Back to summary report — Back to Tax Revision Commission main page — Back to list of research papers
|
Columns DCWatch
Archives Elections Government and People Budget issues Organizations |
WHAT DO WE
KNOW ABOUT THE EFFECT OF TAXES ON ECONOMIC DEVELOPMENT?: Report Prepared for the DC Tax Revision Commission by Stephen T. Mark Institute of Government and Public Affairs University of Illinois at Chicago Therese J. McGuire Institute of Government and Public Affairs and College of Urban Planning and Public Affairs University of Illinois at Chicago Leslie E. Papke Department of Economics Michigan State University First Draft: June 2, 1997 Revised: July 11, 1997 I. Introduction The District of Columbia faces both an immediate fiscal crisis and a longstanding dismally performing economy. Employment declined in the 1990s, and the continuing decline in population has accelerated in recent years. The District's unemployment rate was 8.5 percent in 1996, compared with a regional unemployment rate of just 3.9 percent. The District's tax base is limited by the vast federal presence (over 40 percent of real property is tax exempt) and by federal prohibitions on tax instruments (the District is prohibited from taxing the income of non-resident workers, who comprise 67 percent of the DC workforce). The District's tax rates on business and individuals are the highest in the region. The combination of high tax rates, narrow tax bases, and a poorly performing economy create problems for the District's fiscal and economic future. In this report we draw lessons for reform of the District of Columbia's economic development policy from a vast amount of literature. We begin in section II with a brief description of the present state of the District's and the metro area's economies. In section III we describe the array of economic development tools presently available in the District and in the Maryland and Virginia counties that comprise the metropolitan area. Section IV reviews and draws lessons from the empirical literature that examines inter-regional and intra-regional business location and employment decisions. Section V reviews the literature on enterprise zones, and the final section draws policy conclusions from our survey of the literature. II. The Economy of the DC Metropolitan Area The DC Metropolitan area consists of the District and parts of Maryland and Virginia. For purposes of this study we consider the metropolitan area to be the District, the seven closest counties in Maryland and Virginia, and Alexandria, an independent city in Virginia. The Virginia counties include Arlington, Fairfax, Loudon, and Prince William. The Maryland counties are Charles, Montgomery, and Prince George's. The economy of the District of Columbia deteriorated in the 1990s in a number of ways. First, population decline accelerated, with a 9.1% decline between 1989 and 1994, after a decline of just 4.3% during the full decade of the 1980s (see Table 1). This has harmed the revenue from the individual income tax, which registered a real decline of 13.2% during the five-year period. As Table 2 indicates, individual income tax revenue reached a high in real dollars in 1988. District private sector and government jobs also declined in the 1990s (see Table 4), and the District unemployment rate recovered very little from its recession high in 1992 (see Table 3). At the same time that population declined, District income per capita increased 12.6% in the 1990s (see table 1). In contrast, real income per capita in Montgomery, Prince George's, and Fairfax Counties, as well as that in Alexandria, declined between 1989 and 1994. These trends suggest that the population leaving the District has a mean income below the District mean income, and that this population has been moving to the suburban counties. District income per capita of $30,684 was quite high relative to the national average of $21,696 in 1994. The District's high average income, perhaps, can be explained by its income distribution, which is characterized by a disproportionately high concentration of high income households. Employment also has performed poorly in this decade. Between 1989 and 1994, private employment declined by thirteen thousand jobs, and federal civilian jobs declined by two thousand (see Table 4). The federal decline is a net figure and hides a good bit of turnover. Eleven thousand federal civilian jobs were gained between 1989 and 1992, but thirteen thousand were lost in the city between 1992 and 1994 resulting in a net loss of two thousand. District unemployment rates reflect this poor employment news. As Table 3 shows, District rates have been consistently above metro area rates. The recent era can be divided into three parts. From 1986 to 1990, District unemployment rates were about twice the metro area rates. In 1991 and 1992, the rates increased in both the District and the region reaching high annual averages of 8.6% and 5.3%, respectively, in 1992. Between 1992 and 1996, the unemployment rate declined considerably in the region (to 3.9 percent), but not in the District, where average unemployment for 1996 was 8.5 percent.(1) Table 4 shows that the share of District private sector jobs in the service sector increased from 47.9% in 1974 to 66.2% in 1994. The shares of every other sub-sector of the private economy decreased.(2) Tables 5 and 6 illustrate that the economies in the Maryland and Virginia suburbs also increased their reliance on the service sector, but their employment bases remained better diversified than the District's. The District's share of total metropolitan area employment fell for each major industry between 1974 and 1994 as demonstrated in Table 7. Overall, the District's share of total private employment fell from 34.8% to 23.0% during the twenty-year period. The sharpest decline was for the wholesale trade sector, with a drop of 31.5% to 8.0%. We would expect to see the District's share of employment decline over time as the metropolitan area grows since the vast majority of land available for development is outside of the District. However, it is troubling that these declines in the District's share of total metropolitan employment reflect declines in absolute numbers of jobs in the District for all sectors but services, mining and other (see table 4). Service sector employment in the metro area virtually exploded during the period from 403,483 jobs in 1974 to 1,024,386 jobs in 1994. The District did not maintain its share of service sector jobs, however, as its share fell from 44.1% to 30.4%. Table 8 displays the tax rates imposed in 1995 in the various jurisdictions comprising the metropolitan area. Taxes in Maryland were slightly higher virtually across the board than taxes in Virginia. The District's tax rates were much higher than those found in any of the Washington suburbs. The property tax stands out with the DC rate being approximately twice the rates in the suburbs. III. Economic Development Policy in the District of Columbia The recent fiscal crisis of Washington, DC has yielded a number of economic development initiatives from the District government, the District business community, and the Federal government. To gain perspective on the potential benefits of these proposals, we outline the economic development programs currently in place, compare them with economic development activities in the surrounding jurisdictions, and report on the programs' accomplishments. In newspaper articles and surveys, District business people comment that the City government is uncooperative, unresponsive, uncaring, and "not-so-friendly to business."