State Implementation of SNAP Waivers and Flexibilities During the COVID-19 Pandemic: Perspectives From State Agency Leaders.

OBJECTIVE: To describe state agencies' implementation of the Supplemental Nutrition Assistance Program (SNAP) during the first year of the coronavirus disease of 2019 (COVID-19) pandemic, barriers and facilitators to SNAP implementation, and recommendations to improve SNAP implementation.
DESIGN: Qualitative methodology guided by Bullock's determinants of policy implementation framework using 7 semistructured, virtual focus groups in April 2021.
SETTING: Twenty-six states representing all 7 US Department of Agriculture Food and Nutrition Service regions. PARTICIPANTS: Four focus groups with state-level SNAP administrators and 3 focus groups with state-level SNAP supportive services (Supplemental Nutrition Assistance Program-Education, Employment & Training, and Outreach) supervisors (n = 62). PHENOMENON OF INTEREST: Supplemental Nutrition Assistance Program implementation during the COVID-19 pandemic. ANALYSIS: Thematic analysis using a phronetic iterative approach.
RESULTS: Six primary themes emerged: the policy response, technology needs, collaboration, participant communication, funding realities, and equity. Implementation challenges included the design of waivers in the early pandemic response, inadequate federal guidance and funding, outdated technology, and prepandemic regulations limiting state authority. Modernized technology systems, availability of virtual programming, partnerships, and enhanced benefits facilitated SNAP implementation. CONCLUSIONS AND IMPLICATIONS: Supplemental Nutrition Assistance Program administrators adapted their programs to deliver services virtually during the COVID-19 pandemic. These experiences highlighted the importance of certain policy determinants, such as modernized technology and streamlined application processes, to improve outcomes for SNAP participants and staff.

INTRODUCTION

The coronavirus disease of 2019 (COVID-19) pandemic tested the ability of the US Department of Agriculture (USDA) Supplemental Nutrition Assistance Program (SNAP) to reduce food insecurity and poverty—2 important social determinants of health.1, 2, 3, 4 The SNAP is the nation's largest nutrition assistance program and one of the most effective federal programs in assisting families during economic downturns; it reduces poverty, improves food security, and invests in children's long-term health while promoting economic growth. , Prioritizing equitable access to SNAP and adequate benefits for eligible families may help reduce household food insecurity, help individuals and families meet basic needs, and improve health outcomes for lower-income households.

Opportunities exist to improve SNAP benefit adequacy, expand participation among underserved groups, and reduce the administrative burden of applying for and staying on SNAP. , Households eligible for SNAP receive a monthly benefit intended to supplement household food budgets on the basis of household size, income, and assets. A large body of evidence shows that the SNAP benefit amount has been too low to support households’ nutritional needs, resulting in food hardship, reduced energy intake, and negative health and social consequences. , 8, 9, 10, 11 Groups experiencing disproportionately high risk of food insecurity, including college students, immigrants, people formerly incarcerated, and nondisabled adults without dependents, have either been ineligible for SNAP or subject to time limits contingent on work requirements. In addition, not all people eligible for SNAP enroll, with particularly large gaps in participation among older adults, people with disabilities, and families with immigrants. , Although many states are making strides toward streamlining applications and offering online enrollment or recertification, applying for and staying on SNAP can be challenging. The USDA estimated that in 2011 between 17% and 28% of SNAP-eligible households across 6 states lost benefits because of administrative hurdles, including unclear communication from administering agencies and burdensome interview requirements at recertification.

Opportunities also exist to improve SNAP supportive services, including Supplemental Nutrition Assistance Program-Education (SNAP-Ed), Outreach, and Employment & Training (E&T). The SNAP supportive services aim to promote healthy and active lifestyles (SNAP-Ed) in low-income communities; inform eligible low-income people not participating in SNAP about eligibility requirements and enrollment procedures (Outreach); and provide job training, skill development, and other support (eg, child care, transportation) to SNAP participants seeking employment (E&T).16, 17, 18 The SNAP supportive services are administered by state agencies, often in partnership with collaborating organizations, implementing agencies, or contractors. Because many administrative costs are reimbursed by the federal government, SNAP-Ed, Outreach, and E&T activities must adhere to approved state plans to receive reimbursement.16, 17, 18 These funding realities have, historically, made it difficult for state agencies and their collaborators to reach particularly disconnected or underresourced populations. ,

The COVID-19 pandemic fundamentally altered how SNAP agencies deliver benefits, interact with participants, and provide supportive services, creating an opportunity to study innovation in program implementation. In March-April 2020, as state offices closed and agency staff transitioned to telework, SNAP experienced one of its largest increases in applications in the program's history. Compared with the 2008 recession, during which time SNAP participation increased by no more than 6% over any 3 months, SNAP applications increased by more than 17% between February and May 2020. , In the months that followed, Congress passed legislation (eg, the Families First Coronavirus Response Act), and USDA issued waivers that allowed states to adjust rules on how participants apply for SNAP benefits and engage with SNAP supportive services (ie, SNAP-Ed, Outreach, and E&T). Congressional legislation and USDA guidance put forth in March 2020-April 2021 allowed states to modify SNAP administrative processes (eg, means of verifying participant eligibility and length of certification periods), benefit issuance (eg, household benefit amounts), and supportive services (eg, SNAP-Ed, Outreach, and E&T activities eligible for reimbursement). A detailed description of key legislative and regulatory actions impacting different programmatic and policy areas is provided in Table 1 . These changes were intended to help new and existing participants access SNAP benefits and supportive services while state agencies managed increased caseloads and shifted to remote work.

