The Varieties of Consolidation Experience: A Synthesis and Extension of Local Government Consolidation Models

This article revises and extends Leland and Thurmaier’s (2004a, 2004b) City—County Consolidation (C3) model by synthesizing it with Johnson’s (2004) Theory of Local Constitutional Change (LCC) and Hughes and Lee’s (2002) Evolutionary Consolidation Model (ECM). The result, we find, is a more general model of local government consolidation. Our model is applicable to a wider variety of consolidation types and incorporates a full consideration of varied charter development processes. Ultimately, this allows for acknowledgment of the possibility that consolidation attempts may be halted prior to a referendum campaign and that those attempts may reflect either conflicts of interest or consensual efforts at problem-solving. We focus specifically on Indiana after enactment of the 2006 Government Modernization Act. After enactment of the act, Indiana experienced seven consolidation efforts from 2008 to 2012. Examination of these efforts provides a robust comparative case study of consolidation efforts occurring during a narrow timeframe and under a common institutional context. The study not only illustrates the suitability of our revised and extended model, but it also confirms a number of Leland and Thurmaier’s (2005) findings from their reassessment of the C3 model.

consolidation efforts were soundly rejected by voters. Two of the town-township consolidation efforts were successful, with large majorities approving these referenda. The other town-township and city-township consolidation efforts were terminated before reaching a referendum.
Accounting for these varied consolidation outcomes requires a sound theoretical model. Leland and Thurmaier (2004a) have presented their City-County Consolidation (C3) Model, later updated as their "respecified" C3 model (Leland & Thurmaier, 2004b, p. 315), as a causal model for systematic and comparative analysis of consolidation attempts. Although, as its name implies, the C3 model is most directly applicable to city-county consolidations, it has been applied to other local government reform proposals, including town-township consolidations (e.g.,  and city-city consolidation (e.g., Honadle, 2004).
The current formulation of the C3 model, although useful, is not well suited for the analysis of consolidation efforts that are terminated before reaching a referendum or those that are consensual rather than conflictual. The current model also fails to provide robust insight into the charter development process. The purpose of this study, then, is to revise and extend the current C3 model to encompass a wider variety of consolidation types and outcomes. We also include a fuller account of the entire consolidation process by incorporating Johnson's (2004) Theory of Local Constitutional Change (LCC) and Hughes and Lee's (2002) Evolutionary Consolidation Model (ECM). In doing so, this analysis continues a long tradition of theoretical synthesis that has informed theory building in this area. For example, in developing their theories, both, Leland and Thurmaier (2004a) and Johnson (2004) drew upon earlier work by Rosenbaum and Kammerer (1974) by augmenting the Rosenbaum and Kammerer (R&K) model with components drawn from other models, theories, and frameworks.
In this study, we focus on Indiana local government consolidation attempts that occurred from 2008 to 2012. A research design using cases drawn from a single state has both strengths and weaknesses. On one hand, a single state design may raise concerns about generalizability. Are the findings applicable to cases in other states? Are the results explained primarily by conditions particular to the state from which the cases were drawn? On the other hand, a design focusing on consolidation attempts occurring within one state and within a short time period also has strengths. Such a design allows us to hold constant key variables related to the institutional context (i.e., variables that often vary between states or across time) that could present confounding factors in a multi-state design. Thus, our single state design allows us to focus our examination on the effect of variation in other characteristics, such as the presence of power deflation, the nature of the charter development process, and the type of local government.
Ultimately, our design fulfills conditions necessary for a rigorous comparative case study described by a number of scholars (e.g., Buttolph, Reynolds, & Mycoff, 2008;King, Keohone, & Verba, 1994;Leland & Thurmaier, 2004a;Lijphart,1975). Furthermore, the cases in this study exhibit a wider range of outcomes than most studies (e.g., not only successful and unsuccessful referenda but also attempts that failed to reach a referendum).
In the next section, we situate consolidation within the larger boundary change literature. Next, we compare and contrast existing consolidation frameworks and identify their limitations. In the sections that we follow, we provide brief overviews of Indiana's local government structure and a history of local government consolidation. After which, we provide a discussion of the suggested revisions and extensions and a synthesis of existing consolidation models, which we illustrate using local government consolidation efforts in Indiana. Finally, we analyze Indiana consolidation efforts to present a more comprehensive model of consolidation.
Like Leland and Thurmaier (2004a), Johnson (2004) augmented the R&K model to develop the Theory of Local Constitutional Change (LCC). The LCC and C3 models cover essentially the same range of activities. However, the LCC model considers the creation of the consolidation charter (i.e., local constitution) in greater detail than the C3 model. The C3 model primarily emphasizes the impact of specific charter provisions (e.g., the status of the sheriff or structure of the council) on group mobilization in the referendum campaign rather than the process of developing the charter. In the LCC model, Johnson (2004) isolated constitution setting (i.e., charter development) as a separate stage of the process following the initial agenda setting stage and prior to the final referendum campaign stage. Johnson (2004) suggested that this stage (i.e., not the referendum campaign) was where competition between the demanders of change and the defenders of the status quo began. Thus, a complete understanding of the referendum campaign and outcome requires a full accounting of the charter development process.

