Expert Commentary

SNAP and the effectiveness of work requirements: A research roundup

In this first of a three-part series, we look at recent research on the Supplemental Nutrition Assistance Program, a major component of the farm bill coming up for renewal in September.

(Marques Thomas / Unsplash)

The farm bill is wide-ranging legislation that sets funding and directs priorities for a variety of federal food consumption and production programs in the U.S., including the Supplemental Nutrition Assistance Program, formerly known as food stamps.

Congress usually debates and renews the farm bill every five years. The first farm bill was passed in 1933, with 18 farm bills having been passed in all. The most recent farm bill passed as the Agriculture Improvement Act of 2018 — meaning it expires at the end September, which marks the end of the federal fiscal year.

Over the coming weeks, legislators will make the case for the next farm bill to include funding that supports the interests of their constituents, as well as lobbying and advocacy groups. To help guide journalists in coverage of those debates, The Journalist’s Resource is taking a look at academic research on three pillars of farm bills: SNAP, environmental conservation and rural development.

The research featured in this miniseries can inform the questions that journalists at local, regional and national outlets ask of federal lawmakers.

This week, we’re focusing on SNAP.

SNAP typically gets the largest funding share of any program in the farm bill. The Congressional Budget Office estimates roughly $120 billion in SNAP spending per year over the coming decade. The number of people who get SNAP benefits tends to rise or fall along with the unemployment rate.

“SNAP serves a large and diverse caseload, with substantial shares comprising low-income working families with children, elderly and disabled people, and very low-income individuals with substantial barriers to employment,” according to a November 2022 Congressional Research Service report.

To qualify for SNAP benefits, households have to be at or below 130% of the federal poverty line — this comes out to about $2,500 a month for a family of three, according to a 2023 analysis from the Center on Budget and Policy Priorities, a left-leaning think tank.

States also play a critical role in providing SNAP benefits and determining who qualifies.

“State agencies are responsible for general program administration and ensuring program integrity,” according to USDA. “State agencies determine the eligibility of individuals and households to receive SNAP benefits, and issue monthly allotments of benefits.”

In April 2023, the most recent month for which data is available from the U.S. Department of Agriculture, nearly 42 million people received SNAP benefits. An emergency expansion of benefits during the COVID-19 pandemic, which meant an additional $82 per month on average for qualifying households, expired in March 2023. Nationally, the average monthly SNAP benefit was $230 in 2022, up from $130 in 2019.

SNAP falls under the nutrition policy area of the farm bill, which is administered by USDA. It affects the most number of people of any farm bill program, and it is hotly debated in Congress and in the news media.

SNAP work requirements

Work requirements for SNAP participants usually garner widespread news coverage as the farm bill comes due for reconsideration in Congress. These federally mandated work requirements apply to “able-bodied adults without dependents,” or ABAWDs in government lingo.

SNAP participants who can work and don’t have a dependent are required to work at least 80 hours per month to qualify for food benefits. Participants subject to work requirements don’t necessarily have to work for pay — volunteer work, unpaid work or work in exchange for something other than money all count.

As part of negotiations over the debt ceiling earlier this year, SNAP participants age 18 to 54 will be subject to work requirements by Oct. 1, 2024. Upper ages for work requirements are phasing in over time. The age increased to 50 in early September. It will go up to 52 in October, then to 54 next fall. These caps are legislated to stay in place through 2030.

Some Republicans want to raise the age to 65. Work requirements were first introduced in 1996.

Context from research

It is critical for journalists who want to cover this topic comprehensively to have a sense of recent academic research on the effectiveness of work requirements and whether SNAP in general enables people with low income to purchase food they otherwise might not be able to afford.

Here is important context journalists should consider adding to any story on the SNAP debate:

  • Only about 13% of SNAP participants are subject to work requirements, according to a 2022 USDA report. The figure was 12% in 2019, before the pandemic-related benefit expansion, according to a 2023 analysis from Angela Rachidi, a senior fellow at the American Enterprise Institute, a center-right think tank.
  • At the same time, people aged 50 to 64 are the fastest growing group of SNAP participants, and able bodied people in that age range outnumber those aged 18 to 49, according to Rachidi’s analysis, which was co-written with Thomas O’Rourke of AEI.
  • Demographic context can help frame policy proposals. Pre-pandemic, from October 2019 to February 2020, 38% of SNAP participants were white, 26% were Black, 15% were Hispanic, 3% were Asian and a little over 1% were Native American, according to the USDA. Those figures were roughly the same during the early months of the pandemic.

