Skip to main content

Families, Older People in Every State at Risk Under Proposed SNAP Rule That Would Take Food Away for Not Meeting Work Requirements

As Congress works through federal budget proposals and prepares to reauthorize the Supplemental Nutrition Assistance Program (SNAP) in the pending farm bill, some prominent proposals from House Republicans would harm SNAP participants by taking food away from people who can’t show every month that they meet or are exempt from new work requirements. One bill, from House Agriculture Committee member Dusty Johnson and more than 20 other House Republicans, would dramatically expand SNAP’s already harsh work requirements, putting more than 10 million people — including millions of children — in states nationwide at risk of losing at least some of their food benefits.[1]

The bill’s approach — conditioning access to basic food assistance on documenting hours of work or work activity ― is based on several false assumptions. The first is the notion that people who receive benefits do not work and must be compelled to do so. Most SNAP participants who can work do work, and claims to the contrary are often rooted in prejudices about people based on race, gender, disability status, and class. It also ignores the realities of the low-paid labor market, the lack of child care and paid sick and family leave, how health and disability issues and the need to care for family members affect people’s lives, and ongoing labor market discrimination. And numerous studies have found that SNAP work requirements like these don’t improve employment or earnings, they just cut people off from the food assistance they need to buy groceries.

Johnson’s bill (H.R. 1581, the America Works Act) would expand the population of people subject to SNAP’s existing work requirement, which operates as a time limit where, if over three months someone is unable to document they work or participate in a qualifying employment and training program for at least 20 hours a week, they are cut off from SNAP benefits — and aren’t eligible again until a total of three years has passed.[2] Under H.R. 1581, for the first time adults up to age 65 (instead of the current age of 50) and adult participants who live in a household with school-age children would be subject to these requirements and at risk of losing benefits.[3] The bill also would take away states’ existing flexibilities to waive this punitive work rule in areas with insufficient jobs, a flexibility that every state but one has used since the requirement was put into place in the mid-1990s.[4]

Nationwide more than 10 million people, about 1 in 4 SNAP participants, live in households that would be at risk of losing food assistance benefits under this bill. That includes about 6 million people who would potentially be newly subject to the time limit and at risk of losing SNAP, and about 4 million children who live in families that could have their SNAP benefits reduced, harming the entire household.[5]

Households in every state would be affected. The share of states’ SNAP recipients whose benefits would be at risk ranges from about 1 in 3 in Alaska, the Virgin Islands, Oklahoma, and South Carolina, to about to 1 in 5 in Washington, Vermont, Idaho, and Rhode Island. (See Appendix I, Table 1 for state-level estimates; see Appendix II for an explanation of our methodology.)

Not everyone newly subject to these requirements would lose benefits under the proposal. Many people would, for example, be working more than 20 hours a week and be able to navigate the work verification system, or they would be successfully exempted by their state because of a physical or mental disability. But a very significant number would likely lose benefits because they are out of work or working insufficient hours, the state failed to screen them for an exemption they should have qualified for, or they were unable to navigate the verification system to prove they are working.

Altogether, Millions of Children, Older People, Veterans Risk Losing Benefits

The bill puts at risk the food benefits of:

  • Some 3 million adults up to age 65, primarily parents or grandparents, who live in households with school-age children (adults in households with children under 7 would be exempt). The 4 million children aged 7 to 18 who live in these households would see their household’s food assistance fall if their parents or other adults in the family aren’t able to meet the requirement.
  • About 2 million older adults, aged 50 to 64, who do not have children in their homes. This is a group that has more health issues than younger adults and sometimes are no longer able to do the physical work they did when they were younger or whose skills and work history are not aligned with a changing labor market.
  • Adults who would become newly subject to the time limit in areas with elevated unemployment. The bill would substantially limit states’ current flexibility to temporarily waive this harsh time limit in areas where there aren’t enough jobs: only areas with unemployment over 10 percent would qualify.[6] The Trump Administration estimated that its own proposal limiting this state flexibility, which would have been less restrictive than Johnson’s, would have resulted in more than 700,000 people losing eligibility for SNAP.[7]
  • Examples of other groups that are at risk of losing SNAP benefits include: veterans, unhoused individuals, victims of domestic violence, adults with undiagnosed health conditions, adults who lost jobs when businesses closed or downsized, people working to recover from a substance use disorder, people returning from incarceration who face high barriers to employment, and young adults aging out of foster care, who may be in unstable situations and unable to work a steady 80 hours a month.

