Citation
Environmental nonprofit advocacy : a three-paper examination of organizations, venues, and activities

Material Information

Title:
Environmental nonprofit advocacy : a three-paper examination of organizations, venues, and activities
Creator:
Kagan, Jennifer Anne
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Doctorate ( Doctor of philosophy)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
School of Public Affairs, CU Denver
Degree Disciplines:
Public affairs
Committee Chair:
Heikkila, Tanya
Committee Members:
Weible, Christopher M.
Ely, Todd L.
LeRoux, Kelly

Notes

Abstract:
This dissertation explores the advocacy activities of environmental nonprofits across venues. These organizations are understudied as a distinct type of entity across literatures. While the policy process literature recognizes that environmental nonprofits actively participate in environmental policymaking, the unique characteristics of these organizations are often overlooked. Furthermore, the nonprofit literature recognizes that environmental nonprofits are more likely to advocate than any other type of nonprofit, but few studies have specifically examined the characteristics and activities of these organizations. With these considerations in mind, this dissertation includes three empirical chapters. The first empirical chapter relies on news media to explore differences and similarities between industry and environmental advocacy groups with respect to the venues and issues in which they engage. The second empirical chapter in the dissertation zooms in on environmental nonprofits and explores, through interviews, the types of advocacy activities these organizations use across sectors and how the external contexts and internal characteristics of these organizations may influence their advocacy. The third empirical chapter in the dissertation examines differences between 501(c)(3) and 501(c)(4) organizations in the environmental space using survey responses combined with publicly available nonprofit data. While in service delivery sectors, 501(c)(3) organizations tend to be service providers and 501(c)(4)s are advocacy organizations, the third empirical chapter tests whether this distinction holds true among environmental advocacy organizations. Overall, the dissertation contributes to both the nonprofit and environmental policy literatures by suggesting that environmental nonprofits use a complex web of advocacy activities across multiple venues to influence policy. The dissertation shows that the relationships between organizational and contextual factors, on the one hand, and activities and venues, on the other hand, are more complicated than previously assumed. Finally, the dissertation provides empirical support for differences between environmental nonprofits and other types of nonprofits that warrant future studies about these organizations, their characteristics, and their activities.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
Copyright Jennifer Anne Kagan. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item has the following downloads:


Full Text
ENVIRONMENTAL NONPROFIT ADVOCACY:
A THREE-PAPER EXAMINATION OF ORGANIZATIONS, VENUES, AND ACTIVITIES
by
JENNIFER ANNE KAGAN B.A., New York University, 2000 J.D., University of San Diego School of Law, 2004 M.P.A., University of Colorado Denver, 2013
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Public Affairs Program
2019


©2019
JENNIFER ANNE KAGAN ALL RIGHTS RESERVED
11


This thesis for the Doctor of Philosophy degree by Jennifer Anne Kagan has been approved for the Public Affairs Program by
Tanya Heikkila, Chair Christopher M. Weible Todd L. Ely Kelly LeRoux
Date: August 3, 2019


Kagan, Jennifer Anne (PhD, Public Affairs Program)
Environmental Nonprofit Advocacy: A Three-Paper Examination of Organizations, Venues, and Activities
Thesis directed by Professor Tanya Heikkila
ABSTRACT
This dissertation explores the advocacy activities of environmental nonprofits across venues. These organizations are understudied as a distinct type of entity across literatures.
While the policy process literature recognizes that environmental nonprofits actively participate in environmental policymaking, the unique characteristics of these organizations are often overlooked. Furthermore, the nonprofit literature recognizes that environmental nonprofits are more likely to advocate than any other type of nonprofit, but few studies have specifically examined the characteristics and activities of these organizations.
With these considerations in mind, this dissertation includes three empirical chapters.
The first empirical chapter relies on news media to explore differences and similarities between industry and environmental advocacy groups with respect to the venues and issues in which they engage. The second empirical chapter in the dissertation zooms in on environmental nonprofits and explores, through interviews, the types of advocacy activities these organizations use across sectors and how the external contexts and internal characteristics of these organizations may influence their advocacy. The third empirical chapter in the dissertation examines differences between 501(c)(3) and 501(c)(4) organizations in the environmental space using survey responses combined with publicly available nonprofit data. While in service delivery sectors, 501(c)(3) organizations tend to be service providers and 501(c)(4)s are advocacy organizations,
IV


the third empirical chapter tests whether this distinction holds true among environmental advocacy organizations.
Overall, the dissertation contributes to both the nonprofit and environmental policy literatures by suggesting that environmental nonprofits use a complex web of advocacy activities across multiple venues to influence policy. The dissertation shows that the relationships between organizational and contextual factors, on the one hand, and activities and venues, on the other hand, are more complicated than previously assumed. Finally, the dissertation provides empirical support for differences between environmental nonprofits and other types of nonprofits that warrant future studies about these organizations, their characteristics, and their activities.
The form and content of this abstract are approved. I recommend its publication.
Approved: Tanya Heikkila
v


ACKNOWLEDGEMENTS
I am incredibly grateful to have been a part of the School of Public Affairs at the University of Colorado Denver for the past eight and a half years, first as an MPA student, then as Director of the Wirth Chair in Sustainable Development, and finally as a PhD student.
Thank you in particular to Tanya Heikkila and Chris Weible. I feel so fortunate to have had the opportunity to work with you both. You not only model great scholarship and collegiality, but also how to be good, patient, understanding, and compassionate people. It has been an honor to be part of the WOPPR lab and various research teams throughout my time at SPA, and I truly appreciate all of the support and opportunities I received through working with you both. Thank you also to all the current and former students I have the opportunity to work with through the WOPPR lab. I truly made some lasting friendships throughout this time and am grateful for the connections, laughter, and support.
Thank you to Tanya for serving as my dissertation chair. I could not have asked for a better advisor. You gave me exactly the right amount of support when I needed it and trusted me to get things done when I needed to write, code, interview, or organize my thoughts. I am always amazed by how quickly you turn things around with thoughtful and helpful comments. I have learned so much from you on both personal and professional levels.
Thank you to Chris and to Todd Ely for serving on my dissertation committee. Chris, I always enjoy talking with you, sharing stories, and hearing your perspectives on research. Your insights keep me honest and help me to think outside the box. Todd, I am so grateful that I took 5004 with you back in 2011, because that was really the start of my deeper engagement with SPA, including my work with the Buechner Institute, with Tanya and Chris, and with the Wirth Chair. Your thoughts and comments on my dissertation have been incredibly helpful.
vi


I am also grateful to Kelly LeRoux for serving as the external member of my committee. Kelly, I truly admire your work and feel very fortunate to have had your expertise and input on my dissertation. It is always a pleasure to spend time with you at conferences, and I hope we remain in touch and have the opportunity to work together in the future.
Many other staff and faculty members at SPA have played important roles on my journey. Thank you so much to Dawn Savage for all of your help as my academic advisor throughout my time at SPA. I can always count on you for answers to all of my questions and for fun conversations about chickens and other things. Thank you also to the rest of the staff for your friendliness and help along the way. In addition to my dissertation committee, I have been the beneficiary of much support from other faculty members. Thank you to Mary Guy for your good advice and guidance over the years and for writing opportunities, and thank you to Sebawit Bishu, Christine Martell, John Ronquillo, Will Swann, Sandy Zook, and so many others for advice, support, opportunities to work together, and much more. I will miss you all!
Thank you to my partner, Cheryl. I cannot thank you enough for your support and patience as I have gone through this crazy PhD process. I know I can always count on you for encouragement, lightness, optimism, and unconditional love. I cannot wait for our life together in Hawaii.
Vll


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION............................................................. 1
Nonprofit Organizations and Policymaking..................................2
Overview of Nonprofit Advocacy............................................9
Environmental Nonprofits.................................................11
Overview of Study Context and Methods....................................16
Overview of Remaining Chapters...........................................18
II. ADVOCACY GROUPS AND VENUE SELECTION.....................................21
Overview of Venues in the Policy Literature..............................22
Venue Selection..........................................................25
Advocacy Organizations...................................................28
Study Context: Oil and Gas Policy Debates in Colorado....................30
Methods..................................................................32
Results..................................................................35
Discussion...............................................................52
Conclusion...............................................................55
III. HOW DO ENVIRONMENTAL NONPROFITS ADVOCATE?..............................57
Literature Review........................................................58
Study Context: Oil and Gas Development...................................66
viii


Methods..............................................................68
Results..............................................................69
Discussion and Conclusion............................................88
IV. A QUALITATIVE COMPARATIVE ANALYSIS..................................93
Literature Review....................................................94
Research Context....................................................102
Methods..............................................................103
Advocacy Activities of Environmental Nonprofits......................105
Organizational Characteristics Associated with Advocacy..............107
Discussion...........................................................116
V. CONCLUSION......................................................... 121
Major Takeaways.....................................................123
Study Limitations...................................................128
Future Research Agenda...............................................131
REFERENCES............................................................. 136
APPENDIX
A. SUMMARY OF DISSERTATION CHAPTERS................................... 147
B. POLICY ISSUES IN NEWS ARTICLES..................................... 148
C. ACTIVITY BY JURISDICTION RELATED TO LOCAL AUTHORITY................ 149
D. INTERVIEW PROTOCOL................................................. 152
IX


E. INTERVIEW CODE FORMS, PARTICIPATION BY VENUE AND STATE... 153
F. QCA RESULTS.............................................. 178
x


LIST OF TABLES
TABLE
1. Venue Frequency in Articles........................................................36
2. Policy Issue by Level of Government...............................................39
3. Articles and Advocacy Group Activity by Year......................................40
4. Venues with Mean Participation by Advocacy Group (T-Test Results).................41
5. Policy Issues with Mean Participation by Advocacy Group (T-Test Results)..........42
6. Comparison of State Characteristics................................................68
7. Advocacy Strategy Overview.........................................................70
8. Regulatory Agency Strategies.......................................................73
9. Legislative Strategies.............................................................75
10. Judicial Strategies...............................................................77
11. Ballot Box Strategies.............................................................78
12. Governor’s Office Strategies......................................................80
13. Participation in Advocacy Activities............................................ 106
14. Variable Descriptive Statistics and QCA Calibration..............................Ill
15. Providing Information to Government Officials, Truth Table.......................113
16. Providing Information to Government Officials, Intermediate Solution............. 114
17. Collaborating with People You Disagree with, Truth Table.........................115
18. Collaborating with People You Disagree with, Intermediate Solution...............115
xi


LIST OF FIGURES
FIGURE
1. Venue Frequency in Articles.....................................................36
2. Timeline of Bans, Moratoria, and Local Authority Issues.........................44
3. Activity Across Venues by Year..................................................47
4. Activity Across Venues by Advocacy Group........................................49
5. Local Authority Related Activity at the State Legislature........................50
6. Local Authority Related Activity at the State Ballot Box.........................51
7. Local Authority Related Activity at the Local Ballot Box.........................51
8. Local Authority Related Activity at State Courts.................................52
xii


CHAPTERI
INTRODUCTION
This dissertation explores the advocacy activities of environmental nonprofits with an eye toward enhancing knowledge regarding how these organizations advocate and how advocacy varies by organization and venue. While a number of literatures—such as the policy process, social movement, and nonprofit management literatures—contemplate that nonprofits play a role in environmental policymaking, the literature lacks a thorough understanding of who these organizations are and how they try to influence policy. For example, policy process and social movement literatures contemplate that both formal and informal organizations and groups play a role in environmental policymaking, yet these studies do not examine nonprofit organizations in particular. Studies specific to nonprofit organizations may have implications for nonprofit staff, board members, and donors; however, despite recognizing that environmental nonprofits advocate more than any other type of nonprofit, the nonprofit literature lacks an understanding of these organizations as distinct from other types of nonprofits.
With these considerations in mind, this dissertation includes three empirical chapters.
The first empirical chapter (Chapter 2) relies on news media to explore differences and similarities between industry and environmental advocacy groups with respect to the venues and issues in which they engage. The second empirical chapter in the dissertation (Chapter 3) zooms in on environmental nonprofits and explores, through interviews, the types of advocacy activities these organizations use across sectors and how the external contexts and internal characteristics of these organizations may influence their advocacy. Finally, the third empirical chapter in the dissertation (Chapter 4) examines differences between 501(c)(3) and 501(c)(4) organizations in the environmental space using survey responses combined with publicly available nonprofit data.
1


While in service delivery sectors, 501(c)(3) organizations tend to be service providers and 501(c)(4)s are advocacy organizations, Chapter 4 examines whether there are differences between 501(c)(3) and 501(c)(4) environmental advocacy organizations. The final chapter of the dissertation synthesizes lessons learned and describes a research agenda for further studies about environmental nonprofit advocacy.
Before moving to the empirical chapters of the dissertation, this first chapter lays the general foundation for the studies that follow. This chapter begins by describing that nonprofits are actively involved in policymaking and in doing so, can play an important role in democracy.
I then describe existing literature on nonprofit advocacy and provide rationales for studying environmental nonprofits in particular. The chapter then presents an overview of the dissertation’s study context and methods, before concluding with a short summary of each of the remaining chapters.
Nonprofit Organizations and Policymaking
Before attempting to define and operationalize advocacy based on existing literature, this section describes why nonprofit advocacy is a worthy area of scholarship. Nonprofit organizations play roles in society both as advocates and as service providers. As service providers, there is no question that government reliance on nonprofits for service delivery has increased dramatically in recent years (Boris 2006; Never 2016). For example, beginning in the 1980s and early 1990s with the New Public Management (NPM) movement’s emphasis on contracting out and Osborne and Gaebler’s (1992) prescription that government should steer rather than row, there has been a proliferation of government contracting with the nonprofit sector for services. In addition to their role as service providers, these organizations also often advocate for themselves and their constituents (Hula and Jackson-Elmoore 2001; Minkoff 2002;
2


Fyall 2017). The vast majority of nonprofits register for tax-exempt status pursuant to section 501(c)(3) of the Internal Revenue Service (IRS) code. Section 501(c)(3) organizations are public charities and may only lobby up to a point and are prohibited from engaging in electoral politics. These organizations, however, often advocate via public education, working in coalitions, and limited amounts of lobbying. In addition, a smaller but still significant portion of nonprofits focuses exclusively on advocacy and is not restricted in their activities. By one measure, out of 1.54 million registered nonprofits in 2004, 138,193 were 501(c)(4)s or social welfare organizations, and 86,054 were 501(c)(6)s or business leagues, both of which typically focus on advocacy (Boris 2006). Organizations that primarily advocate, including 501(c)(3)s, 501(c)(4)s, and 501(c)(6)s, are the focus of this dissertation for a couple reasons: 1) Nonprofits play a role in policymaking, and 2) nonprofits have the potential to facilitate democracy.
Nonprofits in the Policy Process
Policy process research refers to “the study of the interactions over time between public policy and its surrounding actors, events, and contexts, as well as the policy or policies’ outcomes” (Weible 2014, 5). Scholarship in this area makes clear that nonprofits are typically involved at some level in policy making (Fyall 2017). For example, the Multiple Streams Framework (MSF) describes that an item comes on to a policy agenda when a problem aligns with a policy solution in the right political environment (Herweg, Zahariadis, and Zohlnhofer 2018). These three streams—the problem, policy, and politics streams—run relatively independently, until a policy entrepreneur couples the streams in a policy window, or at an opportune time (Herweg et al. 2018). Within the MSF, nonprofits may be part of the policy stream, and they may act as policy entrepreneurs (Fyall 2017; Kagan 2019). In the policy stream, policy communities consisting primarily of policy experts generate ideas and policy
3


alternatives (Herweg et al. 2018). Nonprofits, and particularly those that advocate, often participate in these communities and help generate policy alternatives that support their missions. Also, as entrepreneurs, nonprofits often advocate on behalf of specific policies to move them on to policy or decision agendas. When policies are consistent with their missions, nonprofits often devote considerable resources to push those policies forward.
Also, the Advocacy Coalition Framework (ACF) focuses on policy subsystems and the activities, beliefs, and interactions of coalitions within those subsystems (Jenkins-Smith et al. 2018). The ACF focuses on the policy subsystem as the “primary unit of analysis for understanding policy processes” (Jenkins-Smith et al. 2018, 139). Nonprofits are among subsystem actors in that they regularly try to influence subsystem affairs (Jenkins-Smith et al. 2018), and nonprofits frequently advocate for change within coalitions (see Fyall and McGuire 2015). According to the ACF, nonprofits might work within coalitions to influence change by taking advantage of events either within or outside the subsystem, promoting policy-oriented learning, and negotiating agreements either as a coalition member or as a policy broker (Jenkins-Smith et al. 2018).
Other policy process frameworks make similar claims. For example in the Narrative Policy Framework (NPF), nonprofits may use narratives to influence policy or may be actors within narratives (Shanahan, Jones, McBeth, and Radaelli 2018). In the Institutional Analysis and Development Framework (IAD), nonprofits may be actors in action situations or they may provide the institutional frameworks within which action situations occur (Schlager and Cox 2018; Wilson, Ostrom, and Cox 2013). Also, in Punctuated Equilibrium Theory (PET) nonprofits may effect policy change by influencing issue definition, the images associated with issues, and the venues in which issues are debated and considered (Baumgartner et al. 2018).
4


Relatedly, the policy process literature also recognizes that multiple venues exist within a given subsystem (Lubell 2013; Jenkins-Smith et al. 2018). Part of nonprofits’ overarching strategies to influence policy involves venue selection (Pralle 2006), or choosing to participate in venues where organizations believe they are most likely to achieve their policy goals. Thus, not only does the policy process literature contemplate that nonprofits use a variety of advocacy activities to influence policy, but also that they do so across a range of venues.
While the role and impact of nonprofit organization involvement varies from case to case, it is clear that nonprofits actively engage across a range of topics. Even among 501(c)(3)s or charitable organizations, we see organizations engaging in advocacy related to arts and culture, the environment, health issues, civil rights, and more (Berry and Arons 2003; Urban Institute 2015). As such, nonprofit advocacy is an important area of inquiry.
Nonprofits and Democracy
In considering the role that nonprofits play in policymaking, it is also necessary to bear in mind that when nonprofits advocate, they do so not only to further their missions, but also to provide a collective voice for dispersed or underrepresented interests. That is, nonprofits often fill a void in policymaking by supporting citizen engagement, which is a key concern in the field of public affairs. Policy process and public administration fields are framed by, or operate within a paradigm of, democracy and involve questions regarding how to incorporate public preferences and represent citizen interests (deLeon 2007). Indeed, “[cjitizen participation is the cornerstone of democracy” (Roberts 2004), and according to Dahl (2015), effective participation is one of five criteria for democracy with the other four being voting equality, enlightened understanding, control of the agenda, and inclusion of all adults. Thus, democratic institutions
5


must be concerned with how to ensure public access to decision-making venues and processes, and how to incorporate meaningfully public preferences in policy and administration.
In considering these concerns and the potential role that nonprofits play in facilitating democracy, a distinction to bear in mind is that between representative democracy and direct democracy (Box 2009). While many extoll the virtues of direct democracy (e.g., Follett 1918; Roberts 2004), inviting direct participation from an entire polity is often impractical, particularly when meaningful participation is desired. Direct democracy may be easier in smaller units of government, but smaller units are generally less equipped to handle large-scale problems, such as environmental issues. That is, the expansive, transboundary nature of many environmental issues necessitates the involvement of larger democratic units, such as state, national, and even global bodies.
Within these larger units, representative democracy is often invoked. Representative systems, however, also often fall short of facilitating adequate representation (Dahl 2015). First, representatives may have more constituents than they can reasonably connect with in a meaningful way. Second, larger units may be plagued by issues of inequality, which can undermine democracy in a number of ways. According to Shapiro (2015), electoral politics are heavily influenced by money; those with limited resources are less likely to participate; and, even if those with limited resources do participate, the complexity of political systems makes stasis the norm and change difficult to effect.
Nonprofit organizations help mitigate these issues by representing citizen voices and serving as intermediaries between the public and government (see LeRoux 2007, 2009). Indeed, one of the primary purposes of nonprofits is to provide a voice for those whose interests might not otherwise be represented (Child and Gronbjerg 2007; Almog-Bar and Schmid 2014; Mosley
6


2016). In the 1830s, Alexis de Tocqueville, observing the US, noted that associations are essential in democratic societies to overcome the fact that independent citizens otherwise lack political power (de Tocqueville 1966). More recently, according to Putnam (1993), voluntary associations help build social capital, and nonprofits help secure citizen rights, including political, civil, and social rights (see also Smith 2007). According to Child and Gronbjerg (2007), nonprofits are the “primary vehicle” through which individuals come together to engage in advocacy activities and pressure governments to take some desired action (260; see also LeRoux 2007). This is particularly true with respect to environmental issues like climate change, which is an intractable problem, involving interdependent networks and requiring tremendous resources (O’Leary 2015). These complexities make nonprofits all the more relevant, because 1) they may provide an accessible forum in which members of the public can participate, and 2) they can help restore equilibrium when institutions become unbalanced, for example, when an external shock causes an agency to shift resources away from an essential service or issue (see Truman 1951).
A number of scholars have outlined the specific ways in which nonprofits contribute to democracy. In perhaps the most comprehensive description, Fung (2003) describes six ways that associations, including nonprofits, contribute to democracy.1 First, Fung (2003) argues that associations are inherently good in that they support freedom and democracy. Second,
1 While the concept of association is not well-defined, Warren (2001) describes that references to associations “ordinarily mean those kinds of attachments we choose for specific purposes—to further a cause, form a family, play a sport, work through a problem of identity or meaning, get ahead in a career, or resolve a neighborhood problem” (39). Nonprofit organizations are typically conceptualized as a subset of associations (see Dodge and Ospina 2016), and while the purpose here is not to argue that nonprofits necessarily perform all of the functions listed below, the overlap between nonprofits and associations suggests that the ways in which associations may contribute to democracy are also ways in which nonprofits may similarly contribute.
7


associations help develop the attitudes, skills, and behaviors of their members or broader citizen groups that support civic engagement. Third, associations can serve a resistance and power checking function by counterbalancing government or other special interests when necessary. Fourth, associations provide an additional means for citizen interests to be represented in government. Fifth, associations create space for and facilitate public deliberation. Sixth, and finally, associations are increasingly engaged in direct governance, such as by participating in policymaking and service provision. Other scholars have suggested similar ways that nonprofits contribute to democracy, such as by promoting direct citizen involvement in government through education, capacity building, and issue framing; providing a collective voice for otherwise underrepresented interests; and resisting or providing alternative political venues when formal institutions break down (Warren 2003; LeRoux 2009; Dodge and Ospina 2016).
As the above suggests, nonprofits certainly have the potential to improve opportunities for participation and democracy. At the same time, though, nonprofits can also undermine democracy (see Reid 2006). For example, rather than strengthen the voice of those lacking political power, many nonprofits actually amplify the voice of the wealthy and powerful. While nonprofits may serve to aggregate voices, the most disadvantaged communities may not have access to nonprofits (Boris 2006). As described more fully below, this is among the concerns that gave rise to the environmental justice movement. For example, nonprofits can disproportionately influence public policy at times, and it is not always clear who is funding nonprofit efforts (Boris 2006). Furthermore, financial stress and budget pressures may lead organizations to represent funders rather than public interests (Reid 2006).
For these reasons, developing our understanding of environmental nonprofits, including how they try to influence policy and how contextual and organizational factors may influence
8


policy, may aid our understanding of the democratic role that environmental nonprofits play.
This dissertation lays the foundation for this line of research in the environmental nonprofit and environmental policy space.
Overview of Nonprofit Advocacy
With an understanding of the relevance of nonprofit advocacy, this section describes specifically what nonprofit advocacy refers to and how scholars to date have attempted to operationalize the concept. While studies about the advocacy activities2 of nonprofit organizations are not new, perhaps recognizing the growing importance of nonprofit advocacy to counterbalance well-resourced corporate interests, recent years have seen a surge in nonprofit advocacy research. For example, recent scholarly work has focused on the advocacy activities of human service nonprofits and the drivers of those activities (Mosley 2010; Almog-Bar and Schmid 2014), how nonprofits influence policy and effect policy change (Hula and Jackson-Elmoore 2001; Andrews and Edwards 2004; Bryce 2006; Fyall and McGuire 2015; Fyall 2016), and ways of understanding and classifying nonprofit advocacy organizations and their activities (Boris and Mosher-Williams 1998; Child and Gronbjerg 2007).
In the nonprofit literature, advocacy is defined as “the attempt to influence public policy, either directly or indirectly” (Pekkanen and Smith 2014a, 3). Beyond agreement on the definition, though, advocacy as a concept remains difficult to operationalize. For example, in their definition of advocacy, Pekkanen and Smith (2014a) include lobbying, policy advocacy, protesting, and public education. Nicholson-Crotty (2009), on the other hand, describes three common groups of activities, including affiliating with a 501(c)(4) organization, grassroots
2 Unless otherwise noted in the context, this dissertation uses the terms advocacy activities and advocacy strategy interchangeably to refer to specific activities, such as lobbying, testifying, and mobilizing the public, that organizations use to try to influence the policy process.
9


activities, and direct lobbying. A few other typologies exist as well, as described more fully in Chapter 3 (e.g., Berry and Arons 2003; Casey 2011), but there is little consistency between the typologies, and many studies reduce advocacy to a dichotomous variable reflecting whether organizations engage in any type of advocacy. Furthermore, few nonprofits studies to date have incorporated differences in venues into their understandings of advocacy activities.
Despite these issues, though, knowledge regarding nonprofit advocacy has certainly increased over the past couple decades. For example, in a study of 501(c)(3) organizations in Indiana, Child and Gronbjerg (2007) demonstrated that at least some nonprofits across fields advocate, and that environmental and health organizations are more likely than other nonprofits to do so. Berry and Arons (2003) found similar trends looking at a random sample of 501(c)(3) organizations across the US. These authors also demonstrated among the same group of nonprofits, that organizations are most likely (among a list of activities) to provide information to government officials, work in planning and advisory groups, and meet with government officials. They are least likely to testify at legislative or administrative hearings (see also Bass 2007).
Scholars have also begun to elucidate the motivations and factors associated with advocacy. Berry and Arons (2003) found that primary reasons for nonprofit advocacy include protecting or promoting programs that benefit constituencies, and raising awareness about issues. Furthermore, primarily in reliance on resource dependency theory and institutional theory, scholars have identified a number of factors associated with advocacy. Resource dependency theory suggests that organizations require a variety of resources to survive, and they therefore interact strategically in their environments to secure these resources (Eikenberry and Kluver 2004; see also Pfeffer and Salancik 1978). Resource dependency theory consistently suggests
10


that organizational budgets, staff size, and volunteer resources are all associated with higher likelihoods of advocating (e.g., Mosley 2011). According to institutional theory, in order to survive, organizations must conform to the rules, norms, and social expectations that make up their institutional environments (Guo and Acar 2005; see also Meyer and Rowan 1977). Based on institutional theory, we know that factors like institutionalization, professionalization, collaboration, and political environment are associated with advocacy (Mosley 2011; Nicholson -Crotty 2009).
Finally, recognizing the challenges associated with advocacy, scholars have identified major barriers to advocacy. The most significant of these is limited capacity (Bass 2007; Mosley 2014). With respect to capacity issues, organizations, particularly social service providers, often lack the time and money needed to participate effectively in advocacy. Similarly, 501(c)(3) organizations often lack staff with advocacy expertise, making it more difficult for them to engage effectively (Bass 2007; Mosley 2014). Furthermore, while scholars often cite confusion about tax laws as a challenge with nonprofit advocacy (Bass 2007; Berry and Arons 2003), Mosley (2014) found that while organizations are often unclear about what they may legally do, this does not actually prevent them from advocating.
Thus, while scholarship is certainly growing regarding nonprofit advocacy, the majority of studies are limited to 501(c)(3) organizations or social service providers. Thus, the next section describes the rationales for expanding studies of nonprofit advocacy to the environmental field.
Environmental Nonprofits
This dissertation focuses specifically on environmental nonprofits for a few reasons. As described more fully below, environmental nonprofit organizations as a specific group are
11


understudied across literatures. Also, environmental nonprofits have unique implications for society in that they can facilitate stakeholder participation in environmental policymaking, yet they also have a history of promoting social inequities. The rationales for studying environmental nonprofits as a unique subgroup of nonprofits are each discussed in turn.
First, environmental nonprofits are understudied. A limitation of existing studies on nonprofit advocacy is that they typically either use broad strokes to describe advocacy by all types of nonprofits (e.g., Berry and Arons 2003; Pekkanen and Smith 2014b), or they focus specifically on social service providers (e.g., Almog-Bar and Schmid 2014; Fyall 2016). There are certainly good reasons to focus on social service providers. For example, health and human service nonprofits are far more prevalent than other types of nonprofits (Berry and Arons 2003). Also, as Mosley (2014) describes in a study of homeless service providers, these nonprofits provide a critical voice for marginalized communities, including resource constraints, lack of experience with advocacy, and confusion about the legal limits of 501(c)(3) advocacy. Studies exploring the dual role of advocate and provider (see Minkoff 2002; Fyall 2017) among social service organizations remain worthy of inquiry, and through expanding studies of nonprofit advocacy into other substantive areas, including environmental policy contexts, we may learn more about how advocacy occurs in specific contexts as distinct from other contexts. While a few studies to date have differentiated among organizations based on their policy contexts (e.g., Berry and Arons 2003; Child and Gronbjerg 2007), those that do focus on rates of advocacy, rather than on the specific ways that organizations advocate or the specific constraints that affect different types of nonprofits.
The dearth of studies regarding environmental nonprofit advocacy is particularly surprising given the fact that these organizations are more likely than any other type of nonprofit
12


to engage in advocacy (Child and Gronbjerg 2007). These organizations are also different from other types of nonprofits in that they typically lack a clear constituent base, provide nonexcludable benefits, and are often directly opposed by corporations with very deep pockets.
Thus, it is necessary to explore the advocacy activities of environmental nonprofits as distinct from other types of nonprofits, and such an exploration also allows for a deeper examination of nonprofit advocacy activities broadly. That is, while studies to date typically look at whether nonprofits advocate, through examining a population of nonprofits that advocate, we can gain a better understanding of different types of activities and the factors that affect organizational decisions about which activities to use to influence policy.
Furthermore, the importance of environmental policy cannot be overstated. With abundant and significant global issues, including air pollution, ocean pollution, water contamination, and climate change and its effects, environmental policymaking is among the most critical tasks for today’s leaders. Furthermore, environmental issues are considered wicked problems in that they are unstructured, cross-cutting, and relentless, and thus cannot be addressed through traditional problem-solving systems (Weber and Khademian 2008). In response to these issues, there is no question that the environmental movement has grown exponentially by every conceivable measure since WWII (Baumgartner and Jones 2009; see also Boris 2006). While the environmental movement is certainly studied in other literatures—for example, nonprofit organizations may be part of social movements (Reid 2006)—examining environmental nonprofits in particular holds value for nonprofit stakeholders. That is, understanding environmental nonprofits will help inform their staff members about best practices, opportunities, and constraints; help donors who give money and place trust in these organizations
13


better understand the sector; and inform society at large, which confers benefits on these organizations such as through tax exemption, about the benefits and drawbacks to the sector.
The second rationale for studying environmental nonprofits as a distinct group is that stakeholder participation in environmental policymaking and governance is particularly essential because of the shared resources and personal interests at stake (Bingham, Nabatchi, and O’Leary 2005; Durant, Fiorino, and O’Leary 2017). In this context, participatory processes can improve outcomes, empower communities, reduce uncertainties and unexpected resistance, and enhance the knowledge available for assessing and solving environmental problems (Irvin and Stansbury 2004; Pahl-Wostl 2009). For these reasons and because of difficulties with governing at the global level, recent years have seen a trend toward decentralization and increased emphases on participation (Lemos and Agrawal 2006). This trend is facilitated further by empirical evidence that self-governance and co-management are viable solutions to environmental degradation (Ostrom 1990; Lemos and Agrawal 2006). Notably, the work of Elinor Ostrom and others suggest that communities can effectively manage and conserve their resources when certain conditions are present, including that those affected by the rules governing a resource can participate in modifying them (Cox, Arnold, and Tomas 2010). When small resources like village fisheries are at stake, individual participation may be possible, but when larger transboundary resources are involved, stakeholders are aggregated to the agency or organizational level (see Heikkila and Gerlak 2005). Thus, these agencies or organizations, which often include nonprofits, should appropriately represent constituent interests.
In addition to scholarship on public participation in environmental governance, policies across all levels of government emphasize the need for public participation in environmental policymaking. For example, Principle 10 of the Rio Declaration on Environment and
14


