Joshua Tucker: The following is a guest post from Princeton University PhD candidates Andrew Shaver (political science) and Gabriel Tenorio (economics). Shaver previously served in Iraq as a member of a Pentagon task force established to carry out economics-based counterinsurgency programs.
The Islamic State in Iraq and Syria (ISIS) is now in complete or partial control of more than a dozen Iraqi cities in Sunni-dominated areas of Iraq, which has set off an urgent debate over what the Iraqi government might do to reestablish control over its territory and whether the U.S. government should play a role in supporting such efforts.
Many contest the feasibility and ultimate effect of U.S. military intervention. Yet most seem to agree that the Iraqi military can’t go the fight alone, a position apparently confirmed by the Iraqi government’s formal request this week that the United States carry out airstrikes on Iraqi soil against ISIS rebels. Analysis has thus far tended to focus exclusively on the battlefield prospects of Iraqi forces against ISIS and how U.S. assistance might improve them. Yet the attitudes of local citizens within communities contested by ISIS towards the central government may prove critical to the outcome of the current fight.
In considering why ISIS has faced little local resistance as its forces have overrun towns throughout western and central Iraq, scholars have argued that the answer lies largely in discriminatory sectarian policies that are blamed for pushing Sunni leaders “toward supporting insurgency [while] radicaliz[ing] a Sunni community that might have been brought into the system following the civil war.” (See also here and here.) In other words, Maliki has a popularity problem within many of the areas now contested by ISIS.
Why is this important? Less than a decade ago, the United States sought to defeat an Iraqi insurgency that consisted of Al-Qaeda militants fighting alongside Iraqi Sunnis, who then also considered themselves disenfranchised with the collapse of the Sunni Baathist regime. Academic research on this period has found that the combination of military operations and the provision of social services was effective in reducing insurgent violence in affected areas by bringing local Sunnis back into the fold.
Critically, offensive military tactics alone were insufficient. As the authors of the study observe, “the silence of the population, of a substantial portion thereof, is critical for insurgencies. … For many years the residents of Anbar governorate knew who the insurgents were but lacked either the will or violent capacity to resist them. American and Iraqi security forces had the combat power but not the required information [to defeat the insurgents].”
A lack of basic services, including electricity, fuel and water, has remained a significant source of discontent for Iraqi citizens since U.S. forces withdrew in 2011. While such grievances are not sectarian in nature – protests have persisted throughout the country – the combination of sectarian policies and lack of social services may have laid conditions suitable for ISIS’ spread.
To assess how the provision of social services during the Iraq war affected insurgent violence, we examine the relationship between available electricity and insurgent attacks on coalition forces. Using daily electricity production data during the Iraq conflict collected and provided to us by the U.S. State Department for the purposes of this study, we find strong if preliminary evidence that increased electricity supply worked to reduce insurgent violence during the conflict. (For those interested, additional details on empirical strategy appear at the end of the piece.)
This effect is present in during the pre-surge period – when levels of insurgent and sectarian violence steadily increased across the country – as well as in the post-surge period, during which levels of both types of violence quickly tapered off. And the size of the effect is considerable.
Take for example the average national supply of electricity when the surge began in June 2007 (240 MW per 100,000 individuals) and the national average weekly number of violent incidents (0.18 per 100,000 individuals). A 10 percent increase in the supply of electricity from the mean in any given week is associated with a 6.8 percent reduction in violent incidents from the mean the following week, an additional 7.3 percent reduction two weeks later, and a further 6.7 percent incidents after the third week. On the other hand, a 10 percent decrease in the average number of violent incidents in a given week leads to a direct decline of 0.43 percent in the per capita supply of electricity in the following week but has no (statistically significant) effect thereafter. Consistent with a causal story, we find that the effect of electricity on violence is greatest during summer months, when, in the face of temperature well in excess of 100 degrees, demand for electricity is at its peak.
Obama has now embraced the position that conditions in Iraq cannot be remedied without a change in central government policies, arguing last Friday that the problem is “not solely or even primarily a military challenge. … Iraq’s leaders have been unable to overcome too often the mistrust and sectarian differences … ” And he has conditioned U.S. support to the Iraqi regime on “a serious and sincere effort by Iraq’s leaders to set aside sectarian differences … and account for the legitimate interests of all of Iraq’s communities.”
If history is any guide, to counter ISIS’ influence, Iraqi forces should find a strategy that combines offensive military tactics with meeting the needs of citizens within affected communities more effective than the former by itself. The task is likely to grow both more urgent and complicated as Islamic militants commit additional resources to offering social services within communities they have overtaken. Specifically, research suggests that the Iraqi government must not simply provide for citizens but outcompete the insurgents attempting to do so as well.
A note on the empirical strategy: We regress electricity provision on insurgent attacks against coalition and Iraqi forces using a dynamic panel by districts with weekly frequency. To account for the possible selection of greater electricity supply within predictably less violent areas of the country, we use the weekly differences in electricity supply and in the number of violent incidents. We fully expect insurgent violence to affect electricity levels, so our primary explanatory variables consist of four lags of electricity supply (in differences), whose collective effect we measure. We control for lagged increases in violence in order to sort out predictive trends in violence, and we add a series of relevant controls including U.S. and Iraqi government spending electricity infrastructure and U.S. troop levels, as well as pre-war fixed district characteristics, such as ethnicity, income per capita, and access to other public services.