ࡱ > ? A < = > M x bjbj== , W W y l " / / / 8 0 T j0 < W 1 6 " 6 6 6 6 6 6 ZV \V \V \V \V \V \V $ W Y \ V E 6 6 6 6 6 V f7 6 6 V f7 f7 f7 6 4 6 6 T f7 6 ZV f7 f7 > K | L 6 1 0) + / 6 " K L V 0 W K =[ 7 L =[ L f7 "State Enterprise Zone Programs (EZP): A study of the challenges and conditions for maintaining a targeted economic development program"
State Politics and Policy Conference
Mark Cassell
Associate Professor
Kent State University
Robert Turner
Assistant Professor
Skidmore College
One of the enduring questions of state economic development is whether such policies can improve the economic prospects of poor people and communities. Such question have taken on a new sense of urgency with the transformation of the American welfare system towards a work-first system wherein the material well being of individuals is dependent on their ability to their ability to find gainful employment. For the past twenty-five years, many states have adopted enterprise zones to stimulate private sector investment in economically depressed areas. To date, scholars have extensively studied whether enterprise zones attract new investment and create jobs, as well as who benefits from such arrangements. While there are significant doubts about the effectiveness of these programs, virtually all scholars agree that enterprise zones are most effective if they are targeted at economically distressed areas.
However, few scholars have examined whether states have the political will to maintain the strict geographical targeting in enterprise zones. Enterprise zones are a form of redistributive economic development policy, wherein relatively few poor communities are eligible for special tax incentives to attract and retain companies. Economists contend such redistributive policies ensure that the state will maximize the economic impact of the economic development efforts ADDIN EN.CITE Bartik1991131Bartik, Timothy1991Who Benefits from State and Local Economic Development?KalamazooUpjohn Institute Press(Bartik 1991). However, scholars have noted that economic development policies are often awarded to maximize political support at the expense of providing targeted assistance ADDIN EN.CITE Dewar19982330Margaret Dewar1998Why Do State and Local Economic Development Programs Cause So Little Economic DevelopmentEconomic Development Quarterly121(Dewar 1998). Moreover, we know states differ in their political inclination and capacity to pursue redistributive welfare policies (Soss et al 2001) Evidence from Ohio and several other states suggests that after programs are created, states lose the ability or willingness to target assistance to socio-economically depressed areas ADDIN EN.CITE Cassell200324310Cassell, Mark2003Enterprise Zones: Who BenefitsClevelandPolicy Matters OhioOctoberTalanker200323210Talanker, Alyssa and Kate Davis with Greg LeRoy2003Straying From Good Intentions: How States are Weakening Enterprise Zone and Tax Increment Financing ProgramsWashington, DCGood Jobs FirstAugust(Cassell 2003; Talanker 2003). As a result, programs that were originally intended to provide tax incentives to businesses for locating in impoverished neighborhoods now often end up subsidizing economic development in affluent areas.
Although considerable scholarly attention has been devoted to assessing the economic impact of enterprise zones ADDIN EN.CITE Wilder19962180Margaret WilderBarry Rubin1996Rhetoric Versus Reality: A Review of Studies of State Enterprise Zone ProgramsJournal of the American Planning Association424473-491(Wilder and Rubin 1996), no research has been conducted on why the use of enterprise zones as a tool of state economic development has developed differently across states that use the policy. In this paper, we examine two questions. First, when do states adopt enterprise zones to benefit economically depressed regions. Second, when do they transform these policies into a state-wide program, thereby undermining its original redistributive intent. By examining the passage and implementation of enterprise zones, we also seek to contribute to the literature on the state policy innovation. Our results show that by examining the passage and implementation of enterprise zone policy, we can develop a more nuanced understanding of when ideology, economic conditions, and external pressures matter.
Enterprise Zones: A Policy Primer
The case of enterprise zones is an interesting example of cross-national policy learning and policy reinvention over time. The idea for enterprise zones came from a British scholar enamored with the economic dynamism of Hong Kong and Singapore under laissez fair economic policies ADDIN EN.CITE Hall1982830Hall, Peter1982Enterprise Zones: A JustificationInternational Journal of Urban and Regional Research63417-421(Hall 1982). They were subsequently championed by Stuart Butler ADDIN EN.CITE Butler19802731Butler, Stuart1980Enterprise Zones: Pioneering in the Inner CitiesWashington, DCThe Heritage FoundationButler19812291Butler, Stuart1981Enterprise Zones: Greenling the Inner CitiesNew YorkUniverse(1980; 1981) of the Heritage Foundation and were adopted by the Reagan administration as the cornerstone of its urban policy ADDIN EN.CITE Butler19913097Butler, Stuart1991The Conceptual Evolution of Enterprise ZonesRoy E. GreenEnterprise Zones: New Directions in Economic DevelopmentNewbury Park, CASage Publications27-40(Butler 1991). The initial policy idea called for removing the constraints of government regulation and taxation from economically depressed urban areas and letting market forces generate economic growth. However, partisan conflict at the national level prevented any significant policy action until 1993 when the federal Empowerment Zone and Enterprise Community Act was passed under the Clinton Administration and a Democratic Congress.
Despite the partisan gridlock at the national level, the number of states employing enterprise zones has increased dramatically from 0 in 1982 to 42 in 2002, leading one Reagan official to call it the federal governments most successful nonprogram (Cited in Mossberger 2000, p. 54). According to an estimate in 1995, there were approximately 2,840 state zones across the United States ADDIN EN.CITE Wilder19962180Margaret WilderBarry Rubin1996Rhetoric Versus Reality: A Review of Studies of State Enterprise Zone ProgramsJournal of the American Planning Association424473-491(Wilder and Rubin 1996). Moreover, state enterprise zone expenditures are one of the largest component of state economic development budgets ADDIN EN.CITE Peters2002631Alan H. PetersPeter S. Fisher2002State Enterprise Zones: Do They Work?KalamazooW.E. Upjohn Instituteenterprize zones(Peters and Fisher 2002). Why did enterprise zones spread so quickly at the state level despite the stalemate at the national level? Two reasons. First, policy entrepreneurs in the Reagan administration were very active in encouraging states to adopt enterprise zones though appearances with various state organizations and economic development agencies (Mossberger 2000, Ch. 3). Second, the programmatic component behind the enterprise zone label changed dramatically, which in turn, attracted political support across the ideological spectrum.
