Explanations for the Proliferation of Economic Development Corporations Across North Dakota and South Dakota

The rural areas of the United States have experienced a proliferation of quasigovernmental institutions over the past three decades. The formation of such institutions represents an important form of local boundary change. Local boundaries determine service delivery, economic development, and intergovernmental relationships. It remains unclear, though, how the process of boundary change unfolds. Using federal and state data, I examine the ability of four general explanations of boundary change to account for the proliferation of economic development corporations across North Dakota and South Dakota. I find that their creation is not driven by economic change or need, but is more associated with property taxes per capita.


Introduction
State and local governments want to create an environment that encourages new business investments, but they are leery of implementing developmental policies that would increase the general tax burden. This desire for economic growth at a low public cost has led to the proliferation of quasi-governmental institutions, which typically fund themselves through some means other than a direct tax upon all citizens (Sagalyn 2007) 1 . Previously esoteric entities such as business improvement districts have become commonplace (Morcol and Zimmerman 2006) 2 , nearly every state allows the creation of tax increment financing districts to promote private development (Byrne 2006) 3 , and the proliferation of dependent and independent special districts continues at a rapid pace (U.S. Bureau of the Census 2002) 4 . States have also greatly expanded the use of enterprise zones and tax abatement programs in hopes of attracting economic growth to their most distressed areas   5 .
One of the most prevalent quasi-governmental institutions to arise in recent years is the economic development corporation. Legal definitions vary by state, but economic development corporations (EDCs) are generally non-profit corporations organized and overseen by representatives of business, local government, and the public (Wisconsin State Code 2006) 6 in hopes of improving the economic situation of a specific municipality (Texas State Code 2006) 7 or region (Kansas State Code 2006) 8 . EDCs pursue such strategies as publicizing local business opportunities, providing low-interest loans to businesses, encouraging new employment opportunities, making previously-neglected neighborhoods more attractive, and creating regional development plans. EDCs are funded through a number of sources, including membership dues and contributions from participating businesses and governments (North Dakota State Century Code 2008) 9 , income derived from loan payments and service charges (New Mexico State Code 2006) 10 , and the implementation of specially designated sale taxes (Texas State Code 2006) 11 .
Logically, the creation of quasi-governmental institutions such as business improvement districts and economic development corporations should be driven by local need. Depressed areas with inadequate tax bases are the natural beneficiaries of such institutions. However, the available research suggests caution before accepting this conclusion   12 . Indeed,   13 presents evidence that new governmental institutions are often created by private entrepreneurs seeking to benefit themselves rather than the general public. Focusing upon the formation of cities and special districts, Burns asserts that entrepreneurs who successfully instigate and complete the process of incorporation are well-positioned to institutionalize their own values within the new unit, such as a predilection for low taxes or the provision of certain services. These findings serve as a basis for the subsequent study of local boundary change (Foster 1997 14 ;Feiock and Carr 2001 15 ).
Unfortunately, the U.S. Bureau of the Census excludes economic development corporations from its Census of Governments tabulations and only a few states follow their day-to-day activities. As a result, academic research regarding EDCs remains quite limited (Olberding 2002) 16 . This study seeks to derive a better understanding of economic development corporations through an examination of their uneven proliferation across North Dakota and South Dakota. These two states have been selected for analysis due to their limited governmental resources as well as the stark contrast between the economic prospects of their rural counties versus their more urban counties. Both states also experienced a rapid growth in EDC numbers at nearly the same time. The study uses negative binomial distributions to test whether economic conditions drove EDC proliferation during the 1980s and 1990s or if other, more political, considerations were at work. The study relies upon local boundary change theory to give shape to its analysis.