Federal spending on SNAP is miniscule compared with other social safety net programs. By comparison, social security spending was $1.2 trillion in 2022 while Medicare and Medicaid spending was $1.3 trillion, according to the Congressional Budget Office. Federally, SNAP cost about $120 billion in 2022, including $114 billion in benefit payments, according to USDA. Overall, SNAP spending makes up between 1% and 2% of all federal spending yearly.

“While the overall social safety net in recent decades has shifted to provide either work-contingent assistance or assistance targeted to specific categories of people, SNAP is the only social benefits program available to all low-income Americans,” writes Northwestern University economist Dianne Whitmore Schanzenbach in one of the papers featured below.

Keep reading for more insights from recent research on work requirements, online SNAP purchases, the major role state policies play when it comes to SNAP participation, and what happened to SNAP participants in a food desert when a full-service grocery store opened nearby.

Research roundup

Employed in a SNAP? The Impact of Work Requirements on Program Participation and Labor Supply
Colin Gray, et al. American Economic Journal: Economic Policy, February 2023.

The study: Focusing on people who can work and do not have dependents, the authors use a sample of more than 90,000 Virginians who likely would be on SNAP absent work requirements. Virginia suspended SNAP work requirements from 2009 to 2013, during the Great Recession. The authors follow the data on people enrolled in SNAP at the end of this period to see if they stayed in the program after the work requirements were put back in place. The sample represents just 9% of all people in the state receiving SNAP at the time — meaning most people in the program were not able to work or had dependents and were not subject to the work requirements. The work requirements, in other words, affected a small subset of the overall SNAP population.

The findings: A year and a half after the work requirements were reinstated, 37% of prior SNAP participants were no longer on the rolls. People who were homeless or had no income were most affected, the authors find. During the period studied, people age 50 and above were not subject to the work requirements. The authors find a “sharp positive increase in participation at age 50,” which they associate with an overall decrease in participation due to the work requirements of roughly 23 percentage points.

They determine the decline is due to people failing to meet work requirements, not because of excessive paperwork needed to recertify for the program. Crucially, the authors find work requirements do not lead to more labor force “attachment,” a word economists use to describe whether someone has the resources, professional connections and skills to find a job relatively easily.

In other words, work requirements do not appear to make it more or less likely someone will be employed after they are off SNAP, according to the findings. The authors suggest that work requirements may not address “underlying barriers to work,” though they note their results may not apply to other states.

The authors write: “We find that work requirements induce disproportionately higher exit among beneficiaries who are documented to be homeless or to have no earned income prior to the reinstatement of work requirements. In contrast, induced exit is disproportionately lower among those with a history of disability, who are more likely to be exempt from the work requirements.”

Understanding SNAP: An Overview of Recent Research
Diane Whitmore Schanzenbach. Food Policy, January 2023.

The study: The author examines the findings from 17 papers published from 1988 to 2021 that investigate how SNAP benefits affect labor market outcomes nationally and in specific states, as well as personal food consumption, whether the benefits are adequate to meet nutritional needs, the administrative costs of applying for and staying in SNAP, and how SNAP benefits affect individual health and well-being.

The findings: Research finds higher SNAP benefits during the Great Recession led participants to quickly increase their food spending at home, with other research finding “the temporary SNAP increase had a larger per-dollar fiscal stimulus than any other spending increase or tax cut that was enacted to combat the Great Recession.”

In Michigan, research finds people often lose SNAP benefits at times when they are required to report income to an administrative office. Some drop-off may be due to people finding jobs that pay too much for them to qualify, but the research finds about half of people were still eligible and simply missed the deadline. New case management software simplified the process and exits fell 12%, the research finds.

Toward the end of a benefit month, SNAP beneficiaries may be at risk of not eating enough food, defined as less than half of their recommended daily caloric intake, other research finds. Likewise, another paper finds at the end of the month SNAP beneficiaries spend less time grocery shopping and making food at home than earlier in the month. When SNAP payments were disbursed later in the month, further from the time when other expenses such as rent are typically due, food spending is more consistent, other research finds.

