Using Google data to measure the role of Big Food and fast food in South Africa’s obesity epidemic

•Georeferenced Google data on supermarkets and fast food merged with nationally representative survey.•First study in a developing country to model effects of Big Food and fast food on consumer nutrition and health.•Household proximity to supermarkets and fast food restaurants contributes to rising BMI and higher probability of overweight and obesity.•Findings robust also when controlling for socioeconomic status and regional fixed effects. Many developing countries face a rapid increase in overweight and obesity, inasmuch as the prevalence has now nearly converged to levels observed in high-income countries. Among other factors, the rise in obesity is caused by a nutrition transition involving higher affordability and consumption of heavily processed or otherwise unhealthy foods containing high amounts of added sugar, fat, and salt. This development is accompanied by the growing expansion of, and increased access to, large modern food retailers (Big Food) and fast food restaurants. Using a novel methodology, we are able to link proxies of exposure to modern food environments based on Google data with nationally representative micro-level nutrition and health data to examine the influence of Big Food and fast food on overweight and obesity. The micro-level data come from the National Income Dynamics Study (NIDS) in South Africa, a middle-income country with alarming and further rising levels of obesity. We find that proximity to Big Food retailers and fast food restaurants increases overweight and obesity significantly, even after controlling for income and other confounding factors. The results suggest that the shape of food environments needs higher policy attention to promote more healthy food choices, which is true in South Africa and beyond.