(3) O'Cleireacain (1997) concludes that, "the District has no effective economic development strategy." The District has taken some economic development policy initiatives, but apparently they have been poorly received in the business community, and largely ineffective in generating gains in private sector employment. The District's FY 1997 Budget and Financial Plan establishes a new agency, the Business Services and Economic Development Agency (BSEDA). The new agency consolidates the responsibilities previously distributed among eleven different departments.(4) The District's economic development policies include an active Industrial Revenue Bond Program, a group of loan funds governed by federal targeting regulations, three enterprise zone designations, and an array of business services. Industrial Revenue Bonds and Other Loan Programs Private purpose bonds, also known as industrial development and industrial revenue bonds (IRBs), provide a way for private industry to borrow money at below market level interest rates through the tax exempt securities market. The Tax Reform Act of 1986 imposed limits, or caps, for each state on the total amount of IRB issuance, other than debt associated with not-for-profit 501(c)(3) organizations. The Act also limited these bonds to tax exempt institutions and manufacturers. The formula for the caps is the greater of $50 per capita, or $150 million. For the District and some of the smaller states, the $150 million is the applicable maximum. In per capita terms its volume cap comes to over $270, a much greater maximum than the $50 per capita limit that Maryland and Virginia face. The District issues a substantial volume of IRBs. Since the inception of its IRB program in 1985, the District has issued $2.5 billion of IRBs, which is conservatively estimated to represent an annual subsidy of $50 million to the benefitting institutions.(5) By way of comparison, while $129 million of IRBs were issued in the District in fiscal 1996, just $12 million were issued in Fairfax County, $15 million in Prince William County, $68 million in Loudon County, and none in Arlington or Alexandria(6). When issuance declined from $184 million in 1993 to $44 million in 1994 and $52 million in 1995, the District responded with the "Speedy Authorization Act," and issuance grew to $129 million in 1996. It appears that bond issuance has been positively associated with growth in at least one sector of the economy. A large share of the borrowers in this program have been hospitals, and health care is one of the few economic sectors in the District showing growth. In the intervals between 1985 and 1989, and 1989 and 1993, private sector jobs in the District grew by 12.4% and then declined by 3.5%. At the same time, health care jobs saw growth of 55.0% and 40.4%. The District's share of regional health care jobs jumped from 35.1% in 1985 to 42.8% in 1993. Before the IRB program was in place, between 1977 and 1985, the District's share of regional health care jobs declined from 40.0% to 35.1%. The District's success in using IRBs has mixed implications for District revenue. District borrowers have almost all been tax-exempt institutions, primarily hospitals and universities. Therefore, the increase in employment will produce a disproportionately small payback in tax revenue. To the extent employees in the benefitting industries reside in the District, however, some of the growth resulting from the subsidy will yield income tax revenue. Also, the city's retail economy and thus its sales tax revenues benefit from the presence of commuters. Of the District's 52 IRB issues totaling $2.5 billion, just four issues totaling $8.3 million were for manufacturing firms. These manufacturer-subsidizing issuances occurred in 1985, 1986 and 1987. While in the District all of the $129 million in IRB proceeds went to tax-exempt institutions in 1996, of the $32.5 million issued in the City of Baltimore, just $10 million went to a tax-exempt entity and the rest to manufacturers. Of the suburban issuances in 1996, $3 million went to a private entity in Fairfax, and all of Prince William's issuance of $15 million went to a private entity. The District also has five much smaller loan programs, which have lent a total of $4 million to small- and medium-sized businesses and non-profit groups since fiscal 1996. Much of this money comes from the U.S. Department of Housing and Urban Development and can be loaned only to businesses meeting strict criteria such as location in a blighted neighborhood, or creation of jobs. Baltimore reported lending $4 million in economic development loans in fiscal year 1996, so the District's program is not atypical. Enterprise Zones in the District and Surrounding Counties Another legislative effort the District has undertaken is its economic development zone program.(7) The District established three enterprise zones in the 1988 tax code, but administrative regulations were not in place until 1994. Enterprise zone businesses are entitled to three tax credits - two on labor and one on capital. The District credits half of each worker's wages, up to $7,500, for two years, for District residents who earned a low income prior to employment in the zone. An additional credit equal to half of the worker's compensation insurance premium is granted as well. Both of these credits apply against the corporation or unincorporated business income tax. The District also grants a property tax abatement on improvements within zones. The abatement in the first year comes to 80% of the tax due. The credit declines each year, reaching a value of 16% in the fifth year before termination.(8) Maryland has three area zones, two in Prince George's and one in Montgomery County. Zone businesses receive a $3,000 credit for three years for the hiring of a disadvantaged worker, and $500 credit for one year per non-disadvantaged worker hired. Maryland also allows an 80% abatement of the property tax due for five years. The abatement declines to 10% between years six and ten, after which it is terminated.(9) Virginia has one zone in Alexandria. The zone provides businesses a grant of $1,000 per zone resident hired, $500 per non-zone-resident. Both of the Virginia grants can be claimed for three years. There is no abatement of the local property tax in Virginia. But significant state income tax credits are offered to zone businesses. Thirty percent of the value of major new zone construction, up to $125,000, is credited against a business' income tax liability to Virginia. This credit is refundable, so that taxable income can be earned by qualified zone businesses when the credit exceeds the firm's tax liability. Zone firms also receive a credit worth 80% of their income tax liability in the first year of establishment or expansion in the zone, and 60% in the nine following years.