For instance, Costa points out that some 80 percent of murders in Mexico between December 2006 and July 2010 were committed in just 7 percent of that country’s municipalities, mainly in the beleaguered states of Baja California, Chihuahua, Sinaloa, Michoacan and Guerrero. In El Salvador, more than two-thirds of homicides are committed in only 30 (11 percent) of the country’s 262 municipalities. Similar “hotspots” exist in Guatemala, Colombia, Peru and other countries across the region, serving as a reminder that citizen insecurity often has as much to do with local dynamics as it does with national security policy.
The spike in Latin America’s homicide rate has been accompanied by a shift in public perceptions of security. Either because of firsthand experience or increased media attention to the issue, more Latin Americans rate security as their biggest concern than ever before. Costa cites data from public opinion polling project Latinobarometro which indicates that anxiety over crime has been growing steadily since the organization began monitoring the issue in 1995, and finally supplanted economic status region-wide as the number one concern in 2007.
Perhaps the most interesting element of Costa’s paper is his highlighting of the puzzling lack of evidence for the causes of the region’s high homicide rate. Although the prevailing wisdom suggests that much of the violence in Latin America is fueled by the drug trade, the UN’s 2011 Global Study on Homicide claims that only a quarter of homicides in the region are linked to drug trafficking. A look at violence in the Andean region supports this. If drug trafficking were the main cause of violence, one would expect the homicide rates of Peru (5.2 per 100,000) and Bolivia (8.9) -- the second and third largest producers of cocaine, respectively -- to be on par with that of Colombia (33.4).