Consolidation Stopping Points
The LCC and C3 models have differences and similarities with regard to potential termination points in the consolidation process. Both models include the possibility that a consolidation attempt could end early during the agenda setting phase. This could be a result of an effective and appropriate local government response to the crisis climate situation (Johnson 2004;Leland & Thurmaier, 2004a), sufficient opposition from civic elites (Leland & Thurmaier, 2004a), or a "decelerator" event that diminishes rather than intensifies demand for change (Johnson 2004). Similarly, both models allow for the possibility consolidation failure when the consolidation referendum is defeated by voters (Leland &Thurmaier, 2004a;Johnson 2004).
The two models differ with regard to the possibility of termination of the consolidation effort after a consolidation charter committee has been formed, but prior to the consolidation referendum. Leland and Thurmaier's (2004a) C3 model implicitly assumes that once the consolidation charter commission is formed then the process will necessarily continue through to the referendum. Johnson's (2004) LCC model, however, proposes at least two points (one implicit and one explicit) where the consolidation attempt could be terminated prior to a referendum, but after agenda setting. First, because Johnson (2004) makes constitution setting a distinct phase of the process, there is a clear boundary between constitution setting and the referendum campaign-that is, the development of a consolidation charter or constitution. This definition of boundary implies that failure of the consolidation charter commission to approve a charter would result in termination of the process. In a recent study, Schaal, Taylor, and Faulk (2017) found that the consolidation processes in some states allowed the commission and/or the constituent local governments to terminate the consolidation attempt without producing a consolidation charter. Second, Johnson (2004) explicitly considered whether the process could be halted after the charter was approved by the commission, but prior to being placed on a referendum ballot (as had been the case in the two times that the local legislative delegation prevented consolidation proposals for Tallahassee-Leon County, Florida from reaching a ballot).
Indiana consolidation experiences suggest that a charter commission may refuse to approve a charter for reasons not contemplated by the extant models. For example, a commission may develop a charter, but then decide not to approve it because conditions have changed making consolidation less salient to the public interest. The commission may also determine, through the fact finding of the charter development process, that consolidation is not a viable solution to the problems that originally placed consolidation on the agenda. Furthermore, the existence of potential stopping points prior to the referendum can impede efforts to place a consolidation question on the ballot , heightening the importance of accounting for all the opportunities to halt the process.

Revolutionary vs. Evolutionary Consolidation
Both Leland and Thurmaier (2004a) as well as Johnson (2004) declared that local government consolidation is a revolutionary change. This change is initiated and carried out in response to crisis or other situations for which consolidation is proposed as a solution (Leland & Thurmaier, 2005). Hughes and Lee (2002), however, argued that local government consolidation may be the final step in an evolutionary process consisting of a series of smaller cooperative efforts to solve interjurisdictional problems among local governments. While Leland and Thurmaier (2004a) acknowledged the potential influence of existing interlocal agreements to prepare local officials and citizens for discussion about political consolidation, they treated the presence of such agreements more as an environmental variable that affects agenda setting in a revolutionary consolidation process as opposed to an ongoing and evolutionary process of consolidation.
In Hughes and Lee's (2002) ECM, local governments within a metropolitan region solve common problems by harmonizing community expectations with the resources available to solve those problems. Beginning in the first stage, when problems are relatively minor, local governments within a region engage in occasional informal talks about common issues. As problems intensify and resource constraints become more binding, in stage two, local governments engage in more frequent and formal discussions. This results in stage three, interlocal agreements to share facilities, services, and/or authority in particular policy areas. In stage four, the stage before full consolidation, local governments enter into an agreement to share power or decision-making in at least one major policy area. In stage five, the local governments pursue consolidation, "the most permanent cooperative arrangement" (p. 147). Hughes and Lee (2002) suggest that full consolidation is neither inevitable nor necessarily desirable. Their case study of the evolution of intergovernmental cooperation in the Albuquerque, New Mexico area demonstrates that consolidation proposals can be controversial and defeated even after a long history of interlocal cooperation.

Competition vs. Learning in Constitution Formation
Describing the charter development process, Johnson (2004) and Leland as well as Thurmaier (2004a) indicated that this process was as a contest between opposing interests. For Leland and Thurmaier (2004a), they viewed this competition mainly in terms of how the rules of consolidation mobilized consolidation proponents and opponents and influenced the pro-and anti-consolidation referendum campaigns. Changes in tax burdens, structure, and size of the unified council; minority representation on the unified council; and the role of the county sheriff are among the most contentious features of the consolidation charter. These features informed the hypotheses tested in their model to determine factors influencing the success or failure of consolidation referenda. Johnson (2004) suggested that this competition during the constitution setting process provides a preview to the referendum campaign. Pro-and anti-consolidation forces compete to influence the design of the consolidation charter and the voting rules used to adopt it. The distribution of costs and benefits arising from various constitutional provisions influences this competition and ultimately the outcome of consolidation referenda. Groups that stand to benefit from consolidation will attempt to influence the process so that favorable provisions are included. Others will try to defend their interests by arguing in support of the status quo or supporting rules and provisions that make voters less likely to support the charter. Hughes and Lee's (2002) evolutionary model, however, suggests the charter development phase may exhibit much lower levels of conflict. In these situations, the charter development debate is more accurately characterized as a new stage in the process of investigation and learning about community needs in order to determine if consolidation is an appropriate response.

Elements of a Synthesized and Extended Theoretical Model
In summary, we suggest specific elements can be incorporated to synthesize and extend existing consolidation frameworks. The revised model should 1) utilize a three-stage model of the process of agenda-setting, charter development, and referendum campaign stages, facilitating an appropriate focus on the activities and outcomes occurring (specifically during the charter development stage), 2) highlight the possibility that the charter development stage may not necessarily represent a mere contest of interests for favorable charter terms, but it could also serve as a consensual fact finding study of local needs and how consolidation might serve those needs, 3) allow for the possibility that consolidation may be either a revolutionary response to crisis or other major events or it may be an evolutionary process of increasing collaboration over time in response to changes in the regional environment, and 4) incorporate a broader conception of consolidation outcomes, including halting the consolidation effort prior to a referendum. Not only may a consolidation effort be halted in any stage of the process, but a halt may represent a failure or defeat of the pro-consolidation forces. A halt could also represent a consensual and informed decision that consolidation is not in the public interest.
Prior to illustrating the suitability of these elements using Indiana local government consolidation efforts as examples, it is necessary to understand the institutional context in which these efforts occurred. As such, below we provide brief overviews of local government structure in Indiana and the state's history of local government consolidation.