Development, drafted at the 1992 United Nations Conference on Environment and Development, states the following:
Environmental issues are best handled with the participation of all concerned citizens, at the relevant level. At the national level, each individual shall have appropriate access to information concerning the environment that is held by public authorities, including information on hazardous materials and activities in their communities, and the opportunity to participate in decision-making processes (UN 1992).
In the US, among other policies, the National Environmental Policy Act (NEPA) requires federal agencies undertaking projects that significantly affect the environment to provide meaningful opportunities for public participation. Similar initiatives to promote citizen engagement in state and local policymaking exist as well. Thus, understanding and developing environmental nonprofit capacities and strategies can help further objectives related to stakeholder engagement.
Third, environmental justice (EJ) concerns demand that minority and otherwise underrepresented groups be given a voice in environmental policymaking. According to the Environmental Protection Agency (EPA), EJ refers to the “fair treatment and meaningful involvement of all people regardless of race, color, national origin, or income, with respect to the development, implementation, and enforcement of environmental laws, regulations, and policies” (EPA 2017). As the definition suggests, the EJ movement is concerned at least as much with processes as it is with outcomes, emphasizing democratic decision-making, public participation, and community self-determination (Cole and Foster 2001).
The EJ movement began in the late 1970s and early 1980s in response to concerns about discriminatory government and industry policies, and perceptions that existing environmental nonprofits did not represent poor and minority interests (Bullard et al. 2008). Although 40 years have passed, there is no question that environmental injustices persist today. Stories of environmental injustices often grab headlines, as in the cases of the Flint, Michigan water crisis
15


and the Dakota Access Pipeline, and studies consistently show that racial minority and low-income communities still disproportionately suffer from environmental risks (e.g., Bullard et al. 2008; NAACP 2017). Studies to date tend to focus on the substantive side of EJ by focusing on the relationship between community demographics and environmental injustices, and on the drivers of those injustices (Campbell, Kim, and Eckerd 2014). Government policies, however, often focus on the procedural side of EJ and demand that government programs and agencies incorporate EJ and community input (e.g., EPA 2011, 2014). Thus, the literature would benefit from studies about the role that environmental nonprofits play, both in influencing policy and in enhancing the voice of underserved communities.
Overview of Study Context and Methods
To explore the role of nonprofits in environmental policymaking, this study examines oil and gas, or hydraulic fracturing (“fracking”), policy. Fracking is a technique involving the injection of water, sand, and chemical additives into shale rock or other porous formations thousands of feet beneath the earth’s surface to release oil or natural gas (Heikkila et al. 2014).
In the US and around much of the world, fracking is increasingly used in oil and gas development. For example, between 2007 and 2017, the amount of natural gas produced from shale formations increased 10 fold from 1,990,145 million cubic feet to 19,018,457 million cubic feet (EIA 2018).
Not only has the volume of oil and gas produced through fracking increased, but oil and gas development is also encroaching on high population density areas. Close proximity between population centers and oil and gas operations gives rise to a host of local issues, including NIMBYism (not in my backyard); nuisance issues, like traffic, dust, and noise; and concerns about air and water contamination. As a result, communities grapple with how to manage
16


fracking, and some have attempted to ban it. For example, while the state of New York successfully banned fracking, local bans in a number of municipalities like Denton, Texas, and Longmont, Colorado, have been struck down by courts or legislatures (Knight and Gullman 2015; Heikkila and Weible 2016; Jaquith 2017). Local communities with limited authority over oil and gas development may attempt to exercise control by negotiating agreements with oil and gas companies (Zilliox and Smith 2017), regulating nuisance issues, and advocating for increased setbacks and other limits at the state level. Beyond local issues, fracking is part of broader national and global debates about reliance on fossil fuels as energy sources. While those in favor of fracking often cite natural gas as a cleaner alternative to oil and coal, those opposed claim that fracking and the availability of cheap oil and natural gas impede transitions to renewable energy sources, like wind and solar.
Because they involve a wide range of issues across venues and levels of government, fracking debates provide a good context for this study. Fracking is an intensely debated political issue (Heikkila and Weible 2016), drawing participation from industry, government, and nonprofit organizations. Fracking policies are created via judicial decisions, legislation, agency rules, executive orders, and ballot measures, and organizations, including environmental nonprofits, engage in a variety of activities and strategies to influence each of the sources of policy. As such, fracking is an ideal policy context for observing maximum potential advocacy activities.
To examine how nonprofits interact in hydraulic fracturing policy, this dissertation relies on a diverse array of methods, which are explained more fully in each of the subsequent chapters. Data sources include newspaper articles, interviews, and surveys of environmental nonprofits active in oil and gas and related policy. Analyses rely on descriptive and analytic
17


statistics, thematic analysis and constant comparative methods, and qualitative comparative
analysis.
Overview of Remaining Chapters
The final section of this chapter offers a brief overview of the remaining chapters of this dissertation. Each of the three empirical chapters was written to be a standalone publication. As a result, the reader will find some overlap in the literature and study context between chapters. For each chapter, a short summary of the substance and proposed methods is provided. An overview of Chapters 2 through 4 is also provided in Appendix A.
Advocacy Groups and Venue Selection
Chapter 2 presents an analysis of the advocacy groups and venues active in hydraulic fracturing policy in Colorado over a five-year period. Venues are an essential concept in literature on advocacy in that they may both frame the activities that organizations use to influence policy and venue selection is a strategy in and of itself (Pralle 2006). For example, when actors are frustrated with a given decision making body, they may look for an alternative venue. While existing venue shopping literature identifies both venue- and organization-related factors that influence venue selection, the literature lacks an analysis of how organizations can be both proactive and reactive in choosing venues. By examining both industry and environmental group involvement across venues in a particular policy subsystem, this paper demonstrates that venue selection is an ongoing process, influenced by a number of competing factors.
Data for this chapter were derived from the news media. Specifically, data include over 500 articles from two Colorado newspapers published from 2013 to 2017. Articles were coded in NVivo using constant comparative method to identify policy issues, venues, and advocacy groups. Analysis is based on both descriptive and analytic statistics. Results shed light on how
18


industry and environmental advocacy groups differ in their approaches to advocacy, and how activities across venues evolve over time.
How Do Environmental Nonprofits Advocate?
Building on the second chapter and using venues as an overarching framework, Chapter 3 explores the variety of activities that environmental nonprofits use to influence policy. Existing literature describes that organizations use both inside and outside strategies to influence policy. This study expands on existing literature by describing the array of inside, outside, and hybrid strategies that environmental nonprofits use across different venues and how strategies differ by venue. Furthermore, while literature suggests that inside strategies are more common and effective, and that more professionalized organizations tend to use inside strategies, in line with recent research, this study suggests that the relationship between strategies and organizations is more complex. Thus, this study also identifies a few key organizational and contextual factors that may be influencing advocacy decisions.
This chapter relies on semi-structured interviews with executive directors or other senior staff of environmental nonprofits across three states: Colorado, New Mexico, and Oklahoma. While each of these states has significant oil and gas development and ongoing policy debates, these states also have differences in their political contexts and the types of land on which oil and gas development takes place. Interviews were analyzed using thematic analysis. Results demonstrate the types of advocacy activities that groups use across venues; illuminate the similarities and differences between venues; and suggest some of the major factors that influence organizations’ decision about advocacy.
19


Qualitative Comparative Analysis of Advocacy and IRS Designation
Chapter 4 presents an additional look at the advocacy activities of environmental nonprofits, with a particular eye toward differences between 501(c)(3) and 501(c)(4) organizations. Among social service providers, 501(c)(3) organizations are typically service providers and 501(c)(4)s are advocacy organizations. In the environmental space, however, both 501(c)(3) and 501(c)(4) organizations are often advocacy organizations. This study asks two research questions. First, what activities do environmental nonprofits engage in to influence the policy process? Second, which organizational characteristics are associated with advocacy among environmental nonprofits? This study draws again on the advocacy literature and uses Resource Dependency Theory and Institutional Theory to derive theoretical propositions about the relationship between advocacy activities and organizational characteristics.
Chapter 4 relies on survey data from 30 environmental nonprofits about the types of activities they use to further their policy goals and whether they report using these activities effectively. After presenting the results of descriptive statistics, the primary analysis for this chapter uses QCA to examine the organizational factors typically associated with different types of advocacy. Results indicate that organizations report engaging in advocacy at high rates and report doing so effectively. Furthermore, 501(c)(3) and 501(c)(4) organizations report similar trends in advocacy, except that 501(c)(4) organizations are not associated with advocacy that involves collaborating with opponents.
Concluding Chapter
Finally, this dissertation concludes with a discussion of major takeaways and some of the overarching limitations. The concluding chapter also describes a research agenda that builds on the foundations laid in this dissertation.
20


CHAPTER II
ADVOCACY GROUPS AND VENUE SELECTION
Abstract
This chapter presents an analysis of the advocacy groups and venues active in hydraulic fracturing policy in Colorado over a five-year period. Venues are an essential concept in literature on advocacy in that they may both frame the activities that organizations use to influence policy and venue selection is a strategy in and of itself. While existing venue shopping literature identifies both venue- and organization-related factors that influence venue selection, the literature lacks an analysis of how organizations can be both proactive and reactive in choosing venues. By examining both industry and environmental group involvement across venues in a particular policy subsystem, this paper demonstrates that venue selection is an ongoing process, influenced by a number of competing factors. Data for this chapter are drawn from over 500 news articles published in two Colorado newspapers between 2013 and 2017. Articles were coded in NVivo using constant comparative method to identify policy issues, venues, and advocacy groups. Analysis is based on both descriptive and analytic statistics. Results shed light on how industry and environmental advocacy groups differ in their approaches to advocacy, and how activities across venues evolve over time.
Keywords
Venue selection, advocacy groups, environmental policy
21


This chapter presents an analysis of how advocacy groups interact across venues in an oil and gas policy subsystem across a five-year period. To examine the choices that different groups make with respect to venues, this chapter divides advocacy groups into two categories: industry advocacy groups and environmental advocacy groups. There is certainly variation across these groups both in terms of their level of formality or institutionalization and in terms of where they fall on the spectrum ranging from “keep it in the ground” to “drill, baby, drill.” The fact remains, though, that these groups typically find themselves on opposite ends of the negotiating table. Thus, by examining these groups’ involvement across venues, patterns may be discerned in venue selection and policy subsystem evolution.
This chapter asks two primary research questions: 1) What are the different types of venues and policy issues that advocacy groups are involved in within the Colorado oil and gas policy subsystem? 2) Are there differences in how industry advocacy groups and environmental advocacy groups participate across venues? After describing literature on venues, policy subsystems, venue selection, and advocacy groups, the study context and use of news media are discussed. I then present the results of the study along with a discussion of some of their potential implications, particularly for the venue selection literature.
Overview of Venues in the Policy Literature Exploring the venues in which advocacy occurs provides a logical starting place for understanding advocacy. “Policy venues are the institutional locations where authoritative decisions are made concerning a given issue” (Baumgartner and Jones 2009, 32). Venues can be divided both vertically, such as by level of government, and horizontally, such as by branch of government (Holyoke, Brown, and Henig 2012; Buffardi, Pekkanen, and Smith 2015), and as described more fully below, activity in policy subsystems often occurs across numerous venues.
22


Venues not only provide the rules and structures within which advocacy occurs, but organizations may also manipulate venues to achieve their preferred outcomes.
The importance of venues derives from the federalist system and separation of powers that characterize governance in the US. Federalism refers to a division of power between multiple levels of government, and separation of powers refers to our tripartite system of government, designed to prevent any one branch of government from dominating. Together, these features result in a polycentric3 policy system with many centers of decision making that are formally independent of one another (V. Ostrom 1999). Such a system also provides opportunities for multiple decision-making venues to play a role in governing any given policy issue. While historically polycentric systems of governance were thought to stymy real change because of the multiple avenues available for blocking change, more recently, scholars have argued that polycentric systems of governance support change in that they offer multiple decision-making venues and opportunities to shift policy (Baumgartner and Jones 2009; Pralle 2003).
Policy research finds that policy subsystems are comprised of numerous venues and subissues, and that when people and organizations engage in the policy process, they often engage in multiple venues. For example, in the Advocacy Coalition Framework (ACF), the primary unit of analysis is the policy subsystem (Jenkins-Smith et al. 2018). A subsystem is defined by a policy topic, a territorial scope, and the actors involved. Within a given subsystem, coalitions, or groups of actors who share policy core beliefs and coordinate actions, compete across different
3 The term polycentricity was originally used by Michael Polanyi in the context of scientific communities (V. Ostrom 1999). Generally speaking, a polycentric organization is one where “many independent elements are capable of mutual adjustment for ordering their relationships with one another within a general system of rules” (V. Ostrom 1999, 73). V. Ostrom, Tiebout, and Warren (1961) first applied the concept to government.
23


venues over extended periods of time to influence policy change. Also, Ecology of Games Framework contemplates that “governance involves multiple policy games operating simultaneously within a geographically defined policy arena, where a policy game consists of a set of policy actors participating in a rule-governed collective decision making process called a ‘policy institution’” (Lubell 2013, 538). Policy institutions consist of both formal and informal collective choice rules, governing how decisions are made in a given policy institution or venue (Lubell 2013; Ostrom 1990). Thus, the Ecology of Games Framework contemplates that games across multiple venues may occur simultaneously. Recent research also suggests that policy conflicts occur across three polycentric levels of action: political systems, policy subsystems, and policy action situations (Heikkila and Weible 2017; Weible and Heikkila 2017). The political system is the overarching system that governs a territory and has authority over a wide-range of issues. The policy subsystem is as defined above by the ACF. Finally, policy action situations refer to “the diverse arenas within political systems and policy subsystems that include formal and informal venues where policy actors engage, debate, and attempt to address problems around policy issues” (Weible and Heikkila 2017, 26).
Given the availability of multiple venues, venue selection presents a key strategic choice for those interested in influencing policy. When multiple venues have authority over a single given issue, actors may resort to venue shopping, which “refers to the activities of advocacy groups who seek out a decision setting where they can air their grievances with current policy and present alternative policy proposals” (Pralle 2006, 26). Interest in venue shopping originated with Schattschneider (1960), who described that losing groups may increase their chances of success by expanding policy conflicts. More recently, interest in venue shopping traces to Baumgartner and Jones (2009), who described the importance of policy images and venues to
24


policy stability and change. Policy images refer to how the public understands a policy problem (Baumgartner and Jones 2009, 25). Due to bounded rationality and limited attention, policy makers and members of the public typically lack the capacity to hold detailed understandings of policy issues, and therefore rely on heuristics like policy images. Specific policy images resonate differently with different venues and decision makers, which means that as images shift, venues may shift as well.
Venue Selection
At the heart of the venue shopping literature is the idea that actors make strategic choices about venues to maximize their likelihood of obtaining their desired outcome (Ley and Weber 2015). Thus, when actors are frustrated with a given venue and the biases that exist within it, they may look for an alternative venue (Pralle 2003). Sometimes venue shopping results in issue expansion, consistent with Schattschneider (1960), while at other times, the strategy is simply to shift the conflict to a new, more favorable venue (Baumgartner and Jones 2009; Pralle 2006). Furthermore, while venue shopping is sometimes considered a type of strategy among a host of other types of strategies (e.g., Weible and Heikkila 2017), it might also be considered a preliminary choice that then structures and frames the subsequent advocacy activities that groups employ (e.g., Crow et al. 2019).
While traditional venue shopping literature posits that actors make carefully reasoned and deliberate decisions regarding venues, more recently scholars recognize that venue shopping is somewhat ad hoc (Pralle 2003). Most notably, Pralle (2003) demonstrates that venue shopping is more experimental and less deliberate than commonly perceived. Recent studies have elucidated the various constraints that actors face when choosing among competing venues. These constraints can be divided into factors related to venues and factors related to organizations.
25


First, with respect to venue related factors, among the relevant considerations are whether the preferred venue has meaningful authority over the issue at hand, the relationship between venues, ideological alignment between the venue and advocates, the venue’s accessibility to a given group, and the level of control an opponent exerts in a given venue (Holyoke, Brown, and Henig 2012; Ley and Weber 2015). Each of these elements may change over time, giving rise to incentives to shift venues. For example, as noted above, the dominant policy image may change; the venues themselves may change, such as through the election or appointment of new decision makers; and a decision in one venue may influence the authority of another venue (Baumgartner and Jones 2009).
Next, Pralle (2003) argues that venue selection is also influenced by the “inner world of advocacy groups” (255). Most commonly studies highlight the relationship between resource constraints and venue selection (Holyoke, Brown, and Henig 2012). That is, organizations with more financial resources, human resources, and expertise can more easily shift across venues. Furthermore, organizations often have pre-existing preferences based on either their organizational identities or on path dependency (Pralle 2003). With regard to the former, as examples, grassroots organizations may prefer local venues; legal organizations may prefer courts; and anti-establishment organizations may prefer the ballot box. Also, path dependency may lead organizations to become stuck in particular venues, which “is especially likely when groups or individuals develop expertise and skills associated with working in a particular venue” (Pralle 2003, 239). Relatedly, Buffardi, Pekkanen, and Smith (2015) found that nonprofits tend to specialize at certain levels of government, but shop among various venues within each level of government. Finally, venue choice is also influenced by learning. Decisions regarding venue selection are sometimes guided by misinformation or outdated perceptions of venues. Over time,
26


though, sometimes through a trial and error process, actors learn to make better decisions, and learning may occur at the individual, organization, coalition or subsystem level (Pralle 2003; see also Heikkila and Gerlack 2013).
Though scholarship has moved away from assumptions that actors make fully rational decisions about which venues to participate in, the literature continues to assume that actors make proactive decisions about where to place their issues or put their efforts. A much more limited and nascent literature, however, suggests that venue selection may actually be reactive. Ley (2016) distinguishes between activities designed to produce policy change and venue selection by groups with “vested interests.” He then examines how groups seek to insulate their policy victories once achieved. Other related literature describes the importance of veto points to those wishing to block opponents from achieving their goals (Weible, Heikkila, deLeon, and Sabatier 2012). Also, Holyoke, Brown, and Henig (2012) find that actors tend to choose venues in which decision makers are actively working on a given issues, which suggests some form of reactivity.
Despite implicit recognition that venue selection is often reactive and somewhat out of organizations’ control, this side of venue selection remain unexplored in the literature. That is, in high conflict policy subsystems, such as those described by the ACF, PCF, and Ecology of Games Framework, actors and coalition members are constantly engaged in a wide range of policy games or action situations, some of which they choose and others of which are merely a reaction to their opponents’ venues and strategy choices. Thus, this paper attempts to expand on the venue selection literature by exploring how advocacy groups from two competing ideologies participate across venues over a five-year period. After defining and describing advocacy groups, this paper sets up the study context and methods employed.
27


Advocacy Organizations
To explore venues, this study focuses on two competing coalitions4 of advocacy groups: industry advocacy groups and environmental advocacy groups. Advocacy groups include “groups or organizations that make public interest claims either promoting or resisting social change that if implemented, would conflict with the social, cultural, political, or economic interests or values of other constituencies and groups” (Andrews and Edwards 2004, 495). These include diverse groups ranging from large institutionalized nonprofit organizations, to small informal social movement organizations or community groups (Mosley 2011).
These organizations cross a variety of literatures, including interest group, social movement, and nonprofit literatures (Andrews and Edwards 2004), and they are certainly relevant to other literatures, like the policy process literature, as well. For example, interest groups generally refer to voluntary organizations that attempt to influence government. In contrast, the social movement literature typically focuses on informal groups that attempt to effect systemic change from outside government institutions. Finally, the nonprofit literature typically zooms in on more formal nonprofits, which share similar basic structures and adhere to corporate and tax laws.
Three primary types of advocacy groups exist in Colorado’s oil and gas policy subsystem and are included in this study. The first is formal nonprofit organizations. These organizations generally incorporate pursuant to state law and register for tax-exempt status with the IRS. To gain federal tax exemption, nonprofits must register under Title 26 of the IRS code, typically under a subpart of section 501(c). Most relevant to this study are 501(c)(3), (4), and (6)
4 In the ACF, “coalitions are defined by actors who share policy core beliefs and who coordinate their actions in a nontrivial manner to influence a policy subsystem” (Jenkins-Smith et al. 2018, 148).
28


organizations. Environmental nonprofits usually register as either 501(c)(3) or 501(c)(4) organizations. Section 501(c)(3) organizations are public charities or private foundations, and section 501(c)(4) organizations are social welfare organizations. While 501(c)(3) organizations are limited in certain types of advocacy in exchange for tax benefits, section 501(c)(4) organizations are not limited and are often characterized as advocacy organizations. Particularly in the environmental space, though, 501(c)(3) organizations advocate through limited amounts of lobbying, educating the public or local leaders, and engaging with the media, among other things (see Chapter 3).
On the other hand, many industry advocacy groups register as 501(c)(6) organizations. These organizations are business or professional organizations and are typically comprised of organizational or individual members from particular industries or professions. Similar to 501(c)(4) organizations, these organizations are not limited in their advocacy activities. A prime example of this organization type is the Colorado Oil and Gas Association (COGA), which represents the oil and gas industry and “promotes the expansion of Rocky Mountain oil and natural gas markets, supply, and transportation infrastructure through its growing and diverse membership” (COGA 2019).
The second type of advocacy group involved in oil and gas policy in Colorado is informal groups. These organizations vary in their structures, but are typically smaller, volunteer run, and less structured than formal nonprofits. These are often neighborhood or community groups that represent local citizen interests and are sometimes organized via listservs or social media groups. In some cases, these local groups are nested in larger formal nonprofits.
Finally, action groups are those that organize around specific policies, most commonly to support or oppose a ballot measure. These organizations are sometimes registered under section
29


527 as political action committees (PACs), but what distinguishes them is that they are typically short-term organizations that work on a specific policy pending in a particular venue.
Study Context: Oil and Gas Policy Debates in Colorado
To improve our understanding of the activities that advocacy groups use across venues, this study relies on news media coverage of hydraulic fracturing (fracking) policy debates in Colorado. Fracking is a technique involving the injection of water, sand, and chemical additives into shale rock or other porous formations thousands of feet beneath the earth’s surface to release oil or natural gas (Heikkila et al. 2014). In the US and around much of the world, fracking is increasingly used in oil and gas development. For example, between 2007 and 2017, the amount of natural gas produced from shale formations increased 10 fold from 1,990,145 million cubic feet to 19,018,457 million cubic feet (EIA 2018).
Not only has the volume of oil and gas produced through fracking increased, but oil and gas development is also encroaching on high population density areas. Close proximity between population centers and oil and gas operations gives rise to a host of local issues, including NIMBYism (not in my backyard); nuisance issues, like traffic, dust, and noise; and concerns about air and water contamination. As a result, communities grapple with how to manage fracking, and some have attempted to ban it. For example, while the state of New York successfully banned fracking, local bans in a number of municipalities like Denton, Texas, and Longmont, Colorado, have been struck down by courts or legislatures (Knight and Gullman 2015; Heikkila and Weible 2016; Jaquith 2017). Local communities with limited authority over oil and gas development may attempt to exercise control by negotiating agreements with oil and gas companies (Zilliox and Smith 2017), regulating nuisance issues, and advocating for increased setbacks and other limits at the state level. Beyond local issues, fracking is part of
30


broader national and global debates about reliance on fossil fuels as energy sources. While those in favor of fracking often cite natural gas as a cleaner alternative to oil and coal, those opposed claim that fracking and the availability of cheap oil and natural gas impede transitions to renewable energy sources, like wind and solar.
Because they involve a wide range of issues across venues, fracking debates provide a good context for this study. Fracking is an intensely debated political issue (Heikkila and Weible 2016), drawing participation from actors across sectors. Fracking policies are created via judicial decisions, legislation, agency rules, executive orders, and ballot measures, and organizations, including diverse advocacy groups, engage in a variety of activities and strategies to influence each of the sources of policy. As such, fracking is an ideal policy context for observing advocacy activity across a range of venues.
This study focuses specifically on the oil and gas policy subsystem in Colorado.
Colorado ranks eighth in the nation for shale gas production and seventh in the nation for crude oil production. Furthermore, as a “purple” state,5 Colorado regularly grapples with oil and gas issues and has seen its share of fracking related controversy. Colorado is also an ideal state for studying fracking because of the conflicts occurring across venues. For example, local government regulation of fracking has been an issue across the state. In 2012, the city of Longmont banned fracking via voter initiative, and in 2016 the Colorado Supreme Court ruled that the ban was invalid and unenforceable. Relatedly, in 2014, as a compromise to remove two statewide citizen initiatives from the ballot, Governor Hickenlooper established a task force to address state and local authority over fracking. During the time period covered by this study,
5 During the majority of the study period, Colorado’s state house had a Democratic majority and its state senate had a Republican majority. Governor John Hickenlooper is a Democrat, but is widely seen as supportive of the oil and gas industry.
31


numerous bills were introduced at the legislature, such as bills related to setbacks, forced pooling, and incident reporting laws, though bills saw very little success. At the state agency level, the Colorado Oil and Gas Conservation Commission (COGCC) primarily oversees oil and gas activity in the state, though other agencies like the Colorado Department of Public Health and Environment and the Colorado Department of Natural Resources oversee issues relevant to oil and gas development like air pollution and wildlife. Furthermore, while much oil and gas development occurs on private land on the Front Range near population centers, 36% of land in the state is federally owned, and, particularly on the Western Slope, oil and gas development regularly happens on federal land.
Methods
Data Collection
This study relies on the news media as its data source. The news media can be a valuable research tool in that it is publicly available and tracks trends over time, allowing for lower cost longitudinal studies that do not rely on individuals’ memories. At the same time, though, there are certainly limitations associated with the news media. For example, only those issues, actors, and activities that are most visible are likely to gain media coverage, while other, sometimes important, events and activities may go unnoticed by the media. Furthermore, considerable scholarship often notes that the media plays a role in influencing policy either by shaping public opinion or by serving as a tool that policy actors may exploit (Shanahan et al. 2008; Jones and Wolfe 2010).
Despite the benefits and drawbacks, scholars have long recognized the media as a potential resource. For example, Kingdon (2003) stated that “[t]he media report what is going on in government, by and large, rather than having an independent effect on governmental agendas”
32


(59). Recent studies continue to rely on the media. For example, the Narrative Policy Framework (NPF) makes widespread use of the media for understanding the narratives involved in policy conflicts (e.g., Shanahan, McBeth, and Hathaway 2011; Crow and Berggren 2014). Also, more recently, studies increasingly extend beyond traditional news media and look at diverse outlets like social media (e.g., Merry 2016)
Data for this study were drawn from newspaper articles in Colorado. This research uses two newspapers likely to have extensive coverage of oil and gas debates: the Denver Post and the Longmont Times-Call. Denver is the capital and largest city in Colorado. Thus, state-level policy and decision making is centered in Denver, and because they operate within the confines of state law, local debates often tie to state decision-making. Denver is also home to regional federal offices, along with oil and gas industry headquarters and regional offices, making it a hub of governance and decision-making. Longmont, on the other hand, is a center of oil and gas policy activity because it has a high concentration of wells. Longmont also straddles Boulder and Weld counties, which often approach oil and gas policy differently. In 2012, Longmont residents banned hydraulic fracturing within city limits via a citizens’ initiative, which was later ruled invalid and unenforceable by the Colorado Supreme Court.
As part of a larger project, a team of researchers accessed articles through Newsbank and used a two-step process to identify articles related to oil and gas debates in Colorado. The first step used the following search terms to identify potentially relevant articles from each newspaper published from 2007 to 2017: “fracking,” “hydraulic fracturing,” “shale oil,” “shale gas,” “oil drilling,” or “gas drilling.” This search resulted in 1,588 articles from the Denver Post and 1,450 articles from the Longmont Times-Call. Once articles were identified and downloaded, the second step consisted of a manual, or automated when possible, review of articles to ensure that
33


they are relevant to oil and gas policy debates in Colorado. The following categories of articles were excluded from the final population: opinions, editorials, or letters to the editor; articles only tangentially related to oil and gas issues; articles about federal issues, global issues, or issues in another state that do not relate to Colorado; and briefs or short summaries of a series of news topics, only one of which is oil and gas related. After exclusions, the final population includes 908 articles published between 2007 and 2017. Because the time period for this study is 2013-2017, the final sample includes 534 articles.
Data Analysis
To analyze the news articles, I uploaded and hand coded articles in NVivo. Coding was guided by constant comparative analysis, which is an iterative coding process often used in grounded theory studies (see Corbin and Strauss 2015). In a first round of open coding, I coded all 908 articles for major policy issue, primary venue, and date. This round of coding resulted in 78 distinct policy issues and 25 venues. Policy issues and venues were then refined and reapplied in a round of closed coding to the 534 articles that were published in the 2013 to 2017 time frame. The second round of coding included 23 policy issues and 11 venues. A third review of the sample articles was performed to ensure that codes were consistently applied across the articles.
Next, I identified advocacy groups, based on the three categories described above, in the sample articles using Nodes in NVivo. I included all advocacy groups appearing in articles, unless they were only tangentially mentioned or mentioned in part of the article not directly related to fracking. In other words, there was no requirement that advocacy groups be mentioned in relation to a particular venue or activity. While in some cases, articles mentioned that groups participated in a particular way, such as protesting at the capitol or filing a court brief, in many
34


cases references were limited to short quotes or mentions that groups supported or opposed certain policies or actions. Overall, 52 environmental advocacy groups were mentioned in 200 articles, and 15 industry advocacy groups were mentioned in 213 articles.
Results
This section presents preliminary answers to each of the research questions using both descriptive and inferential statistics and graphs. Again, the research questions guiding analysis are as follows: 1) What are the different types of venues and policy issues that advocacy groups are involved in within the Colorado oil and gas policy subsystem? 2) Are there differences in how industry advocacy groups and environmental advocacy groups participate across venues? RQ1: Venues and Policy Issues Summary of Venues
As a preliminary matter, the frequencies with which venues appeared in the Colorado news media from 2013 to 2017 are shown below in Table 1 and Figure 1. Venues were coded at the article level.6 While the majority of articles describe activity in a particular venue, just over a quarter of the articles in the sample are not grounded in a particular venue. These include features about the economic or health impacts of oil and gas on local communities; news about the oil and gas industry, including job loss and creation; and accidents or safety issues on well pads, among other issues.
6 Apart from articles that are not grounded in a particular venue, as described in the text, all articles contain a dominant venue. Many articles focus exclusively on one venue, and even among those articles that mention more than one venue, there is a clear primary venue.
35


