The Regional Differences in Firm Entry in the Service Sector: Using the 1989-97 Longitudinal Establishment and Enterprise Micro-data File

University Of Baltimore, Baltimore MD

Investigators

Abstract

This project examines and analyzes regional differences in the recent rapid growth and evolution of service-producing businesses in the U.S. It will focus on the role of business entry in accounting for these regional differences. Most previous broad-based regional analysis has relied on County Business Patterns (CBP) data, which are limited in several ways. First, they provide only aggregate levels each year, and cannot track changes in a set of establishments over time; secondly, they lack information on firms (or enterprises); and thirdly, much of the data is suppressed for small or sparse areas for detailed industries. The Longitudinal Establishment and Enterprise Microdata (LEEM) file housed at the Center for Economic Studies (CES) covers all establishments with employees for the years 1989-1997, by linking the annual microdata underlying CBP. The service sector will be partitioned into five or six subsectors according to the markets they serve. Using the 1990 Labor Market Area (LMA) definitions to partition the U.S. into local economic units that are large enough to have business entry annually in most of the subsectors, we will prepare regional LMA level data for the services subsectors (including annual data on employment, firm entry and exit) from the LEEM. These data will be aggregated to broad regional and national levels to examine how the distribution of service employment among subsectors and among firm types has changed. Annual gross firm entry rates are calculated to measure levels of entrepreneurial activity in each subsector and region. 'Seasoned' entry rates will be calculated for the more limited set of new firms that have continued in business for at least three years. A model accounting for local differences in both gross and seasoned firm entry rates will be formulated and tested. Other regional data for explanatory variables and ratios will be constructed from publicly available county level databases. We expect to explain regional diversity in startup rates as a function of (1) the size and quality of the pool of potential entrepreneurs, while controlling for (2) the capital requirements, growth, and turbulence rates of the industry, and (3) the population growth rate, and the overall business growth rate of the region. This project's expansion of the investigators' previous research on business demographics using the LEEM would also serve to introduce this major new database for analysis of regional aspects of labor markets, economic development, business demography, and agglomeration.