Jeffrey Bloem is a Ph.D. candidate in Applied Economics at the University of Minnesota.
As previous recap posts highlight the study of violence, conflict, and crime is becoming an increasingly active area of research in development economics. This trend continued at the recent Midwest International Economic Development Conference (MidDev) held at the University of Wisconsin at Madison. In this blog post I discuss six conflict papers on the causes, consequences, and policy responses to conflict and crime.
Understanding the Causes of Conflict
Prominent empirical studies examine the effect of US food aid shipments on conflict in recipient countries (Nunn and Qian 2014) and GDP growth on conflict (Hull and Imai 2015). Paul Christian shows that these results should be handled with extreme caution. The authors clearly demonstrate the challenges and perils of using an instrumental variable with panel data.
Most strikingly, the results from both Nunn and Qian (2014) and Hull and Imai (2015), who use lagged US wheat production and real interest rates respectively as instrumental variables, can be fairly precisely replicated using sales of cassette tapes as an alternative instrumental variable. The best part of this startling paper, however, is a list of recommendations that applied researchers can employ to test whether their results are spurious (blog posts on earlier versions on the Christian & Barrett paper available here and here).
The causes of conflict can arise from both macro sources (e.g., domestic agricultural policies and international trade policies) and micro sources, such as individual frustrations stemming from a variety of market failures. Although the various local institutions surrounding marriage is a classic literature within development economics, little empirical work examines how the marriage market influences the prevalence of conflict via extremist organizations.
Jonah Rexer shows that the market for marriage, particularly as experienced by young men in rural Nigeria, may be an important contributing factor driving the prevalence of violent conflict. The author finds that diminished marriage prospects increases the incidence of Boko Haram violent activity in Nigeria. This finding represents a new and potentially important cause of conflict, but must be interpreted with caution considering the challenges of using instrumental variables as discussed above.
Assessing the Consequences of Conflict
Conflict can have tremendous consequences to a multitude of welfare and economic outcomes, both on firms and economic activity, as well as households’ wellbeing. On the former, Sylvan Herskowitz examines the response of firms in Afghanistan to the presence of conflict and insecurity. The authors use cell phone records as a measure of private sector activity and find that terrorist activity reduces the presence of firms in targeted districts in Afghanistan. This effect can be explained by an increase in firm exit and a reduction in new firm entry.
A brief side note about the use of data is worthwhile. The use of cell phone records to measure or approximate private sector activity represents a creative methodological innovation. In a context that is experiencing conflict and insecurity collecting reliable data can be challenging, expensive, and often times impossible. The authors of this paper show that cell phone records can serve as a useful source of data to understand the behavior of firms.
We now move from the consequences of conflict on firms to the consequences of conflict on individual wellbeing. From 1996 through 2006 Nepal experienced a violent civil war. Lokendra Phandera studies the effect of exposure to this civil war on long-term health outcomes and effects on the next generation. He finds that exposure to this conflict during childhood has a negative effect on height in adulthood. Additionally, a mother’s exposure to the civil war has a negative effect on their child’s development.
These studies are important, not because they are surprising, but because they help us quantify the disruption that the presence and prevalence of conflict can have on both firms and households. Lokendra’s study in Nepal shows that these effects can linger for a remarkable number of years and can even persist across generations.
Evaluating Policy Responses to Conflict
Given the observation that violence, conflict, and crime can have negative and lasting effects on both firms and households, what should policymakers do to help limit these negative consequences?
In Lebanon the United Nations Children’s Fund (UNICEF) and the World Food Programme (WFP) implemented a conditional cash transfer program for Syrian refugees. The cash transfer was designed to cover the cost associated with commuting to school and to compensate households for any forgone income earned by their child. Mitchell Morey shows that although this program lead to no effects on school enrollment, receipt of the cash transfer increased attendance for children who were already enrolled. The effects on attendance are quite large and meaningful. Relative to children who did not receive the cash transfer, children attended over an extra half-day of school per week.
Political corruption is almost universally understood as a harmful phenomenon, and it is a form of criminal activity that is rarely punished. In China, anti-corruption campaigns have grown in popularity in recent years, however it is not clear whether these anti-corruption campaigns are beneficial for the overall economy. Zin Jin studies a sudden “crack down” of political corruption in Heilongjian Province, China. The authors find that this anti-corruption movement had a negative effect on labor productivity and reduces firm entry. Most evidence supports the idea that this “crack down” on corruption reduced in the usefulness of political connections. Whereas prior to the “crack down” political connections lead to productivity gains, after the “crack down” these connections where much less useful.
Given the challenges in understanding the causes of conflict and the long-lasting consequences of conflict much more research is needed to better assist policymaker response. These two studies represent just the beginning of and effort to build a broader understanding of how various policies can be used to best respond to violence, conflict, and crime.