The original theory behind enterprise zones stressed the importance of lowering taxes, limiting regulation, and abandoning minimum wage laws to stimulate indigenous entrepreneurial growth. In 1979, the Illinois legislature passed the first enterprise zone bill in the United States. The bill would have eliminated minimum wage laws, created right to work provisions, reduced zoning restrictions, and curbed health and safety codes in the zones ADDIN EN.CITE Mier19822860Mier, Robert Scott Gelzer1982State Enterprise Zones: The New Frontier?Urban Affairs Quarterly18Sept39-52(Mier and Gelzer 1982). Not surprisingly, the bill was strongly opposed by labor, environmental, and civil rights organizations and was vetoed by governor. Subsequently, enterprise zone policy entrepreneurs shifted the programmatic content of enterprise zone proposals towards a more familiar and politically acceptable set of economic development policies. Instead of laissez faire, states pursued a more interventionist strategy of offering tax incentives to retain or attract manufacturing firms combined with providing public funds for infrastructure revitalization, community building, and greater access to social services. The programmatic compromise created a strange bedfellows political coalition between free market conservatives who are attracted to the laissez fair ideas behinds enterprise zones and urban interests looking to rebuild distressed urban communities ADDIN EN.CITE Guskind19892420Guskind, Robert1989Zeal for the ZonesNational Journal1358-62June 3Gunn19932610Gunn, E. M.1993The Growth of Enterprise Zones, A Policy TransformationPolicy Studies Journal21432-449Wolf19902800Wolf, Michael Allen1990Enterprise Zones: A Decade of DiversityEconomic Development Quarterly413-14enterprize zones(Guskind 1989; Wolf 1990; Gunn 1993). The only common element between the old and new version of enterprise zones was the policy goal, reducing poverty in economically distressed areas (Mossberger 2000).
The exact programmatic components of each states enterprise zone program varied. The typical state program included a mix of investment tax credits, job tax credits, sales tax exemptions or credits, and property tax abatements for firms which located in the zone. Some state programs attempted to reduce the cost of capital with tax credits for investment or property tax abatements while others attempted to reduce the cost of labor through subsidizing jobs and training expenses ADDIN EN.CITE Peters20032347Peters, Alan and Peter Fisher2003Enterprise Zone Incentives: How Effective Are They?Sammis B. White, Richard D. Bingham, and Edward W. WillFinancing Economic Development in the 21st CenturyArmonk, NYM.E. Sharpe(Peters 2003). States also varied in which level of government paid for the program, the local government through property tax abatements or the state through state corporate income tax credits. States also varied in the extent to which assistance to the economically depressed area consisted entirely of incentives to firms or provided additional resources to pay for improved infrastructure or social services ADDIN EN.CITE Erickson19912377Rodney EricksonSusan Friedman1991Comparative Dimensions of State Enterprise Zone PoliciesGreen, Roy E.Enterprise Zones: New Directions in Economic DevelopmentNewbury ParkSage Publications(Erickson and Friedman 1991). Finally, states varied in how actively they managed the local zone, with some appointing a local zone manager and requiring the development of a strategic plan for the zone whereas others did not ADDIN EN.CITE Britnall19912357Michael BritnallRoy E. Green1991Framework for a Comparative Analysis of State-Administered Enterprise Zone ProgramsRoy E. GreenEnterprise Zones: New Directions in Economic DevelopmentNewbury ParkSage Publications(Britnall and Green 1991). While states varied in the types of incentives, the level of additional services, and the degree of local involvement, each state program directed their assistance at economically depressed areas of the state.
The evaluation on whether enterprise zones generate new investment and employment is decidedly mixed, despite the plethora or research ADDIN EN.CITE Wilder19962180Margaret WilderBarry Rubin1996Rhetoric Versus Reality: A Review of Studies of State Enterprise Zone ProgramsJournal of the American Planning Association424473-491Rubin19891820Barry RubinMargaret Wilder1989Urban Enterprise Zones: Employment Impacts and Fiscal IncentivesJournal of the American Planning Association554418-431Papke19932307Papke, Leslie1993What Do We Know About Enterprise ZonesJames PoterbaTax Policy and the EconomyCambridge, MANational Bureau of Economic Research and MIT Press37-49(Rubin and Wilder 1989; Papke 1993; Wilder and Rubin 1996). Public assessments of enterprise zones by state agencies have found a consistent relationship between enterprise zones and increased job growth and investment ADDIN EN.CITE Wilder19962180Margaret WilderBarry Rubin1996Rhetoric Versus Reality: A Review of Studies of State Enterprise Zone ProgramsJournal of the American Planning Association424473-491(Wilder and Rubin 1996). More rigorous econometric models of enterprise zones have found zones have no positive impact on housing market, income, or employment outcomes ADDIN EN.CITE Engberg1999530John EngbergRobert Greenbaum1999State Enterprise Zones and Local Housing MarketsJournal of Housing Research102163-187(Engberg and Greenbaum 1999), the value of incentives or program features have no impact on employment growth ADDIN EN.CITE Bondonio20002310Daniele BondonioJohn Engberg2000Enterprise Zones and Local Employment: Evidence from States' ProgramsRegional Science and Urban Economics305519-49(Bondonio and Engberg 2000), and few poor residents actually get the jobs ADDIN EN.CITE Peters20021571Alan PetersPeter S. Fisher2002State Enterprise Zone Programs: Have They WorkedKalamazoo, MIW.E. Upjohn Institute for Employment ResearchPeters20021571Alan PetersPeter S. Fisher2002State Enterprise Zone Programs: Have They WorkedKalamazoo, MIW.E. Upjohn Institute for Employment ResearchGreenbaum20032810Greenbaum, Robert T.Engberg, John B.2003The Impact of State Enterprise Zones on the Establishment of Urban Manufacturing EstablishmentsJournal of Policy Analysis and Management232315-339enterprize zones(Peters and Fisher 2002; Greenbaum and Engberg 2003). These studies validated one of the original critiques of enterprise zones, that tax incentives alone cannot make up for the locational disadvantages of economically depressed areas that discouraged businesses from investing there in the first place ADDIN EN.CITE Dabney1991420Dabney, Dan1991Do Enterprise Zone Incentives Affect Business Location DecisionsEconomic Development Quarterly54325-334(Dabney 1991).