Table 1. Venue Frequency in Articles.
Venue Number of Percent of Articles Articles
No Venue 140 26%
Local Government 132 25%
State Court 80 15%
State Legislature 41 8%
State Agency 41 8%
State Ballot 32 6%
Local Ballot 21 4%
Governor 19 4%
Federal Combined 28 5%
Figure 1. Venue Frequency in Articles.
The most active policy venue during the time period is local government. Between 2013 and 2017, numerous local governments banned fracking or implemented moratoria and more broadly struggled with regulating fracking and interpreting their authority under state law. Articles about local issues also sometimes fall in other venues, such as in the case of the COGA v. Longmont lawsuit, which played out in state court, but a considerable portion of the articles in
36


this period is covered activity by local governments. While some of this action was by city councils and other activity was in local agencies, because city councils often have to approve agency actions, and because local government action is often discussed broadly, local government activity is combined.
Next most frequently, articles discuss state court activity. In 2012, voters in Longmont enacted a fracking ban via a local ballot initiative, and in 2013, Fort Collins implemented a five-year moratorium. In response to these policies, the Colorado Oil and Gas Association (COGA), joined by the COGCC, sued these municipalities in state court, claiming that local governments lacked the authority to ban fracking because state law preempts local government regulation.
The district court sided with COGA, and in early 2016, the Colorado Supreme Court affirmed, deeming the local policies invalid and unenforceable. Following this decision, however, Boulder County kept a fracking moratorium in place while it worked on adopting new local rules to regulate fracking. This moratorium was extended numerous times, and in 2017, the Colorado Attorney General sued Boulder County as a result.
Issues at the state legislature also arose during this period, although, likely because of the split legislature, this was not the most active venue during this period. State agency activity primarily refers to rulemaking or other activity at the COGCC, which is the primary regulator of the oil and gas industry. Activity in a few other agencies, such as the Colorado Department of Natural Resources and the Colorado Department of Public Health and Environment, is included as well. Smaller numbers of issues arose at the federal level, through state ballot initiatives, the governor’s office, and local ballot initiatives during this period. The small number of federal
37


issues likely reflects the fact that articles were only included in this sample if they included ties
to Colorado.7
Venues and Policy Issues
As the above begins to suggest, there is certainly overlap between venues and issues.
That is, within a given subsystem, while some venues hear a wide range of issues, others primarily focus on just one or two issues. For example, as described above, local governments frequently dealt with bans and moratoria during the relevant time period. While the key questions in this chapter relate to advocacy group participation in venues, a basic understanding of the relationship between venues and sub-issues, also allows for conclusions regarding the factors that may drive participation in venues. As already noted, 23 distinct policy issues were identified in the data. These issues are detailed in Appendix B to the dissertation.
To highlight the complexity of the Colorado oil and gas policy subsystem, Table 2 below shows the frequency with which policy issues occur across levels of government. It also demonstrates that while there is some relationship between venues and policy issues, issues are certainly not confined to a single venue.8 The Table highlights that in some cases, like with fracking generally, industry safety, and procedural matters, issues are heard across venues.
Other issues, however, like public land, are largely confined to just one or two venues. Thus, when actors participate in venues, they may do so because they believe that the venue is friendly, or because an issue already being considered in that venue is of interest to them.
7 The Federal Combined Category includes federal courts, agencies, and Congress.
8 To simplify the table, and because the venue variation is clear, venues are reduced to level of government here.
38


Table 2. Policy Issue by Level of Government.
RQ2: Advocacy Groups
Advocacy Groups
This section examines differences in how industry advocacy groups and environmental advocacy groups participate across venues to contribute to our understanding of venue selection. As a preliminary note, 57% of the articles (307 out of 534) include mentions of advocacy groups. Industry groups appear in 213 articles (40%), and environmental groups appear in 200 articles (37%). It is not clear whether this accurately reflects whether groups were in fact involved. For example, where advocacy groups are not mentioned, it is possible that they were involved but were acting behind the scenes and outside of the media’s eye. Table 3 shows the breakdown of advocacy group mentions across all articles by year and suggests that advocacy group activity in both coalitions has declined on average over time.
39


Table 3. Articles and Advocacy Group Activity by Year.
2013 2014 2015 2016 2017
Number of Articles 122 149 84 75 104
% with Environmental Groups 42.62% 40.94% 36.90% 36% 27.88%
% with Industry Groups 40.16% 45.64% 38.10% 34.67% 36.54%
Next, advocacy group activity by venue is shown below in Table 4. As noted above, associations are based on whether actors appear in articles about particular venues, and not whether the article mentions that actors engaged in any specific way. As the results suggest, advocacy groups in both coalitions participate across venues, and for most venues, they participate at similar rates. Just four venues (or non-venues) show statistically significant differences in rates of participation. First, where issues are not related to any particular venue, articles mention industry groups significantly less than environmental groups. As shown above, these most commonly are articles that talk broadly about the economic effects of fracking, industry safety concerns, or fracking generally, each of which may not relate to particular policy proposals. Thus, it may be that industry groups reserve their activity for pending policies, rather than get involved in general discussions about fracking and its effects. On the other hand, industry groups participate significantly more in state court activity than do environmental nonprofits. In the 2013 to 2017 time frame, this is most likely because COGA sued Longmont and Fort Collins in response to their ban and moratorium, respectively, and this litigation received considerable attention in both Colorado and nationally.
40


Table 4. Venues with Mean Participation by Advocacy Group (T-Test Results).
Venue Number of Articles Environmental Groups Industry Groups
Not Applicable** 140 0.314 0.207
Local Government 132 0.295 0.311
State Court*** 80 0.413 0.775
State Legislature** 41 0.171 0.39
State Agency 41 0.488 0.366
State Ballot 32 0.75 0.688
Local Ballot 21 0.476 0.381
Governor 19 0.368 0.474
Federal Agency* 16 0.688 0.375
Federal Court 6 0.333 0.5
Federal Legislature 6 0.5 0.333
*p < .1 **p < 05 ***p < 01 (two-tailed)
Environmental groups are also significantly less likely than industry groups to be involved in issues at the state legislature. As noted above, this may be related to the fact that throughout much of the study time period, the state house was controlled by Democrats, while the state senate was controlled by Republicans, making it difficult to pass legislation. Environmental groups may have believed other venues were more favorable to their interests. On the other hand, environmental groups are significantly more likely to be involved in federal agencies. This difference may be attributable to the fact that many environmental nonprofits specialize in federal public land issues. In these instances, federal agencies, such as the Bureau of Land Management (BLM), often serve as opponents, rather than oil and gas interests (though these groups are certainly involved as well).
Because there is some relationship between policy issue and venue, with certain issues appearing more frequently in certain venues, Table 5 below shows comparisons between industry group and environmental group involvement across the most frequently mentioned policy issues. In other words, this analysis tests the possibility that issues are the sole drive of
41


venue activity. In the case of policy issues, significant differences exist between the groups in just two areas: bans and moratoria and the anti-fracking movement. With respect to bans and moratoria, industry groups are significantly more likely to be involved. It may be that these issues pose a larger threat to the industry because bans and moratoria may impair oil and gas industry operations, and thus, industry advocacy groups are likely to get involved. Not surprisingly, environmental advocacy groups dominate articles about the anti-fracking movement. Rather than playing an advocacy role, though, these groups are generally the subjects of articles. Thus, policy issues may contribute to differences in participation in state court, which often deals with bans and moratoria. Other differences in venue participation, such as at the state legislature, do not appear to be issue driven. This suggests that, at least in some cases, there are venue-related factors that drive advocacy group participation.
Table 5. Policy Issues with Mean Participation by Advocacy Group (T-Test Results).
Issue Number of Articles Environmental Groups Industry Groups
Ban/Moratorium * * * 131 0.351 0.527
Local Authority 78 0.538 0.654
Fracking generally 72 0.347 0.319
Siting/setbacks 52 0.365 0.288
Industry Safety 32 0.094 0.188
Economy/Jobs 26 0.154 0.115
Public Land 19 0.632 0.368
Procedural 13 0.538 0.384
Air Quality 13 0.538 0.462
Anti-fracking Movement** 12 0.667 0.167
Industry News 12 0.167 0.333
*p < .1 **p < 05 ***p < 01 (two-tailed)
Bans. Moratoria. and Local Authority
To deepen our understanding of advocacy groups and venue, this study next zooms in to a specific policy issue in the Colorado oil and gas policy subsystem and traces the issue across
42


venues and advocacy groups, using the coded news media data. Specifically, this study examines issues related to bans, moratoria, and local authority, which were the most prevalent issues in the data. While these issues were coded separately under two categories (ban/moratorium and local authority), they are in fact closely related in that local authority issues often center on whether local governments have the authority to ban or halt fracking.
Figure 2 below shows a timeline of the major events related to bans, moratoria, and local authority issues in the 2013-2017 time period. The Figure reflects only the subset of articles in this time period related to two previously identified policy topics: ban/moratorium and local authority (209 articles combined). Thus, the line reflects the number of articles that appeared in each quarter on these topics, and the boxes above the timeline show major events based on a manual review of the 209 articles. First, to fully understand the 2013-2017 time period in Colorado in relation to oil and gas development, it is necessary to go back to the end of 2012. In November of 2012, the citizens of Longmont voted to ban oil and gas development that uses hydraulic fracturing within city limits. At this point, Longmont was already a hub of oil and gas conflict. At the time, there was a lawsuit pending against Longmont, initiated by the COGCC, for the city’s enactment of a set of restrictive rules. Thus, the fracking ban secured Longmont’s place as ground zero for state, if not national, fracking policy conflicts. In response to the voter-initiated ban, in December 2012, COGA, later joined by the COGCC, sued Longmont claiming that local governments lack the authority to ban fracking because state law preempts local government regulation.
43


NOv2012 Longmont tan? fracking
OK 2012 2011
COGAsues Other local
Longrnont actions
July 2014 Boulder judge sides with COGA
Aug 2014 Governor compromise
July 2014 State approves ballot measures

Feb 201 & Taak force laaues report
OK 201 & COGCC release* proposed rules
August 2015 Court of Appeals -^Supreme Court
May2016 Supreme Court strikes down ban
Feb 2017 AG sues Boulder
2013
2014
2015
2016
2017
i_ 20
8 »
£
v*
1“
5
= >
Figure 2. Timeline of Bans, Moratoria, and Local Authority Issues.
44


In addition to the pending litigation, action in 2013 and early 2014 primarily involved local efforts to ban or delay fracking, and in 2014, action heated up again when groups from both sides of the issue drafted and began collecting signatures for ballot measures related to local authority. In attempt to thwart these measures, Governor Hickenlooper worked with the Colorado legislature to draft legislation related to local authority. Neither side was happy with proposed bills, though, and in May 2014 the legislative session ended without successfully resolving the issue.
In July 2014, two major events occurred. First, the district court judge in Boulder sided with COGA and ruled that the Longmont ban was unenforceable, though the decision was stayed pending appeal. Second, the state approved two ballot measures for the November ballot, both of which posed significant threats to the industry. One measure would have created an environmental rights amendment to the state constitution, and a second would have increased setbacks, or the distance between well pads and homes, from 500 feet to 2,000 feet. In response to these ballot measures, Governor Hickenlooper successfully struck a compromise with US Representative Jared Polis, who bankrolled the measures, in which Hickenlooper agreed to create a task force to address local authority issues in exchange for Polis withdrawing the ballot measures. The task force began meeting in September 2014 and issued its report in February 2015.
Later that year, in August 2015, the Colorado Court of Appeals requested that the Colorado Supreme Court rule in the Longmont fracking ban case, and the Supreme Court accepted the appeal. As the litigation continued, in December 2015, the COGCC released a set of proposed rules in response to the Task Force’s recommendations. The final major event came in May 2016, when the Colorado Supreme Court ruled in favor of COGA, deeming Longmont’s
45


ban, along with a moratorium in Fort Collins, invalid and unenforceable. When Boulder County continued to extend its moratorium, in February 2017 the Colorado Attorney General sued.
While the peaks and valleys in news media coverage generally follows the events described above, some of the additional patterns in news media coverage derive from local decisions regarding bans and moratoria throughout the time period. In addition to Longmont’s ban, Brighton, Fort Collins, Boulder County, Broomfield, and others all enacted bans or moratoria early in the time period. In addition, a measure to halt fracking in Loveland was voted down by voters in 2014. Boulder County in particular received a lot of press coverage for its activities. Between 2013 and 2017, Boulder County extended its moratorium five times before the Colorado Attorney General sued the county. Boulder ended its moratorium in May 2017. Furthermore, despite the Supreme Court ruling, in late 2017, local governments, including Lafayette and Dacono, enacted moratoria 2017, and Broomfield voters passed a ballot measure giving the local government more local control over oil and gas operations.
Activity by venue. As the above suggests, policy disputes related to local authority issues spanned a number of venues. Throughout the relevant time period, seven venues were involved: the Governor’s office, local ballot boxes, local governments, state ballot boxes, state courts, the state legislature, and state agencies. As shown in Figure 3, the majority of activity occurred in state courts and local governments. Data for the Figure are based on the 209 articles coded as ban/moratorium or local authority, and the lines reflect the raw number of articles by year associated with each venue. The Figure also shows an inverse relationship between local government activity and state court activity. This suggests that while local governments provided venues for resolving fracking disputes early in the time period, once COGA moved the dispute to state court by suing Longmont and later other local jurisdictions, local governments
46


became less active as venues. Interestingly, though, after the Supreme Court decision, which significantly limited local authority, was announced, local government activity increased, as local governments looked for ways to regulate within their newly delimited authority. Similarly, though less dramatically, activity at local ballot boxes was higher early in the time period,
dropped off as other venues took off, and then began to rise again at the end of the time period.
2013 2014 2015 2016 2017
Governor Local Ballot Local Government
State Ballot State Court State Legislature
State Agency
Figure 3. Activity Across Venues by Year.
Patterns are less clear in other venues. For example, state ballot activity saw a dramatic increase in 2014, but was minimal throughout the rest of the time period, and activity in each of the other three venues—Governor’s office, state legislature, and state agencies—was relatively stable and low throughout the time period. The only venues not involved in this issue are federal venues, including federal courts, agencies, and Congress. Because this issue is about state and local control this is a logical result. Taken together, the results in this section may suggest that activity across venues is somewhat of a tradeoff. That is, when activity in one venue is high,
47


activity may be calmer in other venues. It is not clear, though, whether this is because other venues are waiting on decisions or because advocacy groups lack the capacity to work across multiple venues at once.
With respect to advocacy group involvement, Figure 4 shows that across most venues, we see similar levels of engagement by both environmental advocacy groups and industry advocacy groups. This suggests that alignment between group ideology and dominant venue ideology may not be a primary driver of venue selection by advocacy groups. That is, for example, if state agencies tended to favor environmental groups and the state legislature tended to favor industry groups, we would expect to see more environmental group activity in state agencies and more industry group activity at the state legislature, at least if advocacy groups are primarily proactive about their activities. Results below, however, do not support ideology as a major driver of venue selection.
48


60
50 40 30
..ill III
Governor Local Ballot Local State Ballot State Court State State Agency Government legislature
â–  Environmental â–  Industry
Figure 4. Activity Across Venues by Advocacy Group.
The only venue in which there is a dramatic difference between the two coalitions is in state courts. Among the venues involved in this issue, state courts have the highest barrier to entry because of the resources and expertise needed to participate and because participation is restricted by courts. Once litigation has commenced, it is up to courts to decide whether other groups may join the litigation. Furthermore, as Chapter 3 shows, outside advocacy strategies are least likely injudicial venues, meaning that participation is primarily limited to those directly involved in litigation. That is, as described more fully in Chapter 3, in contrast to when policies are pending at regulatory agencies or at the legislature, when issues are before courts, advocacy groups are less likely to run media campaigns, protest, or engage in other grassroots efforts. Thus, these results lend support to existing studies that find a relationship between venues and organizational resources.
49


Patterns in individual venues. Next, by breaking activity apart by venue, and comparing all activity (reflected by number of news articles), industry group activity, and environmental group activity, we can see patterns of advocacy group activity in each venue. Again, data derive from the 209 news articles coded as either ban/moratorium or local authority. While in some cases, activity may be led by advocacy groups, in other cases, advocacy groups seem to be merely reacting to activity in the venue or to activity by the other coalition. The results for each venue are available in Appendix C, and a few examples are shown below.
First, activity at the legislature over the five-year period demonstrates that activity may be happening independently of advocacy groups. For example, Figure 5 shows a spike in activity in early 2014, and at least based on news media coverage, some activity occurs independently of advocacy groups. Furthermore, at least with respect to some activity, legislators may have played a dominate role. For example, the Figure shows that another spike in activity began in mid-2015 before advocacy groups got involved, thereby suggesting that advocacy groups may have reacted to existing activity.
State Legislature
5
4
3
2
1
0

A

<£> & &
*V 'X' \
•# Articles
â– Environmental Groups
â– Industry Groups
Figure 5. Local Authority Related Activity at the State Legislature.
Activity at the state legislature should also be viewed in contrast to state ballot measures where all activity involved advocacy groups, as shown below in Figure 6. Perhaps because the
50


venue does not exist independently of the ballot measures, those within the venue cannot initiate action. Also, while individuals or small groups may have the capacity to initiate a local ballot measure, because the state ballot measures require greater resources, advocacy groups are likely to be involved. Accordingly, we do see some evidence of activity not involving advocacy groups with local ballot measures, as shown below in Figure 7.
State Ballot
10
5
0
y\
6^ & & & <£ & &
'V 'V 'V 'V 'V
â– # Articles
â– Environmental Groups
â– Industry Groups
Figure 6. Local Authority Related Activity at the State Ballot Box.
Local Ballot
8
â– # Articles
â– Environmental Groups
â– Industry Groups
Figure 7. Local Authority Related Activity at the Local Ballot Box.
Finally, at the state court, we see a similar pattern in the sense that little activity occurs that does not involve advocacy groups. See Figure 8 below. In the case of the COGA v. Longmont lawsuit, there is no question that COGA initiated the action, and as shown above, industry groups were more active than environmental groups, likely due to barriers associated
51


with the venue. Thus, while groups follow a similar pattern of activity, engagement by environmental advocacy groups is typically at a lower level than industry group activity.
State Court
# Articles ^—Environmental Groups ^—Industry Groups
Figure 8. Local Authority Related Activity at State Courts.
Thus, through looking at activity across venues in a particular subsystem and then with respect to a specific policy issue, a few lessons may be gleaned. A discussion of the study’s major results are presented in the next section.
Discussion
The results in this study paint a picture of the policy activity within a given subsystem and of the venues and actors at play in that subsystem. The results suggest that in a high conflict policy subsystem such as the one in this study, numerous sub-issues and venues may be involved. As such, advocacy groups have numerous options for how to influence policy. Not only are many venues in play, but across venues, groups can choose from an array of ways to influence the outcomes in those venues. While these options provide opportunities, they also affect and potentially deplete groups’ resources as they attempt to chase policy issues across venues.
This study, which is limited to a five-year period, identifies over 20 distinct issues and 11 venue types in the subsystem. Not only does this indicate the complexity of the policy
52


environment, providing opportunities but also forcing challenging strategic decisions, but it also offers fertile ground for studies related to polycentricity and conflict over extended time periods. That is, because policy subsystems may be comprised of a broad range of venues and sub-issues, advocacy groups must devote significant resources to their advocacy activities, and only rarely see an end to issues. If groups are on the losing side of an issue, they may look for opportunities to expand the conflict or move it to other venues, and if they are on the wining wide, they must be prepared to play defense in some other venue.
With respect to actors, this study zooms in on advocacy groups with conflicting perspectives: industry advocacy groups and environmental advocacy groups. The results indicate that both types of groups are active in the policy subsystem, each appearing in more than one-third of the sample articles. Furthermore, results suggest that these groups participate at similar rates across venues and policy issues. More differences between the groups exist when comparing their activity by venue, as compared with policy issue, thus suggesting that venues are, at least to some degree, a motivating factor in how advocacy groups participate in subsystems.
This study’s results also suggest a few additional points that may serve as useful starting points for future studies. First, this study provides some support for the notion that cumulative activity in a policy subsystem across venues may remain relatively stable over time. As shown above, for example, when state court activity increased in the oil and gas subsystem in Colorado, local government activity decreased. Also, while the data show a clear peak in news media coverage of the subsystem in 2014, there was consistent activity throughout the study’s time period. As such, future studies should consider whether fluctuations in activity happen as a
53


result of capacity issues, the fact that venues may await decisions in other venues before taking action, or both.
Second, two results combine to suggest that factors beyond those typically identified in the literature may be at play in explaining venue selection. The results do not lend support to the idea that ideological alignment between group and venue is a factor in venue participation, because advocacy groups with conflicting ideologies participate at similar rates across venues. Further research is needed to tease out this relationship, though, because it may be that ideology is important in initial venue selection, but that much of venue participation is driven by reactions to opponents’ venue decisions. It also may be that ideology drive the types of strategies (especially inside versus outside strategies) that groups use in each venue. This is not to say that ideology is not a factor, but rather, there may be other factors at play, such as reactivity or differences in strategy selection that are not picked up in this study. The results do support existing knowledge about the role that resources play in venue selection. While across most venues environmental advocacy groups and industry advocacy groups participate at similar rates, industry advocacy groups were significantly more active at state courts, which have the highest barriers to entry. An alternative explanation, though, may be that resources dictate not whether advocacy groups participate, but how they participate. That is, as Chapter 3 suggests, nonprofit organizations are least likely to use outside strategies with respect to courts. Thus, taken together these results suggest that future studies should consider the interaction between venue selection and the use of insider or outside advocacy activities.
Third, results also suggest that organizations may be at least as reactive in venue selection as they are proactive. Two piece of evidence support this finding. First, there is little variation between coalitions with competing ideologies in terms of their participation in venues.
54


Second, examining activity in specific venues shows some evidence that activity sometimes precedes advocacy group involvement. Thus, the venue selection literature would benefit from further exploration of venue selection as a reactive strategy. It is also not clear based on the existing study whether reactivity occurs in relation to venue activity, opponent venue selections, or both.
A few limitations are associated with this study that may provide avenues for future research. First, as noted above, by relying on news media, this study may not capture all subsystem activity and may be less inclusive of events and advocacy groups that occur and work primarily outside of the public eye. Thus, future studies might combine news media with other data sources, such as interviews or reviews of legislative or administrative records. Second, because this study focuses on a single policy subsystem, it is not clear the extent to which results are generalizable. While the use of case studies for theory development is necessary and appropriate, future studies should extend to other contexts as well. Finally, this study collectively examines groups with similar ideologies without accounting for the wide variation that exists among these groups. As noted at the outset, these groups include formal nonprofits, informal groups, and action organizations. Not only may there be wide variation in how these groups group individually advocate, but decisions about venues and advocacy in one groups will likely affect how the other groups advocate. Thus, future studies should look at differences between organizations within a particular ideology as well.
Conclusion
This study maps out a complex and high conflict policy subsystem: oil and gas policy in Colorado. By examining how actors with competing viewpoints interact with venues, we see how venues, issues, and actor participation evolve over time. While some results confirm
55


existing venue shopping literature, others suggest new areas of inquiry. As a starting point, the next chapter examines the inside and outside strategies that nonprofit actors use to influence policy across venues.
56


CHAPTER III
HOW DO ENVIRONMENTAL NONPROFITS ADVOCATE?
AN EXPLORATION OF INSIDE AND OUTSIDE STRATEGIES ACROSS VENUES
Abstract
This chapter explores the advocacy strategies that nonprofit organizations use to influence environmental policy across five venues: legislatures, regulatory agencies, courts, ballot boxes, and governors’ offices. Despite widespread agreement on a definition of advocacy, operationalizing the concept presents challenges. This chapter proposes a venue based approach to understanding advocacy and fills gaps in the literature by describing the activities environmental nonprofits use across venues. Data are based on semi-structured interviews with 22 environmental nonprofits across three states: Colorado, New Mexico, and Oklahoma. Results illuminate the venues and strategies most common to these organizations, along with the organizational and contextual determinants of nonprofit decisions regarding whether and how to get involved in policy advocacy.
Keywords
Environmental policy, nonprofit advocacy, policy venues, advocacy strategies
57


Building on the prior chapter, which examined patterns of advocacy group participation across venues, this chapter develops a deeper understanding of the specific types of advocacy that environmental nonprofits use across different venues. Thus, this chapter describes advocacy activities in each of five venues described in the previous chapter: regulatory agencies, legislatures, courts, ballot boxes, and governors’ offices. As mentioned previously, the existing nonprofit literature suffers from limitations, including that studies of nonprofit advocacy to date primarily focus on social service providers; advocacy is often operationalized using blunt, dichotomous measures; and assumptions often inhere in the extent to which different types of organizations advocate. In an effort to address some of these limitations, this study presents an exploration of the advocacy activities of environmental nonprofits in each of five specific venues and the factors that influence decisions regarding advocacy.
This paper explores two research questions. First, what activities do environmental nonprofits engage in across different venues to influence the policy process? Second, what factors influence the types of advocacy these organizations choose? This paper begins with a review of relevant literature, including the importance of nonprofit advocacy and research to date on advocacy strategies. Next, I present the context for this study: hydraulic fracturing policy debates in Colorado, New Mexico, and Oklahoma. This paper then describes the interviews used for data collection, and the results, including how environmental nonprofits advocate in different venues and the factors that may affect differences in advocacy. The paper concludes with a brief discussion of the findings, limitations, and potential avenues for future research.
Literature Review
While studies about advocacy activities are not new, recent years have seen a surge in nonprofit advocacy research. For example, recent scholarly work has focused on the advocacy
58


activities of human service nonprofits and the drivers of those activities (Mosley 2010; Almog-Bar and Schmid 2014), how nonprofits influence policy and effect policy change (Hula and Jackson-Elmoore 2001; Andrews and Edwards 2004; Bryce 2006; Fyall and McGuire 2015;
Fyall 2016), and ways of understanding and classifying nonprofit advocacy organizations and their activities (Boris and Mosher-Williams 1998; Child and Gronbjerg 2007). Despite this growing body of literature, current knowledge remains limited, particularly because advocacy is difficult to define and operationalize. After an overview of nonprofit advocacy, this section describes literature on advocacy strategies, factors that influence strategies, and limitations of existing studies.
Nonprofit Advocacy
While advocacy is a relevant concept across a variety of literatures, including political science, public policy, nonprofit management, and social movements, the concept remains difficult to operationalize. From a definitional standpoint, advocacy is “the attempt to influence public policy either directly or indirectly” (Pekkanen and Smith 2014a, 3). As this definition suggests, advocacy is a broad concept that may include direct attempts to influence policy, such as lobbying, litigating, or testifying before decision-making bodies, and indirect efforts such as education, media campaigns, or voter mobilization. Historically, studies of nonprofit advocacy have focused primarily on rights based and political organizations, but calls for broader conceptualizations of advocacy, such as that by Boris and Mosher-Williams (1998), have led to a proliferation of research that recognizes the breadth of forms that advocacy can take (e.g., Maclndoe 2014; Almog-Bar 2018).
Nonprofit advocacy is important for at least two reasons. First, nonprofits hold potential benefits for democracy. That is, “[t]hrough advocacy activities, nonprofit and voluntary
59


organizations contribute to democratic governance by representing the interests of citizens and influencing public policy” (Guo and Saxton 2010, 1). Indeed, one of the primary purposes of nonprofits is to provide a voice for those whose interests might not otherwise be represented (Child and Gronbjerg 2007; Almog-Bar and Schmid 2014; Mosley 2016). In the 1830s, Alexis de Tocqueville, observing the US, noted that associations are essential in democratic societies to overcome the fact that independent citizens otherwise lack political power (de Tocqueville 1966). More recently, according to Putnam (1993), voluntary associations help build social capital, and nonprofits help secure citizen rights, including political, civil, and social rights (see also Smith 2007). At the same time, though, others describe that nonprofits may promote special interests and exert disproportionate influence on policymaking (Almog-Bar 2018; Boris 2006). That is, rather than strengthen the voice of those lacking political power, some nonprofits actually amplify the voice of the wealthy and powerful.
Second, nonprofit advocacy is relevant to studies of environmental policy and related policy process literature. Policy process research is “the study of the interactions over time between public policy and its surrounding actors, events, and contexts, as well as the policy or policies’ outcomes” (Weible 2014, 5). Virtually all major policy process theories, such as the Multiple Streams Framework (MSF), the Punctuated Equilibrium Theory (PET), the Advocacy Coalition Framework (ACF), and the Narrative Policy Framework (NPF), contemplate a role for nonprofits (Fyall 2017; Weible and Sabatier 2018). For example, in the MSF, nonprofits may participate in policy communities or act as entrepreneurs (Mason 2014), and in the ACF, nonprofits participate in coalitions to influence change (Fyall and McGuire 2015). While these studies are informative about advocacy, they typically do not examine the nonprofit sector
60


specifically, focusing instead on coalitions or policy entrepreneurs from across sectors (e.g., Mintrom and Norman 2009; Weible, Heikkila, deLeon, and Sabatier 2012).
Advocacy Activities
Numerous studies to date across political science, interest group, social movement, and nonprofit literatures discuss the advocacy strategies of different groups. Among the most common ways to classify advocacy activities is as inside or outside strategies (e.g., Gormley and Cymrot 2006). To align with the common definition of nonprofit advocacy, these might also be considered direct and indirect strategies. Gais and Walker (1991) describe that interest groups typically choose to focus on either “inside” strategies, which leverage relationships within government, or “outside” strategies, which attempt to indirectly influence a specific policy decision or raise awareness around an issue that may lead to policy change in the future. While scholars have since shown that many organizations focus on both types of strategies (e.g., Mosley 2011; Almog-Bar 2018), the distinction remains useful in describing advocacy activities. Specific examples of inside strategies include lobbying, drafting policies, participating in government task forces, and testifying before decision-making bodies (Gais and Walker 1991; Mosley 2011). Examples of outside strategies include talking to the media, protesting, demonstrating, participating in coalitions, educating the public through speakers and events, and issuing reports (Gais and Walker 1991; Mosley 2011).
Perhaps recognizing the difficulties with fitting strategies into these two categories, some scholars have modified or expanded these categories. For example, Mosley (2011) distinguishes between insider tactics, indirect tactics, and outsider tactics, where insider tactics are those that involve “working directly with policy makers and other institutional elites” and indirect tactics include various activities that do not require connections with policy elites (439-440). She
61


describes outsider tactics, such as protests and boycotts, as a subset of indirect tactics. Furthermore, Gormley and Cymrot (2006) rely on the inside-outside strategies concept, but then also include categories that fall between these two typologies, such as coalition work and public policy research.
Others have gone completely beyond the inside-outside dichotomy and developed more nuanced advocacy typologies. For example, Pekkanen and Smith (2014a) distinguish between lobbying, policy advocacy, protesting, and public education. Berry and Arons (2003) describe nine major activities that 501(c)(3) organizations engage in, including more legislative and aggressive tactics like testifying at hearings or lobbying, and more administrative and less aggressive tactics like meeting with government officials or working in planning or advisory groups. Also, Casey (2011) outlines 28 different types of advocacy in eight categories: legal, legislative and administrative, research and policy analysis, coalition building and capacity development, education and mobilization, communication and media outreach, government relations and oversight, and service delivery (see also Schlozman and Tierney 1986; Almog-Bar 2018, 414).
Many of the policy process theories described above also suggest advocacy activities that actors, including nonprofits, may engage in to influence policymaking. For example, Brouwer and Huitema (2018) describe ten advocacy activities across four categories that policy entrepreneurs use to influence water management in the Netherlands. These include attention and support seeking strategies, like demonstrations, rhetorical persuasion, and exploitation of focusing events; linking strategies, such as coalition building, issue linking, and game linking; relational management strategies, including networking and trust building; and arena strategies, such as venue shopping and timing. Also, Gen and Wright (2018) describe how advocacy
62


activities, like motivating the public, lobbying, and building coalitions, are consistent with various policy theories.
Furthermore, a few studies suggest activities specific to hydraulic fracturing policymaking, which is the focus of this study. Weible and Heikkila (2016) examine the activities of policy actors from across sectors involved in hydraulic fracturing conflicts in New York, Colorado, and Texas. In New York, they asked survey respondents about the importance of engaging in a list of activities, and in Colorado and Texas, they asked about the frequency with which actors participate in each activity. Results suggest that across all three states, fracking opponents rated activities higher (either in terms of importance or frequency of engagement) than did fracking proponents. Among the more important or frequently used activities are forming and maintaining coalitions with allies, participating in or organizing public meetings, posting information online, and communicating with the news media. Other studies focus on the importance of discourse and competing framing efforts as central to hydraulic fracturing politics (see Dodge 2017; Dodge and Lee 2017).
Each of the studies described will inform this research by providing useful starting points and prompts for understanding advocacy in the nonprofit context. The studies above, however, typically focus on policy actors or coalitions rather than on nonprofits in particular. Also, while scholars use a variety of tools to determine whether advocacy is present, ultimately studies typically reduce advocacy to a single dichotomous variable reflecting whether organizations advocate at all (e.g., Pekkanen and Smith 2014b; Child and Gronbjerg 2007). Thus, this study zooms in on different types of advocacy across venues in the specific context of environmental nonprofits and environmental policymaking.
63