Explanations for Local Boundary Change
A theoretical perspective that focuses upon the boundaries of local governments may seem rather narrow until one realizes that "...boundaries determine who is included within a jurisdiction and define local arrangements of service provision and production, patterns of economic development, and the exercise of political power" (Feiock and Carr 2001, 383) 17 . From this perspective, then, it is the shared or overlapping boundaries between governments that set the initial parameters for intergovernmental relationships within a county or metropolitan area. One government"s sudden decision to extend its boundaries represents not just an expansion of its service delivery responsibilities and tax base, but a noteworthy projection of political power as well. Likewise, the creation of a new government or quasi-governmental organization alters the local equilibrium. Other governments may react by reevaluating their relationships with one another or creating their own quasi-governmental organizations, which can lead to the eventual transformation of the local political system itself (Perrenod 1984 18 ;Burns 1994 19 2004 21 ); the annexation of unincorporated land by a pre-existing government (Brierly 2004) 22 ; the consolidation of two or more governments into a single unit   23 ; and the legal dissolution of redundant governments (Carr 2004) 24 . Typically, boundary change requires a concerted effort by public and private actors, as well as some involvement by the local electorate.
The literature provides four explanations for the extent and frequency of local boundary change. The first explanation asserts that economic and demographic change will produce a demand for additional public services. These service demands can be met by expanding the size of those local governments already responsible for service delivery, having municipalities annex any needy areas   25 , incorporating new general-purpose governments to provide a complete range of services   26 , or forming special districts and quasi-governmental institutions to offer specific services as required (Smith 1968 27 ;Walsh 1978 28 ;Nelson 1990 29 ;Foster 1997 30 ).
The second explanation claims that state-imposed restrictions upon general-purpose governments can both encourage and discourage boundary change. Indeed, research indicates that state limits upon the ability of municipalities to impose property taxes (MacManus 1981) 31 and create public debt   32 sparked the proliferation of special districts during the 1950s and 1960s as local politicians created new districts to circumvent state restrictions and fund additional services. There is also evidence of a positive relationship between state limits upon the power of municipalities to annex unincorporated areas and the number of special districts within a state (Heikkila and Ely 2003) 33 . Under such limitations, unincorporated areas requiring additional government services cannot expect that their needs will be met via annexation by an already established municipality. These areas are forced into forming additional special districts to alleviate their service demands.
The third explanation proclaims that the creation of governmental units and other such boundary change is not merely an automatic response to economic and demographic fluctuations   34 . Instead, boundary change is often a political act intended to directly benefit a specific group of entrepreneurs. This explanation notes that any group which instigates and successfully completes the process of incorporation is positioned to institutionalize their values within that new government, such as low taxes or the provision and exclusion of certain services. Given these benefits, one would expect an unending series of competitions between various public and private actors to create additional governmental institutions. However, the formation of a new government is not easily done. Indeed, incorporation is often an expensive, time-consuming task with considerable risk of failure   35 . Most actors are better off waiting for someone else to assume this burden, since they will ultimately share in the general benefits provided by a new government anyway. Such circumstances create a conundrum: while it is rational for actors to wait, this also ensures that no one actually attempts to incorporate a new government.
One solution to this paralysis requires an entrepreneur with substantial resources to take on the costs associated with government formation   36 . Such entrepreneurs are undaunted by the prospect of a costly failure and sees the possible institutionalization of their values as a terrific selective benefit. The identification of such entrepreneurs is a source of considerable interest to researchers. Burns emphasizes the role played by real-estate developers and industrial manufacturers in bringing about boundary change while Schneider, Teske and Mintrom (1995) 37 , McCabe (2000) 38 , and   39 stress the importance of "public entrepreneurs." McCabe notes that "...public entrepreneursincluding those who work in universities, civic associations, and businesshave been identified as having an interest in changing the existing local structure for delivering services" (2000, 124) 40 . The creation of a new government is seen as one way of bringing about such change.
The fourth explanation emphasizes the ways in which an initial instance of boundary change can reverberate throughout the local political system and cause other governments to change their own boundaries. These reactions are often done in imitation of the initial boundary change, but can have a preemptive or defensive element as well. For example,   41 described the relationship between the passage of state laws encouraging municipal annexation and the subsequent proliferation of newly incorporated municipalities. She noted that "...where annexation was legal, and citizens thus had reason to worry about being annexed to existing cities with higher taxes" (1994, 80) 42 , citizens were likely to engage in "defensive" incorporations to prevent any unwanted takeovers. Bauroth (2005) 43 found a similar "defensive" dynamic amongst neighboring boroughs in Pennsylvania with the introduction of the municipal authority format during the 1940s.

