Based on the decline in the unemployment rate and other recent trends, it is likely that the official poverty rate will fall from 12.3 percent in 2017 to around 12 percent in 2018. The Census Bureau will release those poverty estimates tomorrow, Tuesday, September 10.

Regardless of where the 2018 rate ends up, it will vastly understate the extent of economic and social deprivation in the United States. But, with a better measure of poverty, one more consistent with most Americans’ understanding of how much income it takes not to be poor, the poverty rate is about 17.7 percent. Roughly 17 million more Americans would be counted as poor than are currently counted as poor using the official measure.  

What reason is there to think that the official poverty measure so massively understates poverty? In a nationally representative survey conducted in 2016 by the conservative American Enterprise Institute (AEI) and the Los Angeles Times, Americans were asked, “[What is the] highest annual income a family of four can have and still be considered poor by the federal government?” The average response was $32,293 — an amount 33 percent higher than the official poverty threshold for a family of four that year ($24,339). When Americans were asked to name the “minimum amount of income” needed to “get along in your community,” the average response was even higher, probably because the word “poverty” is not used. 

These survey responses are much closer to the poverty threshold used by the OECD for international comparisons of poverty in wealthy countries: 50 percent of median disposable income. In 2016, a family of four in the United States needed a disposable income of about $35,000 not to be poor according to this standard. 

Poverty lines tied to a percentage of median income are typically referred to as “relative” measures of poverty. The basic idea is that people with less than half of the income of people at the middle of the income distribution are relatively deprived, even if they’re not starving or homeless, and that relative deprivation has harmful effects on social participation, well-being, and life chances. 

A relative measure of poverty does not pretend to be a purely scientific measure of the minimum income needed for subsistence. This is a good thing because it’s absurd to think that poverty is a purely scientific concept; it’s also a social and political one. 

Related advantages include transparency and relevance, especially compared to the current official measure. The official poverty measure is set at three times the value of an inflation-adjusted “economy food plan” developed by USDA for use in the early 1960s based on data from a 1955 food consumption survey. However relevant and publicly transparent this was as a measure of poverty in the 1960s, it has little meaning today for an American population in which most people were born after 1980. 

If we set the poverty threshold equal to 50 percent of median disposable income, the threshold for one person in 2017 would be $17,652. Using a standard adjustment for economies of scale based on household size, the 2017 relative poverty threshold for a household of two would be $24,964 and $35,304 for a household of four. This compares to official poverty thresholds of $12,228, $15,569, and $25,094, respectively, in 2017. While the official poverty rate in 2017 was 12.3 percent (about 39.7 million people), it would be 17.7 percent that year using the relative measure (about 57 million people). Among children, more than one-in-five (21.2 percent) were poor in 2017 using the relative measure, compared to 17.5 percent using the official measure.

Why is the official poverty line so much lower than contemporary public understanding of poverty and the 50 percent of median standard? Notably, when the official poverty line was first developed in the early 1960s, it was roughly equal to a poverty standard set at 50 percent of median income. But because the measure has never been re-benchmarked to take increases in mainstream living standards into account — and these standards, as measured by median income, increase at a faster rate than inflation over time — the official poverty line has fallen farther and farther behind. 

By contrast, Canada, Ireland, and the United Kingdom, among other wealthy nations, utilize modern, transparent measures that could be easily adopted by the Census Bureau. Canada, for example, uses both a “Low Income Measure” — set at 50 percent of median disposable income” — and a “Market Basket Measure” based upon the cost of a basket of food, clothing, shelter, transportation, and other items needed to have a “modest, basic standard of living.”

In short, the dominant framework for measuring poverty in the United States has become an ideologically conservative one. The official poverty measure conceptualizes poverty only in terms of having an extremely low level of annual income and utilizes poverty thresholds that are adjusted only for inflation rather than for changes in overall living standards. As a result, the official poverty measure has effectively defined deprivation down, moving it further and further away from mainstream living standards over time, as well as from majority public opinion of the minimum amount needed to not live in poverty. The federal government should stop using the official poverty measure for statistical purposes and modernize its approach to measuring low income and social deprivation.