This paper argues that to understand advocacy and its impact on public policy, we need to improve our understanding of how organizations advocate in specific venues. As Buffardi, Pekkanen, and Smith (2015) point out, while much is known about the types of organizations that advocate and the activities that they engage in to influence policy, little is known about the venues in which advocacy occurs. In studies to date, activities tend to not be grounded in a particular venue or discussed with a particular policy type (e.g., agency rule, legislation, ballot measure), but studies often report the frequency with which organizations report engaging in different activities (e.g., Buffardi, Pekkanen, and Smith 2017). For these results to be more meaningful, activities should be linked with particular policy types or venues. That is, perhaps frequencies are dictated more by policy prevalence than by organization preference or competence. Furthermore, certain activities are necessarily linked to pending polices, such as testifying before legislatures, while others activities may be linked to pending policy or to broader attempts to influence public thinking. By breaking advocacy down into component measures and making it venue-specific, studies will benefit from more valid variables.
Factors That Influence Advocacy
As described in Chapter 1, to determine what drives organizations to engage in different types of activities, scholars most commonly depend on institutional theory and resource dependency theory (Mosley 2011). According to institutional theory, in order to survive, organizations must conform to the rules, norms, and social expectations that make up their institutional environments (Guo and Acar 2005; see also Meyer and Rowan 1977). As such, the innerworkings of nonprofit organizations must be understood in the context of these environments (Feeney 1997). Institutional theory has been widely applied in nonprofit studies and used to explain decisions about advocacy (e.g., Schmid, Bar, and Nirel 2008; Suarez and
64


Hwang 2008). Next, resource dependency theory suggests that organizations require a variety of resources to survive, and they therefore interact strategically in their environments to secure these resources (Eikenberry and Kluver 2004; see also Pfeffer and Salancik 1978). Resource dependency theory is also regularly invoked in studies about why and how nonprofits advocate (e.g., Nicholson-Crotty 2009; Mosley 2010; Maclndoe 2014; Fyall 2016).
Building on these theories, studies have identified a number of internal and external variables that influence advocacy across nonprofit organizations or interest groups. For example, according to Gais and Walker (1991), choice of political strategies are a function of four factors: 1) degree of conflict in the political environment; 2) groups’ internal organizational resources; 3) membership characteristics; and 4) principal sources of funding (111). Also, Mosley (2011) includes a series of variables in three categories: institutionalization, dependence on government funding, and capacity. Likewise, Guo and Saxton (2010) found a relationship between certain aspects of constituent representation and the scope and intensity of advocacy among a general sample of nonprofit organizations in Arizona.
While much can be gleaned from existing research, these studies do not examine the specific factors that influence environmental nonprofit advocacy, and there are differences between the environmental nonprofits and other types of nonprofits. For example, many nonprofits often focus exclusively on advocacy; nonprofits advocate across a wide-range of venues and issues; and nonprofits face opponents with very deep pockets, such as oil and gas companies and electric utilities. Thus, this study builds on existing research by exploring the factors that influence advocacy across venues specifically in the environmental policy context.
65


Study Context: Oil and Gas Development
To improve our understanding of the advocacy activities that environmental nonprofits use across venues, this study relies on interviews with organizations active in hydraulic fracturing (fracking) policy debates in the US. Fracking is a technique involving the injection of water, sand, and chemical additives into shale rock or other porous formations thousands of feet beneath the earth’s surface to release oil or natural gas (Heikkila et al. 2014). In the US and around much of the world, fracking is increasingly used in oil and gas development. For example, between 2007 and 2017, the amount of natural gas produced from shale formations increased 10 fold from 1,990,145 million cubic feet to 19,018,457 million cubic feet (EIA 2018).
Not only has the volume of oil and gas produced through fracking increased, but oil and gas development is also encroaching on high population density areas. Close proximity between population centers and oil and gas operations gives rise to a host of local issues, including NIMBYism (not in my backyard); nuisance issues, like traffic, dust, and noise; and concerns about air and water contamination. As a result, communities grapple with how to manage fracking, and some have attempted to ban it. For example, while the state of New York successfully banned fracking, local bans in a number of municipalities like Denton, Texas, and Longmont, Colorado, have been struck down by courts or legislatures (Knight and Gullman 2015; Heikkila and Weible 2016; Jaquith 2017). Local communities with limited authority over oil and gas development may attempt to exercise control by negotiating agreements with oil and gas companies (Zilliox and Smith 2017), regulating nuisance issues, and advocating for increased setbacks and other limits at the state level. Beyond local issues, fracking is part of broader national and global debates about reliance on fossil fuels as energy sources. While those in favor of fracking often cite natural gas as a cleaner alternative to oil and coal, those opposed
66


claim that fracking and the availability of cheap oil and natural gas impede transitions to renewable energy sources, like wind and solar.
Because they involve a wide range of issues across venues and levels of government, fracking debates provide a good context for this study. Fracking is an intensely debated political issue (Heikkila and Weible 2016), drawing participation from industry, government, and nonprofit organizations. Fracking policies are created via judicial decisions, legislation, agency rules, executive orders, and ballot measures, and organizations, including environmental nonprofits, engage in a variety of activities and strategies to influence each of the sources of policy. As such, fracking is an ideal policy context for observing maximum potential advocacy activities.
Organizations selected for this study are environmental nonprofits operating in three states: Colorado, New Mexico, and Oklahoma. These three states were selected because each is ranked in the top 10 for both shale gas and crude oil production, and all three have mid-sized populations, which facilitate understanding policy state-wide. The three states also have differences, though, that provide useful variation and facilitate drawing inferences about the contextual factors that may influence decisions about advocacy. That is, while the issues surrounding oil and gas, including local government authority and air and water pollution, are similar across states, contextual factors lead to differences in venues and in strategies. For example, Oklahoma is a Republican stronghold state, while both New Mexico and Colorado have been leaning Democrat in recent years. Furthermore, while Oklahoma has very little public land, both Colorado and New Mexico are comprised of about 35% federal public land. Oil and gas development in Colorado, however, typically occurs on private land and near population centers, while development in New Mexico is often on public land in more rural areas.
67


Furthermore, while both Colorado and Oklahoma allow citizen initiated ballot measures to make statutory and constitutional changes, New Mexico only allows referenda. A comparison of the three states is provided in Table 6.
Table 6. Comparison of State Characteristics.
Colorado New Mexico Oklahoma
shale formation Niobrara B arnett-W oodford, Lewis Woodford
2019 governor Democrat Democrat Republican
2019 state house Democrat Democrat Republican
2019 state senate Democrat Democrat Republican
2018 governor Democrat Republican Republican
2018 house Democrat Democrat Republican
2018 senate Republican Democrat Republican
population size (millions) 5.61 2.09 3.93
% federal land 35.90% 34.70% 1.60%
allow ballot measures? initiated statutes and amendments referendum only initiated statutes and amendments
shale gas production (billion cubic feet in 2017) 97 592 1290
shale gas state rank 8 7 4
crude oil production (1000 barrels in 2017) 130732 171440 161678
crude oil state rank 7 5 6
Methods
This study relies on qualitative interview data to complement existing studies on nonprofit advocacy activities, which typically rely on survey data (e.g., Bass et al. 2007; Berry and Arons 2003). As Mosley (2011) describes, among the limitations of survey data is that categories are established at the outset and thus necessarily limited to a subset of possible advocacy tactics. By diving deeper into advocacy activities through interviews, respondents are able to describe the full range of activities they engage in across venues.
68


Data were collected from November 2018 through May 2019 via in person, Zoom, and phone interviews with representatives of organizations active in hydraulic fracturing policy debates in Colorado, New Mexico, and Oklahoma. For each organization, interviews were with a senior executive staff member or with the staff member most responsible for oil and gas policy. This paper presents the results of 22 interviews: 9 in Colorado, 8 in New Mexico, and 5 in Oklahoma. Interviewees were selected using snowball and purposive sampling methodologies to identify those organizations most actively engaged in oil and gas policy each of the three states.
Interviews were semi-structured. Respondents were asked to identify roughly the last three oil and gas policy issues in their state that their organization was involved in,9 and were then asked a series of questions about their advocacy activities related to each. The interview questions relevant to this study are included in Appendix D. With interviewee’s consent, the interviews were recorded and transcribed. Interviews were then coded using thematic analysis, which focuses on identifying themes that represent data (Nishishiba, Jones, and Kraner 2014). Transcripts were coded by venue (regulatory, legislative, judicial, ballot measure, governor’s office) and then activities within each venue were identified. Through an iterative process, I identified major themes among activities and then sorted activities by theme. Results are presented in the next section.
Results
This section presents the results of interviews and the advocacy activities that respondents report engaging in across venues. Table 7 below shows an overview of the results from the coding process described above. Once activities were sorted by venue and theme, the
9 In some cases, respondents also described broader energy related issues or issues tangential to oil and gas policy.
69


Table reflects whether each major theme was identified in each venue. An “X” indicates that at least one interviewee mentioned using activities in the relevant theme-venue. The subsections that follow provide more detailed results for each venue, and the final code form reflecting the interviews is available in Appendix E. Overall, the results present a snapshot of recent advocacy activities that organizations report using. Before presenting detailed results by venue, this section describes each category of activity and overarching trends.
Table 7. Advocacy Strategy Overview.
Regulatory
Agencies Legislature Courts Ballot Box Governor’s
Type Major Theme (19) (16) (8) (5) Office (4)
Inside Hybrid Outside Informal Meetings X X X
Formal Participation X X X
Policy Prep X X X
Coalition Work X X X X
Litigation X X
Research X
Member Communication X X X X X
Media X X X X X
Institutional Change X X
Calls for Action X X X X
Official Position X X X
Public Efforts X X X X X
Note: X indicates empirical support for activities used in the relevant theme-venue. The numbers in parentheses in the first row reflect the number of respondents that reported engaging in each venue.
Consistent with recent literature about nonprofit advocacy activities, this paper uses inside and outside strategies as umbrella categories, and also includes a “hybrid” category to reflect the reality that some strategies are difficult to categorize as either inside or outside.
Within each of these three umbrella categories, a few subcategories based on interview responses are presented. Beginning with inside strategies, again, activities in this category depend on
70


direct access to decision makers, typically either through existing relationships or status. Inside strategies can be broken down into Informal Meetings with decision makers or their staff and Formal Participation. The latter refers to situations in which organizations or individuals are invited to participate in formal groups or proceedings, such as agency rulemakings or governors’ task forces.
Outside strategies, on the other hand, include a more diverse group of activities.
Member communication refers to activities that organizations engage in in relation to their members, however defined. Media includes host of activities, typically divided into three categories: earned media, such as talking to reporters or submitting op-eds; social media; and paid media. Institutional change relates to efforts to change the decision-making authority related to a particular policy or venue. Organizations also often make calls for action from others, such as the general public or local leaders. At times, organizations not involved directly in policy activities may opt to take an official position on a measure, and finally public efforts include a wide range of activities that occur in the public sphere, such as protesting, demonstrating, rallying, and hosting or participating in educational events.
Finally, the hybrid category includes a series of activities that do not squarely fall within concepts of inside or outside strategies. These include policy preparation, which may occur at the request of decision makers or outside of decision-making venues, often through the work of coalitions. In addition, a host of coalition work may occur in relation to policies, and while some members of a coalition may have direct access to decision makers, others do not, and thus, coalition members coordinate activities and determine how best to leverage combined resources to influence policy. The literature has alternatively described litigation as both an internal and an external strategy. On the one hand, those dissatisfied with a given venue may use courts as an
71


external resource to influence the primary venue. On the other hand, though, participating in judicial activities require specialized knowledge and skill, and courts may set policy independently of other venues, such as by interpreting other laws. Finally, research has also been characterized as both an internal and an external strategy, depending at least in part on who commissions the research and who is the target audience.
As Table 7 suggests, regulatory activity was most commonly discussed by respondents in this study and respondents report using the broadest range of activities to influence policy in this venue. Collectively, though, respondents report activity across venues, and while some common patterns exist, there are also differences across venues. Each of the specific venues is discussed in turn.
Venue: Regulatory
First, during interviews, respondents discussed activities in regulatory agencies most frequently. Not only was this the most common venue discussed with 19 out of 22 respondents across all three states mentioning activity in this category, but respondents also mentioned the broadest range of agency policies and activities here. For example, in Oklahoma, one respondent mentioned his organization’s work at state agencies on seismicity and produced water issues. In New Mexico, respondents mentioned getting involved in rulemaking at the BLM and regulatory proceedings at the state Public Regulation Commission (NMPRC). Finally, in Colorado, respondents mentioned a variety of agency policies, primarily at the Colorado Oil and Gas Conservation Commission (COGCC), such as the distance between oil and gas wells and school property (school setbacks), but also at other agencies, like the Air Pollution Control Division of the Colorado Department of Public Health and Environment. Activities in each category are shown below in Table 8. The third column presents a summary of the activities described in
72


each category. For all venues, more detailed results, including responses by state and organization type are included in Appendix E.
Table 8. Regulatory Agency Strategies.
Type Major Theme Summary of Activities Described in Interviews
Inside Informal Meetings Provide support to agencies; meet with regulators/agency staff; lobby
Formal Participation Participate in rulemaking; provide expert testimony/technical information
Policy Prep Participate in/convene work groups; draft guidelines; petition for rulemaking
Hybrid Coalition Work Participate in stakeholder meetings/phone calls; work with Native groups; support partners; brief other groups
Litigation Administrative litigation
Research Produce fact sheets, reports; commission research
Member Communication Email blasts; member updates
Media Earned media, social media, paid media
Institutional Change Force action through legislature, sue to force or prevent action
Outside Call for Action Encourage members to attend/speak at meetings, submit comments; request action from local leaders
Official Position Comment on rules; file formal protests
Public Efforts Help citizens file comments; host/participate in educational events and community meetings; support others' advocacy; grasstops organizing; demonstrations and rallies; publish info on website
As Table 8 suggests, respondents described abroad range of inside strategies. Most directly, respondents participated in rulemaking proceedings at various state agencies, provided expert testimony, provided technical information, and supported agencies during rulemaking. Some respondents also mentioned meeting with agency staff outside the context of rulemaking or other formal meetings.
Prior to rulemaking, not only do organizations sometimes petition agencies for rulemaking, but respondents often participate in work groups or stakeholder meetings to draft guidelines or propose rules. Within coalitions, some organizations work with Native groups or provide support to other organizations more directly involved in rulemaking proceedings.
73


Respondents also reported producing or commissioning research and reports. A more limited number of organizations engage in administrative litigation to challenge agency rules or actions.
With respect to outside strategies, organizations engage in a range of media strategies, including earned media, such as press releases, writing op-eds and letters to the editor, and talking with reporters; social media, like Facebook and Twitter; and in more limited circumstances, paid media, such as newspaper and radio ads. Organizations communicate with their members through email blasts and make calls for action from their members and local leaders. They file comments in response to agency actions and encourage and support others in doing so. Organizations also participate in a range of public activities, such as hosting community meetings and educational events, organizing grasstops, demonstrating and rallying, and publishing information on their websites.
Venue: Legislature
The majority of respondents (16 out of 22) mentioned involvement in legislative advocacy. In Colorado, the vast majority of legislation discussed was at the state level, including legislation proposed in prior legislative sessions, such efforts to amend school setbacks and forced pooling laws. Also, many interviews took place during the 2019 legislative session, and in Colorado an omnibus oil and gas bill (SB 19-181) was introduced and later passed to refocus the mission of the COGCC to prioritize healthy and safety; give local governments a stronger role in overseeing oil and gas operations; change forced pooling laws; and create financial assurances and bonding requirements for oil and gas permits to combat issues with orphan wells (Conservation Colorado 2019). Thus, a few interviewees discussed their involvement in the pending legislation. In New Mexico, interviewees most frequently talked about the federal Chaco Cultural Heritage Area Protection Act and its creation of a 10-mile no drilling buffer zone
74


around Chaco Culture National Historical Park in northern New Mexico. Fewer respondents mentioned specific legislation in Oklahoma, and at least according to one respondent, environmental nonprofits in Oklahoma often play defense against policies backed by the oil and gas industry. Advocacy activity results are shown below in Table 9.
Table 9. Legislative Strategies.
Type Major Themes Summary of Activities Described in Interviews
Inside Informal Meetings Lobby; meet with/convene legislators, staff, governor; educate legislators
Formal Participation Testify at hearings
Hybrid Policy Prep Draft legislation; offer input on policy development; convene/participate in groups to create policy proposals/recommendations
Coalition Work Meet in coalitions to discuss policies/strategies; support others
Litigation -
Research -
Member Communication Update members; email blasts; draft member petitions
Media Earned media; social media; paid media
Institutional Change Influence electoral politics; shift to regulatory venue
Outside Call for Action Encourage leaders to lobby; encourage people to call/email their legislators or the governor
Official Position -
Public Efforts Have impacted citizens testify; organize town halls/educational events; rally; collect signatures and deliver petitions; train advocates; post info online
As with regulations, advocacy activities at the legislature consisted of inside, hybrid, and outside strategies. With respect to inside and hybrid strategies, respondents frequently mentioned lobbying; informally meeting with or educating legislators; convening meetings between legislators and other stakeholders; testifying or otherwise participating in hearings; and even drafting or consulting on policy development. Respondents also often reported participating in coalitions to discuss policies, come up with policy proposals, and strategize.
Outside activities include requesting action from others such as by encouraging local leaders to lobby; having impacted members testify at hearings; asking members to email or call their legislators; and setting up meetings between citizens and legislators. Respondents reported
75


engaging in a range of media activities. A few respondents also mentioned efforts to support others’ advocacy, such as by supplying information, providing talking points, and holding training events. Some unique strategies came up with respect to institutional changes at the legislature as well. First, one respondent talked about his organization’s efforts to influence electoral politics. Frustrated by inaction at the legislature, this organization turned its attention to electoral politics and changing the composition of the legislature in hopes of creating better circumstances for future legislative sessions. Second, another respondent described a venue shift to the COGCC following inaction by the split Colorado legislature. Specifically, with respect to school setbacks in Colorado, after failing at the legislature, organizations successfully petitioned the COGCC for a rulemaking on the subject.
Venue: Courts
Courts present an interesting venue in that they are sometimes used to force action in other venues, but they also establish policy in their own right by setting precedent through judicial opinions, declaring that laws or rules are invalid, or interpreting policies. Indeed, a number of environmental nonprofits actually focus specifically on legal activity, and 8 of 22 interviewees mentioned activities in court. As compared with other venues, respondents were most clear about whether they engage injudicial activities. For example, according to one respondent, “We only do that when we’re completely without other options.” In Colorado, respondents typically talked about state-level actions. For example, in 2014, a group of teenagers, supported by the nonprofit Earth Guardians, sued the COGCC in an effort to force the agency to prioritize the environment and public health and safety when issuing oil and gas permits. Ultimately, the Colorado Supreme Court decided in the COGCC’s favor in 2018. In New Mexico, on the other hand, respondents tended to talk about the federal level, such as
76


lawsuits to block federal land leases or to force compliance with NEPA. Activities across categories are shown below in Table 10.
Table 10. Judicial Strategies.
Type Major Themes Summary of Activities Described in Interviews
Inside Informal Meetings -
Formal Participation -
Policy Prep -
Coalition Work Support partners directly involved
Hybrid Litigation File a lawsuit; serve as lead counsel; provide clients/standing declarations
Research -
Member Communication Update members
Media Earned media; social media
Outside Institutional Change -
Call for Action -
Official Position File or sign on to amicus briefs
Public Efforts Encourage people to attend oral arguments
As shown above, a number of organizations mentioned directly engaging in litigation by filing a lawsuit and serving as lead or co-lead counsel on lawsuits. While public interest law firms often take this lead role, grassroots organizations sometimes serve as parties or provide clients for the litigation. Other organizations mentioned filing or signing on to amicus curiae, or friend of the court, briefs in support of their positions. Organizations also mentioned working in coalitions and supporting those directly engaged in litigation. Outside strategies related to judicial activity are more limited than other categories. Organizations explained that they typically avoid trying to sway opinion about pending litigation and limit their activities to updating their members or the media about major events or decisions in the litigation, although one respondent discussed her organization’s efforts to have supporters attend oral arguments or other court proceedings.
77


Venue: Ballot
Ballot measures generally refer to processes through which citizens can vote directly on legislation or constitutional amendments. As mentioned above, while both Colorado and Oklahoma allow citizen initiated ballot measures to make statutory and constitutional changes, New Mexico only allows referenda through which citizens can vote to repeal a statutory measure enacted by the legislature. With respect to ballot measures, 5 out of 22 interviewees mentioned recent activity with ballot measures. Most notably, a number of respondents reported activity surrounding two ballot measures that were on the ballot in Colorado in November 2018. First, Proposition 112 would have required 2,500-foot setbacks from any structures intended for human occupancy, and second, to counter Proposition 112, the oil and gas industry backed Proposition 74. Proposition 74 would have compensated property owners for reductions in the value of their land resulting from state laws. Ultimately both measures failed, but only after over $50 million was spent on the campaigns supporting and opposing the two measures. Activities are shown below in Table 11.
Table 11. Ballot Box Strategies.
Type Major Themes Summary of Activities Described in Interviews
Inside Informal Meetings -
Formal Participation -
Policy Prep Draft proposition; collect signatures
Hybrid Coalition Work Campaign assistance
Litigation -
Research -
Member Communication Updates to members; member petitions
Media Earned media, social media, paid media
Institutional Change -
Outside Call for Action Encourage leaders to take positions; seek endorsements
Official Position Take an official position
Public Efforts Polling/focus groups; create an issue website; canvassing; phone banking; organize educational events/town halls; hand out info
78


Respondents reported a wide range of activities on ballot measures. First, ballot measures are often initiated by advocacy organizations. To this end, respondents reported drafting the language of propositions and collecting and submitting signatures. Once a measure is approved for the ballot, campaigns are typically dominated by outside strategies, because ballot measures themselves are inherently outside strategies. According to one respondent, “as a result of our futile efforts at the state, and in the court, and at our city council meetings, we decided to go ahead and utilize the direct democracy that's available to Coloradans and run a ballot initiative to write our own law.” Thus, there were no insider strategies reported with ballot measures, and outside strategies primarily consist of what Mosley (2011) describes as outsider tactics, which are a subset of indirect tactics. This includes face-to-face or in person work, like door-to-door canvassing; phone banking; holding educational events or town halls; waiving banners at intersections; and handing out fliers. Respondents also discussed a variety of media work, such as through press conferences, press releases, op-eds, letters to the editor, and social media campaigns. Ballot measures also often require that organizations, particularly those most associated with the measure, create a website in support or opposition, and use paid advertising, such as newspaper or digital media advertising and television commercials. Those not directly involved in ballot campaigns may opt to take an official position.
Venue: Governor’s Office
While most executive branch policy comes through regulatory agencies, governors themselves may establish policies through executive orders and overarching plans on specific topics, among other things. Out of the 22 interviewees, just 4 mentioned trying to influence policy through the governor’s office, and almost all activities were mentioned in relation to Colorado Governor John Hickenlooper’s oil and gas task force. As part of a negotiation to
79


withdraw two measures from the ballot in 2014, Governor Hickenlooper created a task force on State and Local Regulation of Oil and Gas Operations. The task force issued its recommendations in 2015, some of which were later debated through rulemaking and legislation. Results are shown below in Table 12.
Table 12. Governor’s Office Strategies.
Type Major Themes Summary of Activities Described in Interviews
Inside Informal Meetings Meet with Governor or staff
Formal Participation Make presentations; participate in policy groups
Policy Prep T-
Hybrid Coalition Work -
Litigation
Research -
Member Communication Member updates
Media Earned media; social media
Outside Institutional Change F-
Call for Action Encourage members to talk to the media
Official Position T-
Public Efforts Educational events; attend/comment at public meetings; rally
With respect to activities in this venue, one organization mentioned having close connections within the governor’s office and, thus, engaging periodically in informal conversations about policies or activities coming out of the governor’s office. As with other types of policies, organizations mentioned hosting events, both while policies are pending and after they have been released to help citizens understand the implications of policies.
Particularly with respect to the Colorado task force, organizations mentioned speaking at task force meetings, providing comments, and rallying. Also, like with other policies, organizations mentioned engaging with their members through email, newsletters, and mailers, and they mentioned a host of media work, including earned media, such as op-eds and letters to the editor, and social media.
80


Venue Nonspecific
In addition to the activities relevant to particular venues described above, organizations are sometimes driven more by overarching issues than by a particular policy proposal. For example, as demonstrated in Chapter 2, advocacy group participation is sometimes driven by the specific sub-issues at play in a given policy subsystem. As such, organizations often engage in advocacy efforts tied to particular issues regardless of whether a policy proposal is pending in a particular jurisdiction. Based on interviews with respondents, most of this work occurs in the public sphere. For example, organizations may periodically engage in direct action, such as through rallies or demonstrations. They may attempt to educate the public or particular groups, such as through events, meetings, and information provision. They may engage their membership or broader constituent groups through social media, and some groups periodically meet with decision makers.
Furthermore, many organizations regularly engage in coalitions related to specific issues that span a variety of venues. For example, a coalition of Native groups and environmental nonprofit organizations meet regularly to discuss strategies and issues related to the Greater Chaco Region in northern New Mexico. The Greater Chaco Region contains sacred, archaeological, and cultural sites, and is home to numerous Native Tribes and Pueblos. Because oil and gas issues in the area span a range of venues, including the federal legislature, the BLM, the NM State Land Office, federal courts, and others, coalition members meet regularly to keep each other apprised and assign responsibility for various types of advocacy. Other organizations outside the Chaco context also described efforts to form partnerships or work with other like-minded organizations.
81


Factors That May Influence Types of Engagement
In addition to describing the types of activities across venues that organizations use to influence policy, interviews also shed light on some of the factors that affect organizations’ advocacy decisions. Existing literature suggests that inside strategies are more common and effective than outside strategies (Berry and Arons 2003; Almog-Bar 2018) and that interest groups typically rely on inside strategies, while grassroots and social movement organizations tend to use outside strategies (Mosley 2011; Crow et al. 2019). As Mosley (2011) points out, however, relationships and distinctions between group types and advocacy activities are becoming more complex, and indeed, interviewees in this study report using both inside and outside strategies. For example, in addition to lobbying at the state capitol, the most professionalized groups mobilize their memberships and post on social media, and on the other hand, even grassroots organizations leverage connections with decision makers. Thus, this section of the paper presents preliminary findings based on interviews regarding some of the factors associated with advocacy decisions. Factors are divided into those internal to organizations and those external to organizations.
Internal Factors
Organizational resources. As noted above, resource dependency theory posits that organizations interact strategically in their environments in order to secure resources, and this theory is often cited in studies regarding advocacy. Accordingly, organizations with greater resources are less concerned about acquiring resources and have more staff time and money available to advocate. Indeed, this study suggests that resources affect the types of advocacy activities in which organizations engage. Specifically, those organizations with less money and staff time available typically must limit their engagement to the most pressing issues.
82


