Roads to development? The distributional effects of market access on wages: evidence from Kenya

This thesis analyses the heterogeneous and distributional effects of changes in trans- port infrastructure on city level wages in Kenya. To do this I develop a consistent theory, dichotomising industries into tradeable and non-tradeable, and adopting a suffcient statistic approach, using market access. To empirically evaluate these questions, I construct two novel data sets digitising historical Statistical Abstracts and Michelin maps between 1992 and 2010. This new data on wages by industry by year by city allows unique insights into the spatial distribution of economic ac- tivity in a Sub-Sahara African country: Kenya. I find that transport investment positively affects wages in Kenya, but that most of the impact is attributed to in- creases in tradeable sector wages, and thus to already relatively more developed cities. I also consider the direct effect of market access on development indicators and find consistent evidence of negative impacts. These results indicate that trans- port infrastructure projects may not be the silver bullet they are often purported to be. Finally, I consider a counterfactual scenario: building the proposed Mombassa- Malaba standard gauge railway. I find that although the project will raise wages in Kenya, it will also increase the spatial wage inequality already present, with the lion's share of benefits going to Nairobi and Mombassa.