Does regionalism beget regionalism? The relationship between norms and regional partnerships for economic development.
Regional governance is on the rise in the United States, according to some scholars and observers (Dodge 1989, 1990; Wallis 1994a). In other words, local government officials are increasingly working together to address interjurisdictional problems and issues. One strategy of regional governance that is being used more often is voluntary groups of local governmental officials--and often business leaders and other citizens--in a region. A number of these "regional partnerships" have been formed specifically to foster the economic development of a multijurisdictional or regional area. Olberding (1997) finds that the number of regional partnerships for economic development increased fivefold during the past decade in large metropolitan areas in the southeastern United States--from three in 1987 to 16 in 1997.
The recent growth in regional partnerships for economic development as a public administration/public policy strategy, however, has not been matched by scholarly research. To date, no study has empirically examined the formation of these regional partnerships across a large number of areas in the United States. Further, no study has looked at the organizational characteristics of a representative sample of regional partnerships for economic development in the United States. Therefore, we have a very limited understanding of why regional partnerships form, how they are structured, and what they do. This article attempts to address the shortcomings by developing a theory of regional partnerships for economic development and then testing it using data from all metropolitan areas in the United States and from a national survey to which executive directors of 133 regional partnerships responded.
The Literature on Regionalism
Two Competing Models of Interlocal Relations
There are two competing models of how local governments in a region relate to one another. On the one hand, there are "interjurisdictional competition" models, which assert that cities rival one another for residents and businesses. Tiebout (1956) presents the notion that there is an optimal number of residents and businesses at which communities produce a bundle of public services at the lowest average cost. A community below the optimum tries to attract residents and businesses; a community at the optimum tries to maintain it; and a community above the optimum does not try to repel residents and businesses, but rather economic factors "push people out of it." (1) If a number of citizens with similar preferences for public services and taxes are not satisfied with the current offerings of local jurisdictions, then new jurisdictions are formed. Proponents argue that multiple, competing cities result in an efficient outcome because it enables citizens and businesses to choose jurisdictions with public services and taxes that most closely match their preferences. In addition, public-choice theory suggests that competition improves democracy because politicians must be responsive to mobile constituents (Roeder 1994).
In contrast to models based on fragmentation and competition, a second set of models is based on "regionalism" by limiting the number of local governments or by fostering coordination and cooperation among them. Public administration traditionalists assert that fewer local governments result in economy-of-scale benefits, greater political accountability, more equitable treatment of citizens, and greater opportunity to address significant problems (Lyons, Lowery, and DeHoog 1992). Further, urban scholars maintain that economic and social linkages among cities in metropolitan areas are strengthening (Hershberg 1996; Wallis 1994a; Florestano and Wilson-Gentry 1994; Savitch et al. 1993; Grell and Gappert 1993; Peirce 1993a,c,d; Hershberg, Magidson and Wernecke 1992). Proponents of regionalism have asserted that a more optimal outcome is achieved when local governments recognize their interdependencies and act in a coordinated way (Barnes and Ledebur 1998; Dodge 1996; Wallis 1994a; Peirce 1993a, c; Grell and Gappert 1992; Barnes and Ledebur 1991; Dodge 1990).
Two Approaches to Regionalism
Regional strategies can be categorized by the number of local government activities they are designed to address and the degree of difficulty in implementing them. The first category includes strategies that attempt to coordinate a large number of services across local governments--or even to coordinate local government in general--and thus are relatively difficult to implement. These "broad-sweeping" strategies include metropolitan government, consolidated government, annexation, and regional councils of government. The second category includes strategies that attempt to coordinate only one or a few services across local governments and thus are relatively easy to implement. These "targeted" strategies include regional partnerships, written agreements among local governments, and special districts.
Strategies to "regionalize" a broad array of local government activities have never been extremely popular and are becoming even less so (Rusk 1995; Dodge 1990). The creation of an additional tier of government at the regional level has not caught on in the United States, as there is only one or a few metropolitan governments? In addition, although proposals to consolidate city and county governments have been approved 21 times, they have been rejected nearly 100 times (Clark 1995; Walker 1987). Finally, the number of regional councils of government has decreased by one-third during the past few decades--from a peak of 669 in 1977 to 450 in 2000. The decline in broad-sweeping regional strategies supports the observation that: "A generation ago, metropolitan-wide governments were frequently proposed as a way to get suburbs to pay their fair share of taxes and to get government services delivered more efficiently. But the political resistance to that approach was fierce, and these days few people believe that rejuggled government is the most promising choice" (Peirce 1993c, 1934).
In contrast, the use of targeted regional strategies has been rising steadily. "The federally mandated regional planning efforts and the consolidation proposals of the past three decades are being replaced by voluntary cooperation among governments and sectors through public-private, `intercommunity partnerships'" (Dodge 1990, 354). The number of regional partnerships for economic development increased fivefold in the southeastern United States from the 1980s to the 1990s (Olberding 1997). In addition, special districts have increased in number from 8,000 in 1942 to more than 30,000 today (Saffell and Basehart 1997).
Regional Partnerships for Economic Development
Literature focusing specifically on regional partnerships for economic development is limited. It consists of a handful of case studies (Raasch and Brooks 1995; Grell and Gappert 1993; Hershberg, Magidson and Wernecke 1992; Coe 1992; Huggins 1992) and one published study that provides information on 13 regional partnerships for economic development in U.S. metropolitan areas (Bennett and Nathanson 1997). Based on a review of these case studies, a few factors seem to influence the formation of regional partnerships for economic development. One factor is a tradition of regionalism among local government officials in a region (Olberding 1997; Heath and Henegar 1994; Grell and Gappert 1993). Olberding (1997) finds that metropolitan areas that had used another regional strategy--consolidated city-county government--were more likely to adopt a regional partnership for economic development than metropolitan areas that did not have consolidated government.
