Collaboration for sustainability? A framework for analyzing government impacts in collaborative-environmental management

Abstract

Citizen participation and empowerment are critical sustainability elements. One increasingly popular form of citizen participation is collaborative environmental management (CEM). This approach has been described as a new way of governing for environmental issues, an alternative to government-centered processes, that empowers stakeholders and citizens to play a dominant role in planning and decision-making. This paper describes a new analytical framework, called the Governmental Impact Framework, for understanding how government affects CEM and the sustainability of outcomes. This framework incorporates institutional analysis to illuminate government-stakeholder relationships and the interplay of biophysical and social factors. Applying the framework to a collaborative land-use planning case in the American state of Ohio indicates that governments are more dominant in collaborative processes than previously thought, and that the channels of influence vary along several dimensions.

Introduction

A key component of sustainability and sustainable development is citizen empowerment in decisions shaping social and environmental conditions. Across a wide range of settings, community participation has been found to affect sustainability prospects. This supports the Brundtland Commission's seminal report on sustainable development (WCED, 1987), which argued that, for communities to articulate and enforce their common interest in sustaining natural resources, community members need to effectively participate in decision making. Moreover, a leading scholar in common pool resource management concluded that including affected individuals in rule making about resources is critical for sustainable human-environment systems (Ostrom, 1990). In addition, the "seeds of change" approach to sustainable development argues that empowering local citizens to participate and take action in their own "backyards" is a prerequisite for sustainable communities (Cuthill, 2002).

One strategy to foster citizen participation and empowerment is the decentralization, or transfer of authority, from central governments to local governments or community organizations. While this may seem straightforward, it often does not happen. Moreover, such divestment neglects possible governmental contributions in expertise, technical information, manpower, institutional networks, or other resources. In fact, sustainability scholars have emphasized the importance of collaboration among a range of individuals, both within and outside formal government structures (Becker et al. 1999; de Jongh & Captain, 1999).

While many government policies have led to unsustainable results, government also assists in fostering sustainability. At a minimum, government can recognize community rights to make resource-use decisions and rules tailored to the local context (Ostrom, 1990). Structures and institutions that increase government transparency and accountability can strengthen citizen empowerment to achieve sustainable results (Lyons et al. 2001). More actively, government might create institutions to encourage individual behavioral changes and encourage policy changes that address local environmental issues (Rich et al. 1995).

An increasingly popular institution for addressing environmental issues is collaborative-environmental management (CEM). Collaboration refers to a process of engaging citizens, along with government officials and other interested stakeholders, in all phases of the policy process. Whereas some forms of participation are directed at gathering citizen input on plans developed by public administrators, collaboration involves engaging stakeholders in priority setting and in the planning, implementation, and evaluation of solutions. It is a process in which diverse stakeholders work together to resolve a conflict or develop and advance a shared vision (Gray, 1989). By coming together and deliberating, stakeholders generate a more comprehensive understanding of problems and possible remedies.1

CEM has arisen in a wide variety of contexts. Sometimes, government agencies promote it to generate outcomes unattainable via more traditional approaches. In the United States, the Environmental Protection Agency has expended millions of dollars through its Section 319 grant program to provide funding for collaborative-watershed projects. States such as Ohio, Colorado, West Virginia, Washington, and California have promoted collaborative-watershed efforts through technical assistance, funding, and provision of personnel (Collins et al. 1998; Schott & Koontz 2002; Steelman & Carmin, 2002; Sabatier et al. 2005). CEM has also been conducted in the Netherlands, Canada, Australia, the United Kingdom, and Southeast Asia (de Jongh & Captain, 1999; Meadowcroft, 1999; Leach & Pelkey, 2001).

Grassroots-citizen organizations have initiated some CEM efforts. In such cases, concerned citizens typically perceive an environmental-management crisis that government has caused or has not adequately addressed. The citizens respond by creating a collaborative group that draws in government officials as participants. One such case was the Applegate partnership in Oregon, as described by Moseley (1999). In the face of litigation and acrimony surrounding northern spotted-owl protection, the Forest Service and Bureau of Land Management dramatically reduced timber-harvesting activities from federal public lands in the Applegate Valley. Residents concerned about jobs, social conditions, and the environment united to advance their shared interests in the public lands, inviting federal officials to participate. Marsten (2001), Snow (2001), Steelman & Carmin (2001), Weber (2003), and others have described similar citizen initiated "grassroots" collaborative efforts.

