Feeding America is getting harder next year
America's poor are getting a lump of coal from Trump and congressional Republicans
Merry Christmas from the Slow Boring team!
As previously discussed, our best information is that relatively few people are going to be spending today reading articles, but we thought this reflection that Caroline wrote on helping the needy here at home and the cruelty of using administrative burdens and known-to-be-ineffective work requirements as a pretext for reducing spending was too good to be relegated to its original destiny as a topper for tomorrow’s mailbag. We decided to run it as a standalone piece — a Christmas gift of bonus content — that we hope you’ll enjoy if you have a little down time on your break.
— Matt
As I got to know the economists and food bank leaders who shaped Feeding America’s auction into the success it is today, I was struck not just by their goodwill, but by their insistence on translating that goodwill into real change in the lives of hungry people. Good intentions were never the scarce resource. The hard work was designing a system that could actually deliver on them at scale.
This is a season of generosity and kindness, but Slow Boring readers know that the federal government is often the institution best positioned to scale those impulses effectively, perhaps second only to markets themselves. As we look ahead to a new year, it’s worth revisiting the Feeding America story with an eye toward lessons for public policy, where design choices can deliver enormous good or inflict real harm.
At the outset, Feeding America was making decisions on behalf of the communities it served. Food banks could accept or reject what the national clearinghouse offered, but they couldn’t express preferences beyond that. The auction changed this dynamic. By giving food banks visibility into what was available and allowing them to signal their needs, it let local knowledge do the work that centralized decision-making could not. Resources were allocated more efficiently not because anyone became more virtuous, but because the system finally let participants compare options and make tradeoffs for themselves.
The same logic already underpins the most effective federal nutrition programs. For the most part, SNAP has historically not told recipients how much protein to eat or which foods best suit their families. Instead, it lets households navigate the private grocery market themselves. Research suggests that programs perform better when they avoid rigid prescriptions and leave room for flexibility. People’s choices often reflect real constraints — of price, time, health, and work — that are hard for a distant system to anticipate.
That’s what makes recent policy changes so striking.
As a result of the One Big Beautiful Bill, states are (beginning this month, and continuing into 2026), implementing expanded SNAP work requirements, extending them to adults ages 55 to 64 and narrowing exemptions that previously protected groups like some veterans and former foster youth. Several states are also imposing new restrictions on what SNAP benefits can be used to purchase. These changes reduce agency precisely where past experience suggests it matters most.
As Rebecca Piazza, a former U.S.D.A. official now at Code for America, told me, these policies function less as work requirements than as work reporting requirements. Past experience shows that they do little to change how many people are working, but instead cause many people who are already working to lose benefits because of the paperwork required to prove compliance. Each new reporting step creates another opportunity for someone to fall out of the program, not because they fail to meet the standard, but because the system makes it harder to demonstrate that they have.
Paperwork requirements matter because they impose real transaction costs.
For households living on the margin, time, attention, and administrative slack are scarce resources. Each new form, deadline, or reporting step competes with work schedules, caregiving, transportation, and housing instability. Those added procedural hurdles don’t just inconvenience people; they predictably reduce participation, even among households that meet the formal requirements.
SNAP serves one in eight Americans. At that scale, even small design changes can determine how many people the program actually reaches. The Feeding America auction succeeded because it trusted participants to know their own needs better than a distant center could. Federal nutrition programs exist to help people afford food. When new requirements make that help harder to access, the result is rarely better behavior. It’s fewer people being served.
Hunger in the US is overhyped when compared with the US 100 years ago, or with poor countries today, but yeah, there's still a problem here, and you nail it. Two obvious facts that neither party wants to face: When you look at the neediest 5%, and especially the neediest 1%, you realize that a large fraction are psychotic or intellectually disabled. Expecting either category to fill out some complicated form is ridiculous. I once took a schizophrenic family friend to the welfare office, and she curled up on the floor and started to cry as the case officer asked her whether she owned any stock or bonds.
Those of us who don't rely on the government for food assistance hate it when large entities burden us with complicated paperwork requirements (IRS, DMV, insurance companies) and come to hate the entities themselves. This change will also make poor people dislike and distrust the federal government more, which is win-win from a Republican perspective.