The bill also restricts states’ ability to maintain eligibility for a limited number of individuals who have used their three months of benefits. Recognizing the harshness of the time limit and its inability to adjust for unexpected or unique circumstances, Congress allowed states to exempt a small percentage (about 12 percent) of people subject to the time limit. States can use these discretionary exemptions in cases when, for example, someone faces a sudden hardship like car trouble or has recently been released from an institution. States are issued these exemptions each year based on their caseload, and those that aren’t used can be rolled over to subsequent years.

The Johnson bill would eliminate the ability to carry over unused exemptions. This would further hamstring a state’s ability to respond to individuals’ specific needs, or to its unique labor market challenges such as a manufacturing plant closure.

Work Reporting Requirements Harm SNAP Participants, Do Not Increase Employment

The approach taken by H.R. 1581 fails to recognize that most adult SNAP participants work when they can; imposes harm on children; and employs a work requirement that research shows fails to increase work. As a result, the bill will increase hardship without improving employment outcomes.

SNAP already limits adults aged 18 to 49 without a child in the household to three months of SNAP benefits in any 36-month period when they cannot show they are employed or in a work or training program for at least 20 hours a week. This work reporting requirement operates as a severe time limit that denies basic food assistance to people who want to work and would accept a job or work program slot. Even SNAP recipients whose state operates few or no employment programs and fails to offer them a spot in a work or training program — which is the case in most states — have their benefits cut off after three months irrespective of whether they are searching diligently for a job.

SNAP’s existing time limit reduces participation in SNAP and fails to increase employment. Claims that the time limit helps individuals increase earnings fail to address published studies that show the opposite. Independent studies have repeatedly shown that SNAP’s work-or-lose-benefits time limit does not increase employment or earnings. It just cuts people off from the food assistance they need to buy groceries.[8] A recent study found that the time limit cut SNAP participation among those subject to it by more than half (53 percent), with no effects on employment and earnings overall.[9] Another recent study found SNAP participation was cut between 7 and 32 percentage points a year after the reinstatement of the time limit, again with no evidence of improved earnings or employment.[10]

Most SNAP participants who can work already do. Before the pandemic, nearly three-quarters of adults participating in SNAP in a typical month worked either that month or within a year of that month of participation. At almost 90 percent of households, the rate was even higher among households with children and at least one working-age adult. Many of the jobs that SNAP participants hold — like as grocery cashiers, cooks, and home health aides — are in essential services.[11] SNAP is a critical support for these workers, as these jobs typically pay low wages, offer unreliable hours, and don’t provide benefits such as paid sick leave. Other SNAP participants are only temporarily between jobs, often turning to SNAP after a job loss to help them while they seek new work.

Some working-age adults are unable to work at various times. While 81 percent of adults who don’t receive disability benefits (which have very stringent eligibility requirements) and don’t have young children at home worked in the previous year, those who reported not working most commonly cited attending school, caregiving responsibilities, early retirement, illness, or disability as the reasons for not working.[12]

Workers who do not consistently work 80 hours a month are at risk of losing assistance they need to afford food at the grocery store under this bill. Many workers in low-paying jobs face a volatile labor market, irregular work schedules, and little control over how much they can work. Research shows one year’s success in the labor market does not translate into continued success in the next year.[13] SNAP provides some insurance for workers who lose jobs or can only find limited work. Moreover, workers who are working but are unable to document that they consistently work at least 80 hours a month are also at risk of losing food assistance. Not all workers get regular pay stubs or documentation, including many task-based gig workers, such as drivers and delivery workers. Moreover, the bureaucratic systems for verifying employment can result in people losing SNAP despite working. When Arkansas imposed a work-reporting requirement on Medicaid beneficiaries, confusing rules and complex reporting requirements led to about one-quarter of all those subject to the requirements losing coverage in the first seven months.[14]

Some people who should be exempted from the time limit because of a disability may not be properly screened, resulting in them losing SNAP benefits. Many people who don’t receive disability benefits have health conditions that impede their ability to work for the required number of hours. For them to be exempted from the requirements, a caseworker has to screen them properly and exempt them in the computer system properly. Many recipients won’t know to ask for the screening and caseworkers may be unaware of a person’s relevant health issues.

