The political mean season has returned, with all the usual slanders of social programs that protect Americans from the worst kinds of hardship and deprivation. The poors! They’re behaving badly! And government handouts paid for with your tax dollars are to blame. In the end, we are told, the well-intentioned policies of meddling do-gooders do nothing but encourage bad behaviors and self-defeating choices.
The latest target in this campaign is the Supplemental Nutrition Assistance Program (SNAP, more commonly called food stamps).
In a New York Times story over the weekend, Anahad O’Connor massages and misreports a USDA study to reinforce some of the worst stereotypes about food stamps. For his trouble, the editors placed it on the front page. Readers of the newspaper of record learn that the end result of tax dollars spent on food assistance is a grocery cart full of soda. No exaggeration. The inside headline for the story is “What’s in the Shopping Cart of a Food Stamp Household? Lots of Sugary Soda,” and the front-page illustration shows a shopping cart containing almost nothing but two-liter pop bottles.
O’Connor tells us that “the No. 1 purchases by SNAP households are soft drinks, which account for about 10 percent of the dollars they spend on food.” Milk is number one among non-SNAP households, we are told, not soft drinks.
And O’Connor does not hold back in building his indictment of the program. He begins by framing these and other alleged facts with a quote that informs readers, “SNAP is a multibillion-dollar taxpayer subsidy of the soda industry.” The story doubles down on this misleading image by ending with a discussion of how the big soda companies lobby to keep the SNAP funds flowing — and with a quote asserting, “This is the first time we’ve had confirmation that this massive taxpayer program is promoting all the wrong kinds of foods.”
Let’s be clear here: this is nonsense. It’s a political hack job against a program that helps millions of Americans feed themselves, and we should all be outraged that the New York Times has disguised it as a piece of factual news reporting on its front page.
There are two major problems here. First, O’Connor misrepresents the findings of the USDA report. Second, O’Connor’s article is a case study in the dark arts of making biased reporting appear even-handed. Let’s start with the facts.
Even if we relied solely on O’Connor’s massaged numbers, we’d have to conclude that 90 percent of what people buy with SNAP is not soda (which O’Connor says makes up 10 percent of purchases). But what does the USDA report actually say? Strangely, neither the author nor the New York Times saw fit to link to the report in the story, online or in print. I had to look through agency reports to find it.
Spoiler alert: the report does not state that SNAP changes what people buy at the grocery store; it does not suggest any effect on buying soda; and its findings differ considerably from the picture O’Connor paints.
O’Connor did not simply miss details buried in obscure tables. He misreported basic statistics, selectively chose to ignore the report’s major findings, and cherry-picked a few facts to build a misleading case. Here are the top three items in the report’s summary of its major findings:
- There were no major differences in the expenditure patterns of SNAP and non-SNAP households, no matter how the data were categorized.
- The top 10 summary categories and the top 7 commodities by expenditure were the same for SNAP and non-SNAP households, although ranked in slightly different orders.
- Less healthy food items were common purchases for both SNAP and non-SNAP households. Sweetened beverages, prepared desserts and salty snacks were among the top 10 summary categories for both groups. Expenditures were greater for sweetened beverages compared to all milk for both groups, as well.
Later, the report adds these bullet points to its summary:
- Overall, there were few differences between SNAP and non-SNAP household expenditures by USDA Food Pattern categories. Expenditure shares for each of the USDA Food Pattern categories… varied by no more than 3 cents per dollar when comparing SNAP and non-SNAP households.
- Protein foods represented the largest expenditure share for both household types, while proportionally more was spent on fruits and vegetables than on solid fats and added sugars, grains or dairy.
No fair reading of this report can support the Times’ portrayal. Even the central fact O’Connor builds the article around — the claim that soda makes up 10 percent of SNAP household purchases — seems to have been ginned up.
Exhibit 6 of the report states that soft drinks comprised 5.44 percent of SNAP household expenditures. To get the 10 percent figure, my guess is that O’Connor looked to Exhibit 5, which incorporates soft drinks into the broader summary category of all sweetened beverages. That category makes up 9.25 percent of SNAP household purchases, a percentage well below the “proteins” category of meat, poultry, and seafood (19.19 percent).
