The Supplemental Nutritional Assistance Program (SNAP), often referred to as food stamps, provides low-income families with in-kind transfers to purchase groceries at selected grocery stores. In 2019, over 36 million U.S. citizens were enrolled to receive SNAP benefits in any given month with over $60 billion in assistance given out in total.
Despite the large role that SNAP plays in the social safety net in the U.S., many eligible U.S. citizens did not receive SNAP benefits. Unrath (2021) investigates the role that reporting requirements and eligibility checks play in incomplete take-up of SNAP benefits. The paper uses monthly enrollment histories for all SNAP participants in California between 2005 and 2020 and takes advantage of a 2013 policy change in which California households went from being required to submit eligibility reports every 3 months to every 6 months to estimate the MVPF of eliminating reporting requirements for SNAP.
MVPF = 2.1
The paper follows the framework introduced in Finkelstein and Notowidigdo (2019) and divides the population into (relatively) low-income and high-income participants. The relatively low-income individuals are those with likely higher need for SNAP; the relatively high-income individuals are those with likely less need for SNAP.
The net costs are comprised of two components.
(i) For each type, the sum of (for each month of induced enrollment, the average monthly benefit received by that type plus the income tax revenue lost due to labor supply responses by that type to SNAP enrollment, multiplied by the change in enrollment by that type in that month relative to enrollment).
minus (ii) the cost to the government of administering an additional eligibility verification.
The paper considers the reform’s effects on retention from four to six months after enrollment to reflect the change from quarterly to semi-annual eligibility checks, which means that part (i) has three pieces for each type: one for each of months 4, 5, and 6 since enrollment.
The paper estimates the average monthly benefit received by the low-income type to be $416 and $348 for the high-income type.
The paper uses estimates from Hendren and Sprung-Keyser (2020) and estimates that the fiscal externality is $0.16 for every $1 in SNAP benefits received by adults, and -$0.11 for every $1 in SNAP benefits received by children. The paper assumes no fiscal externality for seniors. The paper calculates a net fiscal externality for each type by multiplying these fiscal externalities by the share of adults/children/seniors for each type, and then multiplying that by the average benefit amount. This yields a fiscal externality of $3.24 for the low-income type and $12.60 for the high-income type.
The sum of the change in enrollments between months 4 and 6 was 0.24 for the low-income type and 0.29 for the high-income type.
For (ii), the paper assumes the public cost of administering a quarterly report is about $100, and that this cost is the same between types.
Using the paper’s estimates of the change in enrollment after the 2013 reform, the net costs are calculated as ($416+$3.24)(0.24) +($348+$12.60)(0.29) – $100 = $105.19.
As described in the Costs section, the paper uses the framework introduced in Finkelstein and Notowidigdo (2019) and divides the population into low-income and high-income participants (i.e., those with relatively more or less need for SNAP).
The willingness to pay for the reduced reporting requirements is comprised of two components:
(i) For each group, the sum of the (for each month of induced enrollment, the average monthly benefit received by that type plus, multiplied by the change in enrollment by that type in that month relative to enrollment).
plus (ii) the private cost of completing the eligibility requirements.
The paper estimates the average monthly benefit received by the low-income type to be $416 and $348 for the high-income type. The sum of the change in enrollments between months 4 and 6 was 0.24 for the low-income type and 0.29 for the high-income type.
The paper follows Finkelstein and Notowidigdo (2019) and assumes the private opportunity cost for completing the eligibility requirements is twice the minimum wage. The paper uses the minimum wage in California in 2013, implying a personal cost of $20 for both types.
Using the paper’s estimates of the change in enrollment after the 2013 reform, the overall willingness to pay is calculated as $416(0.24) + $348(0.29) + $20 = $220.76.
The paper estimates the MVPF of the 2013 reform that reduced reporting requirements for SNAP as $220.76 / $105.19 = 2.09
Amy Finkelstein and Matthew J. Notowidigdo (2019). “Take-up and Targeting: Experimental Evidence from SNAP.” The Quarterly Journal of Economics, 134(3): 1505–1556. https://doi.org/10.1093/qje/qjz013
Nathaniel Hendren and Ben Sprung-Keyser (2020). “A Unified Welfare Analysis of Government Policies.” The Quarterly Journal of Economics, 135(3): 1209–1318. https://doi.org/10.1093/qje/qjaa006
Matthew Unrath (2023). “Targeting, Screening, and Retention: Evidence from the Supplemental Nutrition Assistance Program in California.” Working Paper. https://mattunrath.github.io/files/research/Unrath_SNAP.pdf