Failing to properly screen people with health conditions has been a problem with SNAP’s existing time limit. About 1 in 3 individuals identified by the Franklin County SNAP agency in Ohio as subject to the 20-hour work rule were later identified by a community group trying to help these individuals meet the requirement as having a physical or mental condition that impacted their ability to work, a condition that likely should have made them exempt from the requirement.[15] And when three states imposed a similar work reporting requirement in Medicaid in 2018, evidence suggests that people who were working and people with serious health needs who should have been eligible for exemptions lost coverage or were at risk of losing coverage due to red tape. Large numbers of beneficiaries in both states reported that they didn’t know about the work requirement or whether it applied to them.[16]

Taking assistance away from parents harms children. Children are harmed by policies that take away their parents’ benefits even if the children technically can continue to receive benefits. If an adult in the household can’t prove they are meeting the requirement, the whole household will receive less in SNAP benefits and will have fewer SNAP benefits than they need to afford food for the family. The impacts of food insecurity and hunger on children are well documented, with even short periods of food insecurity resulting in harm to their mental and physiological development and increased risk of adverse health outcomes.[17]

Taking assistance away from people who don’t show they are complying with a work requirement punishes them for the racial and gender inequities in our nation’s economy and labor market. As noted, non-elderly adults who are likely to participate in SNAP are generally working. But the structural disadvantages and discriminatory practices that individuals face, along lines of race, gender, disability, and other areas of identity, lead to higher unemployment, lower wages, occupational segregation, and limits on career advancement and upward economic mobility.[18] Taking assistance away from people unable to document compliance with a work requirement ignores the fact that substantial hiring discrimination persists in the United States. A recent literature review found that employers shown otherwise-equivalent resumes from fictitious job applicants are generally between 19 percent and 30 percent more likely to follow up with applicants who have typically “white” names than those with typically “Black” names.[19]

The jobs that public benefit program enrollees are able to get typically pay low wages and fail to provide basic benefits like paid sick and family and medical leave.[20] Lost wages and job loss often occur as a result of an individual’s or family member’s health issue, the latter of which especially affects women, who disproportionately serve as family caregivers.

The Johnson bill ignores these realities. It defines work for purposes of SNAP eligibility as half time — at least 80 hours of work a month. People with physical or mental limitations or who have caregiving responsibilities for family members they do not live with may be unable to work half time and would face the loss of food assistance despite working to the best of their capacity. The bill places the burden on people receiving food assistance to individually document their work or exemption and suggests that they and not their circumstances are the sole factor behind their need for help. It also suggests they lack the capacity to manage their own advancement and, if compelled to work, will be lifted out of poverty. But we know that SNAP itself has anti-poverty effects and, as explained below, helps people work.[21]

SNAP supports workers, including those who are paid low wages and those temporarily out of work. When a worker’s income drops low enough to qualify for SNAP, the program’s entitlement structure guarantees that food assistance will be available. This helps those recently unemployed while they look for new work and supports workers in very low-paying jobs. SNAP’s benefit formula favors earned income over other income through an earned income deduction and phases out benefits gradually as income rises, thus encouraging workers to seek greater income through higher wages and increased hours.

SNAP also supports individuals’ ability to succeed by mitigating the impacts of food insecurity on recipients’ health. Research shows a strong correlation between food insecurity and chronic health conditions among working-age adults. People facing food insecurity are more likely to postpone needed medical care or skip medication due to cost. When individuals who are food insecure do access health care, they are more likely to visit an emergency room or be admitted to (and stay longer at) a hospital compared to those who are food secure. As a result, food insecurity is associated with higher health care costs, whereas SNAP participation is associated with better access to preventative health care, improved medication adherence, and reduced health care costs.[22]

These positive impacts on participants’ health and well-being, in turn, can help people stay at work, return to work, and position them to get ahead.[23] SNAP participation among children is also associated with improved outcomes, including greater likelihood of graduating high school and improved earnings as adults.[24]

Appendix I:

Estimates on People Whose SNAP Benefits Would Be at Risk Under America Works Act

The state figures below are likely to be lower bounds of the number of people who would be at risk of losing food assistance under the Johnson bill in most states because they do not include estimates of the number of people who would be at risk of losing SNAP because of two of the bill’s provisions: the limit on state flexibility to apply for area waivers from the time limit and the ending of states’ ability to carry over discretionary exemptions.

The national totals for the number of people at risk of losing SNAP and newly subject to the time limit each include 700,000 people, which is based on a Trump Administration estimate of an earlier regulatory proposal to limit state discretion on area waivers that was less restrictive. We have not yet independently estimated that figure.