Local Government Structure and Consolidation in Indiana
Local government in Indiana consists of counties, municipalities (i.e., cities and towns), townships, school districts, and special districts. The number of municipalities has remained stable over the past several decades, increasing from 564 in 1982 to 569 in 2012. Towns are more numerous than cities, making up approximately 80% of municipalities in the state.
During 2012, there were 1,006 Indiana townships, which are general purpose governments. Each township was led by an elected trustee and township board, offering a limited range of services. Their primary services were overseeing volunteer fire departments and emergency medical services, providing poor relief, maintaining cemeteries, and operating parks. Townships funded services through their property taxing authority. All cities and towns were located within one or more townships.
Indianapolis and Marion County were consolidated in 1970 through a 1969 Act of the General Assembly. Sometimes called "Unigov," Indianapolis-Marion County is Indiana's sole instance of city-county consolidation. Parks (1993, 1995) provided information on the background and structure of the consolidated government, while both Segedy and Lyons (2001) and Rosentraub (2000) evaluated its various aspects.
Amid growing concern with the effectiveness of Indiana local governments and inefficiencies resulting from the number and layering of local governments in the state, the Indiana General Assembly passed the Government Modernization Act (GMA) in 2006. 1 This act enables the reorganization of political subdivisions, defining a uniform process for local government consolidation. The process allows for two different approaches to forming the reorganization committee responsible for developing a plan of reorganization (i.e., the consolidation charter). One approach is for the two or more governing bodies to enact identical resolutions to create a reorganization committee. The other approach requires reorganization proponents to file petitions supporting the creation of a reorganization committee. A petition with signatures of at least five percent of the voters in the subdivision must also be filed with each political subdivision to be included in the reorganization proposal.
Once the reorganization committee completes the plan of reorganization, the participating political subdivisions' legislative bodies can either adopt the plan, adopt it with revisions, or reject it. If they adopt identical plans, the adoptions are certified, and the plan and certifications are filed with the county recorder. Following a review by the Department of Local Government Finance, the county election board places the referendum question on the ballot for the next general or municipal election covering all precincts within the reorganizing political subdivisions. If the participating political subdivisions fail to adopt a final plan of reorganization, then citizens can petition for the approval of, and voting on, a final plan of reorganization.
Before the GMA became law, local government consolidation in Indiana required special legislation. In 1974, a consolidation referendum was held in Evansville and Vanderburgh County, but it was rejected by voters. In 1990, a consolidation plan was drafted by a citizen's committee, but it was tabled by the Evansville mayor and Vanderburgh County commissioners without seeking special legislation. In 2006, another citizen's committee drafted a consolidation plan, but legislation providing for a referendum did not pass the legislature (League of Women Voters of Southwestern Indiana, Undated). Key legislators felt that with the passage of the GMA, there was no need for special legislation for one community (Whitson, 2006). Consequently, after 1974 there were no additional consolidation referenda prior to enactment of the GMA. Schaal, Taylor, and Faulk (2017) found that states with general legislation tend to have more consolidation referenda than states requiring special legislation. Indiana's consolidation history provides additional support for this relationship. Since the GMA was passed, more consolidation attempts have occurred for a variety of government types in Indiana, providing further evidence that state authorization of consolidation under general law facilitates consolidation attempts. These seven local government consolidation efforts occurred within a four-year period and a common institutional context. They were also governed by the GMA. These common conditions allowed us to compare their varying crisis climates, charter development processes, and consolidation outcomes to illustrate the utility of an extended and synthesized consolidation model. In the next section, we describe the variety of crisis climates predating the Indiana consolidation efforts.

Crisis Climates in Indiana Consolidation Attempts
Drawing on the Rosenbaum and Kammerer model, the original C3 model proposed that consolidation attempts arise from a crisis climate creating a demand for change (Leland & Thurmaier, 2004a). Leland and Thurmaier (2004a) further suggested that when local government response to crisis is ineffective, power deflation occurs. As a result, local governments consider consolidation. In a subsequent assessment of the C3 model, Leland and Thurmaier (2005) determined that civic problems are frequently present in communities that ultimately consolidate, but that they do not always rise to crisis level. In the Local Constitutional Change model, Johnson (2004) suggested that consolidation may also be an attempt to create new institutional arrangements in response to current needs rather than a result of power deflation. Leland and Thurmaier (2004a) explained that a crisis climate is characterized by the onset of civic problems, which can develop slowly or suddenly. This creates a demand for response by local government. Three of the seven Indiana cases exhibited climates of fiscal crisis when changes (or potential changes) to local revenue streams threatened the quality and/or quantity of local government service provision. Another three cases exhibited annexation and development related crisis climates when annexation battles with neighboring communities threatened their future potential for growth and expansion. The seventh case had no discernable crisis (see Table 1).
Three of the Indiana consolidation attempts were characterized by power deflation in that the local governments' unsuccessful initial responses to their crisis climates contributed to a "growing lack of confidence" (Rosenbaum & Kammerer, 1974, p. 25) in the local governments, which in turn led to calls for consolidation. Three other cases appear to fit the Johnson (2004) model, where consolidation is considered a proactive approach to meeting community needs in the face of changing conditions, rather than a response following failure.

