Every year, the world economy invests a large amount of resources to improve or develop transport infrastructure. How should these investments be allocated to maximize social welfare? In this proposal, I propose to develop and apply new methods to study optimal transport networks in general-equilibrium models of international trade, urban economics and economic geography. The methodology will build on recent work (Fajgelbaum and Schaal, 2017), in which my coauthor and I studied the network design problem in a general neoclassical trade framework.
In the first project, I develop a new framework to analyze optimal infrastructure investment in an urban setting. The model features people commuting between residential areas and business districts as well as a choice over the mode of transportation. We plan to evaluate the framework to historical data about specific cities.
In the second project, I propose and implement an new algorithm to compute optimal transport networks in the presence of increasing returns to transport, a likely prominent feature of real-world networks. The algorithm applies a branch-and-bound method in a series of geometric programming relaxations of the problem.
In the third project, I study the dynamic evolution of actual transport networks using satellite data from the US, India and Mexico. In the spirit of Hsieh and Klenow (2007), I use the model to measure distortions in the placement of roads between rich and poor countries.
In the fourth project, I study the inefficiencies and welfare losses associated with political economy frictions among governments and planning agencies. I use the model to identify inefficiencies and relate them to measures of institutions and political outcomes.
In the final project, I propose a new explanation behind the Zipf’s law distribution of city sizes. I show that Zipf’s law may result from particular topological properties of optimal transport networks that allocate resources efficiently in space.
Fields of science
Call for proposal
See other projects for this call