In addition, case studies suggest that need may influence the formation of regional partnerships for economic development. High unemployment or the loss of a large number of jobs has motivated local governments to work together in some regions (Coe 1992; Huggins 1992). Further, limited resources of local governments can influence regional cooperation. For example, Ehrenhalt (1995) says that local government officials in the Pittsburgh metropolitan area were motivated to work together because some cities were struggling to finance police, street maintenance, and other public services, and a few cities had even gone bankrupt. Another dimension of need that may impact the formation of regional partnerships for economic development is the degree to which cities in a metropolitan area are similar economically. Some scholars have found that large economic differences among cities in a region demonstrate a need for a regional approach (Olberding 1997; Nunn and Rosentraub 1996; Hershberg, Magidson, and Wernecke 1992); however, others have concluded that large differences in local economies make it more difficult to pursue a regional strategy (Peirce 1993a; Dye 1964).
As with the formation of regional partnerships for economic development, their organizational structure and processes have received limited attention from scholars. Case studies of regional partnerships have described the membership, staff, resources, and activities of a few individual groups (Raasch and Brooks 1995; Grell and Gappert 1993; Coe 1992; Hershberg, Magidson, and Wernecke 1992; Huggins 1992). In addition, Bennett and Nathanson (1997) survey 13 regional partnerships for economic development and find tremendous variation in their structure and processes. For example, their staffs ranged from three to about 70, while their budgets ranged from $750,000 to $6 million (Bennett and Nathanson 1997). The authors report that these organizations' activities included direct contact with prospects, media relations, direct mail, special events, trade shows, international trade shows, and advertising. Olberding (1997) finds that 16 regional partnerships in the southeastern United States also collect and analyze data, manage programs to assist small business owners and entrepreneurs, and plan and prioritize improvements to the region's infrastructure and work force.
Theory of Regional Partnerships for Economic Development
Scholars have long recognized the difficulty of achieving and sustaining voluntary cooperation among a large number of individuals with no central authority--the so-called "dilemma of collective action" (Axelrod 1986, 1997; Ring and Van de Ven 1994; Van de Ven 1976a; Olson 1965). Axelrod (1986, 1997) argues that, in situations with large numbers of individuals or groups, the solution to this dilemma is cooperative norms--or the extent to which parties usually act in a collaborative or coordinated manner or are expected to act in such a manner: "Today, norms still govern much of our political and social lives.... Large numbers of individuals and even nations often display a great degree of coordinated behavior that serves to regulate conflict. When this coordinated behavior takes place without the intervention of a central authority to police the behavior, we tend to attribute the coordinated behavior and the resulting regulation of conflict to the existence of norms" (Axelrod 1997, 47).
Other scholars have also concluded that cooperative norms--or something conceptually similar--are critical for shifting from competitive to cooperative behavior. For example, in her comprehensive review of the cooperation literature, Ostrom (1998) concludes that the key determinant of cooperation is "norms of reciprocity"--or the tendency of individuals "to react to the positive actions of others with positive responses and the negative actions of others with negative responses" (10). In addition, she maintains that "two ways out of social dilemmas" are face-to-face communication among parties and rules, sanctions, and institutions that support and enforce cooperation. Similarly, Axelrod (1986, 1997) asserts that certain mechanisms support norms and thus help to achieve and sustain cooperation among large numbers or individuals or organizations. One of these mechanisms is voluntary groups.
The norms-based theory of cooperation is appropriate for the study of regional partnerships for economic development. These regional partnerships are voluntary groups formed to support collaborative behavior in a large-number situation in which there is no central authority. The norms-based theory of regional partnerships for economic development predicts that cooperative norms in a metropolitan region are positively related to the formation of regional partnerships. In addition, there are predictions regarding the relationship of norms to various organizational characteristics of the regional partnerships (see figure 1). These predictions are discussed in greater detail below.
[FIGURE 1 OMITTED]
There are three groups whose norms may be relevant to the formation of regional partnerships. Case studies have documented that two groups have been actively involved in the formation and maintenance of regional partnerships for economic development: government officials and business leaders (Raasch and Brooks 1995; Grell and Gappert 1993; Coe 1992; Hershberg, Magidson, and Wernecke 1992; Huggins 1992). These findings are consistent with literature on economic development policy in general, which has recognized that the key actors are government officials and business leaders (Gray and Lowery 1990; Eisinger 1988; Peterson 1981; Lindblom 1977). In addition, there is a third group of individuals who may influence regional partnerships for economic development: citizens. Rusk (1995) finds that civic involvement played "a vital role" in the adoption of another regional strategy--city-county consolidation. Further, Cable, Feiock, and Kim (1993) conclude that citizen access to governmental institutions has some effect on economic development policies adopted by local governments. Based on this literature, the following hypothesis is made:
[H.sub.i]: Cooperative norms in government, in the private sector, and in the citizenry are positively related to the formation of regional partnerships for economic development.
The norms-based theory can be enriched by considering the environment in which local governments, businesses, and citizens act. Particularly relevant to regional partnerships for economic development are the needs of the region. Case studies of regional partnerships for economic development have found that economic need is a key factor in the formation of regional partnerships (Coe 1992; Huggins 1992). In addition, political scientists have asserted that social needs influence public policy outcomes (Peterson 1981, 1995; Dye 1990). For example, Dye (1990) finds that the need for education--measured as the percentage of school-aged children in the state--was the most important determinant of state educational expenditures. Further, organizational scholars have identified need--or, to an extreme, threat to survival--as an important determinant of cooperation among organizations (Browning, Beyer, and Shetler 1995; Astley 1984). In light of these findings, a second hypothesis is made:
[H.sub.2]: Need is positively related to the formation of regional partnerships for economic development.
Based largely on the work of Van de Ven (Ring and Van de Ven 1994; Van de Ven 1976a), it is asserted that cooperative norms influence not only the formation of interorganizational relationships, but also their structure and processes. Van de Ven (1976a) identifies three dimensions of structure in interorganizational relationships: (1) formalization through agreements or contracts; (2) complexity, or the number of organizations involved in the relationship and the number of projects and tasks undertaken through the interorganizational relationship; and (3) centralization, or the degree of concerted decision making by a committee or board of individuals. (3) In addition, he identifies two dimensions of process in interorganizational relationships: (1) intensity of resource flows, or the amount of resources moving among organizations; and (2) intensity of information flows, or the frequency of communication among organizations. Van de Ven (1976a) makes specific hypotheses about each of the dimensions. Based on his work, a final hypothesis is made:
[H.sub.3]: Cooperative norms are related to the structure and processes of regional partnerships for economic development. Specifically, cooperative norms are negatively related to the formalization and centralization of regional partnerships and positively related to their membership complexity, task complexity, intensity of resource flows, and intensity of information flows.