Fiscal Crises
In 2008, the Indiana General Assembly enacted property tax caps. These caps were to be phased in beginning in 2009 and fully implemented in 2010. They were expected to reduce property tax revenue for nearly every local government unit in the state. However, the revenue reductions were uneven, with impacts ranging from minimal to severe (Faulk, 2013;Taylor, 2015). From December 2007 until June 2009, the nation was in an economic recession (National Bureau of Economic Research, 2010). The consolidation activity described in this study occurred either during the recession or in the early part of the recovery. Thus, the recession is a factor in the fiscal climate affecting local governments during the period under study. Although Indiana is generally recognized as a state that was particularly hard hit by the recession, there was substantial local variation in the impact of the recession across the state. In June of 2009, at the trough of the national recession, Indiana's statewide unemployment rate was 10.8%. At the same time, local unemployment rates in the communities under study ranged from a low of 7.7% in Hendricks County (Avon, Brownsburg) to 12.0% in Delaware County (Muncie, Yorktown). In two of the consolidation attempts (Muncie-Delaware and Evansville-Vanderburgh), fiscal crises arising from imposition of the tax caps were significant factors in placing consolidation on local agendas. In a third case (Avon-Washington), a different fiscal crisis that predated both the tax caps and the recession played a role.
In Muncie and Delaware County the impacts of the tax caps were quite large. Muncie's property tax revenues were reduced by more than 30%; and, Delaware County's revenues fell more than 17%. Delaware County also experienced a large recession related increase in the local unemployment rate from 5.2% in December 2007 to 12.0% in June 2009. Responding to the tax caps, the mayor of Muncie laid off 32 firefighters and five police officers as a cost cutting measure (Smith 2009). The ensuing controversy regarding the layoffs created a crisis climate where the potential for efficiency via consolidation was attractive to the local taxpayer group that organized the petition drive to require Muncie and Delaware County to appoint a reorganization committee (Werner, 2010). In Evansville and Vanderburgh County the impacts were more modest, with city and county property tax revenues reduced by four to five percent. The impact of the recession was also more modest, with the local unemployment rate increasing from 4.5% in December 2007 to 8.4% in June 2009. Uncertainty regarding the impact of the tax caps, however, led the county to not renew a homestead property tax exemption (Langhorne, 2009b). In the face of the ensuing controversy, the homestead exemption was later reinstated, but the combined impact of the exemption and the tax caps was a $15 million budget shortfall for Evansville and Vanderburgh County. Concern about the budget shortfall created a crisis climate prompting the local newspaper's editorial board to call for consideration of local government consolidation ("Evansville, Vanderburgh County facing," 2009). The president of the local League of Women Voters referred to a need for greater government efficiency when explaining the reason for their petition drive (Langhorne, 2009a).
Although the Town of Avon and Washington Township each experienced tax cap revenue reductions of about eight percent, their consolidation effort was spurred by a fiscal problem predating the tax caps. The township had a longstanding practice of relying on emergency loans and emergency levies to fund the township fire department serving both the town and township. In June 2010, a group of local citizens filed a petition forcing the Indiana Department of Local Government Finance (DLGF) to review the legality of the most recent emergency loan for the fire department. Although DLGF upheld the loan, the township trustee was concerned that if the policy of funding the fire department with emergency loans was ever invalidated, then layoffs would be required (Beasor, 2010). The potential invalidation of the emergency loans and levies created the crisis climate that led local officials to consider consolidation. Minutes from a meeting of the Greater Avon Study Committee (2012a) and the Plan of Reorganization (Greater Avon Study Committee, 2012b) confirm the concern by town and township officials about the continued viability of fire department emergency loans.
In these three cases, one can credibly argue that the consolidation effort was a result of power deflation following a fiscal crises. In Muncie-Delaware County and Evansville-Vanderburgh County, local officials were viewed as having responded ineffectively to the imposition of property tax caps and citizen groups mobilized to force them to consider consolidation. In Avon-Washington, local officials were viewed as ineffective in that they were continually relying on emergency loans and levies to ensure that the township fire department was adequately funded. Citizen attempts to invalidate the emergency loans led the town and township to explore consolidation as a means to fix the fire department funding problem.

Annexation and Development Related Crises
Annexation is the most common form of boundary change. Annexation efforts by Indiana municipalities tend to be controversial as rural residents are often resistant to receiving the higher level of service, with corresponding higher taxes, afforded by a city or town government. Indiana law allows landowners affected by an involuntary annexation to protest it in court (Indiana Chamber of Commerce, 2013). In some communities, development concerns and annexation battles with neighboring communities contributed to the crisis climates. The Town of Yorktown and the City of Muncie had been engaged in an annexation battle dating back to 2005 (Yencer, 2005;Slabaugh, 2005). This territorial conflict intensified when the effort to consolidate Muncie and Delaware County got underway. Yorktown and Mount Pleasant officials initiated their own consolidation effort, in part, to minimize the impact on Yorktown and Mount Pleasant in the event that the Muncie-Delaware consolidation effort was successful (Yorktown-Mount Pleasant Reorganization Committee, 2010).
Property development concerns combined with an annexation battle created the crisis climate in Zionsville. Rapid development just outside town limits governed by Boone County's more permissive development standards concerned Zionsville officials (Woodson, 2008). Town and township officials were also concerned that aggressive annexation by nearby Whitestown would curtail Zionsville's future expansion (Annis, 2007;Olson, 2007).
Similarly, a battle for control over fast growing unincorporated areas in White River Township drove the Greenwood-White River consolidation attempt. Merger proponents argued that extending city services to the densely populated, unincorporated areas of the township would provide increases in both population and tax base that would allow the city to compete with other cities for corporate offices and high end retail. The City of Greenwood was also embroiled in an annexation dispute with the Town of Bargersville, which had attempted to annex property along a commercial corridor in the fast growing suburbs (McLaughlin, 2009).
The Greenwood-White River effort was characterized by a high level of conflict and was ultimately unsuccessful, whereas the Zionsville-Eagle-Union and Yorktown-Mount Pleasant efforts were more consensual and ultimately successful. Despite this difference, it is not apparent that any of these three consolidation attempts were characterized by power deflation. In all three cases, municipal and township officials initiated the consolidation effort as a proactive response to annexation and development related issues. Applying the boundary change framework developed by Feiock and Carr (2001), boundary change entrepreneurs were willing to engage in the collective action processes that were available to potentially bring about consolidation with a more desirable entity rather than wait for annexation by a less desirable entity.

A Non-Crisis Climate
Unlike the other consolidation attempts, the Brownsburg-Brown-Lincoln consolidation attempt did not appear to be motivated by a specific crisis in the community. The reorganization plan introduction indicated that the effort was primarily a means to explore "efficiencies in governance" for town and township citizens (Brownsburg Reorganization Committee, 2011). One member of the reorganization committee stated that the goal of the committee was "to determine if [consolidation] makes sense" (Essett, 2011).
The plan also mentioned a secondary purpose of protecting against annexation to allow citizens to "control their own destiny" (Brownsburg Reorganization Committee, 2011); and, a news report indicated that concern about development just outside the town limits, governed by laxer county zoning regulations, was also an issue of importance to some consolidation proponents (Doan, 2012A). There is no indication in committee records or contemporaneous news reports, however, of a specific threat or crisis.
Considered together, these cases confirm Leland and Thurmaier's (2005) conclusion that neither the presence of a crisis climate nor power deflation are necessary preconditions for a successful consolidation referendum. Of the six cases with a crisis climate, only three also exhibited power deflation. Two of the three cases that exhibited power deflation resulted in failed referenda. The third was halted prior to a referendum. Of the other three cases with a crisis climate, two resulted in a successful referendum and one terminated without holding a referendum. The one case in which there was no identifiable crisis climate terminated with no referendum (see Table 1).
Whatever crisis or situation places consolidation on the local agenda, the charter formulation process may either represent a contest of interests for favorable charter terms or serve as a consensual study of local needs and how consolidation might serve them. These dissimilar charter development processes are explored in the next section.