When asked about how her organization decides where to put its efforts, one respondent stated, “It’s really based on where the fire is hottest.... Because our resources are limited, it really is based on where the fire is hot.” Similarly, another respondent described, “We just can't engage in everything, because we are a small shop, you know, we only have as much time and resources as we have.” While virtually all organizations mentioned working in coalitions, those with fewer resources emphasized both the importance of other organizations’ work. “We're very good at a few things, and we should leave it to the others to do the rest.” Also, to supplement their efforts, some organizations also focus on training other organizations and individuals to do advocacy work.
Organizational identity. Based on interviews, organizations’ identities also play a role in the choices they make about advocacy. While their missions and identities as environmental nonprofits often dictate the types of issues they get involved in, other facets of their identities may influence their choices around venues and activities.
Whether organizations identify as grassroots or community organizations influences the types of advocacy activities that they choose. That is, while more professionalized environmental nonprofits make decisions internally, lead their own efforts, and inform their members of their activities after the fact, other organizations clearly identify as community organizations and emphasize that their members take the lead in their advocacy activities. While the line between these two types of organizations is not always clear, the extent to which an organization identifies as one of these two types helps drive decisions about advocacy. As one respondent described:
The members of the organization, the impacted residents . . . who are members on the ground, they're the ones who dictate our policy and our strategy and tactics.
Our staff are there to advise and support, but we actually aren’t decision makers on our campaigns. We always have like issue committees of impacted people, of
83


members who lead our efforts. . . . And so those are the folks who are turning out
to speak, not our organizers.
Thus, at least in the case of some organizations, advocacy is more limited to outside strategies and supporting those most impacted by policies.
One the other hand, more professionalized organizations tend to make decisions internally based on their expertise and organizational mission. For example, according to one respondent, “Everything we do is driven by the mission of the organization and then the vision that we've set for each one of our programs.” As another respondent described, “It's not that we try to dictate what we believe is best to everyone, but we really do try to come up with the smartest policies we can.”
This result is also supported by findings in other studies, such as Guo and Saxton (2010), who found a link between representation and advocacy. Thus, we may expect that an organization that primarily identifies as a grassroots or community-based organization may approach advocacy differently than an organization without strong connections to its members or constituents.
Expertise. Furthermore, an organization’s expertise, based on an organization’s collective expertise or on individual staff expertise, influences their decisions about where and how to get involved in advocacy. In describing her organization’s involvement at the federal level, one respondent explained, “And it's also where our expertise is. We're really good on the federal laws: ESA, NEPA, Clean Water, Clean Air.” Expertise may originate with an organization’s founding, or it may result from path dependency in that organizations acquire their identities and skills over time.
The role of expertise is most pronounced among organizations that identify as environmental or public interest law firms or among organizations with significant legal
84


expertise. These organizations typically focus on litigation and are often involved in lawsuits or regulatory actions, particularly on federal land. These organizations may also be more engaged in rulemakings or other inside activities where their legal expertise is as an asset, and they may have more limited involvement in other types of activities. According to one respondent, “Our theory of change has been, in the past, entirely in the courts, and it's only recently that we've started organizing, so I would say the vast majority of our work is in the courts.”
IRS status. The advocacy consequences of how an organization registers for tax exemption under the IRS code are explored more fully in Chapter 4. The interviews in this study reveal few differences between organizations based on their IRS status. The majority of organizations in this study (14 of 22) are registered under section 501(c)(3), which is somewhat surprising given that 501(c)(3) organizations face some restrictions in their advocacy. Section 501(c)(4) organizations are typically considered advocacy organizations, but only four organizations in this study are registered under that subpart. Two organizations in Oklahoma are registered under section 501(c)(6) as business or professional groups, which likely reflects the political context in that state as described below, and two organizations are not registered for federal tax exemption. The only noticeable difference in advocacy between any of the above categories based on interviews is that a couple respondents mentioned the importance of their 501(c)(4) status because it allows them to engage in electoral politics. These respondents were distinguishing themselves from 501(c)(3) organizations, which are strictly prohibited from supporting or opposing political candidates.
External Factors
Political climate. First, state political environment plays a role in determining organizations’ advocacy. This point is most pronounced in Oklahoma where very few
85


environmental nonprofits are directly engaged in oil and gas policy. Not only is it difficult to identify nonprofits in Oklahoma that are active in oil and gas policy issues, but the few that are active report very little success from their activities. According to one respondent, very few of the “big greens” are represented in Oklahoma, which “is partly because of the dominance of oil and gas. It’s partly the culture. Oklahoma’s culture doesn’t really support anything resembling activism.” Rather than attempt to engage directly with oil and gas policy at the legislature or otherwise, nonprofits in Oklahoma typically report focusing their efforts on general activities where they can find common ground. “Oklahomans in general don’t like to be told what to do, and they definitely don’t want government telling them what to do. So, when you can create a safe place to learn and act toward things that are better for the community, it’s a swell of incredible people trying to do good work.”
Furthermore, Colorado provides an interesting within state comparison where organizations on the Front Range often make different decisions about policy than those on the Western Slope. Large amounts of drilling happen both on the Front Range and on the Western Slope, but the population is much larger on the Front Range, encompassing cities like Denver and Boulder. Thus, industries are more diverse on the Front Range, and local economies are less dependent on oil and gas. As one Western Slope respondent explained: “We know that it produces jobs up here that are important paying jobs, where the workers down on the Front Range are just part of the mix. Amazon hires more people on the Front Range than oil and gas. Here only maybe the school district hires more people than oil and gas does.” Thus, those on the Western Slope tend to focus on more cooperative strategies.
Level of conflict. The level of conflict surrounding a particular issue influences the types of advocacy that an organization uses to influence policy. Even within a given policy
86


subsystem, like oil and gas policy, the amount of conflict varies across time and specific policy
issue. Thus, respondents report that when particularly conflictual issues come up, they may turn
to different strategies. For example, according to one respondent describing what she meant by
“get more aggressive” in the Greater Chaco Region:
That means we try to get more press. I do a lot more writing. I do blogging. I do more outreach. We meet regularly. We will, if it looks like something bad is happening, like 3 million more acres and you know, 89 more parcels are coming up, we meet more regularly. We try to think of more aggressive ways to combat this, because BLM has changed their rules pretty much every other month. We have had to meet more regularly on how to dodge and how to become—they figured they can get rid of us, but they haven't been able to.
Also, as noted above, conflict within a given venue may cause groups to shift venues, such as by
resorting to ballot measures.
These results are consistent with scholarship on conflict and advocacy. For example, according to Gais and Walker (1991), “[t]he most important factor determining the level of political activity engaged in by interest groups is the amount of conflict they experience in their immediate political environment” (104; see also Weible and Heikkila 2017; Schattschneider 1960). Relatedly, Gormley and Cymrot (2006) found that the presence of enemies in an organization’s political environment is a particularly strong factor in triggering the use of advocacy activities.
Venues. Finally, the venues themselves or the actors’ perceptions of venues influence how organizations advocate. Most notably, organizations report different approaches to advocacy when issues are in front of courts. For example, “I have a pretty robust earned media strategy in terms of getting that information out there. We don't want to appear to be trying to influence the court at all. So just getting that out in front of the public and on social media and stuff like that, that it's happening.” According to another respondent, “And then sort of once it
87


goes into the black hole of litigation, it's sort of, you know, not much happens. So it's sort of hard to maintain much sort of engagement and interest from our members.”
Beyond courts, though, other venues or stages of policymaking also influence advocacy. For example, one respondent noted the importance of inside or more internal strategies with respect to federal legislation. “I think for the purposes of federal legislation, it makes more sense to kind of work within a group.” And considering outside strategies, another respondent said,
“In organizing, it's best to get the public involved at like comment periods or commenting.”
Thus, within a given venue or across venues, organizations may use multiple activities in conjunction with one another and as part of an overarching strategy to influence policy.
Discussion and Conclusion
As the above suggests, environmental nonprofits in Colorado, New Mexico, and Oklahoma are active across a range of venues in oil and gas or hydraulic fracturing policy. At least in recent years, respondents report being most engaged in regulatory issues, and within regulatory agencies, respondents report using the broadest range of activities to influence policy, including inside strategies; hybrid strategies, like coalition work and litigation; and outside strategies, such as media campaigns and direct action. While activity at regulatory agencies and legislatures is somewhat similar, respondents describe different approaches to other venues. For example, when issues are pending before courts, few organizations are directly involved in litigation. Other organizations typically limit their activities to supporting those directly involved and keep their members and the public apprised. With respect to the ballot box, ballot measures are often borne out of frustration with other venues, and thus, respondents report that their work falls almost exclusively in the outside category, including a range of media activity and public measures. Respondents report much more limited activity at governors’ offices,
88


Full Text

PAGE 1

ENVIRONMENTAL NONPRO FIT ADVOCACY: A THREE PAPER EXAMIN ATION OF ORGANIZATIONS , VENUES, AND ACTIVITIES by JENNIFER ANNE KAGAN B.A., New York University, 2000 J.D., University of San Diego School of Law, 2004 M.P.A., University of Colorado Denver, 2013 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Public Affairs Program 2019

PAGE 2

ii © 2019 JENNIFER ANNE KAGAN ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Doctor of Philosophy degree by Jennifer Anne Kagan has been approved for the Public Affairs Program b y Tanya Heikkila, Chair Christopher M. Weible Todd L. Ely Kelly LeRoux Date: August 3, 2019

PAGE 4

iv Kagan, Jennifer Anne (PhD, Public Affairs Program ) Environmental Nonprofit Advocacy: A Three Paper Examination of Organizations, Venues, and Activities Thesis directed by Professor Tanya Heikkila ABSTRACT This dissertation explores the advocacy activities of environmental nonprofits across venues. These organizations are understudied as a distinct type of entity across literatures. While the policy process literature recognizes that environmental nonprofits actively participate in environmental policymaking, the unique characteristics of these organizations are often overlooked. Furthermore, the nonprofit literature recognizes that environmental nonprofits are more likely to advocate than any oth er type of nonprofit, but few studies have specifically examined the characteristics and activities of these organizations. With these considerations in mind, this dissertation includes three empirical chapters. The first empirical chapter relies on news media to explore differences and similarities between industry and environmental advocacy groups with respect to the venues and issues in which they engage. The second empirical chapter in the dissertation zooms in on environmental nonprofits and explores, through interviews, the types of advocacy activities these organizations use across sectors and how the external contexts and internal characteristics of these organizations may influence their advocacy. The third empirical chapter in the dissertation examines differences between 501(c)(3) and 501(c)(4) organizations in the environmental space using survey responses combined with publicly available nonprofit data. While in service delivery sectors, 501(c)(3) organizations tend to be service providers and 501(c)(4)s are advocacy organizations,

PAGE 5

v the third empirical chapter tests whether this distinction holds true among environ mental advocacy organizations. Overall, the dissertation contributes to both the nonprofit and environmental policy literatures by suggesting that environmental nonprofits use a complex web of advocacy activities across multiple venues to influence polic y. The dissertation shows that the relationships between organizational and contextual factors, on the one hand, and activities and venues, on the other hand, are more complicated than previously assumed. Finally, the dissertation provides empirical supp ort for differences between environmental nonprofits and other types of nonprofits that warrant future studies about these organizations, their characteristics, and their activities. The form and content of this abstract are approved. I recommend its publ ication. Approved: Tanya Heikkila

PAGE 6

vi ACKNOWLEDGEMENTS I am incredibly grateful to have been a part of the School of Public Affairs at the University of Colorado Denver for the past eight and a half years, first as an MPA student, then as Director of the W irth Chair in Sustainable Development, and finally as a PhD student. Thank you in particular to Tanya Heikkila and Chris Weible. I feel so fortunate to have had the opportunity to work with you both. You not only model great scholarship and collegiality , but also how to be good, patient, understanding, and compassionate people. It has been an honor to be part of the WOPPR lab and various research teams throughout my time at SPA , and I truly appreciate all of the support and opportunities I received th rough working with you both . Thank you also to all the current and former students I have the opportunity to work with through the WOPPR lab . I truly made some lasting friendships throughout this time and am grateful for the connections, laughter, and support. Thank you to Tanya for serving as my dissertation chair. I could not have asked for a better advisor. You gave me exactly the right amount of support when I needed it and trusted me to get things done when I needed to write, code, interview, or organize my though ts. I am always amazed by how quickly you turn things around with thoughtful and helpful comments. I have learned so much from you on both personal and professional levels . Thank you to Chris and to Todd Ely for serving on my dissertation committee. Chris, I always enjoy talking with you , sharing stories, and hearing your perspectives on research. You r insights keep me honest and help me to think outside the box. Todd, I am so grateful that I took 5004 with you back in 2011, because that was really the start of my deeper engagement with SPA, including my work with the Buechner Institute , with Tanya and Chris , and with the Wirth Chair. Your thoughts and comments on my dissertation have been incredibly helpful.

PAGE 7

vii I am also grateful to Kelly LeRoux for serving as the external member of my committee. Kelly, I truly admire your work and feel very fortunate to hav e had your expertise and input o n my dissertation. It is always a pleasure to spend time with you at conferences, and I hope we remain in touch and have the opportunity to work together in the future. Many other staff and faculty members at SPA have played important roles on my journey. Thank you so much to Dawn Savage for all of your help as my academic advisor throughout my time at SPA. I can always count on you for answers to all of my questions and for fun conversations about chickens and other things . Thank you also to the rest of the staff for your friendliness and help along the way . In addition to my dissertation committee, I have been the beneficiary of much support from other faculty members . Thank you to Mary Guy for your good advice and guidance over the years and for writing opportunities, and thank you to Sebawit Bishu, Christine Martell, John Ronquillo, Will Swann, Sandy Zook, and so many others for advice, support, opp ortunities to work t ogether, and much more. I will miss you all! Thank you to my partner, Cheryl. I cannot thank you enough for your support and patience as I have gone through this crazy PhD process. I know I can always count on you for encouragement , lightness, optimism, and unconditional love . I cannot wait for our life together in Hawaii.

PAGE 8

viii TABLE OF CONTENTS CHAPTER I. INTRODUCTION ................................ ................................ ................................ ............... 1 Nonprofit Organizations and Policymaking ................................ ................................ ....... 2 Overview of Nonprofit Advocacy ................................ ................................ ...................... 9 Environmental Nonprofits ................................ ................................ ................................ 11 Overview of Study Context and Methods ................................ ................................ ......... 16 Overview of Remaining Chapters ................................ ................................ ..................... 18 II. ADVOCACY GROUPS AND VENUE SELECTION ................................ .................... 21 Overview of Venues in the Policy Literature ................................ ................................ ... 22 Venue Selection ................................ ................................ ................................ ................ 25 Advocacy Organizations ................................ ................................ ................................ ... 28 Study Context: Oil and Gas Policy Debates in Colorado ................................ ................. 30 Methods ................................ ................................ ................................ ............................. 32 Results ................................ ................................ ................................ ............................... 35 Discussion ................................ ................................ ................................ ......................... 52 Conclusion ................................ ................................ ................................ ........................ 55 III. HOW DO ENVIRONMENTAL NONPROFITS ADVOCATE ? ................................ .... 57 Literature Review ................................ ................................ ................................ .............. 58 Study Context: Oil and Gas Development ................................ ................................ ........ 66

PAGE 9

ix Methods ................................ ................................ ................................ ............................. 68 Results ................................ ................................ ................................ ............................... 69 Discussion and Conclusion ................................ ................................ ............................... 88 IV. A QUALITATIVE COMPARATIVE ANALYSIS ................................ ......................... 93 Literature Review ................................ ................................ ................................ .............. 94 Research Context ................................ ................................ ................................ ............ 102 Methods ................................ ................................ ................................ ........................... 103 Advocacy Activities of Environmental Nonprofits ................................ ........................ 105 Organizational Characteristics Associated with Advocacy ................................ ............ 107 Discussion ................................ ................................ ................................ ....................... 116 V. CONCLUSION ................................ ................................ ................................ ............... 121 Major Takeaways ................................ ................................ ................................ ............ 123 Study Limitations ................................ ................................ ................................ ............ 128 Future Research Agenda ................................ ................................ ................................ . 131 REFERENCES ................................ ................................ ................................ ..................... 136 APPENDIX A. SUMMARY OF DISSERTATION CHAPTERS ................................ ........................... 147 B. POLICY ISSUES IN NEWS ARTICLES ................................ ................................ ...... 148 C. ACTIVITY BY JURISDICTION R ELATED TO LOCAL AUTHORITY ................... 149 D. INTERVIEW PROTOCOL ................................ ................................ ............................ 152

PAGE 10

x E. INTERVIEW CODE FORMS, PARTICIPATION BY VENUE AND STATE ............ 153 F. QCA RESULTS ................................ ................................ ................................ .............. 178

PAGE 11

xi LIST OF TABLES TABLE 1. Venue Frequency in Articles. ................................ ................................ ............................ 36 2. Policy Issue by Level of Government. ................................ ................................ ................ 39 3. Articles and Advocacy Group Activity by Year. ................................ ................................ 40 4. Venues with Mean Participation by Advocacy Group (T Test Results). ........................... 41 5. Policy Issues with Mean Participation by Advocacy Group (T Test Results). .................. 42 6. Comparison of State Characteristics. ................................ ................................ .................. 68 7. Advocacy Strategy Overview. ................................ ................................ ............................ 70 8. Regulatory Agency Strategies. ................................ ................................ ........................... 73 9. Legislative Strategies. ................................ ................................ ................................ ......... 75 10. Judicial Strategies. ................................ ................................ ................................ ............ 77 11. Ball ot Box Strategies. ................................ ................................ ................................ ....... 78 ................................ ................................ ........................... 80 13. Participation in Advocacy Activities. ................................ ................................ ............. 106 14. Variable Descriptive Statistics and QCA Calibration. ................................ .................... 111 15. Providing Information to Government Officials, Truth Table. ................................ ...... 113 16. Providing Information to Government Officials, Intermediate Solution. ....................... 114 17. Collaborating with People You Disagree with, Truth Table. ................................ ......... 115 18. Collaborating with People You Disagree with, Intermediate Solution. ......................... 115

PAGE 12

xii LIST OF FIGURES FIGURE 1. Venue Frequency in Articles. ................................ ................................ ............................. 36 2. Timeline of Bans, Moratoria, and Local Authority Issues. ................................ ............... 44 3. Activity Across Venues by Year. ................................ ................................ ....................... 47 4. Activity Across Venues by Advocacy Group. ................................ ................................ .... 49 5. Local Authority Related Activity at the State Legislature. ................................ ................. 50 6. Local Authority Related Activity at the State Ballot Box. ................................ ................. 51 7. Local Authority Related Activity at the Local Ballot Box. ................................ ................ 51 8. Local Authority Related Activity at State Courts. ................................ .............................. 52

PAGE 13

1 CHAPTER I INTRODUCTION This dissertation explores the advocacy activities of environmental nonprofits with an eye toward enhancing knowledge regarding how these organizations advocate and how advocacy varies by organization and venue. While a number of literatures such as the policy process, social movement, and nonprofit management literatures con template that nonprofits play a role in environmental policymaking, the literature lacks a thorough understanding of who these organizations are and how they try to influence policy. For example, policy process and social movement literatures contemplate that both formal and informal organizations and groups play a role in environmental policymaking, yet these studies do not examine nonprofit organizations in particular. Studies specific to nonprofit organizations may have implications for nonprofit staff, board members, and donors ; however, despite recognizing that environmental nonprofits advocate more than any other type of nonprofit, the nonprofit literature lacks an understanding of these organizations as distinct from other types of nonprofits. With these considerations in mind, this dissertation includes three empirical chapters . The first empirical chapter (Chapter 2) relies on news media to explore differences and similarities between industry and environmental advocacy groups with respect to the venues and issues in which they engage . The second empirical chapter in the dissertation (Chapter 3) zooms in on environmental nonprofits and explores , through interviews , the types of advocacy activities these organizations use across sectors and how the external contexts and internal characteristics of these organizations may influence their advocacy. Finally, the third empirical chapter in the dissertation (Chapter 4) examines differences between 501(c)(3) and 501(c)(4) organizations in the environmental space using survey responses combined with publicly available nonprofit data .

PAGE 14

2 W hile in service delivery sec tors, 501(c)(3) organizations tend to be service providers and 501(c)(4) s are advocacy organizations, Chapter 4 examines whether there are differences between 501(c)(3) and 501(c)(4) environmental advocacy organizations. The final chapter of the dissertat ion synthesizes lessons learned and describes a research agenda for further studies about environmental nonprofit advocacy. Before moving to the empirical chapters of the dissertation, this first chapter lays the general foundati on for the studies that follow. This chapter begins by describing that nonprofits are actively involved in policymaking and in doing so, can play an important role in democracy . I then describe existing literature on nonprofit advocacy and provide ration ales for studying environmental nonprofits in particular. The chapter then presents an overview of the remaining chapters. Nonprofit Organizations and Policyma king Before attempting to define and operationalize advocacy based on existing literature, this section describ es why nonprofit advocacy is a worthy area of scholarship. Nonprofit organizations play roles in society both as advocates and as service provid ers. As service providers, there is no question that government reliance on nonprofits for service delivery has increased dramatically in recent years (Boris 2006; Never 2016). For example, b eginning in the 1980s and early 1990s with the New Public Manag rather than row, there has been a proliferation of government contracting with the nonprofit sector for services. In addition to their role as service providers, these organizations also often advocate for themselves and their constituents (Hula and Jackson Elmoore 2001; Minkoff 2002;

PAGE 15

3 Fyall 2017). The vast majority of nonprofits register for tax exempt status pursuant to section 50 1(c)(3) of the Internal Revenue Service ( IRS ) code. Section 501(c)(3) organizations are public charities and may only lobby up to a point and are prohibited from engaging in electoral politics. These organizations, however, often advocate via public educ ation, working in coalitions, and limited amounts of lobbying. In addition, a smaller but still significant portion of nonprofits focuses exclusively on advocacy and is not restricted in their activities. By one measure, out of 1.54 million registered no nprofits in 2004, 138,193 were 501(c)(4)s or social welfare organizations, and 86,054 were 501(c)(6)s or business leagues, both of which typically focus on advocacy (Boris 2006). Organizations that primarily advocate, including 501(c)(3)s, 501(c)(4)s, and 501(c)(6)s, are the focus of this dissertation for a couple reasons : 1) N onprofits play a role in policymaking, and 2) nonprofits have the potential to facilitate democracy. N onprofits in the Policy Process the interactions over time between public Scholarship in this area makes clear that nonprofits are typically involved at some level in policy making (Fyall 2017). For example , the Multiple Streams Framework (MSF) describes that an item comes on to a policy agenda when a problem aligns with a policy solution in the right political environment (Herweg, Zahariadis, and Zohlnhöfer 2018). Th ese three streams the problem, policy, and politics streams run relatively independently, until a policy entrepreneur couples the streams in a policy window, or at an opportune time (Herweg et al. 2018). Within the MSF, nonprofits may be part of the polic y stream, and they may act as policy entrepreneurs (Fyall 2017 ; Kagan 2019 ). In the policy stream, policy communities consisting primarily of policy experts generate ideas and policy

PAGE 16

4 alternatives (Herweg et al. 2018). Nonprofits, and particularly those t hat advocate, often participate in these communities and help generate policy alternatives that support their missions. Also, as entrepreneurs, nonprofits often advocate on behalf of specific policies to move them on to policy or decision agendas. When po licies are consistent with their missions, nonprofits often devote considerable resources to push those policies forward. Also , the Advocacy Coalition Framework (ACF) focuses on policy subsystems and the activities, beliefs, and interactions of coalitions within those subsystems (Jenkins Smith et al. Smith et al. 2018, 139). Nonprofits are among subsystem actors in that they regular ly try to influence subsystem affairs (Jenkins Smith et al. 2018), and nonprofits frequently advocate for change within coalitions (see Fyall and McGuire 2015). According to the ACF, nonprofits might work within coalitions to influence change by taking ad vantage of events either within or outside the subsystem, promoting policy oriented learning, and negotiating agreements either as a coalition member or as a policy broker (Jenkins Smith et al. 2018). Other policy process frameworks make similar claims. F or example in the Narrative Policy Framework (NPF), nonprofits may use narratives to influence policy or may be actors within narratives ( Shanahan, Jones, McBeth, and Radaelli 2018 ). In the Institutional Analysis and Development Framework (IAD), nonprofit s may be actors in action situations or they may provide the institutional frameworks within which action situations occur ( Schlager and Cox 2018 ; Wilson, Ostrom, and Cox 2013 ). Also, in Punctuated Equilibrium Theory (PET) nonprofits may effect policy cha nge by influencing issue definition, the images associated with issues, and the venues in which issues are debated and considered (Baumgartner et al. 2018).

PAGE 17

5 Relatedly, the policy process literature also recognizes that multiple venues exist within a given subsystem (Lubell 2013; Jenkins strategies to influence policy involves venue selection (Pralle 2006), or choosing to participate in venues where organizations believe they are most likely to achieve the ir policy goals. Thus, not only does the policy process literature contemplate that nonprofits use a variety of advocacy activities to influence policy, but also that they do so across a range of venues. While the role and impact of nonprofit organization involvement varies from case to case, it is clear that nonprofits actively engage across a range of topics. Even among 501(c)(3)s or charitable organizations, we see organizations engaging in advocacy related to arts and culture, the environment, health issues, civil rights, and more (Berry and Arons 2003; Urban Institute 2015 ). As such, nonprofit advocacy is an important area of inquiry. Nonprofits and D emocracy In considering the role that nonprofits play in policymaking, it is also necessary to bear i n mind that when nonprofits advocate, they do so not only to further their missions, but also to provide a collective voice for disperse d or underrepresented interests. That is, nonprofits often fill a void in policymaking by supporting citizen engagement , which is a key concern in the field of public affairs. Policy process and public administration fields are framed by, or operate within a paradigm of, democracy and involve questions regarding how to incorporate public preferences and represent citizen is one of five criteria for democracy with the other four being voting equality, enlightened u nderstanding, control of the agenda, and inclusion of all adults. Thus, democratic institutions

PAGE 18

6 must be concerned with how to ensure public access to decision making venues and processes, and how to incorporate meaningfully public preferences in policy an d administration. In considering these concerns and the potential role that nonprofits play in facilitating democracy, a distinction to bear in mind is that between representative democracy and direct democracy (Box 2009). While many extoll the virtues of direct democracy (e.g., Follett 1918; Roberts 2004), inviting direct participation from an entire polity is often impractical, particularly when meaningful participation is desired. Direct democracy may be easier in smaller units of government, but small er units are generally less equipped to handle large scale problems, such as environmental issues. That is, the expansive, transboundary nature of many environmental issues necessitates the involvement of larger democratic units, such as state, national, and even global bodies. Within these larger units, representative democracy is often invoked. Representative systems, however, also often fall short of facilitating adequate representation (Dahl 2015). First, representatives may have more constituents th an they can reasonably connect with in a meaningful way. Second, larger units may be plagued by issues of inequality, which can undermine democracy in a number of ways. According to Shapiro (2015), electoral politics are heavily influenced by money; thos e with limited resources are less likely to participate; and, even if those with limited resources do participate, the complexity of political systems makes stasis the norm and change difficult to effect. Nonprofit organizations help mitigate these issues by representing citizen voices and serving as intermediaries between the public and government (see LeRoux 2007, 2009). Indeed, one of the primary purposes of nonprofits is to provide a voice for those whose interests might not otherwise be represented ( Child and Grønbjerg 2007; Almog Bar and Schmid 2014; Mosley

PAGE 19

7 2016). In the 1830s, Alexis de Tocqueville, observing the US, noted that associations are essential in democratic societies to overcome the fact that independent citizens otherwise lack political power (de Tocqueville 1966). More recently, according to Putnam (1993), voluntary associations help build social capital, and nonprofits help secure citizen rights, including political, civil, and social rights (see also Smith 2007). According to Child and Grønbjerg in advocacy activities and pressure governments to take some desired action (260; see also LeRoux 2007). This is particularly true with respect to environmental issues like climate change, which is an intractable problem , involving int erdependent networks and requiring tremendous they may provide an accessible forum in which members of the public can participate, and 2) they can help restore equilibrium when institutions beco me unbalanced , for example, when an external shock causes an agency to shift resources aw ay from an essential service or issue (see Truman 1951). A number of scholars have outlined the specific ways in which nonprofits contribute to democracy. In perhaps the most comprehensive description, Fung (2003) describes six ways that associations , including nonprofits, contribute to democracy. 1 First, Fung (2003) argues that associations are inherently good in that they support freedom and democracy. Second, 1 While the concept of association is not well defined, Warren (2001) describes that references to to further a cause, form a family, play a sport, work through a probl em of identity or meaning, get typically conceptualized as a subset of associations (see Dodge and Ospina 2016), and while the purpose here is not to argue that nonpro fits necessarily perform all of the functions listed below, the overlap between nonprofits and associations suggests that the ways in which associations may contribute to democracy are also ways in which nonprofits may similarly contribute.

PAGE 20

8 ass ociations help develop the attitudes, skills, and behaviors of their members or broader citizen groups that support civic engagement. Third, associations can serve a resistance and power checking function by counterbalancing government or other special in terests when necessary. Fourth, associations provide an additional means for citizen interests to be represented in government. Fifth, associations create space for and facilitate public deliberation. Sixth, and finally, associations are increasingly en gaged in direct governance, such as by participating in policymaking and service provision. Other scholars have suggested similar ways that nonprofits contribute to democracy, such as by promoting direct citizen involvement in government through education , capacity building, and issue framing; providing a collective voice for otherwise underrepresented interests; and resisting or providing alternative political venues when formal institutions break down (Warren 2003; LeRoux 2009; Dodge and Ospina 2016). As the above suggests, nonprofits certainly have the potential to improve opportunities for participation and democracy. At the same time, though, nonprofits can also undermine democracy (see Reid 2006) . For example, rather than strengthen the voice of tho se lacking political power, many nonprofits actually amplify the voice of the wealthy and powerful. While nonprofits may serve to aggregate voices, the most disadvantaged communities may not have access to nonprofits (Boris 2006). As described more fully below, this is among the concerns that gave rise to the environmental justice movement. For example, nonprofits can disproportionately influence public policy at times, and it is not always clear who is funding nonprofit efforts (Boris 2006). Furthermor e, financial stress and budget pressures may lead organizations to represent funders rather than public interests (Reid 2006). For these reasons, developing our understanding of environmental nonprofits, including how they try to influence policy and how contextual and organizational factors may influence

PAGE 21

9 policy, may aid our understanding of the democratic role that environmental nonprofits play. This dissertation lays the foundation for this line of research in the environmental nonprofit and environmen tal policy space. Overview of Nonprofit Advocacy With an understanding of the relevance of nonprofit advocacy, this section describes specifically what nonprofit advocacy refers to and how scholars to date have attempted to operationalize the concept. Whi le studies about the advocacy activities 2 of nonprofit organizations are not new, perhaps recognizing the growing importance of nonprofit advocacy to counterbalance well resourced corporate interests, recent years have seen a surge in nonprofit advocacy re search. For example, recent scholarly work has focused on the advocacy activities of human service nonprofits and the drivers of those activities (Mosley 2010; Almog Bar and Schmid 2014), how nonprofits influence policy and effect policy change (Hula and Jackson Elmoore 2001; Andrews and Edwards 2004; Bryce 2006; Fyall and McGuire 2015; Fyall 2016), and ways of understanding and classifying nonprofit advocacy organizations and their activities (Boris and Mosher Williams 1998; Child and Grønbjerg 2007). I Beyond agre ement on the definition, though , advocacy as a concept remains difficult to operationalize . For example, in their definition of advocacy, Pekkanen and Smith (2014a) include lobbying, policy advocacy, protesting, and public education. Nicholson Crotty (2009), on the other hand, describes three common groups of activities, including affiliating with a 501(c)(4) organization, grassroots 2 Unless othe rwise noted in the context, this dissertation uses the terms advocacy activities and advocacy strategy interchangeably to refer to specific activities, such as lobbying, testifying, and mobilizing the public, that organizations use to try to influence the policy process.

