A Consideration of the Evaluation of Demonstration Projects to End Childhood Hunger (EDECH)

Abstract

Food insecurity, which occurs when food intake is reduced due to limited resources, has become a leading indicator of well-being in the United States for 2 central reasons. First, more than 37 million Americans lived in food insecure households in 2018.1Coleman-Jensen A. Rabbitt M. Gregory C. Singh A. Household Food Security in the United States in 2018. ERR-270. US Department of Agriculture, Economic Research Service, Washington, DC2019Google Scholar Of these, almost 13 million are in the more severe category of very low food security. Second, there is a well-established set of negative health outcomes associated with food insecurity.2Gundersen C. Ziliak J. Food insecurity and health outcomes.Health Aff. 2015; 34: 1830-1839Crossref PubMed Scopus (551) Google Scholar These lead to higher health care costs3Berkowitz S. Basu S. Meigs J. Seligman H. Food insecurity and health care expenditures in the United States, 2011-2013.Health Serv Res. 2017; 53: 1600-1620Crossref PubMed Scopus (96) Google Scholar and higher rates of mortality.4Gundersen C. Tarasuk V. Cheng J. de Oliveira C. Kurdyak P. Food insecurity status and mortality among adults in Ontario, Canada.PLoS One. 2018; 13e0202642Crossref PubMed Scopus (42) Google Scholar The magnitude and subsequent consequences of food insecurity would be higher in the absence of the Supplemental Nutrition Assistance Program (SNAP) (https://www.fns.usda.gov/snap/supplemental-nutrition-assistance-program). Its success in reducing food insecurity has been demonstrated in numerous studies during the past 2 decades.5Gregory C. Smith T. Salience, food security and SNAP receipt.J Policy Anal Manag. 2019; 38: 124-154Crossref PubMed Scopus (24) Google Scholar, 6Gundersen C. Kreider B. Pepper J. Partial identification methods for evaluating food assistance programs: A case study of the causal impact of SNAP on food insecurity.Am J Agric Econ. 2017; 99: 875-893Crossref Scopus (62) Google Scholar, 7Mabli J. Ohls J. Supplemental Nutrition Assistance Program Participation is associated with an increase in household food security in a national evaluation.J Nutr. 2015; 145: 344-351Crossref PubMed Scopus (60) Google Scholar, 8Swann C. Household history, SNAP participation, and food insecurity.Food Policy. 2017; 73: 1-9Crossref Scopus (32) Google Scholar Despite the success of SNAP, a high proportion of SNAP recipients are food insecure. For example, 50.1% of SNAP recipients were food insecure in 2017 vs 23.4% of eligible nonrecipients.1Coleman-Jensen A. Rabbitt M. Gregory C. Singh A. Household Food Security in the United States in 2018. ERR-270. US Department of Agriculture, Economic Research Service, Washington, DC2019Google Scholar One approach to reducing these food insecurity rates among SNAP recipients is to increase benefit levels. Ziliak9Ziliak J. Modernizing SNAP Benefits. Policy Proposal 2016-06, The Hamilton Project. Brookings Institution, Washington DC2016Google Scholar presented arguments in favor of an across-the-board increase in SNAP benefit levels by indexing benefits to the US Department of Agriculture’s Low-Cost Food Plan rather than the Thrifty Food Plan (https://fns-prod.azureedge.net/sites/default/files/media/file/CostofFoodApr2020.pdf). Another method would be to identify the dollars needed by SNAP recipients to be food secure and then increase SNAP benefits to bridge this gap.10Gundersen C. Kreider B. Pepper J. Reconstructing the Supplemental Nutrition Assistance Program to more effectively alleviate food insecurity in the United States.RSF J Soc Sci. 2018; 4: 113-130Google Scholar Two recent demonstration projects commissioned by the Food and Nutrition Service (FNS) of the US Department of Agriculture considered the impact of different approaches to increasing SNAP benefits on food insecurity. In the first of the 2 interventions examined, the Kentucky Ticket to Healthy Food (TTHF) project,11Chojnacki G.J. Gothro A.G. Gleason P.M. Forrestal S.G. A randomized controlled trial measuring effects of extra Supplemental Nutrition Assistance Program (SNAP) benefits on child food security in low-income families in rural Kentucky.J Acad Nutr Diet. 2021; 121: S9-S21Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar the amount of SNAP benefits was increased by providing additional SNAP benefits to cover transportation costs to reach retail food outlets and increasing the earnings deduction when calculating net income and, hence, benefit levels. The increase in SNAP benefits to reflect transportation costs implicitly recognizes food acquisition cost as a determinant of food insecurity,12Courtemanche C. Carden A. Zhou X. Ndirangu M. Do Walmart supercenters improve food security?.Appl Econ Perspect Policy. 2019; 41: 177-198Crossref Scopus (14) Google Scholar,13Gregory C. Coleman-Jensen A. Do high food prices increase food insecurity in the United States?.Appl Econ Perspect Policy. 2013; 35: 679-707Crossref Scopus (83) Google Scholar insofar as it considers transportation costs as adding to the total cost of food. The second demonstration examined was the Nevada Healthy, Hunger Free Kids (HHFK) project.14Gleason P. Kleinman R. Chojnacki G. Briefel R. Forrestal S. Measuring the effects of a demonstration to reduce childhood food insecurity: A randomized controlled trial of the Nevada Healthy, Hunger Free Kids project.J Acad Nutr Diet. 2021; 121: S22-S33Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar This directed increased benefits to households with children under the age of 5 years with incomes below 75% of the poverty level. The empirical approach used to evaluate both projects was excellent and appropriate to answer the questions posed by FNS. A key finding in both projects is that the expansion of benefits did not lead to statistically significant declines in food insecurity. In what follows, I will show why this is not surprising, given what we know about the level of benefit needed to reduce food insecurity. I then turn to how moving beyond a binary measure of food insecurity might have shown that these increases in SNAP benefits did lead to improvements in well-being. I conclude with some insights that one can draw by contrasting the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) with SNAP in the HHFK report. One way to understand the levels of need of food insecure households is to ask them about the “dollars needed to be food secure.”15Gundersen C. Ribar D. Food insecurity and insufficiency at low levels of food expenditures.Rev Inc Wealth. 