Charter Development as Contest of Interests vs. Consensual Fact Finding
Johnson (2004) considers charter development as a separate stage of the consolidation process. Leland and Thurmaier (2004a) consider it as the first part of the referendum stage. Despite this difference, both theories consider the impact of charter provisions on the eventual consolidation referendum. Johnson (2004) views the charter development stage as a contest of interests between the policy entrepreneurs who advocate for new institutional arrangements under consolidation and the defenders of the status quo. Leland and Thurmaier (2004a) note that particular charter provisions will have both supporters and opponents and view the specific charter provisions primarily as independent variables affecting the likelihood that the consolidation referendum is successful.
Two Indiana consolidation attempts, Evansville-Vanderburgh and Greenwood-White River, exhibited the type of contentious interest-advancing behavior by supporters and opponents described by Johnson. In Muncie-Delaware, consolidation opponents did not engage in the charter development process but became active once the referendum campaign began. In the town-township consolidation efforts (Avon-Washington, Brownsburg-Brown-Lincoln, Yorktown-Mount Pleasant, and Zionsville-Eagle-Union) the process was consensual, rather than contentious, representing more of a fact finding mission than a contest for interests.

Contentious Contests of Interests
Analysis of the Indiana cases confirms that like other cases examined in the literature (e.g., Leland & Thurmaier, 2004a), the approach to services (particularly public safety), the distribution of taxes, and land use regulations are potentially contentious issues that influence the continuation of the consolidation process and the success or failure of the referendum, if reached. The Evansville-Vanderburgh reorganization plan included provisions intended to reduce opposition to consolidation by minimizing the fiscal and regulatory impacts on the nonurban areas. It included multiple tax and service districts to enable nonurban residents to avoid receiving, and paying for, the more intensive services provided to urban residents. It also preserved, at least initially, the less restrictive county land use ordinances that were in effect in the nonurban areas of the county (City of Evansville-Vanderburgh County Reorganization Committee, 2011).
Despite these provisions, the fiscal impact of consolidation in Evansville-Vanderburgh became a matter of controversy between consolidation advocates and opponents. One important concern was the potential for shifting the costs of sheriff patrols from urban to nonurban taxpayers, which other analysis has shown to strongly influence the success of consolidation referenda (see for instance, Leland & Thurmaier, 2004a). The reorganization plan ultimately specified that the police patrol function would remain unconsolidated. The city police department would continue to serve the urban areas, and the county sheriff department would continue to patrol the nonurban areas (City of Evansville-Vanderburgh County Reorganization Committee, 2011). In the pre-consolidation situation, city property owners, as county residents, helped fund the rural sheriff's patrols, but would cease to do so in the consolidated government. After consolidation, all sheriff patrol costs would be borne by residents of the nonurban area, leading to a substantial post-consolidation tax increase for nonurban taxpayers (Crowe Horwath, 2010). This tax shift generated conflict during charter development and a subsequent referendum campaign (Gootee, 2011; "Evansville residents help," 2011).
The decision not to consolidate the law enforcement agencies was itself a matter of substantial conflict during the charter development phase in the Evansville-Vanderburgh consolidation effort. Early in the charter development process, the city police chief and county sheriff each submitted proposals for consideration by the public safety subcommittee. Under the chief's proposal, the two agencies would remain separate, but the police department would be responsible for countywide law enforcement operations, leaving the sheriff's department responsible only for operation of the county jail, court security, and process serving.
The sheriff proposed merging the two agencies into one, under the supervision of the sheriff (Langhorne, 2010a). The public safety subcommittee approved the sheriff's proposal, but the conflict persisted (Langhorne, 2010b). The police chief and local chapter of the Fraternal Order of Police (FOP) continued to argue against the sheriff's proposal, claiming it would lead to increased costs, reduced efficiency, and conflict between the mayor and sheriff that could negatively impact law enforcement (Langhorne, 2010b). The sheriff continued to argue that accountability for public safety properly belonged to an elected sheriff, rather than an appointed police chief (Langhorne, 2010c).
Although the sheriff's proposal was included in the plan of reorganization submitted by the committee to the city and county (City of Evansville-Vanderburgh County Reorganization Committee, 2011), the city council and county commission used their power granted under the GMA to remove the law enforcement merger from the plan out of concern that it would lead to strong opposition from city residents (Gootee, 2011). Despite this change to the plan, the sheriff supported consolidation during the referendum campaign, while the Fraternal Order of Police continued to oppose it (Langhorne, 2012).
In the Greenwood-White River consolidation effort the committee took a different approach to fiscal impacts. Although the Greenwood-White River plan included urban and rural tax and service districts, nearly all urban services would be extended to the rural service district. The Greenwood Police Department would be expanded to serve the rural district. The Greenwood Board of Public Works and Safety would take responsibility for street and sidewalk maintenance in the rural district. The White River Township Fire Department would continue providing fire protection within the rural district, but the plan proposed a future merger of the city and township fire departments. Solid waste and yard waste collection were the only urban services not proposed for extension to the rural district (White River Township and City of Greenwood Reorganization Committee, 2009a).
This plan also did little to minimize the regulatory impact on residents of the unincorporated portion of the township. It proposed an update of Greenwood's comprehensive plan to govern land use in the rural district after consolidation, but incorporation of "right to farm" policies protecting agricultural land was the only concession made to rural landowners in the plan of reorganization (White River Township and City of Greenwood Reorganization Committee, 2009a).
This approach to services, taxes, and land use regulation created additional potential for conflict in a situation that was already contentious because of the ongoing annexation battle with the Town of Bargersville. Many township residents who would be affected by the consolidation had expressed a preference to be left alone, but if they had to choose would prefer to be annexed by Bargersville than consolidated with Greenwood (White River Township and City of Greenwood Reorganization Committee, 2009c).
Because the reorganization plan extended so many urban services to the rural district, the proposed Greenwood-White River reorganization took on many aspects of an annexation, rather than a consolidation, further aggravating the existing conflict. The fiscal analysis included in the reorganization plan compared property tax rates in the City of Greenwood before and after the proposed consolidation, which demonstrated that the larger, postconsolidation city would have a lower tax rate than the city prior to consolidation. The analysis ignored, however, the increased taxes to be levied on residents of the unincorporated portion of the township to fund the higher level of service (White River Township and City of Greenwood Reorganization Committee, 2009a).
The White River Township trustee funded an independent fiscal analysis which found that residents of the unincorporated portion of the township would experience tax increases of 19% to 27% after consolidation. Although the record doesn't reflect any instance in which the county sheriff weighed in on the consolidation proposal, the trustee's analysis questioned whether the proposed expansion of the city police department would be sufficient to provide the same level of coverage township residents were currently receiving from the county sheriff's department (Pete, 2009).
In the Muncie-Delaware case, interest based conflict emerged only toward the very end of the charter development process. As in other cases, the reorganization committee included provisions intended to mitigate opposition to consolidation. It provided for multiple taxing districts and grandfathered the existing county land use regulations (Muncie-Delaware County Government Reorganization Committee, 2010). The committee also decided early in the process to leave the police department and sheriff department separate to avoid controversy (E. Kelly, personal communication, September 26, 2016). In any event, there was no substantial conflict among the members of the reorganization committee and their deliberations were largely consensual. Local officials and groups opposed to consolidation waited until the committee had finished its work before taking concrete actions to oppose the consolidation effort.
After receiving the plan of reorganization for review, the Muncie City Council and Delaware County Commission commissioned a fiscal analysis that found (similar to the Evansville-Vanderburgh case) that the costs of sheriff patrol would be shifted to nonurban property owners, increasing tax rates in the nonurban area (Crowe Horwath, 2011b). The City Council and County Commission then used their authority to amend the plan of reorganization to add a double supermajority requirement for passage, making voter approval unlikely (Roysdon & Walker, 2012). It is possible that without this ability to insert a poison pill at the last minute, consolidation opponents may have been more active during the drafting of the plan of reorganization. In any event, once the referendum campaign began, the contours of conflict were similar to those in Evansville-Vanderburgh.

