Skip to:

Ongoing

Economic Consequences of Violence

rtxeleu.jpg

A man sits outside a closed restaurant in downtown Mexico City April 30, 2009. Mexico began shutting down parts of its economy on Thursday to slow the spread of a new flu strain as officials urged increased worldwide precautions against an imminent pandemic. With its tourism industry savaged, shoppers staying home and exports to the U.S. in steep decline, Mexico could find itself in the longest, deepest recession it has seen in years, according to analysts.
Photo credit: 
REUTERS/Jorge Dan

Researchers

Senior Fellow
  • Associate Professor, Political Science
Research Staff, CDDRL
CDDRL Postdoctoral Fellow, 2012-13

The Poverty and Governance Program has an ongoing research agenda aiming to evaluate the impact of drug-related violence on economic activity. We start with the premise that drug cartels are organized to maximize profits from trafficking drugs illegally into the United States. Under certain conditions narco-traffickers operate, produce, and traffic drugs without using coercive strategies against citizens or businesses. In these situations drug trafficking could be carried out under moderate levels of drug-related violence and therefore has limited impact on economic activity.

However, when a drug related conflict sparks and escalates between and within drug trafficking organizations, cartels and criminal cells might find it in their interest to diversify their portfolios of activities to include other types of crimes such as kidnapping, extortion, human trafficking, and oil theft, among others. These illegal activities are expected to have a much more direct impact on society and therefore on economic activity.

In one of the studies prepared for the Inter-American Development Bank (IDB) we measured the economic costs of drug-related violence in Mexico. The study was one of the first analyses proposing electricity consumption as an indicator of the level of municipal economic activity in Mexico from 2006 to 2010. The study also measures the effects of drug related violence on labor indicators, including employment, small businesses, and labor income.

Total Homicides for Selected Municipalities: 1998-2010

During the government of Felipe Calderón, the economy grew at an average rate of 1.84%, the lowest rate of the last four administrations. Although this low performance could be partly attributed to the increasing violence in the country, it is difficult to differentiate the effects of such violence from other factors that strongly affected Mexico’s economy. The financial crisis in the United States in 2008 and 2009 was followed by a severe contraction of the Gross Domestic Product in Latin America. In the same years, the Mexican government cut public spending significantly, and the country was impacted by the Influenza (AH1N1) epidemic that paralyzed economic activity for several weeks.

To evaluate the impact of violence on the economy, the study uses two identification strategies. First, we use an instrumental variable specification to model the marginal impact of violence. We use the instrument proposed by Mejía and Castillo (2012) based on historical seizures of cocaine in Colombia interacted with the distance of the Mexican border towns to the United States. The study finds that marginal increases of violence have negative effects on labor participation and on the proportion of employed people in a municipality. The marginal effect of violence is also negative and substantive for the proportion of business owners and for labor income, but not for energy consumption.

Average Electricity Consumption by Treatment Group The figure shows the average electricity consumption of the treated municipalities and their synthetic controls. Time is normalized to zero indicating the beginning of an inter-cartel conflict (turf war).

However, it is likely that there is a violence threshold above which a substantive deterioration of the economic activity should be expected. The inter-cartel war for control of certain drug trafficking routes and plazas generally comes paired with substantial increases in both executions between members of rival criminal organizations and criminal activities affecting the general population. Under these conditions, it is likely that a critical number of economic agents will stop their businesses operations and families will move into more secure areas, reducing the total production of goods and services and the economic activity of the region.The instrumental variable strategy identifies the marginal effects of homicide rates on economic variables. Nevertheless, it is possible that the violence does not affect the economy in a linear form. Under conditions of moderateviolence, firms might internalize the costs associated with such environment, including extortion payments, robberies, and increasing security technology, among others.

Electricity Consumption in Ciudad Juárez, 2002 - 2010 The figure shows the average electricity consumption in Ciudad Juárez and its synthetic control before and after the start of an inter-cartel conflict in 2007.

Our results indicate that wars between rival drug cartels for strategic routes or territories in the last six years have had a profound impact on the local economy in Mexico. Moreover, by proposing “threshold” effects of violence on economic activity, the study provides guidelines for future research aiming to understand the relationship between violence and economic activity.To measure the effects of inter-narco drug wars on local economies, we used the synthetic control methodology proposed by Abadie and Gardeazabal (2003) and and Abadie, Diamond and Hainmueller (2010). We define the beginning of a “turf war” in a municipality as the moment when violence exceeds an upper band defined around historical trends of violence in territory. The analysis indicates that those municipalities that experienced dramatic increases in violence between 2006 and 2010 significantly reduced their energy consumption. In particular, affected municipalities consumed on average 4.19% and 7.4% less electricity per capita than their synthetic counterfactuals during the first and second year after the start of a war between cartels.

Updated: Dec. 2013

Publications