AP Photo/Jose Luis Magana
Just how many Americans are poor? The Trump administration wants you to believe that just 3 percent of the U.S. population is poor. The Council of Economic Advisers made that claim in a little-noticed report published earlier this summer, as part of a coordinated effort to justify harsh new restrictions on government assistance programs.
This bad-faith estimate emerged more from a desire to hurt the poor than to engage in honest policymaking on the issue of poverty. In March, the Congressional Budget Office found that nearly half of social safety net payments are going to people that the federal government once considered “middle class.”
Think about that: At a time when even the middle class is starting to look poor, the administration argues that not even the poor are poor so that federal officials can move to cut programs that both groups now rely on to stay afloat.
So what’s behind this numbers game?
How we measure poverty in the United States has actually been the subject of a long debate that arises out of the tendency by policymakers, demographers, and others to undercount poor people.
The story begins in 1963. The Census Bureau developed the “official poverty measure” (OPM) that year—that’s the “poverty line” that Americans hear so much about. It was calculated at three times the food budget of a family of a given size, based on food expenses data from 1955.
Today, not even the Census Bureau views the OPM as an accurate measure of who is poor in the U.S., even though it continues to churn out the number each year. As Vox’s Dylan Matthews put it in 2014, “The way we measure poverty is based on a 51-year-old analysis of 59-year-old data on food consumption, with no changes other than inflation adjustment. That’s bananas.” As Matthews points out, the OPM is an outdated, overly blunt instrument.
Starting in the 1990s, the Census Bureau developed the Supplemental Poverty Measure (SPM) as an alternative. Like the OPM, the Supplemental Poverty Measure looks at the expenses of a given family, but it takes into account many other details—such as the costs of clothing, shelter, utilities, the impact of government benefits programs, and much more.
This measure, while still an imperfect undercount, is an improvement over the OPM in that it takes into account factors like the impact of regional differences in certain indicators such as the cost of living and homeownership status. The federal government continues to use the OPM to establish guidelines for government programs and influence other policy.
Enter Trump. Faced with these two imperfect ways of counting poor people in the United States, the Council of Economic Advisers attempts to argue that not only are the OPM and the SPM both wrong, but they both heavily overcount the extent of poverty. Rather than looking at the income of poor people to define poverty, the council argues that the level of consumption by families is a more accurate measure. That is, poor people are only poor if they spend less than some minimum threshold of dollars.
The startling claim that poverty in the United States had steadily shrunk to 3 percent by 2016 doesn’t hold up.
The foundation for that proposal is a 2017 academic paper by economists Bruce D. Meyer and James X. Sullivan. Their measure of what they call “consumption poverty” anchors the dollar amount threshold to a value that produces a rate equivalent to the official poverty rate in 1980, with adjustments for inflation. They concluded that “consumption poverty” had fallen to 3 percent, a claim then echoed by the White House.
The authors of the study note that the choice of 1980 is “arbitrary.” But if the researchers linked the poverty rate by consumption to the official poverty rate in 2015 (which would produce a figure more anchored to current costs), they would arrive at a poverty rate of 12.6 percent—four times the White House estimate!
But a consumption-based measure of poverty doesn’t work either. University of Michigan researchers evaluated the OPM, the SPM, and the Meyers-Sullivan consumption model by looking at common real-life poverty indicators (like, say, not being able to afford food).
They compared rates of food insecurity, the inability to meet essential household expenses, and unemployment with the rates of poverty over time estimated by the OPM, the SPM, and the consumption-based model. How did these rates of material deprivation, they wondered, compare with the different estimates of who was “poor”?
The consumption model favored by the White House, they found, would lead to the erroneous conclusion that poverty was lower during the Great Recession than in the early 2000s—even though food insecurity, non-food material, and medical hardships were all higher.
While not perfect, the OPM and the SPM tracked well with actual measures of deprivation.
The startling claim that poverty in the United States had steadily shrunk to 3 percent by 2016 doesn’t hold up. But the Council of Economic Advisers is using it to argue that there are fewer poor people than ever, so government programs to assist the poor are no longer necessary. Their claim is breathtaking in its sheer audacity.
But it’s worse than that, since all three of these measures still significantly undercount the poor. That was the unmistakable conclusion of “The Souls of Poor Folk,” a 50-year audit of poverty in the United States that Aaron Noffke and I conducted for the Institute for Policy Studies and the Poor People’s Campaign.
We looked at a much larger group of people affected by issues of poverty, systemic racism, militarism and the war economy, and ecological devastation—a measure that meant doubling the SPM threshold.
By that doubled measure, we found that 42.5 percent of Americans—or about 140 million people in the U.S.—are poor or low-income.
The poor or low-income threshold equates to about $47,000 a year for a family of four, with significant differences between geographic location and homeowner status. (For instance, renting in Silicon Valley compared to rural Alabama would require an extra $33,000 a year to not be counted as poor or low-income.)
We looked at this broader category for a simple reason: The experience of people struggling to meet basic economic needs extends beyond the income thresholds that officially determine poverty. Measuring poverty first requires defining the minimum set of requirements that enable an individual or family to enjoy a modest but dignified way of human life. American poverty rates should reflect how many people are living outside of those parameters.
For instance, 55 percent of Americans struggle to pay medical bills, 40 percent of adults have less than $400 in savings, and nearly half of renters can’t afford housing costs. A poverty rate of 12.7 percent, the OPM for 2016—or 14 percent, the SPM for 2016—excludes vast chunks of these individuals. Doubling the SPM threshold does not.
Defining poverty in narrow terms—whether through the OPM, the SPM, or through Trump’s farcical claim—sends this message: Debilitating economic conditions are perfectly acceptable to the broader society.
So why did Trump’s Council of Economic Advisers put out a bogus statistical measure on poverty that undercounts poor people in the United States? Simply put, the report was designed to help stigmatize the poor and low-income people; to attack the programs that barely keep them afloat today, and to provide a glossy academic cover for the entire disturbing exercise.
Americans should not pretend poverty doesn’t exist. Instead, people need to demand that the federal government expand programs that have make inroads against poverty, such as Social Security, the Earned Income Tax Credit, and SNAP benefits—while creating new programs to reduce household debt, promote wage growth, and to guarantee basic rights of economic well-being.
Saurav Sarkar and Aaron Noffke led the research team for “The Souls of Poor Folk,” an audit of poverty, systemic racism, militarism, and ecological devastation in the United States, for the Institute for Policy Studies and the Poor People’s Campaign.
