Average and quantile treatment effects of the American Folic Acid Fortification: an evaluation in a quasi- experimental framework

The American program of folic acid fortification is generally thought to have increased the average amount of serum folate in the population and, hence, widely considered as a successful public health intervention. We use several waves of the “National Health and Nutrition Examination Survey” (NHANES) to evaluate the causal impact of the fortification of ready-to-eat cereals on serum folate concentration, using a quasi experimental framework. First, we compute the average treatment effect by using matching methods to solve the problem of selection on observables, finding a strong selection into treatment mainly based on race-ethnicity and education. Second, we assess the distributional impact of the fortification by computing quantile treatment effects, under different assumptions on the dependence between the distributions of potential outcomes, and we find significant variation in the impact of fortification across the population, thus rejecting the common effects model. Fortification appears to have had the least (though still modestly beneficial) impact among those that most needed it and the biggest impact among those that needed it least, thus suggesting the presence of folate overconsumption in the latter group, with potential adverse health effects. Third, by controlling our estimates for the concentration of beta-carotene, we find support for the hypothesis that part of the increase in serum folate concentration can be explained by changes in diet, leaving a smaller attributable effect to the fortification itself.