Road Building and Violence

Infrastructure development has long been considered a privileged peace-building tool. The underlying rationale is that developing infrastructures furthers economic growth which in turn reduces violence. In this project we empirically test the validity of this assumption by examining the effect of road development on violence in Iraq between 2003 and 2016 with a novel dataset. To do so, we build data of the Iraqi road network over the period which we combine with data of investments in the road sector, violence and economic development, we find evidence that road-building is linked to an increase in violence. This result entices us to question the prevalent postulate related to infrastructure development and identify the conflicting mechanisms at play. Furthermore, closer analysis of the effect of road building specified by road types argues for a heterogeneous effect of road network on violence according to road size.
This paper builds on the empirical literature studying the impact of infrastructure-building and specifically of road construction on development (e.g. Alder et al., 2014; Donaldson and Hornbeck, 2016; Banerjee, Duflo, and Qian, 2012 and Burgess et al., 2015). The studies estimating the contribution of transportation infrastructure to regional or local development have built from Eaton and Kortum (2002)'s general equilibrium trade framework. The model predicts that new infrastructures (highways and railroads) decrease trade costs and increase overall income level.
Transportation infrastructure has also given rise to a stream of research which can be considered as more spatially and geographically focused but which nonetheless retains an economic dimension. These studies look at the impact of transportation on the territorial organization of economic activity as well as its impact on urbanization. In 2007, the analysis of limited access to American highways provided evidence of their construction's impact on central city population decline (Baum-Snow 2007). Recently, Baum-Snow et al. 2012 show that the spatial configuration of highways and railroads participated in the decentralization of Chinese cities population.
Our project contributes to a better understanding of development spending efficiency in a context of insurgency and argues for a more informed implementation of infrastructure development programs in countries under menace of insurgency.
We are planning to pursue an extension of this project with different dependant variables, notably electricity and power investments. The question would be to analyse the effect on violence of electricity coverage development.