How Poverty in the United States Is Measured and Why It Matters

A major goal of the White House’s Build Back Better Act is to reduce poverty in America. But how do we define and measure poverty—and what are the consequences?

The COVID-19 pandemic has sparked renewed interest in the social safety net and the government’s role in reducing poverty. Following record-breaking unemployment and economic instability, the U.S. government allocated billions of dollars in emergency relief through the CARES Act, the Consolidated Appropriation Act, and the American Rescue Plan. Recently, the Biden administration proposed a $1 trillion reconciliation bill—the Build Back Better Act—to support American children and families in need.

How those Build Back Better Act dollars are distributed will depend partly on how lawmakers decide to measure poverty. The current bill uses multiple poverty measures to determine eligibility and allocate funds. For example, access to child care is prioritized for “families with income that does not exceed 200 percent of the Federal poverty level.” Children are eligible for Medicaid if they live in families with income that does not exceed 133% of the poverty line. Certain programs to help cover home heating and cooling costs are available to low-income households at or below 80% of the area median income or 200% of the poverty level.

Understanding the differences between poverty measures is important because these variations determine the number of people who may be eligible to receive benefits and guide decisionmakers in their efforts to provide economic relief.

How the Official Poverty Measure Is Calculated

When U.S. lawmakers speak about lifting people out of poverty, they are typically referring to the official poverty measure, which first originated as a tool in Lyndon Johnson’s War on Poverty in the 1960s.1 Before the “war” could begin, the Johnson administration needed a way to count the number of people in poverty, deliver aid, and measure the effectiveness of anti-poverty policies. Johnson’s new Office of Economic Opportunity settled on a definition based on an article written by Social Security Administration economist Mollie Orshansky. She developed the measure using the Department of Agriculture’s food plans, based on the minimum amount of food needed to have a nutritionally “fair” to “good” diet.2 The definition has remained largely unaltered since the 1960s.

Today, there are two versions of the official poverty measure: 1) Poverty thresholds, produced by the U.S. Census Bureau, and 2) Federal poverty guidelines (FPG), produced by the Department of Health and Human Services (HHS).3 Poverty thresholds are used to measure poverty, whereas poverty guidelines are used to determine eligibility for certain federal programs. Poverty threshold calculations are affected by peoples’ age and family composition. Poverty guidelines simply use the number of people in a family unit to determine poverty status.4 They also vary by geography; Alaska and Hawaii have separate poverty guidelines. Both poverty thresholds and guidelines are updated each year based on annual changes in the consumer price index. And both are only calculated for people whose poverty status can be determined: People living in prisons, nursing homes, college dorms, military barracks, and other unconventional housing situations are not factored into the poverty estimates.5

The Census Bureau reported that 37.2 million people, or 11.4% of the U.S. population, were living in poverty in 2020, based on official poverty thresholds.6 The federal government does not publish estimates on the number of people living below the HHS poverty guidelines, although these statistics are used to estimate the number of people eligible for particular programs.

Developing a Better Poverty Measure

Most researchers, policymakers, and others who study poverty would agree that the official measure is flawed. The official poverty measure is based on a family’s pretax cash income and does not include non-cash benefits from housing subsidies, the Supplemental Nutrition Assistance Program (SNAP), or other forms of government relief.7 Costs related to housing, clothing, transportation, and other expenses commonly considered basic human needs are not considered. And the official measure does not account for variations in the cost of living across the country.

In 2010, the Census Bureau, working with the Office of Management and Budget, produced the Supplemental Poverty Measure (SPM) to address the shortcomings of the official poverty measure. Like the official measure, there are two components to the SPM: the minimum cost of basic needs (expenses) and the income that a family generates (financial resources). But the SPM also factors in non-cash transfers, like SNAP benefits and tax credits; expenses related to medical care, child care, taxes, and food; geographic cost-of-living differences; and whether a family is renting or paying a mortgage. The SPM also covers a broader population; in particular, the definition of families is expanded to include unmarried partners and their children, unrelated children under age 15, and foster children.

The SPM uses estimates from the Consumer Expenditure Survey to calculate average spending on various resources.8 Unlike the federal poverty guidelines, the SPM is not intended to be used to determine program eligibility.

Table 1: Comparing the Official Poverty Measure and the Supplemental Poverty Measure

Official poverty measure Supplemental poverty measure
Resources considered Pretax cash income Combined cash value of SNAP benefits, housing subsidies, Earned Income Tax Credit (EITC), Social Security, and more
Expenses considered Three times the cost of a minimum-food diet in 1963 (in today’s prices) Combined expenses related to payroll taxes, utilities, housing, medical costs, transportation, and more
Other adjustments None (level is consistent across the United States) Geographic cost of living and rental unit adjustment


In 2020, 29.8 million Americans were living below the SPM poverty threshold (9.1% of the population).9

Defining Need Based on “Multiples” of the Poverty Guidelines or Median Income

In addition to the official poverty measure and SPM, many programs use multiples of the federal poverty guidelines to determine eligibility. For example, SNAP eligibility guidelines are 130% of the FPG, and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) caps eligibility at 185% of the FPG.10 Families with incomes below 200% of the federal poverty threshold—$52,492 for two adults and two related children in 2020—are often classified as “low-income.” Families are classified as being in “deep poverty” if their income falls below 50% of the poverty guidelines ($13,123 for a family of four).11

In 2020, 17.9 million people fell below 50% of the poverty threshold (5.5% of the population), and 89.7 million people lived below the 200% poverty threshold (27.5% of the population).12

To better adjust for geographic differences, the U.S. Department of Housing and Urban Development (HUD) uses a state’s median family income as the basis for its definition of low-income. HUD defines low-income as 80% of a state’s median family income and very low-income as 50% of the state’s median family income.13

How Are Poverty Data Collected?