(10) To compare these disparate deals, consider a hypothetical firm. Suppose $100,000 of improvements are made to a building currently worth $250,000, and this building is located in an enterprise zone in either the District, Prince George's County, or Alexandria. As a percentage of assessed value, the current tax rates are 2.15% in the District, 1.05% in Prince George's County, and 1.07% in Alexandria. As Table 9 shows, keeping the full 80% abatement in place for five years in the Maryland zone produces a sizable discount over the DC abatement. Further, the Maryland abatement continues for another five years. It is also apparent that the difference in property tax rates has a strong impact on this comparison. Beginning in year three the tax due on the improvement in the District's zone is more than the tax due on a similar improvement outside the zone in Prince George's County. Suppose the firm plans to hire two disadvantaged zone residents. In the District, these workers would bring credits worth $30,000 over two years, plus twice the annual worker's compensation premium. In Maryland, the hiring firm would receive credits worth $18,000 over three years, in Virginia, $6,000. These figures reflect undiscounted sums of credits allocated over two or three years. This per worker credit may more than offset the District disadvantage in property tax for enterprise zone participants, depending on the amount of new construction and the number of new hires.(11) Were the hypothetical firm located in a Virginia zone, it would receive a $30,000 credit against its corporate income tax in the first year, based on the 30% credit for real property improvements. The additional credit of 80% of its corporate income liability in its first year and 60% for the next nine years would probably make the Virginia exemptions worth more than the District or the Maryland ones, depending, again, on how the firm's expenditures are distributed. Also available to zone firms in Maryland and Virginia are statewide job-creation tax credits available against business income taxes for firms creating new jobs in those states. In Maryland, firms that in the course of two years create sixty or more sufficiently high paying jobs are entitled to a credit equal to the lesser of 21/2% of the wages paid, or $1,000 per job.(12) In Virginia, businesses that create over 100 new jobs may claim a credit of $1,000 per job in excess of the 100 job minimum.(13) The enterprise zone incentives in Maryland and Virginia have had moderate use. Alexandria's zone, created in 1994, has four businesses claiming credits for 52 new jobs.(14) The two zones in Prince George's County have also yielded four claimant businesses, responsible for 152 new jobs, since their beginnings in 1995 and the mid 1980s.(15) The District economic development zone program, in the code since 1988, and open for business since 1994, has yielded one qualified business, the Good Hope Marketplace. The project is under construction and no new-jobs figure is available (O'Cleireacain, 1997). The greater generosity of the Virginia enterprise zone program and the greater success of the Maryland program suggest two differences between the financial position of the District and that of the two states. Because District tax revenues have been declining in the 1990s, District authorities may feel they cannot afford the generous credits the zone program would provide were businesses to participate in greater numbers. Also, insofar as District non-residents benefit from economic development aid, the District gets less political and financial gain from zone participation. The fact that 67% of District workers are non-residents suggests a large proportion of its entrepreneurs are as well.(16) It seems certain many of the tax breaks to the business community would accrue to non-residents. Business Services The District's reorganization of the economic development activities into the new BSEDA reflects a concern about management and responsiveness. All of the suburban counties have economic development agencies. Their missions are inherently easier than the District's. They publicize their extraordinary growth and attempt to reinforce it. The Fairfax County Economic Development Authority concentrates its efforts on marketing, with a third of its $3.8 million budget spent on advertising.(17) It boasts a World Wide Web site and advertisements in the Economist and on CNN. The group provides the information that economic development agencies customarily do, such as the identification of available space for arriving firms and their employees, but it sets itself apart from the crowd with services such as a newsletter for small, minority and woman-owned businesses and a planned conference in 1998 on information technology.(18) Fairfax is the economic star of the metropolitan area. In the 1980s it gained 227,000 jobs, accounting for 31% of the growth in jobs in the region. Between 1990 and 1994 it gained nearly 21,000 jobs, more than the increase in the region as a whole (about 17,000, U.S. Department of Commerce, Bureau of Economic Analysis). It is impossible to say whether its sophisticated economic development agency is a factor in this performance or a luxury afforded by Fairfax's prosperity. The business services the BSEDA plans to perform include a One-Stop Business Center. The center is seen as a multi-purpose facility for existing and prospective businesses. It will issue licenses and permits and provide regulatory information. BSEDA will also have a Business Development component, charged with marketing the District as a business location and serving as a point of contact for DC businesses contemplating relocation. According to the District's FY 1997 Budget and Multiyear Plan, a Housing and Community Development group will administer the economic development policies of the District, including its enterprise zones. Tax Breaks to Individual Firms Recently the District negotiated its first firm-specific tax deferral, for a firm called the Bureau of National Affairs, Inc. This corporation agreed to remain in the District, where it employs 1,100 people, in exchange for a ten-year deferral of its real property taxes. The bill was passed by the City Council, but included a requirement that the mayor submit legislation to the council establishing standards for the granting of economic development incentives. The mayor must submit this new legislation to the council by September 16, 1997. IV. Inter-regional and Intra-regional Econometric Studies The question, "Do taxes matter?", has captured the attention of numerous researchers. From individual firm location decisions to aggregate regional employment growth rates, researchers have attempted to determine whether taxes and other policy variables are significant factors in explaining why some regions outperform others. Answering the question is of obvious importance to policy makers who must impose taxes to finance public expenditures, but desire to minimize the harm to the economic development of their regions. In order to provide an answer to the