Research Design and Methodology
Analyses of the Formation of Regional Partnerships
The first set of analyses is designed to explain why regional partnerships for economic development are formed. The population comprises all areas in the United States defined by the Census Bureau as consolidated metropolitan statistical areas (CMSAs), metropolitan statistical areas (MSAs), and New England county metropolitan areas (NECMAs) from 1980 to 1997--the period during which data were collected. A total of 244 observations are included in the first set of analyses.
No listing of regional partnerships for economic development per se had existed to this point; therefore, regional partnerships were identified using a systematic and comprehensive search process. Listings of economic development groups in the following publications were consulted to identify those that are regional in scope: Site Selection magazine, the Worldwide Chamber of Commerce Directory, and the National Council for Urban Economic Development's Trends in Economic Development Organizations. In addition, a search of Internet Web sites was conducted using various combinations of keywords such as "regional" and "economic development. (4) Finally, officials in state economic development departments who work with local governments were contacted by telephone and interviewed regarding groups in their state that "fit" the definition of regional partnerships for economic development. Some state officials provided information over the telephone, while others sent lists of regional partnerships. This process identified 191 regional partnerships for economic development in 147 metropolitan areas, indicating that some metropolitan areas had more than one regional partnership.
The first set of analyses consists of two models, each of which has a unique dependent variable but the same independent variables. In the first model, the dependent variable is whether or not a metropolitan area had at least one regional partnership for economic development in 1997 (Regional Partnership Existence). In the second model, the dependent variable is the total number of regional partnerships in 1997 (Total Regional Partnerships). In both models, key determinants are cooperative norms--in government, in the private sector, and in the citizenry--as well as need. The two models are represented by the following equations:
Regional Partnership [Existence.sub.i] = [G.sub.i] + Business [Associations.sub.i] + Civic [Associations.sub.i] + [E.sub.i] + e; and Total Regional [Partnerships.sub.i] = [G.sub.i] + Business [Associations.sub.i] + Civic [Associations.sub.i] + [E.sub.i] + e.
[G.sub.i] is a vector of variables measuring cooperative norms in government in metropolitan area i. Based on the notion that a higher number of governments results in greater competition (Sbragia 2000), three variables in this vector measure government fragmentation and thus are inversely scaled measures of cooperative norms. They include Fragmentation of City Government, or the total number of municipal, town, and township governments in each metropolitan area in 1982; Fragmentation of County Government, or the total number of county governments in 1982; and Fragmentation of Special Districts, or the total number of special-purpose governments in 1982. Three additional variables in [G.sub.i] measure the degree to which other types of regional strategies have been pursued in each metropolitan area. They include Consolidated Government, whether metropolitan area i had a consolidated city-county government; COG Strength, the total number of employees of regional councils for government in 1987--the first year for which these data were available; and COG for Economic Development, which equals 2 if a metropolitan area had a COG designated as an economic development district by the federal government, 1 if a metropolitan area had a COG that was not so designated, and 0 if a metropolitan area did not have a COG in 1987--again, the first year for which these data were available.
Business Associations Per [Capita.sub.i] is a measure of cooperative norms in the private sector. It is the total number of chambers of commerce relative to the total population in metropolitan area i in 1987--the first year for which these data were available. Ideally, there would be multiple variables to measure cooperative norms in the private sector, such as the total number of members in all chambers and the total amount of resources of all chambers; however, accurate data were not available for these measures. Civic Associations Per [Capita.sub.i] is a measure of cooperative norms in the citizenry. It is the total number of membership organizations--or "social structures of cooperation" (Putnam 1993, 89)--relative to the total population in each metropolitan area in 1980.
[E.sub.i] is a vector of environmental factors and control variables. It consists of the following variables measuring economic need in metropolitan area i: the Employment Rate in 1984; Per Capita Income in 1983, adjusted to 1994 dollars; Economic Disparity, the absolute difference between central-city per capita income and suburban per capita income relative to the metropolitan area per capita income in 1983; and Government Revenues Per Capita, the total revenues of all local governments in 1982 relative to the population of the metropolitan area. [E.sub.i] also includes one variable measuring another type of need, need for regionalism. Regional Lifestyle is the percentage of citizens in metropolitan area i who lived in one county but worked in another in 1980. In addition, this vector includes the following control variables: Local Authority for Regional Strategies, an index of power granted by the state for local governments to pursue regional strategies, such as consolidation, annexation and interlocal agreements (5); Population, the total number of residents of metropolitan area i in 1984; Population Density, the total population in 1984 relative to the amount of land in metropolitan area i; and three dummy variables measuring location in a particular region of the country (for instance, Northeast, Midwest, and West).
Because the dependent variable in the first model (Regional Partnership Existence) is dichotomous, logistic regression is used to analyze the relationships between the independent variables and the probability that a regional partnership is formed (Dielman 1996). Because the dependent variable in the second model (Total Regional Partnerships) is continuous, multiple regression is used to analyze the relationships between the independent variables and the number of regional partnerships (Dielman 1996).
Both models were examined for multicollinearity among independent variables, heteroscedastic residuals, and outliers and leverage points. In the first analysis, in which Regional Partnership Existence is the dependent variable, 21 of the 244 metropolitan areas had to be dropped because of missing data, leaving 223 observations. (6) In addition, based on a review of standardized residuals and Cook's D, two observations were determined to unduly influence the regression analysis and thus were dropped, leaving 221 observations in the final analysis. In the second analysis, in which Total Regional Partnerships is the dependent variable, all 244 metropolitan areas were included. Based on a review of standardized residuals and Cook's D, one observation was determined to have an excessive influence on the regression analysis and thus was dropped, leaving 243 observations in the final analysis.
Finally, it should be noted that the earliest data available for a number of the independent variables in both formation models is for 1987; however, the mail survey of executive directors determined that some regional partnerships had been formed before this time. Therefore, in an effort to mitigate any causality problems, the models are run both on all metropolitan areas--those without regional partnerships, those with partnerships formed after 1987, and those with partnerships formed 1987 and earlier (N=223 for the first model)--and on a reduced, "time-ordered" population that includes only metropolitan areas without regional partnerships and those with partnerships known to have been formed after 1987 (N=192 for the first model).
