Food Deserts or Mirages?

A food desert is a concept that is simple, powerful, and widely accepted as a key contributor to poor nutrition and obesity. The idea springs from the observation that areas with poor access to fresh and affordable foods suffer from high rates of obesity. But when you scratch below the surface of evidence for the role that food deserts play and for interventions that can help, you might find yourself wondering if food deserts are, in fact, seductive mirages. Three new publications from economists — two studies and a review paper — shed light on the state of evidence about this problem.

In the one of two studies on the subject published in the March issue of Applied Economic Perspectives and Policy, Kyureghian and colleagues looked at the effects of food store access and income on fruit and vegetable purchases. They combined national data on retail food availability with data on actual purchases in a mixed effect analysis. They found that access to a supermarket outlets did not affect fruit and vegetable purchases, but income did. Even so, income and access were inadequate to explain wide variations in fruit and vegetable purchases. Policies that only address access and affordability, they concluded, will likely be futile.

In the second study, Alviola and colleagues analyzed data from Arkansas to determine the effect of food deserts on childhood obesity rates. They found no statistically meaningful relationship between food deserts and obesity rates in Arkansas school districts. Their finding was consistent across different models and analyses, regardless of the inclusion of urban data. The authors acknowledge the importance and enormous attention given to food deserts in discussing their findings and they stop short of saying that food desert initiatives should be discarded. But, they conclude, “There is no question that more work is needed to definitively identify the effect of food desert areas on childhood obesity.”

Betsy Donald presents a thoughtful review, rethinking the problem of food deserts, in the March issue of the Journal of Economic Geography. She points out the complexity of the problem and the lack of rigorous pre/post evidence from retail interventions that will be necessary to advance serious policy changes. She also points out that improved access alone will not be adequate to foster healthier diets. Affordability is perhaps more important and tougher to solve than access. And retail food distribution is changing so fast that it’s easy for research to become quickly irrelevant.

Commenting on the evidence for alleviating food deserts as a key to reducing obesity rates, David Allison said “this may be another beautiful hypothesis slain by ugly facts.” Allison, a Distinguished Professor and Associate Dean for Science at the University of Alabama at Birmingham, pointed to recent, separate studies by Brian Elbel, Adam Drewnowski, and Richard Mattes that call suppositions about food deserts into question. For initiatives against food deserts to help, he said, would require a number of cascading events to line up:

  1. Increased access that leads to more purchases of beneficial foods and less of others
  2. Shifts in actual food consumption patterns
  3. Reduced caloric intake
  4. All while holding in check other factors that could neutralize the benefit

This paints a picture of inadequate evidence to be confident that fixing food deserts will reduce obesity rates.

Click here to access the  Kyureghian study, here to access the Alviola study, and here to access the Donald review.

Nevada Desert Mirage image by Brocken Inaglory / Wikimedia