LAUSR

The introduction of Seguro Popular (SP)- providing health insurance to over 50 million Mexicans since the early 2000s-represents a large shift in health care delivery to the Mexican population. And yet, its impact on Mexico's marginalized communities has been little studied, and its impact on health is unclear. Using a survey of poor urban Mexicans and Mexican Ministry of Health administrative data, this article investigates SP's effect on those at the most risk for health disparities by looking at the impact of the programme on demand for preventive care services, especially among women, children and the indigenous. Three outcomes important to Mexico's burden of disease are explored: general physical exams, diabetes screening, and cervical cancer screening. Ordinary least square regressions show that the introduction of SP is associated with an increase in demand for all three services, but these results are likely biased due to selection into the programme. I then use the staggered geographic roll out of SP between 2004 and 2007 to identify the causal impact of the programme on demand. I use length of exposure to SP as an instrumental variable to predict SP affiliation in 2009. Two stage least squares estimates of the causal impact of SP on demand for preventive care services finds that SP affiliation increases adult demand for physicals, but does not affect demand for diabetes screening. Additionally, I find that female and child SP affiliates are less likely to demand physicals, while affiliates who identify as indigenous are less likely to demand physicals but more likely to demand cervical cancer screenings.