A Partnership of the | | |
SEARCH     HOME     CONTACT US Quick Connect     MEDIA     POLICY / PRACTITIONERS     STUDENTS
 

Overview

Poverty in the US

Poverty on the West Coast

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: