Questions on “How Congress Can Fix Food Stamps”

The National Review in July published a short opinion piece in July ‘24 entitled How Congress Can Fix Food Stamps. Sam Adolphsen, the author, argues that the cost of SNAP benefits is too high. He comments that “about 17.6 million food-stamp recipients are able-bodied adults, and about two-thirds don’t work,” which leads him to list a series of Republican-led, state-level policies that require SNAP recipients to participate in job training or employment programs. These policy decisions, Adolphsen argues throughout the article, are “common-sense”, “wildly popular among voters”, “turning a handout into the hand up that [SNAP benefits were] always supposed to be,” and are “protecting taxpayers while promoting work over welfare”.

However, Adolphsen fails to demonstrate key assumptions in his ideologically-driven thesis: do work requirements actually give Americans an opportunity to no longer live in poverty? Is “moving [people] from welfare to work” an effective way to lower poverty?

To be clear, the essay doesn’t spend any time focusing on policies that achieve an outcome where families are highly food secure. In his own words, “Republican’s top priority should be reining in this out-of-control welfare program,” referring to the high cost and enrollment. He uses the word “crisis” to describe the $110 billion plus cost per year forecasted for 2024 to 2034 (since revised, see footnote), the fact that an average of 42 million people monthly participated in SNAP in FY 2023, and that two-thirds of able-bodied adults without dependents (ABAWDs) are not working.¹

A generous interpretation is that he believes that getting participants eligible to work into training or employment programs means they won’t live in poverty any longer. But having a job does not mean making enough to live. Cynically, however, I feel that the opinion piece is dismissive of this, discussing reforms born from a desire to cut costs, rather than proposing changes that would assist the Americans who rely on SNAP benefits. And while I do believe there is a place in critical discourse for both approaches, Adolphsen’s essay could have done a bit more to highlight root causes, rather than just the costs, of high SNAP participation.

So I’ll confine this to these two questions: 1) why SNAP participation and program costs continue to be high, and 2) what are the effects of work requirements as part of the SNAP benefits?

How (and why) have SNAP participation and costs changed since 2019?

Adolphsen gives a dramatic and imprecise one-sentence explanation:

President Biden has dramatically expanded the food-stamp program, exemplifying his penchant for spending binges, government handouts, and economic ineptitude.

In the aftermath of the COVID-19 pandemic, we undoubtedly are spending more on SNAP benefits, especially as participation gradually returns to pre-COVID levels. And Adolphsen is right to split out costs from participation throughout his essay; here, we’ll start with participation in the program.

SNAP participation spikes during crises, as expected.

In 2019, the average monthly SNAP participation was about 35 million people; in 2023, that number was up to just over 42 million driven by the global crisis that saw unemployment and food costs soar. By design, the costs of SNAP will decrease or increase when the U.S. economy is doing better or worse, respectively. In December 2020, roughly 30 million adults reported they did not have enough to eat over the last 7 days; a recent USDA report found that food insecurity has continued to rise since 2021. As of 2024, grocery prices have not fallen to pre-2020 levels (and they aren’t expected to), and the unemployment rate (after falling from its spike during government shutdowns in 2020 by mid-2022) is creeping back up. As the USDA Economic Research Service’s Key Statistics page points out,

Historically, SNAP spending and the participation rate tend to track the unemployment rate and poverty rate…

A 1-percentage point increase in the unemployment rate is associated with an additional 2 million–3 million additional SNAP participants. Changes in program policy and administrative practices tend to augment the rise in SNAP participation during economic downturns and contribute to the continued rise in participation during the early stages of economic recovery.

Notably, the source study only uses data from 1976-2010, so I would be curious to see an updated version of this. Additionally, an overview article from the Pew Research Center indicates that increased food stamp never returned to pre-2008 levels after the Great Recession.² I would like to see a study about recovery, e.g., how long would it take to return to pre- crisis levels of SNAP enrollment. However, I don’t see any clear evidence that continued high levels on SNAP participation are abnormal. I think the article could do a better job of addressing this and provide better insight on how SNAP participation adjusts after social crises.

Benefits spending also increased to match prices

The Thrifty Food Plan (TFP) is the lowest-cost of food plans designed to estimate grocery spending for a family of four, which is guidance used to set the maximum benefit distribution for SNAP benefits.³ It is updated in line with inflation, but received a more comprehensive update in 2021.

The Congress-passed 2018 Farm Bill (signed by former President Trump) mandated that the USDA to reevaluate the TFP every 5 years, with the first report due before 2022. Only 13 Republican Senators and 44 Republican House members (19% of Republican house votes made) voted against this bill.