Clearly, some enterprise zones work, others dont. A review of state sponsored enterprise zone programs found estimates of employment growth ranged from 0 3,076 while estimates of retained jobs ranged from 0-10,890 ADDIN EN.CITE Wilder19962180Margaret WilderBarry Rubin1996Rhetoric Versus Reality: A Review of Studies of State Enterprise Zone ProgramsJournal of the American Planning Association424473-491(Wilder and Rubin 1996). Some have suggested the positive impact of enterprise zones is because of a policy placebo effect, where the business community takes the mere existence of the zone program as a signal that the city and state have made a commitment to the zone area. That perceived commitment has a positive influence on decisions about investment and location ADDIN EN.CITE Wilder19962180Margaret WilderBarry Rubin1996Rhetoric Versus Reality: A Review of Studies of State Enterprise Zone ProgramsJournal of the American Planning Association424473-491(Wilder and Rubin 1996). In short, it seems fair to conclude that enterprise zones may have a positive impact on employment growth, although the size of this effect is difficult to ascertain.
Although considerable scholarly attention has been devoted to assessing the economic impact of enterprise zones ADDIN EN.CITE Wilder19962180Margaret WilderBarry Rubin1996Rhetoric Versus Reality: A Review of Studies of State Enterprise Zone ProgramsJournal of the American Planning Association424473-491(Wilder and Rubin 1996), no research has been conducted on why states use of enterprise zones differs. Enterprise zones are part of a larger question over how states should distribute their scarce economic development assistance within the state. The central policy question is whether they should adopt a balanced growth or targeted approach to economic development ADDIN EN.CITE Eisinger1988501Eisinger, Peter1988The Rise of the Entrepreneurial StateMadisonUniversity of Wisconsin Press(Eisinger 1988). Under a balanced growth strategy, state development assistance is made available to every community in the state, rather than limited to the neediest jurisdictions ADDIN EN.CITE ACIR197924510ACIR1979The States and Distressed CommunitiesWashington, DCAdvisory Commission on Intergovernmental Relations(ACIR 1979). Proponents of a balanced growth approach argue it makes sense economically and politically. Economic development efforts should attempt to maximize economic growth in the state, and economically distressed areas are viewed as a poor investment for the state ADDIN EN.CITE ACIR198128410ACIR1981Regional Growth: Interstate Tax CompetitionWashington, DCAdvisory Commission on Intergovernmental Relations(ACIR 1981). Similarly, a balanced approach premised on inclusiveness helps avoid political charges of favoring upstate versus downstate, or urban versus rural interests ADDIN EN.CITE Eisinger1988501Eisinger, Peter1988The Rise of the Entrepreneurial StateMadisonUniversity of Wisconsin Press(Eisinger 1988).
By contrast, proponents of a targeted approach contend states should focus their economic development policies on economically depressed areas. The underlying justification for a targeted strategies, like enterprise zones, is twofold. First, the transformation of the American economy from a manufacturing and goods to a service economy has resulted in a spatial mismatch where jobs are in suburban areas and disadvantaged urban minorities are trapped in declining labor markets in the inner city ADDIN EN.CITE Kasarda19902380Kasarda, John D.1990City Jobs and Residents on a Collision Course: The Urban Underclass DilemmaEconomic Development Quarterly44(Kasarda 1990). The result is geographically concentrated pockets of economically distressed areas. Second, economists suggest the economic benefits of creating jobs are greater if the economic development policies create jobs in poor, high-unemployment communities ADDIN EN.CITE Bartik1991131Bartik, Timothy1991Who Benefits from State and Local Economic Development?KalamazooUpjohn Institute Press(Bartik 1991). An enterprise zone program that is effectively targeted on a economically depressed area will disproportionately benefit low income individuals.
The decision to use enterprise zones to geographically target assistance at poor areas in a state represented a major shift in the political and policy debate over how states should allocated their economic development resources within the state. Historically, states economic development programs have resembled the balanced growth approach ADDIN EN.CITE Eisinger1988501Eisinger, Peter1988The Rise of the Entrepreneurial StateMadisonUniversity of Wisconsin Press(Eisinger 1988). The majority of state economic development policies, such as capital subsidies and labor policies are available to eligible firms anywhere in the state. While a few states in the 1970s and early 1980s did have some geographically targeted programs such as industrial revenue bonds, industrial parks, and tax increment financing, these programs did not use explicit and rigorous economic distress criteria in determining eligibility. As a result, prior to enterprise zones, most geographically targeted economic development programs were not explicitly targeted at helping economically depressed places, and by extension poor people.
By contrast, state enterprise zones used explicit criteria for determining whether an area was eligible for enterprise zone designation to target their economic development efforts to the most needy areas. States economic distress criteria varied from the poverty rates, unemployment, median income, population decline, welfare caseloads, or low tax capacity. ADDIN EN.CITE HUD199224410HUD1992State Enterprise Zone UpdateWashington DCUS Department of Housing and Urban Development28-31(HUD 1992). Moreover, most states established a competitive rather an automatic designation process for conferring enterprise zone status. While states did not abandon their statewide incentives and balanced growth strategy, the adoption of enterprise zones signified that states were making careful targeting an important complementary strategy (Eisinger 1988, 189). The adoption of enterprise zones represented a significant change in a states practice and philosophy of economic development.