Table 1

Description of SNAP COVID-19 Waivers and Program Modifications Discussed in SNAP Administration Focus Groups

Waiver or Program Modification	Description
Policy area 1: Certification periods
Certification period	The amount of time a client's SNAP case remains active, typically 12 months, with a 6-month periodic report to identify household changes that may affect benefit levels or eligibility. Some states provide longer certification periods (24–36 mo) for households comprising only older adults and/or persons with a disability
Extension of certification periods and adjustment of periodic reporting waiver	Originally authorized through the FFCRA (March 2020),2,23 this waiver allowed states to extend certification periods and periodic reports, with the FNS guiding to extend periods by 6 mo. This allowed for cases to remain active without additional verification. The continuing resolution (Continuing Appropriations Act of 2021) in October 202029 provided further flexibility for states to adjust the length and duration of extensions without the preapproval of the Secretary of Agriculture
Policy area 2: Interviews
Face-to-face interview waiver	Originally authorized through the FFCRA,2 this waiver allowed states not to provide the option to hold a face-to-face interview with an applicant. The Continuing Resolution in October 202029 allowed states to use this waiver without preapproval from the Secretary of Agriculture
Initial and Recertification Interview Waiver	Originally authorized through the FFCRA, this waiver allowed states to approve benefit receipt without completion of an interview given identity had been verified, and all mandatory components of an application were complete. The Continuing Resolution in October 2020 allowed states to use this waiver without preapproval from the Secretary of Agriculture
Policy area 3: Food access and purchasing
Emergency allotments	Originally authorized through the FFCRA, states could provide supplemental SNAP benefits to raise all cases to the maximum household benefit level for months in which a state and federal public health emergency were in place. Food and Nutrition Service3 later amended guidance to permit a benefit increase for all households, including those already at the maximum benefit, and to provide a transitional month of benefits after authority for emergency allotments expires
15% benefit increase	Originally authorized through the Consolidated Appropriations Act of 2021 (December 2020)30 and extended through the American Rescue Plan (March 2021),31 this benefit increase provided a universal increase of 15% to all households from January 2021 to September 2021
Expansion of college student eligibility	Authorized through the Consolidated Appropriations Act of 2021,30 college students who were eligible for work-study or have no expected family contribution were temporarily granted SNAP eligibility through the duration of the federal public health emergency
Temporary suspension of nondisabled adults without dependent's work requirements	Authorized through the FFCRA, this allowed states to partially suspend work requirements for SNAP recipients that are nondisabled adults without dependents by suspending the time limit of benefit receipt through the month after the month in which the federal public health emergency declaration ends
Policy area 4: Application processing and verifications
Common application	An application that integrates application components and determines eligibility for SNAP and at least 1 other safety net program (eg, Medicaid)
Integrated eligibility system	An eligibility system (ie, a computing software) that integrates the administration of management of multiple safety net programs
Legacy system	An outdated eligibility system remains in use by some state agencies because of limited resources (staff and financial) which prevents upgrades
Modernized system	An updated eligibility system that uses more recent, industry-standard technology for SNAP case processing and management compared with legacy systems
Audio recordings for telephonic signatures waiver	Authorized through the FFCRA, this waiver allowed states to document in writing the verbal attestation of a client in place of an audio-recorded verbal attestation of a client (a signature is required for a SNAP application to be complete)

COVID-19 indicates coronavirus disease 2019; FFCRA, Families First Coronavirus Response Act; FNS, Food and Nutrition Service; SNAP, Supplemental Nutrition Assistance Program.

Prior work has shown significant variation in state waiver uptake and implementation within and across states during the COVID-19 pandemic. For example, many states were approved to extend SNAP certification periods in spring 2020, but some only applied this waiver to certain households, and much discontinued use of the waiver after several months. In addition, some waivers authorized during the pandemic, such as the ability to waive interview requirements for certain households, were viewed by state SNAP administrators as opportunities to increase program participation and administrative efficiency during normal circumstances (ie, beyond the national state of emergency).

Although groups have documented the types of program modifications available to and used by state agencies during the COVID-19 pandemic, there are gaps in knowledge about how SNAP and its supportive services were implemented, why states implemented certain program changes and not others, and the barriers and facilitators to implementing the nation's largest nutrition assistance program during a period of diminished staff capacity and increased program need.24, 25, 26 In this study, we draw from the implementation science field to address these gaps. Implementation science seeks to understand the process of translating research into practice and policy, identify determinants of implementation outcomes, and evaluate implementation success. To address the multilevel (ie, federal, state, and local stakeholders) and multifaceted complexities of policy implementation, Bullock and colleagues developed the determinants of the policy implementation framework, which provides a theoretical underpinning to policy implementation drawing from the fields of implementation science and public policy. This study uses a qualitative approach grounded in the determinants of policy implementation framework to describe: (1) how state agencies implemented SNAP and its supportive services in the first year of the COVID-19 pandemic, (2) determinants (ie, barriers and facilitators) of SNAP implementation, and (3) recommendations from state-level administrators about how to better prepare SNAP for future emergencies and longer-term.

METHODS

Participants and Recruitment

Focus groups were conducted with state-level SNAP administrators or supervisors working in 1 of 4 programmatic areas: (1) administration (ie, application processing, verification, recertification, and benefit issuance); (2) SNAP-Ed, (3) Outreach, or (4) E&T. We purposively recruited individuals holding state-level managerial or supervisory positions within the programmatic areas of interest; job titles included managers, directors, coordinators, and supervisors. Before sending invitations to participate, we classified all 50 states and Washington DC into 2 mutually exclusive groups: high waiver and low waiver utilizers. Waiver use was determined by summing the number of unique program waivers requested and approved in each state through January 2021 on the basis of data collected in a prior survey and documentation by the USDA. , We also gathered data on state characteristics associated with SNAP administration, including USDA Food and Nutrition Service (FNS) region, state vs county administration (some states administer SNAP through county-level government but are supervised at the state-level), baseline application processing technology, size of the SNAP caseload before the pandemic, and state governor political party affiliation (given historical partisan support for SNAP). Data on state characteristics were collected through public sources, including the USDA, US Census Bureau, Center for Budget and Policy Priorities, National Association of County and City Health Officials, and Code for America. , 33, 34, 35, 36 In March 2021, the American Public Human Services Association (APHSA) —a bipartisan membership association representing state and local human services agencies—emailed all state agencies that administer SNAP, inviting individuals holding state-level managerial or supervisory roles in the programmatic areas of interest to participate in a focus group. One week later, a second invitation was sent to states that had not yet responded. To ensure representation from diverse perspectives, we sent a second invitation to some states so that our sample reflected a range of waiver utilization and other characteristics (eg, the political party of leadership, FNS region).

