Unequal Access to Food

From wide-open rural lands to densely built inner cities, low-income Americans in every state struggle to find fresh and healthy food in their communities. But is the obvious fix—building more grocery stores—the most effective one?

New research from Elisabeth Paulson, PhD ’21, offers a different, more nuanced look at the issue. Paulson’s data-driven mathematical model and accompanying empirical analysis grouped families receiving Supplemental Nutrition Assistance Program (SNAP) benefits based on their shopping and nutrition habits and how far they lived from grocery stores.

Households that lived within half a mile of a market and did not appear to value healthy eating significantly benefited from easy access to food, Paulson’s research showed. These families, according to the USDA’s National Household Food Acquisition and Purchase Survey, did not follow food pyramid guidelines or frequently read nutrition labels. Still, they bought more produce and shopped more often than other families that shared similar nutrition behaviors but lived over a half-mile from a market.

Yet, Paulson found that distance to a grocery store, whether near or far, only made a marginal difference for people who valued healthy diets. This group, Paulson says, was willing to travel outside their neighborhood to buy fruits and vegetables.

Local and federal governments provide financial incentives for businesses to open grocery stores in enclaves that have limited produce options, known as food deserts. To have the greatest impact, Paulson wrote in her paper, the construction of grocery stores should be in “low-value nutrition” neighborhoods.

While much of her research has focused on access-based interventions, Paulson cautions against taking a single-solution approach to the issue.

“This is not just a problem of access,” says Paulson, whose research is under review. “If our goal is to promote healthy diets, the higher-level question is: What other intervention, or combination of interventions, could be used to achieve that goal?”

Paulson is currently tackling this higher-level problem. She is devising a data-driven model that gives families a personalized set of incentives—access, education, and price—based on their traits, such as nutrition habits or food preferences. This model will enable government agencies to adjust and craft policies at more individualized levels using consumer behavior trends.

She also has partnered with the Massachusetts Department of Transitional Assistance. Its Healthy Incentives Program offers SNAP rebates on purchases of fruits and vegetables made at participating farmers markets, farm stands, and community-supported agriculture. Paulson will advise policymakers on how to expand the program by applying methods used in her research as well as other principles, including “network effects.”

Paulson researched food deserts at the suggestion of her advisors, Retsef Levi (J. Spencer Standish (1945) Professor of Management; Professor, Operations Management; Co-Director, Leaders for Global Operations Program) and Georgia Perakis (William F. Pounds Professor of Management; Professor, Operations Management and Operations Research and Statistics; EMBA Faculty Director; Co-Director, Operations Research Center), who are also co-authors of the study. Paulson had collaborated with Levi and Perakis on making food supply chains more sustainable and reducing waste in the farm supply chain, but knew little about food deserts. Still, the opportunity to work on a critical problem using advanced analytics appealed to her.

The daughter of a statistics professor and a software engineer, Paulson loves the rigors of theoretical math. As an undergraduate student, she immersed herself in game theory, especially strategies about deception. She even dedicated her senior thesis to the topic. “Game theoretic models were fun and interesting to analyze, but the real-world applications were not as straightforward,” she says.

Then, there was the type of math she performed as a data scientist in her first job after college. While her work helped shape health care policy, computing data summaries and running regressions lacked the intellectual challenge she was craving. She searched for doctoral programs and found the MIT Operations Research Center (ORC), an interdisciplinary graduate program founded in 1953. Paulson saw that ORC students were working on sophisticated, data-driven solutions affecting businesses and society, and wanted to be a part of it. During her five years at MIT and ORC, Paulson distinguished herself as a graduate student, earning a National Science Foundation graduate research fellowship and receiving several honors for her work.

Paulson is spending the 2021–22 school year as a postdoctoral scholar at Stanford University’s Immigration Policy Lab. She is doing research and data analysis for GeoMatch, a machine learning tool that pairs immigrants to locations within host countries based on their education, work skills, and personal characteristics. Then, in July 2022, Paulson will resume her food policy research as an assistant professor at Harvard Business School.

In the spirit of “mens et manus” (“mind and hand”), Paulson aims to strengthen her partnerships with nonprofit and policy groups. She credits Levi and Perakis for reminding her to think outside of the academic world.

“They always kept me working toward impact,” Paulson says. “It has shaped how I view my research going forward.”