Yes, we needed economists to tell us this. A new working paper
http://papers.nber.org/papers/w16152#fromrss
published by the National Bureau of Economic Research finds “strong evidence for a revenge effect” when examining the relationship between civilian casualties caused by the U.S.-led military coalition in Afghanistan and radicalization after such incidents occur. The paper even estimates of how many insurgent attacks to expect after each civilian death. Those findings, however intuitive, might resolve an internal military debate about the counter-productivity of civilian casualties — and possibly fuel calls for withdrawal.
“When ISAF units kill civilians,” the research team finds, referring to the U.S.-led coalition in Afghanistan, “this increases the number of willing combatants, leading to an increase in insurgent attacks.” According to their model, every innocent civilian killed by ISAF predicts an “additional 0.03 attacks per 1,000 population in the next 6-week period.” In a district of 83,000 people, then, the average of two civilian casualties killed in ISAF-initiated military action leads to six additional insurgent attacks in the following six weeks.
The team doesn’t examine the effect of CIA drone strikes in neighboring Pakistan, the subject of fierce debate concerning both the level of civilian deaths the strikes generate and their radicalizing effect.
A team of four economists — Stanford’s Luke N. Condra and Joseph H. Felter, the London School of Economics’ Radha K. Iyengar, and Princeton’s Jacob N. Shapiro — used the International Security Assistance Force’s own civilian-casualty data to reach their conclusions, breaking it down by district to examine further violence in the area in which civilians died. They examined the effect of over 4000 civilian deaths from January 2009 to March 2010 by looking at the sometimes-lagging indications of reprisal attacks in the same areas. To be clear, the team’s research is inferential, creating a statistical model to examine spikes in violence following civilian-casualty incidents, rather than interviewing insurgents as to their specific motivations.
But in their study, the researchers found that there’s a greater spike in violence after ISAF-caused civilian deaths than after insurgent-caused ones. “An incident which results in 10 civilian casualties will generate about 1 additional IED attack in the following 2 months,” the researchers write. “The effect for insurgents is much weaker and not jointly significant.”
In other words, even if the insurgents possess a “total disregard for human life and the Afghan people,” as an ISAF press release reacting to this weekend’s insurgent bombings in Herat put it, Afghans effectively would rather be killed by other Afghans than foreigners.
That’s not all. The researchers found that ISAF-caused civilian casualties corollate with long-term radicalization in Afghanistan. Plotting reprisal incidents of violence in areas where civilians died at coalition hands, the data showed that “that the Coalition effect is enduring, peaking 16 weeks after the event. This confirms the intuition that civilian casualties by ISAF forces predict greater violence through a long-run effect.” That’s consistent with intuitions that civilian casualties “are affecting future violence through increased recruitment into insurgent groups,” although they find no direct evidence for such a thing. Interestingly, the researchers found the opposite to be the case in Iraq: U.S.-caused civilian casualties are more likely to cause short-term retaliatory spikes than they are violence over the long term. (Yet.)...