Economic Development Corporations in North Dakota and South Dakota
The creation of an economic development corporation in North Dakotaknown legally as a "certified nonprofit development corporation"is a relatively straightforward legal process (North Dakota Century Code 2008, 10-33-124) 44 . Any number of North Dakota residents can come together and file articles of incorporation with the secretary of state"s office. These articles specify the corporation name, the nature of its business, the principal place where it will transact that business, the names of its officers and directors, and the means through which the corporation will be governed. To qualify as a "certified nonprofit development corporation," the incorporators must also provide a statement that the corporation "...has adopted a resolution to invest a majority of membership payments, dues, or contributions received in primary sector businesses" (North Dakota Century Code 2008, 10-33-124-2c) 45 . This statement allows a new corporation considerable discretion as to how it will actually encourage economic development. Most corporations, though, engage in such activities as offering low-interest loans to new businesses, marketing the local economy to outside business interests, and overseeing employee training programs. However, none of the corporation"s income can be distributed directly to its members or governing officers.
Certified nonprofit development corporations can receive income from a variety of public and private sources, but they generally rely upon the Certified Nonprofit Development Corporation Investment Credit for funding (North Dakota Office of State Tax Commissioner 2008) 46 . The state allows an income tax credit of up to $2,000 to any individual, estate, trust, or corporation that buys a membership, pays dues, or otherwise contributes funds to a certified nonprofit development corporation (North Dakota Century Code 2008, 10-33-124-4) 47 .
This study searched the North Dakota secretary of state"s corporate registration records (2008) 48 to derive a list of all nonprofit corporations engaged in economic development activities, as well as their years of incorporation and central locations. The corporations under examination here focus their efforts on areas less than the size of a county, typically cities and townships. All nonprofit corporations devoted to low-income housing, though, were excluded from the analysis. The data is illustrated in Figure 1. The incorporation of an economic development corporation in South Dakota is also an uncomplicated legal process (South Dakota Codified Laws 2008,   49 . Three or more South Dakota residents can file articles of incorporation of a non-profit corporation with the secretary of state"s office. These articles specify the corporation name, the nature of its business, the place where it will transact that business, its officers, and the means through which the corporation will be governed. However, most non-profits devoted to economic development are actually incorporated as what the state code calls "local industrial development corporations." These corporations are specifically "...formed for the purpose of furthering the economic development of a community and its environs, and with authority to promote and assist in the growth and development of small business concerns in the areas covered by its operation." (South Dakota Codified Laws 2008, 7-29-24) 50 . Incorporating as an industrial development corporation offers a number of benefits to a non-profit, such as: exemptions from local property taxes and certain state taxes; exemptions from certain fees and bonding requirements; and a formalized process of leasing property from a county (South Dakota Codified Laws 2008, Chapter 5) 51 . Perhaps the most important benefit, though, is that the state and local governments perceive industrial development corporations as legitimate partners in developing the local economy. This is a level of respect that regular non-profit corporations devoted to civic improvements do not necessarily receive.
This study also searched the South Dakota secretary of state"s corporate registration records (2008) 52 to tabulate a list of nonprofit corporations engaged in economic development activities, as well as their years of incorporation and central locations. These corporations focused their efforts on areas less than the size of a county, typically cities or a portion of the county. The data is illustrated in Figure 2.

