:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/58534181/GettyImages_545133226.0.jpg)
On January 24, the New York Times published an opinion piece by Angus Deaton, a Nobel Memorial Prize-winning economist, in which he claimed that millions of Americans — specifically, 3.2 million to 5.4 million, depending on the poverty line used — “are as destitute as the world’s poorest people.” This is simply wrong.
To the extent that this empirical claim, by an esteemed Princeton scholar, leads policymakers to reduce international aid, or causes charitable donors to redirect their money away from the world’s most impoverished people, it is also dangerous.
Deaton flirts with the idea of redirecting money in his piece. “In my own giving, I have prioritized the faraway poor over the poor at home,” he writes. But recently, as the result of “insightful new data,” he has “come to doubt both the reasoning and the empirical support” for that view. He similarly questions why the World Bank, USAID, and Oxfam prioritize non-Americans (even as he also notes, “None of this means that we should close out ‘others’ and look after only our own”).
Shameful levels of poverty do exist in the United States
It is true that America has serious problems of poverty and inequality. These inequities have resulted in quite shocking outcomes for America’s poorest. For example, 34 percent of households surveyed in Lowndes County, Alabama, recently tested positive for hookworm. Hookworm transmission occurs by way of feces and is easy to avoid if one has modern sanitation, but in the same county, 42 percent of the sampled households were exposed to raw sewage within their home. This is a national disgrace, and it highlights the dire plight of America’s poor.
Nevertheless, it is incorrect and misleading to draw an equivalence between poverty in America and poverty in low-income countries. It is only through the misinterpretation of poverty statistics that one can equate the two. Let me explain how Deaton is misusing data here (and he is not the only one to make this error).
Let’s start with the purely economic side of poverty. In order to measure poverty, we need to survey people and record how much they “earn.” There are two main ways of doing this. The first, common in low-income countries, is to ask people about their consumption and then derive a dollar figure from their answers. The second approach, more common in high-income countries, is to simply ask people about their income.
There are many problems in comparing data across these different types of surveys. The largest is that poor people in rich countries often receive many non-cash benefits that boost consumption without boosting income — for instance, in the US, the Supplemental Nutrition Assistance Program (SNAP).
In one analysis, the non-cash benefits provided to American households with near-zero income increase their household consumption by an average of about $20 a day. A very well-regarded book on the analysis of household surveys notes that “survey-based measures of income are often substantially less than survey-based measures of consumption […] even in industrialized countries.”
The World Bank, which runs many of these surveys, has noted the dangers in comparing income and consumption-based poverty figures. In one report, its experts observe that many of the people who “declare zero income on a survey” have “a consumption level that is not zero.”
Nevertheless, people keep making this mistake. For example, Kathryn Edin and Luke Shaefer, of Johns Hopkins and the University of Michigan, have claimed that millions of Americans live on less than $2 a day, the threshold used by many international agencies for determining extreme poverty. They use income-based surveys to measure poverty and ignore programs like SNAP. (When Laurence Chandy, now of UNICEF, and Cory Smith, now an MIT PhD student, redid Edin and Shaefer’s calculations using a more comparable consumption survey, they found that almost nobody in the United States lives on $2 a day.)
Even if the extreme poverty level in America is set at $4 a day, Deaton’s claim doesn’t hold up
Deaton makes the same fundamental error: His American poverty figures measure income, but the poverty figures for poor countries measure consumption. Citing the Oxford economist Robert Allen, Deaton also argues that the extreme poverty line for Americans should be higher than $2 a day, perhaps even as high as $4 a day, because “[t]here are necessities of life in rich, cold, urban and individualistic countries that are less needed in poor countries.” For instance, people in warm countries may not need housing, he says, and “a poor agricultural laborer in the tropics can get by with little clothing and transportation.”
These are debatable claims (and Allen’s work on the subject has come under a lot of scrutiny), but even if we grant a higher $4 a day poverty line for Americans but use apples-to-apples consumption-based poverty measures, then it turns out that America still has only a tiny fraction of its population in extreme poverty. The fact that anyone in the US lives on less than $4 day is a genuine tragedy, but Deaton’s count of 3 million to 5 million Americans in extreme poverty is off by an order of magnitude.
There are also problems with Deaton’s claims regarding the health of Americans. Anne Case (also of Princeton) and Deaton have claimed to find “a marked increase in the all-cause mortality of middle-aged white non-Hispanic men and women in the United States between 1999 and 2013.” Deaton repeats this claim in the new op-ed. But a closer look at the data reveals that rising mortality rates appear to be confined to middle-aged, white, non-Hispanic women, especially those in the American South.
It’s a bit curious why we talk so much about the health problems of non-Hispanic white people, and so little about the fact that African Americans still have lower life expectancy than whites, despite making major health gains. Regardless, America could do much better on health, and Case and Deaton’s results are worthy of serious attention. Still, Deaton generalizes these issues in misleading ways.
Economic measures — even the accurate ones — don’t capture every aspect of poverty
Finally, the very act of living in America provides many benefits that are not generally captured in poverty measures but that enable one to live a better life. America is not experiencing civil war. The American political system is highly imperfect and under stress, but it is considerably better at protecting liberties, providing services, and enabling representation than the political systems in many low-income countries. These often intangible benefits help people lead fuller lives, even if they are often not considered when discussing poverty, and they overwhelmingly lean in America’s favor.
America’s wealth also means that it can help the poor within its own borders without cutting foreign development aid — an idea Deaton seems to put on the table — which only amounts to about 1 percent of the federal budget in any case. It can do so by reorienting some of the remaining 99 percent of the federal budget to better help the poor, by reorienting portions of state and local resources, and by raising revenue in ways that lean relatively more on the rich.
In sum, America indeed has very serious problems with poverty and inequality. But it is wildly inaccurate to claim that millions of Americans “are as destitute as the world’s poorest people.”
Ryan Briggs is an assistant professor of political science at Virginia Tech. His research focuses on the political economy of poverty alleviation. Find him on Twitter @ryancbriggs.
The Big Idea is Vox’s home for smart discussion of the most important issues and ideas in politics, science, and culture — typically by outside contributors. If you have an idea for a piece, pitch us at thebigidea@vox.com.