question, we survey the vast empirical literature on this topic in a search for consensus findings and for lessons for policy makers. Our job is made easier by the fact that several surveys of the relevant literature have been conducted in recent years. In essence, we survey the surveys and provide our own interpretation of the findings and conclusions reached by the authors. By econometric studies we refer to statistical analyses that relate a variable of interest (the dependent variable), such as branch plant openings or aggregate employment growth, to several variables (the independent variables), such as electricity costs or quality of the labor force, that are theoretically expected to influence the variable of interest. This method of analysis enables the researcher to determine, in a rigorous way, which of the independent variables are statistically significant factors for explaining the dependent variable. These types of studies provide systematic evidence based on data and verifiable facts, which is more compelling than anecdotal evidence or information from interviews with business managers, who may have an incentive to exaggerate the importance of taxes. In this discussion we say that a variable "matters" if it has a statistically significant relationship to the dependent variable. If the effect of the variable is not statistically different from zero, we say that the variable "does not matter." Statistical significance should not be confused with the size or importance of the effect. A variable can be statistically significant, and thus matter, but the estimated coefficient and elasticity may be small, in which case we would say that the variable is not an important factor. The different studies can be divided into two types: studies that compare one region to another (where region is often a state, county or metropolitan area), and studies that compare areas to one another within one region (such as municipalities in an SMSA). This division is sensible because the set of important factors is likely to differ for inter-regional location decisions compared to intra-regional decisions. In particular, because many labor market and cost factors are constant within a given region, taxes, which vary from one locale to another within a region, might be expected to be important factors in intra-regional analyses. We rely on three recent surveys of the relevant literature. Two of these appeared in the March/April 1997 issue of the New England Economic Review as part of a proceedings of a symposium held by the Federal Reserve Bank of Boston on "The Effects of State and Local Public Policies on Economic Development." Even though Wasylenko (1997) focuses on taxes and Fisher (1997) on expenditures, the two articles reference many studies in common. One such reference is a third survey of the literature by Bartik (1991). We select for our review 16 articles discussed by Wasylenko, Fisher, or Bartik. We consider these studies to be important contributions to the literature and illustrative of our main points. We do not provide a comprehensive review; this has been done by the three authors mentioned and by many others. Rather, our aim is to use a representative set of articles to draw a consensus set of findings and policy implications. The 16 articles are listed and briefly described in Table 10, where we have categorized them as inter-regional or intra-regional studies. In the table, we describe the dependent variable, we indicate whether the unit of observation is states, metro areas, or local governments, and we indicate whether tax variables (and spending variables in those studies that include them) are found to be statistically significant factors. The inter-regional studies examine differences in economic development across states or SMSAs. The studies differ in terms of their measures of economic activity and in the list of explanatory factors examined. The intra-regional studies examine differences in location activity or employment across areas within a given metropolitan area. Many, but not all, inter-regional studies find that taxes are a statistically significant factor (for example, the two most recent studies, Hines, 1996, which examines the location of foreign branch plants in the U.S., and Tannenwald, 1996, which examines investment by manufacturing companies, come to opposite conclusions about the effect of taxes), but the findings are not always robust to changes in specification, time period, or measurement. As noted in Table 10, the finding of a significant tax effect in Wasylenko and McGuire (1985) could not be replicated by the same authors using more recent data (McGuire and Wasylenko, 1987 and Carroll and Wasylenko, 1991). Authors of intra-regional studies more consistently find a significant effect of tax differentials on local economic activity. However, in both types of studies, when taxes are statistically significant, the size of the effect is not large. Other factors, such as labor costs or labor quality, tend to be more important. These empirical findings are supported by the evidence from surveys of firms, which consistently place taxes low on the list of critical factors in making location decisions. The fact that contradictory results for tax and spending variables can be found not only across the different studies but within the same study gives one pause. It may be that a spending variable is a positive factor for manufacturing, but a negative factor for wholesale trade; or that a tax variable may be a negative factor if the dependent variable is measured one way, but not a factor if it is measured another; or the results may change as the sample size is varied (each of these examples exists in at least one study listed in Table 10). In other words, it is not appropriate based on one or more of these studies to place much confidence in a bottom-line conclusion that "taxes matter" or "taxes do not matter." A different conclusion from this literature is that there is a role for publicly-provided services in shaping a conducive environment for business. Many of the studies examine both taxes and government expenditures and find that spending on education, highways, and other types of services likely to be valued by firms has a positive effect on economic activity. Indeed, some studies find that an increase in taxes coupled with an increase in spending on desirable services is a net plus for economic development. There is also a practical lesson in the empirical results on taxes and spending. While the regression results may indicate that a cut in taxes would result in a (statistically significant but small) boost in economic activity, this is a marginal effect; the result depends on holding all other factors, including government spending, constant. For state and local governments facing balanced budget rules, such an experiment is not feasible. In summary, there are several policy lessons to be learned from inter-regional and intra-regional econometric studies. First, many studies find that the level of taxes matters, i.e., that tax variables are statistically significant determinants of economic activity, but many others do not, casting doubt on how the