Does the Net Work? A Natural Experiment on the Connectedness of Ugandan Local Governments

My research pipeline investigates how the adoption of novel technologies affects firms’ behavior and entrepreneurship. My main project focuses on how firms respond to a stark technology enhancement of public service provision in urban Uganda. The idea is to ask how the digitization of public services may streamline the administrative architecture of local governments and how this affects business creation, growth and public service management. I address this question by looking at the construction of the National Data Transmission Backbone Infrastructure and e-Government Infrastructure (NBI/EGI) project, a massive Ugandan government investment in public service provision whose implementation lasted about a decade.At this stage, I have a neat identification strategy for estimating the causal impact of shifting public services to an e-system on firms’ output. By tracking down the implementation reports of the NBI/EGI investment and by establishing a direct link with the Ugandan Ministry of Information and Communication Technology, I created a geo-spatial dataset containing the treatment allocation across time at municipality level. For identification, I leverage two facts: (i) the NBI/EGI has been rolled out in four waves covering different regions of the country; (ii) the expiration dates of the ongoing telecommunication contracts of local government authorities when the municipality is served by the backbone network is quasi-random. I use (i) to estimate a multi-period Difference-in-Differences, while (ii) serves as an instrumental variable for e-service provision in a given municipality.I expect the introduction of this new technology to deeply affect the network structure of public service providers; also government authorities shall gain in e?iciency because of the automation of a set of bureaucratic tasks. As to firms, I expect changes over both the extensive and the intensive margins in the formal market: a more efficient public infrastructure reduces fixed costs for creating or perpetuating a business as well as for declaring it to the public authorities.The objective is now to access the complementary data sources necessary to implement the estimation and develop further the empirical analysis to trace out the mechanisms driving the treatment effects. I hope to do so as a visiting researcher at Columbia Business School (CBS), under the guidance of Jonas Hjort, who has invited me for the academic year 2020-21. I had the chance to work for Jonas prior grad school; he has guided me toward the PhD in Zurich and already provided me with important feedback on this project at an earlier stage. In addition, I believe the exposure to the Development group of Columbia would tremendously benefit my research given their expertise in firms and technology. The combination of the faculty from CBS and the Department of Economics makes Columbia the ideal cutting-edge environment for my visiting.