SNAP beneficiaries report better health and fewer visits to the doctor when they are able to use the program, research finds. In another paper, elementary and middle school students in South Carolina score slightly lower on math tests when it’s been more than 26 days since their families received SNAP benefits. For children, food programs such as SNAP general improve nutrition and reduce the likelihood of hunger, research finds.

The author writes: “[SNAP] alleviates food hardship, and in some cases improves nutrition quality. While negative work incentive effects are inescapable, the impacts on employment and hours worked are modest. That may be partly because program participation improves a range of health outcomes in both adults and children.”

An Analysis of SNAP Online Purchasing Behavior in California: A Review of the First 7 Months of Program Implementation and Lessons Learned
Isabelle Foster, et al. American Journal of Health Promotion, October 2022.

The study: This research is among the first to analyze an online SNAP purchasing program.  In April 2020, California joined the ranks of states that offer SNAP participants the ability to buy groceries online. (A pilot program included in the 2014 farm bill made online buying available in a handful of states, but the COVID-19 pandemic accelerated the rollout. Today, every state and the District of Columbia is authorized to offer SNAP purchases online.)

The goal of the pilot was to give SNAP participants a way to buy food with their benefit when there were few or no local grocers that accept SNAP. The authors note some challenges SNAP participants face in buying food online, including that their benefit cannot be used to cover shipping and delivery. They analyze the number of purchases per month, excluding returns or invalid transactions, across California’s 58 counties during the first 7 months of CalFresh, the name for the online SNAP program there. Amazon and Walmart were the only approved CalFresh retailers during this time.

The findings: Per county there were a median of 665 monthly online SNAP purchases from April to October 2020. There were about 1.7 million CalFresh transactions during this time, worth about $112 million and accounting for nearly 1% of all SNAP purchases in the state. Across the period studied there were about 875,000 purchases worth $46 million, with an average basket value of $53 from Amazon. From Walmart, there were 893,000 purchases worth $66 million, with an average basket value of $73.

CalFresh participation consistently increased from April to August for urban, suburban and rural counties, but purchases slowed in rural areas after August. SNAP participants living in the four rural California counties had the lowest rate of online food purchases. There were no Walmart stores in the rural counties during the period studied, and Amazon Fresh did not deliver to them. The authors note that before CalFresh, one in five SNAP dollars was spent at Walmart, so it may have been easier for the retailer to advertise the new online shopping option to its existing SNAP customers.

The authors write: “Consistent with in-person purchases, shopping trends at each retailer varied across the benefit month. Spending decreased during the third week of each month, and then increased through the first and second weeks. This likely coincides with California’s benefits issuance timeline, which staggers CalFresh dispersal from the first to the 10th day of the month.”

Do SNAP Work Requirements Work?
Timothy Harris. Economic Inquiry, September 2020.

The study: Starting in 2010, as part of recovery efforts following the economic shock of the Great Recession, states were allowed to apply for waivers from federal SNAP work requirements if their economies were not doing well — for example, if the state unemployment rate was over 10% for a year. Fourteen states kept work requirements in place, even though they qualified for waivers, according to the author.

States could also apply on behalf of individual counties. From 2011 to 2017, economic conditions improved in some counties with waived work requirements, meaning SNAP participants in those areas had to again comply with the work requirements to receive benefits. The author uses these geographic and time variations to study employment differences between people who get SNAP — with and without work requirements.

The findings: Among SNAP participants who were able to work, did not have dependents and were subject to work requirements, employment rates were 1.3 percentage points higher than those who got SNAP without the work requirement.

Likewise, SNAP participation fell 1.7 percentage points with the work requirement — workers without a high school diploma were the most likely to lose benefits. In sum, more people were employed and fewer were on the SNAP rolls with the work requirement than without. The author notes that the yearly dollar value of SNAP is small compared with other social welfare programs, such as Medicaid, and that people will have stronger incentives to find work larger the value of a lost benefit.

The author writes: “Arguably, ABAWDs should be the most responsive to work requirements as they do not have dependents at home, have no disabilities, and are of working age. Policies that seek to expand work requirements to other households — such as those with dependents — likely will have smaller employment effects than those found in this study.”

The Downs and Ups of the SNAP Caseload: What Matters?
Stacy Dickert-Conlin, Katie Fitzpatrick, Brian Stacy and Laura Tiehen. Applied Economic Perspectives and Policy, September 2020.

