Who is poor?

Who is poor?

Alternative Ways to Think About
"Who is poor?"*

In addition to the official U.S. absolute income poverty measure, a wide range of other indicators of economic poverty have been proposed and implemented, such as the following:

Relative Income Poverty

Relative income measures compare the income of a family to a norm reflecting the economic position of the overall society, adjusted for price-level changes. Because overall measures of social well-being, such as median income, tend to increase over time, the poverty standard also tends to increase.

Consumption Poverty

For many households, income may temporarily dip below the poverty line because of something that happened that year, such as unemployment or a bad harvest. An alternative is to use annual family consumption rather than annual income as a more permanent indicator of resources.

"Capability" Poverty

This indicator identifies the poor as those who do not have the capability to secure a sufficient level of resources to meet a minimum standard. A capability measure of poverty focuses attention on policies that foster economic independence.

Asset Poverty

In addition to income, assets or wealth also play an important role in understanding the level and composition of poverty in the United States. Assets create opportunities, secure a standard of living, and often pass class status on to the next generation. In this sense, asset poverty comes closer in meaning and significance to traditional ideas of economic well-being and access to life chances.

Subjective Poverty

Some researchers have measured poverty by relying on the subjective responses of individuals to questions about their perceptions of economic position or well-being, relative to some norm. The effectiveness of subjective poverty measures is limited by small sample sizes and wide variation around the average, which makes it difficult to set a generally accepted poverty threshold.

*This sidebar draws from an essay by Robert Haveman, John Bascom Professor of Economics and Public Affairs Emeritus, former IRP Director, and IRP Affiliate at the University of Wisconsin–Madison, "What Does It Mean to be Poor in a Rich Society?" and a related lecture (Flash Video).

In the United States, the Census Bureau determines who is poor using an economic poverty measure, by calculating official poverty estimates using a set of dollar-value thresholds, or minimum amount of income needed to meet basic needs, that vary by family size and composition. The measure is defined as three times the amount of (cash) income that a family is thought to need for a "thrifty food basket" given its size and composition.

This measure has remained mostly unchanged since it was introduced in the 1960s, other than to be annually adjusted for inflation using the Consumer Price Index. (See FAQ #5, "What is the Consumer Price Index and how is it used?" for a description of this index.) The official poverty measure is useful for measuring long-term poverty trends. (To read more about the U.S. poverty measure, see FAQ #2, "How is poverty measured in the United States?") However, it excludes non-cash (in-kind) transfers, taxes, and tax credits. It also does not account for any categories of expenditures, nor for regional or local cost of living.

The Census Bureau releases official poverty data in an annual report, the most recent of which was released in September 2015: Income and Poverty in the United States: 2014. The report indicates that in 2014, some 46.7 million people, or 14.8 percent of the total U.S. population, lived in poverty. Neither the number of people in poverty nor the poverty rate were statistically different from 2013 estimates.

However, the number of people in poverty did increase from 2013 to 2014 for certain groups, including unrelated individuals, people between ages 18 and 64 with a disability, people with a bachelor's degree or higher, and married-couple families.

Notably, whereas the United States uses an absolute poverty measure that accounts only for cash income, there are alternative approaches to defining and measuring poverty (see Sidebar). Furthermore, some poverty experts assert that income poverty measures ignore noneconomic considerations that may affect a person's well-being, such as living in unsafe surroundings or having a chronic medical condition. This failure to consider noneconomic conditions when determining who is poor may result in policy efforts that overlook important aspects of what it means to be poor.

To produce a more refined and complex statistic than the official administrative poverty measure, the Census Bureau introduced a new measure in 2010, the Supplemental Poverty Measure (SPM), which defines family, resources, and expenses differently than the official measure (e.g., counting federal program benefits such as food assistance and tax credits and subtracting work and medical out-of-pocket expenses when figuring resources). The SPM complements but does not replace the official measure, which continues to be used to determine official poverty rates and federal program eligibility.

The Census Bureau released the most recent SPM poverty statistics in September 2015: The Supplemental Poverty Measure: 2014. These are shown in comparison to official poverty rates for the total population and by age group in Figure 1. All differences are statistically significant, with the SPM rate higher among the entire population, lower among children (because the SPM counts public benefits that target children most intensively), higher among working-age adults, and much higher among the elderly, whose out-of-pocket medical expenses, counted by the SPM but not the official measure, are typically much greater than those of other age groups.

Figure 1. Differences in poverty rates between the official and SPM measures are significant: 2014

The following sections explore the most recent poverty estimates based on official U.S. poverty thresholds for different groups in the population, including people of color, children, women as compared to men, families, and the elderly.

Income-to-Poverty Ratios

An important consideration when identifying who is poor is that the official poverty statistics most discussed and quoted in the media indicate only how many people are living below their poverty threshold. Therefore, for example, a family living with income barely over their poverty threshold—and so by all accounts struggling financially—is not counted as "poor" in official statistics.

