Annual report on US consumption poverty: 2017 – AEI – American Enterprise Institute: Freedom, Opportunity, Enterprise2nd November 2018
Download the Report
This report presents estimates of consumption and income based poverty in the United States derived from information collected in the U.S. Bureau of Labor Statistics’ Consumer Expenditure Survey and the U.S. Census Bureau’s Current Population Survey. A poverty rate visualization tool, additional results, and resources can be found at povertymeasurement.org.
Summary of findings
- Using the standard of living of the poor in 1980, the consumption poverty rate fell from 13.0 percent in 1980 to 2.8 percent in 2017, while the official poverty rate fell by 0.7 percentage points over that period. 0.2 percentage points of the overall drop in consumption poverty occurred between 2016 and 2017.
- Using the standard of living of the poor in 2015, the consumption poverty rate fell from 32.9 percent in 1980 to 12.1 percent in 2017. Using this higher standard, 0.6 percentage points of the overall drop in consumption poverty occurred between 2016 and 2017. Note: The standard of living of those at the poverty line has changed over time because the official poverty thresholds are adjusted using a biased price index. For simplicity, our main results are reported using the 1980 historical standard. The choice of year for the historical standard is arbitrary, as was the initial official standard, a point which we discuss more fully below.
- The 2017 decline in the consumption poverty rate continues a trend that started in 2011, when poverty began to fall after rising for two years during the Great Recession.
- Three factors explain why consumption poverty shows a long-term decline but the official poverty measure does not: the consumption poverty measure 1) is constructed using a bias-corrected measure of inflation; 2) implicitly incorporates taxes and in-kind transfers in family resources by using consumption; and 3) avoids the bias due to the under-reporting of certain types of income that are commonly received by those with low reported income.
The Office of Management and Budget established the procedure for measuring the official poverty rate in the United States through a Policy Directive in 1978. This official rate is determined by comparing the pre-tax money income of a family or a single unrelated individual to poverty thresholds that vary by family size and composition. For example, in 2017, the poverty threshold for a one-parent, two-child family is $19,749. The underlying data on pre-tax money income come from the Current Population Survey Annual Social and Economic Supplement. If a family has income below the poverty cutoff for that size family, all family members are classified as poor. Except for a few minor changes, the only adjustment to these thresholds over the past five decades has been for inflation using the Consumer Price Index for all Urban Consumers (CPI-U).
The release of this report is motivated by several longstanding criticisms of the Official Poverty Measure (OPM). Many criticisms can be found in sources such as Citro and Michael (1995), Blank (2008), and U.S. Census Bureau (2016b), but two are probably of greatest importance. First, the price index that the OPM relies on to adjust the poverty thresholds for inflation, the CPI-U, is known to overstate the extent of inflation (e.g. Hausman 2008). This problem can be addressed by using an unbiased price index.
Second, the OPM does not reflect in-kind transfers and tax credits that have grown over time, such as the Supplemental Nutrition Assistance Program (SNAP), housing benefits and the Earned Income Tax Credit (EITC). The first problem means that the poverty cutoffs rise too quickly over time, leading more and more people to be below the cutoff in the absence of countervailing increases in income. The implication of the second problem is that the OPM fails to reflect the full array of resources, cash and noncash, that families can use to meet their needs.
A potential solution to the second problem is to include SNAP, housing, tax credits, and other benefits in the measure of income used to determine poverty status. This is the approach taken in the Census Bureau’s Supplemental Poverty Measure (SPM). Unfortunately, the third problem with the OPM, income underreporting, plagues the SPM as well leading. The survey data sources for government benefits suffer from substantial reporting problems, and consequently they substantially understate the in-kind and tax benefits mentioned earlier: SNAP, housing benefits, the EITC (Meyer, Mok and Sullivan 2015; Meyer and Mittag forthcoming). Some forms of income included in the OPM are also sharply under-reported such as cash welfare, pension income (Bee and Mitchell 2017), and earnings for those at the very bottom (Meyer, Wu, Mooers and Medalia (2018). Thus, the SPM only addresses one of three major problems with the OPM, and because it makes a partial correction for one problem, while leaving the others in place it can do more harm than good (Meyer and Sullivan 2012; Meyer, Wu, Mooers and Medalia 2018).