PAGE 22

10 activities, and direct lobbying. A few other typologies exist as well, as described more fully in Chapter 3 (e.g., Berry and Arons 2003; Casey 2011), but there is little consistency between the typologies , and ma ny studies reduce advocacy to a dichotomous variable reflecting whether organizations engage in any type of advocacy. Furthermore, few nonprofits studies to date have incorporated differences in venues into their understandings of advocacy activities. Des pite these issues, though, knowledge regarding nonprofit advocacy has certainly increased over the past couple decades. For example, in a study of 501(c)(3) organizations in Indiana, Child and Gronbjerg (2007) demonstrated that at least some nonprofits ac ross fields advocate, and that environmental and health organizations are more likely than other nonprofits to do so. Berry and Arons (2003) found similar trends looking at a random sample of 501(c)(3) organizations across the US. These authors also demo nstrated among the same group of nonprofits, that organizations are most likely (among a list of activities) to provide information to government officials, work in planning and advisory groups, and meet with government officials. They are least likely to testify at legislative or administrative hearings (see also Bass 2007). Scholars have also begun to elucidate the motivations and factors associated with advocacy. Berry and Arons (2003) found that primary reasons for nonprofit advocacy include protectin g or promoting programs that benefit constituencies, and raising awareness about issues. Furthermore, primarily in reliance on resource dependency theory and institutional theory, scholars have identified a number of factors associated with advocacy. R es ource dependency theory suggests that organizations require a variety of resources to survive, and they therefore interact strategically in their environments to secure these resources (Eikenberry and Kluver 2004; see also Pfeffer and Salancik 1978). Reso urce dependency theory consistently suggests

PAGE 23

11 that organizational budgets, staff size, and volunteer resources are all associated with higher likelihoods of advocating (e.g., Mosley 2011). According to institutional theory, in order to survive, organizatio ns must conform to the rules, norms, and social expectations that make up their institutional environments (Guo and Acar 2005; see also Meyer and Rowan 1977). Based on institutional theory, we know that factors like institutionalization, professionalizati on, collaboration, and political environment are associated with advocacy (Mosley 2011; Nicholson Crotty 2009). Finally, recognizing the challenges associated with advocacy, scholars have identified major barriers to advocacy. The most significant of thes e is limited capacity (Bass 2007; Mosley 2014). With respect to capacity issues, organizations, particularly social service providers, often lack the time and money needed to participate effectively in advocacy. Similarly, 501(c)(3) organizations often l ack staff with advocacy expertise, making it more difficult for them to engage effectively (Bass 2007; Mosley 2014). Furthermore, while scholars often cite confusion about tax laws as a challenge with nonprofit advocacy (Bass 2007; Berry and Arons 2003), Mosley (2014) found that while organizations are often unclear about what they may legally do, this does not actually prevent them from advocating. Thus, while scholarship is certainly growing regarding nonprofit advocacy, the majority of studies are limit ed to 501(c)(3) organizations or social service providers. Thus, the next section describes the rationales for expanding studies of nonprofit advocacy to the environmental field. Environmental N onprofits This dissertation focuses specifically on environmental nonprofits for a few reasons. As described more fully below, environmental nonprofit organizations as a specific group are

PAGE 24

12 understudied across literatures. Also, environmental nonprofits have unique implications for society in that they can facilitate stakeholder participation in environmental policymaking, yet they also have a history of promoting social inequities. The rationales for studying environmental nonprofits as a unique subgroup of nonpro fits are each discussed in turn. First, environmental nonprofits are understudied. A limitation of existing studies on nonprofit advocacy is that they typically either use broad strokes to describe advocacy by all types of nonprofits (e.g., Berry and Aron s 2003; Pekkanen and Smith 2014b), or they focus specifically on social service providers (e.g., Almog Bar and Schmid 2014; Fyall 2016). There are certainly good reasons to focus on social service providers. For example, health and human service nonprofi ts are far more prevalent than other types of nonprofits (Berry and Arons 2003). Also, as Mosley (2014) describes in a study of homeless service providers, these nonprofits provide a critical voice for marginalized communities, including resource constrai nts, lack of experience with advocacy, and confusion about the legal limits of 501(c)(3) advocacy. Studies exploring the dual role of advocate and provider (see Minkoff 2002; Fyall 2017) among social service organizations remain worthy of inquiry , and thr ough expanding studies of nonprofit advocacy into other substantive areas, including environmental policy contexts, we may learn more about how advocacy occurs in specific contexts as distinct from other contexts. While a few studies to date have differen tiated among organizations based on their policy contexts (e.g., Berry and Arons 2003; Child and Grønbjerg 2007), those that do focus on rates of advocacy, rather than on the specific ways that organizations advocate or the specific constraints that affect different types of nonprofits. The dearth of studies regarding environmental nonprofit advocacy is particularly surprising given the fact that these organizations are more likely than any other type of nonprofit

PAGE 25

13 to engage in advocacy (Child and Grønbjerg 2007). These organizations are also different from other types of nonprofits in that they typically lack a clear constituent base , provide non excludable benefits, and are often directly opposed by corporations with very deep pockets. Thus, it is necessa ry to explore the advocacy activities of environmental nonprofits as distinct from other types of nonprofits, and such an exploration also allows for a deeper examination of nonprofit advocacy activities broadly. That is, while studies to date typically l ook at whether nonprofits advocate, through examining a population of nonprofits that advocate, we can gain a better understanding of different types of activities and the factors that affect organizational decisions about which activities to use to influe nce policy. Furthermore, the importance of environmental policy cannot be overstated. With abundant and significant global issues, including air pollution, ocean pollution, water contamination, and climate change and its effects, environmental policymak ing is among the problems in that they are unstructured, cross cutting, and relentless, and thus cannot be addressed through traditional problem solving syste ms (Weber and Khademian 2008). In response to these issues, there is no question that the environmental movement has grown exponentially by every conceivable measure since WWII (Baumgartner and Jones 2009 ; see also Boris 2006 ). While t he environmental mo vement is certainly studied in other literatures for example, nonprofit organizations may be part of social movements ( Reid 2006 ) examining environmental nonprofits in particular holds value for nonprofit stakeholders. That is, understanding environmental nonprofits will help inform their staff members about best practices, opportunities, and constraints; help donors who give money and place trust in these organizations

PAGE 26

14 better understand the sector ; and inform society at large, which confers benefits on th ese organizations such as through tax exemption , about the benefits and drawbacks to the sector . The second rationale for studying environmental nonprofits as a distinct group is that stakeholder participation in environmental policymaking and governance i s particularly essential outcomes, empower communities, reduce u ncertainties and unexpected resistance, and enhance the knowledge available for assessing and solving environmental problems (Irvin and Stansbury 2004; Pahl Wostl 2009). For these reasons and because of difficulties with governing at the global level, rec ent years have seen a trend toward decentralization and increased emphases on participation (Lemos and Agrawal 2006). This trend is facilitated further by empirical evidence that self governance and co management are viable solutions to environmental degr adation (Ostrom 1990; Lemos and Agrawal 2006). Notably, the work of Elinor Ostrom and others suggest that communities can effectively manage and conserve their resources when certain conditions are present, including that those affected by the rules gover ning a resource can participate in modifying them (Cox, Arnold, and Tomás 2010). When small resources like village fisheries are at stake, individual participation may be possible, but when larger transboundary resources are involved, stakeholders are agg regated to the agency or organizational level (see Heikkila and Gerlak 2005). Thus, these agencies or organizations, which often include nonprofits, should appropriately represent constituent interests. In addition to scholarship on public participation i n environmental governance, policies across all levels of government emphasize the need for public participation in environmental policymaking. For example, Principle 10 of the Rio Declaration on Environment and

PAGE 27

15 Development, drafted at the 1992 United Nat ions Conference on Environment and Development, states the following: Environmental issues are best handled with the participation of all concerned citizens, at the relevant level. At the national level, each individual shall have appropriate access to in formation concerning the environment that is held by public authorities, including information on hazardous materials and activities in their communities, and the opportunity to participate in decision making processes (UN 1992). In the US, among other po licies, the National Environmental Policy Act (NEPA) requires federal agencies undertaking projects that significantly affect the environment to provide meaningful opportunities for public participation. Similar initiatives to promote citizen engagement i n state and local policymaking exist as well. Thus, understanding and developing environmental nonprofit capacities and strategies can help further objectives related to stakeholder engagement. Third, environmental justice (EJ) concerns demand that minori ty and otherwise underrepresented groups be given a voice in environmental policymaking. According to the involvement of all people regardless of race, color, national origin, or income, with respect to the development, implementation, and enforcement of environmental laws, regulations, and much with processes as it is with outcom es , emphasizing democratic decision making, public participation, and community self determination (Cole and Foster 2001). The EJ movement began in the late 1970s and early 1980s in response to concerns about discriminatory government and industry policies , and perceptions that existing environmental nonprofits did not represent poor and minority interests (Bullard et al. 2008). Although 40 years have passed, t here is no question that environmental injustices persist today. Stories of environmental injust ices often grab headlines, as in the cases of the Flint, Michigan water crisis

PAGE 28

16 and the Dakota Access Pipeline, and studies consistently show that racial minority and low income communities still disproportionately suffer from environmental risks (e.g., Bul lard et al. 2008; NAACP 2017). S tudies to date tend to focus on the substanti ve side of EJ by focusing on the relationship between community demographics and environmental injustices , and on the drivers of those injustices (Campbell, Kim, and Eckerd 2014) . G overnment policies , however, often focus on the procedural side of EJ and demand that government programs and agencies incorporate EJ and community input (e.g., EPA 2011, 2014) . Thus, the literature would benefit from studies about the role that envir onmental nonprofits play, both in influencing policy and in enhancing the voice of underserved communities. Overview of Study Context and M ethods To explore the role of nonprofits in environmental policymaking, this study examines oil and gas, or hydraulic Fracking is a technique involving the injection of water, sand, and chemical additives into shale rock or other porous formations . 2014). In the US and around much of the world, fracking is increasingly used in oil and gas development. For example, between 2007 and 2017, the amount of natural gas produced from shale formations increased 10 fold from 1,990,145 million cubic feet to 19,018,457 million cubic feet (EIA 2018). Not only has the volume of oil and gas produced through fracking increased, but oil and gas development is also encroaching on high population density areas. Close proximity between population centers and oil a nd gas operations gives rise to a host of local issues, including NIMBYism (not in my backyard); nuisance issues, like traffic, dust, and noise; and concerns about air and water contamination. As a result, communities grapple with how to manage

PAGE 29

17 fracking, and some have attempted to ban it. For example, while the state of New York successfully banned fracking, local bans in a number of municipalities like Denton, Texas, and Longmont, Colorado, have been struck down by courts or legislatures (Knight and Gull man 2015; Heikkila and Weible 2016; Jaquith 2017). Local communities with limited authority over oil and gas development may attempt to exercise control by negotiating agreements with oil and gas companies (Zilliox and Smith 2017), regulating nuisance iss ues, and advocating for increased setbacks and other limits at the state level. Beyond local issues, fracking is part of broader national and global debates about reliance on fossil fuels as energy sources. While those in favor of fracking often cite nat ural gas as a cleaner alternative to oil and coal, those opposed claim that fracking and the availability of cheap oil and natural gas impede transitions to renewable energy sources, like wind and solar. Because they involve a wide range of issues across v enues and levels of government, fracking debates provide a good context for this study. Fracking is an intensely debated political issue (Heikkila and Weible 2016), drawing participation from industry, government, and nonprofit organizations. Fracking po licies are created via judicial decisions, legislation, agency rules, executive orders, and ballot measures, and organizations, including environmental nonprofits, engage in a variety of activities and strategies to influence each of the sources of policy. As such, fracking is an ideal policy context for observing maximum potential advocacy activities. To examine how nonprofits interact in hydraulic fracturing policy, this dissertation relies on a diverse array of methods, which are explained more fully in each of the subsequent chapters. Data sources include newspaper articles, interviews, and surveys of environmental nonprofits active in oil and gas and related policy. Analyses rely on descriptive and analytic

PAGE 30

18 statistics, thematic analysis and constant comparative methods, and qualitative comparative analysis. Overview of Remaining C hapters The final section of this chapter offers a brief overview of the remaining chapters of this dissertation. Each of the three empirical chapters was written to be a standalone publication. As a result, the reader will find some overlap in the literature and study context between chapters. For each chapter, a short summary of the substance and proposed methods is provided. An overview of Chapter s 2 through 4 is also provided in Appendix A. Advocacy Groups and Venue Selection Chapter 2 presents an analysis of the advocacy groups and venues active in hydraulic fracturing policy in Colorado over a five year period . Venues are an essential concept in literature on advocacy in that they may both frame the activities that organizations use to influence policy and venue selection is a strategy in and of itself (Pralle 2006). For example, when actors are frustrated wi th a given decision making body, they may look for an alternative venue. While existing venue shopping literature identifies both venue and organization related factors that influence venue selection, the literature lacks an analysis of how organizations can be both proactive and reactive in choosing venues. By examining both industry and environmental group involvement across venues in a particular policy subsystem, this paper demonstrates that venue selection is an ongoing process, influenced by a numb er of competing factors. Data for this chapter were derived from the news media. Specifically, data include over 500 articles from two Colorado newspapers published from 2013 to 2017. Articles were coded in NVivo using constant comparative method to iden tify policy issue s, venues, and advocacy groups. Analysis is based on both descriptive and analytic statistics. Results shed light on how

PAGE 31

19 industry and environmental advocacy groups differ in their approaches to advocacy, and how activities across venues evolve over time. How Do Environmental Nonprofits Advocate? Building on the second chapter and using venues as an overarching framework, Chapter 3 explore s the variety of activities that environmental nonprofits use to influence policy . Existing literatur e describes that organizations use both inside and outside strategies to influence policy. This study expands on existing literature by describing the array of inside, outside, and hybrid strategies that environmental nonprofits use across different venue s and how strategies differ by venue. Furthermore, while literature suggests that inside strategies are more common and effective, and that more professionalized organizations tend to use inside strategies, in line with recent research, this study suggest s that the relationship between strategies and organizations is more complex. Thus, this study also identifies a few key organizational and contextual factors that may be influencing advocacy decisions. This chapter relies on semi structured interviews with executive directors or other senior staff of environmental nonprofits across three states: Colorado, New Mexico, and Oklahoma. While each of these states has significant oil and gas development and ongoing policy debates, these states also have diffe rences in their political contexts and the types of land on which oil and gas development takes place. Interviews were analyzed using thematic analysis. Results demonstrate the types of advocacy activities that groups use across venues; illuminate the si milarities and differences between venues; and suggest some of the major factors that influence

PAGE 32

20 Qualitative Comparative Analysis of Advocacy and IRS Designation Chapter 4 presents an additional look at the advocacy activities of environmental nonprofits, with a particular eye toward differences between 501(c)(3) and 501(c)(4) organizations. Among social service providers, 501(c)(3) organizations are typically service providers and 501(c)(4)s are advocacy organizatio ns. In the environmental space, however, both 501(c)(3) and 501(c)(4) organizations are often advocacy organizations. This study asks two research questions. First, what activities do environmental nonprofits engage in to influence the policy process? Second, which organizational characteristi cs are associated with advocacy among environmental nonprofits? This study draws again on the advocacy literature and uses Resource Dependency Theory and Institutional Theory to derive theoretical propositions abo ut the relationship between advocacy activities and organizational characteristics. Chapter 4 relies on survey data from 30 environmental nonprofits about the types of activities they use to further their policy goals and whether the y report using these a ctivities effectively. After presenting the results of descriptive statistics, t he primary analysis for this chapter uses QCA to examine the organizational factors typically associated with different types of advocacy. Results indicate that organizations report engaging in advocacy at high rates and report doing so effectively. Furthermore, 501(c)(3) and 501(c)(4) organizations report similar trends in advocacy , except that 501(c)(4) organizations are not associated with advocacy that involves collaborat ing with opponents. Concluding Chapter Finally, this dissertation concludes with a discussion of major takeaways and some of the overarching limitations . The concluding chapter also describes a research agenda that builds on the foundations laid in this dissertation.

PAGE 33

21 CHAPTER II ADVOCACY GROUPS AND VENUE SELECTION Abstract This chapter presents an analysis of the advocacy groups and venues active in hydraulic fracturing policy in Colorado over a five year period. Venues are an essential concept in lite rature on advocacy in that they may both frame the activities that organizations use to influence policy and venue selection is a strategy in and of itself. While existing venue shopping literature identifies both venue and organization related factors t hat influence venue selection, the literature lacks an analysis of how organizations can be both proactive and reactive in choosing venues. By examining both industry and environmental group involvement across venues in a particular policy subsystem, this paper demonstrates that venue selection is an ongoing process, influenced by a number of competing factors. Data for this chapter are drawn from over 500 news articles published in two Colorado newspapers between 2013 and 2017. Articles were coded in NV ivo using constant comparative method to identify policy issues, venues, and advocacy groups. Analysis is based on both descriptive and analytic statistics. Results shed light on how industry and environmental advocacy groups differ in their approaches t o advocacy, and how activities across venues evolve over time. Keywords Venue selection, advocacy groups, environmental policy

PAGE 34

22 This chapter presents an analysis of how advocacy groups interact across venues in an oil and gas policy subsystem across a five year period. To examine the choices that different groups make with respect to venues, this chapter divides advocacy groups in to two categories: industry advocacy groups and environmental advocacy groups. There is certainly variation across these groups both in terms of their level of formality or institutionalization and in terms of where they fall on the spectrum ranging from remains, though, that these groups typically find themselves on opposite ends of the negotiating table. Thus, by examining in ven ue selection and policy subsystem evolution. This chapter asks two primary research questions: 1) What are the different types of venues and policy issues that advocacy groups are involved in within the Colorado oil and gas policy subsystem? 2) Are there differences in how industry advocacy groups and environmental advocacy g roups participate across venues? After describing literature on venues, policy subsystems, venue selection, and advocacy groups, the study context and use of news media are discussed. I then present the results of the study along with a discussion of some of their potential implications, particularly for the venue selection literature. Overview of Venues in the Policy Literature Exploring the venues in which advocacy occurs provides a logical starting place for divided both vertically, such as by level of government, and horizontally, such as by branch of government (Holyoke, Brown, and Henig 2012; Buffardi, Pekkanen, and Smith 2015), and as described more fully below, activity in policy subsystems often occurs across numerous venues.

PAGE 35

23 Venues not only p rovide the rules and structures within which advocacy occurs, but organizations may also manipulate venues to achieve their preferred outcomes. The importance of venues derive s from the federalist system and separation of powers that characterize governance in the US. Federalism refers to a division of power between multiple levels of government, and separation of powers refers to our tripartite system of government, designed t o prevent any one branch of government from dominating. Together, these features result in a polycentric 3 policy system with many centers of decision making that are formally independent of one another (V. Ostrom 1999). Such a system also provides opport unities for multiple decision making venues to play a role in governing any given policy issue. While historically polycentric systems of governance were thought to stymy real change because of the multiple avenues available for blocking change, more rece ntly, scholars have argued that polycentric systems of governance support change in that they offer multiple decision making venues and opportunities to shift policy (Baumgartner and Jones 2009 ; Pralle 2003). Policy research finds that policy subsystems a re comprised o f numerous venues and sub issues, and that when people and organizations engage in the policy process, they often engage in multiple venues . For example, in the Advocacy Coalition Framework (ACF), the primary unit of analysis is the policy s ubsystem (Jenkins Smith et al. 2018). A subsystem is defined by a policy topic, a territorial scope, and the actors involved. Within a given subsystem, coalitions, or groups of actors who share policy core beliefs and coordinate actions, compete across d ifferent 3 The term polycentricity was originally used by Michael Polanyi in the context of scientific communities (V. Ostrom 1999). Generally speaking, a polycentric organization is one where ment for ordering their relationships and Warren (1961) first applied the concept to government.

PAGE 36

24 venues over extended periods of time to influence policy change. Also, Ecology of Games F ramework simultaneously within a geographically defined policy arena, where a policy g ame consists of a set of policy actors participating in a rule governed collective decision making process called a P olicy institutions consist of both formal and informal collective choice rules, governing how de cisions are made in a given policy institution or venue (Lubell 2013; Ostrom 1990). Thus, the Ecology of Games Framework contemplates that games across multiple venues may occur simultaneously. Recent research also suggests that policy conflicts occur ac ross three polycentric levels of action: political systems, policy subsystems, and policy action situations ( Heikkila and Weible 2017 ; Weible and Heikkila 2017 ). T he political system is the overarching system that governs a territory and has authority ove r a wide range of issues. The policy subsystem is as defined above by the ACF. Finally, policy action situations and informal venues where policy actors engag e, debate, and attempt to address problems around Given the availability of multiple venues, venue selection presents a key strategic choice for those interested in influencing policy. When multiple venues h ave authority over a single groups who seek out a decision setting where they can air their grievances with current policy and present alternative policy proposal with Schattschneider (1960), who described that losing groups may increase their chances of success by expanding policy conflicts. More recently, interest in venue shopping trace s to Baumgartner and Jones (2009 ), who described the importance of policy images and venues to

PAGE 37

25 policy stability and change. Policy images refer to how the public understands a policy problem (Baumgartner and Jones 2009 , 25). Due to bounded rationality and limited attent ion, policy makers and members of the public typically lack the capacity to hold detailed understandings of policy issues, and therefore rely on heuristics like policy images. Specific policy images resonate differently with different venues and decision makers, which means that as images shift, venues may shift as well. Venue Selection At the heart of the venue shopping literature is the idea that actors make strategic choices about venues to maximize their likelihood of obtaining their desired outcome (Ley and Weber 2015). Thus, when actors are frustrated with a given venue and the biases that exist within it, they may look for an alternative venue (Pralle 2003). Sometimes venue shopping results in issue expansion, consistent with Schattschneider (19 60), while at other times, the strategy is simply to shift the conflict to a new, more favorable venue (Baumgartner and Jon es 2009 ; Pralle 2006). Furthermore, while venue shopping is sometimes considered a type of strategy among a host of other types of s trategies (e.g., Weible and Heikkila 2017), it might also be considered a preliminary choice that then structures and frames the subsequent advocacy activities that groups employ ( e.g., Crow et al. 2019). While traditional venue shopping literature posi ts that actors make carefully reasoned and deliberate decisions regarding venues, more recently scholars recognize that venue shopping is somewhat ad hoc (Pralle 2003). Most notably, Pralle (2003) demonstrates that venue shopping is more experimental and less deliberate than commonly perceived. Recent studies have elucidated the various constraints that actors face when choosing among competing venues. These constraints can be divided into factors related to venues and factors related to organizations.

PAGE 38

26 First, with respect to venue related factors, among the relevant considerations are whether the preferred venue has meaningful authority over the issue at hand, the relationship between venues, ideological alignment between the venue and advocates, the ve given group, and the level of control an opponent exerts in a given venue (Holyoke, Brown, and Henig 2012; Ley and Weber 2015). Each of these elements may change over time, giving rise to incentives to shift venues. For example, as noted above, the dominant policy image may change; the venues themselves may change, such as through the election or appointment of new decision makers; and a decision in one venue may influence the authority of another v enue (Baumgartner and Jones 2009 ). constraints and venue selection (Holyoke, Brown, and Henig 2012). That is, organizations with more financial resources, human resources, and expertise can more e asily shift across venues . Furthermore, organizations often have pre existing preferences based on either their organizational identities or on path dependency (Pra lle 2003). With regard to the former, as examples, grassroots organizations may prefer local venues; legal organizations may prefer courts; and anti establishment organizations may prefer the ballot box. Also, path dependency may lead organizations to be (Pralle 2003, 239). Relatedly, Buffardi, Pekkanen, and Smith (2015) found that nonprofits t end to specialize at certain levels of government, but shop among various venues within each level of government. Finally, venue choice is also influenced by learning. Decisions regarding venue selection are sometimes guided by misinformation or outdated perceptions of venues. Over time,

PAGE 39

27 though, sometimes through a trial and error process, actors learn to make better decisions, and learning may occur at the individual, organization, coalition or subsystem level ( Pralle 2003 ; see also Heikkila and Gerlack 2013). Though scholarship has moved away from assumptions that actors make fully rational decisions about which venues to participate in, the literature continues to assume that actors make proactive decisions about where to place their issues or put the ir efforts. A much more limited and nascent literature, however, suggests that venue selection may actually be reactive. Ley (2016) distinguishes between activities designed to produce policy change and venue He then examines how groups seek to insulate their policy victories once achieved. Other related literature describes the importance of veto points to those wishing to block opponents from achieving their goals (Weible, Heikkila, deLeon, and Sabatier 2012). Also , Holyoke, Brown, and Henig (2012) find that actors tend to choose venues in which decisi on makers are actively working on a given issues, which suggests some form of reactivity. Despite implicit recognition that venue selection is often reactive and somewhat out of literature. That is, in high conflict policy subsystems, such as those described by the ACF, PCF, and Ecology of Games Framework, actors and coalition members are constantly engaged in a wide range of policy games or action situations, some of which they choose and others of which are merely a the venue selection literature by exploring how advocacy groups from two competing ideologies participate across venu es over a five year period. After defining and describing advocacy groups, this paper sets up the study context and methods employed.

PAGE 40

28 Advocacy Organizations To explore venues, this study focuses on two competing coalitions 4 of advocacy groups: industry ad vocacy groups and environmental advocacy groups. Advocacy groups include change that if implemented, would conflict with the social, cultural, political, or eco nomic include diverse groups ranging from large institutionalized nonprofit organizations , to small informal social movement organizations or community groups ( Mosley 2011). These organizations cross a variety of literatures, including interest group , social movement, and nonprofit literatures (Andrews and Edwards 2004) , and they are certainly relevant to other literatures, like the policy process literature, as well. For example, interest groups generally refer to voluntary organizations that attempt to influence government. In contrast, the social movement literature typically focuses on informal groups that attempt to effect systemic change from outside gover nment institutions. Finally, the nonprofit literature typically zooms in on more formal nonprofits, which share similar basic structures and adhere to corporate and tax laws. ubsystem and are included in this study. The first is formal nonprofit organizations. These organizations generally incor porate purs uant to state law and register for tax exempt status with the IRS. To gain federal tax exemption, nonprofits must registe r under Title 26 of the IRS code, typically under a subpart of section 501(c). Most relevant to this study are 501(c)(3), (4), and (6) 4 core beliefs and who coordinate Smith et al. 2018, 148).

PAGE 41

29 organizations. Environmental nonprofits usually register as either 501(c)(3) or 501(c)(4) organizations. Section 501(c )(3) organizations are public charities or private foundations, and section 501(c)(4) organizations are social welfare organizations. While 501(c)(3) organizations are limited in certain types of advocacy in exchange for tax benefits, section 501(c)(4) or ganizations are not limited and are often characterized as advocacy organizations. Particularly in the environmental space, though, 501(c)(3) organizations advocate through limited amounts of lobbying, educating the public or local leaders, and engaging w ith the media, among other things (see Chapter 3). On the other hand, many industry advocacy groups register as 501(c)(6) organizations. These organizations are business or professional organizations and are typically comprised of organizational or indivi dual members from particular industries or professions. Similar to 501(c)(4) organizations, these organizations are not limited in their advocacy activities. A prime example of this organization type is the Colorado Oil and Gas Association (COGA) , which natural gas markets, supply, and transportation infrastructure through its growing and diverse The second type of advocacy group involved in oil and gas policy in Colorado is informal groups. These organizations vary in their structures, but are typically smaller, volunteer run, and less structured than formal nonprofits. These are often neighborhood or community groups that represent local c itizen interests and are sometimes organized via listservs or social media groups. In some cases, these local groups are neste d in larger formal nonprofits. Finally, action groups are those that organize around specific policies, most commonly to support or oppose a ballot measure. These organizat ions are sometimes registered under section

PAGE 42

30 527 as political action committees (PACs) , but what distinguishes them is that they are typically shor t term organizations that work on a specific policy pending in a particular venue. Study Context: Oil and Gas Policy Debates in Colorado To improve our understanding of the activities that advocacy groups use across venues, this study relies on news media coverage of hydraulic fracturing (fracking) policy debates in Colorado. Fracking is a technique involving the injection of water, sand, and chemical additives ase oil or natural gas (Heikkila et al. 2014). In the US and around much of the world, fracking is increasingly used in oil and gas development. For example, between 2007 and 2017, the amount of natural gas produced from shale formations increased 10 fol d from 1,990,145 million cubic feet to 19,018,457 million cubic feet (EIA 2018). Not only has the volume of oil and gas produced through fracking increased, but oil and gas development is also encroaching on high population density areas. Close proximit y between population centers and oil and gas operations gives rise to a host of local issues, including NIMBYism (not in my backyard); nuisance issues, like traffic, dust, and noise; and concerns about air and water contamination. As a result, communities grapple with how to manage fracking, and some have attempted to ban it. For example, while the state of New York successfully banned fracking, local bans in a number of municipalities like Denton, Texas, and Longmont, Colorado, have been struck down by c ourts or legislatures (Knight and Gullman 2015; Heikkila and Weible 2016; Jaquith 2017). Local communities with limited authority over oil and gas development may attempt to exercise control by negotiating agreements with oil and gas companies (Zilliox an d Smith 2017), regulating nuisance issues, and advocating for increased setbacks and other limits at the state level. Beyond local issues, fracking is part of

PAGE 43

31 broader national and global debates about reliance on fossil fuels as energy sources. While tho se in favor of fracking often cite natural gas as a cleaner alternative to oil and coal, those opposed claim that fracking and the availability of cheap oil and natural gas impede transitions to renewable energy sources, like wind and solar. Because they i nvolve a wide range of issues across venues, fracking debates provide a good context for this study. Fracking is an intensely debated political issue (Heikkila and Weible 2016), drawing participation from actors across sectors . Fracking policies are crea ted via judicial decisions, legislation, agency rules, executive orders, and ballot measures, and organizations, including diverse advocacy groups , engage in a variety of activities and strategies to influence each of the sources of policy. As such, frack ing is an ideal policy context for observing advocacy activity across a range of venues . This study focuses specifically on the oil and gas policy subsystem in Colorado. Colorado ranks eighth in the nation for shale gas production and seventh in the natio n for crude 5 Colorado regularly grapples with oil and gas issues and has seen its share of fracking related controversy. Colorado is also an ideal state for studying fracking because of the conflicts occu rring across venues. For example, local government regulation of fracking has been an issue across the state. In 2012, the city of Longmont banned fracking via voter initiative, and in 2016 the Colorado Supreme Court ruled that the ban was invalid and un enforceable. Relatedly, in 2014, as a compromise to remove two statewide citizen initiatives from the ballot, Governor Hickenlooper established a task force to address state and local authority over fracking. During the time period covered by this study, 5 During the majority of the study period emocratic major ity and its state senate had a R epublican majority. G overnor John Hickenlooper is a D emocrat, but is widely seen as supportive of the oil and gas industry.