Analyses of the Structure and Processes of Regional Partnerships
The second set of analyses is designed to explain variation in the structure and processes of regional partnerships for economic development. In order to obtain information on structure and processes, a survey was designed and sent to executive directors of the 191 regional partnerships identified in 1997. Executive directors of 133 regional partnerships (70 percent) completed and returned surveys, making up the sample for the second set of analyses
The survey asked questions regarding the number of written agreements developed during the formation of the regional partnership, the total number of individuals involved in the regional partnership, the types of data collection and dissemination activities, the types of economic development activities, the annual budget, and the number of board meetings per year. The survey also included seven items related to cooperative norms among local government officials in the region and seven items related to norms among business leaders in the region. (see the appendix) The items were adapted from measures of norms that were developed and tested in research of industrial buyer-seller relationships (Doney and Cannon 1997; Heide and John 1992). Each item was followed with a 7-point Likert scale, with 1 being completely inaccurate and 7 being completely accurate.
Principal component analysis was conducted on the seven items measuring norms among local government officials. It determined that all of the items captured one component or dimension--presumably, cooperative norms in local government. Therefore, one variable was developed from these seven items and labeled Government Norms. Similarly, principal component analysis was conducted on the seven items measuring norms among business leaders, and it concluded they captured one component or dimension--presumably, cooperative norms in the private sector. Thus, one variable was developed from these seven items and labeled Private-Sector Norms.
The structure and process analyses consist of seven models, each of which has a unique dependent variable but the same independent variables. The dependent variables are the seven structure and process variables identified by Van de Ven (1976a). In all seven models, key determinants are cooperative norms--in government, in the private sector, and in the citizenry. The are represented by the following equation:
Structure/[Process.sub.i] = Government [Norms.sub.i] + Private Sector [Norms.sub.i] + Civic [Associations.sub.i] + [E.sub.i] + [S.sub.i] + e.
Structure/[Process.sub.i] is a vector of variables measuring organizational structure and processes of regional partnerships for economic development in 1997. Based on Van de Ven's work, this vector includes the following variables: Formalization, an index of the number of written documents developed during formation of the regional partnership; Centralization of Decision Making, the inverse of the number of persons who serve on the governing board; Membership Complexity, the total number of members in the regional partnership; Information Task Complexity, the number of data collection and dissemination activities undertaken by the partnership; Economic Task Complexity, the number of economic development activities undertaken by the partnership; Resource Flow Intensity, the regional partnership's total annual revenues; and Information Flow Intensity, the number of meetings held by the regional partnership's governing board.
Government [Norms.sub.i] is the variable developed during the principal component analysis of seven survey items designed to measure cooperative norms among local governmental officials in each metropolitan area. Private-Sector [Norms.sub.i] is the variable developed during the principal component analysis of seven survey items designed to measure cooperative norms among business leaders in each metropolitan area. Again, Civic Associations Per [Capita.sub.i]--a measure of citizens' norms--equals the total number of membership organizations relative to the total population in each metropolitan area in 1980.
[E.sub.i] is a vector of environmental factors and control variables. As in the formation analyses, [E.sub.i] in the structure and process analyses includes Employment Rate, Per Capita Income, Economic Disparity, Government Revenues Per Capita, Regional Lifestyle, Population, and Population Density. (It does not include Local Authority for Regional Strategies or the regional dummy variables.)
For each of the seven models, the six structure and process variables, which are not the dependent variable, are included in [S.sub.i] as independent variables. For example, in the Formalization model, [S.sub.i] includes Centralization of Decision Making, Membership Complexity, Information Task Complexity, Economic Task Complexity, Resource Flow Intensity, and Information Flow Intensity. In all seven models, [S.sub.i] also includes Age of Regional Partnership, measured in years. This vector is built into the models because it is expected that the inclusion of these variables will help to explain variation across regional partnerships and reduce specification bias.
Because all of the seven dependent variables are continuous, the statistical technique used to analyze the relationships between the independent variables and the dependent variable in each model is multiple regression (Dielman 1996).
Each of the structure and process models was examined and tested for multicollinearity among independent variables, heteroscedastic residuals, and outliers and leverage points. Variation inflation factors for all seven models were well below 7; therefore, multicollinearity was minimal. The three models with Centralization, Membership Complexity, and Resource Flow Intensity as dependent variables initially had heteroskedastic residuals, indicating potential violations of the assumptions of common variance and normality. Thus, each of these dependent variables was transformed using the natural log. After these transformations, probability plots indicated the residuals were homoskedastic. Finally, standardized residuals and Cook's D were used to identify observations as potential outliers and levers. At least one outlier or lever was identified in all of the models and dropped from the analysis.
Findings from the Formation Analyses
As stated earlier, 147 of the 244 metropolitan areas included in the analysis have at least one regional partnership for economic development. Some metropolitan areas have more than one. For example, there are six regional partnerships in Los Angeles-Riverside-Orange County, California, and in Portland-Salem, Oregon. Survey results confirm an upward trend in the number of regional partnerships during the past two decades, as some observers have noted (Olberding 1997; Rusk 1995; Dodge 1990). Of the 133 regional partnerships for which surveys were completed and returned, 108 (about 80 percent) were formed during 1980-97, while 28 (about 20 percent) were created during the 40 years prior to 1980.
Table 1 presents results of the analyses of the formation of regional partnerships for economic development in U.S. metropolitan areas. The first model--with Regional Partnership Existence as the dependent variable--is significant at the 0.000 level ([chi square] = 62.391). Overall, the model accurately predicts which metropolitan areas have regional partnerships and which do not in 72 percent of the cases. It has an [R.sup.2] equal to 0.336, meaning the model explains about one-third of the variation in whether a metropolitan area has a regional partnership for economic development. The second model--with Total Regional Partnerships as the dependent variable--is significant at the 0.000 level (F = 10.693). The [R.sup.2] of 0.477 indicates this model explains about one-half of the variation in the total number of regional partnerships for economic development in each metropolitan area. (7)
Results indicate some support of the hypothesis that cooperative norms are positively related to the formation of regional partnerships for economic development ([H.sub.1]). In the first model in which Regional Partnership Existence is the dependent variable, seven of the eight variables measuring cooperative norms have coefficients with signs in the predicted direction. Two of the seven--Fragmentation of City Government and Fragmentation of Special Districts--are significant at the p <. 10 level or better. In the second model, in which Total Regional Partnerships is the dependent variable, six of the eight variables measuring cooperative norms have coefficients with signs in the predicted direction. Three of the six--Fragmentation of City Government, COG Strength, and Civic Associations Per Capita--are significant at the p <. 10 level or better.
