Matthew Desmond, a Princeton sociologist who has won a MacArthur "genius" grant as well as a Pulitzer for his previous book, Evicted, devotes his latest work to expounding a "theory" that will explain "why there is so much poverty in this land of abundance," the United States. According to Desmond, the American people as a whole are its cause.
Without offering a precise definition or measure of poverty, Desmond laments that America, the world's richest country, has "more poverty than any other advanced democracy." (That the United States is also by far the most populous advanced democracy might itself help to explain this alleged fact.) Relying chiefly on a report he had published in 2015, along with statistics issued by the Census Bureau and the OECD, Desmond asserts that "almost one in nine Americans" live in poverty, with over 38 million unable to "afford basic necessities" and another 108 million "getting by on $55,000 a year or less," "stuck in that space between poverty and security."
Of course a large majority of the world's population—including India, Pakistan, Indonesia, China, most of Africa, and much of Latin America—would envy an individual or even a family with an income half that large. And as Desmond acknowledges, the official definition of poverty, developed by Mollie Orshansky, an economist for the Social Security Administration in the 1960s, was a seat-of-the-pants calculation. Since poverty meant lacking enough income to cover life's basic necessities, with food the greatest need, she judged that because food expenditures (at the time) consumed roughly a third of an average family's budget, the poverty threshold as of 1965 should be set at $3,000 a year (assuming the family would need to spend $1,000 on food). This entailed that in 1965, 50 million Americans, including 22 million children, were poor. As Desmond observes, the government's Official Poverty Measure (OPM) is still based on Orshansky's calculations, adjusted for inflation, so that the poverty line in 2022 was set at $13,500 for a single person and $27,750 for a family of four.
Continued reliance on Orshansky's calculations, as economists like Nicholas Eberstadt and Phil Gramm have pointed out, has led to increasing distortions—for instance, higher agricultural productivity has reduced the proportion of income needed for food, and the national figure abstracts from wide geographic differences. And as Orshansky herself acknowledged, the OPM greatly overestimates poverty by counting only cash income, excluding such forms of public aid as housing assistance, Medicaid, and the "refundable" Earned Income Tax Credit (all of which have expanded since the 1960s).
While the OPM has been accompanied since 2011 by a Supplemental Poverty Measure, with a national panel of "experts" proposing an update last May that would supplant it, the panel's proposal itself suffers from serious problems of ideological bias, as explained in a recent analysis by Scott Winship of the American Enterprise Institute.
Another deficiency is that the OPM gives a misleadingly static impression of poverty, as if the same people remained at the lowest level throughout their lives. As Eberstadt noted in an essay on "The Mismeasurement of Poverty," given the dynamic character of the American economy, long-term rather than temporary poverty "appears to be the lot of only a tiny minority of the people counted as poor by the official U.S. poverty metric." Consider the fact that mostly poor immigrants are regularly arriving to fill bottom rungs of the economic ladder, while others move up—and many penurious graduate students will certainly rise in economic status once they complete their studies.
While Desmond uses the OPM to deny that there has been any lasting reduction in poverty over the past half-century, his own long note cites evidence that belies that claim. Notably, he cites research using an "anchored" poverty measure (that is, one based on actual living standards, rather than cash income), showing that poverty fell by roughly 40 percent over the past 50 years, and a "widely cited" report by a nonpartisan organization that found child poverty had declined by 59 percent just between 1993 and 2019. Desmond even acknowledges the evidence that government antipoverty programs have thus "borne fruit." But he carps that the progress took place mostly during two periods totaling 15 years rather than being steady, and adds his "feeling" that what really matters is that urban poverty has grown proportionately since 2000 (which means that non-urban poverty declined by even more than the overall average). Clearly, Desmond is grasping at straws to find statistics that justify his feeling.
In reality, the data that Desmond cites to estimate the extent of poverty are thoroughly flawed, as is demonstrated in the 2019 AEI "Report on U.S. Consumption Poverty" by economists Bruce Meyer and James X. Sullivan. This report corrects the "official" poverty measure by relying on patterns of consumption rather than far less reliable estimates of income. The OPM exaggerates the extent of inflation from 1980 to 2018; as already noted, it excludes non-cash sources of support such as housing vouchers, Medicaid, food stamps, and tax credits; and it ignores the underreporting of income by those classified as poor. By contrast, Meyer and Sulllivan conclude, between 1980 and 2018 the consumption poverty rate fell by over 10 percentage points—from 13 percent in 1980 to only 2.8 percent in 2018. (This despite the fact that the "official" rate fell by only 1.2 points during that period.)
