Electronic Thesis and Dissertation Repository

Degree

Doctor of Philosophy

Program

Business

Supervisor

Beamish, Paul W.

Abstract

This dissertation examines the characteristics, profitability, and survival of multinational enterprise (MNE) foreign direct investment (FDI) in North American “global” cities (GCs), such as Los Angeles, New York, and Toronto. Across GCs and their metropolitan areas (Metros), MNEs often co-locate with their home country and co-industry peers in “co-ethnic” and “co-ethnic, co-industry” (CECI) clusters. Despite their substantial influence on the world economy GCs are relatively underexplored as location units of analysis in International Business (IB) research. Accordingly, I address three research questions. First, how do subsidiary and MNE characteristics differ between GCs, Metros, and other locations? Second, how does subsidiary profitability and survival differ between GCs, Metros, and other locations? Third, how does co-ethnic and CECI cluster membership influence subsidiary profitability and survival? For analysis, I use a sample comprising 2,863 unique Japanese subsidiaries in North America across 1,605 MNEs over the years 1990-2013. I apply a multi-level longitudinal analysis model and determine spatially significant clusters using geo-coding, proximal distance, and density analysis. In the first essay (Chapter 2), I use internalization theory and the eclectic paradigm to explain how subsidiary level FDI characteristics and MNE level assets may differ between GCs, Metros, and other locations. The results largely support my arguments. The second essay (Chapter 3) examines subsidiary profitability in GCs and Metros and co-ethnic and CECI clusters. I posit and find that subsidiary profitability aligns with location and ecosystem advantages. The third essay (Chapter 4) is an extension to Chapter 3 and examines subsidiary survival. For GCs and Metros, I find as hypothesized that the location drivers of profitability lead to higher exit rates. Different from my arguments, co-ethnic clusters have no effect on exit rates, and the positive impact of CECI clusters is limited to locations outside of GCs and Metros. My dissertation responds to calls for a fuller treatment of the global city phenomenon; and for bridging IB research with economic geography. It informs the eclectic paradigm at a sub-national level, adds to conceptual work on MNE clusters, and provides a large sample, longitudinal baseline to inform subsequent theoretical and empirical research.

Share

COinS