The negative and significant coefficients on Fragmentation of City Government in both models and on Fragmentation of Special Districts in the first model indicate that fragmentation of local government in a metropolitan area is inversely related to the formation of regional partnerships for economic development. In other words, the greater the degree of fragmentation and competition, the lesser the degree of regional coordination and cooperation, and the less likely regional partnerships for economic development are formed. Also as expected, the coefficient on COG Strength is positive and significant m the second model. This indicates that the stronger another type of regional strategy--the regional council of government--the more regional partnerships are formed.
In addition, one finding supports the expectation that citizens' norms are related to the formation of regional partnerships for economic development. The positive and significant coefficient on Civic Associations Per Capita in the second model indicates the more "social structures of cooperation" in a metropolitan area (Putnam 1993), the more regional partnerships for economic development are formed.
Further, results of the first analysis provide some empirical support that need is positively related to the formation of regional partnerships for economic development ([H.sub.2]). As expected, there is a negative and significant coefficient on Employment Rate in the first model. This indicates that the lower the employment level in a metropolitan area, the greater the economic need and the more likely that a regional partnership for economic development is formed. (It should be noted that, in the second model, the coefficient on Employment Rate is negative and nearly significant with a p-value of .12.) Also, there is a negative and significant coefficient on Government Revenues Per Capita in the second model. This indicates that the fewer financial resources local governments have, the greater the economic need and the more regional partnerships for economic development are formed. (It should be noted that, in the first model, the coefficient on Government Revenues Per Capita is negative and nearly significant with a p-value of. 11.)
Finally, there are positive and significant coefficients on Regional Lifestyle in both models. These findings indicate that the greater the degree to which citizens commute across jurisdictional lines in a metropolitan area, the greater their need for a regional strategy and the more likely regional partnerships for economic development are formed.
Results of the Structure and Process Analyses
The survey results indicate that regional partnerships vary noticeably in their structure and process characteristics. For example, membership in regional partnerships ranged in number from six to 1,739 in 1997, while financial resources ranged from zero to $15 million. In addition, governing boards of regional partnerships held a minimum of zero to a maximum of 24 meetings in 1997. Further, the number of economic development programs implemented by regional partnerships ranged from one to 18 in 1997, while the number of data gathering and dissemination programs ranged from zero to nine.
Because the hypothesis being tested is that norms are related to the structure and processes of regional partnerships, results of the seven analyses are displayed together in table 2. All models are significant at the 0.000 level. The models differ in the amount of variance they explain--from 22.7 percent in the Information Flow Intensity model to 53.0 percent in the Economic Task Complexity model.
There is weak support that cooperative norms are related to the organizational structure and processes of regional partnerships for economic development ([H.sub.3]). Of the 21 variables measuring norms (three in each of the seven models), only six are significant at the p < . 10 level or better. Four of the six significant variables have signs in the predicted direction.
Of the three sets of norms, those in government provide the most support of [H.sub.3]. Government Norms--the variable developed during principal component analysis based on responses by executive directors of regional partnerships to seven survey items--is significant and in the predicted direction in three of the seven models. (8) In the model estimating Formalization of regional partnerships for economic development (Formal in table 2), the negative and significant sign on Government Norms indicates that metropolitan areas with strong cooperative norms in government have regional partnerships that are less formalized through written documentation. In the model estimating Centralization of Decision Making (LNCENT in table 2), the negative and significant sign on Government Norms indicates that metropolitan areas with strong cooperative norms in government have regional partnerships with less centralized decision making--that is, more members on the governing board. In the model estimating Information Flow Intensity (Infoflow in table 2), the positive and significant sign on Government Norms indicates that metropolitan areas with strong cooperative norms in government have regional partnerships with more intense information flows through meetings of the governing board.
The first set of analyses confirms the observation from case studies that a tradition of regional cooperation--or strong cooperative norms--is related to the formation of regional partnerships for economic development. What does this conclusion mean for local governmental officials and others who are considering a regional partnership as a strategy to enhance economic development? Do the findings imply that individuals in areas with highly fragmented and competing local governments--indicating weak cooperative norms--should not attempt this strategy? Not quite. They suggest that forming a regional partnership may be more difficult in such areas. But the results also indicate that weak cooperative norms may be overcome by other factors, particularly economic need and a need for regionalism. Further, there may be other factors that could not be included in this national study but may have a significant influence on the formation of regional partnerships, particularly the role of individual leaders. "It is not natural for leaders in any community to think extraterritorially.... It takes a special sensitivity to explore the implications of an issue on other communities and to initiate conversations with the leaders of these communities" (Dodge 1989, 8).
Individuals who are interested in forming a regional partnership but are located in an area with weak cooperative norms may want to attempt this strategy under the following conditions: (1) if and when economic need is relatively high, such as after the closing of a large manufacturing plant in the area or another event that results in a noticeable loss of jobs; (2) if and when a large percentage of residents have a "regional lifestyle," thus indicating a need for a regional approach to economic development; and (3) if and when key government and business leaders in the area have "bought into" the concept of a regional partnership and have vocalized their support for this strategy publicly.
The second set of analyses reveals that not every regional partnership is alike. Far from it. Regional partnerships vary greatly in their formalization, centralization of decision-making authority, total membership, number of economic development programs, data collection and dissemination activities, financial resources, and annual meetings. In addition, the second set of analyses finds that cooperative norms are weakly related to the structure and processes of regional partnerships for economic development. What are the practical implications of these findings? The most conclusive observation that can be drawn is that while cooperative norms seem to influence the formation of regional partnerships for economic development, they have a limited impact on how they look and what they do. Thus, regarding the influence of cooperative norms, it seems that the hurdle for practitioners interested in the strategy of a regional partnership may lie in the initial formation. Beyond that, norms seem to have only a subtle influence on the organizational structure and processes of regional partnerships.