Estimated Number of People in Each State Whose SNAP Benefits Would Be at Risk Under H.R. 1581, The America Works Act
    SNAP households with children, with no child under age 7 
StateTotal number of SNAP participants in households at risk of losing SNAPShare of all SNAP participants in households at risk of losing SNAPNumber of SNAP participants newly subject to time limitNumber of adults aged 
18–64 without disabilities, newly subject to time limit
Number of children in household with adults as described in prior columnNumber of adults aged 50-64 without disabilities in households with no children newly subject to time limit
District of Columbia25,00025%15,0007,0009,00010,000
New Hampshire15,00022%8,0005,0007,0003,000
New Jersey150,00021%70,00045,00065,00025,000
New Mexico125,00028%65,00040,00050,00025,000
New York610,00023%335,000195,000230,000140,000
North Carolina345,00027%175,000115,000150,00060,000
North Dakota10,00022%5,0004,0005,0002,000
Rhode Island30,00021%15,00010,00010,0006,000
South Carolina175,00030%80,00055,00090,00025,000
South Dakota20,00023%9,0006,0009,0003,000
Virgin Islands7,00031%3,0002,0003,0001,000
West Virginia75,00025%40,00020,00030,00015,000
U.S. TotalMore than 10 million*About 26%About 6 million*3 million4 million2 million

* Includes an estimated 700,000 at risk of losing SNAP due to being newly subject to the time limit as a result of the elimination of the state waiver in areas with insufficient jobs. This number is based on a Trump Administration estimate of an earlier regulatory proposal to limit state discretion on area waivers that was less restrictive. We have not yet independently estimated that figure. 

Note: Estimates were developed using SNAP administrative data for fiscal year 2019, the most recent full year of data that are available. Estimates have been rounded to the nearest thousand for figures below 10,000 and to the nearest 5,000 for larger figures. See methodology. 

Source: CBPP analysis of 2019 SNAP Quality Control data.

Appendix II:


Most of the state and national numbers in this paper are based on the SNAP Quality Control (QC) data for federal fiscal year 2019, which includes detailed information on the economic and other circumstances of participating SNAP households in each state that is collected as part of SNAP’s QC system. Under SNAP’s QC system, states annually pull a representative sample of SNAP cases (totaling about 50,000 cases nationally over the year), and an independent state reviewer checks the accuracy of the original eligibility worker’s eligibility and benefit decision within federal guidelines. Federal officials then re-review a subsample of about a third of cases. The system produces overpayment and underpayment error rates.

The data from the sample also are published as a report for the public and are available as public use microdata for researchers.[25] The data for 2019 are the most recent full year of data available because Congress temporarily suspended the QC system for parts of fiscal years 2020 and 2021 during the COVID-19 pandemic. Fiscal year 2019 is very likely more similar to future years (during which time this provision could be in effect) than 2020 or 2021 would have been because of the high unemployment, temporary changes to SNAP eligibility and benefits, and other disruptions that occurred because of the pandemic during those years. In fiscal year 2019 there were approximately 37 million SNAP participants, which is similar to the 38 million average annual number of SNAP participants in the Congressional Budget Office’s current baseline forecasts over the 2024 to 2033 period. These estimates assume that the demographic composition of the SNAP caseload in the future will be similar to what it was in 2019. To not provide an overly precise estimate, we have rounded the 2019 figures to the nearest 5,000 when the figures were over 10,000 and to the nearest 1,000 when at 10,000 or below.

Using this 2019 QC data, CBPP is able to identify the number of people who newly would be subject to the time limit based on their age, household composition, and whether they receive disability-related benefits. This is the group we classify as “at risk” of losing food assistance. We are not, however, able to model all of the exemptions individuals may qualify for (such as a work-limiting condition or pregnancy), nor whether they would be able to document their work hours. So, as we point out, not all of the individuals at risk of losing SNAP would lose benefits under the proposal, but many would.

We have not yet estimated the effects of the proposed policy that would substantially limit states’ ability to apply for a waiver from the time limit for areas with “insufficient jobs.” The bill would limit these waivers to areas with unemployment rates over 10 percent. This is a more restrictive policy than the one that the Trump Administration proposed through regulation. It estimated that its less restrictive policy would have resulted in more than 700,000 people losing eligibility for SNAP. We have included that figure in the national totals because it represents a lower bound for the number of people who would lose eligibility under H.R. 1581 under the change to waiver policy. State-level estimates of this change are especially difficult to estimate because they depend on future economic conditions and state decisions about whether or not to apply for waivers. The table includes the Trump Administration’s earlier estimate in the national totals for the number of people who would be at risk of losing SNAP and the number newly subject to the time limit. Note that 15 states have state legislation prohibiting the state from applying for area waivers, so, without changes in state law no people in these states would lose SNAP because of the change in waiver policy.