This study was deemed exempt and did not require Institutional Review Board oversight per the Johns Hopkins Bloomberg School of Public Health Institutional Review Board. With this determination, we still obtained verbal consent from all participants before data collection and audio recording. Individuals were not compensated for participation.

Focus Group Methodology

Seven focus groups were held via Zoom in April 2021 and ranged from 60–90 minutes in length. We selected the focus group methodology as this allows participants to engage in dynamic discussion over shared common experiences (ie, the implementation of SNAP waivers and flexibilities), generating a greater understanding of attitudes, perceptions, and experiences with the phenomenon of interest. Four groups focused on SNAP administration, and 1 each focused on SNAP-Ed, E&T, or Outreach. The 4 focus groups on SNAP administration were organized according to waiver uptake: 2 groups included individuals from states classified as high utilizers, and 2 groups included individuals from states classified as low utilizers. We constructed 1 focus group each on SNAP-Ed, E&T, and Outreach to supplement information shared in the SNAP administration focus groups.

Some participants brought colleagues to listen to the focus group, but no more than 2 individuals from each state verbally responded to questions on behalf of their team. A study team member facilitated and moderated focus groups in alignment with a semistructured guide. The moderator was a graduate student researcher with advanced training in qualitative methods and previous experience conducting focus groups and interviews for research. This researcher also led the development of the moderator guide. Three additional study team members trained in qualitative methods attended focus groups to co-facilitate and take notes; team members completed a 1-hour training about roles and expectations. Co-facilitators were instructed to send probing questions of interest directly to the moderator via private chat to incorporate into the discussion as appropriate. Participants were invited to have their video on and were encouraged to share thoughts verbally instead of through the chat feature of Zoom (Zoom Video Communications, Version 5.11.6, 2021). One co-facilitator monitored the chat should any participant choose to use this available feature; the moderator read all comments in the chat aloud and incorporated them into questioning. The moderator had no prior relationship with focus group participants. Two co-facilitators in focus groups were APHSA staff members who built prior rapport with participants through professional workgroups and meetings.

Moderator Guide

Four moderator guides were developed for the focus groups on the basis of findings from a prior survey, which 43 state agencies completed from December 2020 to January 2021 and asked about SNAP waiver use and program adaptations in the first 12 months of the pandemic. Questions asked in each guide were informed by the determinants of policy implementation framework by Bullock et al, which describes the factors affecting policy implementation and the relationships between those factors. Specifically, the framework defines 8 determinants of policy implementation, including (1) characteristics of the policy, (2) policy formulation process, (3) vertical public administration and thickness of hierarchy, (4) network/interorganizational relationships, (5) implementing agency responses; (6) attributes and response from those affected by the policy, (7) timing/sequencing, and (8) external environment or policy context. We drew from these 8 determinants to inform our questioning. Guides were tailored to each SNAP service area and reviewed by 4 SNAP policy experts at APHSA to ensure accuracy and relevance to the state of SNAP policy at the time of the focus groups. Questions included in the moderator guide were not revised during data collection.

The guide used in the 4 SNAP administration focus groups included questions in 4 policy areas: (1) certification periods, (2) interviews, (3) food access and purchasing, and (4) application processing and verifications (Table 1). Policy area 1 (certification periods) asked about waivers that allowed states to extend SNAP certification periods. Policy area 2 (interviews) asked about waivers that allowed states to adjust interview requirements during initial application and recertification. Policy area 3 (food access and purchasing) asked about program modifications allowing states to provide supplemental benefits (emergency allotments), increase SNAP benefits by 15%, and expand SNAP eligibility among college students. Policy area 4 (application processing and verifications) asked about the use of technology and waivers to facilitate remote application processing, including telephonic signature collection. Consistent with the framework by Bullock and colleagues, in each policy area questions were asked about how states adapted operations using waivers (the implementing agency response), implementation challenges and successes encountered (network/interorganizational relationships, timing/sequencing, and external environment), impacts of federal policy changes (the policy formulation process and vertical public administration and thickness of hierarchy), recommendations for future emergencies, and implications for the future of SNAP (characteristics of the policy). The guides for SNAP-Ed, E&T, and Outreach focus groups included questions about how states modified supportive services during the pandemic, implementation challenges encountered when modifying services, implementation successes achieved by state leadership and staff, and recommendations for the future. For example, the study team asked about program adaptations required to create virtual nutrition education curricula or to communicate frequent program changes with participants remotely.

Data Management and Analysis

Focus groups were audio-recorded, transcribed verbatim by a professional transcription company, and checked for accuracy by 2 graduate-level research assistants. Comments provided via the Zoom chat feature (on average fewer than 6 chat comments per focus group) were added to transcripts by the 2 graduate-level research assistants on the basis of the time stamp provided. We imported transcripts into ATLAS.ti (version 8.4.5, ATLAS.ti GmbH, Berlin, Germany, 2022), in which we conducted all coding and generated code reports to aid in thematic analysis. Data were organized and thematically analyzed by 2 researchers using a phronetic iterative approach, which uses prior literature to shape the framing of emergent themes. Consistent with this approach, the 2 researchers, listened to all recordings, completed critical readings, line-by-line coded a subset of 3 transcripts (1 SNAP administration, 1 SNAP-Ed, and 1 SNAP-E&T), and co-created one 89-item codebook organized into 8 code groups on the basis of the research objectives and literature. The phronetic iterative approach uses inductive and deductive coding methods over 2 cycles (primary and secondary cycles) to organize the data; we used codes that both emerged from critical readings and codes that were informed by the determinants of policy implementation framework by Bullock et al to organize the data. Our primary-cycle codes drew from the 8 policy determinants described by Bullock et al. For example, the code of inconsistent guidance was informed by the determinant of the vertical public administration and thickness of hierarchy, whereas the determinant informed the code of timeliness for timing/sequencing. Through secondary-cycle coding, we collapsed codes into groups to begin to address our research questions. Inductive coding is also a central component of the phronetic iterative process; through inductive coding, we identified equity-focused codes to also include and consider in our analysis despite this not being a component of the framework by Bullock et al.

