Do enterprise zones create jobs? Evidence from California’s enterprise zone program

https://doi.org/10.1016/j.jue.2010.01.002Get rights and content

Abstract

We use new establishment-level data and geographic mapping methods to improve upon evaluations of the effectiveness of state enterprise zones, focusing on California’s program. Because zone boundaries do not follow census tracts or zip codes, we created digitized maps of original zone boundaries and later expansions. We combine these maps with geocoded observations on most businesses located in California. The evidence indicates that enterprise zones do not increase employment. We also find no shift of employment toward the lower-wage workers targeted by enterprise zone incentives. We conclude that the program is ineffective in achieving its primary goals.

Introduction

State and federal enterprise zone programs are the principal means by which governments try to directly promote economic development in specific locations, typically economically-distressed urban areas. There is considerable debate over the effectiveness of enterprise zones in spurring job creation (e.g., Peters and Fisher, 2002). Evaluations of enterprise zone programs face several challenges, including precisely identifying the targeted areas, selecting appropriate comparison or control areas, distinguishing the effects of enterprise zones from other geographically-targeted policies, and choosing outcomes in line with program incentives and goals (e.g., Boarnet, 2001).

We evaluate the effectiveness of California’s enterprise zone program, using new data sources that permit us to meet many of these challenges. The first data source is detailed GIS maps we constructed of the precise boundaries of enterprise zones as they evolved over time. With maps of both initial designations and expansion areas, we define the control areas in multiple ways, allowing us to perform sensitivity tests and to examine whether spillovers affect our results. The second is the National Establishment Time-Series (NETS) database, which includes employment and location information on nearly all business establishments in California in the period 1992–2004. By constructing precisely geocoded location information and combining this information with GIS maps, we can measure employment, the number of establishments, and other characteristics of these establishments in each year, in each enterprise zone and in appropriate control areas. In addition, we incorporate information on other geographically-targeted policies to try to isolate the effect of the state enterprise zone program.

Section snippets

Limitations of previous research on enterprise zones

Most existing research evaluating the effects of enterprise zones assesses their effects on jobs, businesses, or zone residents. Typically, these studies compare outcomes like employment (e.g., Bondonio and Engberg, 2000, O’Keefe, 2004) or number of establishments (Dabney, 1991) across enterprise zones and comparable regions where zone incentives do not apply. The results vary across studies. Many studies fail to find employment effects of enterprise zones, although some of the work (e.g.,

California’s enterprise zone program

California’s enterprise zone program has multiple goals: attracting jobs and businesses and raising employment is one goal, while others include reducing poverty and unemployment and raising incomes in target areas.6 These multiple goals – job creation and improving residents’ circumstances – stem from the 1996 merger of two precursor programs that gave rise to the current enterprise zone program: the Enterprise

Data, mapping, and methods

The challenges faced by research on enterprise zones played a central role in shaping the methods and approaches we use in this paper. With regard to the first challenge – precise identification of zone boundaries – we digitally map California’s enterprise zones street-by-street rather than approximating by using census tracts, zip codes, etc. Of course the precise geographic contours of enterprise zones that we create are only useful if we can map business establishments or employment into

Enterprise zone employment and establishments in the context of the state’s economy

Table 3 presents descriptive information on the enterprise zones we study. Column (1) reports employment in each enterprise zone in our sample as of 2004. The zones are sorted from highest to lowest employment levels. As reported at the bottom of column (1), overall employment statewide in these enterprise zones is about 1.38 million, and employment in the control rings used in our empirical analysis (extending 1000 ft from the zone boundaries) is about 580,000. Overall, employment in the

Discussion and conclusions

Our analysis of California’s enterprise zone program cannot reject the hypothesis that the program fails to increase employment. We arrived at this conclusion after drawing precise enterprise zone boundaries digitally, mapping nearly all businesses in the state, accounting for other geographically-targeted policies, and comparing employment growth in enterprise zones with carefully considered control areas. We do not assess the effect of the program on unemployment or poverty, but it is hard to

Acknowledgments

We thank Marco Anderson, Eric Becker, Amy Ewing, Toni Feinstein, Matthew Gelbman, Ethan Jennings, Davin Reed, Mark Vasquez, and especially Jennifer Graves, Marisol Cuellar-Mejia, Ingrid Lefebvre-Hoang, and Smith Williams for outstanding research assistance. We also thank Toni Symonds and Frank Luera for many helpful discussions about California’s enterprise zone program, and Daria Burnes, Joel Elvery, Mark Ibele, Magnus Lofstrom, Suzanne O’Keefe, Michael Teitz, seminar participants at UCI and

References (38)

  • Richard V. Burkhauser et al.

    Who gets what from employment pay or play mandates?

    Risk Management and Insurance Review

    (2008)
  • Busso, Matias, Kline, Patrick, 2007. Do local economic development programs work? Evidence from the federal empowerment...
  • California Franchise Tax Board, 2006. Annual Report 2006. <http://www.ftb.ca.gov/aboutftb/annrpt/2006/2006AR.pdf>...
  • California Redevelopment Association, 2008. What can redevelopment do?...
  • A. Colin Cameron et al.

    Bootstrap-based improvements for inference with clustered errors

    Review of Economics and Statistics

    (2008)
  • Dan Y. Dabney

    Do enterprise zone incentives affect business location decisions?

    Economic Development Quarterly

    (1991)
  • Michael Dardia

    Subsidizing Redevelopment in California

    (1998)
  • Steven J. Davis et al.

    Job Creation and Destruction

    (1996)
  • David E. Dowall

    An evaluation of California’s enterprise zone programs

    Economic Development Quarterly

    (1996)
  • Cited by (0)

    1

    Fax: +1 415 291 4401.

    View full text