Shotaro Beppu

Transportation technology allows more market integration. The benefit could be unequal due to the adoption of this technology. This paper studies the rapid adoption of steamships in the late 19th century and its effect on the First Globalization and the Great Divergence. For this, this paper uses a novel deep-learning method to digitize historical shipping data to analyze the transition from sailing to steamships and the change in worldwide shipping patterns. Using the change in duration, this paper shows that the advent of steamships increased trade and growth on average. However, colonized countries experienced insignificant gains. To understand the mechanism, this paper incorporates differences in shipping technology to trade with heterogeneous firms. The estimate points out that the advantage of steamships might not have been captured due to the high cost of adoption for trade links involving colonies. This provides further insights into how shipping technology affects economic activity through trade and how the fixed cost of adoption matters in gains from such technology.

International trade and conflicts are key aspects of this world. In particular, weapons are traded across the world and civil wars take place in multiple regions. However, not much is known about how those two are related. This paper combines methods in international trade, network analysis, and literature in both empirical and theoretical conflict to state the importance of this overlooked relationship. This paper uses the shift-share design and network measure derived from the Cournot competition in multiple markets to find significant weapons reallocation to other countries after a ceasefire. This paper then tests insights from a bargaining model with an imperfectly observed claim to illustrate how this exogenous increase in weapons may lead to conflicts. Using the predicted change in weapons due to the reallocation as an instrument, an increase in weapons led to more conflicts. Transparency of weapons trade, however, can counteract this rise in conflict.

In recent decades the world has seen an increasing number of regional trade agreements in force. Coinciding with increasing tariffs between the US and China, whether this phenomenon promotes protectionism whereby countries trade less with countries outside or functions as a stepping stone for global free trade is a central concern for answering how international trade evolves in the future. This paper gives insight for answering this question by investigating the stability of such regional trading blocs. Inspired by research in the theory of strategic network formation, the model formulated in this paper shows that, contrary to main findings from canonical models, trading blocs are stable in possibly many cases. The results imply that the importance of countries not belonging to trading blocs will increase its importance in the future. Furthermore, changes induced by globalisation do not necessarily lead to global free trade.

Health and macroeconomic outcomes varied substantially across prefectures in Japan during the COVID-19 crisis. Using an estimated macro-epidemiological model as well as the idea of revealed preference, we compute the marginal rate of substitution (MRS) and the conditional trade-off curve between health and economic outcomes in each prefecture. We find that there is a large heterogeneity in the MRS as well as the location and shape of the conditional trade-off curve.

The progress of science increasingly relies on machine learning (ML) and machines work alongside humans in various domains of science. This study investigates the team structure of ML-related projects and analyzes the contribution of ML to scientific knowledge production under different team structure, drawing on bibliometric analyses of 25,000 scientific publications in various disciplines. Our regression analyses suggest that (1) interdisciplinary collaboration between domain scientists and computer scientists as well as the engagement of interdisciplinary individuals who have expertise in both domain and computer sciences are common in ML-related projects; (2) the engagement of interdisciplinary individuals seem more important in achieving high impact and novel discoveries, especially when a project employs computational and domain approaches interdependently; and (3) the contribution of ML and its implication to team structure depend on the depth of ML.

Using census and commuting matrix (LODES), I estimate agglomeration effect for cities in the Midwest using nonlinear estimation akin to RDD.

Using a pilot RCT study which made a portion of users switch their default search engine, I estimate how default setting matters in which search engine they use and how this effect differs by different search engines and browsers as well as perception in quality.