Health workforce reallocation in the aftermath of conflict: Evidence from Colombia

While a large literature has studied the effects of violent conflict on health outcomes, little is known about how violence reduction can affect a key driver of post-conflict recovery, namely the quantity and type of healthcare workers. By leveraging a permanent ceasefire that ended over five decades of armed conflict between the Colombian government and the FARC insurgency, we study the extent to which conflict termination affected the share of different types of health workers in areas more exposed to FARC violence relative to other places. Based on administrative records on the location of all formal healthcare workers in Colombia and using a difference-in-differences strategy, we find that a municipality that experienced one standard deviation higher FARC violence intensity relative to the rest of the country witnessed 13.4% post-ceasefire differential decrease in the share of employed healthcare workers per 1,000 people. We find a stronger decrease among vocational nurses and a weaker decrease among physicians. We show that this effect is likely explained by lifting mobility restrictions in previously violent areas, and document that, because the net reduction in healthcare workers increased the within-municipality share of (higher-educated) physicians, it did not translate into a deterioration of mortality rates or healthcare service provision.