Consensual Fact Finding
The four town-township consolidation processes exhibited minimal conflict. In all four cases, town and township officials, rather than a community group, initiated the consolidation process. In each of the four cases, the reorganization committee included common charter provisions intended to minimize the fiscal and regulatory impact on residents and businesses outside the existing town limits, such as multiple tax districts and grandfathered county land use regulations in the non-urban areas (Greater Avon Study Committee, 2012b; Brownsburg Reorganization Committee, 2011; Communities of Zionsville Area for Better Government, 2008a; Yorktown-Mt. Pleasant Township Reorganization Committee, 2011c).
Three of the four town-township consolidations (Avon-Washington, Zionsville-Eagle-Union, and Yorktown-Mt. Pleasant) projected only very small fiscal impacts. The only high cost service the towns had in common with the townships was fire protection and in each case the communities already had longstanding agreements to fund their fire departments. The remaining township services, such as park and cemetery maintenance and poor relief, would be transferred to the town government with no expectation that service costs would change. Property owners and residents in the urban service districts consisting of the area within the pre-consolidation town limits would continue receiving and paying for the higher level of service provided prior to consolidation. The newly consolidated governments would enter into memorandums of understanding (MOUs) with the county governments to continue providing the pre-consolidation level of sheriff patrol and road maintenance to the nonurban areas (Greater Avon Study Committee, 2012b; Brownsburg Reorganization Committee, 2011; Communities of Zionsville Area for Better Government, 2008a;Yorktown-Mt. Pleasant Township Reorganization Committee, 2011c).
With respect to the sheriff patrol, it is not clear whether the reorganization committees made this decision to avoid opposition by the sheriff or to minimize the impact on residents of the unincorporated area (or for both reasons). As a result, the Zionsville-Eagle-Union and Yorktown-Mt. Pleasant consolidations were projected to produce small cost reductions for both urban and nonurban residents, primarily due to the elimination of the township trustee's salary (Crowe Horwath, 2011a;Woodson, 2007). The fiscal analysis presented as part of the Avon-Washington plan of reorganization projected zero net impact on property tax rates for the areas affected by the consolidation (Greater Avon Study Committee, 2012b). In the fourth case, the Brownsburg-Brown-Lincoln reorganization committee did not prepare a fiscal analysis, but because it adopted provisions very similar to those contained in the other towntownship consolidations, there was no indication of substantial fiscal impacts (Brownsburg Reorganization Committee, 2011).
With negligible fiscal impacts and no expensive duplicate services to sort out, the charter development processes in all four town-township consolidations proceeded without conflict. In Zionsville-Eagle-Union and Yorktown-Mount Pleasant the reorganization committees focused their discussion on the most orderly and rational methods for merging town and township services (see, for example, Yorktown-Mt. Pleasant Township Reorganization Committee, 2011a, 2011b; Communities of Zionsville Area for Better Government, 2007Government, , 2008b. In addition to the normal consolidation considerations that characterized the other towntownship consolidation efforts, the fire protection situation in the Avon-Washington case was a focus of charter development efforts. As noted earlier, concerns about the continued viability of the emergency loans and levies funding the fire department was a major motivating factor that led the town and township to consider consolidation (Greater Avon Study Committee 2012a, 2012b). To resolve this issue, the Avon-Washington reorganization plan proposed that the Town of Avon use its financial reserves to purchase fire department capital equipment owned by Washington Township, which would then use the proceeds of the purchase to retire outstanding emergency loans. As part of the consolidation process, the town and township would apply to the Department of Local Government Finance for approval to replace the emergency levy with a permanent operating levy to fund fire protection services for the consolidated government (Greater Avon Study Committee, 2012b).
These town-township consolidation efforts illustrate that charter development need not be a contentious contest of interests. In the four town-township consolidation efforts, town and township officials developed reorganization plans in a consensual fashion with the goal of solving common problems. In Zionsville-Eagle-Union and Yorktown-Mount Pleasant, the primary problem was an annexation threat from a nearby community. In Avon-Washington, the issue was continued sustainable funding for the fire department. The Brownsburg-Brown-Lincoln consolidation effort was motivated by no specific identifiable crisis and was, instead, a general exploration of the potential benefits of consolidation.
While the Zionsville-Eagle-Union and Yorktown-Mount Pleasant efforts proceeded to referenda and were each approved by a large margin, the Avon-Washington and Brownsburg-Brown-Lincoln consolidation efforts were terminated without being placed on a referendum ballot. In each case, the decisions to proceed to a referendum or to terminate the process were based on facts determined during the charter development process. The Greenwood-White River effort was also terminated without a referendum, but for different reasons. The varied reasons for these terminations are the subject of the next section.