Whether citizen-initiated or government sparked, CEM efforts often have been described as a new way of governing and an alternative to government-centered policy. But governments rarely leave the picture entirely. They are often key stakeholders and, even if not, provide the institutional and political setting within which CEM efforts play out. As more citizens and government officials grapple with increasingly complex, multi-media, multi-jurisdictional environmental-management issues, and as more parties engage in collaborative efforts, we need to examine more carefully governmental roles in CEM, and how these roles affect sustainability.

This paper uses the analytical framework developed by Koontz et al. (2004) to examine a collaborative-management effort in the United States. After explaining how the framework combines environmental and social elements, the paper then applies it to a multiple-case study of collaborative planning for farmland preservation. Although sustainability may be most often associated with forests and watersheds, productive farmland represents a critical resource at the human-environment interface that has long been viewed as important for social and ecological sustainability (Olson & Lyson, 1999). The paper concludes with a discussion of how collaboration in farmland preservation suggests broader implications for sustainability and government involvement in other collaborative efforts.

A New Analytical Framework

The analytical framework employed here draws on prior research about the factors affecting CEM processes and outcomes, combined with insights from institutional analysis. In particular, the Institutional Analysis and Development (IAD) framework developed by Ostrom and colleagues (1994) provides a key foundation. The IAD framework suggests sets of variables likely to affect outcomes arising from human interaction in light of biophysical, cultural, and institutional contexts. As a framework rather than a theory, IAD organizes inquiry and can be animated by particular theories to match a given setting. For example, scholars attempting to understand behavior within markets may employ neoclassical microeconomic theory, while those who study behavior in hierarchies may draw on principal-agent theories.

The IAD framework emphasizes the importance of rules and institutions. Rules are prescriptions that forbid, permit, or require certain actions in particular contexts and specify the sanctions authorized if the prescriptions are violated (Crawford & Ostrom, 1995). By creating, enforcing, and changing rules, a group of individuals may be able to overcome collective-action dilemmas. More broadly, people create other institutions, such as norms and shared strategies that influence individual choices. Fundamentally, human action both affects, and is affected by, institutions. Thus, the IAD framework draws the analyst's attention to the interactions between institutions and individual decision making, as well as the interactions of these variables with aspects of the physical world and community culture. The focal point in the IAD framework is the "action arena" where participants decide among diverse actions (Ostrom, 1990; Ostrom et al. 1994; Imperial, 1999).

The IAD framework provides a solid foundation for questions about interorganizational and interpersonal relationships relating to the environment and decisionmaking. From this basis, plus additional insights from other scholarship, Koontz et al. (2004) developed a framework for analyzing governmental impacts on collaborative-environmental management. This new framework specifies that governments, both as institutions and as actors, affect CEM efforts. Government impacts CEM through three primary channels: issue definition, resources, and structure/decision processes. Figure 1 illustrates the analytical framework.

[Image omitted: see PDF]

Figure 1 Framework for Analyzing Governmental Impacts on Collaborative-Environmental Management (Koontz et al. 2004).

Governmental actors are people with attitudes, beliefs, skills, and values who engage in decision making and interactions that can influence collaborative processes and outcomes. Governmental institutions are the rules, structures, laws, norms, and socio-cultural processes of the administrative state that shape human action. While governmental institutions provide constraints and opportunities within which governmental actors work, at the same time governmental actors may shape and alter governmental institutions.

Issue definition refers to how an issue is framed, what set of solutions is considered, and the issue's scale. Political scientists and policy scholars have long examined the struggles over framing an issue, which can greatly affect which stakeholders become involved and how the issue works through political and administrative processes (Schattschneider, 1960; Snow & Benford, 1988). The biophysical scale of an issue is of particular importance for environmental management, because it involves multiple scales across a landscape. For example, on a fine scale, watershed-management efforts may involve tributaries or portions of stream reaches, but these are embedded in broader-scale units, including sub-basins and basins that may cover thousands of square miles. Choices about the ecological scale of management can have profound effects on collaborative processes and outcomes (Thomas, 1999).