Three transcripts (2 SNAP administration, 1 SNAP-E&T) were independently coded by the 2 researchers using the codebook. Following the phronetic iterative process, the 2 researchers visually compared agreement in coded segments and discussed the interpretation of codes used. The codebook was revised until consensus was reached across all code groups and codes, resulting in the final codebook. One researcher coded the remaining 4 transcripts (2 SNAP Administration, 1 SNAP-Ed, and 1 SNAP Outreach). Consistent with the phronetic iterative approach, we analyzed code occurrence by research question and used Bullock et al to understand how our findings fit within their theorized determinants of policy implementation and existing literature.

Themes, successes, challenges, and recommendations were summarized in memos by 2 researchers and discussed with the research team (all of whom had co-facilitated focus groups). After the research team reached a consensus on the main findings, to promote trustworthiness in our findings, we invited study participants to engage with and add to the interpreted data via two 60-minute meetings held in July and August 2021. These meetings included 74 state-level SNAP administrators and supervisors working in administration, SNAP-Ed, E&T, and Outreach and represented 15 out of 26 states included in our sample. Administrators and supervisors from states not represented in our sample were also invited to attend debriefings and react to the results presented. The SNAP administrators and supervisors agreed with the themes and recommendations summarized but provided suggestions on word choice and administrative details (eg, funding mechanisms for program operations). These minor revisions were incorporated, and the results were finalized.

RESULTS

State Characteristics

We conducted 7 focus groups with 62 unique participants representing 26 states from all 7 FNS regions (Table 2 ). Half (50%) of states represented had democratic governors, and a half (50%) had republican governors (Table 3 ). State populations ranged from less than 1,000,000 people (15%) to more than 11,000,000 people (15%). The SNAP caseloads across the states ranged from having less than 100,000 people enrolled in SNAP (12%) to more than 1,000,000 enrolled (23%) as of August 2020. Many (46%) states enrolled at least 90% of eligible people in SNAP, but fewer (12%) enrolled less than 75% of those eligible. Although SNAP is federally regulated and state-supervised, it is administered by state or county agencies; most states represented in the sample were state-administered (85%) vs county-administered (15%). Before the pandemic, most states represented (92%) offered an online application (vs paper), and most (85%) offered a combined application for SNAP and at least 1 other program (eg, Medicaid).

Table 2

SNAP Focus Group Participation by Food and Nutrition Service Region and Programmatic Area

Region	SNAP Administration	SNAP-Ed	SNAP-E&T	SNAP Outreach
Mid-Atlantic	1 (2)	0 (0)	2 (3)	0 (0)
Mountain Plains	4 (5)	3 (4)	1 (9)a	1 (1)
Midwest	4 (5)	1 (1)	1 (1)	1 (2)
Northeast	2 (3)	2 (2)	0 (0)	1 (1)
Southeast	3 (3)	1 (1)	0 (0)	0 (0)
Southwest	1 (1)	0 (0)	1 (3)	0 (0)
Western	5 (7)	2 (3)	2 (5)	2 (2)
Totalb	20 (26)	9 (11)	7 (21)	5 (6)

SNAP indicates Supplemental Nutrition Assistance Program; SNAP-Ed, Supplemental Nutrition Assistance Program-Education; SNAP-E&T, Supplemental Nutrition Assistance Program Employment & Training.

Multiple supervisory staff from the same state joined the focus group to listen, but only 2 state-level SNAP leaders were instructed to answer questions on behalf of the team

Unique states across focus groups total n = 26 (some states were represented in multiple focus areas). Unique participants across focus groups total n = 62 (2 participants joined 2 different programmatic areas given their state-level roles and responsibilities).

Note: Presented values are n states (n participants).

Table 3

Characteristics of States (n = 26) Included in 7 Focus Groups About SNAP Implementation During the COVID-19 Pandemic

Characteristic	n (%)
Total	26 (100)
Governor political partya
Democratic	13 (50)
Republican	13 (50)
Populationb
< 1,000,000	4 (15)
1,000,000–3,000,000	5 (19)
3,000,001–5,000,000	3 (12)
5,000,001–7,000,000	6 (23)
7,000,001–11,000,000	4 (15)
> 11,000,0001	4 (15)
SNAP administration
SNAP caseload sizec
< 100,000	3 (12)
100,000–500,000	9 (35)
500,001–1,000,000	8 (31)
> 1,000,000	6 (23)
SNAP coveraged,e
< 75%	3 (12)
75% to 90%	11 (42)
> 90%	12 (46)
County vs state administrationf
County	4 (15)
State	22 (85)
SNAP Application Access
Onlineg	24 (92)
Combinedg,h	22 (85)

COVID-19 indicates coronavirus disease 2019; SNAP, Supplemental Nutrition Assistance Program.

Political party was determined by the party affiliation of the governor as of February 2021

Per 2020 Census data of the US Census Bureau

Per US Department of Agriculture, Food and Nutrition Services, August 2020

Per Center for Budget and Policy Priorities, January 2021

SNAP coverage is defined as the percent of SNAP-eligible individuals enrolled in SNAP

Per National Association of Counties

Per Code for America, 2019

A combined application is defined as the ability of a participant to complete one application for SNAP and at least one other safety net program for eligibility to be determined (eg, Medicaid, Special Supplemental Nutrition Program for Women, Infants, and Children, Temporary Assistance for Needy Families, etc). Combined applications are available online to participants.

Themes

We identified 6 themes from the 7 focus groups, including (1) the policy response, (2) technology needs, (3) collaboration, (4) participant communication, (5) funding realities, and (6) equity. Of the 6 themes, 5 aligned with Bullock et al, and 1 (equity) emerged from inductive coding and analysis. A description of each theme (Table 4 ), the successes and challenges of implementing SNAP during the COVID-19 pandemic (Table 5 ), and recommendations from focus group participants to better prepare SNAP for future emergencies and longer-term (Table 6 ) are described below. Hereafter, participants are referred to as SNAP administrators (participants in the SNAP administration focus groups) and supervisors (participants in the SNAP-Ed, Outreach, and E&T focus groups).