literature would answer the simple question posed at the beginning of this section. Even when taxes do matter, the magnitude of the effect is small, and other factors are more important in explaining differences in economic activity across space. Finally, in creating an environment conducive to economic development, how governments spend their tax revenues may be more important than the level of taxes. Our overarching conclusion is that cutting taxes is not a panacea for a poorly performing regional economy. As McGuire (1992) stated in her review of Bartik (1991), the "...message to policy makers is that the effects of state and local tax policy are so uncertain that concern over this issue [of economic development] should not be a driving force in general fiscal policy decisions" (page 458). There is a related, much smaller, literature on the effects of economic development incentives that warrants brief mention here. These studies involve estimation of models similar to those summarized above, but an additional variable, some measure of the extent of utilization of targeted tax breaks and loan subsidy programs, is included in the regressions. McHone (1984) investigates the effect of eight different industrial development incentives on the relative employment performance of counties in 26 multi-state SMSAs. The incentive variables are measured at the state level and all but two are measured as binary variables based on his assessment of a given county's advantage with the particular incentive program. He finds weak evidence of an effect of several of the variables, including a variable indicating whether a state has a separate agency or authority to administer the industrial development bond (IDB) program. His result for state and local taxes, however, is contrary to expectations in that he finds that taxes are positively related to employment growth. Marlin (1990) finds that the volume of IDB loans issued in a state is positively related to Gross State Product (GSP). This finding is marred by the fact that the model he estimates omits variables likely to be important and includes variables that are poorly measured (e.g., local taxes). De Bartolome and Spiegel (1997) examine manufacturing employment change in the states between 1990 and 1993. Their key independent variable is spending by the state development agency, which has a positive, statistically significant effect on manufacturing employment. Interpretation of this finding for policy makers is difficult given that the authors do not explore different categories and programs of spending by state development agencies. V. An Overview of Enterprise Zone Programs Enterprise zone (EZ) programs are geographically targeted tax, expenditure, and regulatory inducements that have been part of subnational economic development strategies since the early 1980s. By 1993, 37 states and the District of Columbia had established some form of EZ initiative. Recently, the federal government initiated a program of Empowerment Zones that is based on the state models. While they differ in specifics, all the programs provide tax preferences to capital and/or labor and other development incentives to encourage investment expansion or location, and to enhance employment opportunities for residents in depressed areas. A survey by Erickson and Friedman (1991) shows that most zones use tax incentives: 51 percent offer sales or use tax credits, job creation and wage credits; 49 percent offer employer income tax credits; 43 percent offer selective hiring credits; 37 percent offer investment credits. In addition, 20 states made property tax reduction available at the option of the local government. In her survey of spatially targeted economic development strategies, Ladd (1994) compares zone programs to other policy approaches to combat urban distress. She describes three broad approaches: people-oriented strategies, place-based people strategies, and pure place strategies. Most state enterprise zone programs are purely place-based or place-based people strategies in that place-specific assistance is used to help the residents of distressed urban areas. The tax incentives provided to the Empowerment Zones and Enterprise Communities established by Federal legislation are an example of this type of strategy. Tax incentives include a 20 percent tax credit covering the first $15,000 of wages and certain types of training businesses provide to employees who live and work in the zone, as well as tax and financial incentives for investment in the zone. Like most economic development policy that relies on tax incentives, the use of enterprise zones is controversial. EZs have been criticized on the grounds that they will be ineffective and inefficient in stimulating new economic activity. This criticism is part of a long-standing debate on the effects of inter-site tax differentials on the location of capital investment. If any tax-induced investment only represents relocation from another state, then tax competition is a zero-sum game for the country as a whole. In addition, the preferential treatment of certain types of investment or employment within EZs may induce decisions which would be uneconomic in the absence of the tax incentives. Further, critics point out that while the usual stated goal of EZ programs is to increase wages and employment, typically the capital incentives are larger than the wage subsidies. (For example, investment tax credits will be stated in percentage terms, while wage subsidies are capped at a dollar amount.). Fisher and Peters (1997) find that typical capital incentives have much larger effects on the price of capital goods than average labor incentives have on wages in the 20 state EZ programs they analyzed. Gravelle (1992) notes that subsides to capital increase employment in only a roundabout fashion. A capital subsidy causes firms to substitute capital for labor. Only when the capital subsidy induces more production does employment increase. Therefore, if the goal of the EZ program is to increase local employment, subsidies to labor are more appropriate Proponents of geographically targeted subsidies argue that the EZ investment may give individuals employment experience that enhance their long run employability. Thus, even a relatively short run economic development program may have long run effects (Bartik, 1991). In addition, concern with net capital investment may be less relevant for EZs since redistribution even within the state may be an end in itself. If investment is relocated from local labor markets with low unemployment to local labor markets with high unemployment, the incentives may generate efficiency gains for the economy as under-utilized resources are tapped. Efficiency gains may also result if reductions in unemployment produce positive externalities such as reductions in social unrest. It is also possible that, in addition to encouraging existing businesses to locate in particular geographic areas, the incentives may induce the creation of new businesses that would not otherwise have been started. Such new businesses could produce taxable profits and incomes that would reduce the revenue cost of the incentives. A