Dashboard

Text last updated

24 de Septiembre, 2026

| Dottie Rosenbaum , Joseph Llobrera , Catlin Nchako y Luis Nuñez

SNAP Tracker: People Are Losing Food Assistance as the Harmful 2025 Republican Reconciliation Law Is Implemented

SNAP Participation Dropping Following Enactment of Harmful 2025 Republican Reconciliation Law

Percentage change in SNAP participants between and

Note: Some states appear to have anomalous data. See “Chart Notes” below.

Source: CBPP calculations from USDA and state program data

Center on Budget and Policy Priorities | cbpp.org

Key Takeaways

Between the Republican reconciliation law’s July 2025 enactment and June 2026 (the last month with data on all states):

SNAP participation is down by over 5 million people, or 13 percent, in USDA data.
SNAP participation has dropped in every state except Alaska, including by 5 percent or more in 48 states, by 10 percent or more in 24 states, and close to 20 percent or more in 5 states, in USDA data.
SNAP participation has fallen below the pre-pandemic level of nearly 37 million in February 2020, and has reached its lowest point in 17 years.
The last time there was such a steep decrease in participation in this short period of time (other than temporary spikes following natural disasters) was nearly three decades ago, after Congress enacted very deep cuts to SNAP (then the Food Stamp Program) in 1996. SNAP participation dropped by 13 percent (3.1 million people) in the 11 months from March 1997 to February 1998.
In the 28 states with publicly available data or that have shared unpublished data, the number of children receiving SNAP has fallen by 1.5 million since July 2025. It’s very likely that nationwide, the number of children receiving SNAP has dropped by about 2 million since July 2025.^[1]
Unemployment nationwide has been flat at about 4 percent since July 2025, BLS data show. At best, it’s down in 28 states and the District of Columbia by a percentage point. It’s very unlikely that reduced need is driving the decline in SNAP participation.

These participation declines began before some of the major changes from the harmful 2025 Republican reconciliation law were fully in effect. We expect these harms to grow as states fully implement eligibility restrictions and take actions to lower their error rates.

The harmful Republican reconciliation law enacted July 4, 2025, poses great harm to families, including higher food costs, poverty, and hunger. It makes deep cuts to federal funding for food assistance through the Supplemental Nutrition Assistance Program (SNAP), shifting significant new costs to states. It dramatically expands SNAP’s already harsh and ineffective provision taking away people’s benefits for not meeting a work requirement. And it ended eligibility for many people with a lawful immigration status.

We estimate that SNAP participation nationwide fell by over 5 million people (nearly 13 percent) between the law’s July 2025 enactment and June 2026, based on the latest month of data from the U.S. Department of Agriculture (USDA).

The declines are especially pronounced in some states. SNAP participation fell in Arizona by 52 percent from July 2025 to June 2026. The decline is 46 percent from July 2025 to August 2026, using more recent state data. The caseload declines are close to 20 percent or more in Arizona, Illinois, Louisiana, Nevada, and Oklahoma, and — using more recent state data — in Florida, Kansas, and Nebraska.

The declines started before the harmful 2025 Republican reconciliation law’s enactment, suggesting factors at play in addition to that law. But in almost all states, declines in SNAP participation accelerated after the 2025 Republican reconciliation law, and we expect that trend to continue. (Some states have had very small increases in participation in recent months, but remain substantially below their level in July 2025.) Use the map and charts below to see details.

In the 28 states with available data, the number of children receiving SNAP food assistance has fallen by 1.5 million since July 2025. Put another way, children account for close to half of the nearly 3.6 million people who are no longer receiving SNAP in those states. (See table.)

Starting in October 2027, most states will have to pay between 5 and 15 percent of SNAP benefit costs, totaling hundreds of millions of dollars a year in many states. And the amount a state will have to pay will be based on fiscal year 2025 or 2026 error rates. The magnitude of the cost shift and the urgency surrounding error rates incentivized states to take drastic measures to reduce their payment error rates quickly and cut program costs, even if it meant delaying or improperly denying benefits to eligible people. We’re likely seeing some of those initial effects in SNAP participation data.