Theoretical and empirical work on regional partnerships has been limited despite the tremendous increase in their use as a public administration/public policy strategy during the past few decades. This article is the first attempt to examine the formation, structure, and processes of regional partnerships for economic development. However, it does not look at the performance or effectiveness of these regional partnerships. In other words, do metropolitan areas with regional partnerships experience greater economic development than metropolitan areas without them? This research question--one of the most important from a public administration perspective--is the focus of a working paper (Olberding 2001).
In addition, there are other avenues for additional research of regional partnerships. One is to examine the involvement of government officials and business leaders in regional partnerships for economic development and the degree to which the public-private mix influences their activities and performance. In addition, a follow-up study could be conducted in a few years to determine whether the number of regional partnerships has continued to increase in the twenty-first century, whether it has stayed about the same, or whether it has decreased. (The last-named observation would indicate these groups were simply a fad of the 1980s and 1990s.) A follow-up study could also examine the impact of cooperative norms and need on the maintenance of regional partnerships and on their organizational characteristics. A final direction for future research is to study voluntary groups that have been formed in .policy arenas other than economic development, such as environmental management. In short, the phenomenon of regional partnerships seems to be a fertile area for future research.
Survey Items Measuring Cooperative Norms
A. Among local government officials
1. Local governmental officials in my metropolitan area are committed to positive change for the entire region.
2. In my metropolitan area, it is expected that local officials in one jurisdiction keep those in other jurisdictions informed of changes that may affect them.
3. Local governmental officials from different jurisdictions in this region keep promises that they make to one another.
4. Economic problems in my metropolitan area are addressed jointly by local officials across the region.
5. Local officials from various jurisdictions across this region frequently communicate face-to-face with one another.
6. In my metropolitan area, local governmental officials from different jurisdictions trust one another.
7. Local officials across this region area are interested in improving only their own jurisdiction (negatively worded).
B. Among business leaders
1. Economic problems in my metropolitan area are addressed jointly by business leaders across the region.
2. In my metropolitan area, business leaders trust one another.
3. Business leaders across this region frequently communicate face-to-face with one another.
4. Business leaders in this region are interested in improving only the jurisdiction in which their business is located (negatively worded).
5. In my metropolitan area, it is expected that business leaders keep each other informed of changes that may affect them.
6. Business leaders across this region keep promises that they make to one another.
7. Business leaders in my metropolitan area are committed to positive change far the entire region.
Table 1 Results of the First Set of Analyses: The Formation of Regional Partnerships Dependent variables REGPART Independent Expected variables direction betas FRAGCITY - -0.019 ** FRAGCNTY - 0.161 FRAGSPEC - -0.009 * CONSOL + 0.887 COGED + 0.151 COGSTRNG + 0.005 BUSPC + 13101.64 CIVPC + 1282.79 EMPRATE - -12.816 * INCOME - 6.87E-05 REGLIFE + 6.933 *** GOVREVPC +/- -0.001 ECONDISP +/- 0.702 LOCAUTH + 0.172 ** POP . 2.80E-06 *** POPDEN . 0.001 NEAST . 2.023 ** MIDWEST . 1.446 ** WEST . 1.470 ** *** = p < .01 Percent correct = 71.7 ** = p < .05 [X.sup.2] = 62.391 * = p<.10 [R.sup.2] = 0.336 N = 219 N = 243 TOTPART Independent variables Betas FRAGCITY FRAGCNTY -6.55E-03 *** FRAGSPEC -1.05E-04 CONSOL 1.42E-03 COGED -0.109 COGSTRNG 3.47E-02 BUSPC 1.11 E-02 *** CIVPC 1497.73 EMPRATE 401.70 * INCOME -2.778 REGLIFE 4.76E-05 GOVREVPC 1.973 *** ECONDISP -1.83E-04 * LOCAUTH 0.561 POP 3.50E-02 * POPDEN 3.35E-07 *** NEAST 2.29E-04 MIDWEST 0.418 ** WEST 0.384 ** *** = p < .01 0.560 *** ** = p < .05 Sig. < .000 * = p<.10 F = 10.693 N = 219 [R.sup.2] = 0.477 Note: In the first model, REGPART equals 1 if a metropolitan area had at least one regional partnership for economic development in 1997 and 0 if it did not. Logistic regression was used to explain variation in REGPART. In the second model, TOTPART equals the total number of regional partnerships for economic development in each metropolitan area in 1997. Multiple regression was used to explain variation in TOTPART. Table 2 Results of the Second Set of Analyses: The Structure and Processes of Regional Partnerships Dependent variables FORMAL LNCENTR LNMEM Independent variables Standardized betas GOVNORM -0.22 ** -0.19 ** -0.06 PRIVNORM 0.26 *** 0.16 * 0.29 *** CIVPC 1.00E-03 0.13 0.05 FORMAL NI -0.11 -0.08 CENTRAL -0.11 NI -0.31 *** MEMBER -0.26 *** -0.35 *** NI INFOTASK -0.05 -0.10 0.04 ECONTASK 0.37 *** -0.17 * 0.30 *** RESFLOW 0.28 *** 0.09 0.08 INFOFLOW 0.06 0.27 *** 0.07 AGE -0.04 -0.16 * 0.15 * EMPRATE -0.07 0.20 ** 0.21 ** INCOME -0.09 -0.11 0.13 REGLIFE -0.08 -0.08 -0.03 ECONDISP -0.07 -0.15 * -0.23 *** GOVRESPC 0.26 ** -0.20 * 0.13 POP -0.17 -0.16 -0.04 POPDENS 0.17 * 0.12 0.05 Model sig. <.000 <.000 <.000 F-statistic 3.159 3.011 4.802 [R.sup.2] 0.324 0.310 0.415 N 130 132 133 Dependent variables INFOTASK ECONTASK LNRES INFOFLOW Independent variables Standardized betas GOVNORM 0.12 -0.04 -0.10 0.27 *** PRIVNORM -0.08 3.00E-03 -0.01 -0.08 CIVPC 0.05 0.09 -0.14 -0.10 FORMAL -0.06 0.32 *** 0.03 0.06 CENTRAL -0.07 -0.01 -0.16 * 0.16 * MEMBER 0.07 0.18 ** 0.11 0.08 INFOTASK NI 0.47 *** 0.25 ** 0.07 ECONTASK 0.55 *** NI 0.26 ** 0.16 RESFLOW 0.18 ** 0.02 NI -0.04 INFOFLOW 0.11 0.10 0.02 NI AGE -0.17 ** -9.00E-03 0.16 * 0.17 * EMPRATE 0.03 -0.02 0.06 -0.09 INCOME 0.06 -0.11 0.04 -0.04 REGLIFE 0.03 -0.06 0.17 ** -0.05 ECONDISP -0.11 0.08 0.03 0.16 * GOVRESPC -0.06 -0.18 ** 0.02 0.21 * POP -0.20 * 0.29 *** 7.00E-03 -0.02 POPDENS 0.07 -0.16 * 0.08 0.08 Model sig. <.000 <.000 <.000 <.000 F-statistic 5.473 7.444 3.774 1.97 [R.sup.2] 0.452 0.53 0.366 0.227 N 131 130 129 132 NI = Not included in model *** = p < .01 level ** = p < .05 level * = p <. 10 level
(1.) Tiebout is skeptical of the plausibility of the last scenario, saying, "The case of the city that is too large and tries to get rid of residents is more difficult to imagine" (1956, 420).