However, recent evidence suggests that state enterprise zones have shifted from a targeted program aimed at helping poor communities to statewide business attraction and retention program by easing the economic distress criteria and increasing the number of zones in the state. Studies of enterprise zone programs at the local level in Louisville ADDIN EN.CITE Lambert20012590Thomas E. LambertPaul A. Coomes2001An Evaluation of the Effectiveness of Louisville's Enterprise ZoneEconomic Development Quarterly152168-180(Lambert and Coomes 2001) and Philadelphia ADDIN EN.CITE Wallace19992600Wallace, Sherri Leronda1999A Case Study of the Enterprise Zone Program: "EZ" Avenue to Minority Economic Development?Economic Development Quarterly199259-65(Wallace 1999) found local governments enterprise zones are no longer targeted at helping blighted urban neighborhoods and employing the disadvantaged, but are being implemented as a de facto city-wide business retention and expansion strategy. A recent study of enterprise zones found that virtually every state has weakened the eligibility requirements to permit their use in non-blighted or affluent areas ADDIN EN.CITE Talanker200323210Talanker, Alyssa and Kate Davis with Greg LeRoy2003Straying From Good Intentions: How States are Weakening Enterprise Zone and Tax Increment Financing ProgramsWashington, DCGood Jobs FirstAugust(Talanker 2003). As a result, programs that were originally intended to provide tax incentives to businesses for locating in impoverished neighborhoods now often end up subsidizing economic development in affluent areas ADDIN EN.CITE Cassell200324310Cassell, Mark2003Enterprise Zones: Who BenefitsClevelandPolicy Matters OhioOctober(Cassell 2003). Scholars of other economic development policies have noted that incentives are often awarded to maximize political support at the expense of helping economically distressed areas ADDIN EN.CITE Lowi19641190Lowi, Theodore1964American Business, Public Policy, Case-Studies, and Political TheoryWorld Politics164Dewar19982330Margaret Dewar1998Why Do State and Local Economic Development Programs Cause So Little Economic DevelopmentEconomic Development Quarterly121(Lowi 1964; Dewar 1998). In other words, states and localities are increasingly abandoning the original targeted approach of enterprise zones to aid economically depressed areas in favor of a statewide, balanced growth approach.
The Creation and Expansion of State Enterprise Zone Programs
The evolution of state enterprise zone programs raises two interesting questions. When do states create enterprise zone proposals programs targeted to help poor places, and when do they change their policies from a targeted program towards a balanced growth program benefiting more affluent areas. What is of interest to us is not whether enterprise zones help poor people or poor places, but when do states make the political decision to aid them at the expense of other areas? The data on the creation of an enterprise zone program in a state and the expansion of the number of enterprise zones were taken from a dataset collected by John Engberg and Fisher and Peters ADDIN EN.CITE Fisher2002631Alan H. PetersPeter S. Fisher2002State Enterprise Zones: Do They Work?KalamazooW.E. Upjohn Instituteenterprize zonesEngberg1999530John EngbergRobert Greenbaum1999State Enterprise Zones and Local Housing MarketsJournal of Housing Research102163-187(Engberg and Greenbaum 1999; Peters and Fisher 2002) and supplemented with our analysis of some of the missing states. It includes information on the year the state adopted enterprise zones and expanded the number of zones in the state for every state between 1982 and 1996.
Four Competing Models of Enterprise Zone Policies
State policy innovation and change is theorized to be the product of internal and external pressures ADDIN EN.CITE Berry19982627Francis Stokes BerryWilliam D. Berry1998Innovation and Diffusion Models in Policy ResearchPaul A. SabatierTheories of the Policy ProcessBoulder, COWestview Press169-200(Berry and Berry 1998). Internal pressures include the views of political elites, changes in public opinion, the strength of the advocacy coalition, and socio-economic conditions. External influences are hypothesized to include policy learning or economic competition from neighboring states. Drawing upon this body of research, we develop four sets of hypotheses to explain the creation and subsequent proliferation of enterprise zones. Partisan or ideological models would suggest that changes in a states ideological or partisan makeup should lead to a change in economic development policy. Socioeconomic models suggest state policy choices are the product of societal actors response to changing economic and social conditions. Institutional models would suggest that differing institutional arrangements privilege certain interests in their ability to pursue their policy goals. Finally, an external model suggests that policy changes in neighboring states affect a states policy decisions ADDIN EN.CITE Walker19692790Walker, Jack L.1969The Diffusion of Innovations Among the American StatesAmerican Political Science Review67880-99diffusion(Walker 1969). Each is laid out in greater detail below.
Partisan or Ideological Model
To date, most assessments of state economic development policies have found economic development is a bipartisan policy issue, pursued equally by both parties ADDIN EN.CITE Eisinger1988501Eisinger, Peter1988The Rise of the Entrepreneurial StateMadisonUniversity of Wisconsin PressCable19982500Greg CableRichard Feiock1998The Adoption of State Economic Development Programs: An Event History AnalysisJournal of Politics(Eisinger 1988; Cable and Feiock 1998). Explanations for the lack of partisan differences point to one of two sets of explanations. Structuralist arguments point to the fierce competition among states for investment, jobs, and workers in an era of highly mobile capital, diminished federal funding, and increased international competition ADDIN EN.CITE Peterson19812391Peterson, Paul1981City LimitsChicagoUniversity of Chicago PressPeterson19952991Peterson, Paul1995The Price of FederalismWashington, DCBrookings(Peterson 1981; Peterson 1995). Political explanations point to elected officials to pursue policies designed to provide immediate economic and electoral gratification through credit-claiming ADDIN EN.CITE Eisinger1995520Eisinger, Peter1995State Economic Development Policy In The 1990s: Politics and Policy LearningEconomic Development Quarterly9146-158Wolman19882970Wolman, Harold1988Local Economic Development Policy: What Explains the Divergence Between Policy Analysis and Political BehaviorJournal of Urban Affairs1019-28Elkin19873021Elkin, Stephen L.1987City and Regime in the American RepublicChicagoUniversity of Chicago PressFeiock19863030Feiock, RichardClingermayer, James1986Municipal Representation, Executive Power, and Economic Development Policy AdoptionPolicy Studies Journal15211-30(Feiock and Clingermayer 1986; Elkin 1987; Wolman 1988; Eisinger 1995). However, economic development policies have important distributional consequences in terms of who benefits. Even policies which produce similar levels of economic growth can have significant distributional differences for who benefits ADDIN EN.CITE Leatherman20042417Leatherman, John C. Marcouiller, David W.2004Regional Economic Modeling and the Study of Distributional IssuesLaura Reese and David FasenfestCritical Evaluation of Economic PoliciesDetroitWayne State University Press144-72(Leatherman and Marcouiller 2004). Enterprise zones are an important example of an economic development policy with potentially redistributive implications.
