This page is optimized for a taller screen. Please rotate your device or increase the size of your browser window.

Want Healthier Eating? Evidence Shows Limiting Choice Works

May 31, 2017

grocery storeThe federal government runs two complementary - but different - programs to help poorer Americans feed their families. SNAP, the Supplemental Nutrition Assistance Program (formerly called the Food Stamp Program), is open to poor and near-poor families and allows them to use the benefits to purchase nearly any foods (prominent exceptions include alcohol, cigarettes, and prepared foods). WIC, the Special Supplemental Feeding Program for Women, Infants, and Children, is only open to households with a pregnant woman or a young child (under five years old) and the benefit can only be used to purchase specific quantities of specific “good” foods (e.g., skim milk, whole grains, fruits and vegetables).

Drawing on her personal experience in a piece for Slate[1], Heather Gilligan called WIC “terrible” and railed against the limitations on foods that can be purchased: [I]f the federal government really wants to help parents, it should trust them to make their own purchasing decisions instead of forcing them to spend hours navigating the intricacies of WIC, as it currently exists. Luckily, we already have a program in place that respects the dignity and autonomy of its recipients. It’s called SNAP and it works … Food stamps should be the model for WIC, not the other way around. Gilligan supports these personal experiences by noting the lack of strong evidence that WIC improves nutrition[2]—and thus, a reason to terminate the program.

Rigorous Evidence from 16 U.S. Communities

Research by my colleagues at Abt Associates and Mathematica Policy Research and myself[3] suggests that the policy choices are much more subtle. With funding from Congress and under contract from the U.S. Department of Agriculture’s Food and Nutrition Service, we implemented an evaluation of the Summer Electronic Benefits Transfer for Children Demonstration (SEBTC), among the largest randomized trials of an American nutrition program ever conducted−nearly 100,000 households. In demonstration sites, SEBTC provided poor and near-poor households with an electronic benefit transfer card (like a debit card) worth $60 per school-age child, per month, during the summer. This was approximately the value of the food provided by school meals—which are not available during the summer when school is not in session.

Crucially for our purposes, sites were allowed to choose between operating the program according to SNAP rules (i.e., households could purchase almost any food) or according to WIC rules (i.e., households could only purchase specified quantities of specified good foods). Within each site, households were randomly assigned to either the $60 per child benefit or no benefit. Toward the end of the summer, we surveyed both households that were and were not offered the benefit. The survey specifically asked about children’s intake of specified foods.

Research Shows Benefits of Restrictions

The results are simple and dramatic. Giving households more food assistance according to SNAP rules—i.e., allowing them to purchase almost any food they choose—improves dietary intake a little.

  • Fruit and vegetable intake increases by 9 percent;
  • Whole grain intake increases by 12 percent;
  • Dairy intake increases by 5 percent;
  • There is no impact—positive or negative—on intake of sugar, sugar sweetened beverages, or sugar from cereals.

In contrast, giving households the same amount of assistance according to WIC rules—i.e., requiring them to purchase specified quantities of specified good foods—improves dietary intake a lot.

  • Fruit and vegetable intake increases by 19 percent;
  • Whole grain intake increases by 53 percent;
  • Dairy intake increases by 16 percent.

These increases were two, four, and three times the increase from simply giving more food assistance (i.e., SNAP rules).

The takeaway: Restrictions matter—a lot!

Finally, with WIC rules, intake of sugar, sugar sweetened beverages, or sugar from cereals decreases slightly. Overall sugar intake declines by 5 percent and intake of sugar sweetened beverages declines by 14 percent! With SNAP rules, there is no change in intake of these bad foods.

Policy Decisions Limiting Choice – There’s A Trade-Off

These results confirm what should be common sense: If you give a household food assistance which can only be used to purchase good foods, members of the household will eat more of those good foods. Perhaps more surprisingly, if you give a household more good foods, household members will eat less bad foods.

Yes, under the WIC model households will use only about two-thirds of the food assistance; while under the SNAP model, households will use nearly all of the assistance than they would under the SNAP model. As a result, the WIC model is moderately cheaper. Even so, distributing the benefit according to WIC rules yields no difference in food security and much larger improvements in nutritional intake.

Yes, the WIC program’s rules are not ideal. Some retailers exploit the rules for their profit. For households, complying with the rules is time consuming and requires attention to detail. Those rules can probably be tweaked to address some of the problems.

Finally, yes, the WIC program is paternalistic. Rather than trusting families to buy good foods, the WIC program only allows benefits to be used on good foods. Such paternalism is annoying to program participants and rightly troubling to many citizens and policy experts.

Our results imply that there is a real trade-off: A little bit of inconvenience and paternalism yields much better nutritional outcomes; outcomes that have been hard to achieve by other policy means.
The research is clear. Providing moderately more food assistance doesn’t do much. Providing rebates for purchasing healthy foods goes a long way.[4] Limiting choice has the greatest impact. Given this trade-off, if getting poor households to eat more healthy foods is a major policy goal, we should give serious consideration to limiting choice in food assistance.

[1] “The Federal Government’s Program to Feed Pregnant Women and Children Is Terrible”, Heather Tirado Gilligan, Nov. 30 2016 11:08 AM. [2] U.S. Department of Agriculture, Food and Nutrition Service, Office of Research and Analysis, Effects of the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC): A Review of Recent Research by Silvie Colman, Ira P. Nichols-Barrer, Julie E. Redline, Barbara L. Devaney, Sara V. Ansell, and Ted Joyce. Project Officer Janis Johnston. Report WIC-12-WM. Alexandria, VA: January 2012. [3] Collins, Ann and Jacob Alex Klerman. 2017. “Improving Nutrition by Increasing SNAP Benefits.” Special Issue of American Journal of Preventive Medicine. [4] Collins Ann, M., Ronette Briefel, Jacob Alex Klerman, Gretchen Rowe, Anne Wolf, Anne Gordon, Julia Lyskawa, and Syeda Fatima. 2016. Summer Electronic Benefit Transfer for Children (SEBTC) Demonstration: Summary Report.

Work With Us
Ready to change people's lives? We want to hear from you.
We do more than solve the challenges our clients have today. We collaborate to solve the challenges of tomorrow.