The core of Desmond's argument, however, doesn't depend on statistics, but on the claim that Americans as a whole are engaged in "exploiting" the poor, a practice that will not end until we resolve to become "poverty abolitionists." Unlike most social scientists, Desmond rejects attributions of the persistence of real poverty to factors like the rise of single motherhood (an issue identified by Daniel Moynihan in his 1965 report The Negro Family—a problem that has greatly increased since then among African Americans, even as it has spread among the white lower class as well). Nor does he concern himself with the rotten state of inner-city schools: He discusses public education only to lament the insufficient degree of racial integration, rather than recommending the expansion of charter schools and school-choice programs, which offer poor families a real alternative to union-run public schools. (Instead, he favors policies to increase unionization throughout the economy.)
Desmond justifies his attribution of poverty to "exploitation" by viewing the economy as a zero-sum game, in which every gain by people in higher economic classes must have been won by depriving poor people of some good to which they were entitled. He accuses better-off people of hoarding wealth through their "excesses"—for instance, by buying "cheap goods and service" produced by "the working poor," investing in companies that produce such goods, enjoying free checking accounts that are "subsidized" by overdraft fees paid by others, and electing officials who "fabricate stories" about not making the poor dependent on government, "sinful[ly]" warning that "proposals to reduce poverty … would cost too much." (Actually, "cheap" goods sold by stores like Walmart and Costco are far more likely to be purchased by the less well-off; overdraft fees are easily avoided by balancing one's checkbook; and Desmond offers no remedy for the real problem of dependency.)
Disregarding the vast scope of programs at all levels of government that already serve the poor, Desmond is confident that much larger such programs could be financed by "crack[ing] down on corporations and families who cheat on their taxes" and compelling the well-off to "take" less from the government. To do this, he recommends "bump[ing] up the top marginal [federal] tax rate … to 50 percent, as it was in 1986; or 70 percent, as it was in 1975," while jacking up the corporate rate to 46 percent, "as it was from 1979 through 1986."
Desmond seems unaware of the many deductions from taxable income that were eliminated by the 1986 Tax Reform Act—unlimited medical deductions; liberal home-office deductions; tax averaging; and so on, which made the "official" top rates merely nominal. And he denies that any "serious social scientist" thinks that raising taxes on the rich or international corporations would have a disincentivizing effect, reducing national wealth and the resources to raise everyone's living standards. Note, however, that Ronald Reagan's 1981 tax cut, the 1986 reform, and Donald Trump's 2017 cut all unleashed economic booms, with African-American poverty falling to a record low in the Trump years. Do "serious social scientists" ignore these facts?
Another of Desmond's proposals, taxing capital gains at the same rate as regular income, ignores the fact that since the former aren't indexed for inflation, seeming gains may not constitute actual increases in wealth—and the tax code allows only limited opportunities to balance them against capital losses. And he endorses broadening the scope of antipoverty programs so as to benefit the middle class as well, just to make it politically difficult (as FDR planned with regard to Social Security) to repeal them. But Desmond refuses to call his proposals redistributive, since he "hate[s] the word." Instead, they are simply compensation for the government's considerable "effort" devoted to enriching "the American aristocracy."
Other elements of Desmond's agenda include such conventional left-liberal proposals as indexing the minimum wage for inflation or even making it illegal for companies to pay "degrading wages." (Currently, only 2 percent of American workers earn the minimum wage, and raising it would reduce the opportunity for teenagers and recent immigrants to obtain needed work experience.) To his credit, Desmond criticizes "snob zoning" laws, a point on which advocates of free markets agree.
But perhaps the most striking aspect of Desmond's argument is his claim that people who object to expanded government welfare programs are hypocritically ignoring the fact that they themselves "benefit most from government largesse" in the form of tax "breaks" like the mortgage-interest deduction (greatly reduced by the 2017 Tax Cut and Jobs Act), while also enjoying "employer-sponsored health insurance." In effect, Desmond starts with the premise that all wealth belongs in principle to the government, so that any reduction in the taxes one pays, or any benefits one receives on the job, amounts to "taking" sums, part of which rightfully belong to other people. This is the opposite of the liberal principles articulated in the Declaration of Independence.
Near the end, citing the nation's "bounty," Desmond expresses agreement with an "ecologist" who advocates "an economy of abundance … in which wealth means having enough to share," and "where the gratification of meeting your family needs is not poisoned by destroying that possibility for someone else"—as if that is what non-poor Americans were currently doing. While "poverty," according to Desmond, "is the dream killer," he concludes that "[w]e don't need to outsmart this problem. We need to out-hate it." This statement helps explain many of the intellectual flaws of Poverty, by America. Whether it justifies the cost of a Princeton education I must leave to tuition-paying parents to decide.
Poverty, by America
by Matthew Desmond
Crown, 304 pp., $28
David Lewis Schaefer is a professor of political science at the College of the Holy Cross.