Resources for collaboration divide into three broad categories: human, technical, and financial. Human resources include a collaborative effort's volunteers, leaders, and staff members. These individuals possess knowledge, skills, and experience that may advance collaboration. In many cases, an entrepreneurial leader must be available to establish a new collaborative group (see Moseley, 1999). Technical resources refer to knowledge about the environment, and about the local context, that can inform collaborative efforts. Such capabilities may be found among group members or garnered from external sources. Finally, financial resources are the funding and in-kind contributions that allow a group to conduct business and perform activities. The amount and sources of funding shapes group activities substantially. For instance, member donations may be applied to a wide variety of activities, while grants may stipulate the activities to be performed (Steelman & Carmin, 2001).

Group structure refers to the way membership and activities are organized. Some groups rely on a strong leader while others create shared leadership through advisory boards or executive councils. Group structure includes the administrative processes that coordinate work, such as monthly meetings and standing or ad-hoc committees. Group decision-making processes are the means of aggregating individual preferences into decisions. A common decision rule is consensus, though not all collaborative groups use this approach, and different rules may be used for different types of group decisions. Group decision-making processes may be used to select participants, create plans, choose strategies, and allocate resources, among other things. A group's authority to make externally binding decisions is important. Often, collaborative groups are advisory only, in that policymakers (government agencies and elected officials) can accept or reject any recommendations, but in some instances collaborative groups may be granted policy authority.

CEM outcomes are notoriously difficult to evaluate. Many scholars argue that the ultimate measure of success is ecological conditions (Schweik & Thomas, 2002; Conley & Moote, 2003; Sabatier et al. 2005). Environmental sustainability, by its very nature, suggests conditions that endure. But it is extremely difficult to link CEM processes to ecological conditions due to long time horizons and multiple interacting variables. Moreover, few CEM groups monitor their activities' environmental-quality effects; instead they focus on more tangible outcomes like plans, projects, best-management practices, and policy changes. In this study, the focus is on these more concrete outcomes, especially farmland preservation-plan creation and implementation.

In addition to environmental conditions, social conditions are a core element of sustainability. Thus, measurement can include social outcomes such as individual well-being, equity, and enhanced community cohesion regarding pressing problems. Prior studies indicate that collaboration may lead to improved trust and understanding among diverse stakeholders, as well as to increased legitimacy in the broader community for particular activities. In this study, social-outcome measures focus on community-capacity enhancement to solve problems, including network building and understanding.

Applying the Framework to Farmland Preservation

Farmland is an important natural resource critical to both human societies and ecological systems. But 1.2 million acres of productive farmland in the United States is converted each year to development (American Farmland Trust, 2006). Conversion of prime farmland to housing, roads, parking lots, and other physical infrastructure has harmed rural cultures and economies in many parts of the country (Daniels, 1999; Olson & Lyson, 1999) and degraded biodiversity, wildlife habitat, and water quality (Reilly 1996; Heimlich & Anderson 2001; Hansen et al. 2005; Theobald et al. 2005). A number of states have responded by encouraging preservation planning. The state of Ohio, for instance, created a grant program to encourage local-stakeholder collaboration to preserve farmland, as described below.

The Ohio Farmland Preservation Planning Program

In June 1998, the Ohio Department of Development's Office of Housing and Community Partnerships (OHCP) announced a matching-grant program encouraging rural counties to prepare local farmland-preservation plans. The OHCP was following a recommendation from the Ohio Farmland Preservation Task Force (OFPTF), formed in 1996 to develop locally-based tools for promoting farmland preservation. The program's objective was to encourage counties to "gather appropriate data from which local goals [could] be established relative to the agricultural industry and farmland" (Graves, 1998). Goal setting was seen as a prerequisite for strategic planning. By encouraging local planning that included farmers, the OFPTF hoped to build grassroots support to sway state legislators to support farmland preservation (Wise, 2000). The Ohio Farmland Preservation Planning Program (OFPPP) was independent from other local land use-planning processes. It should be noted that Ohio has no statewide-planning mandate and local governments decide whether or not to do land-use planning. In some counties with broader programs, farmland preservation-task forces interfaced with these prior initiatives.