While the intellectual forefathers of state enterprise zone programs were laissez faire conservatives drawn to its free market orientation, we assume that the liberals and Democrats are most likely to be supportive of enterprise zone programs at the state level because they are targeted at helping economically depressed areas. We predict Democrats would be more supportive of making enterprise zones a component of the states economic development strategy for political, ideological and policy reasons. The political base of state Democratic parties is in cities which would benefit from the passage of enterprise zone legislation. Ideologically, targeted economic development policies, like enterprise zones, are viewed as redistributive rather than developmental by policy-makers in nature ADDIN EN.CITE Tao19993080Tao, Jill L.Feiock, Richard1999Directing Benefits to Need: Evaluating the Distributive Consequences of Urban Economic DevelopmentEconomic Development Quarterly13155-65(Tao and Feiock 1999). The Democratic party has historically been more supportive of redistributive policies to support the interests of low income individuals than the Republican party. From a policy perspective, while Democrats would prefer not to offer tax incentives to corporations, they realize that corporate investment must be induced not commanded ADDIN EN.CITE Lindblom19771141Lindblom, Charles1977Politics and Markets: The World's Political-Economic SystemsNew YorkBasic Books(Lindblom 1977) and thus attempt to limit these incentives to benefit the economically distressed for both policy and political reasons. In a speech calling for the creation of more federal enterprise zones, President Clinton exemplified the Democratic perspective on the redistributive and political nature of enterprise zones, by stating that his enterprise zone program had created jobs and that "[t]he best poverty program, the best anticrime program, the best urban program is still a job for every person who will work ADDIN EN.CITE Clinton19992895Clinton, William J.1999What They're Saying About EZsEZ GAZETTE: A Newsletter on Empowerment Zones, Enterprise Communities, and Enterprise ZonesRichmondFebruaryhttp://law.richmond.edu/ezproj/feb99/what.htm(Clinton 1999). In contrast, state Republican parties are assumed to prefer that the states economic development strategy reflect the balanced growth approach to economic development, as opposed to targeting them for economically distressed areas. Such a strategy would spread the enterprise zone benefits to the more affluent portions of the state which also coincide with their political base.
We include two measure of partisan or ideological control of state government in our models. First, we developed a measure of the partisan control of state government by determining the percentage of the three lawmaking institutions, governor, senate, and assembly, controlled by Democrats. A "1" indicates absolute Democratic control of state government, while a "0" represents absolute Republican control of state government. Second, we also included a measure of state government ideology that is weighted based on the relative power of the governor, and the two major parties in each house of the legislature ADDIN EN.CITE Berry19982880Berry, William D.Ringquist, Evan J.Fording, Richard C.Hanson, Russell L.1998Measuring Citizen and Government Ideology in the American StatesAmerican Journal of Political Science42337-348(Berry, Ringquist et al. 1998). For the adoption model, we assume that greater Democratic control of state government and/or liberal state government ideology increases the likelihood a state will adopt an enterprise zone program. For the expansion model, we hypothesize that states with a more conservative state ideology will be more likely to increase the number of enterprise zones in the state, thereby changing the program from a targeted one benefiting poor areas to a balanced growth one benefiting the entire state.
Socioeconomic Model
Socioeconomic explanations would suggest that the decision to create or expand a state enterprise zone program is a result of underlying social and economic conditions. Scholars have noted that urban ADDIN EN.CITE Green19892530Gary P. GreenArnold Fleishmann1989Analyzing Local Strategies for Promoting Economic GrowthPolicy Studies Journal17557-573Clingermayer19902550James ClingermayerRichard Feiock1990The Adoption of Four Economic Development Policies in Large CitiesPolicy Studies Journal18539-52(Green and Fleishmann 1989; Clingermayer and Feiock 1990) and state ADDIN EN.CITE Cable19982500Greg CableRichard Feiock1998The Adoption of State Economic Development Programs: An Event History AnalysisJournal of PoliticsBowman19882830Ann O'M Bowman1988Competition for Economic Development Among Southeastern CitiesUrban Affairs Quarterly23,511-27Rubin19872510Herbert RubinIrene Rubin1987Economic Development Incentives: The Poor Pay MoreUrban Affairs Quarterly22.32-62Brace1994191Brace, Paul1994The Political Economies of the American StatesWashingtonCQ Press(Rubin and Rubin 1987; Bowman 1988; Brace 1994; Cable and Feiock 1998) decisions makers are more likely to adopt new economic development programs under conditions of economic duress, although these previous studies did not explicitly distinguish between redistributive versus developmental policies. We hypothesize that as economic conditions for a state worsen, officials are more likely to respond by enacting enterprise zone programs. We include a short and long term measure of state economic performance. For short term, we use the annual percentage of the workforce in a state that is officially considered by the federal government as unemployed. For long term economic performance, we use the percent increase in the total number of manufacturing jobs from 1982-1995. While both variables measure state level economic conditions, we assume any downturn in either variable is likely to hit economically distressed areas harder than the rest of the state.
We also include a measure of the percentage of state population that lives in an urban area. We assume that the greater the percentage of the state which lives in urban areas, there will be more political pressure on officials to design economic development programs to address their needs. The larger the urban population, the more likely it is a state will adopt an enterprise zone program and increase the number of enterprise zones.
Institutional Model
The decision to target a states economic development resources on the most needy areas of the state, at the expanse of the majority is likely to promote institutional conflict between the governor and the legislature. This conflict is likely to be most evident in the decision to increase the number of enterprise zones in the state. The legislature would prefer to increase the number of zones to make benefits available in every district ADDIN EN.CITE Mossberger20002701Karen Mossberger2000The Politics of Ideas and the Spread of Enterprise ZonesWashington, DCGeorgetown University Press(Mossberger 2000). However, governors, since they are held accountable by voters for a host of policy outcomes are likely to prefer a more efficient use of the states enterprise zone resources by targeting only the most needy areas. However, governors vary in their institutional capacity to meet their policy goals. We assume that greater executive centralization and state institutional capacity is associated with greater capacity to target economic development resources ADDIN EN.CITE Johnson1982971Chalmers Johnson1982MITI and the Japanese MiracleStanford, CAStanford University PressKatzenstein1978987Katzenstein, Peter1978Conclusion: Domestic Structures and Strategies of Foreign Economic PolicyPeter J. KatzensteinBetween Power and Plenty: Foreign Economic Policies of Advanced Industrialized StatesMadison, WIUniversity of Wisconsin PressZysman19832281Zysman, John1983Governments, Markets, and GrowthIthacaCornell University PressReich19831661Reich, Robert1983The Next American FrontierNew YorkNew York Times Books(Katzenstein 1978; Johnson 1982; Reich 1983; Zysman 1983). Executive centralization measures a governor's level of institutional power ADDIN EN.CITE Bowman19882720Bowman, Ann O'MRichard C. Kearney1988Dimensions of State Government CapabilityWestern Political Quarterly412341-62(Bowman and Kearney 1988). Our measure of state institutional capacity is also drawn from an overall measure of bureaucratic capacity based on levels of staffing and spending ADDIN EN.CITE Bowman19882720Bowman, Ann O'MRichard C. Kearney1988Dimensions of State Government CapabilityWestern Political Quarterly412341-62(Bowman and Kearney 1988). We hypothesize that states with greater executive centralization and institutional capacity will be better able to prevent the expansion in the number of enterprise zones in the state.