Halting the Consolidation Process
Three of the consolidation attempts examined (Avon-Washington, Brownsburg-Brown-Lincoln, and Greenwood-White River) were halted after development of a reorganization plan, but prior to placing the consolidation on a referendum ballot. In the cases of Avon-Washington and Brownsburg-Brown-Lincoln, officials utilized legal provisions allowing them to determine that consolidation was not in the public interest and halt the process, as contemplated in Johnson's (2004) model. The Greenwood-White River consolidation was halted largely due to public official error, but the three cases highlight the possibility that consolidations will not necessarily proceed to a referendum once placed on the agenda, as implied by the City-County Consolidation model (Leland & Thurmaier, 2004a).
In the Avon-Washington case, the Indiana General Assembly provided a decelerating event that resolved the crisis that motivated the consolidation effort. After the reorganization plan was drafted, but before it had been approved by the town council and township board, the Indiana General Assembly enacted legislation allowing township fire departments to convert emergency levies to operating levies (Washington Township Board and Avon Town Council, 2013). This change in law resolved the crisis, leading the town council and township board to reject the reorganization plan as unnecessary (Coggeshall, 2013).
A different concern regarding fire protection derailed the Brownsburg-Brown-Lincoln consolidation attempt. Fire protection is provided by a fire territory created to serve the town and both townships. The plan of reorganization proposed that the fire territory would be dissolved, and a new fire department created to serve the newly consolidated government (Brownsburg Reorganization Committee, 2011). Officials later determined, however, that the new local government would not be able to implement a levy to replace the fire territory's equipment replacement levy, which generated more than $580,000 per year for capital purchases, or about six percent of the fire territory budget. With no viable method to replace that revenue, the town and township elected to reject the consolidation plan and halt the process (Doan, 2012b).
The Greenwood-White River consolidation was ultimately halted by a combination of elite actions, some of which were unintended. The reorganization plan was originally approved by the township board and city council in December 2009, with the referendum set for May 2010 (White, 2009). The plan was later amended to postpone the referendum until November 2010 because the township board and Greenwood mayor felt that additional time was needed to educate the public about the plan (White River Township and City of Greenwood Reorganization Committee, 2009b). Unfortunately, for consolidation supporters, local officials missed the August deadline to place the referendum on the November ballot. Then, during the November elections, new members were elected to the White River Township Board, who rescinded support for the consolidation proposal as their first official act (McLaughlin, 2011).
The terminations of these consolidation efforts highlight the necessity for accounting for multiple potential stopping points in a comprehensive consolidation analytical framework. These cases also underscore the possibility that officials may utilize an opportunity to halt the process when they agree that the consolidation is not in the public interest or that consolidation opponents may use stopping points as a weapon in the conflict to defeat the consolidation. The ability to explore consolidation without necessarily committing to holding a referendum also supports the view that consolidation may be a deliberative and evolutionary process, rather than a conflict driven revolutionary process. This distinction is discussed further in the next section. Hughes and Lee (2002) claim that consolidation may represent the culmination of an evolutionary problem-solving process among the local governments, rather than a revolutionary effort in response to crisis. We argue that the four town-township consolidation efforts, both the successful and the terminated, exemplify the type of evolutionary processes described by Hughes and Lee (2002).

Evolutionary vs. Revolutionary Processes
In Hughes and Lee's (2002) Evolutionary Consolidation Model, local governments explore collaboration as a problem-solving effort when governmental capacity falls short of citizens' expectations. These collaborative attempts begin with informal discussions about a limited number of common issues and progress through ever higher levels of collaborative activity, potentially resulting in full governmental consolidation. These collaborative efforts serve both as fact finding processes, enabling local officials to learn more about local problems and potential solutions, and trust building exercises, allowing local officials to put aside fears of losing control.
The Avon-Washington and Brownsburg-Brown-Lincoln consolidation attempts most closely resemble the situation described by Hughes and Lee (2002). The Avon-Washington discussion of consolidation was explicitly an attempt to solve a fire funding problem common to the town and township governments (Beasor, 2010;Greater Avon Study Committee, 2012A). Furthermore, the fact that they had this problem in common was due to prior collaborative efforts that had resulted in joint funding and oversight of the fire department serving both governments.
In Brownsburg-Brown-Lincoln, the discussion was not focused on a particular problem, but rather an exploration of general concerns common to the two governments (Brownsburg Reorganization Committee, 2011;Doan, 2012a). Similar to Avon-Washington, however, the exploration of consolidation and its impact on the governments followed longstanding collaborative activities related to fire protection.
The Zionsville-Eagle-Union and Yorktown-Mount Pleasant consolidation efforts each possessed a revolutionary quality in the fact they were driven largely by annexation related crisis climates. They both, however, also possessed evolutionary characteristics in that they were consensual fact finding processes and, like the other town-township discussions, built on prior collaborative action in fire protection.