Table 4

Description of Themes From 7 Focus Groups About SNAP Implementation During the COVID-19 Pandemic

Theme	Description
Policy response	Waivers and program flexibilities authorized by Congress and offered by the US Department of Agriculture's Food and Nutrition Service were critical to meet staff and participant needs, but design and implementation requirements provided by FNS impacted the ability to fully optimize implementation
Technology needs	The availability, access, and ability to adapt technology were necessary to support the shift to telework, remote case processing, and delivery of virtual services
Collaboration	Peer-to-peer learning across state agencies, partnerships with professional organizations, and existing close relationships with community organizations helped agency staff interpret guidance, communicate with participants, and leverage funding
Participant communication	States communicated with participants through various channels to quickly relay information about program changes, but frequent policy changes contributed to participant confusion
Funding realities	To meet the administrative realities of SNAP operations, additional funding was needed throughout the pandemic to support administrative costs and develop expanded supportive services offered through SNAP-Ed and SNAP-E&T
Equity	Preexisting concerns about access to adequate SNAP and supportive services among underserved populations were magnified by the COVID-19 pandemic, emphasizing the need to expand eligibility for certain groups and make benefits more accessible to all who are eligible

COVID-19 indicates coronavirus disease 2019; SNAP, Supplemental Nutrition Assistance Program; SNAP-Ed, Supplemental Nutrition Assistance Program-Education; SNAP-E&T, Supplemental Nutrition Assistance Program Employment and Training.

Table 5

SNAP COVID-19 Waiver Implementation Success and Challenges Described by State-Level SNAP Leaders

Theme	Successes	Challenges	Illustrative Quotes
Policy response	Meeting caseload demands	Competing agency priorities	“[The pandemic] severely diminished our workforce capacity while we tried to stand up remote functionality…[We] saw, in April, a 400% application increase, and there's no doubt in my mind we would not have been able to meet the need if we didn't have the ability to extend the certification period.”—SNAP Administration, State 16
	Streamlining operations	Payment accuracy and program integrity concerns	“…there was a strong ‘no’ from upper management and everyone that we would not waive interviews. That it would be too likely for fraud and QC errors they said, ‘No, that's not an option.’ We're not going to do that.” —SNAP Administration, State 4
	Continuing Resolution offering state flexibility	Rigid waiver and policy guidance	“Our tribal partners were hit hard by the pandemic. Good food access became an emergent need. SNAP-Ed played a big hand in being a part of getting food out to their communities. And we felt very restricted by what SNAP-Ed was able to participate in to get food out [...] Fidelity of curricula is not even something we're really thinking about right now. We're just trying to get food access and education out to people in the way that will most reach them.” —SNAP-Ed, State 11
Technology needs	Online application portals	Time required to adapt technology systems	“A lot of our customers are dual recipients so they could be receiving Medicaid or federally or state-funded cash assistance. So it wouldn't make sense for us to push out or remove the recertification requirements for SNAP without doing the same for other programs because otherwise we're not really gaining any advantages in terms of workload and from the customer perspective they still have to do the same. We had to push out those programs as well. It took quite a bit to coordinate that.” —SNAP Administration, State 25
	Modernized eligibility technology systems	Outdated technology systems	“Honestly, it may have had to do with system constraints. We have an older legacy system and we're looking at building or purchasing a new system. And with the waivers and adjustments that we've already taken that may have put an additional strain on our system because we're also doing emergency supplements and P-EBT. And then we were doing D-SNAP, so I'm exhausted.” —SNAP Administration, State 17
Collaboration	Peer learning facilitated by professional organizations	Confusing communication and guidance from federal agencies	“The initial guidance that they [Food and Nutrition Service] sent out at the beginning was confusing at best. And then when you throw P-EBT in there on top of all that. I mean, give me a break. But, yes, the guidance initially was confusing and then when they changed gears in the middle and had us do it a different way was even more confusing” —SNAP Administration, State 19
	Community organizations and institutions reaching participants	Limited technical assistance provided by federal partners	“We have been working with our Department of Higher Education on expanding outreach into colleges and universities […] We have also been working cooperatively on how can we message and educate to colleges and universities the importance of applying for SNAP and how that can support their students. Some colleges and universities have our outreach in their campuses. They have the specific people who do SNAP outreach” —SNAP Outreach, State 8
Participant communication	Online communication platforms	Short duration of waivers creating participant confusion	“We have a Facebook page, [and] it didn't matter what we posted in April. All the comments were, ‘Where's the extra food assistance?’ With FNS sending out the new guidance on April 1, we weren't able to start issuing those emergency allotments until [later in April]. So there was a lot of feedback from our customers. Recently I got a customer feedback that was sent to FNS that said, ‘why did my benefits go down? I didn't get any more income this month, but it went down significantly.’ And that was just because they had not got the emergency allotment yet. These folks believe that this is supposed to be their allotment amount. So the messaging there is rough.” —SNAP Administration, State 25
	Translation of outreach materials		“[The Food Bank] takes all our correspondence, all of our press releases and then adapts them for clients and [translates] them in other languages.” —SNAP Administration, State 10
	Virtual platforms promoting larger geographic reach		“We're definitely going to keep virtual services…some of our rural areas don't have as many providers available to participants. They can receive services from any provider in the state [with virtual providers]. So, if one urban area has a lot of providers... participants in another area can access those services.” —SNAP-E&T, State 26
Funding realities	Leveraging external grants to expand supportive services	Limited reimbursement allowed to meet pandemic realities	“During the pandemic, the E&T final rule was released, which included a couple of paragraphs that seemed to imply that for the supervised job search activity, specifically, you could now use 50/50 funds to purchase technology items for participants to keep. But the language, I think one interpretation of the language is that it was limited just to that single activity or component which would be really weird in SNAP-E&T policy world. So I think we're still awaiting final policy guidance from FNS that says definitively yes, that can only be used to purchase laptops for participants enrolled in that particular component. Or no it can be used for participants in any SNAP-E&T component. So that's kind of what we're waiting on.” —SNAP-E&T, State 1
	Federal emergency funding	Constrained state fiscal realities	“The amount of money that is going to take to update the technology, I don't see it as possible at least within the next 10, 15, 20 years. I see this as generations ahead. But financially as desirable as it would be I don't think it's reasonable. I don't think financially any of us would be able to afford it. [My state] is in a pretty bad way from the pandemic. I mean it was rough. And our budget is—we're bouncing back but without those federal funds, who knows what would be going on right now.” —SNAP Administration, State 22
Equity	Universal benefit increase	Delayed responses to meet needs of underserved populations	“It pains me to think about our poorest households getting no [benefit] supplements whatsoever for almost an entire year. You're not going to get anything positive from me on that portion of the emergency allotment. But I will say I think the idea of emergency allotments was an incredibly beneficial option that we had available to us. I just wish it had been done [the] way that is being done now all along.” —SNAP Administration, State 16
	Suspending work requirements	Reaching populations in the digital divide	“I think there are some issues around this type of population-based policies that we have in place for students, for felons, for people who are paying child support. There are rules that target certain populations that I think are highly problematic in terms of access when you're already a needs-based program [...] And I think there are operational challenges that we need to address around things like language access, physical access, technology and ensuring you have every door open to clients in person, over the phone, and online.” —SNAP Administration, State 23
	Expanding college student eligibility	Lack of transition plans for waivers that expanded access and benefits	“And what an important piece of the puzzle, right? All these folks that have gotten used to this certain benefit and gotten accustomed to all of that and then we take it away […] So I think we all have to—we have to be cognizant about what people have been accustomed to over the past year. And then just to take it away? So a transition plan is definitely something that's needed.” —SNAP Administration, State 13

