Constructive COIN

2CommentsJoin

Sign-up for free weekly updates from ForeignAffairs.com.

“We’ve got a government in a box, ready to roll in,” said General Stanley McChrystal, the top American commander in Afghanistan, of plans for holding the hard-won town of Marja in February. As Lieutenant General Mohammed Karimi, the deputy chief of staff of the Afghan army, explained, “We want to show people that we can deliver police, and services, and development. We want to convince the Afghans that the government is for them.”

Only a few years ago, the strategy for U.S. troops in Afghanistan and Iraq focused on targeting insurgents. The strategy failed; critics now call it “mowing the grass,” since soldiers would repeatedly clear areas of insurgents only to see them reappear afterward. In the face of steadily increasing violence in Iraq from 2003 through 2006, the approach was abandoned. Today the U.S. military's counterinsurgency strategy centers on protecting the population, with a special emphasis on political and economic development.

The objective is often termed “winning the hearts and minds” of the population, and its logic is simple. Insurgents cannot operate without civilians learning something about their location and identities: recruiting, raising funds, and preparing weapons and explosives all entail observable actions. The more effective the government is at providing services to civilians, the more likely it is that they will prefer the government over the insurgents. In turn, the population will provide the government with a steady stream of tactically useful information -- calls to anonymous tip lines about the location of weapons caches, for example -- that the government can use not only to defeat the insurgents but also to prevent their reappearance.

But does that development-based strategy work? The assumption that development helps governments defeat insurgents is testable: one can check if supplying government services is associated with lower levels of violence. In a study we conducted using data on reconstruction spending and violence in Iraq, we found that the provision of certain government services does lead to a reduction in violence.

To measure the provision of government services, we collected data on reconstruction expenditures through the Commander’s Emergency Response Program for the period of 2004–8. CERP funds are administered by U.S. forces and spent locally, mostly on small projects implemented by Iraqi contractors. To measure violence, we used data on violent incidents against coalition and Iraqi forces.

Looking across districts in Iraq, the raw correlation between violence and CERP spending is positive; more money is spent in the more violent districts. That is not because CERP spending increases violence but rather that CERP administrators spend strategically: more violent places attract more spending. Once we control for a region’s proclivity for violence (by comparing changes in violent incidents to changes in spending), CERP spending appears to reduce violence. That effect was especially strong in the period after the surge in U.S. troop levels, from 2007 on. Importantly, the effects are not limited to reconstruction spending by the military; the Community Stabilization Program, which was administered by the U.S. Agency for International Development, also focused on small projects and had a similar violence-reducing effect.

But this approach has not always worked. Before 2007, it was unsuccessful, likely because the forces administering it tended to spend less time in their communities and so knew less about them. In fact, much of the reconstruction aid in Iraq has not helped reduce violence; CERP and related programs account for only $3.1 billion of the $25.3 billion of U.S. reconstruction spending in Iraq that we observed. The remainder -- including spending on large projects and programs carried out primarily by foreign contractors -- appears to have had no violence-reducing effect. Our data suggest that reconstruction money works when projects are small, troops have a good working relationship with noncombatants, projects are chosen in consultation with local officials, programs are administered by local contractors, and a provincial reconstruction team is nearby to provide guidance.

Does job creation help? Much of U.S. aid to countries emerging from internal conflict is premised on the idea that increasing employment opportunities for potential fighters will reduce terrorism and insurgent violence. Yet despite the massive expenditures involved, the relationship between governance and economic development in conflict zones is not well researched. The assumptions underpinning the job creation argument -- that insurgency is a full-time job, that it is a low-skill occupation, and that the key limit on violence is the number of fighters insurgent organizations can recruit -- are questionable in many conflicts and have largely gone untested.