Conflict and Trade: What Can Tensors tell us about Trade Tensions?

• In Event: Globalization and Conflict

Abstract

Understanding the relationship between trade and conflict has never been more urgent given the U.S.’ recent retreat from its traditional role as a champion of free trade and the accompanying rise in tensions in its foreign relations. In this paper, we seek to move beyond merely verifying that a relationship between trade and conflict exists toward exploring the conditioning factors which may shape this relationship. To do so, we seek to make three contributions to the existing literature. First, we employ a multilinear tensor regression (MLTR) framework which allows us to not only control for, but to explicitly model changes in relational data over time. This method allows us to model the problem of endogeneity in the relationship between trade and conflict, which researchers in the literature have long struggled to resolve, head on. In using this network-based approach, we are able to estimate different dependencies that previously have not been explicitly modeled, such as reciprocity and transitivity, which are theoretically quite important to understanding trade and conflict more generally, among different dyads across time.

Second, by using data on conflictual statements made by state leaders and trade, we are also better able to capture a more fine-grained analysis of the relationship between trade and conflict. Virtually all previous studies which explore the relationship between trade and armed conflict (e.g. military interstate disputes, war), that is, when the level of tensions has already become quite severe. By investigating the relationship between trade and the tenor of words exchanged between countries, we hope to advance our understanding of the micro-mechanisms through which trade and conflict are related more broadly. To that end, we have collected monthly level data on country statements of conflictual statements from of 161 countries from March 2001 to December 2014.

Finally, we will explore the relationship between trade and conflict using three different conceptions of trade: openness (the ratio of trade to total economic output); vulnerability (proportion of trade between a pair of states as a share of total trade of each country) and gain (i.e., gains from trade). Each conceptualization captures different aspects of the trade conflict relationship but most papers deal with each of these issues separately and there seem to be few if any studies which provide a rigorous comparison of them. Doing so will advance our understanding of the many different ways trade and conflict can interact as well as the relative and absolute importance of any given conceptualization of trade dependence. Given that trade and conflict are two of the primary ways through which states interact with each other, furthering our understanding of how trade and conflict might relate to each other is fundamental for furthering our understanding international relations more broadly.

Authors

• Cindy Cheng, Bavarian School of Public Policy

• Shahryar Minhas, Michigan State University