A More Comprehensive Consolidation Model
The Indiana consolidation cases reviewed here demonstrate that existing consolidation models can be synthesized and extended to create a more comprehensive consolidation model. Figures  2, 3, and 4 present graphic depictions of the three stages of the consolidation process: agenda setting, charter development, and referendum campaign. In the agenda setting stage illustrated in Figure 2, consolidation achieves informal agenda status when suggested as a solution to a community problem. Consolidation may be proposed as a proactive response to conditions in a crisis climate, a result of power deflation after other crisis responses have failed, or in cases of no crisis, as a possible next step in ongoing collaborative efforts by the constituent local governments. The process within this stage may be conflictual, when arising from power deflation, or consensual, such as when it is presented as a proactive crisis response or as the next step in ongoing collaborative efforts. If a charter commission is appointed, then the consolidation effort proceeds to the charter development stage.
During the charter development stage illustrated in Figure 3, the charter commission deliberates about the structure and other characteristics of the proposed consolidated government, perhaps with influence from elites and consolidation entrepreneurs. The nature of the deliberation may be either conflictual or consensual, depending on how the proposal was initiated in the agenda-setting stage. In conflictual processes, the parties will tend to frame the consolidation discussion and advance and defend charter terms in ways that serve their interests. In consensual processes, those involved in the deliberations will tend to adopt a fact finding and problem-solving approach to determine if consolidation is a viable solution to common problems. In either type of process, consolidation advocates will seek terms that tend to minimize conflict and promote approval. If the charter commission produces a consolidation charter and the governing bodies place the consolidation question on a referendum ballot, then the consolidation effort proceeds to the referendum campaign.  In the referendum campaign stage illustrated in Figure 4, supporters and opponents of consolidation, including elites, entrepreneurs, and the press may attempt to influence the election outcome. If the process has been conflictual to this point, then the conflict will likely continue. If the process has been consensual, then it may continue to be so, or opponents may become active during this final stage. If there is conflict, it will largely concern the projected impacts of the consolidation, such as fiscal impacts. Ultimately, the referendum is held, and the consolidation proposal is either approved or defeated.

Discussion and Conclusion
This article extends the literature on local government boundary change by examining the full spectrum of local government consolidation outcomes. We augment Thurmaier's (2004a, 2004b) City-County Consolidation model by synthesizing it with Johnson's (2004) Theory of Local Constitutional Change and Hughes and Lee's (2002) Evolutionary Consolidation Model. The key features of the extended model are related to a heightened focus on charter development. Not only do we include charter development as a separately identified step in the local government consolidation process, but we also allow for the possibility that charter development may be a consensual investigation of community needs and solutions, rather than a conflictual contest of interests. Consolidation efforts in which charter development is characterized by investigation and learning are more likely to be evolutionary, as envisioned by Hughes and Lee (2002), than revolutionary, as envisioned by Thurmaier (2004a, 2004b) and Johnson (2004).
The extended model also contemplates additional points at which the consolidation process might be halted creating the potential for a wider variety of consolidation outcomes. A comparative case study of seven local government consolidation attempts occurring from 2008 to 2012 under the common institutional context of Indiana's Government Modernization Act, illustrates the suitability of our revisions to the C3 model and demonstrates the applicability of the model to other forms of consolidation in addition to city-county.
The findings of this study underscore the role and importance of conflict and consensus in the consolidation process. We find that when conditions allow consolidation to be approached as a consensual attempt at problem-solving, then the prospects of a successful consolidation effort are enhanced, compared to efforts which are conflictual. Four of the seven cases were characterized by consensus. Of these four, two (Yorktown-Mount Pleasant and Zionsville-Eagle-Union) were approved by a large margin in the referendum. The other two consensual cases were halted without going to referendum because the reorganization committee decided either that consolidation was unnecessary (Avon-Washington) or against the public interest (Brownsburg-Brown-Lincoln).
In the other three cases, the consolidation effort was highly conflictual. In none of these three cases were consolidation proponents successful. In two cases (Muncie-Delaware and Evansville-Vanderburgh), the consolidation proposal was soundly defeated at referendum. In the third case (Greenwood-White River), the process was halted when officials inadvertently missed the deadline to place the question on the ballot. Officials had originally planned to place the question on the ballot in an earlier election but decided to delay because the high level of conflict highlighted the need for more time to build support.
The three conflictual cases also corroborate that taxation is a major issue in consolidation politics regardless of the level of government. In the two consolidation efforts that were defeated at referendum and the one that was halted by a missed deadline, the potential for shifting tax burdens among taxpayers was a primary point of controversy leading to the conflict.
Our findings provide additional insight into Leland and Thurmaier's (2005) finding that consolidation proposals based on efficiency and cost savings arguments are rarely successful. In the two cases with unsuccessful referenda (Muncie-Delaware and Evansville-Vanderburgh), consolidation advocates promoted consolidation as a means to efficiency and cost savings and, as Leland and Thurmaier (2005) have noted in other cases, voters did not buy their argument. On the contrary, we confirm Leland and Thurmaier's (2005) finding that economic development concerns are more likely to motivate voters to support consolidation than promises of greater efficiency. In the cases of the two successful consolidation referenda (Zionsville-Eagle-Union and Yorktown-Mount Pleasant), local officials presented consolidation as the solution to problems of development.
We also confirm Leland and Thurmaier's (2005) finding that neither a crisis climate nor power deflation are necessary preconditions for local governments to consider consolidation. Out of six cases with an identified crisis, only two proceeded to a successful referendum and neither of those involved power deflation. Of the three cases, with power deflation, none were successful. Two were defeated at referendum and the third was halted because the crisis was resolved without consolidation.
For practitioners, we recommend they familiarize themselves with stages of Hughes and Lee's (2002) Evolutionary Consolidation Model and the consensual problem-solving approach adopted by officials in Zionsville-Eagle-Union, Yorktown-Mount Pleasant, Avon-Washington, and Brownsburg-Brown-Lincoln. Consolidation may or may not be the appropriate solution to problems facing local governments within a county or region, but discussion and exploration of cooperative efforts among nearby local governments may lead to productive solutions.
Finally, this study continues the tradition of synthesis that has informed theory building in the study of local government consolidation. The extended and revised model presented here allows for the analysis of a wider variety of local government consolidations, including those involving municipal and township governments and accounts for varied charter development processes. It also allows for multiple potential stopping points, permitting the incorporation of the wide variety of consolidation process that exist across different states. This analysis is particularly relevant for local governments in states which have the township layer of government. It is our hope that scholars and policy makers will find this revised theory useful in the study, design, and implementation of future consolidation efforts.