External Model
Finally, a states enterprise zone policy could be influenced by their neighboring states policies through competition between the states for jobs and investment. As Berry and Berry note, it is when policy adoptions are attempts to compete with other states that the likelihood of regionally focused, rather than nationally based, diffusion seems greatest (1998, 175). Since enterprise zones are primarily viewed as a redistributive policy rather than a developmental policy, we do not expect that the neighboring states having an enterprise zone will affect the adoption decision. However, we expect that as more of its neighbors adopt an enterprise zone program, a state will be more likely to expand the number of zones to attract the investment. Competitive pressure is measured using a dummy variable where a 1 indicates a state borders another state with an enterprise zone program and a 0 indicates a state borders no state with an enterprise zone program.
Modeling The Passage of Enterprise Zone Legislation Using Event History Analysis
To assess our four sets of hypotheses on the passage of state enterprise zone legislation, we performed event history analysis. In the case of adoption, the data consist of one observation for each state for each year from 1982 to 1996 or the year in which the state adopted the program. We choose 1982 as the initial year since this was the first year the a state enacted an enterprise zone policy. The event of interest is the passage of enterprise zone legislation for each state-year. The passage variable equals 0 for every year prior to passage and 1 for the year of passage. No more state- years are included in the dataset for the state after the year of passage. In event history analysis, the risk set is the individuals in the sample that are at risk of an event occurring. The risk set decreases over time as individuals in the sample experience the event ADDIN EN.CITE Allison19822917Allison, Paul D.1982Discrete-Time Methods for the Analysis of Event HistoriesSamuel LeinhardtSociological Methodology 1982San FranciscoJosey-BassAllison19842901Allison, Paul D.1984Event History Analysis: Regression for Longitudinal Event DataBeverly HillsSage(Allison 1982; Allison 1984). We also include an annual dummy variable to control for maturation effects and ensure the slope estimate for a regional effect is not biased in a positive direction ADDIN EN.CITE Beck19982560Nathan BeckJonathan KatzRichard Tucker1998Taking Time Seriously: Time Series Cross Section Analysis with a Binary Dependent VariableAmerican Journal of Political Science421260-88Allison19842901Allison, Paul D.1984Event History Analysis: Regression for Longitudinal Event DataBeverly HillsSage(Allison 1984; Beck, Katz et al. 1998).
Because are data were a cross-sectional pooled time series for all states that had adopted the enterprise zone program, we use the xtprobit procedure in the statistical software program, STATA. In a regular probit model, one assumes that all observations are independent of each other. For this model, however, for each panel (in this case states) the observations (for 15 years) are correlated. The xtprobit proceduretakes into account the correlation as well as the variation within and between states. The results are presented in Table 1 below.
The model suggests that the adoption of enterprise zones is driven by a distinctive set of political and economic factors. The parameter estimate for the partisan control of state government is positive in sign and statistically significant, demonstrating that greater control of state governing institutions by Democrats significantly increases the likelihood of a state adopting an enterprise zone program. As a double check on our findings, we also ran a cross-sectional time series OLS model with panel corrected standard errors. Our measure of government control was again statistically significant and positive. The results confirm our hypothesis that Democrat-controlled state governments are more likely to favor targeted tax incentives. Similarly, the greater the urban population in the state, the greater the likelihood of a state adopting the program.
Economic conditions are also important, but not as other studies of state economic development policy have suggested. As unemployment increases, states are less, not more, likely to adopt an enterprise zone. This counter-intuitive finding can be explained by the targeted nature of economic benefits conferred by enterprise zones. As economic conditions worsen statewide, it becomes more difficult politically to provide special benefits for economically distressed areas at the expense of the rest of the state. Instead, it is more likely a state will adopt a balanced growth strategy aimed at promoting economic growth in the entire state. During economic good times, states are more amenable to pursuing redistributive, targeted economic development policies like enterprise zones. Long term economic trends in sate manufacturing growth have not statistically significant impact.
State capacity also appears to influence the adoption of an enterprise zone program. The parameter estimate for executive centralization is statistically significant and negative, suggesting support for the hypothesis that strong governorships are able to slow the adoption of an economic development program that targets a subpopulation in the state. This finding suggests that the predominant political force in providing geographically specific benefits is the legislature, not the executive branch. The parameter estimates for staff and compensation are not statistically significant, which suggest the decision to adopt enterprise zones is a political one, not affected by the issue of state institutional capacity.
Finally, as expected, we find little evidence of a positive or negative regional diffusion effect for the adoption of enterprise zone programs. The parameter estimate for the dummy variable that captures whether a states neighbor has an Enterprise Zone Program is negative, but not statistically significant. In other words, the policy learning or competitive pressures from neighboring states have no apparent effect on a states own enterprise zone policy choices.
When do states pursue redistributive targeted economic development policies? Our results suggest the decision to adopt an enterprise zone program is predominantly a political one. It is driven foremost by Democratic control of the state governing institutions, and to a lesser degree by the political pressures created the size of the states urban population. The greater the extent of Democratic control of state governing institutions, the more likely the state is to adopt a set of economic development policies that target assistance to economically distressed areas. Our findings also suggest that states are more likely to adopt redistributive economic development programs during economic good times, when unemployment is low not high. Finally, our results also suggest that states with weaker executives are more likely to adopt the program, reflecting the political influence of urban legislators to pursue economic development policies that benefit their constituencies, as opposed to the entire state. In conclusion, the decision to adopt a redistributive targeted economic development policy is the product of a distinct set of political and economic conditions, conditions which are likely to change over time as the program is implemented.