COVID-19 indicates coronavirus disease 2019; D-SNAP, Disaster-Supplemental Nutrition Assistance Program; FNS, Food Nutrition and Service; P-EBT, Pandemic-Electronic Benefit Transfer; QC, quality control; SNAP, Supplemental Nutrition Assistance Program; SNAP-Ed, Supplemental Nutrition Assistance Program-Education; SNAP-E&T, Supplemental Nutrition Assistance Program Employment and Training.

Table 6

Recommendations for the US Department of Agriculture From State-Level SNAP Administrators and Supervisors to Improve SNAP Implementation

Theme	Recommendations
Policy response	Provide waivers that are longer and can be adapted to state contexts, such as the model authorized by the continuing resolution
	Simplify reporting requirements that are paired with waivers to reduce the administrative burden
	Allow states to continue to waive quality control reviews in the fiscal year immediately following the end of the national emergency to prevent penalties resulting from actions taken during the pandemic response
Technology needs	Prioritize the modernization of eligibility systems
	Provide timely waiver approvals to allow for technology system changes to be made
Collaboration	Reevaluate SNAP-Ed and E&T rules and guidance to allow providers to easily adapt services to best meet the needs of communities and partnering organizations
Participant communication	Provide guidance and best practices to optimize virtual services, participant engagement, and evaluation
Funding realities	Grant flexibility to reimbursable services to promote participant access to needed technology to complete job training and skill development to achieve economic well-being
	Authorize additional federal emergency administrative funding to states as soon as possible within emergency response
Equity	Establish policy mechanisms to quickly expand eligibility to people to promote benefit access and food security among those in need
	Provide a benefit increase to all SNAP households early in an emergency response linked with an economic downturn and provide a transition plan as the economy recovers and stabilizes
	Create infrastructure to quickly deploy outreach support to those challenging to reach (eg, those in the digital divide)

SNAP indicates Supplemental Nutrition Assistance Program; SNAP-Ed, Supplemental Nutrition Assistance Program-Education; SNAP-E&T, Supplemental Nutrition Assistance Program Employment and Training.

Policy response

The SNAP administrators largely agreed that waivers and flexibilities offered by FNS helped with caseload management, but rigid guidance and lack of state authority created implementation challenges. Many SNAP administrators described unprecedented application increases at the start of the pandemic, making waivers designed to streamline application, recertification, and verification processes crucial. Specifically, many SNAP administrators agreed that the waiver of face-to-face interviews, the waiver of interviews at the time of initial application and recertification, the extension of certification periods and periodic reports, and the waiver of audio recordings for telephonic signatures all helped to streamline operations (Table 1). Interview waivers allowed SNAP administrators to streamline the application and recertification procedures and process larger case volumes when demand was high while reducing administrative burdens for participants. The extension of certification periods and periodic reports allowed SNAP administrators to continue to provide benefits to existing participants without recertifying their eligibility, ensuring continued benefit access for households and freeing up administrative resources to process new applications. Telephonic signatures without the required audio recording were described as a low-tech option for providing remote verification services to applicants.

Although these waivers were beneficial, they were not universally used by all SNAP administrators. Some SNAP administrators described their operations as less impacted by the pandemic (eg, the technology was available to more easily transition to telework or SNAP caseloads did not increase dramatically), so changes to case processing were not needed. In other cases, competing priorities made it challenging to implement SNAP waivers effectively. For example, several SNAP administrators described experiencing natural disasters during the pandemic, requiring activation of the Disaster-Supplemental Nutrition Assistance Program (D-SNAP)—a food benefit provided to low-income households during natural disasters—and every state implemented Pandemic-Electronic Benefit Transfer (P-EBT)—a new program administered jointly by SNAP, child nutrition agencies, and schools to provide EBT benefits in lieu of in-person school meals. Some SNAP administrators found it difficult or impossible to implement waivers, which required additional reporting and information technology adjustments, without additional staff and IT resources.

Rigid policies and guidance provided by USDA created challenges for SNAP administrators, SNAP-Ed supervisors, and E&T supervisors. State agencies can face penalties if payment error rates exceed certain levels set by the USDA. Many SNAP administrators reported concerns that the combined effects of staff teleworking, management of increased caseloads, and use of waiver flexibilities would result in financial liabilities for their agencies. Congress's decision to temporarily waive quality control review requirements (ie, reporting requirements to generate payment error rates) for 1 year was described as helping SNAP administrators focus on adapting SNAP operations during the pandemic without fear of consequences. Program integrity remained a priority, with some SNAP administrators explaining that they maintained quality control reviews to ensure proper program oversight and preserve staff skills. Several SNAP administrators described challenges in implementing the early waivers to extend certification periods because of how the waivers were designed. To use the waiver, state agencies were required to extend all recertifications by 6 months, which was described as creating twice the amount of work when extended cases became due. In addition, in states that aligned SNAP certification periods with jointly administered programs (eg, Medicaid), extending certification periods for 1 program but not the others reduced efficiency for program staff and participants. The continuing resolution passed in October 2020 offered flexibility in how states applied this waiver to their caseloads and allowed waiver uptake without USDA Secretary approval; SNAP administrators described this legislative action as resolving many of the challenges of early waivers.

