Essays on development economics and Japanese economic history

Abstract

This thesis consists of three independent chapters on development economics and Japanese economic history. The first chapter analyzes the effect of railroad construction in the Meiji period (1868–1912) on technology adoption and modern economic development. By digitizing a novel data set that measures the use of steam engines at the factory level and determining the cost-minimizing path between destinations as an identification strategy, I find that railroad access led to the increased adoption of steam power by factories, which in turn induced structural change and urbanization. My results support the view that railroad network construction was key to modern economic growth in pre-First World War Japan. The second chapter analyzes the effect of time horizon on local public investment in the Edo period (1615–1868). I use a unique event in Japanese history during this period to identify the effect. In 1651, the sudden death of the executive leader of the Tokyo government reduced the transfer risk of local lords, especially for insiders, who supported the Tokyo government during the war of 1600. Using a newly digitized data set and a difference-in-differences strategy, I find that after 1651, regions owned by insiders increased the number of public projects more than regions owned by the other lords. I discuss other possible channels to interpret the effect of tenure risk, but I find no strong support for these alternative channels and conclude that the results support a longer time horizon effect. The third chapter provides more general background and a complete description of the data availability in Japan in the 17th–20th centuries, to discuss future research directions. It would aid reexamination of the history of Japan and other East Asian countries, which have experienced different economic and political paths.