AMES, Iowa - Forty million Americans, including 6.5 million children, are food insecure, according to the U.S. Department of Agriculture, which means they do not have enough food for an active, healthy life.
Many rely on the Supplemental Nutrition Assistance Program (SNAP) - the largest food assistance program for low-income families - to help make ends meet. Still, 51.2 percent of households receiving SNAP benefits, commonly known as food stamps, were food insecure in 2016.
Given the extent of food insecurity, a team of Iowa State University economists developed a methodology to analyze potential redundancies between SNAP and the Special Supplemental Nutrition Program for Women, Infants and Children (WIC), the third-largest food assistance program in the U.S. Their research, published in the Southern Economic Journal, provides evidence that the programs are in fact complementary, not redundant. They found that participating in both SNAP and WIC compared to SNAP alone increases food security by at least 2 percentage points and potentially as much as 24 percentage points.
"Our findings can help policymakers design more efficient programs to meet food needs," said Helen Jensen, ISU professor emeritus of economics. "We know low-income families often participate in more than one food assistance program, and we find the combination of SNAP and WIC helps reduce food insecurity for participating households."
Challenges of measuring program effects
The programs are similar, but serve different needs. WIC covers specific foods to meet the nutritional needs of pregnant women and new mothers as well as infants and young children. Participants also receive nutrition counseling and referrals for health services, such as prenatal programs. In comparison, eligible households can use SNAP benefits to buy most food items. All households included in the study were potentially eligible for both programs, but they chose whether or not to participate.
This "self-selection" is one reason it is difficult for researchers to ascertain whether a program causes a change in food insecurity. WIC and SNAP benefits are not randomly assigned, so any differences in food security outcomes between participants and nonparticipants could be due to actual causal impacts of the programs or unobserved differences between households that apply for benefits and those that do not.
If households at greatest risk of becoming food insecure are most likely to apply - for example, in the case of a job loss - it might falsely appear the programs are ineffective in alleviating food insecurity, the researchers said. In fact, while participants may be less food secure than eligible nonparticipants, participants may still be more food secure than they would have been in a world without the programs.
Another challenge for researchers is that households are known to systematically underreport benefits, often because they don't want to admit they are receiving government assistance.
"For these reasons, traditional econometric methods lead to misleading estimates," said Oleksandr Zhylyevskyy, associate professor of economics. "With that in mind, we developed a methodology that allows us to more accurately measure the true effects of WIC and SNAP."
The researchers applied their methodology to data from the USDA's National Household Food Acquisition and Purchase Survey or FoodAPS, which provides self-reported household participation in SNAP and WIC and validated data for SNAP participation. The study included 460 households that were income-eligible for both programs. They were surveyed for one week.
On average, these households were families of four with two children, one under the age of 6. The average monthly income was about $1,600. More than 75 percent rented a home or apartment, 26 percent did not own or lease a vehicle and 11 percent had used a food pantry within the past 30 days.
FoodAPS matched survey responses about SNAP participation with official administrative records to identify response errors, but no similar verification was available for WIC. The ISU researchers say the new methodology was specifically designed to handle this type of scenario in which researchers can corroborate answers for some survey questions, but not others.
"Our goal was to strike a balance between making assumptions that are weak enough to be credible, but strong enough to be informative," said Brent Kreider, professor of economics. "Policymakers may ask whether these programs actually work or merely increase government spending without reducing food insecurity. We find WIC helps even when SNAP is already in place."
The study was funded by the National Bureau of Economic Research.
Brent Kreider, email@example.com, 515-294-6237
Oleksandr Zhylyevskyy, firstname.lastname@example.org, 515-294-6311