The OFPPP made grants of up to US$10,000 available to each of the 81 counties eligible for Federal Small Cities Community Development Block Grant funds. Funding was contingent on the counties providing a 1:1 match in dollars or in-kind and each jurisdiction establishing a county farmland-preservation task-force that included a "cross section of interests" (Graves, 1998). The state granted the funds to county commissioners who were then responsible for creating the task forces. The grants came with very few strings attached, requiring only that each task force produce a plan that included soils information, submit it to the OHCP by December 31, 1999, and include a "cross section of interests" in creating the plan. This flexibility reflected the understanding that different counties had different capacity and readiness to address farmland preservation (Wise, 2000).

Of the 81 eligible counties, 61 participated in the grant program. The non-participating counties were disproportionately located in the economically challenged Appalachian region of southeast Ohio. The underrepresentation of this area stemmed from the low salience of farmland preservation and lack of available matching funds (Wise, 2000).

A notable, but not that unusual, aspect of the grant program was its decentralized nature. Task forces were free to operate as they saw fit, with little oversight from state officials or interaction with other task forces. Thus, local stakeholders were allowed to choose topics for investigation and emphasis in a bottom-up fashion (within the bounds of farmland preservation), as recommended by Becker et al. (1999). While county commissioners served as the official grant recipients, in practice many delegated responsibility for convening the task force to other governmental actors or to citizens. The individual leaders organized and managed the task forces in different ways across the state. In a few instances, counties folded the task forces into ongoing land-planning efforts, but most of them were stand-alone organizations.

Like many other grant programs, the OFPPP is a government institution whereby the granting agency provides financial resources and then steps aside as the grantees craft processes tailored to their local circumstances. Less politically charged than regulations, grants may be especially attractive, and they appear to support widely held commitments to vest greater authority in local decision making. But government-grant programs, as institutions, may have a bigger impact on collaboration than conventionally thought. To investigate OFPPP impacts on task-force process and outcomes, a multiple-case study was conducted.

Research Methods

The results described below come from a comparative case study of fifteen farmland preservation-task forces participating in the OFPPP. The principal investigator selected cases to ensure a wide range of geographic locations and urbanization levels within the American state of Ohio. First, at least two cases come from each of the following regions: central, northwest, northeast, southwest, and southeast. Second, although all of the counties in the program are designated as rural, they exhibit a range of urbanization levels. Following Sharp & Vinland (2000), at least two cases come from each non-urban county category: non-metropolitan, small metropolitan fringe, small metropolitan core, and large metropolitan fringe.

To gather information about task-force processes and outcomes, the research team interviewed two to four key informants for each task-force with each interview lasting about one hour. Outside the task forces, the research team interviewed a state official who worked closely with the county-task forces, as well as county commissioners in each of the fifteen counties two years after the task-force plans were completed. Although the research team did not attend any task-force meetings, the principal investigator was present at a statewide farmland-preservation conference that included representatives from most of the 61 grant-receiving counties. An additional data source was documentary, with the research team conducting content analysis of fifteen completed farmland-preservation plans. Other documents included task force meeting-attendance records and expenditure reports. These data yielded information about how governmental actors and institutions affected issue definition, resources, group structure and decision processes, and subsequently outcomes.

Results

Issue Definition: Government grants targeted to particular issues, such as farmland preservation, play an integral role in issue definition. The OFPPP established objectives centering on preserving farmland, a fairly narrow construction, compared to a more general land-use definition, or even such alternatives as green space or open space. Framing the issue in terms of agricultural landscapes and farming issues is not too surprising since the grant program's impetus was a recommendation of the OFPTF that included members with high stakes in farmland preservation. Thus, county task-force leaders invited participants primarily from the agricultural community, rather than a more diverse range of affected interests. In particular, environmental and urban interests often lacked representation on the task forces (see Table 1), diminishing support, beyond the agricultural community, for many task-force recommendations.

