Synergies in International Investment

In the increasingly interconnected global financial market, understanding incentives in large transactions has important implications for global stability and welfare. My research considers large international transactions in two contexts. The first context considered is currency markets, where dollar dumping announcements are analyzed in the context of other similar large scale exchange rate impacting events. I argue that dollar dumping announcements share many similarities with both central bank interventions and financial crises. Further, I argue that correct anticipation of these events by traders can lead to significant gains at the expense of less informed market participants. The second context looks to foreign direct investment choices in semiconductor factories, which typically have a minimum efficient scale of anywhere from ;I motivate the first essay, "Does Dollar Dumping Affect Relative Currency Values?," by addressing the issue that various reports in the media identified dollar dumping as the principal cause of the US dollar depreciation from 2000 to 2008. The study examines whether dollar dumping has an effect on potential overnight profits in foreign exchange markets via an event study and finds the effects to be considerable. A preliminary analysis shows successful anticipation of a dollar dumping event results in a mean overnight return of 7%. Further analysis shows that dollar dumping is an important source of exchange rate dynamics sharing many similarities with central bank interventions and economic crises.;The second essay, "Synergies in High Tech Production: A New Model of Complex FDI and Sourcing," analyzes a recent trend in semiconductor manufacturing. Namely, why have many semiconductor firms begun shifting their skilled-labor intensive fabrication facilities to Southeast Asia? I develop a partial equilibrium model of location choice with three stages of production and market incentives internal to the firm. My model predicts that given sufficient global demand, firms will locate the third stage of production, assembly, abroad in search of cheaper labor and the second stage of production, manufacturing, in a different location, usually an entirely different country, to minimize cross border transactions costs. I then confirm the relevance and validity of my model by running a logit estimation using country, industry, firm, facility, and assembly data on the semiconductor industry to provide support for third country low-cost labor seeking, in contrast to the typical third country product market seeking related factors, influencing location choice of semiconductor second stage production. Taken together, these models of different stages of production combined with small differences in cross country border related costs help to explain the counter intuitive movement of high-skilled labor stages of production to areas of lower high-skilled labor concentration.;The third and final essay uses a large longitudinal data set on the semiconductor industry to determine which, if any, of three major multinational enterprise paradigms, Dunning's Eclectic, Vernon's Product Cycle , or the Uppsala Learning to Internationalize, best characterize observed foreign expansion in semiconductors. Although elements of both Dunning's and Vernon's paradigms are apparent, it is clear that other factors are also at work. As the semiconductor industry is often generalized to encompass pharmaceuticals and other high-tech industries, this essay advocates for a revision of the theory of the multinational enterprise for high tech, research and development intensive industries. Furthermore, it argues that in a world of decreasing macro-level transactions costs, industry specific factors are increasingly prominent, rendering sweeping generalizations about the causes of FDI less useful.