thorough discussion of these argument is beyond the scope of this review. Instead, this section focuses primarily on empirical studies of zone performance. First, we discuss key issues in evaluating zone performance. Next, we give a brief guide to the literature and summarize the finding of a recent survey article. Finally, we review studies of zone performance in three states -- Maryland, Indiana, and New Jersey. Why is Zone Performance Difficult to Measure? Zone evaluation depends primarily on two factors -- program goals and the nature of the available data. Often, the legislation is unclear about whether the goal of the zone program is to increase net employment or investment. Studies typically assume that the intent of the legislation is to create new jobs in the zone, not merely relocate jobs from outside of the zone. These jobs may be full-time, part-time, or of limited duration, since the legislation typically does not specify the type or duration of job it is intended to create. In practice, zone success is frequently measured by the amount of investment undertaken after the designation, the increase in the number of firms in the zone, and the change in zone employment. Cost is measured by direct government spending and foregone revenue per job created (or, if the goal of the program is zone-resident employment, cost per zone-resident job). Determining which new jobs result from zone designation presents a practical difficulty. The key methodological issue is how to separate the effects of zone designation from jobs and investment that arise from other factors -- for example, general upturns in the economy or in the area surrounding the zone. Alternatively, which of the measured changes in jobs and investment are attributable solely to the zone program? Survey or case study methodologies provide useful information on zone participation, but they cannot definitively answer this question. Firm managers' estimates of net job creation or investment are subjective, and even candid managers may have difficulty attributing a certain number of new jobs to zone incentives alone. Surveys of zone administrators are even more problematic, since the responses may be self-serving and the relative candor of the respondents may differ. Further, surveying existing participants several years into the zone program creates selection bias that tends to inflate evidence of zone success. Firms that left the zone during the program are not surveyed. The remaining firms tend to be winners that make the zone program look successful. The study of the New Jersey zone program by Rubin (1990) illustrates this problem. Econometric analysis is better suited to performing the "but for the zone" experiment. If the zone sites were randomly selected, the effect of the program could be measured by comparing the performance of the experimental and control groups. Actual EZ designation, of course, is based on economic performance, so the data are non-experimental. This problem is called sample selection and can be addressed with a variety of techniques (see Papke, 1994). In particular, researchers must account for the possibility that EZ designation might be correlated with unobservable factors that also might effect economic performance. For example, when evaluating the effect of zone designation on employment, one must control for the fact that typical zones have high unemployment rates or one may conclude that designation increases unemployment. Or, if zone administrators select the fastest-growing areas for designation, this selection would bias the analysis. Without proper controls, one might conclude that zone designation causes faster growth. Most EZ programs have not been evaluated using appropriate econometric techniques. We next provide a brief summary of the literature to date and conclude with a discussion of three evaluations of state programs, two of which employ proper methods. A Brief Guide to the Literature Most of the literature consists of descriptive studies of zone programs. While descriptive EZ studies do not attempt to measure the effectiveness of EZs, they provide useful lessons on EZ construction. Informative surveys of these studies are found in Britnall and Green (1988), Funkhouser and Lorenz (1987), and Ladd (1994). One descriptive study in particular is interesting for its comparison of tax incentives versus administrative assistance. Elling and Sheldon (1991) examine the effectiveness of tax incentives in 47 zones in Illinois, Indiana, Kentucky, and Ohio. They describe three forms of incentives: direct tax savings, direct nontax savings, and deregulation. They compare administrative strategies across the zones. Regression analysis relates the characteristics of each program to the number of firms that apply for zone benefits each year. They find that the main contributors to high application rates are the administrative resources devoted to the zone, and the services, such as technical assistance, the zone provides. Their conclusion is that tax and financial incentives appear to be an ineffective part of the zone programs. The most detailed survey is provided by Ladd (1994). She compares EZ programs to other spatially targeted development strategies and provides a context for zone evaluation. She notes that while some zone programs are implemented to help small businesses, they include tax abatements that are not useful for small firms (reductions in the corporate tax rate, for example, are not useful for sole proprietorships or partnerships). She compares supply-side tax reductions, in general, with more active governmental programs, such as government sponsored job training programs. She concludes that "... experience to date with enterprise zones provides a reasonably clear indication that ... the zones have not proved to be a cost-effective means of providing jobs" (page 207). Specific State Programs Three state programs are of particular interest. Maryland's EZ program is one of the first, and its counties border the D.C. area. Indiana and New Jersey have programs that also began in the early 1980s, and have been evaluated using the preferred econometric techniques discussed above. Maryland The U.S. General Accounting Office (1988) reviews Maryland's enterprise zone program after four years of operation. The Maryland zone businesses were offered investment credits, and general employment credits (that is, they were offered for disadvantaged and nondisadvantaged workers alike). Most of the GAO findings are based on an event study of employment levels in three EZs in the program. Two of the selected zones were considered "best instances" as defined by duration and efficient administration. The GAO finds that while employment increased between 8 percent (63 workers) and 76 percent (555 workers) for participating businesses in the three zones, follow-up interviews with employers indicated that factors other than the program seemed to account for these differences. Indiana Papke (1994) analyzes the effects of the Indiana EZ program on investment and unemployment. Several specifications are used to separate the effects of zone designation from other influences. She examines the effects