The Expansion of State Enterprise Zone Programs
Traditionally, policy diffusion research has focused almost exclusively on the passage of a piece of legislation such as state lotteries ADDIN EN.CITE Berry19902710Berry, Frances StokesBerry, William D.1990State Lottery Adoptions: An Event History AnalysisAmerican Political Science Review84395-415(Berry and Berry 1990), sales and income taxes ADDIN EN.CITE Berry19922490Frances Stokes Berry William D. Berry1992Tax Innovation in the States: Capitalizing on Political OpportunityAmerican Journal of Political Science363715-42diffusion(Berry and Berry 1992), living will laws ADDIN EN.CITE Glick19912660Henry GlickScott P. Hayes1991Innovation and Reinvention in State Policy Making: Theory and Evolution of Living World LawsJournal of Politics533835-850Hayes19972750Hayes, Scott P.Glick, Henry1997The Role of Agenda Setting in Policy Innovation: An Event History Analysis of Living Will LawsAmerican Politics Quarterly254497-516(Glick and Hayes 1991; Hayes and Glick 1997), pre-Roe abortion laws ADDIN EN.CITE Mooney19952580Christopher Z. MooneyMei-Hsien Lee1995Legislative Morality in the American States: The Case of Pre-Rose Abortion Regulation ReformAmerican Journal of Political Science39599-627diffusion(Mooney and Lee 1995), and state insurance reform ADDIN EN.CITE Balla20012640Stephen Balla2001Interstate Professional Associations and the Diffusion of Policy InnovationsAmerican Politics Research293221-245(Balla 2001). The policy innovation literature has suggests that once states adopt an innovation, they never revise or reject it ADDIN EN.CITE Mooney20012870Mooney, Christopher Z.2001Modeling Regional Effects on State Policy DiffusionPolitical Research Quarterly54March103-24(Mooney 2001), although Glick and Hays study of living will laws emphasized the importance of the value of reinvention as part of the innovation process. (1991, 848). This approach may be problematic because most policy change happens through the modification of existing laws ADDIN EN.CITE Jones19942851Jones, Charles O.1994The Presidency in a Separated SystemWashington, DCBrookings Institution Press(Jones 1994) or changes in the administration of the existing laws ADDIN EN.CITE Hogwood19832681Brian HogwoodB. Guy Peters1983Policy DynamicsNew YorkSt. Martins Press(Hogwood and Peters 1983). The resulting methodological bias may distort our understanding of the policy process by under-representing cases of policy change via modification or abandonment ADDIN EN.CITE Blomquist19982657Blomquist, William1998The Policy Process and Large N Comparative StudiesSabatier, Paul A.Theories of the Policy ProcessBoulder, COWestview201-223(Blomquist 1998). For example, Minstroms analysis of the legislative consideration of school choice and passage of school choice legislation are affected by slightly different political dynamics ADDIN EN.CITE Minstrom19972630Michael Minstrom1997Policy Entrepreneurs and the Diffusion of InnovationAmerican Journal of Political Science413738-770Minstrom19982570Michael MinstromSandra Vergari1998Policy Networks and Innovation Diffusion: The Case of State Education ReformsJournal of Politics60126-48diffusion(Minstrom 1997; Minstrom and Vergari 1998).
Focusing exclusively on the passage of enterprise zone legislation would produce a misleading understanding of the policy process because many states have systematically undermined the original targeted nature of the legislation through legislative and administrative changes. As noted before, a recent study found most states have eased the eligibility requirements of their enterprise zone programs which has transformed the policy from being targeted at helping economically depressed areas to a business attraction and retention program benefiting the entire state ADDIN EN.CITE Talanker200323210Talanker, Alyssa and Kate Davis with Greg LeRoy2003Straying From Good Intentions: How States are Weakening Enterprise Zone and Tax Increment Financing ProgramsWashington, DCGood Jobs FirstAugust(Talanker 2003). Interviews with state enterprise zone officials revealed most programs focus on helping poor communities was gradually eroded over time by a mix of incremental administrative and legislative changes.
By increasing the number of zones in the state, the program becomes less targeted to helping economically distressed areas and more of a statewide incentive program pitting the poorest communities against more affluent ones, and thereby undermining the original intent to stimulate growth in poor areas. For example, as the City of Louisville and Jefferson County governments expanded the original enterprise zone twice to include the regional airport, major industrial sites and affluent neighborhoods, it lost its original focus on a small distressed area, it ADDIN EN.CITE Lambert20012590Thomas E. LambertPaul A. Coomes2001An Evaluation of the Effectiveness of Louisville's Enterprise ZoneEconomic Development Quarterly152168-180(Lambert and Coomes 2001). On average, the states in our sample increased the number of zones in their states by 250 percent. Thus, we decided to use the increase in the number of enterprise zones in the state as the best indicator for whether a program remained targeted at poor communities or not.
Unlike the creation of enterprise zones model, our dependent variable in the expansion of enterprise zones is the number of new enterprise zones in the state each year. This longitudinal data on policy outcomes is a better measure of assessing policy change because it allows us to assess the implementation of the policy, rather than intended effects of the legislation. Moreover, the cross-sectional fixed-effects model allows us to account for the cumulative impact of policy changes on policy outcomes, as opposed to event history analysis which would force us to identify a single outcome at one point of time.
The methodology used in the expansion model is also different. Because are data are time series cross-sectional data (TSCS) with a small time period (T) relative to the number of cross-sectional units (N), we use ordinary least squares (OLS) with panel corrected standard errors. This was done on the advice of Beck and Katz (1995) who demonstrate that OLS with panel corrected standard errors still does well when TCable19982500Greg CableRichard Feiock1998The Adoption of State Economic Development Programs: An Event History AnalysisJournal of Politics(Cable and Feiock 1998). In other words, states with higher unemployment rates are able to keep their enterprise zones targeted on the most economically distressed areas. Given the high costs of enterprise zones ADDIN EN.CITE Peters2002631Alan H. PetersPeter S. Fisher2002State Enterprise Zones: Do They Work?KalamazooW.E. Upjohn Instituteenterprize zones(Peters and Fisher 2002), states may not be able to afford expanding them in periods of rising unemployment. As the economy improves and state unemployment decreases, states are more likely to increase the number of zones to presumably include less economically distressed areas. What was once a geographically targeted program becomes more of a statewide incentive program. The variable for long term manufacturing growth is negative and statistically significant, suggesting with positive long term economic vitality create fewer enterprise zones than their counterparts whose economies are struggling. This dynamic reveals an interesting paradox about state economic development capacity. It is harder, not easier, to target public assistance to poorer communities during good times than bad.