As a result of the reevaluation process, the TFP was increased in FY 2022, which was a separate increase from any COVID-19 relief funds, and based on data collected on the costs of items recommended within the plan. From the above linked memo, it found that the cost of the recommended diet was 21% higher than the cost allotted for within the benefits plan at the time. This is a permanent increase, unless a new law is passed that entirely removes it.

A key point to note is that this TFP reevaluation, unlike previous times, was not required to be cost neutral (page vii of the Thrifty Food Plan, 2021, accessible from the USDA). This is a pivotal point not discussed in Adolphsen’s essay, but does provide context to his complaints about rising costs, and raises questions about whether Biden is truly responsible for this change.

In a Fact Sheet released by the Biden Administration, it notes:

Therefore, as directed by the 2018 Farm Bill, the President will ask USDA to consider beginning the process of revising the Thrifty Food Plan to better reflect the modern cost of a healthy basic diet.

This memo was released January 22, 2021. Yet the TFP 2021 report states that “the process to update the Thrifty Food Plan began several years ago…”⁴ The TFP 2021 report only notes President Biden’s contributions in two places (in a 125 page report):

President Joseph R. Biden committed via Executive Order 14002 to prioritize the review.⁵

President Joseph R. Biden emphasized the commitment to the reevaluation in Executive Order 14002 on January 22, 2021.⁶

The cited Executive Order 14002 is incredibly vague, and the TFP reevaluation had to be completed by 2022 anyway. Here is a New York Times article framing the TFP update positively for the Democrats, Biden Administration Prompts Largest Permanent Increase In Food Stamps, yet goes on to note that Biden’s only contribution was to “[urge] the department to speed up the process,” which happened about 10 months until the report was due anyway. Plus, the Secretary of Agriculture at the time, Tom Vilsack, is quoted goading Republicans: “‘A majority of Republicans in the Senate and House voted for this,’ Mr. Vilsack said. ‘They deserve credit.’”

Apparently Republicans did not realize that what they were voting for allowed the USDA to drop the cost-neutral requirement. The same New York Times article states:

In allowing the plan’s value to rise, officials argued they were following the 2018 law, which required new standards but did not specify whether costs should stay the same…

But as hints of the benefit increase spread last week, Republicans pushed back. Former Representative Mike Conaway, a Republican and chief author of the 2018 law, said it was written “assuming the precedent of cost-neutrality would be followed” and warned the administration against “unilateral overreach.”

So this may have just been a poorly-written law. It is also hard to know to what extent this was debated internally, e.g., was this a ‘loophole’ that the evaluators took advantage of, or was this the widely accepted interpretation of the law at the time of writing in the USDA? I would be very interested in learning more about who identified this and the process internally to come to these conclusions.

A February 2023 hearing saw Republicans denouncing and blaming FNS (Food & Nutrition Security) leadership for this change. Testimony from the Deputy Undersecretary indicates a “fundamental difference of perspective regarding what the 2018 Farm Bill directed for the re-evaluation” between the USDA and a Government Accountability Office report released in 2022.⁷

It is this GAO report that we get the closest to practical answers. In response to Biden’s Executive Order and Fact Sheet, “FNS officials said they hired additional staff, stopped work on other projects, and worked overtime.”⁸ As to the cost-neutral claim, a footnote in the GAO report notes that “USDA OGC officials told [GAO] that holding costs constant during prior TFP reevaluations was an administrative decision made by the Secretary of Agriculture, not a legal requirement,” which seems at odd with the claims made in the TFP 2021 report.⁹ There is not further substantive discussion of this interpretation, but I did find a Senate Fact sheet reiterating that this complied with the law.

Putting the history and any assignation of responsibility aside, how did this change affect costs? From a 2022 CBO report, the change in the TFP is said to be “main source” of a $266 billion increase in estimated outlays from 2022 to 2031.¹⁰

What happens when work requirements are enforced for eligible SNAP participants?

A 2023 report from the Center on Budget and Policy Priorities states that: “strong research evidence on SNAP’s existing work-reporting requirement shows that it does not increase employment or earnings but does cause many people to lose food assistance.” Despite this, the 2023 Fiscal Responsibility Act doubled down on work requirements, limiting the waivers that states took advantage of during the COVID-19 pandemic. The maximum age for work requirements also was increased for ABAWDs, from 52 to 54.

Adolphsen claims that “in Kansas, the Republican-controlled legislature passed veto-proof bills in 2023, expanding the age range of able-bodied adults required to participate in job training,” and goes on to claim that “participation in such programs has more than tripled.”

