@phdthesis{oai:ynu.repo.nii.ac.jp:00007681,
 author = {Mutsvangwa, Simba},
 month = {Sep},
 note = {This thesis presents three essays in international trade economics. It considers three cases in which international trade and trade costs are linked to conflicts. First chapter, “International Trade, Conflict and the Distance Puzzle: A Structural Gravity Model”, empirically analyzes the relationship between conflict (both intrastate and militarized interstate conflict) and international trade using a structural gravity model. Conflict can be expected to increase international trade cost by a big margin, and hence making it important to fully understand how it affects trade. Using year-by-year cross country regressions, this chapter also focuses on how the distance variable, which proxies the trade cost in gravity model of trade, behaves over time using global dataset from 1962 through 2001. This is analyzed when the effect of conflict is included in the trade cost function of the structural gravity model. This non decreasing distance effect in the gravity model is called the distance puzzle. Costs linked to conflict found to have a substantial negative effect on international trade. Militarized interstate conflict reduces trade by 61% (in tariff equivalent terms) and this is about double the effect of intrastate conflict which has 32%. This chapter also found that due to conflict, high income countries’ trade is affected more negatively than low income countries although they can quickly recover. However, on the other hand we found an unexpected distance trend. Although the distance puzzle is not completely solved by using the structural gravity model, the trade cost is stable, that is, over time it is neither increasing nor decreasing by a significant margin. Distance coefficient is constant under the structural gravity model while increasing when the standard gravity model is applied. This chapter concludes that the distance puzzle lies in the structure of the gravity model used and not in the omitted variables. Second chapter, co-authored with Craig, R. Parsons, “International Trade Cost and Conflict” tries to answer the question of how large is the cost of conflict on trade cost? The effect of conflict on trade may, at first seem apparent. Such violent disruption must surely reduce trade, ceteris paribus. Some empirical findings in the literature find a negative effect of conflict on trade. This chapter adds to the nascent literature in two ways. First, much of the literature is focused on the effect of conflict on bilateral trade. In this chapter, we separately examine the effect on trade by both intrastate conflict (civil war) and interstate conflict. Second contribution is the measure of trade costs used. We use the Novy (2013)1 measure of trade costs. The novelty of the trade model, which is based on micro-models of trade, is that what is important is to compare internal trade to international trade between any two countries. As such, we are measuring the effect of the conflicts on the “trade costs” between countries. We confirmed the negative effects of both types of conflict on trade. We find, in our sample of 110 countries, that interstate conflict raises bilateral trade costs by approximately 21.6% (in tariff equivalent terms), while intrastate conflict raises the trade costs by only 7%. As such, interstate is roughly three times as damaging to trade. Third chapter, “International Trade and Trade Cost using Non-CES Preferences: Translog Gravity Model” studies the effect of conflicts on trade. In contrary to most previous literature on this issue, this chapter empirically analyzes the relationship between conflict (Militarized Interstate Conflict) and international trade using a non-Constant Elasticity of Substitution (CES) based gravity model following Novy (2013)2. Like the first chapter, this section also analyzes the distance puzzle (sometimes called the missing globalization puzzle) of international economics, in this case, when translog gravity model is applied. Using a micro founded gravity equation which is based on a translog demand system this chapter sheds more light on the non-decreasing distance coefficient of the gravity model using data from 1970 through 2001. The missing globalization in the gravity model may be due to the CES preferences based part of the model. Trade is sensitive to trade costs if the exporting country provides a small share of the destination country’s imports. Using the non-CES gravity model, this paper found that the distance puzzle is solved while using the standard gravity model, the absolute distance coefficient is increasing. The results are the same despite the inclusion of conflict effect. In general, given that there is no significant difference in the absolute distance coefficients despite including the effects of conflicts, this shows that the distance puzzle is not present due to the omitted variables, in this case conflict effects. However, since the distance puzzle varnishes after using the translog gravity model it shows that the puzzle lies in the structure of the gravity model.},
 school = {横浜国立大学},
 title = {Essays on Conflict, International Trade and Trade Cost},
 year = {2016}
}