The study: The 2002 and 2008 farm bills gave states wide leeway in deciding who qualifies for SNAP benefits. The “legislative and regulatory changes gave states increased flexibility to simplify program administration and increase program access, especially for low income working families,” the authors write.

They use detailed administrative data to track how SNAP participation changed from 1990 to 2016 as states enacted policies at various times related to eligibility, stigma and transaction costs.

These policy changes included:

  • Removing personal vehicles from asset tallies used to determine eligibility — by 2016, every state allowed applicants to not count at least one vehicle as an asset.
  • Increased outreach beginning in the 2000s, as the federal government began reimbursing states for half their costs related to public education, including TV and radio advertising.
  • Differences by state in the number of months SNAP participants were given between having to recertify their eligibility.
  • Fingerprinting requirements for SNAP participants. By 2001, Arizona, California, Texas and New York required fingerprinting. Those states made up roughly 25% of the national SNAP caseload. No states currently require fingerprints for SNAP.
  • The rollout beginning in 1989 in Maryland of electronic benefit transfer cards, which look like traditional credit cards. Every state used these cards by 2004.

The authors also simulate what SNAP caseloads would have been like if eligibility requirements had been the same across states from 2000 to 2016. In this paper, changes in “caseloads” refers to changes the number of SNAP participants.  

The findings: When states mandated shorter timeframes between recertification periods their caseloads declined, relative to when states offered longer timeframes.

EBT rollout, conversely, increased caseloads, as stigma fell related to using SNAP benefits. Exempting one or more vehicles from asset tests also slightly increased caseloads after two years. Fingerprinting requirements led to a relatively sharp drop in caseloads over two years. Advertising on TV and radio increased the proportion of SNAP participants in states that rolled out those campaigns.

From 2000 to 2016, if the federal government had mandated states keep eligibility requirements in place in 2000, the national growth in SNAP participation would have been 38% smaller, the authors estimate.

The authors write: “Broad-based categorical eligibility policies were a subject of heated debate in the farm bills of 2014 and 2018 and our estimates suggest that they do have an important relationship with caseload growth, increasing the caseload by approximately 10% after two years. Work requirements for ABAWDs, another policy receiving a great deal of attention, is negatively related to caseload growth, suggesting that increasing these requirements would lead to steep caseload declines.”

SNAP Participants Improved Food Security and Diet After a Full-Service Supermarket Opened in an Urban Food Desert
Jonathan Cantor, et al. Health Affairs, August 2020.

The study: The authors examine what happened to nutrition in low-income households when two federal programs, SNAP and the Healthy Food Financing Initiative, overlapped in one U.S. city — Pittsburgh.

HFFI in part focuses on funding new, large, full-service supermarkets in food deserts. In cities, food deserts are usually defined as areas where about one third of the population lives more than one mile from a supermarket. The authors note past research that finds smaller food stores that accept SNAP tend to stock fewer nutritious, healthy offerings than larger grocers.

They surveyed a sample of 195 SNAP participants in Pittsburgh’s Hill District before and after the HFFI-funded supermarket opened there in October 2013. The Hill District had been a food desert for decades prior to the opening of the new supermarkert. For comparison, the authors also surveyed 85 SNAP participants in another Pittsburgh food desert neighborhood with similar demographics — roughly 95% Black and 80% female — that did not get a new supermarket during the study period.

The findings: Survey participants in the neighborhood where the new supermarket opened reported feeling more secure in their ability to get enough food, while this measure was unchanged in the comparison neighborhood. They also reduced their sugar intake, compared with the comparison neighborhood, by nearly 3.5 teaspoons per day.

(The full-service supermarket in the Hill District, a Shop ‘n Save, closed in March 2019. In April 2023, the city approved the development of a new supermarket in the area, with the owners aiming to open by the end of the year.)

The authors write: “This is the first study to indicate that the introduction of a full-service supermarket into a food desert can improve both food security and diet among SNAP participants. The results indicate that such placement may be an effective policy to heighten the impact of a long-standing federal food and nutrition program — in this instance, SNAP — in communities with limited access to healthy food choices.”

Further reading

The Supplemental Nutrition Assistance Program: History, Politics and Public Health Implications
Marion Nestle. American Journal of Public Health, November 2019.

US Farm Bills and the ‘National Interest’: An Historical Research Paper
Nadine Lehrer. Renewable Agriculture and Food Systems, July 2018.

About The Author