(2.) The number depends on one's definition of "metropolitan government." Some scholars say that Metro Miami-Dade County is the only metropolitan government in the United States (Saffell and Basehart 1997). Others, however, consider Portland and Minneapolis-St. Paul to have metropolitan governments because their regional councils of government have broad authority and taxing powers (Dodge 1992).
(3.) According to Van de Ven (1976a, b), most organizational theorists agree that formalization, complexity, and centralization are the major dimensions of social structure.
(4.) Because a primary activity of regional partnerships for economic development is marketing their metropolitan areas to prospective businesses, many of them operate Web sites to promote the partnerships and the region.
(5.) Local Authority For Regional Strategies is an index ranging from 0 to 13. In developing this index, metropolitan areas were given one point for each of the following regional strategies authorized by state government: annexation, consolidation of cities, consolidation of cities and counties, and interlocal service agreements. In addition, metropolitan areas were given one point for each of the following actions not required by state government for annexation: a public hearing, referendum and majority approval in annexing city, referendum and majority approval (or written consent) in the area to be annexed, and approval of the county governing authority. They were given one point for each of the following actions not required by state government for the consolidation of cities: referendum and majority approval of only one city, and referendum and majority approval of all cities. Finally, metropolitan areas were given one point for each of the following actions not required by state government for city-county consolidation: referendum and majority approval of all affected cities, referendum and majority approval of county, and referendum and majority approval of unincorporated area in county.
(6) The use of SPSS for logistic regression--unlike multiple regression--does not enable the replacement of missing values with the mean.
(7) These results are very similar to the results of the model run on the reduced, time-ordered population, which includes only metropolitan areas without regional partnerships and those with regional partnerships known to have been formed after 1987. The primary difference is that the analysis of the time-ordered population results in fewer significant variables because of a smaller sample size.
(8.) While Private Sector Norms is significant in three of the seven models, two of the coefficients have signs in the direction opposite of what was predicted.
Astley, W. Graham. 1984. Toward an Appreciation of Collective Strategy. Academy of Management Review 9(3): 526-35.
Axelrod, Robert. 1986. An Evolutionary Approach to Norms. American Political Science Review 80(4): 1095-111.
--. 1997. The Complexity of Cooperation: Agent-based Models of Competition and Collaboration. Princeton, NJ: Princeton University Press.
Barnes, William R., and Larry C. Ledebur. 1991. Toward a New Political Economy of Metropolitan Regions. Environment and Planning 9: 127-41.
--. 1998. The New Regional Economies: The U.S. Common Market and the Global Economy. Thousand Oaks, CA: Sage Publications.
Bennett, Frederick C., and Josef Nathanson. 1997. Metropolitan Alliances. Commentary (Spring): 34-38.
Browning, Larry D., Janice M. Beyer, and Judy C. Shetler. 1995. Building Cooperation in a Competitive Industry: SEMATECH and the Semiconductor Industry. Academy of Management Journal 38(1): 113-51.
Cable, Gregory, Richard C. Feiock, and Jaehoon Kim. 1993. The Consequences of Instituionalized Access for Economic Development Policy Making in U.S. Cities. Economic Development Quarterly 7(1): 91-97.
Clark, Charles S. 1995. Revitalizing the Cities: Is Regional Planning the Answer? CQ Researcher 5(38): 897-905.
Coe, Benjamin P. 1992. Tug Hill, New York Progress Through Cooperation in a Rural Region. National Civic Review 81 (4): 449-65.
Dielman, Terry E. 1996. Applied Regression Analysis for Business and Economics. 2nd ed. Belmont, CA: Wadsworth Publishing Company.
Dodge, William. 1989. The Emergence of Intercommunity Partnerships in the 1980s. National Civic Review 78(1): 5-14.
--. 1990. Regional Problem Solving in the 1990s: Experimentation with Local Governance for the 21st Century. National Civic Review 79(4): 354-66.
--. 1992. Strategic Intercommunity Governance Networks ("SIGNETs" of Economic Competitiveness in the 1990s). National Civic Review 81 (4): 403-17.
--. 1996. Regional Excellence: Governing Together to Compete Globally and Flourish Locally. Washington, DC: National League of Cities.
Doney, Patricia M., and Joseph P. Cannon. 1997. An Examination of Trust in Buyer-Seller Relationships. Journal of Marketing 61(2): 35-51.
Dye, Thomas R. 1964. Urban Political Integration: Conditions Associated with Annexation in American Cities. Midwest Journal of Political Science (November): 430-46.
--. 1990. American Federalism: Competition Among Governments. Lexington, MA: Lexington Books.
Ehrenhalt, Alan. 1995. Cooperate or Die. Governing (September): 28-32.