The Census Bureau provides another way to describe a person's economic well-being by gauging how close to or far from their poverty threshold a family's income rests using an income-to-poverty ratio. The official poverty report (see above link) includes information on income-to-poverty ratios for households living with incomes from 50 percent below to 200 percent above their threshold. See the latest income-to-poverty ratios in Table 5 of: Income and Poverty in the United States: 2014.

These ratios are useful, for example, for estimating the percentage of Americans living in deep poverty, defined as income less than half the threshold, and near poverty, defined as living with income no more than 25 percent above the threshold. According to the 2014 income-to-poverty ratios, 6.6 percent of Americans were living in deep poverty. Among children, the rate is higher: 9.3 percent of children were living under 50 percent of their family's poverty threshold.

Near poverty rates (income between 100 and 124 percent of the poverty threshold) are much higher: 19.4 percent of the overall population; 27.0 percent of children; 17.1 percent of whites; 32.5 percent of blacks; and 31.5 percent of people of Hispanic origin have income no more than 124 percent above their poverty threshold.

Does poverty "discriminate"?

The official national poverty rate represents an average over the entire population, and does not really indicate who is well-off and who is worse off. Blacks and Hispanics, for example, have poverty rates that greatly exceed the overall average. The poverty rate for all blacks and Hispanics remained near 30 percent during the 1980s and mid-1990s, while the overall poverty rate was closer to 15 percent. Thereafter the poverty rates of these two groups began to fall, though they remained high.

In 2000, the rate for blacks dropped to 22.1 percent and for Hispanics to 21.2 percent—the lowest rate for both groups since the United States began measuring poverty. Meanwhile, the overall poverty rate was about half those rates, at 11.3 percent. By 2014, however, the poverty rate for blacks had risen to 26.2 percent and for persons of Hispanic origin, who can be of any race, the poverty rate was up to 23.6 percent. The rate for the overall population was 14.8 percent.

The 2014 poverty rate for Asians was 12.0 percent, and whites not of Hispanic origin had a poverty rate of 10.1 percent. See Figure 2 for long-term changes in the U.S. poverty rate by racial or ethnic group, which reveals a striking trend among blacks, whose poverty rate decreased from 41.8 percent in 1966 to 26.2 percent in 2014. Nonetheless, the poverty rate among blacks is more than two times greater than the 12.7 percent poor rate for whites.

Figure 2: U.S. poverty rates vary significantly by race and ethnicity: 1959−2014

How many children are poor?

In 2014, children represented 23.3 percent of the overall U.S. population, and some 15.5 million of them, 21.1 percent of all children in the United States, were poor. More than 4.6 million of these poor children were under 5 years old. Of the 15.5 million poor children, 6.5 million were living in deep poverty, which is defined as an annual income of less than half the poverty threshold.

As already noted, poverty rates vary considerably by race. Among children, 18 percent of black children and 12.9 percent of Hispanic children were in deep poverty, compared to 9.3 percent of children overall. Children of color consistently have much higher poverty rates than white children, as shown in the line graph below (as is the case in poverty rates among adults as well, as shown in Figure 2 above).

National statistics on child poverty, depicted in Figure 3 for all races and by race and Hispanic origin, are readily available but it is difficult to obtain accurate, summary data concerning child poverty at the state and local levels, and among particular groups. Therefore, non-Census Bureau-produced state estimates often draw from a number of different data sources, some official, some produced by private agencies and organizations. To increase the accuracy and reliability of estimates for small areas and groups, agencies and organizations often resort to three-year averages.

Figure 3: U.S. child poverty rates also vary a lot by race and Hispanic origin: 1959-2014

The number of respondents within each state is too small to allow for the production of official state child poverty figures based on Current Population Survey data, so the Census Bureau recommends using the American Community Survey (ACS) for state-level child poverty estimates. Since 2006, the ACS has released subnational estimates of income and poverty for all places, counties, and metropolitan areas with a population of at least 65,000, as well as for the nation and individual states.

ACS data are presented in one-year, three-year, and five-year (beginning in late 2010) estimates, which cover areas with populations over 65,000, over 20,000, and all areas, respectively. Figure 4 below shows 2014 child poverty rates for the United States as a whole and for individual states using ACS data (there is a two-tenths of a percentage point difference between the overall child poverty estimate—21.3 percent—and that found using CPS data—21.1 percent).

Figure 4. Child poverty rates in the U.S. overall and state by rank vary considerably: 2014

Another source of data on child poverty in addition to the Census Bureau is the Children's Defense Fund, which publishes "Children in the States" fact sheets on child well-being for each state. It ranks states according to indicators such as prenatal care, infant mortality, and per-pupil expenditures in public schools as well as child poverty estimates.

The Kids Count project of the Annie E. Casey Foundation reports many indicators of American children's health and well-being each year for the entire country and for each state, in various publications and through its online Data Center.