Explanations for the Uneven Proliferation of Economic Development Corporations in North Dakota and South Dakota: Model and Hypotheses
The dependent variables in this study are the total number of economic development corporations created in North Dakota and South Dakota during the periods of 1987 -1992 and 1992 -1997. The study tests the extent to which the four general explanations for boundary change can account for the uneven proliferation of EDCs within the various local political systems. Thus, the study is not so much concerned in why a specific, individual EDC arose. Instead, it is interested with how the many economic and political forces across a local political system encouraged or discouraged EDC formation. Consequently, the study defines "local political system" as the individual county.
The model used in this study is composed of four sets of independent variables, each set representing a general explanation for local boundary change:

New Economic Development Corporations = service demands + institutional constraints upon local government + entrepreneurship + other local boundary changes.
Some 23 of North Dakota"s 53 counties experienced the creation of at least one EDC between 1987 and 1992 while 21 counties saw at least one new EDC between 1992 and 1997. Likewise, 24 of South Dakota"s 66 counties added at least one EDC between 1987 and 1992 while 15 counties had at least one new EDC form between 1992 and 1997. Given the number of counties in both states receiving a score of "zero" on the dependent variables, the data used throughout this study violates the normal distribution assumptions necessary for ordinary least squares regression. The variances of the dependent variables are also larger than the means, which implies an over-dispersed count variable. As a result, the study uses a negative binomial distribution to examine its hypotheses.

Hypothesis One: The greater the demand for local services within a county, the greater the number of new EDCs.
Counties experiencing economic and demographic change, whether positive or negative, will have to deal with fresh demands by business people, property owners, and residents for additional services. These forceful demands can be met through the existing network of local governments or via the creation of additional cities and special districts (Bollens 1961 53 ; Wood 1961 54 ; Smith 1964 55 ). However, government incorporation is often a long process that does not provide immediate relief to service demands   56 . As a result, there will be a general incentive to maximize the contributions of the pre-existing governments in the hopes that this will be sufficient until more long-term solutions are available. This incentive will lead to the formation of new economic development corporations to assist in the provision of services. The process of creating a new governmental organization can be expensive, risky, and timeconsuming   66 . Only entrepreneurial groups with considerable political and financial resources are likely to see this process through to a successful conclusion   67 . These groups do not expend their resources just to make slight adjustments to the local political system, though. Instead, they seek a reconfiguration of the existing governing arrangements to their own benefit.   68 identified private real-estate developers and manufacturers as the most likely groups to push for boundary change. Her research indicates that private developers call for the creation of special districts to provide services and infrastructure for their properties while manufacturers help incorporate municipalities to keep tax rates low. There is evidence that these groups maintain some influence over these governments for many years after incorporation (Bauroth 2007) 69 .
The measure of local entrepreneurship used in this study is derived from the U.S. Bureau of Census (2001) 70 : the Percentage of Local Earnings Derived from Finance, Insurance, and Real Estate in 1987 and 1992 for each county. This measure is calculated by dividing Earnings in Finance, Insurance, and Real Estate with Total Earnings in All Industries.

Hypothesis Four: The greater the overall levels of boundary change within a county, the greater the number of new EDCs.
Boundary change does not occur in isolation   71 . Shared and overlapping boundaries set the parameters for intergovernmental relationships so that one government"s decision to extend or retract its boundaries reverberates throughout the local political system. Other governments may react by reconfiguring their own boundaries or otherwise enhancing their power, which can lead to a transformation of the local political system (Perrenod 1984 72 ;Burns 1994 73 ). Such a dynamic will lead to the creation of new EDCs as municipalities use all available options in their ongoing strategy to improve their positions.
Boundary change is represented through Change in the Number of Special Districts by county from 1982 to 1987 and from 1987 to 1992. The definition of a special district used in this study follows that of the U.S. Bureau of the Census: special district governments are independent, special-purpose governmental units (other than school district governments) that exist as separate entities with substantial administrative and fiscal independence from general-purpose governments...In order to be counted as a special district government, rather than be classified as a subordinate agency, an entity must possess three attributes: existence as an organized entity, governmental character, and substantial autonomy. (2002, vii) 74 .
The study uses Bureau"s Census of Governments data to identify the number of special districts in each county. It should be emphasized that the Bureau does not consider North Dakota or South Dakota economic development corporations to be special districts (2002) 75 .
A primary difference between the two states is the presence of Job Development Authorities in North Dakota. In 1985, the North Dakota legislature gave county governments the authorization to levy a property tax of up to 4 mills for the purpose of encouraging job creation (North Dakota Century Code 2008) 76 . However, the counties themselves do not maintain absolute control over these revenues. Instead, a county seeking to levy such a tax must create a public agency called a jobs development authority, or JDA, to oversee its dispersal. The intent of the legislature was to create an agency that would "...assist rural communities to diversify their economic bases so the communities would be less dependent on agriculture as the base of the communities" economies" (North Dakota Legislative Council 2007, 2) 77 . The JDA would implement this policy, in part, by entering into contracts with local economic development corporations. To account for this and other statutory differences, a dummy measure for "North Dakota county is included in the model as a control.
Finally, the study includes the number of municipalities within a county for 1987 and 1992. This measure also serves as a control for local population.