Table 1 Summary of governmental impacts on collaborative processes and outcomes.

[Image omitted: see PDF]

click table for larger view

The grant program established the biophysical boundaries as the county level in rural areas (urban counties were not eligible). The county-level spatial scale was appropriate in several regards. First, the grant-program creators were aware that existing soil data from the Natural Resource Conservation Service and Ohio Department of Natural Resources were aggregated at the county level, and thus readily available for county-level planning. Second, counties in Ohio are small enough to make task-force meetings accessible to any county resident. Third, as an important political subdivision of the state, the county is where most of the land-use authority resides. This meant that task-force recommendations could be readily transmitted to the appropriate government officials. In fact, the incorporation of task-force recommendations into county land-use plans was one goal of the program (Wise, 2000).

However, by focusing only on the county scale, task forces were hindered on several fronts. In many instances, agricultural issues and farmland preservation were affected by land-use trends in other counties, yet neighboring counties were not included in the task-force deliberations. For example, watershed boundaries typically do not fall along county lines, yet land uses upstream can influence the quality of downstream water and land substantially. As another example, growing urban populations and associated development in nearby counties can pressure land use in a given county, yet such forces were beyond the purview of task-force plans. It is also worth mentioning that, in some locations, land-use authority rests not with counties but with smaller townships, so the county is not always the most appropriate level for policy planning.

The OFPPP, a statewide governmental institution, was added to a set of existing governmental land-use institutions that substantially affected issue definition, especially through limits on the set of feasible policy solutions. In locations with land-use zoning ordinances in place, zoning was seen as a powerful tool for farmland-preservation plans. But in locales without existing land-use zoning ordinances, task-force members were acutely aware of the negative political ramifications of recommending zoning. As one member said, "We did not use the 'Z' word...The more rural you get, the more resistant people are to zoning."

Resources: Through the state-grant program, counties received important financial resources in the form of US$10,000 awards. Task forces could use the funds to establish groups, provide meeting space and resources, obtain technical information, and create plans. The flexibility of the grants allowed each task force to tailor its expenditures to its needs. From the state's perspective, the 1:1 match requirement created leverage to bring local resources to bear on the planning process. Task forces in a few counties garnered more than the US$10,000 match through the entrepreneurial efforts of local political leaders and agency personnel who linked farmland preservation to other land use-planning efforts underway in their jurisdictions.

The OFPPP did not directly provide any human resources from the state. Rather, county commissioners were delegated authority to oversee the creation of the task forces. Such devolution of authority can help to build local capacity and increase trust, especially in a state such as Ohio with a strong tradition of local control ("home rule") in land-use matters (see Thomas, 1999). In practice, it can also hinder attempts to draw in diverse interests that are not presently part of the local-power structure. Instead of seeking leadership beyond traditional-government officials, most task forces were led by local-government actors.

Human resources include not only group leaders, but members as well. By defining the issue as farmland preservation and by delegating responsibility to county commissioners, the state program encouraged participation by farmers and local government officials in most task forces. While these members may have had the greatest stake in farmland preservation, they also recognized that the issue did not generate much concern or policy agreement in the broader community.

The OFPPP grant guidelines had minimal technical requirements--only that the eventual plan include a soils map. But task forces typically went beyond this requirement, seeking to incorporate technical data. The state program did not systematically provide such information, although state officials did cooperate in convening an annual farmland-preservation conference that was well attended by task-force members from across the state. Few of the task forces turned to their peers for technical data as the grant provisions did not explicitly require or encourage them to do so. Instead, the task forces gathered information largely through trial and error. Data included in final farmland-preservation plans tended to draw on material from the census of agriculture that was not customized to fit the local conditions. Rather, the plans typically just listed categories used in the census. The ready availability of government sources made them attractive to task forces seeking technical data.