Organizational scholars have stressed the importance of state organizational capacity to
resist political pressures and retaining an enterprise zone program focused on economically distressed areas (Mossberger 2000). Our analysis finds mixed evidence for this. We had hypothesized that governors would be inclined to minimize the number of zones so as to maximize economic impact while minimizing the draining on the state treasury, whereas the legislature would seek to increase the number of zones. The parameter estimate for executive centralization is statistically significant and negative, meaning the more centralized the executive branch is, the fewer the number of enterprise zones created. Our conversations with state enterprise zone coordinators revealed that they are always having to fight the legislature about eligibility standards (for zones), and thus we predicted that states with greater bureaucratic capacity, measured by staffing, would be less likely to expand their enterprise zone programs. However, the parameter estimate for staffing and compensation is positive, but not statistically significant.
Finally, we examined what role, if any, cross border competition or regional diffusion played in influencing a states enterprise zone program. As noted earlier, we found little evidence to support the regional diffusion thesis that states adopt enterprise zone programs to compete with their neighbors for investment and jobs. We do, however, find evidence that states implementation of their enterprise zone programs are influenced by neighboring states policy choices. The parameter estimate for whether a neighboring state has an enterprise zone program is statistically significant and positive, suggesting that states that border other states with enterprise zone program are more much more likely to increase the number of zones in the state. The finding highlights the existence of a race-to-the-bottom competition among states that negatively impacts their ability to target their economic development policies to help poor areas.
Whereas the creation of enterprise zone programs was a function of Democratic control of state government and good economic times, our results suggest that expansion of the number of zones is driven by an entirely different set of political and economic dynamics. The more conservative the state government ideology and low unemployment lead to an increase in the number of enterprise zones in the state, thereby undermining the redistributive intent of the program. Interstate competition also leads to an increase in the number of zones, as states increase the number of enterprise zones in increasingly less distressed areas to compete with neighboring states to attract investment and jobs. Higher levels of executive centralization, greater liberal state government ideology, and long term manufacturing growth are associated with the creation of fewer enterprise zones.
Conclusion
In conclusion, a states ability to target their economic development policies to help the most economically distressed areas of their states, as measured by the creation and expansion of their enterprise zone programs, is a function of partisan control of state government, and less about socio-economic and administrative factors. Enterprise zone legislation, which typically provides major tax incentives and other financial assistance to induce firms to expand their operations or relocate to the most economically distressed areas of the state, is most likely to get enacted as Democrats control more of the state governing institutions. However, the targeted aspect of these programs is mostly likely to be undermined by an increase in the number of zones in the state as a states conservative state government ideology increases. These results suggest party control of government matters for who gets what and when in state enterprise zone policies.
Our findings also have implications for the state policy innovation and diffusion literature. Had we confined our analysis to the passage of enterprise zone legislation, our results would suggests that Democratic control of state institutions, the size of urban population, and low unemployment are the set of political and socio-economic conditions that are most conducive to a state pursuing a redistributive economic development policy. This picture would be incomplete, since we know policies can be changed, undermined, and even reversed through subsequent legislative, administrative, and budgeting changes. By extending our analysis to include the implementation of the enterprise zone program, we discovered the dynamic nature of the policy process and developed a more pessimistic account of states capacity to pursue redistributive economic development policies that target economically distressed areas. While economic competition from neighboring states has no impact on the creation of enterprise zone programs, the presence of an enterprise zone program in a neighboring state created a race-to-the-bottom dynamic where states actively increased the number of zones to compete with their neighbors, at the expense of targeting their economic development efforts at poor communities. Our analysis also suggested that the more conservative the state ideology, the less likely a state is to keep its enterprise zone focused on distressed communities.
The result is a less than optimistic assessment of whether states can pursue redistributive policies like enterprise zones that target state economic development efforts at economically distressed communities. While states can enact such policies during periods of Democratic control of state institutions and times of economic prosperity, they are gradually undermined by increases in conservative state ideology and periods of continued low unemployment which spur states to increase the number of zones. The result is that programs that were originally intende d t o p r o v i d e t a x i n c e n t i v e s t o b u s i n e s s e s f o r l o c a t i n g i n i m p o v e r i s h e d n e i g h b o r h o o d s a r e i n c r e a s i n g l y l i k e l y t o s u b s i d i z e e c o n o m i c d e v e l o p m e n t i n a f f l u e n t a r e a s .
T a b l e 1 . T h e D e t e r m i n a n t s o f S t a t e A d o p t i o n o f a n E n t e r p r i s e Z o n e P r o g r a m
V a r i a b l e ( s e) p>|z|PartisanDemocrat control of state government.8.52***
(2.09).000Socio-economic Percentage of the state that lives in urban areas..17***
(.05).001Unemployment - percentage of the population considered unemployed by the federal government-.22**
(.125).08Percentage change in the number of manufacturing jobs from 1982 to 1995-.18
(1.05).85Administrative Staffing and compensation.26
(.26).98Executive centralization-1.20***
(.46).01Competitive Does a neighbor have an EZP?-.25
(.63).69Year1982
-8.40
(925).9931983 -6.21
(925).9951984-5.21
(925).9961985
-21.99
(925)1.01986-3.42
(925).9971987-1.87
(925).99819881.12
(925).99919892.96
(925).9971993-8.16
(925)11994-10.75
(925)119955.66
(925).995Constant-14.37
(925).998Number of observations379 Number of groups50Chi-Square = 21.78 Source: Author's analysis. Unstandardized logistic regression coefficients are presented, standard errors are in p a r e n t h e s e s ; p r o b a b i l i t i e s b a s e d o n a 2 - t a i l e d t e s t . * p <