Finally, rigid policies guiding SNAP-Ed and E&T before the pandemic created challenges. Because states are only reimbursed for approved activities, SNAP-Ed and E&T supervisors described being limited in their abilities to adapt their services to the changing needs of participants. For example, in several states, participation in nutrition education declined during the pandemic. SNAP-Ed supervisors described wanting to redeploy resources to assist community partners with food distribution, but this activity was not allowable under USDA guidelines.

Technology needs

The SNAP administrators and SNAP-Ed, E&T, and Outreach supervisors all described challenges in adapting their in-person programming to a virtual setting; however, those with more modernized infrastructure fared better than those with older systems. Before the pandemic, investments in online customer portals were described as making it easier for state agencies to transition to remote procedures. In some cases, SNAP administrators with older eligibility systems (also called legacy systems) described being unable to use waivers because of the time and information technology resources required to make modifications. This was described as especially problematic when waivers were approved by FNS close in time to the expected implementation date. At times, the combination of outdated eligibility systems and late FNS approvals prevented SNAP administrators from taking advantage of waivers that would have benefited participants and staff.

Collaboration

The SNAP administrators and SNAP-Ed, E&T, and Outreach supervisors all described leveraging partnerships to overcome challenges created by federal policy and guidance. Some SNAP-Ed and E&T supervisors turned to partner organizations (eg, food banks or workforce development organizations) to leverage external grants to develop and expand supportive services. For example, partnerships allowed food pantries to supply ingredients (not reimbursable through SNAP) to participants for virtual cooking classes held by SNAP-Ed educators. In many states, community job training organizations purchased laptops for E&T participants to access virtual job training resources. Many SNAP administrators shared that they participated in regional and national workgroups that provided assistance interpreting FNS guidance or drafting waiver requests.

Participant communication

State agencies and their partners (eg, food banks) were required to communicate frequent policy and programmatic changes to SNAP participants. Many SNAP administrators described this as particularly difficult when FNS approved waivers for only a short period. For example, 1 SNAP administrator described how monthly approvals for emergency allotments led to confusion among SNAP participants despite frequent communications from the state. This was described as often leaving participants uncertain about their benefit amount in a given month and when supplemental benefits would be received. SNAP administrators and SNAP-Ed, E&T, and Outreach supervisors all described relying on state websites to communicate with participants, whereas some also used new communication platforms, such as social media. Some SNAP administrators kept essential staff onsite to serve participants without internet access. All described communication with participants through community partners, such as food pantries and higher education systems, as helpful. Furthermore, SNAP-Ed and E&T supervisors described some benefits of new virtual platforms (eg, Zoom) for facilitating collaboration among staff across the state and engaging participants in geographically disparate areas with poor access to in-person services.

Funding realities

The SNAP administrators described needing additional funding to support the transition to telework, adjust service delivery, and implement waivers. The unanticipated implementation of P-EBT placed a significant financial burden on agencies. SNAP administrators said they continued to fund operations through the 50% uncapped federal formal match, but constrained state fiscal realities made it challenging to leverage existing state funds to meet new program needs. The Consolidated Appropriations Act in December 2020 and American Rescue Plan in March 2021 provided additional financial resources, but many SNAP administrators shared that this relief came too late.

Supplemental Nutrition Assistance Program-Education and E&T supervisors turned to partner organizations to leverage external grants to develop and expand supportive services that could not be reimbursed because of prepandemic policies. For example, at the start of the pandemic, E&T supervisors could not use the reimbursement to purchase laptops for participants; grants from community organizations were described as helping to meet this need. The FNS later issued a new E&T rule to expand reimbursable services to include laptops for certain tasks (ie, job searches).

Equity

All SNAP administrators agreed that benefit access and adequacy improved because of waivers issued by FNS and Congressional actions, but a more organized response was needed sooner. In particular, many SNAP administrators shared that they requested to expand college student eligibility earlier in the pandemic but were denied by FNS. Benefit increases provided through emergency allotments were described by nearly all SNAP administrators as confusing and inequitable because they excluded the lowest income households until April 2021; the universal benefit increase of 15% authorized from January—September 2021 was described as a clearer and more equitable approach.

Many SNAP administrators and SNAP-Ed, Outreach, and E&T supervisors described the need to consider underserved populations beyond the COVID-19 pandemic. For example, addressing the digital divide (ie, those without internet access and/or digital literacy) was important in all focus groups. This was essential to ensure all participants, regardless of technology access and literacy, had access to SNAP and supportive services. Furthermore, many SNAP administrators shared that a transition plan (eg, slowly decreasing benefits received through emergency allotments) was needed to continue to support participants throughout pandemic recovery. Beyond emergency operations, SNAP administrators and SNAP-Ed, E&T, and Outreach supervisors all described a desire to modify existing procedures and generate new best practices. Some SNAP administrators wished for more flexibility to waive interview requirements for certain households permanently. Many SNAP administrators shared that they desired to better align SNAP with other human services programs to promote greater benefit coverage among eligible households. Some SNAP-Ed, E&T, and Outreach supervisors wanted to better align SNAP supportive services with each other to improve participants’ experiences and engagement.

Recommendations

Determinants of implementation are modifiable, and administrators and supervisors provide recommendations to improve future SNAP operations (Table 6). The SNAP administrators recommended that FNS design waivers adaptable to individual state contexts, reduce administrative burdens of waivers, and provide additional technical assistance on waiver implementation. The SNAP administrators also recommended prioritizing technology system upgrades and establishing mechanisms to quickly expand benefit access and streamline operations in times of national emergency. The SNAP-Ed and E&T supervisors recommended granting flexibility to expand the scope of work and reimbursable supportive services to best serve both SNAP participants and partnering organizations.