2011; 57: 704-726Crossref Scopus (26) Google Scholar In the Current Population Survey (https://www.census.gov/programs-surveys/cps.html), after the food expenditure questions but before the 18-item scale in the Core Food Security Module, households are asked “In order to buy just enough food to meet (your needs/the needs of your household), would you need to spend more than you do now, or could you spend less?” Households responding that they need more money for food were asked the following question: “About how much more would you need to spend each week to buy just enough food to meet the needs of your household?” This is the “resource gap.” In 2017, the average resource gap, per week, for food insecure households was $34.43 and for food insecure SNAP households it was $43.43. (Author’s calculation based on data from the December supplement of the 2017 Current Population Survey.) When the estimated resource gap is multiplied by 4, the result is approximate monthly resource gaps of $137.72 and $173.72, respectively. In contrast, the average monthly increases in SNAP benefit levels for TTHF was $22 and for HHFK was $44. In both cases, this is far below the amount needed for the average household to escape food insecurity. While the increase in SNAP benefits resulting from the demonstration did not raise enough of these households above the food security threshold to register an impact, there may still be positive impacts on food insecurity. Consider the limitations of a prevalence measure of food insecurity, that is, the standard measure of food insecurity whereby a household is either food secure or food insecure. This approach treats all food-insecure households identically. For example, if one believes that responding affirmatively to, for example, 15 questions means a household has a higher degree of food insecurity than one responding to, for example, 6 questions, the food insecurity rate does not differentiate between these 2 cases. This has implications for measures at the population level, but it also has implications for evaluation of increases in benefits akin to those found in these 2 projects. Suppose that these 2 households now respond affirmatively to 13 and 4 questions, respectively. Unambiguously, both of these households are better off due to the receipt of increased SNAP benefits. While further breakdowns of food security status into categories such as low food security and very low food security would help move toward a more nuanced perspective on improvements in food security status, in the example above, these 2 households do not escape either very low food security or low food security, respectively. An alternative approach would be to measure the depth of food insecurity.16Balistreri K. A decade of change: Measuring the extent, depth and severity of food insecurity.J Fam Econ Issues. 2016; 37: 373-382Crossref PubMed Scopus (15) Google Scholar,17Gundersen C. Measuring the extent, depth, and severity of food insecurity: An application to American Indians in the United States.J Popul Econ. 2008; 21: 191-215Crossref Scopus (76) Google Scholar Under this measure, both the proportion of households that are food insecure and the number of affirmative responses to the Core Food Security Module are incorporated. So, if the depth of food insecurity was used instead of the standard binary measure, this would allow for the assessment of potential improvements in food insecurity status due to higher benefit levels. Future evaluations done by FNS and other organizations, therefore, may wish to consider the depth along with the prevalence of food insecurity. Unlike the TTHF project, the HHFK project also included steps to increase participation in other assistance programs, especially the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC; https://www.fns.usda.gov/wic). WIC has nearly universal participation among eligible families with infants in the United States. The proportion of households participating in WIC falls substantially as children age, with participation rates among eligible 4-year-olds at about 16%.18Thorn B. Tadler C. Huret N. et al.WIC Participant and Program Characteristics 2014. US Department of Agriculture, Food and Nutrition Service, Alexandria, VA2015Google Scholar In contrast, SNAP participation rates among households with children are, depending on the metric used, between 75% and 95%. The HHFK project further illustrates this divide between SNAP and WIC. All of the households in this study are eligible for WIC (and SNAP) and are those that FNS is most interested in enrolling in WIC. The participation rates are similar to other WIC-eligible populations—44% in the treatment group and 40% in the control group. Moreover, despite efforts to increase participation in WIC through this project, the number of new enrollees in WIC was small. If FNS does want to have increased participation rates among children in WIC, alongside ongoing efforts, they may want to investigate changes to the program. These changes could include efforts to make participation in the program easier by not requiring multiple visits to WIC offices to maintain benefits. In my opinion, the appeal of WIC could also be enhanced by not limiting client choices to products that are scarce on store shelves (eg, nonstandard sizes of bread loaves) and by expanding the choice of healthful foods to more accurately reflect the consumption patterns of low-income children (eg, by allowing the purchase of white rice along with brown rice). SNAP is the most important component of the social safety net against hunger in the United States. Future researchers may wish to examine the relative impacts of larger increases in SNAP benefit levels, to use more sensitive measures of food insecurity, and to consider the reasons why the participation rates in SNAP are so much higher than in WIC. C. Gundersen is Agricultural, Consumer, and Enviornmental Sciences (ACES) distinguished professor, Department of Agricultural and Consumer Economics, University of Illinois, Urbana.