Eisinger, Peter K. 1988. Rise of the Entrepreneurial State. Madison, WI: University of Wisconsin Press.
Florestano, Patricia S., and Laura Wilson-Gentry. 1994. The Acceptability of Regionalism in Solving State and Local Problems. SPECTRUM: The Journal of State Government 67(3): 26-33.
Gray, Virginia, and David Lowery. 1990. The Corporatist Foundations of State Industrial Policy. Social Science Quarterly 71(1): 3-24.
Grell, Jan, and Gary Gappert. 1993. The New Civic Infrastructure: Intersectoral Collaboration and the Decision-Making Process. National Civic Review 82(2): 140-48.
Heath, Mark D., and Edward G. Henegar. 1994. Success Story! The Carolinas Partnership. Economic Development Review 12(1): 80-84.
Heide, Jan B., and George John. 1992. Do Norms Matter in Marketing Relationships? Journal of Marketing 56(2): 3244.
Hershberg, Theodore. 1996. Regional Cooperation: Strategies and Incentives for Global Competitiveness and Urban Reform. National Civic Review 85(2): 25-30.
Hershberg, Theodore, Pam Magidson, and Mary Lou Wernecke. 1992. Promoting Regional Cooperation in Southeastern Pennsylvania. National Civic Review 81 (4): 418-34.
Huggins, Michael. 1992. Momentum 21: Regional Cooperation in the Chippewa Valley. National Civic Review 81(4): 43548.
Lindblom, Charles. 1977. Politics and Markets. New York: Basic Books.
Lyons, Willam E., David Lowery, and Ruth Hoogland DeHoog. 1992. The Politics of Dissatisfaction: Citizens, Services and Urban Institutions. Armonk, NY: M.E. Sharpe.
National Council for Urban Economic Development. 1995. Trends in Economic Development Organizations. Washington, DC: National Council for Urban Economic Development.
Nunn, Samuel, and Mark S. Rosentraub. 1996. Metropolitan Fiscal Equalization: Distilling Lessons from Four U.S. Programs. State and Local Government Review 28(2): 90-102.
Olberding, Julie Cencula. 1997. Regional Alliances for Economic Development: Why Cooperative Structures Are Adopted in a Competitive Environment. Paper presented at the Southeastern Conference for Public Administration, September 25-28, Knoxville, Tennessee.
--. 2001. Do Regional Partnerships for Economic Development Really Increase Employment and Income? Paper presented at the annual meeting of the American Society for Public Administration, March 10-13, Newark, New Jersey.
Olson, Mancur, Jr. 1965. The Logic of Collective Action. Cambridge, MA: Harvard University Press.
Ostrom, Elinor. 1998. A Behavioral Approach to the Rational Choice Theory of Collective Action. American Political Science Review 92(1): 1-22.
Peirce, Neal R. 1993a. Citistates: How Urban America Can Prospet' in a Competitive World. Washington, DC: Seven Locks Press.
--. 1993b. At HUD, the Byword is Partnership. National Journal, April 3, 828.
--. 1993c. The Dawn of the Age of the "Citistates." National Journal, July 31, 1934.
--. 1993d. There's a New Way to Define Regions. National Journal, August 14, 2045.
Peterson, Paul E. 1981. City Limits. Chicago: University of Chicago Press.
--. 1995. The Price of Federalism. Washington, DC: Brookings Institution.
Putnam, Robert D. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press.
Raasch, Robert, and James Brooks. 1995. Regional Alliances Improve Economic Advantage. Nation's Cities Weekly, September 18.
Ring, Peter Smith, and Andrew H. Van de Ven. 1994. Developmental Processes of Cooperative Interorganizational Relationships. Academy of Management Review 19(1): 90-1 18.
Roeder, Phillip W. 1994. Public Opinion and Policy Leadership in the American States. Tuscaloosa, AL: University of Alabama Press.
Rusk, David. 1995. Cities Without Suburbs. Washington, DC: Woodrow Wilson Center Press.
Saffell, David C., and Harry Basehart. 1997. Governing States and Cities. New York: McGraw-Hill.
Savitch, H.V., David Collins, Daniel Sanders, and John P. Markham. 1993. Ties That Bind: Central Cities, Suburbs and the New Metropolitan Region. Economic Development Quarterly 7(1): 341-57.
Sbragia, A.M. 2000. Entrepreneurial Cities, U.S. Federalism, and Economic Development. In American Intergovernmental Relations: Foundations, Perspectives and Issues, 3rd ed., edited by Lawrence J. O'Toole, Jr., 217-28. Washington, DC: CQ Press.
Tiebout, Charles M. 1956. A Pure Theory of Local Expenditures. Journal of Political Economy (October): 416-24.
Van de Ven, Andrew H. 1976a. On the Nature, Formation, and Maintenance of Relations Among Organizations. Academy of Management Review 1 (4): 24-36.
--. 1976b. A Framework for Organization Assessment. Academy of Management Review 1 (1): 64-78.
Walker, David B. 1987. Snow White and the 17 Dwarfs: From Metropolitan Cooperation to Governance. National Civic Review 76(1): 14-27.
Wallis, Allan D. 1994a. Evolving Structures and Challenges of Metropolitan Regions. National Civic Review 83(1): 40-53.
--. 1994b. Inventing Regionalism: A Two-phase Approach. National Civic Review 83(4): 447-68.
Worldwide Chamber of Commerce Directory. 1998. Worldwide Chamber of Commerce Directory. Loveland, TX: Worldwide Chamber of Commerce Directory.
Julie Cencula Olberding recently earned a doctoral degree from the Martin School of Public Policy and Administration at the University of Kentucky. Her research of regional partnerships for economic development also examines their effectiveness--specifically, whether metropolitan areas with regional partnerships have experienced greater increases in employment and income than metropolitan areas without them. Her research and teaching interests involve state and local government, intergovernmental relations, and community and economic development. Email: firstname.lastname@example.org.
|Printer friendly Cite/link Email Feedback|
|Author:||Olberding, Julie Cencula|
|Publication:||Public Administration Review|
|Date:||Jul 1, 2002|
|Previous Article:||County service delivery: does government structure matter?|
|Next Article:||Gubernatorial use of the item veto for narrative deletion.|