So, it appears O’Connor inflated the soft drink estimate of 5.44 percent to 9.25 percent by equating soda with the broader summary category of sweetened beverages, and then rounded 9.25 up to 10 percent for good measure. (O’Connor later updated the story to note the discrepancy between sweetened beverages and soft drinks, albeit without a correction.)
And that claim about milk, not soda, being number one among non-SNAP households? Here, O’Connor is just scoring a political point based on the slimmest of rank-order differences: for non-SNAP households, Exhibit 6 reports that fluid milk products make up 4.03 percent of purchases while soft drinks make up 4.01 percent. Two one-hundredths of a percentage point. That’s what O’Connor uses to paint non-SNAP households as the normal milk people, and offer them up as the foil to the bad soda people on SNAP.
Here’s what the report actually finds.
First, out of each dollar paid to the grocery store, non-SNAP households spent 4 cents on milk and 4 cents on soft drinks. Households receiving SNAP spent 4 cents on milk and 5 cents on soft drinks. Consider those numbers and look again at the shopping cart full of soda that graced the New York Times’ front page.
Second, the life conditions and characteristics of SNAP recipients differ considerably from the huge category of “all other households,” but the data analyzed for the study made it impossible for the researchers to distinguish program effects from these group differences in any way. As the study makes clear, no conclusions can be drawn about whether the program is responsible for even the tiny gap observed for soda purchases.
Third, because SNAP benefits are too low to cover the full food costs of households, people who receive SNAP spend a mix of program benefits and other income on food. This is critical to understand because, as the authors of the study note, the data provide no indication of whether soda purchases were made with SNAP dollars, work earnings, or some other source of income. If someone spent all their SNAP benefits on meats and vegetables and then used their own money to buy a six-pack of soda, they were classified as a SNAP household purchasing soda. As a result, the study tells us nothing about what people actually used SNAP benefits for, only what they bought in general.
The study’s authors are admirably clear about all these facts and limitations, and O’Connor clearly knew about them. Indeed, he provides himself with some political cover by quickly mentioning a few of these points deep in the article. What O’Connor does not do is fulfill the journalist’s obligation to explain how the facts fit together, or avoid using a factual news article to leave a misleading impression.
The article quotes well-established food stamp critics whose views were on the record long before the recent USDA study, but, in a research field where their views are widely opposed, does not provide space for a single researcher with a contrasting viewpoint.
O’Connor also singles out soda-company lobbying as the reason why USDA officials have been resistant to commodity-specific limitations on what households can buy with SNAP benefits. Nothing is said about the many legitimate reasons why lots of people who work in the field hold this commitment — from the belief recipients should have dignity as equal citizens to the implementation burdens having an extensive list of verboten items would introduce.
To be fair, O’Connor does eventually acknowledge that “sweetened beverages are a common purchase in all households across America.” But this is really just one of the oldest tricks used in news stories that disparage stigmatized social groups based on behaviors that are common in the population as a whole. Headline and lede: “What’s in the Shopping Cart of a Food Stamp Household? Lots of Sugary Soda.” Somewhere below: All Americans are drinking lots of sugary soda. (It’s an easy game to play: Headline and lede: “Children from Low-Income Families See No Point to Homework.” Somewhere below: Rightly or wrongly, kids in general don’t like homework and tend to see it as pointless.)
What also makes O’Connor’s article so troubling is that he wraps the usual scurrilous myths about SNAP in a veneer of health promotion — a framing that’s sure to win over some left-leaning readers who’d otherwise recoil at the usual trumped-up claims about food stamps. Yet in the end, O’Connor’s health paternalism doesn’t just run aground morally, but empirically: the study provides no evidence that SNAP encourages soda purchasing, and no evidence that SNAP funds (as opposed to personal funds) were used to buy soft drinks.
O’Connor writes a lot about sugar, and not much about social policy. So perhaps his main target here is the sugar industry. If so, he has thrown millions of food-insecure Americans — most of whom work or have significant disabilities — under the bus to advance his agenda.
Just as political attacks on social protections are on the rise, the article panders to the worst stereotypes of “welfare,” ignoring the SNAP program’s many successes. In the process, it tells people who imagine the worst about food stamps that they’ve been right all along. Facts be damned.