This article investigates the impact on economic activity of one of the most successful mobile phone-based financial innovations in the world. Combining data from the expansion of the mobile money agent network and light density data derived from night-time satellite imagery, I estimate the impact on local economic performance of the mobile money innovation in Kenya in the six years that followed its launch. The strategies that I pursue exploit the variation in local areas that gained access to mobile money services at different times as well as the high resolution of the data, which allows for the inclusion of hyper-local fixed effects. Areas with access to mobile money services are found to grow faster, especially if they are initially richer, urban, and connected to roads and banks. Access to mobile money services appears to be a complement to, rather than a substitute for, other alternatives that enable people to connect, trade, and allocate investments within their networks.
On the Dispensability of New Transportation Technologies: Evidence from the Heterogeneous Impact of Railroads in Nigeria, with Dozie Okoye and Roland Pongou. Submitted.
Exploring heterogeneity in the long-run impacts of a technology is a first step towards understanding conditions under which this technology is conducive to long-run economic development. This article shows that colonial railroads in Nigeria have long-lasting impacts on individual and local development in the North, but virtually no impact in the South neither in the short run nor in the long run. In the short run, this heterogeneous impact can be accounted for by differences in initial access to coastal ports and associated export markets. In the long run, it is explained by path-dependence, given that the railroads are no longer functioning. We highlight the fact that the railway had no impact in areas that had greater pre-railway access to ports of export, and that these areas adopted the railway at very low rates because railroads were more expensive compared to roads and rivers. Our analyses rule out the possibility that the heterogeneous impacts are driven by cohort effects, presence of major roads, early cities, or missionary activity, or by crude oil production.