While he did not provide a source or context, I found reports from the Kansas Department for Children and Families. I think the bills he is referring to should have went into effect in July 2023. So my best guess is that the FY 2024 Participation Report should show this change; page 17 shows the “Food Assistance Employment & Training Program Caseload and Expenditures” and indicates July 2023 had 798 adults, with June 2024 having 1,266 participants, which is not triple. The FY23 monthly average was 580 adults; the FY24 average was 1,307. I also thought perhaps he combined training programs, but the “SNAP WORK PROGRAMS (GOALS)” (page 22) shows a large drop in participants from 557 in FY23 to 273 in FY24.

There is a PDF from the Kansas that reports on FY 2023 food assistance participation (which ends in July 2023); he may be referring to page 12, a table titled “Food Assistance Employment & Training Program Caseload and Expenditures”, which shows the number of adults enrolled each month increasing from 281 in July 2022 to 843 in June 2023 - which is exactly tripled, amusingly - but the following chart, “Food Assistance Employment & Training Program - Historical Caseload and Expenditures Detail” shows that the average number of adults enrolled monthly increased only from 424 in FY22 to 580 in FY23 (page 13).

On page 15, there is a SNAP Work Programs (GOALS) table, which is confusing, as it shows “Avg Monthly Adults Receiving Assistance” increase from 39 in FY22 to 557 in FY23! Yet the “Avg Monthly Adults Enrolled,” (the next column over) shows a decrease from 495 in FY22 to zero in FY23.

This may be a reporting error, as page 22 of this document has a chart “SNAP WORK PROGRAMS (GOALS) Monthly Average Number of Adults Receiving Assistance” that shows 495 in FY22 and 557 in FY23. So the numbers may be simply mixed up in the table. It doesn’t seem to be corrected in the FY24 report either.

So, I’m not quite able to pinpoint what exactly Adolphsen is measuring. The number of ABAWDs enrolled, however, did decrease from 13,504 in July 2023 to 9,964 by June 2024 (page 8), but to what extent that can be attributed to enforced work requirements, I’m not sure. I found one supporting news article that discusses wider decreases in program participation in FY 2023 after previous requirements. Another article from the Kansas Reflector notes how a paid lobbyist supporting work requirements up to age 60 could not provide any evidence it would work or that previous reforms were effective.¹¹ I did see discussion that food pantry usage was increasing in 2021, but the USDA 2023 food insecurity report does not show a statistically significant change between 2018-2020 (11.3%) to 2021-2023 (10.6%).¹²

Similarly, I found next to nothing on the impact of Idaho’s changes. This is despite the Governor “urging caution” and recommending that legislators “closely monitor the impacts.” I see some news articles on Louisiana’s enforcement; here, 15,000 recipients lost benefits in the summer of ‘24. Again, the article mentions the department “has no information about the 15,000 people who were disenrolled since the fall and whether they have found employment, or what sort of wages they are earning if they did, a spokesperson said.” This is surprising to me that the state did not take the chance to follow up to identify outcomes, but again, the critic in me feels that this demonstrates that legislators thought whether people were successful in finding quality jobs was secondary to the true purpose of the legislation: to just decrease participation in SNAP. Again, data (collected and pushed by the FGA) doesn’t actually show any clear successes.¹³

In the second to last paragraph, the author claims that work requirements for SNAP benefits are “wildly popular,” noting that “80 percent of swing-state Wisconsin voters cast a ballot in favor of welfare work requirements,” without linking to any background on the measure.¹⁴ Checking Ballotpedia shows that “casting a ballot…” really meant less than 2 million people answered a vague advisory question that was not legally binding. The question in full: “shall able-bodied, childless adults be required to look for work in order to receive taxpayer-funded welfare benefits?” The Wisconsin State Journal, which advised voting yes, also recognized the lack of merit in the question, stating: “the question is too cute by half in seeking to manufacture outrage.” I don’t think saying it was a question designed to be manipulated as a marketing tool is out of line.

Again, studies that have looked at this find very little evidence that work requirements lead to stable, well-paying jobs. An NBC News report compiled 12 studies, and all reviewed found decreased SNAP enrollment while 50% found no impact on employment, 25% found mixed to mostly none, and 25% found positive effects.

Lastly, the piece aside, it is frustrating to read the some of the outwardly hateful comments posted along with the article. Very few are well-reasoned, and most rely on gross generalizations and are very unkind. After some time, the comments appear closed to non-subscribers.

As something to follow up with, we can see the author’s opinions in his testimony to Congress from 2020.