A new study suggests that programs like food stamps and housing vouchers cut poverty by almost twice as much as we thought they did.
According to standard census numbers, the poverty rate in New York from 2008 to 2011 was 13.6 percent, before taking these programs into account. The programs the study examines — food stamps (a.k.a. SNAP), welfare (a.k.a. TANF), state-level general assistance programs, and housing aid — dropped that down to 10.8 percent. But the study suggests the real number was even lower: a mere 8.3 percent. If that's true, then the estimated poverty-fighting power of these programs has been dramatically understated for years.
People really don't like telling the census they get government help
The official data on poverty comes from the Current Population Survey's (CPS) Annual Social and Economic Supplement (ASEC), a yearly survey that reaches about 100,000 addresses and yields detailed information on household incomes, education levels, health insurance, and use of government assistance programs. In theory, it should tell us everything we need to know. But all surveys come with a risk of error, and ASEC, in particular, has been getting worse in recent years. Fewer people in the sample fill it out, and fewer people who fill out some of it fill out the income portion. This means it misses a lot of money people are receiving from the government. "The underreporting of income from government programs is a problem that is well-known in the CPS, and which has grown over time," Christopher Wimer, a poverty researcher at Columbia and lead author of a landmark 2013 paper on the war on poverty, writes in an email.
Luckily, there's another source of data on how much money households are getting from the government: the government itself. So University of Chicago economist Bruce Meyer, who has produced a lot of the research identifying this problem, and CERGE-EI's Nikolas Mittag got microdata from the New York State Office of Temporary and Disability Assistance and the federal Department of Housing and Urban Development to see how much various households in New York state were getting from food stamps, welfare, and housing assistance. They then matched that data with the ASEC survey data, so they could compare what respondents said they were getting in government assistance with what they were actually getting.
They found that more than a third of people getting housing assistance, 40 percent of people getting food stamps, and 60 percent of people getting TANF or state-based General Assistance didn't say so when surveyed for ASEC. And even people who said they got assistance underestimated how much they were getting, by 6 percent for food stamps, 40 percent for TANF/General Assistance, and 74 percent for housing assistance. In other words: Our survey data on how many people are using these programs, and how much they're getting, is really, really wrong.
How wrong? Well, from 2008 to 2011 in New York, the average poverty rate before taking these programs into account was 13.6 percent. According to the survey data, food stamps/welfare/housing assistance dropped that down to 10.8 percent. But according to the more accurate administrative data, those programs actually cut the rate to 8.3 percent. Using the right numbers, in other words, nearly doubles the poverty-fighting power of these programs. The differences for "deep poverty" — that is, the share of households living on 50 percent of the poverty line or less — are a bit smaller, but for "near poverty" (the share of households living on 150 percent of the poverty line or less) they're even bigger, as the above chart suggests.
The survey's shortcomings are even greater when you look at households headed by a single mother, a particularly vulnerable group with a poverty rate of 37.5 percent before these programs are taken into account. The survey data suggests that welfare/food stamps/housing assistance cut that down to 30.2 percent. The administrative data suggests they actually cut it to 19.2 percent.
The big lesson: We've made more progress than we thought
Papers like this can seem a tad technical, but they're part of a broader trend in which sociologists and economists are reevaluating the legacy of the war on poverty and the safety net that it produced, and finding that we've had a lot more success than we'd thought. You can see this in Wimer's paper from 2013, which found a massive reduction in poverty from 1967 to 2012, but only when you take government programs into account. The green line shows how poverty shifted before taking taxes and transfers into account; the blue line shows how it changes after adding in those programs:
You can also see this trend in a 2012 paper by Meyer and James X. Sullivan, which measured poverty by looking at how much people reported consuming, as opposed to how much they reported earning. That makes it easier to take into account what government programs help you buy. Meyer and Sullivan found that poverty, measured thusly, fell from 30.8 percent in 1960-'61 to 4.5 percent in 2010.
And you can see the trend in a piece in the New York Review of Books from earlier this year by Christopher Jencks, a hugely respected poverty researcher at Harvard's Kennedy School. Jencks did some back-of-the-envelope math to deal with the fact that the official poverty rate doesn't even include non-cash benefits like food stamps, that it uses a less accurate inflation index, and that it doesn't include tax benefits like the earned income tax credit. He found that after you do that, the poverty rate in 2013 was more like 4.8 percent — far lower than either the 14.5 percent official number that year or the more directly comparable 19 percent figure from 1964, before the war on poverty.
Findings like this shouldn't breed complacency. There are still millions of Americans living in truly desperate poverty (and billions of people around the world in even worse poverty), so these programs clearly haven't accomplished everything we want them to. But the emerging, positive view of transfer programs is an indication that we know how to tackle this problem, and that even greater transfers could conceivably end poverty in America altogether.