on the local labor market and on two types of capital investment -- inventories, which were targeted by the investment incentives, and investment in machinery and equipment, which would likely coincide with increased economic development. The estimates indicate that the Indiana EZ program permanently increased the value of inventories by about 8 percent in the zones relative to what it would have been without the program. However, the value of machinery and equipment was reduced by about 13 percent. Based on this evidence, there has been a shift in the type of capital investment. Zone designation also appears to have had a large impact on the local labor market. Unemployment claims declined by about 19 percent following designation. At the mean of unemployment claims, this is about 1,500 fewer claims per year. Since the employment incentives were relatively modest, this improvement may reflect a demonstration effect described by zone administrators. Papke (1993) uses Census data from 1980 and 1990 to compare the economic well-being of Indiana zone residents before and after the EZ program. In spite of the reduction in unemployment rates in the zones, the income numbers suggest that zone residents were not appreciably better off with the Indiana EZ program. Ongoing research by Papke examining migration patterns in and out of the Indiana zones may provide an explanation for these findings if she can determine whether the reduction in unemployment claims found near the Indiana zones are due to increased employment or to migration of unemployed workers away from the zones. New Jersey Boarnet and Bogart (1996) apply econometric techniques from Papke (1994) to the New Jersey EZ program. The New Jersey tax incentives include credits against the state corporation business tax if a firm hires new full-time employees, reduced sales taxes on some purchases, reduced sales taxes on retail sales, and a reduction in unemployment insurance taxes. Boarnet and Bogart find no evidence that the program had a positive effect on municipal employment or on municipal property values. How can these two experiences be reconciled? Of course, the two state programs differ in incentives and in administration and that may account for the differences. But further research may indicate that these two evaluations are not inconsistent with one another. Tax incentives provided to capital may increase investment as Papke (1994) finds. But, as discussed above, the link from capital subsidies to increased employment and improved economic well-being of zone residents is an indirect one. Thus, Boarnet and Bogart (1996) find no change in employment, and Papke (1993) finds no change in the incomes of zone residents. VI. Policy Lessons from the Empirical Literature The studies of the effects of taxes on economic development and business location appear to have come to a consensus that taxes "matter." But that consensus is not wholly satisfying for two related reasons. First, while the majority of the results from the many high quality empirical studies of this issue find taxes to be a statistically significant factor, several others come to the opposite conclusion. Second, because of the somewhat scatter-shot nature of the findings, it is difficult for researchers to give advice to policy makers, who are anxious to know whether their tax policies and tax incentives are likely to be effective economic development tools. In summary, while most researchers find taxes to be a statistically significant factor in business location and expansion decisions, the economic effect of taxes tends both to be small and to be less important than other factors. Labor force availability and quality, for example, appear to be more important for explaining differences across locations in economic activity. How tax revenues are spent tends to be important, important enough that high relative taxes may not be a deterrent to economic growth if the revenues are used to finance services of value to business, such as education and transportation infrastructure. The studies do make clear that a policy of cutting taxes to induce economic growth is not likely to be efficient or cost effective in the general case. In specific cases, where a city's taxes have gotten far out of line or a state's industrial base is particularly sensitive to a specific tax, reductions in taxes may be warranted. But the evidence does not support the blanket use of tax incentives in the name of economic development. The small number of credible empirical studies of enterprise zones has come to a preliminary, incomplete conclusion that the tax incentives offered in zones may well affect the location of capital, but not necessarily employment. The employment and welfare of zone residents do not seem to be improved measurably by zone designation. The tentative policy lesson we can draw from this still evolving literature is that the effectiveness of the tax incentives offered in enterprise zones depends on the policy goal. If the idea is to increase investment in the zone, the evidence indicates that site-specific tax breaks to capital may be effective. If the goal is to improve the employment or welfare of zone residents, the evidence appears to indicate that zone programs, as currently structured, are not sufficient. 1. Recently he DC unemployment rate improved relative to the metropolitan area, with the District rate at 6.8 percent and the rate for the region at 3.4 percent in May, 1997. 2. Actually, jobs in the unidentified industries in the "Other" category also increased. 3. Stephen C. Perry, "Factors Which Influence Business to Stay in or Leave the District of Columbia," George Washington University, Center for the Advancement of Small Business, 1994; "Todd Allan Takes a Walk to Maryland, and It Didn't Need to Happen," Bob Levey's Washington, Washington Post, November 12, 1993; "Playing Games with D.C. Business," Washington Post, May 23, 1997. 4. Government of the District of Columbia, A Vision for America's First City: FY 1997 Budget and Multiyear Plan, (May, 1996). 5. The interest rate on tax exempt debt will be less than that payable on taxable debt, and the difference will be determined by the marginal tax rate. If the marginal tax rate is 30% and the market interest rate is 8%, a borrower would pay 8% on a taxable bond, but only 5.6% on tax exempt debt. The estimate of an annual subsidy of $50 million is based on the assumption of a two percentage point difference between taxable and tax-exempt interest rates. 6. Reliable totals for the Maryland counties in the region could not be obtained. 7. The District refers to enterprise zones as economic development zones. We will use the term enterprise zone. 8. DC Tax Code, Title 5, Chapter 14, and Title 47, Chapter 18. 9. Maryland Department of Business and Economic Development, Business Location Portfolio. 10. Literature provided by Alexandria Economic Development Partnership. 11. The ratio of the value of commercial property to private sector workers in the District is $46,000 per worker, so it is reasonable to associate an improvement worth $100,000 with the hiring of two workers. 12. Literature from Maryland Department of Business and Economic Development. 