Are Taquerias Healthier than American Fast-Food Restaurants?

More than one third of the U.S. adult population is obese (35.7%). Obesity is associated with higher risks of heart disease, stroke, type II diabetes and certain types of cancer. Despite the increasing attention to obesity in the academic and public arena, not much is known about its causes and remedies. Fast-food restaurants and food-deserts have been blamed to be important factors of the obesity epidemic. While there is consensus that fast-food is less healthy and can importantly contribute to obesity, it is less clear whether exposure to fast-food restaurants and food deserts affect health. A few studies have attempted to analyze the causal effects of proximity to fast-food restaurants, but different strategies led to different conclusions on the magnitude and the significance of the effects on weight gain and obesity rates. The debate on the effects of fast-food exposure remains open. Furthermore, less is known about the role of different types of restaurants. Given the evidence on the healthier dietary habits of first-generation Mexicans and on the unhealthy assimilation in weight gain and obesity among Hispanics, one may wonder whether proximity to ethnic restaurants (eg. Mexican restaurants) might provide a more traditional and healthier alternative to American-style fast-food restaurants. Using survey data show that child and parent body mass index (BMI) are lowest in Mexican-American families who select Mexican restaurants.  This paper contributes to previous studies by analyzing the relationship between proximity to different types of restaurants (fast-food, traditional Mexican restaurants etc.) and a broad set of maternal and child's health outcomes. In particular, I focus on excessive maternal weight gain, which has been linked to post-partum obesity and adverse health outcomes. Focusing on pregnant women allows me to exploit the large sample of administrative records drawn from the Vital Statistics of Florida to use mother fixed effects. This allows me to account for time invariant individual characteristics that might be correlated with both residential location and health outcomes. Similarly to Currie et al. (2010), I use data on the exact geographic location of restaurants and analyze how the availability of fast-food and ethnically defined restaurants is related to maternal and children health outcomes. The restaurant data are merged with information drawn from the Vital Statistics data for the metropolitan areas of Miami, covering the universe of births occurring in this area between 1990 and 2009. This strategy allows me to identify precisely even small effects exploiting the large size of the data. The results of this paper suggest that proximity to Mexican restaurants is associated with a lower likelihood of excessive weight gain among US born mothers during pregnancy. Effects appear to be larger among black, low-skilled and young mothers, though these differences are not precisely estimated. No protective effect of Mexican restaurants was found for the foreign-born. Fast-food proximity has no significant effects on maternal weight gain. Proximity to fast-food or Mexican restaurant are not significantly associated with other maternal and child's health outcomes.