There are two primary sources for U.S. poverty statistics: the Current Population Survey and the American Community Survey.

The official annual estimates of national poverty levels are calculated from the Annual Social and Economic Supplement to the Current Population Survey (CPS ASEC). A joint program between the Census Bureau and the Bureau of Labor Statistics, the CPS ASEC collects information about income and benefits received during the previous calendar year and is best used for national-level estimates. The Census Bureau recommends using multiyear averages to compare CPS ASEC data across states due to the relatively small sample size (approximately 100,000 households per year).14

Every year, the Census Bureau’s American Community Survey (ACS) surveys about 3.5 million addresses on a wide range of topics, including their income the year prior to the survey. Unlike the CPS ASEC, ACS 1-year estimates can be used at the state level. Five-year ACS estimates can be used to measure poverty for geographic areas as small as block groups, although margins of error may be too large for smaller geographic areas or population subgroups.

The Census Bureau also produces single-year poverty estimates for counties and school districts through its Small Area Income and Poverty Estimates (SAIPE) program. SAIPE data are model-based estimates that combine ACS data with administrative records and are the best source for up-to-date poverty statistics for substate areas.

State-Specific Poverty Measures

Many organizations have developed custom poverty measures to study economic disparities and monitor trends in their states. The Public Policy Institute of California and the Stanford University Center on Poverty and Inequality developed the California Poverty Measure (CPM) to better understand poverty in California and how patterns vary across counties (see map). Like the SPM, the CPM adjusts for differences in geographic costs of living and Earned Income Tax Credit benefits. However, the CPM also accounts for non-cash benefits like CalFresh and CalWORKs that are unique to California.15 Similar measures have been developed in New York and Oregon.16

The Self-Sufficiency Standard is another measure of economic need that has gained attention in recent years. Originally developed for counties in Washington state, the Self-Sufficiency Standard is now used in more than 40 states to determine the amount of income working families need to make ends meet.17

Measuring Poverty During the Coronavirus Pandemic

One of the challenges with all of these poverty measures is timeliness. Policymakers need up-to-date information to make informed funding decisions, but poverty data typically lag behind other key economic measures like unemployment, inflation, stock-market, and consumer-confidence indicators. Official poverty measure estimates reflecting the potential impact of the COVID-19 pandemic were not available until September 2021—18 months after the pandemic was declared a national emergency.

The official poverty measure is over 50 years old, but—despite its flaws—remains a widely used measure of Americans’ economic well-being. This article outlines a few of the alternative measures decisionmakers are currently using to assess economic need. However, COVID-19 showed us how quickly people’s economic circumstances can change during a crisis. To provide accurate, up-to-date poverty data for decision-making, researchers need to consider alternative methods of evaluating and tracking economic well-being based on real-time data from administrative records and other nontraditional sources.

For more information about the surveys the Census Bureau uses to measure poverty, read their page on surveys and programs.






1 Gordon M. Fisher, “The Development of the Orshansky Poverty Thresholds and Their Subsequent History as the Official U.S. Poverty Measure,” Census Working Papers (September 1997): 1-2.

2 Fisher, “The Development of the Orshansky Poverty Thresholds and Their Subsequent History as the Official U.S. Poverty Measure.”

3 U.S. Census Bureau, “Poverty Thresholds”; and U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, “2021 Poverty Guidelines.”

4 Amanda Lee, “U.S. Poverty Thresholds and Poverty Guidelines: What’s the Difference?” Population Reference Bureau,  March 25, 2018.

5 U.S. Census Bureau, “How the Census Bureau Measures Poverty,” November 22, 2021.

6 U.S. Census Bureau, “POV-01. Age and Sex of All People, Family Members and Unrelated Individuals, Below 100% of Poverty — All Races (1).” Note: The poverty universe includes all members of the resident civilian noninstitutionalized population of the United States, excluding unrelated children under age 15.

7 Liana E. Fox and Kalee Burns, “What’s the Difference Between the Supplemental and Official Poverty Measures?Random Samplings (blog), U.S. Census Bureau, September 9 2021.

8 Fox and Burns, “What’s the Difference Between the Supplemental and Official Poverty Measures?”

9 Liana Fox and Kalee Burns, The Supplemental Poverty Measure: 2020, U.S. Census Bureau, September 14, 2021.

10 U.S. Department of Agriculture, “SNAP Eligibility”; and U.S. Department of Agriculture, “WIC Eligibility Requirements.”

11 The Annie E. Casey Foundation, Kids Count Data Center, “Children in Extreme Poverty (50 Percent Poverty) in the United States,” September 2020; and U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, Financial Condition and Health Care Burdens of People in Deep Poverty, July 2015.

12 U.S. Census Bureau, “POV-01. Age and Sex of All People, Family Members and Unrelated Individuals, Below 50% of Poverty—All Races (1)”; and U.S. Census Bureau, “POV-01. Age and Sex of All People, Family Members and Unrelated Individuals, Below 200% of Poverty—All Races (1).”

13 Administration for Children and Families, Office of Community Services, Low Income Home Energy Assistance Program (LIHEAP) Fact Sheet, 2021.

14 U.S. Census Bureau, “Differences Between Available Surveys/Programs for Poverty.”

15 Caroline Danielson et al., The California Poverty Measure: A New Look at the Social Safety Net, Public Policy Institute of California, October 2013.

16 Oregon State University, “The Oregon Poverty Measure”; and New York City Government, Appendix B: Deriving a Poverty Threshold for New York City, 2018.

17 University of Washington, Center for Women’s Welfare, “Overview: What Is the Self-Sufficiency Standard?