Collaborative Research: The Clustering of U.S. Business Enterprises in the 1990s

University Of North Carolina At Chapel Hill, Chapel Hill NC

Investigators

Abstract

This project will utilize confidential enterprise-level data for the contiguous U.S. and point process modeling techniques to document basic trends in clustering of U.S. businesses between 1989 and 1999 (by sector, degree of technology intensity, size, etc.). The project will investigate whether enterprises cluster to take advantage of key externalities postulated in the scientific and policy literatures. In addition to a suite of descriptive spatial analysis techniques, the study uses two complementary methodologies, D-functions and spatial duration models, that employ case-control designs to isolate spillover effects from other factors influencing clustering. The study will explore whether clustering related to spillovers, at least in some industries or groups of industries, appears to be strengthening or weakening during the 1990s. A growing number of federal, state, and local economic development policy initiatives are premised on, or attempt to leverage, specific advantages enterprises gain from locating in close spatial proximity. Such policy initiatives include industry clusters, technology transfer, and business network programs. At the same time, other public policies and programs anticipate that, in the information age, telecommunications technologies make the geographic concentration of business establishments less necessary. The scientific and theoretical literature is undecided on whether geographic concentration is a less important strategic consideration for businesses today than in earlier industrial eras. This project will inform scientific and policy debates by providing important baseline empirical information about clustering in different U.S. sectors during the 1990s.