Group structure and decision processes: The state-grant rules were silent on matters of group structure and decision process, so government institutions were not influential here. However, government actors led many of the task forces and their engagement in this regard led to notable impacts on this variable. Groups created a variety of structures, ranging from none to many committees and monthly to quarterly meetings with a range of agenda items. Decision processes in some task forces were consensus-based, while in others majority rule was used. The one decision rule common to all of the taskforces was their advisory function; no task-force decisions or recommendations were binding outside of the group. Thus, the task forces' ability to affect policy rested with their persuasiveness to policymakers.

Outcomes: The primary environmental-management outcome for the OFPPP planning effort was the final farmland-preservation plan. Given the diversity of resources, group structures, and decision processes, and the fact that different counties faced different political, social, economic, and land-use circumstances, it is not surprising that final plans varied considerably across the fifteen task forces. As a simple measure of complexity, the plan page counts ranged from 14 to 331, and the number of elements suggested by the state-program guidelines (e.g., number of farms in the county, geographic location of farmed acres in the county) ranged from one to twelve. The single technical requirement, that the plan contain a county-soil map, was met by just twelve of the fifteen final plans. In terms of plan implementation, two years after plan completion, commissioners in eight of the fifteen counties could point to particular recommendations that had been adopted into policy, ranging from incorporation into "smart growth" and comprehensive planning, to funding decisions for purchasing land use-development rights, to changing zoning ordinances.

These environmental outcomes are linked, in part, to the local-governmental actors who played leadership roles on many task forces. Group structure and decision-process choices shaped how they functioned and created their plans. The state-grant program, as a governmental institution, influenced plan contents by defining the issue as farmland preservation and setting the biophysical scale as the county, and by encouraging governmental actors to participate. The flexibility in how funds could be spent and in group structure and decision process led to a diverse array of plan contents. Local zoning ordinances, as a governmental institution, influenced the recommended set of feasible policy alternatives.

In addition to environmental outcomes, task forces generated many social outcomes. Social capital was often enhanced, with many members more willing to participate in future community problem-solving. At the same time, however, some task-force members' motivation was dampened by a lack of implementation. Members who expected their recommendations to lead to change were frustrated when policy enactment proved slow. Another social outcome, network building, was widely realized with key informants in twelve of the fifteen task forces reporting improved network ties after the planning process. These ties were often prompted by governmental actors who shared their connections with people and organizations outside the task forces. Finally, in all fifteen cases, key informants reported that the grant program led to increased stakeholder interaction and greater understanding of farmland-preservation issues and possible solutions.

Discussion

As described above, the framework for analyzing governmental impact can help us to understand how government as both institution and actor can affect CEM through issue definition, resources, and group structure and decision processes. In the case of farmland-preservation planning, the OFPPP, a governmental institution, sparked the initial collaborative efforts, defined the issue, and provided the bulk of the financial resources for many groups. The program also interfaced with local-governmental institutions--zoning ordinances--which constrained the recommended set of feasible policy alternatives.

The role of government as actor was played not by state officials, but by a wide range of local-government officials who were involved in creating and maintaining task forces and planning activities. In many groups, governmental actors provided critical human, technical, and financial resources. Moreover, governmental actors participating in the task forces were involved in establishing group structures and processes. The importance of local-governmental institutions and actors in shaping the task forces indicates that, even as one government seeks to spark collaboration through purely institutional means, the way in which that collaboration plays out can depend on governmental actors and institutions at other levels.

Overall, the farmland-preservation study illustrates that, when government resources encourage collaboration they typically have wide-ranging impacts beyond resource levels. While a grantor may desire to let its grantees work with flexibility and be responsive to local conditions, in fact the recipient is not likely to work "outside of government." Rather, government actors and institutions at multiple levels will affect local collaborative efforts. Thus, collaboration is not an alternative to government; rather, it occurs within constraints and opportunities created by existing governmental institutions.

While governmental institutions often significantly affect collaborative activities, it is important to note that governmental actors can mediate the impact of institutional initiatives. For example, entrepreneurial government actors in some of the county-task forces were able to tap into other government institutions for additional resources.