PAGE 44

32 numerous bills were introduced at the legislature, such as bills related to setbacks, forced pooling, and incident reporting laws, though bills saw very little success. At the state agency level, the Colorado Oil and Gas Conservation Commission (COGCC) p rimarily oversees oil and gas activity in the state, though other agencies like the Colorado Department of Public Health and Environment and the Colorado Department of Natural Resources oversee issues relevant to oil and gas development like air pollution and wildlife. Furthermore, while much oil and gas development occurs on private land on the Front Range near population centers, 36% of land in the state is federally owned, and, particularly on the Western Slope, oil and gas development regularly happens on federal land. Methods Data C ollection This study relies on the news media as its data source. The news media can be a valuable research tool in that it is publicly available and tracks trends over time, allowing for lower cost longitudinal studies tha are certainly limitations associated with the news media. For example, only those issues, actors, and activities that are most visible are likely to gain media coverage, while other, sometimes important, events and activities may go unnoticed by the media. Furthermore, considerable scholarship often notes that the media plays a role in influencing policy either by shaping public opinion or by serving as a tool that policy actors may exploit (Shanahan et al. 2008; Jones and Wolfe 2010). Despite the benefits and drawbacks, scholars have long recognized the media as a in government, by and l

PAGE 45

33 (59). Recent studies continue to rely on the media. For example, the Narrative Policy Framework (NPF) makes widespread use of the media for understanding the narratives involved in p olicy conflicts (e.g., Shanahan, McBeth, and Hathaway 2011; Crow and Berggren 2014). Also, more recently, studies increasingly extend beyond traditional news media and look at diverse outlets like social media (e.g., Merry 2016) Data for this study were d rawn from newspaper articles in Colorado. This research uses two newspapers likely to have extensive coverage of oil and gas debates: the Denver Post and the Longmont Times Call . Denver is the capital and largest city in Colorado . Thus, s tate level policy and decision making is centered in Denver , and because they operate within the confines of state law, local debates often tie to state decision making. Denver is also home to regional federal offices , along with oil and gas industry headquarters an d regional offices , making it a hub of governance and decision making. Longmont , on the other hand, is a center of oil and gas policy activity because it has a high concentration of wells. Longmont also straddles Boulder and Weld counties, which often ap proach oil and gas policy differently. I n 2012, Longmont residents banned hydraulic fracturing within city limits ruled invalid and unenforceable by the Colorado Supreme Court. As part of a larger project, a tea m of researchers accessed articles through Newsbank and used a two step process to identify articles related to oil and gas debates in Colorado. The first step used the following search terms to identify potentially relevant articles from each newspaper p Denver Post and 1,450 articles from the Longmont Times Call . Once articles were i dentified and downloaded, the second step consisted of a manual, or automated when possible, review of articles to ensure that

PAGE 46

34 they are relevant to oil and gas policy debates in Colorado. The following categories of articles were excluded from the final p opulation: opinions, editorials, or letters to the editor; articles only tangentially related to oil and gas issues; articles about federal issues, global issues, or issues in another state that do not relate to Colorado; and briefs or short summaries of a series of news topics, only one of which is oil and gas related. After exclusions, t he final population includes 908 articles published between 2007 and 2017. Because the time period for this study is 2013 2017, the final sample includes 534 articles . D ata A nalysis To analyze the news articles, I uploaded and hand coded articles in NVivo. Coding was guided by constant comparative analysis, which is an iterative coding process often used in grounded theory studies (see Corbin and Strauss 2015). In a fir st round of open coding, I coded all 908 articles for major policy issue , primary venue, and date. This round of coding resulted in 78 distinct policy issues and 25 venues. Policy issue s and venues were then refined and reapplied in a r ound of closed cod ing to the 534 articles that were publ ished in the 2013 to 2017 time frame. The second round of coding included 23 policy issue s and 11 venues. A third review of the sample articles was performed to ensure that codes were consistently applied across the articles. Next, I identified advocacy groups , based on the three categories described above, in the sample articles usi ng Nodes in NVivo. I included all advocacy groups appearing in articles, unless they were only tangentially mentioned or mentioned in part of the article not directly related to fracking. In other words, there was no requirement that advocacy groups be mentioned in relation to a particular venue or activity. While in some cases, articles mentioned that groups participated in a particul ar way, such as protesting at the capitol or filing a court brief, in many

PAGE 47

35 cases references were limited to short quotes or mentions that groups supported or opposed certain policies or actions. Overall, 52 environmental advocacy groups were mentioned in 200 articles, and 15 industry advocacy groups were mentioned in 213 articles. Results This section presents preliminary answers to each of the research questions using both descriptive and inferential statistics and graphs. Again, the research questions guiding analysis are as follows: 1) What are the different types of venues and policy issues that advocacy groups are involved in within the Colorado oil and gas policy subsystem ? 2) Are there differences in how industry advocacy groups and environmental advocacy groups participate across venues? RQ1: Venues and Policy I ssues Summary of Venues As a preliminary matter, t he frequencies with which venues appeared in t he Colorado news media from 2013 to 2017 are shown below in Table 1 and Figure 1 . Venues were coded at the article level. 6 While the majority of articles describe activity in a particular venue, just over a quarter of the articles in the sample are not grounded in a particular venue. These include features about the economic or health impact s of oil and gas on local communities; news about the oil and gas industry, including job loss and creation; and accidents or safety issues on well pads, among other issue s. 6 Apart from articles that are not grounded in a particular venue, as described in the text, all articles contain a dominant venue. Many articles focus exclusively on one venue, and even among those articles that mention more than one venue, there is a clear primary venue.

PAGE 48

36 Table 1 . Venue Frequency in A rticles. Venue Number of Articles Percent of Articles No Venue 140 26% Local Government 132 25% State Court 80 15% State Legislature 41 8% State Agency 41 8% State Ballot 32 6% Local Ballot 21 4% Governor 19 4% Federal Combined 28 5% Figure 1 . Venue F requency in A rticles . The most active policy venue during the time period is local government. Between 2013 and 2017, numerous local governments banned fracking or implemented moratoria and more broadly struggled with regulating fracking and interpreting their authority under state law. Art icles about local issues also sometimes fall in other venues, such as in the case of the COGA v. Longmont lawsuit, which played out in state court, but a considerable portion of the articles in No Venue Local Government State Court State Legislature State Agency State Ballot Local Ballot Governor Federal Combined

PAGE 49

37 this period is covered activity by local governments. While s ome of this action was by city councils and other activity was in local agencies, because city councils often have to approve agency actions, and because local government action is often discussed broadly, local g overnment activity is combined. Next most f requently, articles discuss state court activity. In 2012, voters in Longmont enacted a fracking ban via a local ballot initiative, and in 2013, Fort Collins implemented a five year moratorium. In response to these policies, the Colorado Oil and Gas Asso ciation (COGA), joined by the COGCC, sued these municipalities in state court, claiming that local governments lacked the authority to ban fracking because state law preempts local government regulation . The district court sided with COGA, and in early 20 16, the Colorado Supreme Court affirmed , deeming the local policies invalid and unenforceable. Following this decision, however, Boulder County kept a fracking moratorium in place while it worked on adopting new local rules to regulate fracking. This mor atorium was extended numerous times, and in 2017, the Colorado Attorney General sued Boulder County as a result. Issues at the state legislature also arose during this period, although, likely because of the split legislature, this was not the most active venue during this period. State agency activity primarily refers to rulemaking or other activity at the COGCC, which is the primary regulator of the oil and gas industry. Activity in a few other agencies, such as the Colorado Department of Natural Resour ces and the Colorado Department of Public Health and Environment, is included as well. Smaller numbers of issues arose at the federal level, through state ballot initiatives, the ll number of federal

PAGE 50

38 issues likely reflects the fact that articles were only included in this sample if they included ties to Colorado. 7 Venues and P olicy Issue s As the above begins to suggest, there is certainly overlap between venues and issues. That is , within a given subsystem, while some venues hear a wide range of issues, others primarily focus on just one or two issues. For example, as described above, local governments frequently dealt with bans and moratoria during the relevant time period. Whil e the key questions in this chapter relate to advocacy group participation in venues, a basic understanding of the relationship between venues and sub issues, also allows for conclusions regarding the factors that may drive participation in venues. As alr eady noted , 23 distinct policy issues were identified in the data. These issues are detai led in Appendix B to the dissertation . To highlight the complexity of the Colorado oil an d gas policy subsystem, Table 2 below shows the frequency with which policy issue s occur across levels of government . It also demonstrates that while there is some relationship between venues and policy issues, issues are certainly not confined to a single venue. 8 The Table highlights that in some cases, like with fracking generally, industry safety, and procedural matters, issues are heard across venues. Other issues, however, like public land, are largely confined to just one or two venues. Thus, when actors participate in venues, they may do so because they believe that the venue is friendly, or because an issue already being considered in that venue is of interest to them. 7 The Federal Combined Category includes federal courts, agencies, and Congress. 8 To simplify the table, and be cause the venue variation is clear, venues are reduced to level of government here.

PAGE 51

39 Table 2 . Policy Issue by Level of Government. Local State Federal No Venue Total Ban/Moratorium 75 54 2 131 Local Authority 15 58 5 78 Fracking Generally 30 22 2 18 72 Siting/Setbacks 13 23 16 52 Industry Safety 3 7 1 21 32 Economy/Jobs 1 1 24 26 Public Land 1 18 19 Air Quality 5 1 7 13 Procedural 2 9 1 1 13 Anti fracking Movement 5 2 5 12 Health Impacts 5 1 6 12 Industry News 12 12 Other 8 27 4 23 62 Total 153 213 28 140 534 RQ2: Advocacy Groups Advocacy G roups This section examines differences in how industry advocacy groups and environmental advocacy groups participate across venues to contribute to our understanding of venue selecti on. As a preliminary note, 57% of the articles (307 out of 534 ) include mentions of advocacy groups . Industry groups appear in 213 articles (40 %), and e nvironmental groups appea r in 200 articles (37 %). It is not clear whether this accurately reflects whether groups were in fact involved . For example, where advocacy groups are not mentioned, it is possible that they were involved but were acting behind the scenes and outside of Table 3 shows the breakdown of advocacy group mentions across all articles by year and suggests that advocacy group activity in both coalitions has declined on average over time.

PAGE 52

40 Table 3 . Articles and Advocacy Group Activity by Year. 2013 2014 2015 2016 2017 Number of Articles 122 149 84 75 104 % with Environmental Groups 42.62% 40.94% 36.90% 36% 27.88% % with Industry Groups 40.16% 45.64% 38.10% 34.67% 36.54% Next, a dvocacy group activity by venue is shown below in Table 4 . As noted above, associations are based on whether actors appear in articles about particular venues, and not whether the article mentions that actors engaged in any specific way. As the results suggest, advocacy groups in both coalitions participate acros s venues, and for most venues, they participate at similar rates. Just four venues (or non venues) show statistically significant differences in rates of participation. First, where issues are not related to any particular venue, articles mention industr y groups significantly less than environmental groups. As shown above, these most commonly are articles that talk broadly about the economic effects of fracking, industry safety concerns, or fracking generally, each of which may not relate to particular p olicy proposals. Thus, i t may be that industry groups reserve their activity for pending policies , r ather than get involved in general discussions about fracking and its effects . On the other hand, i ndustry groups participate significantly more in state court activity than do environmental nonprofits. In the 2013 to 2017 time frame, this is most likely because COGA sued Longmont and Fort Collins in response to their ban and moratorium, respectively, and this litigation received considerable attention in both Colorado and nationally.

PAGE 53

41 Table 4 . Venues with Mean Pa rticipation by Advocacy Group (T Test R esults). Venue Number of Articles Environmental Groups Industry Groups Not Applicable** 140 0.314 0.207 Local Government 132 0.295 0.311 State Court*** 80 0.413 0.775 State Legislature** 41 0.171 0.39 State Agency 41 0.488 0.366 State Ballot 32 0.75 0.688 Local Ballot 21 0.476 0.381 Governor 19 0.368 0.474 Federal Agency* 16 0.688 0.375 Federal Court 6 0.333 0.5 Federal Legislature 6 0.5 0.333 *p < .1 **p < .05 ***p < .01 (two tailed) Environmental groups are also significantly less likely than industry groups to be involved in issues at the state legislature. As noted above, this may be related to the fact that throughout much of the s tudy time period, the state house was controlled by D emocrats, while the state senate was controlled by R epublicans, making it difficu lt to pass legislation. E nvironmental groups may have believed other venues were more favorable to their interests. On t he other hand, environmental groups are significantly more likely to be involved in federal agencies. This difference may be attributable to the fact that many environmental nonprofits specialize in federal public land issues. In these instances, federal agencies, such as the Bureau of Land Management ( BLM ) , often serve as opponents, rather than oil and gas interests (though these groups are certainly involved as well). Because there is some relationship between policy issue and venue , with certain issues appear ing more frequently in certain venues , Table 5 below shows comparisons between industry group and environmental group involvement across the most frequently mentioned policy issues. In other words, this analysis tests the possibility that issues ar e the sole drive of

PAGE 54

42 venue activity. In the case of policy issue s, significant differences exist between the groups in just two areas: bans and moratoria and the anti fracking movement. W ith respect to bans and moratoria, industry groups are significantly more likely to be involved. It may be that these issues pose a larger thre at to the industry because bans and moratoria may impair oil and gas industry operation s , and thus, industry adv ocacy groups are likely to get involved. Not surprisingly, environmental advocacy groups dominate articles about the anti fracking movement. Rather than playing an advocacy role, though, these groups are generally the subjects of articles. Thus, policy issues may contribute to differences in participation in state court, which often deals with bans and moratoria. Other differences in venue participation, such as at the state legislature, do not appear to be issue driven. This suggests that, at least in some cases, there are venue related factors that drive advocacy group participation. Table 5 . Policy Issue s with Mean Pa rticipation by Advocacy Group (T Test R esults) . Issue Number of Articles Environmental Groups Industry Groups Ban/Moratorium*** 131 0.351 0.527 Local Authority 78 0.538 0.654 Fracking generally 72 0.347 0.319 Siting/setbacks 52 0.365 0.288 Industry Safety 32 0.094 0.188 Economy/Jobs 26 0.154 0.115 Public Land 19 0.632 0.368 Procedural 13 0.538 0.384 Air Quality 13 0.538 0.462 Anti fracking Movement** 12 0.667 0.167 Industry News 12 0.167 0.333 *p < .1 **p < .05 ***p < .01 (two tailed) Bans, Moratoria, and Local A uthority To deepen our understanding of advocacy groups and venue, this study next zooms in to a specific policy issue in the Colorado oil and gas policy subsystem and traces the issue across

PAGE 55

43 venues and advocacy groups , using the coded news media data . Specifically, this study examines issues related to bans, moratoria, and local authority, which were the most prevalent issues in the data. While these issues were code d separately under two categories (ban/moratorium and local authority) , they are in fac t closely related in that local authority issues often center on whether local governments have the authority to ban or halt fracking. Figure 2 below shows a timeline of the major events related to bans, moratoria, and local authority issues in the 2013 2 017 time period . The F igure reflect s only the subset of articles in this time period related to two previously identified policy topics: ban/moratorium and local authority (209 articles combined). Thus, the line reflects the number of articles that appea red in each quarter on these topics, and the boxes above the timeline show major events based on a manual review of the 209 articles. First, to fully understand the 2013 2017 time period in Colorado in relation to oil and gas development, it is necessary t o go back to the end of 2012. In November of 2012 , the citizens of Longmont voted to ban oil and gas development that uses hydraulic fracturing within city limits. At this point, Longmont was already a hub of oil and gas conflict. At the time , there was a lawsuit pending against Longmont, initiated by the COGCC, for place as ground zero for state, if not national, fracking policy conflicts. In response to the voter initiated ban, in December 2012, COGA, later joined by the COGCC, sued Longmont claiming that local governments lack the authority to ban fracking because state law preempts local government regulation.

PAGE 56

44 Figure 2 . Timeline of Bans, Moratoria, and Local Authority Issues.

PAGE 57

45 In addition to the pending litigation, action in 2013 and early 2014 primarily involved local efforts to ban or delay fracking, and in 2014, action heated up again when groups from both sides of the issue drafted and began collecting signatures for ballot measures related to local authority. In attempt to thwart these measures, Governor Hickenlooper worked with the Colorado legislature to draft legislation related to local authority. Neither side was happy with proposed bills , though, and in May 2014 the legislative session ended without successfully resolving the issue. In July 2014, two major events occurred. First, the district court judge in Boulder sided with COGA and ruled that the Longmont ban was unenforceable, though the dec ision was stayed pe nding appeal . Second, the state approved two ballot measures for the November ballot, both of which posed significant threats to the industry. One measure would have created an environmental rights amendment to the state constitution, and a second would have increased setbacks, or the distance between well pads and homes, from 500 feet to 2,000 feet. In response to these ballot measures, Governor Hickenlooper successfully struck a compromise with US Representative Jared Polis, who bankrolled the measures , in which Hickenlooper agreed to create a task fo rce to address local authority issues in exchange for Polis withdrawing the ballot measures. The task force began meeting in September 2014 and issued its report in February 2015. Later that year, in Augu st 2015 , the Colorado Court of Appeals requested that the Colorado Supreme Court rule in the Longmont fracking ban case, and the Supreme Court accepted the appeal . As the litigation continued, in December 2015 , the COGCC released a set of proposed rules i

PAGE 58

46 ban, along with a moratorium in Fort Collins, invalid and unenforceable. When Boulder Count y continue d to extend i ts moratorium, in February 2017 the Colorado Attorney G eneral sued. While the peaks and valleys in news media coverage generally follows the events described above, some of the additional patterns in news media coverage derive from local ban, Brighton, Fort Collins, Boulder County, Broomfield, and others all enacted bans or moratoria early in the time period. In addition, a measure to halt fracking in Loveland was voted down by voters in 2014. Boulder County in particular received a lot of press coverage for its activities. Between 2013 and 2017, Boulder County extended its moratorium five times before the Colorado Attorney General sued th e county. Boulder ended its moratorium in May 2017. Furthermore, despite the Supreme Court ruling, in late 2017, local governments, including Lafayette and Dacono, enacted moratoria 2017 , and Broomfield voters passed a ballot measure giving the local gov ernment more local control over oil and gas operations. Activity by v enue . As the above suggests, policy disputes related to local authority issues spanned a number of venues. Throughout the relevant time period, seven venues were involved: the Governor courts, the state legislature, and state agencies. As shown in Figure 3 , the majority of activity occurred in state courts and local governments. Data for the Figure are based on the 209 articles coded as ban/moratorium or local authority, and the lines reflect the raw number of articles by year associated with each venue. The Figure also shows an inverse relationship between local government activity and state court activity. T his suggests that while local governments provided venues for resolving fracking disputes early in the time period, once COGA moved the dispute t o state court by suing Longmont and later other local jurisdictions, local governments

PAGE 59

47 became less active as ve nues. Interestingly, though, after the Supreme Court decision, which significantly limited local authority, was announced, local government activity increased, as local governments looked for ways to regulate within their newly delimited authority. Simil arly, though less dramatically, activity at local ballot boxes was higher early in the time period, dropped off as other venues took off, and then began to rise again at the end of the time period. Figure 3 . Activity Across Venues by Year. Patterns are less clear in other venues. For example, state ballot activity saw a dramatic increase in 2014, but was minimal throughout the rest of the time period, and activity in each of the other three venues was relatively stable and low throughout the time period. The only venues not involved in this issue are federal venues, including federal courts, agencies, and Congress. Because this issue is about state and local control this is a logical result. Tak en tog ether, the results in this section may suggest that activity across venues is somewhat of a tradeoff. That is, when activity in one venue is high, 0 5 10 15 20 25 30 2013 2014 2015 2016 2017 Number of Articles Governor Local Ballot Local Government State Ballot State Court State Legislature State Agency

PAGE 60

48 activity may be calmer in other venues. It is not clear , though, whether this is because other venues are waiting on decisions or because advocacy groups lack the capacity to work across multiple venues at once. With respect to advoc acy group involvement, Figure 4 shows that across most venues, we see similar levels of engagement by both environmental ad vocacy groups and industry advocacy groups. This suggests that alignment between group ideology and dominant venue ideology may not be a primary driver of venue selection by advocacy groups . That is, for example, if state agencies tended to favor environ mental groups and the state legislature tended to favor industry groups, we would expect to see more environmental group activity in state agencies and more industry group activity at the state legislature, at least if advocacy groups are primarily proacti ve about their activities. Results below, however, do not support ideology as a major driver of venue selection.

PAGE 61

49 Figure 4 . Activity Across Venues by Advocacy Group. The only venue in which there is a dramatic difference between the two coalition s is in state courts. A mong the venues involved in this issue, state courts have the highest barrier to entry because of the resources and expertise needed to participate and because participation is restricted by courts. O nce litigation has commenced, it is up to courts to decide whether other groups may join the litigation. Furthermore, as Chapter 3 shows, outside advocacy strategies are least likely in judicial venues, meaning that participation is primarily limited to those directly involved in litigatio n. That is, as described more fully in Chapter 3, in contrast to when policies are pending at regulatory agencies or at the legislature, when issues are before courts, advocacy groups are less likely to run media campaigns, protest, or engage in other gra ssroots efforts. Thus, these results lend support to existing studies that find a relationship between venues and organizational resources. 0 10 20 30 40 50 60 Governor Local Ballot Local Government State Ballot State Court State legislature State Agency Environmental Industry

PAGE 62

50 Patterns in individual v enues. Next, by breaking activity apart by venue , and comparing all activity (reflected by number of news articles), industry group activity , and envir onmental group activity , we can see patterns of advocacy group act ivity in each venue. Again, data derive from the 209 news articles coded as either ban/moratorium or local authority. W hile i n some cases, activity may be led by advocacy groups, in other cases, advocacy groups seem to be merely reacting to activity in the venue or to activity by the other coalition. The results for each ve nue are available in Appendix C , and a few examples are shown below. First, activity at the legislature over the five year period demonstrates that activity may be happening independently of advocacy groups. For example, Figure 5 shows a spike in activity in early 2014, and at least based on news media cover age, some activity occurs independently of advocacy groups. Furthermore , at least with respect to some activity, legislators may have played a dominate role. For example , the F igure shows that another spike in activity began in mid 2015 before advocacy g roups got involved, thereby suggesting that advocacy groups may have reacted to existing activity. Figure 5 . Local Authority Related Activity at the State Legislature. Activity at the state legislature should also be viewed in contrast to state ballot m easures where all activity involved advocacy groups , as shown below in Figure 6 . Perhaps because the 0 1 2 3 4 5 State Legislature # Articles Environmental Groups Industry Groups

PAGE 63

51 venue does not exist independently of the ballot measures, those within the venue cannot initiate action. Also, while individuals or small groups may hav e the capacity to initiate a local ballot measure, because the state ballot measures require greater resources, advocacy groups are likely to be invo lved. Accordingly, we do see some evidence of activity not involving advocacy groups with local ballot mea sures , as shown below in Figure 7 . Figure 6 . Local Authority Related Activity at the State Ballot Box . Figure 7 . Local Authority Related Activity at the Local Ballot Box . Finally, at the state court, we see a similar pattern in the sense that little activity occurs that does not involve advocacy groups. See Figure 8 below. In the case of the COGA v. Longmont lawsuit, there is no question that COGA initiated the action, an d as shown above, industry groups were more active than environmental groups, likely due to barriers associated 0 5 10 State Ballot # Articles Environmental Groups Industry Groups 0 2 4 6 8 Local Ballot # Articles Environmental Groups Industry Groups

PAGE 64

52 with the venue. Thus, while groups follow a similar pattern of activity, engagement by environmental advocacy groups is typically at a lower le vel than industry group activity. Figure 8 . Local Authority Related Activity at State Courts . Thus, through looking at activity across venues in a particular subsystem and then with respect to a specific policy issue, a few lessons may be gleaned. A major results are presented in the next section. Discussion The results in this study pain t a picture of the policy activity within a given subsystem and of the venues and actors at p lay in that subsystem. T he results suggest th at in a high conflict policy subsystem such as the one in this study , numerous sub issues a nd venues may be involved. As such, advocacy groups have numerous options for how to influence policy. Not only are many venues in play, but across venues, groups can choose from an array of ways to influence the outcomes in those venues. While these options provide opportunities, they also venues. T his study, which is li mited to a five year period, identifies over 20 distinct issues a nd 11 venue types in the subsystem . Not only does this indicate the complexity of the policy 0 5 10 15 State Court # Articles Environmental Groups Industry Groups

PAGE 65

53 environment, providing opportunities but also forcing challenging strategic decisions, but it als o offers fertile ground for studies related to polycentricity and conflict over extended time period s . That is, because policy subsystems may be comprised of a broad range of venues and sub issues, advocacy groups must devote significant resources to thei r advocacy activities, and only rarely see an end to issues. If groups are on the losing side of an issue, they may look for opportunities to expand the conflict or move it to other venues, and if they are on the wining wide, they must be prepared to play defense in some other venue. With respect to actors, this study zooms in on advocacy groups with conflicting perspectives: industry advocacy groups and environmental advoc acy groups. The results indicate that both types of groups are active in the policy subsystem, each appearing in more than one third of the sample articles. Furthermore, results suggest that these groups particip ate at similar rates across venues and policy issues. More differences between the groups exist when comparing their activity by venue, as compared with policy issue, thus suggesting that venues are, at least to some degree, a motivating factor in how advocacy groups participate in subsystems. rting points for future studies. First, this study provides some support for the notion that cumulative activity in a policy subsystem across venues may remain relatively stable over time. As shown above, for example, when state court activity increased in the oil and gas subsystem in Colorado , local government activity decreased. Also, while the data show a clear peak in news media coverage of the subsystem in 2014, period. As such, future studi es should consider whether fluctuations in activity happen as a

PAGE 66

54 result of capacity issues, the fact that venues may await decisions in other venues before taking action, or both. Second, two results combine to suggest that factors beyond those typically id entified in the literature may be at play in explaining venue selection . The results do not lend support to the idea that ideological alignment between group and venue is a factor in venue participation, because advocacy groups with conflicting ideologies participate at similar rates across venues. Further research is needed to tease out this relationship, though, because it may be that ideology is important in initial venue selection, but that much of venue participation is driven by reactions to opponen (especially inside versus outside strategies) that groups use in each venue. This is not to say that ideology is not a factor, but rather, there may be other factors at play, such as reactivity or differences in strategy selection that are not picked up in this study. The results do support existing knowledge about the role that resources play in venue selection . While across most venues environmental advocacy groups and ind ustry advocacy groups participate at similar r ates, industry advocacy groups we re significantly more active at state courts, which have the highest barriers to entry. An alternative explanation, though, may be that resources dictate not whether advocacy g roups participate, but how they participate. That is, as Chapter 3 suggests, nonprofit organizations are least likely to use outside strategies with respect to courts. Thus, taken together these results suggest that future studies should consider the interaction between venue selection and the use of insider or outside advocacy activities. Third, results also suggest that organizations may be at least as reactive in venue selection as they are proactive. Two piece of evidence support this finding. First, there is little variation between coalitions with competing ideologies in terms of their participation in venue s .

PAGE 67

55 Second, examining activity in specific venues shows some evidence that activity sometimes precedes advocacy group involvement. Thus, t he venue selection literature would benefit from further exploration of venue selection as a reactive strategy. It is also not clear based on the existing study whether reactivity occurs in relation to venue activity, opponent venue selections, or both. A few limitations are associated with this study that may provide avenues for future research. First, as noted above, by relying on news media, this study may not capture all subsystem activity and may be less inclusive of events and advocacy groups that o ccur and work primarily outside of the public eye . Thus, future studies might combine news media with other data sources, such as interviews or reviews of legislative or administrative records. Second, because this study focuses on a single policy subsys tem , it is not clear the extent to which results are generalizable. While the use of case studies for theory development is necessary and appropriate , future studies should extend to other contexts as well . Finally, this study collectively examines group s with similar ideologies without accounting for the wide variation that exists among these groups. As noted at the outset , these groups include formal nonprofits, informal groups, and action organizations. Not only may there be wide variation in how the se groups group individually advocate, but decisions about venues and advocacy in one groups will likely affect how the other groups advocate. Thus, future studies should look at differences between organizations within a particular ideology as well. Conc lusion This study maps out a complex and high conflict policy subsystem: oil and gas policy in Colorado. By examining how actors with competing viewpoints interact with venues, we see how venues, issues, and actor participation evolve over time. While s ome results confirm

PAGE 68

56 existing venue shopping literature, others suggest new areas of inquiry. As a starting point, the next chapter examines the inside and outside strategies that nonprofit actors use to influence policy across venues.