DISCUSSION

We aimed to describe how state agencies implemented SNAP and its supportive services in the first year of the COVID-19 pandemic, the determinants of implementation (ie, barriers and facilitators), and recommendations from state-level administrators about how to better prepare SNAP for future emergencies and longer-term. Our study was informed by the determinants of policy implementation framework by Bullock et al, which describes implementation factors that influence policy outcomes and the relationships between them. The SNAP administrators agreed that SNAP waivers and flexibilities authorized in response to the COVID-19 pandemic enabled them to manage unprecedented caseload increases and expand access to SNAP. The ability to extend certification periods, waive interviews, and accept telephonic signatures without audio recording was described as crucial. Modernized technology systems, virtual programming, partnerships, and expanded access to benefits were identified as facilitators of implementation. Administrators and supportive service supervisors identified implementation challenges that stemmed from the initial design of waivers, federal guidance, outdated technology, and existing regulations that limited the authority given to states. In future emergencies, SNAP administrators and supportive service supervisors recommended that waivers be adaptable to variations in program administration at the state and local levels. Longer-term investments in technology were recommended to streamline verification and recertification processes, benefit delivery, communications, and access to supportive services. These strategies to modernize the delivery of SNAP were identified as ways to promote equity among SNAP participants—a theme that was not informed by Bullock et al, but emerged through inductive coding and our analysis.

The SNAP is just 1 part of a larger federal nutrition safety net that includes the National School Lunch Program (NSLP), the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), and, most recently, P-EBT. These programs experienced many similar challenges implementing program waivers and flexibilities during the COVID-19 pandemic that can be connected to Bullock et al, including challenges related to the external environment and policy context (eg, supply chain disruptions and technology limitations) and implementing agency responses (eg, shift to remote services and redeployment of financial or staff resources). , A recent study of the NSLP found that school foodservice providers had difficulty switching from in-person meal service to meals-to-go because of insufficient operating budgets, reduced revenues, and inadequate staff capacity. , A study of WIC found that participation declined 9% in the first 9 months of the pandemic in states that required WIC participants to reload benefits in-person instead of remotely. , Limited technology was also identified as a barrier to implementing P-EBT, resulting in delayed issuance of benefits to many households in need. , By contrast, waivers that eased administrative burdens because of their features (eg, promoting remote services) were generally well-received by food assistance administrators. , For example, the Los Angeles County WIC Program saw a significant increase in participation and a reduction in gaps in participation by race and ethnicity during the pandemic. This change was attributed, in part, to the transition to remote services, which was perceived as making program participation easier for families. ,

In this study, SNAP administrators and supportive service supervisors recommended several solutions to address technology gaps, identified as a major external barrier to successful implementation outcomes. Many SNAP administrators recommended investing in modernized, integrated eligibility systems, which can increase efficiency for agency staff, respond more quickly to program modifications, and simplify participant reporting requirements. , The SNAP-Ed supervisors and E&T supervisors were interested in additional technical assistance from the USDA to optimize virtual services and better reach populations with poor access to in-person programming. Past research suggests virtual programming is particularly useful in rural areas, in which people often reside far from local offices and lack adequate staffing. Although evidence overall is limited, early research suggests that ensuring access to technology among participants through partnerships, adapting curricula to engage in the virtual environment, and providing adequate training to all providers promotes successful delivery and engagement with virtual programming.

This study has strengths and limitations to consider. First, this work represents the perspectives of state-level SNAP administrators and supportive service supervisors. Although this is an important group for understanding SNAP operations and policy implementation, other perspectives were not captured, including those of federal policymakers and SNAP participants. Second, although we recruited state-level SNAP administrators and supervisors from states with varying characteristics (eg, number of waivers used, population size, technology systems available, etc), not all 50 states were represented in our sample. We did confirm findings with state-level SNAP administrators and supervisors before finalizing to ensure an accurate representation of their experiences. Third, we conducted 1 focus group each with state-level supervisors from SNAP-Ed, Outreach, and E&T. Therefore, we may not have reached data saturation within these programmatic areas. Fourth, the focus groups did not ask about all SNAP adaptations made during the pandemic. For example, the moderator guide did not include quality control review waivers and P-EBT as distinct questions. However, they did emerge through focus groups and were probed about when appropriate. Finally, although SNAP policy experts reviewed our guide for relevancy and accuracy, we did not pilot the guide with informants outside of the research team before data collection.

IMPLICATIONS FOR RESEARCH AND PRACTICE

Findings from this study support Congressional and USDA policy actions to prepare for future national public health emergencies. First, codifying waivers for future emergencies activated through automatic triggers (eg, a national public health emergency) could provide states the advanced planning and clarity needed to rapidly adapt services to meet the needs of SNAP participants during crises. Streamlining application processes and providing universal benefits would promote equitable access and adequacy in emergencies. The Disaster-Supplemental Nutrition Assistance Program is an example of an established emergency food assistance program with streamlined processes; this program can be considered a precedent to establish a mechanism to quickly deliver SNAP in public health emergencies beyond disasters. Increased administrative costs were common challenges faced by SNAP administrators and implementers of other nutrition assistance programs, such as NSLP. Congress could authorize supplemental emergency administrative funding to meet increased staffing and technology demands in future emergencies.

Findings from this study also support policy actions to streamline and modernize SNAP implementation beyond emergency settings. Congressional appropriations for USDA to allocate investments in technology and virtual services could increase participation in SNAP by removing administrative barriers and expanding access to supportive services, particularly for individuals with poor access to in-person programming. Increasing reimbursement rates for technology investments could expedite the modernization and integration of eligibility systems, creating administrative efficiency for SNAP and other human services programs.

Authorizing demonstration projects to test and evaluate streamlined procedures (eg, eliminating interviews for certain households and simplifying recertification processes) would generate evidence to understand the impact on agency staff, participants, and operations, including payment accuracy, retention of benefits, and equitable access to benefits across different populations. Research and technical assistance in providing SNAP-E&T and SNAP-Ed programming are needed to assess the effectiveness of virtual services in different contexts (eg, rural vs urban contexts, among populations with differing levels of digital literacy, etc). The 2023 Farm Bill offers an opportunity to build evidence through demonstration projects, fund innovative approaches to improve SNAP benefit access and participant experience, and prioritize rigorous evaluation of these efforts.