Analysis
While the literature recognizes the general importance of quasi-governmental organizations for state and local policy, empirical research is still quite limited. The first step for this study, then, is to examine North Dakota and South Dakota on the relevant variables by county. It should be emphasized that these figures are county means, not statewide averages. The results are presented in Table 1. As can be seen, North Dakota counties had an average population that was approximately 20 percent larger than South Dakota counties. However, North Dakota counties were also experiencing a more rapid decline in population. Both sets of counties underwent a sizable growth in income, though it was greater in South Dakota. North Dakota counties tended to have more municipalities and special districts, but were seeing smaller growth in their special district numbers. South Dakota counties collected higher property tax revenues per capita, while a greater portion of local earnings in North Dakota counties were derived from the finance, insurance, and real estate industries. Finally, North Dakota counties had more economic development corporations incorporate between 1992 and 1997 than South Dakota counties, though both underwent mean growth.
These measures indicate that the economic and demographic condition were fairly volatile in both states during the period under study.
Having examined conditions at the county level, the model can now be employed to consider the extent to which the four general explanations of local boundary change account for the uneven distribution of new EDCs by county. The Negative Binomial Distribution results for 1987-1992 and 1992-1997 are found in Table 2. Districts by county has a significant and positive association with the formation of EDCs for 1987-1992. Consequently, the addition of a single new special district is associated with the creation of an additional 0.31 EDCs during this period. In addition, the number of municipalities within a county has a significant impact for 1987-1992: each additional municipality within a county is related to the formation of 0.095 new EDCs. These results provide some support for Hypothesis Four: boundary change reverberates across the local political system and can bring about additional instances of boundary changes, such as the formation of quasi-governmental institutions. However, both variables proved insignificant for 1992-1997.
Finally, the North Dakota dummy variable was only significant during the period of 1992-1997. Counties scoring "1" on this control measure experienced the creation of an additional 0.86 EDCs.

Conclusion
Shared and overlapping boundaries between local governments set the initial parameters for intergovernmental relationships within a county or metropolitan area   80 .
One government"s decision to extend its boundaries is not a politically neutral event. Instead, it often represents a significant projection of power and responsibilities (Perrenod 1984 81 ;Burns 1994 82 ). The literature provides four general explanations for the prevalence of local boundary change: demographic and economic change produces strong demands for additional services; state statutory limits upon local government behavior forces local politicians into circumventing these statutes; resourceful public and private entrepreneurs instigate boundary change for their own advantage; and, a single instance of boundary change can reverberate throughout the local political system and bring about further boundary changes. This study examined the ability of these explanations to account for the uneven proliferation of economic development corporations across North Dakota and South Dakota by county.
Some important conclusions can be derived from the analysis. First, unemployment and population growth had no impact upon the creation of economic development corporations. These results suggest that the creation and use of such quasi-governmental entities in North Dakota and South Dakota was not driven by in local service demands.
Second, state constraints upon local government in terms of property taxes do affect the creation of additional EDCs. It appears that the proliferation of EDCs was driven at least in part by a desire to circumvent state statutes.
Third, the relative economic strength of finance, insurance, and real estate establishments has no impact upon EDC formation, at least in these two states.
Finally, boundary change does not appear to occur in isolation. The presence of many municipalities as well as the formation of an increasing number of special districts within a county encourages the further proliferation of EDCs, even after controlling for other relevant variables. This finding indicates that the instability caused by such boundary change reverberates throughout the local political system.