In the years following the 9/11 attacks, a new generation of scholars have begun to study the economic, political, religious, ethnic, and cultural factors that motivate and shape asymmetric conflicts in locations around the globe. To date, economic methodologies have yielded subtle and interesting insights about how a variety of economic and political factors influence conflict and human behavior. Recently, Ethan Bueno de Mesquita of the University of Chicago Harris School of Public Policy convened a group of scholars who have collected data from such remote and far-flung conflict zones as Afghanistan, the Democratic Republic of the Congo Iraq, Pakistan, the favelas of Rio, Somalia, and Vietnam. Together, scholars demonstrated the promise of this emerging field and the value of applying data-driven methodologies to gain valuable insights that can inform policymaking.
Data gathering can be an enormous challenge in conflict zones, often involving not just time and resources but also personal risk and creative thinking about how to access what already exists. Through the work they presented, researchers demonstrated how innovative approaches to data collection and analysis can open the door to new methodologies and fresh insights. Discussions explored opportunities to apply similar data collection strategies to other contexts.
Political Structures
A number of research papers explored how existing and evolving political structures—or the lack thereof—play a central role in fomenting and shaping violent conflict.
Somalia is a largely ungoverned, low-information environment where piracy threatens local trade and clans largely police themselves and settle interclan disputes. As such, it is provides a natural testing ground to examine the signals that often drive conflicts, the dynamics of these conflicts, and the economic impact on local industries like livestock trading and fishing. Researchers developed a model and used National Geospatial Intelligence Agency data to investigate whether simple cooperative agreements might be self-enforcing under specific conditions, and the impact that variables like trade volume and income have on conflicts and responses.
For many years, the favelas of Rio de Janeiro were large, ungoverned territories that existed without street names or property rights; armed groups and drug gangs dominated the neighborhoods. But that changed starting in 2008, when 9,000 police endeavored to retake 200 slums populated by 600,000 residents. Using data collected by The Instituto de Seguranca Publica, aggregated monthly crime and violence data within pacification zones, and data from a local crime hotline, researchers were able to track such variables as drug and gun seizures and arrests, police killings and homicides, theft, and migration of criminal activity to surrounding districts. This data shed new insights into both the short- and long-term impact of police control.
Another paper focused on the structure of factional coalitions in Afghanistan and Pakistan. Researchers combined field data and media reports with ideas from machine learning to identify and map clusters of violence geographically and examine which factions were controlling this violence. Using data from the Worldwide Incidents Tracking system and BFRS data from Pakistan, they studied how insurgent groups are organized, what ethnic composition they have, where they operate, when they attack, and what motivates and constrains them. The study’s scalable methodology also holds promise for studying alternative contexts as the Philippines, Colombia, and territorial settings like neighborhood gangs.
Throughout recorded history, interstate territory disputes have increased propensity for war, raised military spending, depressed financial flows and impeded democratization. One study examined border disputes dating back to Europe in the 1600s to consider what forms conflicts take, and why they arise in some places rather than others. Historically, claim–making activities are confined to a very few actual years, reflect historical border precedents, are dependent on the value of the land resources in question, and depend on what other systematic uncertainty actors are facing.
Human Behavior
Another set of papers demonstrated how empirical research can yield fascinating insight into what motivates human behavior and action on both sides of a conflict.
For years, policymakers have debated whether humanitarian aid provided in civilian casualty zones helps short-circuit radicalization, or feeds additional violence. Researchers identified a natural control group in war-torn Afghan villages that received support from USAID between 2011 and 2013. Interviews with villagers revealed that acceptance or rejection of aid were shaped by two factors—cognitive biases regarding who was responsible for the air strikes and civilian casualties, and who dominated the space where the humanitarian assistance was provided.
One study in Pakistan gathered data first-hand through a carefully constructed experiment designed to study anti-American sentiments. Through a disguised survey of residents representing a broad range of Pakistani ethnic groups, the study examined how variables like financial costs and changes in social contexts reflect on both “sense of self” and individual decision making. The study asked under what conditions (private vs public) individuals would betray their anti-American beliefs and accept money for participating in the survey. This painstaking process led to insights about the potential that US-sponsored programs may have to change embedded individual and social attitudes.
Getting policing right in the third world is a significant problem, and poor policing can drag down economies and harm state/citizen relationships. One study examined this dynamic in the Democratic Republic of the Congo, where the state is a coercion-wielding organization characterized by rampant corruption. Researchers studied eight police bureaus to observe how this corruption is organized and executed. They observed police officer behavior at 15 intersections over 30 days, and the daily data collected from both police stations and the Ministry of Finance revealed a culture of bribery that dictated police income levels and disrupted civilian life. The study also examined whether an alternative wage structure could change this paradigm.
The Vietnam War was the first conflict to employ data science to influence tactical decision making, but until recently no scholars had applied recent insights from the Iraq campaign to understand military decisions in Vietnam. One researcher gained extraordinary access to government-owned data collected in the field and examined how this data guided the military’s counterinsurgency strategy. She studied the relative effectiveness of the US Army’s top-down approach to bombing at-risk villages (as well as the imprecise data-driven process guiding those decisions) to the bottom-up approach the Marines employed to win hearts and minds within local villages by emphasizing public works. Her findings offer valuable insights into counterinsurgency strategies in effect today.
This was the first Becker Friedman Institute conference to explore the economic analysis of conflict in different regions and contexts around the world. The University of Chicago is poised to become a center of seminal research in this area with the establishment of the Pearson Institute for the Study and Resolution of Global Conflicts and The Pearson Global Forum. Housed at the Harris School of Public Policy and funded by the Thomas L. Pearson and the Pearson Family Members Foundation, the new institute has appointed two inaugural faculty members: Chris Blattman and Oeindrila Dube, both trained economists. They join an active community of scholars at Harris and across campus studying the politics and economics of conflict, violence, poverty, crime. This offers rich opportunities for cross-disciplinary collaboration, as the two institutes provide an intellectual destination for the best scholars working in this field.
—Rob Squire