The effect of school meals on the well-being of multigenerational households receiving SNAP benefits

Year: 2019

Research Center: Tufts University/University of Connecticut (UConn) RIDGE Program

Investigator: Laurito, Agustina

Institution: University of Illinois at Chicago

Project Contact:
Agustina Laurito
University of Illinois at Chicago
Department of Public Administration
412 S. Peoria St., 136
Chicago, IL 60607
Phone: 312-996-3179


This project investigates the effect of school meals on the food security of multigenerational households and the food acquisitions and health of individual members of those households. Since the Great Recession, the number of multigenerational households (households with three generations) or skipped generation households (grandparents and grandchildren only) has increased substantially. At the same time, food insecurity has remained relatively high among seniors (aged 60 years old or more) and older adults (aged between 50 and 59 years old), and it is higher among those in households with grandchildren present. Multigenerational households often experience more food insecurity and other hardships, yet there is limited research that examines how the different food assistance programs work together to ameliorate food insecurity and diet among these households, and the well-being of all members of multigenerational households.

This is an important question because multigenerational households are complex and diverse. Some of these household include senior members (aged 60 years old or more), while others are headed by grandparents younger than 60 years old. The role of SNAP and school meals may differ between these households either because of the age of the children in those households, because of access to different resources relative to non-multigenerational households with children, or because of SNAP eligibility and benefit rules that differ between households with a senior and non-senior member. This paper seeks to address this gap in the literature by focusing on the effect of school meals on the food insecurity and health of multigenerational households and individual members of those households, including seniors. In doing so, this project wants to highlight the importance of taking into account the complexity of household composition when estimating the effects of social policies. Specifically, the project answers these key research questions:

  1. What is the effect of school meals on 30-day adult household food insecurity in multigenerational households?
    1. Does the effect of school meals differ between households with a senior member and those headed by non-senior grandparents?
  2. What is the effect of school meals on individual meal acquisitions and self-reported health of multigenerational household members?
  3. Does the use of informal food sources and discounts differ by household type when school is in session and not?
This project relies on restricted data from the National Food Acquisition and Purchase Survey (FoodAPS). FoodAPS is a nationally representative survey that tracks food acquisitions and purchases for all households and individual household members, including children, for 7 days. FoodAPS includes a large sample of 1,581 SNAP households whose eligibility was confirmed through self-reports and administrative data matching. This study’s sample includes 131 multigenerational households that receive SNAP benefits and 740 individuals in these households. The empirical strategy exploits the categorical eligibility of children in SNAP households for up to 2 school meals a day, and plausibly exogenous variation in the timing of FoodAPS data collection relative to when school is in session. The identifying assumption is that the timing of data collection should be uncorrelated with household characteristics. Thus, households interviewed when school was in session or on break should be similar except for their access to school meals. Results from this study should yield credible causal estimates of the effect of school meals on the food insecurity and diet of multigenerational households receiving SNAP benefits for all, and by household type. Individual-level analyses estimate of the role of school meals in improving self-reported health and their effect on individual daily and weekly meal acquisitions. Daily analyses are restricted to weekdays and include a sample of 2,907 individual-day observations.

Results show that access to school meals lowers 30-day adult food insecurity among multigenerational households by 16 percentage points. School meals also lower the probability that households report running out of food and having no money to buy more. This project also finds positive effects of school meals on household’s healthy eating index calculated using food at home acquisitions. Correlational analyses of the number of school lunches acquired show a positive association between additional lunches per child and lower food insecurity. Each additional meal per child correlates with a decline in food insecurity of roughly 5 percentage points. This finding suggests that increasing school meal participation might result in substantial declines in food insecurity.

This study also finds improvements in self-reported health among adults aged 29 to 59 years old and seniors. Meal acquisition analyses show children (aged 5 to 14 years old) and seniors are less likely to miss any breakfast, lunch, or dinner on a given day when school is in session, while suggestive evidence shows adults are less likely to miss any of these meals in a given week.

While school meals play an important role for both households with seniors and those with non-senior grandparents, this last set of households is less likely to use informal food sources (food banks, churches, family and friends) and coupons for food at home purchases when school is in session. This is not the case for households with seniors.

Results from this paper are important for policy in a various ways. First, the results provide further evidence of the spillover of school meals on adult members of households, including those with seniors and older adults showing their potential for improving senior food insecurity and well- being. Second, the results highlight the importance of taking into account family complexity in estimating the effects and interactions of different social policies. Potential policy implications include expanding access to school meals during summer months and expanding policies that increase school meal participation such as Community Eligibility Provisions.