The official poverty measure (OPM) in the U.S. fell from 22.4 percent of the population in 1959 to 10.5 percent in 2019, which suggests that economic insecurity—at least as measured by the portion of people who are in poverty—has fallen. These gains, however, were not steady or evenly shared. Figure 1 shows the percent of people in the U.S. who have been in poverty since 1959 (there are no data for certain subgroups—1959 and 1967, which is linearly interpolated with a dotted line).The figure also shows poverty rate by age groups—under 18, 18–64, and 65 and over. For most of the period here, age sixty-five was the full retirement age for Social Security Old Age Insurance benefits.
For children and adults under sixty-five, the reduction in poverty occurred in the initial period of measurement, from 1959 to 1973. In the nearly five decades since, the poverty rate has fluctuated between 11 percent and 15 percent. For adults over sixty-five, in contrast, the proportion in poverty has been on a slow and steady decrease, falling to 8.9 percent in 2019.
Breaking down poverty by race reveals deep disparities. In 2019—a record year of poverty lows under the OPM—18.8 percent of the Black population, 15.7 percent of the Hispanic population, 7.3 percent of the non-Hispanic White, and 7.1 percent of the Asian population were in poverty. Over the thirty years from 1989 to 2018, OPM poverty rates averaged 26.4 percent for the Black population, 24.4 percent for the Hispanic population, 11.9 percent for the Asian population, and 8.8 percent for the non-Hispanic White population.
Reaching an overall poverty rate of 10.5 percent in 2019 was a milestone, but experts expect data to show an increase in poverty in 2020 due to the COVID-19 pandemic.
The official poverty rate cannot be interpreted as an exhaustive estimate of economic insecurity. There are two key problems.
First, the official measure is generally recognized as flawed. Specifically, it counts the number of people who have cash income below a certain threshold, but that threshold only increases with inflation. Wages rise faster than prices (this is how standards of living increase), so although the threshold increases every year, the actual number becomes less meaningful as a measure of poverty relative to trends in the economy. In 1959, for example, the family of four threshold was $2,973, which represented 55 percent of median family income. In 2019, the threshold was $26,172, or 38 percent of median family income.
The other problem with the official poverty rate is that it only measures pre-tax cash income, and therefore does not capture post-tax transfers or noncash support programs. So Social Security, a cash benefit not subject to the income tax, is counted, but the Earned Income Tax Credit (EITC) and Supplemental Nutrition Assistance Program (SNAP, initially Food Stamps) are not. This means that the official poverty measure accurately reflects cash income but not all income resources. The U.S. Census Bureau developed the Supplemental Poverty Measure (SPM) to take those post-tax transfers and in-kind benefits into account (among numerous other changes). Prior research found that under the SPM accounting, these income support programs have powerful anti-poverty effects that have resulted in significant reductions in poverty rates in recent decades. Still, that accounting found that 11.7 percent of the population was poor in 2019.
An alternative definition of poverty is used by the Organisation of Economic Co-operation and Development (OECD); it defines poverty as 50 percent of median income in a given country. According to this definition, the poverty rate in the U.S. after taxes and transfers was 17.8 percent in 2017.
If we were to assume that poverty was an exhaustive measure of economic insecurity in the U.S., we would then be left with the conclusion that the U.S. labor market has made little progress in reducing economic insecurity over the past forty years and has been greatly bolstered by public support programs.
But it is incorrect to equate poverty with insecurity. Poverty measures income, not precarity, though it is arguably the best measure of precarity at our disposal. It should be considered necessary but not sufficient; that is, even if we assume that nearly all individuals in poverty are economically insecure, there are individuals who are not in poverty who are also economically insecure.
During the pandemic, it was not solely the 10.5 percent of the population that was in poverty at the start of 2020 that was affected by the recession. In December 2020, 37.5 percent of adults reported that it had been at least somewhat difficult paying for “usual household expenses,”, 13.7 percent of adults were in households where there was not enough food to eat sometimes or often over the previous week,, and 18.1 percent of renter-occupied households were behind on rent. Of those behind on rent, more than 52 percent reported it as at least somewhat likely that they would be evicted in the next two months. Precarity extended far above the threshold for poverty.