State	Most Recent Month of Data	Change in Total Participation		Change in Child Participation
State	Most Recent Month of Data	Count	Percent	Count	Percent	Share of State’s Total Participation Change
Alabama	May 2026	-81,838	-11%	-23,324	-7%	29%
Arizona	August 2026	-416,924	-46%	-172,777	-46%	41%
Arkansas	July 2026	-24,813	-10%	-8,687	-9%	35%
California	May 2026	-333,062	-6%	-156,440	-9%	47%
Colorado	June 2026	-33,681	-5%	-12,598	-5%	37%
Connecticut	June 2026	-56,044	-16%	-15,668	-14%	28%
Florida	July 2026	-647,822	-22%	-247,073	-22%	38%
Georgia	June 2026	-133,135	-9%	-52,242	-8%	39%
Idaho	June 2026	-9,332	-7%	-4,756	-9%	51%
Kansas	August 2026	-41,240	-22%	-18,024	-21%	44%
Louisiana	June 2026	-170,227	-21%	-82,064	-23%	48%
Maine	August 2026	-22,866	-13%	-6,897	-13%	30%
Maryland	July 2026	-33,869	-5%	-16,418	-6%	48%
Massachusetts	July 2026	-188,295	-18%	-67,260	-20%	36%
Michigan	June 2026	-138,337	-10%	-46,366	-9%	34%
Minnesota	May 2026	-17,077	-4%	-13,082	-7%	77%
Missouri	July 2026	-39,707	-6%	-15,860	-6%	40%
Montana	June 2026	-6,583	-8%	-2,598	-8%	39%
Nebraska	June 2026	-24,532	-16%	-11,934	-17%	49%
New Hampshire	June 2026	-2,454	-3%	-945	-4%	39%
New Jersey	June 2026	-53,523	-7%	-21,339	-6%	40%
New Mexico	July 2026	-22,899	-5%	-12,634	-7%	55%
Ohio	May 2026	-107,573	-8%	-50,477	-9%	47%
Pennsylvania	August 2026	-235,748	-12%	-67,836	-10%	29%
South Dakota	August 2026	-4,688	-6%	-2,725	-8%	58%
Texas	August 2026	-689,325	-20%	-357,875	-21%	52%
Vermont	June 2026	-5,332	-8%	-1,812	-10%	34%
Washington	May 2026	-57,319	-6%	-24,939	-8%	44%
Total	-	-3,598,245	-13%	-1,514,650	-14%	42%
* Ohio defines children as individuals under the age of 22 who receive SNAP benefits and live with a parent. Notes: States shown are those that publicly report SNAP participation data for children, or do not currently publish participation data publicly but have shared unpublished data. Kentucky also provides child participation data but is omitted due to data quality concerns. Minnesota’s published state data on total participation are not comparable to its data on children, because the former includes recipients of state-funded benefits. As a result, we calculated shares in this table using USDA program data on total participation in Minnesota. For three states (Colorado, Idaho, and Nebraska), we calculate shares using the unpublished data on total participation that the states shared, although that data differs slightly from the same data on total participation that the agencies publish online. Source: CBPP calculations of SNAP state agency child participation data as of September 23, 2026

In addition, the harmful 2025 Republican reconciliation law expanded SNAP’s already harsh and ineffective work requirement, extending it to older adults up to age 65 and to parents and other caregivers of children age 14 and older. It also substantially narrowed states’ ability to waive the time limit for areas with few job opportunities and eliminated exemptions for veterans, people experiencing homelessness, and former foster youth. Research consistently shows the work requirement cuts many people off SNAP without increasing employment or earnings, indicating that many of these recipients are likely to be cut off from SNAP benefits as states ramp up implementation.

We are tracking state SNAP participation data to identify trends likely attributable to the 2025 Republican reconciliation law. This tracker shows the most recent federal data on SNAP participation and often more recent data from state agencies administering the SNAP program, as well as unemployment data from the Bureau of Labor Statistics (BLS). This tracker is initially focused on SNAP participation and will be regularly updated as new data become available. Additional indicators may be added over time.

The Congressional Budget Office (CBO) estimates that the harmful 2025 Republican reconciliation law cuts SNAP by nearly $187 billion through 2034, the largest cut to the program in history, and that more than 2.4 million people in a typical month would be cut from SNAP. When also including people losing a substantial amount, but not all, of their SNAP benefit, we estimated based on CBO’s figures that 4 million people in a typical month will lose out, once the changes are fully implemented. The actual drops in SNAP participation in the first year after the law’s enactment have already far exceeded CBO’s estimates.

The Harmful 2025 Republican Reconciliation Law’s Impact on SNAP Administration

Several of the harmful 2025 Republican reconciliation law’s provisions directly eliminate eligibility for SNAP, including the expansion of the work requirement to new groups of recipients, the significant limitations on work requirement waivers for areas with higher unemployment, and new restrictions that take SNAP away from people with certain lawful immigration statuses.

The 2025 Republican reconciliation law is also substantially influencing how state agencies administer SNAP. To minimize the amount they will be required to pay under the reconciliation law’s cost-shift provisions, states are incentivized to adopt policies that will reduce error rates quickly. Error rates are measures of states’ over- and under-payments in SNAP, mostly resulting from unintentional mistakes by state agencies and households, not fraud. The states’ error rates in fiscal years 2025 and 2026 will determine how much they pay in fiscal year 2028, the first year the cost-shift provision is in effect.