End Notes

[1] Katie Bergh and Dottie Rosenbaum, “House Republicans’ Proposals Could Take Food Away from Millions of Low-Income Individuals and Families,” CBPP, March 20, 2023,

[2] Given that SNAP’s 20-hour work requirement for adults without a child in the house operates as a time limit since many subject to it have no way of meeting it, we use the terms “work requirement” and “time limit” interchangeably.

[3] America Works Act of 2023, H.R. 1581,; and Rep. Dusty Johnson, “Johnson Leads Work Requirements Fight in the U.S. House: The America Works Act closes loophole states have used to ignore existing work requirements,” March 14, 2023,

[4] America Works Act of 2023.

[5] Figures are based on CBPP analysis of the 2019 SNAP Household Characteristics data. In 2019 there were approximately 37 million SNAP participants, which is similar to the 38 million average annual number of SNAP participants in the Congressional Budget Office’s current baseline forecasts over the 2024 to 2033 period.

[6]Current rules allow states several ways to document there are insufficient jobs, including relying on the Labor Department’s “Labor Surplus Area” designation, a criterion that several federal agencies use to prioritize government contracts or assistance based on elevated unemployment rates.

[7] We have not independently estimated the impact of the Trump Administration’s proposed rule.

[8] See, for example, Wenhui Feng, “The Effects of Changing SNAP Work Requirement on the Health and Employment Outcomes of Able-Bodied Adults without Dependents,” Journal of the American Nutrition Association, Vol. 41, No. 3, February 2021,

[9] Colin Gray et al., “Employed in a SNAP? The Impact of Work Requirements on Program Participation and Labor Supply,” American Economic Journal: Economic Policy, Vol. 15, No. 1, pp. 306-41,

[10] Tracy Vericker et al., “The Impact of ABAWD Time Limit Reinstatement on SNAP Participation and Employment,” Journal of Nutrition Education and Behavior, March 2, 2023, S1499-4046(23)00008-8,

[11]For details on the most common jobs of SNAP participants by state, see

[12] Kristin F. Butcher and Diane Whitmore Schanzenbach, “Most Workers in Low-Wage Labor Market Work Substantial Hours, in Volatile Jobs,” CBPP, July 24, 2018,

[13] Ibid.

[14] Jennifer Wagner and Jessica Schubel, “States’ Experiences Confirm Harmful Effects of Medicaid Work Requirements,” CBPP, updated Nov. 18, 2020,

[15] See Ohio Association of Food Banks, “Franklin County Work Experience Program: Comprehensive Report, Able-bodied Adults Without Dependents,” 2014,

[16] CBPP, “Taking Away Medicaid for Not Meeting Work Requirements Harms People with Disabilities,” updated March 10, 2022,

[17] Steven Carlson and Joseph Llobrera, “SNAP Is Linked With Improved Health Outcomes and Lower Health Care Costs,” CBPP, December 14, 2022,

[18] Butcher and Schanzenbach.

[19] S. Michael Gaddis et al., “Discrimination Against Black and Hispanic Americans is Highest in Hiring and Housing Contexts: A Meta-Analysis of Correspondence Audits,” Social Science Research Network, Vol. 30, No. 20, December 2021,

[20] Kathleen Romig and Kathleen Bryant, “A National Paid Leave Program Would Help Workers, Families,” CBPP, April 27, 2021,

[21] CBPP, “Chartbook: SNAP Helps Struggling Families Put Food on the Table,” November 7, 2019,

[22] Carlson and Llobrera.

[23] Christian A. Gregory and Partha Deb, “Does SNAP Improve Your Health?” Food Policy, Vol. 50, January 2015, pp. 11-19, See also Carlson and Llobrera.

[24] Sharon Parrott et al., “Building an Equitable Recovery Requires Investing in Children, Supporting Workers and Expanding Health Coverage,” CBPP, March 24, 2021,

[25] Katheryn Cronquist, “Characteristics of Supplemental Nutrition Assistance Program Households: Fiscal Year 2019, USDA, Food and Nutrition Service, 2021, and SNAP Quality Control Data,