to determine eligibility for certain means-tested programs. The
poverty guidelines are not used by cash assistance programs,
housing assistance programs, or large portions of Medicaid. However,
they are used by a smaller portion of Medicaid and by food
stamps, the National School Lunch Program, and a number of medium-sized
or smaller non-entitlement programs, including Head Start. Because
some legislators and administrators have deemed the poverty line to be
too low, some programs use percentage multiples of the poverty guidelines – for
instance, 125 percent, 185 percent or even 300 percent – to determine
program eligibility. (See http://aspe.hhs.gov/poverty/faq.shtml#programs.) For
a detailed history of the current official poverty measure,
see this Poverty Measurement Working Paper on the Census Bureau
web site: http://www.census.gov/hhes/www/povmeas/papers/orshansky.html.
2005 Poverty Thresholds |
| |
Related children under 18 years |
Size of family unit |
|
|
|
|
|
|
|
|
|
|
| |
Weighted average threshold |
None |
One |
Two |
Three |
Four |
Five |
Six |
Seven |
Eight or more |
One person (unrelated individual) |
9,973 |
|
|
|
|
|
|
|
|
|
Under age 65 |
10,160 |
10,160 |
|
|
|
|
|
|
|
|
Age 65 and over |
9,367 |
9,367 |
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
|
Two persons |
12,755 |
|
|
|
|
|
|
|
|
|
Householder under 65 |
13,145 |
13,078 |
13,461 |
|
|
|
|
|
|
|
Householder 65 and over |
11,815 |
11,805 |
13,410 |
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
|
Three persons |
15,577 |
15,277 |
15,720 |
15,735 |
|
|
|
|
|
|
Four persons |
19,971 |
20,144 |
20,474 |
19,806 |
19,874 |
|
|
|
|
|
Five persons |
23,613 |
24,293 |
24,646 |
23,891 |
23,307 |
22,951 |
|
|
|
|
Six persons |
26,683 |
27,941 |
28,052 |
27,474 |
26,920 |
26,096 |
25,608 |
|
|
|
Seven persons |
30,249 |
32,150 |
32,350 |
31,658 |
31,176 |
30,277 |
29,229 |
28,079 |
|
|
Eight persons |
33,610 |
35,957 |
36,274 |
35,621 |
35,049 |
34,237 |
33,207 |
32,135 |
31,862 |
|
Nine persons or more |
40,288 |
43,254 |
43,463 |
42,885 |
42,400 |
41,603 |
40,507 |
39,515 |
39,270 |
37,757 |
Issues in Poverty Measurement
Since the poverty thresholds were developed over four decades ago, there
have been numerous proposals to revise them (or the income
definition used with them) in one way or another. Some of these proposals
have called for revisions in the poverty measure that would
reflect contemporary economic and social conditions. One of the
most detailed proposals, made in a 1995 report by a poverty
panel appointed by the National Research Council (NRC), proposed
a new methodology for developing an official poverty measure
for the United States. (The
full report of the NRC panel is available on the Census Bureau
website,
here. In
2004 the Committee on National Statistics (CNSTAT) convened
a workshop to review federal research on the NRC poverty panel’s
recommendations and evaluate progress in moving toward a new
measure of poverty. The
proceedings from the CNSTAT workshop are available on the Institute
for Research on Poverty website, here
and a summary of workshop recommendations is available here.
In 2004 and 2005, the University of Maryland, the U.S. Department of Commerce,
the U.S. Department of Health and Human Services (HHS) and
the American Enterprise Institute sponsored a seminar series
to explore the limitations of the current federal poverty measure and to
identify alternative approaches for gauging the material well-being of
low income Americans. The
seminar papers are available at the University of Maryland,
Welfare Reform Academy website, here.
Debates have focused on how income should be defined, how the poverty
thresholds should be determined, and in some cases (e.g., the NRC poverty
panel) on both. With respect to income definition, one of the key
criticisms is that noncash benefits such as food stamps and housing subsidies
are not counted as income, although they can be a significant portion of
certain families’ resources. Second, some analysts argue that
after-tax income rather than before-tax income should be used in measuring
poverty, since taxes are a nondiscretionary expense; using after-tax income
would also result in taking into account the effects of the Earned Income
Tax Credit (EITC). Third, some argue that with the rise of cohabitation,
the practice of tallying separately the income of unrelated individuals
sharing a household may fail to reflect economies of scale experienced
by cohabiters.
With respect to the poverty thresholds, some researchers take issue with
the choice of the multiplier (three), noting that food costs no longer
represent one-third of the family budget in light of rising housing costs
and out of pocket medical expenses. James Ziliak, director of the
University of Kentucky’s Center for Poverty Research points out that
American families now spend about one-sixth of their income on food (The
Plain Dealer, July 30, 2006). Moreover, at the time the poverty thresholds
were developed, far fewer women were working. With the increased labor
participation of women and the rise in single parent households, the costs
of commuting and child care are now a sizeable portion of the family budget. Another
frequently heard criticism of the thresholds is that they do not account
for nontrivial differences in the cost of living across regions and metropolitan
areas.
In a less widely known part of the poverty measurement debates, since
the early 1990s, analysts in the U.S. have developed dozens
of ‘basic
needs budgets’ or ‘family budgets’ (item-by-item ‘market
baskets’) for working families in various states and localities. These
budgets have been developed to reflect a ‘no frills’ or ‘basic
family survival’ standard of living. Dollar totals for these
budgets are significantly higher than the corresponding current
official poverty thresholds. For examples, see the Economic Policy
Institute website, here,
and here,
and, for Self-Sufficiency Standards for over 30 states, click
here.
Thus far, there is little consensus as to how or even if, a different
approach to poverty measurement should be implemented. Some contend
that the general approach to defining the poverty level is valid, but requires
revision to improve the measurement of both family resources and current
expenses. Others argue that the poverty line should be raised to
reflect the experiences and minimal needs of American families under current
economic and social conditions. Still others argue that the official
poverty line is not an accurate reflection of deprivation, because many
who are below poverty actually have many material resources, such as TVs,
cars, microwaves, etc. Moreover, they contend that self-reported
income is unreliable and often does not include money earned in the informal
sector. From their perspective, poverty should be evaluated on the
basis of living conditions and consumption patterns, and income verified
from external sources. Ultimately, formal authority for the decision
of whether and how to revise the official poverty measure rests with the
Office of Management and Budget.
Related Links:
|