Many of the policies the states are already adopting in an effort to lower error rates are likely to result in eligible households losing SNAP food assistance because of administrative burdens. This can include policies such as requiring more frequent verification of income and expenses, which may make it more difficult for eligible households to access benefits or result in processing delays. Unfortunately, given the way SNAP error rates are calculated, wrongly denying or delaying benefits to an eligible household is not considered an “error.”

SNAP participants in some states already were facing long application delays and call center wait times before the 2025 Republican reconciliation law’s enactment. The law’s short implementation timeline and cut to the federal investment in state administrative expenses are making it hard for state agencies to adjust their operations. As a result, SNAP participation may decline faster in the coming months than might be expected from the law’s implementation, resulting in increased poverty, food insecurity, and hunger.

State SNAP Participation and Unemployment Trends

SNAP Participation in

Source: USDA and state agency program data.

Center on Budget and Policy Priorities | cbpp.org

SNAP participation may have dropped in part due to the harmful 2025 Republican reconciliation law, even before some of its major provisions have taken effect. Larger-than-expected participation declines, for example, may indicate situations where states are implementing policies in an effort to reduce error rates that have the effect of broadly reducing SNAP enrollment. Other constraints, such as staffing shortages, may also be at play. SNAP also faced an unprecedented disruption during the government shutdown late in 2025, during which the Trump Administration illegally refused to deliver benefits.

This chart includes the number of monthly SNAP participants that states report to USDA and, when applicable, more recent data that states post on their websites. The latter data are sometimes slightly different from USDA’s due to differences in how they both count SNAP enrollment, but the trends are typically the same.

For information on participation irregularities, see “Chart Notes” box below.

Change in SNAP Participation Since Harmful 2025 Republican Reconciliation Law Enactment

Source: CBPP calculations from USDA program data.

Center on Budget and Policy Priorities | cbpp.org

Monthly Unemployment Rate in

Note: Unemployment data for October 2025 were not published due to federal government shutdown.

Source: U.S. Bureau of Labor Statistics.

Center on Budget and Policy Priorities | cbpp.org

This chart shows state unemployment rates from August 2024 through the most recent month of data available from the U.S. Bureau of Labor Statistics (BLS).

Changes in SNAP participation are driven by changes in the number of people eligible for the program and the share of eligible people who successfully enroll. SNAP responds to changes in poverty and the economy. As unemployment increases, poverty increases and more people become eligible for SNAP benefits. States with rising unemployment may see an increase in demand for food assistance. States with falling SNAP participation while unemployment is rising or holding steady may indicate that an increasing number of eligible families cannot access SNAP to help keep food on their table. Data from the Urban Institute indicate that food insecurity remained high in 2025, just as states began to implement deep cuts to SNAP and health care coverage .

Download the data: https://apps.cbpp.org/fa2-24-26/data/downloadable_data.xlsx

Chart Notes

USDA SNAP participation data may include disaster assistance (D-SNAP). USDA provided D-SNAP in the fall of 2024 to Georgia, Florida, North Carolina, South Carolina, Tennessee, and Virginia due to hurricanes Helene and Milton; to Louisiana due to hurricane Francine; and to New Mexico due to severe storms. It provided D-SNAP to California in February 2025 due to wildfires, and Kentucky and West Virginia in the spring of 2025 and Arkansas and New Mexico in the summer of 2025, all due to severe storms. States do not often include disaster SNAP in the data they post.

USDA and BLS routinely revise historical data when a new month of data are released. As a result, the data in this tracker may change slightly as the tracker is updated over time. The most recent month of USDA data are usually preliminary. State agency data may be preliminary and subject to later revisions.

State SNAP participation trends may show some anomalies due to irregularities in the federal or state agency data.

USDA data for North Dakota show an anomalous blip in October 2025 (so we replaced data for that month with participation data verified through the state agency). State agency SNAP participation data for Ohio appear to be misreported for the month of October 2025, so we have omitted them. For Minnesota, the number of people the state reports includes a small number of people who receive state-funded food benefits. In addition, Minnesota, under unique authority known as MFIP, combines SNAP food benefits with cash assistance under the Temporary Assistance for Needy Families program for some families.

Puerto Rico does not participate in SNAP. It receives a block grant for nutrition assistance.

Más sobre este tema

End Notes

^[1] Among states that have provided participation data for children, the share of the state’s total SNAP participation decline that is from fewer children participating ranged from about 30 percent to about 77 percent. To calculate this national estimate, we conservatively took the low end of this range and assumed that the decline in the number of children comprised roughly 30 percent of the total participation decline in states where we do not have participation data for children. For half of the 28 states the number of children participating in SNAP is posted publicly. For 14 states ― Alabama, Arkansas, California, Colorado, Connecticut, Georgia, Idaho, Minnesota, Montana, Nebraska, New Hampshire, Pennsylvania, Vermont and Washington ― CBPP is using unpublished data from the state agency.