Another implication of this study is the importance of stakeholder expectations. The OFPPP created a purely advisory role for the county-task forces. Yet some key informants expressed disappointment at the limited degree to which their recommendations were adopted by county commissioners. For task-force members who had expended considerable time and energy in the collaborative process, seeing their recommendations "gathering dust on a shelf" reduced their enthusiasm for future collaboration. Obviously, this diminishes the social outcome of increasing community capacity to work together to solve future problems. In fact, such collaborative "fatigue" has been noted in other research studies (e.g., Korfmacher, 1998).

These results lead to suggestions for public managers and policymakers looking to CEM to solve environmental problems. For managers, it is important to be explicit about citizens' roles in the policy process, especially the degree to which collaborative recommendations will be binding. Participants who expect that their recommendations will carry weight in policymaking can become disillusioned and distrustful if they are subsequently treated only as advisory, rather than empowered as Cuthill (2002) and others recommend.

For policymakers, it is important to consider the costs and benefits of collaborative approaches and to choose when and where to support such efforts. Collaboration is more likely to succeed when agencies and actors are willing to adopt more flexible approaches to planning and implementation, as compared with traditional regulatory approaches. This can foster the Brundtland Commission's recommendation to provide community members with opportunities to effectively participate in decision-making processes (WCED, 1987). Allowing local groups to address issues in contextually relevant ways is likely to yield outcomes that vary across locations. Some agencies with strong traditions of technocratic expertise may have a culture and bureaucratic structures that resist collaboration.

Conclusion

Citizen participation in policy and decision making is an important element of sustainability. Government and citizens have been collaborating in the environmental arena for a number of years, yet careful empirical study across cases is just beginning. While each case of collaboration is born of a specific context, with a particular set of stakeholders facing particular issues, social scientists need to seek underlying concepts to better inform public managers, policymakers, and citizens about the links between government and collaborative efforts, and how these efforts affect sustainability.

This paper has presented an analytical framework, developed from prior research and cross-case analysis, to orient researchers to particular sets of variables for investigation. The framework suggests governmental actors and institutions influence CEM processes primarily through issue definition, resources, and group structure and decision processes, which ultimately affect environmental and social outcomes. These dimensions provide guidance for understanding the workings of a farmland-preservation program implemented in the American state of Ohio. While government-as-institution determined issue definition and provided critical financial resources, government-as-actor shaped human and technical resources and group structure and decision processes. In the end, high plan variability reflected differences in local contexts, even as the program led to substantial social outcomes across most of the groups.

The results highlight the thorny issue of citizen roles and expectations, including whether group recommendations will be binding. Since citizen trust and future capacity to solve collective issues depends, in part, on congruence between expectations and reality, managers should clearly articulate the degree to which authority will be shared with citizens. Some governmental actors, especially those that emphasize technocratic expertise, may resist such sharing of authority.

If we are to make informed choices about collaborative solutions for environmental and social problems, we need analytical tools to empirically examine this growing approach. Unpacking the term "government" into the concepts of actors and institutions can help to clarify governmental roles and influence on collaborative efforts. Thus, the framework presented here should provide useful guidance for understanding CEM and its contributions to sustainability.

Acknowledgement

The author acknowledges contributions from Toddi Steelman, JoAnn Carmin, Katrina Smith Korfmacher, Cassandra Moseley, and Craig Thomas who contributed to the broader initiative of which this research is a part. Research assistance was provided by Rachel Brakeman, Travis Masters, Sara Nikolic, and Josh Stephens. Funding for this project came from The Ohio State University School of Environment and Natural Resources, the C. William Swank Program in Rural-Urban Policy, and the Ohio Agricultural Research and Development Center 's Research Enhancement Competitive Grant Program. An earlier version of this paper was presented at the Institutional Analysis for Environmental Decisionmaking Workshop held on 28-29 January 2005 at the U.S. Geological Survey's Fort Collins Science Center.

------------

Notes

1 For key works addressing CEM see Cortner & Moote, 1999; Wondolleck & Yaffee, 2000; Brick et al. 2001; Leach et al. 2002; Koontz et al. 2004; Sabatier et al. 2005.