A Note about Poverty Data

Official national figures on poverty in the United States derive from the Annual Social and Economic Supplement (ASEC) to the Current Population Survey (CPS). Because the old decennial census long-form questionnaires, which asked about household income, were given to a very large sample (one of every six households), the decennial censuses through 2000 provide good, if dated, information about the percentage of children who were poor not only nationally but in geographic areas as small as census tracts (which on average contain 4,000 to 5,000 people) as well.

The American Community Survey (ACS) replaced the long form in 2006; therefore, the 2010 census did not include a long form and does not provide income and poverty estimates, for any age group. From 2006 forward, annual ACS estimates can be compared to the Census 1990 and Census 2000 estimates.

Does poverty vary by other characteristics?

Poverty Rates by Sex

In 2014, 16.1 percent of females and 13.4 percent of males were in poverty. Gender differences in poverty rates were more pronounced for the elderly, 12.1 percent of women and 7.4 percent of men were living under the poverty threshold. For women between the ages of 18 and 64, the poverty rate was 15.3 percent while the rate for men ages 18 to 64 was 11.6 percent. For children under age 18, there was no statistical difference between the poverty rate for girls (21.1 percent) and the rate for boys (21.2 percent). Long-term trends in poverty rates by sex, from 1966 to 2014, are shown in Figure 5.

Figure 5: U.S. poverty rates are consistently higher among women as compared to men: 1966–2014

Poverty Rates by Family Type

The family poverty rate and number of families in poverty in 2014 were 11.6 percent and 9.5  million, respectively, which in both cases is not a statistically significant change from 2013 (see Figure 6). However, married-couple families saw increases in 2014 in both the poverty rate, 6.2 percent, and number in poverty, 3.7 million, up from 5.7 percent and 3.4 million in 2013.

Figure 6: Families headed by a single mother are by far the poorest family type, with families headed by a single father also high compared to married-couple families, but half those of single-mother families: 2013 and 2014

The poverty rate for families headed by a single woman (with no husband present) did not increase significantly, at 30.6 percent in 2014, but the number in poverty decreased to 4.8 million in 2014, down from 5.2 million in 2013. Male-householder families (with no wife present) had a much lower poverty rate than that for female-headed families, 15.7 percent in 2014, representing 1.0 million families.

Poverty Levels by Residence Area Type

As shown in Figure 7, poverty rates remained about the same across metropolitan areas in 2014 (14.5 percent poor in 2012 and in 2014). Suburban areas saw an increase, with poverty rates reaching their highest level (11.8 percent) since the mid-1960s. Rural poverty rates decreased to 16.5 percent. Long-term trends show central city poverty rates rising above rural and small town rates starting in the mid-1970s.

Figure 7: U.S. poverty rates vary significantly by residence area type: 1967−2014

Poverty by Region

Poverty rates vary considerably across regions of the United States. In the South, more than 19.5 million people were living in poverty in 2014, representing a poverty rate of 16.5 percent. The Northeast experienced a slight drop in poverty from 13.0 percent in 2013 to 12.6 percent in 2014. In the West, the poverty rate was 15.2 percent; in the Midwest, the poverty rate was 13.0 percent in 2014.

Despite the relative stability in these regions—none of the four regions experienced a significant change in the poverty rate or the number in poverty between 2013 and 2014—the regional numbers mask significant growth within particular subregions, states, and cities. Figure 8 shows each region's respective poverty rates over time.

Figure 8: U.S. poverty rates by region vary less than those by residence area type: 1969−2014

Further Information

U.S. Census Bureau

American FactFinder

Main Poverty Page

2014 Poverty Data Highlights
Description of Income and Poverty Data Sources
Living in Near Poverty in the United States: 1966–2012
How the Census Bureau Measures Poverty
Supplemental Poverty Measure
Expert Meeting on Income, Poverty, and Health Insurance, March 20, 2015

U.S. Department of Health and Human Services

Poverty Estimates, Trends, and Analysis

Summary of 2014 Current Population Survey Data
Indicators of Welfare Dependence: Annual Report to Congress, 2015
Financial Condition and Health Care Burdens of People in Deep Poverty

Related IRP Resources

"Leveraging Big Data to Help Restore the American Dream," Focus 32 (1: Spring/Summer 2015).
Neil Damron, "Brain Drain: A Child's Brain on Poverty," Poverty Fact Sheet #8, Institute for Research on Poverty, University of Wisconsin–Madison, March 2015.
Robert Haveman, "What Does It Mean to Be Poor in a Rich Society?," 2008, Robert J. Lampman Memorial Lecture (Slides | Video | Article).
Lawrence F. Katz, "Reducing Inequality: Neighborhood and School Interventions," Focus 31 (2: Fall/Winter 2014–15).
Alexandra K. Murphy and Scott W. Allard, "The Changing Geography of Poverty," Focus 32 (1: Spring/Summer 2015).
Dan Simon, "Poor and in Poor Health," Poverty Fact Sheet #3, Institute for Research on Poverty, University of Wisconsin–Madison, November 2013.
Dan Simon, "Is the American Dream Still Attainable?" Poverty Fact Sheet #6, Institute for Research on Poverty, University of Wisconsin–Madison, September 2014.