13. 1996 Virginia Corporation Income Tax Form. 14. Based on conversations with Virginia Department of Housing and Community Development. 15. Based on conversations with the Prince George's County Economic Development Corporation. 16. U.S. Bureau of the Census, Journey to Work Division. 17. Fairfax County Economic Development Authority, Financial Statements. 18. Fairfax County Economic Development Authority, 1996 Annual Report. References Bartik, Timothy J. 1985. "Business Location Decisions in the United States: Estimates of the Effects of Unionization, Taxes and Other Characteristics of States." Journal of Business and Economic Statistics, vol. 3, no. 1 (January), pp. 14-22. Bartik, Timothy J. 1991. Who Benefits From State and Local Economic Development Policies? Kalamazoo, Michigan: W.E. Upjohn Institute for Employment Research. Boarnet, M.G. and W.T. Bogart. 1996. "Enterprise Zones and Employment: Evidence from New Jersey." Journal of Urban Economics, vol. 40, pp. 198-215. Brintnall, M. and R. Green. 1988. "Comparing State Enterprise Zone Programs: Variations in Structure and Coverage." Economic Development Quarterly, vol. 2, no.1, pp. 50-68. Carlton, Dennis. 1983. "The Location and Employment Choices of New Firms: An Econometric Model with Discrete and Continuous Endogenous Variables." The Review of Economics and Statistics, vol. 65, pp. 440-49. Carroll, Robert, and Michael Wasylenko. 1991. "Employment Growth and State Government Fiscal Behavior 1973 to 1987." Chapter 2 in Therese McGuire and Dana Wolfe Naimark (eds.), State and Local Finance for the 1990's: A Case Study of Arizona. Tempe, AZ: Arizona State University School of Public Affairs. Charney, Alberta. 1983 "Intraurban Manufacturing Location Decisions and Local Tax Differentials." Journal of Urban Economics, vol. 14 pp. 184-205. Dalenberg, Douglas R. and Mark D. Partridge. 1995. "The Effects of Taxes, Expenditures, and Public Infrastructure on Metropolitan Area Employment." Journal of Regional Science, vol. 35, no. 4 (November), pp. 617-40. de Bartolhome, Charles A.M. and Mark M. Spiegel. 1997. "Does State Economic Development Spending Increase Manufacturing Employment." Journal of Urban Economics, vol. 41, pp. 153-75. Elling, R.C., and A.W. Sheldon. 1991. "Determinants of Enterprise Zone Success: A Four State Perspective." in Enterprise Zones: New Directions in Economic Development, R.E. Green, ed., Newbury Park, California: Sage Publications, pp. 136-154. Erickson, R.A. and S.W. Friedman. 1991. "Comparative Dimensions of State Enterprise Zone Policies." in Enterprise Zones: New Directions in Economic Development, R.E. Green, ed., Newbury Park, California: Sage Publications, pp. 155-176. Erickson, Rodney and Michael Wasylenko. 1980. "Firm Location and Site Selection in Suburban Municipalities." Journal of Urban Economics, vol. 8, pp. 69-85. Fisher, Ronald C. 1997. "The Effects of State and Local Public Services on Economic Development." New England Economic Review, (March/April), pp. 53-67. Fisher, Peter S. and Alan H. Peters. 1997. "Tax and Spending Incentives and Enterprise Zones." New England Economic Review, (March/April), pp. 109-130. Fox, William. 1981. "Fiscal Differentials and Industrial Location: Some Empirical Evidence." Urban Studies, vol 18, pp. 105-11. Funkhouser, R. and E. Lorenz. 1987. " Fiscal and Employment Impacts of Enterprise Zones." Atlantic Economic Journal, vol. 15, No. 2, pp. 62-76. Gravelle, J.G. 1992. "Enterprise Zones: The Design of Tax Incentives." CRS Report for Congress 92-476 S, Congressional Research Service, The Library of Congress, June 3. Helms, L. Jay. 1985. "The Effect of State and Local Taxes on Economic Growth: A Time Series-Cross Section Approach." The Review of Economics and Statistics, vol. LXVII (November), pp. 574-82. Hines, Jr., James R. 1996. "Altered States: Taxes and the Location of Foreign Direct Investment in America." The American Economic Review, vol. 86, no. 5, pp. 1076-94. Ladd, H.F. 1994. "Spatially Targeted Economic Development Strategies: Do They Work?" Cityscape, vol. 1, no. 1 (August), pp. 93-218. Luce, Thomas F. Jr. 1990. "Local Taxes, Public Services, and the Intrametropolitan Location of Firms and Households." Public Finance Quarterly, vol. 22, no. 2 (April), pp. 139-67. Marlin, Matthew R. 1990. "The Effectiveness of Economic Development Subsidies." Economic Development Quarterly, vol. 4, no. 1, pp. 15-22. McGuire, Therese. 1985. "Are Local Taxes Important in the Intrametropolitan Location of Firms? An Empirical Analysis of the Minneapolis-St. Paul Metropolitan Area." Journal of Urban Economics, vol. 18, pp. 226-34. McGuire, Therese J. 1992. Review of Who Benefits from State and Local Economic Development Policies? By Timothy J. Bartik. National Tax Journal, vol. 45, no. 4, pp. 457-59. McGuire, Therese J. and Michael Wasylenko. 1987. "Employment Growth and State Government Fiscal Behavior: A Report on Economic Development for States from 1973 to 1984." Report Prepared for the New Jersey State an Local Expenditure, Revenue, and Policy Commission, July. McHone, W. Warren. 1984. "State Industrial Development Incentives and Employment Growth in Multistate SMSAs." Growth and Change, pp. 8-15. Mofidi, Alaeddin and Joe A. Stone. 1990. "Do State and Local Taxes Affect Economic Growth?" The Review of Economics and Statistics, vol. 72, no. 4 (November), pp. 686-91. O'Cleireacain, Carol. 1997. The Orphaned Capital: Adopting the Right Revenues for the District of Columbia. Brookings Institute Press. Washington, D.C. Papke, Leslie E. 1987. "Subnational Taxation and Capital Mobility: Estimates of Tax-Price Elasticities." National Tax Journal, vol. XL, no. 2, pp. 191-203. Papke, Leslie E. 1991. "Interstate Business Tax Differentials and New Firm Location: Evidence from Panel Data." Journal of Public Economics, vol. 45, pp. 47-68. Papke, Leslie E. 1993. "What do we know about enterprise zones?" in: J.M. Poterba, ed., Tax Policy and the Economy, vol. 7, MIT Press, Cambridge, pp. 37-71. -----. 1994. "Tax Policy and Urban Development: Evidence from the Indiana Enterprise Zone Program." Journal of Public Economics, vol. 54, pp. 37-49. Rubin, M. 1990. "Urban Enterprise Zones: Do They Work? Evidence from New Jersey." Public Budgeting and Finance, vol. 10, pp. 3-17. Tannenwald, Robert. 1996. "State Business Tax Climate: How Should it Be Measured and How Important Is It?" Federal Reserve Bank of Boston, New England Economic Review, (January/February), pp. 23-38. U.S. General Accounting Office. 1988. Enterprise Zones: Lessons from the Maryland Experience, GAO/PEMD-89-2. Wasylenko, Michael. 1980. "Evidence of Fiscal Differentials and Intrametropolitan Firm Relocation." Land Economics, vol 56, pp. 339-49. Wasylenko, Michael. 1997. "Taxation and Economic Development: The State of the Economic Literature." New England Economic Review, (March/April), pp. 37-52. Wasylenko, Michael and Therese McGuire. 1985. "Jobs and Taxes: the Effect of Business Climate on States' Employment Growth Rates." National Tax Journal, vol. 38, no. 4 (December), pp. 497-512. |
Send mail with questions or comments to webmaster@dcwatch.com
Web site copyright ©DCWatch (ISSN 1546-4296)