PAGE 69

57 CHAPTER III HOW DO ENVIRONMENTAL NON PROFITS ADVOCATE? AN EXPLORATION OF IN SIDE AND OUTSIDE STR ATEGIES ACROSS VENUE S Abstract This chapter explores the advocacy strategies that nonprofit organizations use to influence environmental policy across five venues : legislatures, regulatory agencies, courts, ballot boxes, . Despite widespread agreement on a definition of advocacy, operationalizing the concept presents challenges. This chapter proposes a venue based approach to understanding advocacy and fil ls gaps in the literature by describing the activities environmental nonprofits use across venues. Data are based on se mi structured interviews with 22 environmental nonprofits across three states : Colorado, New Mexico, and Oklahoma . Results illuminate t he venues and strategies most common to these organizations, along with the organizational and contextual determinants of nonprofit decisions regarding whether and how to get involved in policy advocacy. Keywords Environmental policy, nonprofit advocacy , policy venues, advocacy strategies

PAGE 70

58 Building on the prior chapter, which examined patterns of advocacy group participation across venues, this chapter develops a deeper understanding of the specific types of advocacy that environmental nonprofits use ac ross different venues. Thus, this chapter describes advocacy activities in each of five venues described in the previous chapter : regulatory agencies, the existing nonpr ofit literature suffers from limitations, including that studies of nonprofit advocacy to date primarily focus on social service providers; advocacy is often operationalized using blunt, dichotomous measures; and assumptions often inhere in the extent to w hich different types of organizations advocate. In an effort to address some of these lim itations, this study presents an exploration of the advocacy activit ies of environmental nonprofits in each of five specific venues and the factors that influence dec isions regarding advocacy . This paper explores two research questions. First, what activities do environmental nonprofits engage in across different venues to influence the policy process? Second, what factors influence the types of advocacy these organi zations choose? This paper begins with a review of relevant literature, including the importance of nonprofit advocacy and research to date on advocacy strategies. Next, I present the context for this study: hydraulic fracturing policy debates in Colorad o, New Mexico, and Oklahoma. This paper then describes the interviews used for data collection, and the results, including how environmental nonprofits advocate in different venues and the factors that may affect differences in advocacy. The paper conclu des with a brief discussion of the findings, limitations, and potential avenues for future research. Literature R eview While studies about advocacy activities are not new, recent years have seen a surge in nonprofit advocacy research. For example, recent scholarly work has focused on the advocacy

PAGE 71

59 activities of human service nonprofits and the drivers of those activities (Mosley 2010; Almog Bar and Schmid 2014), how nonprofits influence policy and effect policy change (Hula and Jackson Elmoore 2001; Andrews and Edwards 2004; Bryce 2006; Fyall and McGuire 2015; Fyall 2016), and ways of understanding and classifying nonprofit advocacy organizations and their activities (Boris and Mosher Williams 1998; Child and Grønbjerg 2007). Despite this growing body of li terature, current knowledge remains limited, particularly because advocacy is difficult to define and operationalize. After an overview of nonprofit advocacy, this section describes literature on advocacy strategies , factors that influence strategies, and limitations of existing studies. Nonprofit A dvocacy While advocacy is a relevant concept across a variety of literatures, including political science, public policy, nonprofit management, and social movements, the concept remains difficult to operationalize. From a definitional standpoint, advocacy is suggests, advocacy is a broad concept th at may include direct attempts to influence policy, such as lobbying, litigating, or testifying before decision making bodies, and indirect efforts such as education, media campaigns, or voter mobilization. Historically, studies of nonprofit advocacy have focused primarily on rights based and political organizations, but calls for broader conceptualizations of advocacy, such as that by Boris and Mosher Williams (1998), have led to a proliferation of research that recognizes the breadth of forms that advoca cy can take (e.g., MacIndoe 2014; Almog Bar 2018). Nonprofit advocacy is important for at least two reasons. First, nonprofits hold potential

PAGE 72

60 organizations contribut e to democratic governance by representing the interests of citizens and nonprofits is to provide a voice for those whose interests might not otherwise be represen ted (Child and Grønbjerg 2007; Almog Bar and Schmid 2014; Mosley 2016). In the 1830s, Alexis de Tocqueville, observing the US, noted that associations are essential in democratic societies to overcome the fact that independent citizens otherwise lack poli tical power (de Tocqueville 1966). More recently, according to Putnam (1993), voluntary associations help build social capital, and nonprofits help secure citizen rights, including political, civil, and social rights (see also Smith 2007). At the same ti me, though, o thers describe that nonprofits may promote special interests and exert disproportionate influence on policymaking (Almog Bar 2018; Boris 2006). That is, rather than strengthen the voice of those lacking political power, some nonprofits actual ly amplify the voice of the wealthy and powerful. Second, nonprofit advocacy is relevant to studies of environmental policy and related between public policy and its surrounding actors, events, and contexts, as well as the policy or Multiple Streams Framework (MSF), the Punctuated Equilibrium Theory (PET), the Advoc acy Coalition Framework (ACF), and the Narrative Policy Framework (NPF), contemplate a role for nonprofits (Fyall 2017; Weible and Sabatier 2018). For example, in the MSF, nonprofits may participate in policy communities or act as entrepreneurs (Mason 201 4), and in t he ACF, nonprofits participate in coalitions to influence change (Fyall and McGuire 2015). While these studies are informative about advocacy, they typically do not examine the nonprofit sector

PAGE 73

61 specifically, focusing instead on coalitions or p olicy entrepreneurs from across sectors (e.g., Mintrom and Norman 2009; Weible, Heikkila, deLeon, and Sabatier 2012). Advocac y A ctivities Numerous studies to date across political science, interest group, social movement, and nonprofit literatures discuss the advocacy strategies of different groups. Among the most common ways to classify advocacy activities is as inside or outside strategies (e.g., Gormley and Cymrot 2006). To align with the common definition of nonprofit advocacy, these might also be con sidered direct and indirect strategies. Gais and Walker (1991) describe that interest groups uence a specific policy decision or raise awareness around an issue that may lead to policy change in the future. While scholars have since shown that many organizations focus on both types of strategies (e.g., Mosley 2011; Almog Bar 2018), the distinctio n remains useful in describing advocacy activities. Specific examples of inside strategi es include lobbying, drafting policies, participating in government task forces, and testifying before decision making bodies (Gais and Walker 1991; Mosley 2011). Exa mples of outside strategies include talking to the media, protesting, demonstrating, participating in coalitions, educating the public through speakers and events, and issuing reports (Gais and Walker 1991; Mosley 2011). Perhaps recognizing the difficultie s with fitting strategies into these two categories, some scholars have modified or expanded these categories. For example, Mosley (2011) distinguishe s between insider tactics, indirect tactics , and outsider tactics , where insider tactics are those that i include various activities that do not require connections with policy elites (439 440). She

PAGE 74

62 describes outsider tactics, such as protests and boycotts, as a s ubset of indirect tactics. Furthermore, Gormley and C ymrot (2006) rely on the inside outside strategies concept , but then also include categories that fall between these two typologies, such as coalition work and public policy research. Others have gone c ompletely beyond the inside outside dichotomy and developed more nuanced advocacy typologies. For example, Pekkanen and Smith (2014a) distinguish between lobbying, policy advocacy, protesting, and public education. Berry and Arons (2003) describe nine ma jor activities that 501(c)(3) organizations engage in, including more legislative and aggressive tactics like testifying at hearings or lobbying, and more administrative and less aggressive tactics like meeting with government officials or working in plann ing or advisory groups. Also, Casey (2011) outlines 28 different types of advocacy in eight categories: legal, legislative and administrative, research and policy analysis, coalition building and capacity development, education and mobilization, communica tion and media outreach, government relations and oversight, and service delivery (see also Schlozman and Tierney 1986; Almog Bar 2018, 414). Many of the policy process theories described above also suggest advocacy activities that actors, including nonpr ofits, may engage in to influence policymaking. For example , Brouwer and Huitema (2018) describe ten advocacy activities across four categories that policy entrepreneurs use to influence water management in the Netherlands. These include a ttention and su pport seeking strategies , like demonstrations, rhetorical persuasion, and exp loitation of focusing events; l inking strategies , such as coalition building, issue linking, and game linking; r elational management strategies , including networking and trust bui lding; and arena strategies , such as venue shopping and timing. Also, Gen and Wright (2018) describe how advocacy

PAGE 75

63 activities, like motivating the public, lobbying, and building coalitions, are consistent with various policy theories. Furthermore, a few studies suggest activities specific to hydraulic fracturing policymaking, which is the focus of this study. Weible and Heikkila (2016) examine the activities of policy actors from across sectors involved in hydraulic fracturing conflicts in New York, Colo rado, and Texas. In New York, they asked survey respondents about the importance of engaging in a list of activities, and in Colorado and Texas, they asked about the frequency with which actors participate in each activity. Results suggest that across al l three states, fracking opponents rated activities higher (either in terms of importance or frequency of engagement) than did fracking proponents. Among the more important or frequently used activities are forming and maintaining coalitions with allies, participating in or organizing public meetings, posting information online , and communicating with the news media. Other studies focus on the importance of discourse and competing framing efforts as central to hydraulic fracturing politics (see Dodge 2017 ; Dodge and Lee 2017). Each of the studies described will inform this research by providing useful starting points and prompts for understanding advocacy in the nonprofit context. The studies above, however, typically focus on policy actors or coalition s rather than on nonprofits in particular. Also, while scholars use a variety of tools to determine whether advocacy is present, ultimately studies typically reduce advocacy to a single dichotomous variable reflecting whether organizations advocate at all (e.g., Pekkanen and Smith 2014b; Child and Grønbjerg 2007) . Thus, this study zooms in on different types of advocacy across venues in the specific context of environmental nonprofits and environmental policymaking.

PAGE 76

64 This paper argues that to understand advocacy and its impact on public policy, we need to improve our understanding of how organizations advocate in specific venues. As Buffardi, Pekkanen, and Smith (2015) point out, while much is known about the types of organizations that advocate and the activities that they engage in to influence policy, little is known about the ve nues in which advocacy occurs. In studies to date, activities tend to not be grounded in a particular venue or discussed with a particular policy type (e.g., agency rule, legi slation, ballot measure), but studies often report the frequency with which organizations report engaging in different activities (e.g., Buffardi, Pekkanen, and Smith 2017). For these results to be more meaningful, activities should be linked with particu lar policy types or venues. That is, perhaps frequencies are dictated more by policy prevalence than by organization preference or competence . Furthermore, certain activities are necessarily linked to pending polices, such as testifying before legislatur es, while others activities may be linked to pending policy or to broader attempts to influence public thinking. By breaking advocacy down into component measures and making it venue specific, studies will benefit from more valid variables. Factors That I nfluence A dvocacy As described in Chapter 1, t o determine what drives organizations to engage in different types of activities, scholars most commonly depend on institutional theory and resource dependency theory (Mosley 2011). According to institutional theory, in order to survive, organizations must conform to the rules, norms, and social expectations that make up their institutional environments (Guo and Acar 2005; see also Meyer and Rowan 1977). As such, the innerworkings of nonprofit organizations mu st be understood in the context of these environments (Feeney 1997). Institutional theory has been widely applied in nonprofit studies and used to explain decisions about advocacy (e.g., Schmid, Bar, and Nirel 2008; Suárez and

PAGE 77

65 Hwang 2008). Next, resource dependency theory suggests that organizations require a variety of resources to survive, and they therefore interact strategically in their environments to secure these resources (Eikenberry and Kluver 2004; see also Pfeffer and Salancik 1978). Resource dependency theory is also regularly invoked in studies about why and how nonprofits advocate (e.g., Nicholson Crotty 2009; Mosley 2010; MacIndoe 2014; Fyall 2016). Building on these theories, studies have identified a number of internal and external variab les that influence advocacy across nonprofit organizations or interest groups. For example, according to Gais and Walker (1991), choice of political strategies are a function of four factors: internal organizational resources; 3) membership characteristics; and 4) principal sources of funding (111). Also, Mosley (2011) includes a series of variables in three categories: institutionalization, dependence on government funding, and capacity. Lik ewise, Guo and Saxton (2010) found a relationship between certain aspects of constituent representation and the scope and intensity of advocacy among a general sample of nonprof it organizations in Arizona. While much can be g leaned from existing research, these studies do not examine the specific factors that influence environmental nonprofit advocacy, and there are differences between the environmental nonprofits and other types of nonprofits. For example, many nonprofits often focus exclusively on advocacy; nonprofits advocate across a wide range of venues and issues; and nonprofits face opponents with very deep pockets, such as oil and gas companies and electric utilities. Thus, this study builds on existing research by exploring the factors t hat influence advocacy across venues specifically in the environmental policy context.

PAGE 78

66 Study Context : Oil and Gas Development To improve our understanding of the advocacy activities that environmental nonprofits use across venues, this study relies on interviews with organizations active in hydraulic fracturing (fracking) policy debates in the US. Fracking is a technique involving the injection of water, sand, and chemical additives into shale rock or other porous formations thousands of feet beneath t around much of the world, fracking is increasingly used in oil and gas development. For example, between 2007 and 2017, the amount of natural gas produced from shale f ormations increased 10 fold from 1,990,145 million cubic feet to 19,018,457 million cubic feet (EIA 2018). Not only has the volume of oil and gas produced through fracking increased, but oil and gas development is also encroaching on high population dens ity areas. Close proximity between population centers and oil and gas operations gives rise to a host of local issues, including NIMBYism (not in my backyard); nuisance issues, like traffic, dust, and noise; and concerns about air and water contamination. As a result, communities grapple with how to manage fracking, and some have attempted to ban it. For example, while the state of New York successfully banned fracking, local bans in a number of municipalities like Denton, Texas, and Longmont, Colorado, have been struck down by courts or legislatures (Knight and Gullman 2015; Heikkila and Weible 2016; Jaquith 2017). Local communities with limited authority over oil and gas development may attempt to exercise control by negotiating agreements with oil and gas companies (Zilliox and Smith 2017), regulating nuisance issues, and advocating for increased setbacks and other limits at the state level. Beyond local issues, fracking is part of broader national and global debates about reliance on fossil fuels as energy sources. While those in favor of fracking often cite natural gas as a cleaner alternative to oil and coal, those opposed

PAGE 79

67 claim that fracking and the availability of cheap oil and natural gas impede transitions to renewable energy sources, like wind and solar. Because they involve a wide range of issues across venues and levels of government, fracking debates provide a good context for this study. Fracking is an intensely debated political issue (Heikkila and Weible 2016), drawing participation from industry, government, and nonprofit organizations. Fracking policies are created via judicial decisions, legislation, agency rules, executive orders, and ballot measures, and organizations, including environmental nonprofits, engage in a variety of activ ities and strategies to influence each of the sources of policy. As such, fracking is an ideal policy context for observing maximum potential advocacy activities. Organizations selected for this study are environmental nonprofits operating in three states : Colorado, New Mexico, and Oklahoma. These three states were selected because each is ranked in the top 10 for both shale gas and crude oil production, and all three have mid sized populations, which facilitate understanding policy state wide. The three states also have differences, though, that provide useful variation and facilitate drawing inferences about the contextual factors that may influence decisions about advocacy. That is, while the issues surrounding oil and gas, including local government authority and air and water pollution, are similar across states, contextual factors lead to differences in venues and in strategies . For example, Oklahoma is a R epublican stronghold state, while both New Mexico and Colorado have been leaning D emocrat in recent years. Furthermore, while Oklahoma has very little public land, both Colorado and New Mexico are comprised of about 35% federal public land. Oil and gas development in Colorado, however , t ypically occurs on private land and near population centers , while development in New Mexico is often on public land in more rural areas.

PAGE 80

68 Furthermore, while both Colorado and Oklahoma allow citizen initiated ballot measures to make statutory and constitutional changes, New Mexico only allows referenda. A compari son of the thre e states is provided in Tabl e 6 . Table 6 . Comparison of State Characteristics. Colorado New Mexico Oklahoma shale formation Niobrara Barnett Woodford, Lewis Woodford 2019 governor D emocrat D emocrat R epublican 2019 state house D emocrat D emocrat R epublican 2019 state senate D emocrat D emocrat R epublican 2018 governor D emocrat R epublican R epublican 2018 house D emocrat D emocrat R epublican 2018 senate R epublican D emocrat R epublican population size (millions) 5.61 2.09 3.93 % federal land 35.90% 34.70% 1.60% allow ballot measures? initiated statutes and amendments referendum only initiated statutes and amendments shale gas production (billion cubic feet in 2017) 97 592 1290 shale gas state rank 8 7 4 crude oil production (1000 barrels in 2017) 130732 171440 161678 crude oil state rank 7 5 6 Methods This study relies on qualitative interview data to complement existing studies on nonprofit advocacy activities, which typically rely on survey data (e.g., Bass et al. 2007; Berry and Arons 2003). A s Mosley (2011) describes, a mong the limitations of survey data is that categories are established at the outset and thus necessarily limited to a subset of possible advocacy tactics. By diving deeper into advocacy activities through in terviews, respondents are able to describe the full range of activities they engage in across venues.

PAGE 81

69 Data were collected from November 2018 through May 2019 via in person, Zoom, and phone interviews with representatives of organizations active in hydraulic fracturing policy debates in Colorado, New Mexico, and Oklahoma. For each organization, interviews were with a senior executive staff member or with the staff member most responsible for oil and gas policy. This paper presents the results of 22 interviews : 9 in Colorado, 8 in N ew Mexico, and 5 in Oklahoma. Interviewees were selected using snowball and purposive sampling methodologies to identify thos e organizations most actively engaged in oil and gas policy each of the three states. Interviews were semi structured. Respondents were asked to identify roughly the last three oil and gas policy issues in their state that their organization was involved in, 9 and were then asked a series of questions about their advocacy activities related to each. The interview questions relevant to this s tudy are included in Appendix D . interviews were recorded and transcribed. Intervie ws were then coded using thematic analysis , which focuses on identifying themes that represent data (Nishishiba, Jones, and Kraner 2014). office) and then activ ities within each venue were identified. Through an iterative process, I identified major themes among activities and then sorted activities by theme. Results are presented in the next section. Results This section presents the results of interviews and the advocacy activities that respondents report engaging in across venues. Table 7 below shows an overview of the results from the coding process described above. Once activities were sorted by venue an d theme, the 9 In some cases, respondents also described broader energy related issues or issues tangential to oil and gas policy.

PAGE 82

70 least one interviewee mentioned using activities in the relevant theme venue. The subsections that follow provide more detailed results for each v enue, and the final code form reflecting the intervi ews is available in Appendix E . Overall , the results present a snapshot of recent advocacy activities that organizations report using. Before presenting detailed results by venue, this section describes each category of activity and overarching trends. Table 7 . Advocacy Strategy Overview. Type Major Theme Regulatory Agencies (19) Legislature (16) Courts (8) Ballot Box (5) Office (4) Inside Informal Meetings X X X Formal Participation X X X Hybrid Policy Prep X X X Coalition Work X X X X Litigation X X Research X Outside Member Communication X X X X X Media X X X X X Institutional Change X X Calls for Action X X X X Official Position X X X Public Efforts X X X X X Note: X indicates empirical support for activities used in the relevant theme venue. The numbers in parentheses in the first row reflect the number of respondents that reported engaging in each venue. Consistent with recent literature about nonprofit advocacy activities, this paper uses reflect the reality that some strategies are difficult to categorize as ei ther inside or outside. Within each of these three umbrella categories, a few subcategories based on interview responses are presented. Beginning with inside strategies, again, activities in this category depend on

PAGE 83

71 direct access to decision makers, typic ally either through existing relationships or status. Inside strategies can be broken down into Informal Meetings with decision makers or their staff and Formal Participation . The latter refers to situations in which organizations or individuals are invi task forces. Outside strategies, on the other hand, include a more diverse group of activities. Member communication refers to activities that organizations enga ge in in relation to their members, however defined. Media includes host of activities, typically divided into three categories: earned media, such as talking to reporters or submitting op eds; social media; and paid media. Institutional change relates t o efforts to change the decision making authority related to a particular policy or venue. Organizations also often make calls for action from others, such as the general public or local leaders. At time s , organizations not involved directly in policy ac tivities may opt to take an official position on a measure, and finally public efforts include a wide range of activities that occur in the public sphere, such as protesting, demonstrating, rallying, and hosting or participating in educational events. Fin ally, the hybrid category includes a series of activities that do not squarely fall within concepts of inside or outside strategies. These include policy preparation , which may occur at the request of decision makers or outside of decision making venues, often through the work of coalitions. In addition, a host of coalition work may occur in relation to policies, and while some members of a coalition may have direct access to decision makers, others do not, and thus, coalition members coordinate activitie s and determine how best to leverage combined resources to influence policy. The literature has alternatively described litigation as both an internal and an external strategy. On the one hand, those dissatisfied with a given venue may use courts as an

PAGE 84

72 e xternal resource to influence the primary venue. On the other hand, though, participating in judicial activities require specialized knowledge and skill, and courts may set policy independently of other venues, such as by interpreting other laws. Finally , research has also been characterized as both an internal and an external strategy, depending at least in part on who commissions the research and who is the target audience. As Table 7 suggests, regulatory activity was most commonly discussed by respond ents in this study and respondents report using the broadest range of activities to influence policy in this venue. Collectively, though, respondents report activity across venues, and while some common patterns exist, there are also differences across ve nues. Each of the specific venues is discussed in turn. Venue: Regulatory First, during interviews, respondents discussed activities in regulatory agencies most frequently . Not only was this the mos t common venue discussed with 19 out of 22 respondents across all three states mentioning activity in this category, but respondents also mentioned the broadest range of agency policies and activities here. For example, in Oklahoma, one respondent mentio state agenci es on seismicity and produced water issues. In New Mexico, respondents mentioned getting involved in rulemaking at the BLM and regulatory proceedings at the state Public Regulation Commission (NMPRC). Finally, in Colorado, respondents mentioned a variety of agency policies, primarily at the Colorado Oil and Gas Conservation Commission (COGCC), such as the distance between oil and gas wells and school property (school setbacks), but also at other agencies, like the Air Pollution Control Division of the Col orado Department of Public Health and Environment. Activities in each cate gory are shown below in Table 8 . The third column presents a summary of the activities described in

PAGE 85

73 each category. For all venues, more detailed results, including responses by st ate and organization type are included in Appendix E . Table 8 . Regulatory Agency Strategies. Type Major Theme Summary of Activities Described in Interviews Inside Informal Meetings Provide support to agencies; meet with regulators/agency staff; lobby Formal Participation Participate in rulemaking; provide expert testimony/technical information Hybrid Policy Prep Participate in/convene work groups; draft guidelines; petition for rulemaking Coalition Work Participate in stakeholder meetings/phone calls; work with Native groups; support partners; brief other groups Litigation Administrative litigation Research Produce fact sheets, reports; commission research Outside Member Communication Email blasts; member updates Media Earned media, social media, paid media Institutional Change Force action through legislature, sue to force or prevent action Call for Action Encourage members to attend/speak at meetings, submit comments; request action from local leaders Official Position Comment on rules; file formal protests Public Efforts Help citizens file comments; host/participate in educational grasstops organizing; demonstrations and rallies; publish info on website As Table 8 suggests, r espondents described a broad range of inside strategies. Most directly, respondents participated in rulemaking proceedings at various state agencies, provided expert testimony, provided technical information, and supported agencies during rule making. Some respondents also mentioned meeting with agency staff outsi de the context of rulemaking or other formal meetings. Prior to rulemaking, not only do organizations sometimes petition agencies for rulemaking, but respondents often participate in w ork groups or stakeholder meetings to draft guidelines or propose rules. Within coalitions, some organizations work with Native groups or provide support to other organizations more directly involved in rulemaking proceedings.

PAGE 86

74 Respondents also report ed p roducing or commissioning research and reports. A more limited number of organizations engage in administrative litigation to challenge agency rules or actions. With respect to outside strategies, organizations engage in a range of media strategies, inclu ding earned media, such as press releases, writing op eds and letters to the editor, and talking with reporters; social media , like Facebook and Twitter; and in more limited circumstances, paid media, such as newspaper and radio ads. Organizations communi cate with their members through email blasts and make calls for action from their members and local leaders. They file comments in response to agency actions and encourage and support others in doing so. Organizations also participate in a range of publi c activities, such as hosting community meetings and educational events, organizing grasstops, demonstrating and rallying, and publishing information on their websites. Venue: Legislature The majority of respondents (16 out of 22 ) mentioned involvement in legislative advocacy. In Colorado, the vast majority of legislation discussed was at the state level, including legislation proposed in prior legislative sessions, such efforts to amend school setbacks and forced pooling laws. Also, many interviews took place during the 2019 legislative session, and in Colorado an omnibus oil and gas bill (SB19 181) was introduced and later passed to refocus the mission of the COGCC to prioritize healthy and safety; give local governments a stronger role in overseeing oi l and gas operations; change forced pooling laws; and create financial assurances and bonding requirements for oil and gas permits to combat issues with orphan wells (Conservation Colorado 2019). Thus, a few interviewees discussed their involvement in the pending legislation. In New Mexico, interviewees most frequently talked about the federal Chaco Cultural Heritage Area Protection Act and its creation of a 10 mile no drilling buffer zone

PAGE 87

75 around Chaco Culture National Historical Park in northern New Mexi co. Fewer respondents mentioned specific legislation in Oklahoma, and at least according to one respondent, environmental nonprofits in Oklahoma often play defense against policies backed by the oil and gas industry. Advocacy activity r es ults are shown b elow in Table 9 . Table 9 . Legislative Strategies . Type Major Themes Summary of Activities Described in Interviews Inside Informal Meetings Lobby; meet with/convene legislators, staff, governor; educate legislators Formal Participation Testify at hearings Hybrid Policy Prep Draft legislation; offer input on policy development; convene/participate in groups to create policy proposals/recommendations Coalition Work Meet in coalitions to discuss policies/strategies; support others Litigation Research Outside Member Communication Update members; email blasts; draft member petitions Media Earned media; social media; paid media Institutional Change Influence electoral politics; shift to regulatory venue Call for Action Encourage leaders to lobby; encourage people to call/email their legislators or the governor Official Position Public Efforts Have impacted citizens testify; organize town halls/educational events; rally; collect signatures and deliver petitions; train advocates; post info online As with regulation s , advocacy activities at the legislature consisted of inside, hybrid, and outside strategies. With respect to inside and hybrid strategies, respondents frequently mentioned lobbying; informally meeting with or educating legislators; convening meetings between legislators and other stakeholders; testifying or otherwise participating in hearings; and even drafting or consulting on policy development. Respondents also often reported participating in coalitions to discuss po licies, come up with policy proposal s, and strategize . Outside activities include requesting action from others such as by encouraging local leaders to lobby; having impacted members testify at hearings; asking members to email or call their legislators; and setting up meetings between citizens and legislators. Respondents reported

PAGE 88

76 engaging in a range of media activities. A few respondents also mentioned efforts to support holding training events. Some unique strategies came up with respect to institutional changes at the electoral politics. Frustrated by inaction at the legis lature, this organization turned its attention to electoral politics and changing the composition of the legislature in hopes of creating better circumstances for future legislative sessions. Second, another respondent described a venue shift to the COGCC following inaction by the split Colorado legislature. Specifically, with respect to school setbacks in Colorado, after failing at the legislature, organizations successfully petitioned the COGCC for a rulemaking on the subject. Venue: Courts Courts prese nt an interesting venue in that they are sometimes used to force action in other venues, but they also establish policy in their own right by setting precedent through judicial opinions, declaring that laws or rules are invalid, or interpreting policies. Indeed, a number of environmental nonprofits actually focus specificall y on legal activity, and 8 of 22 interviewees mentioned activities in court. As compared with other venues, respondents were most clear about whether they engage in judicial activities . For example , according to one respondents typically talked about state level actions. For example, in 2014, a group of teenagers, supported by the nonprofit Earth Guardians, sued the COGCC in an effort to force the agency to prioritize the environment and public health and safety when issuing oil and gas New Mexico, on the oth er hand, respondents tended to talk about the federal level, such as

PAGE 89

77 lawsuits to block federal land leases or to force compliance with NEPA. Activities across catego ries are shown below in Table 10 . Table 10 . Judicial Strategies. Type Major Themes Summar y of Activities Described in Interviews Inside Informal Meetings Formal Participation Hybrid Policy Prep Coalition Work Support partners directly involved Litigation File a lawsuit; serve as lead counsel; provide clients/standing declarations Research Outside Member Communication Update members Media Earned media; social media Institutional Change Call for Action Official Position File or sign on to amicus briefs Public Efforts Encourage people to attend oral arguments A s shown above, a number of organizations mentioned directly engaging in litigation by filing a lawsuit and serving as lead or co lead counsel on lawsuits. While public interest law firms often take this lead role, grassroots organizations sometimes serve as parties or provide clients for the litigation. Other organizations mentioned filing or signing on to amicus curiae, or friend of the court, briefs in support of their positions. Organizations also mentioned working in coalitions and supporting those dir ectly engaged in litigation. Outside strategies related to judicial activity are more limited than other categories. Organization s explained that they typically avoid trying to sway opinion about pending litigation and limit their activities to updating their members or the media about major events or decisions in the litigation, although one respondent discussed her organization s efforts to have supporters attend oral arguments or other court proceedings.

PAGE 90

78 Venue: Ballot Ballot measures generally refer to processes through which citizens can vote directly on legislation or constitutional amendments. As mentioned above, while both Colorado and Oklahoma allow citizen initiated ballot measures to make statutory and constitutional changes, New Mexico only all ows referenda through which citizens can vote to repeal a statutory measure enacted by the legislature. With respect to ballot measures, 5 out of 22 interviewees mentioned recent activity with ballot measures. Most notably, a number of respondents report ed activity surrounding two ballot measures that were on the ballot in Colorado in November 2018. First, Proposition 112 would have required 2 , 500 foot setbacks from any structures intended for human occupancy, and second, to counter Proposition 112, the oil and gas industry backed Proposition 74. Proposition 74 would have compensated property owner s for reductions in the value of their land resulting from state laws. Ultimately both measures failed, but only after over $50 million was spent on the campa igns supporting and opposing the two measures. Activities are shown bel ow in Table 11 . Table 11 . Ballot Box Strategies. Type Major Themes Summary of Activities Described in Interviews Inside Informal Meetings Formal Participation Hybrid Policy Prep Draft proposition; collect signatures Coalition Work Campaign assistance Litigation Research Outside Member Communication Updates to members; member petitions Media Earned media, social media, paid media Institutional Change Call for Action Encourage leaders to take positions; seek endorsements Official Position Take an official position Public Efforts Polling/focus groups; create an issue website; canvassing; phone banking; organize educational events/town halls; hand out info

PAGE 91

79 Respondents reported a wide range of activities on ballot measures. First, ballot measures are often initiated by advocacy organizations. To this end, respondents reported drafting the language of propositions and collecting and submitting signatures. On c e a measure is approved for the ballot, campaigns are typically dominated by outside strategies, because ballot measures futile efforts at the state, and in the court, and at our city council meetings, we decided to go ahead and utilize the direct democracy that's available to Coloradans and run a ballot initiative to nd outside strategies primarily consist of what Mosley (2011) describes as outsider tactics, which are a subset of indirect tactics . This include s face to face or in person work, like door to door canvassing; phone banking; holding educational events or t own halls; waiving banners at intersections; and handing out fliers. Respondents also discussed a variety of media work, such as through press conferences, press releases, op eds , letters to the editor, and social media campaigns. Ballot measures also of ten require that organizations, particularly those most associated with the measure, create a w ebsite in support or opposition, and use paid advertising, such as newspaper or digital media advertising and television commercials. Those not directly involve d in ballot campaigns may opt to take an official position. While most executive branch policy comes through regulatory agencies, governors themselves may establish policies through executive orders and overarching plans on specifi c topics, among other things. Ou t of the 22 interviewees, just 4 mentioned trying to influence part of a negotiation to

PAGE 92

80 withdraw two measures from the ballot in 2014, Governor Hickenlooper created a task force on State and Local Regulation of Oil and Gas Operations. The task force issued its recommendations in 2015, some of which were later debated through rulemaking and legislation. Res ults are shown below in Table 12 . Table 12 Type Major Themes Summary of Activities Described in Interviews Inside Informal Meetings Meet with Governor or staff Formal Participation Make presentations; participate in policy groups Hybrid Policy Prep Coalition Work Litigation Research Outside Member Communication Member updates Media Earned media; social media Institutional Change Call for Action Encourage members to talk to the media Official Position Public Efforts Educational events; attend/comment at public meetings; rally With respect to activities in this venue , one organization mentioned having close and, thus, engaging periodically in informal types of policies, organizations mentioned hosting events, both while poli cies are pending and after they ha ve been released to help citizens understand the implications of policies. Particularly with respect to the Colorado task force, organizations mentioned speaking at task force meetings, providing comments, and rallying. Also , like with other policies, organizati ons mentioned engaging with their members through email, newsletters, and mailers, and they mentioned a host of media work, inc luding earned media, such as op eds and letters to the editor, and social media.

PAGE 93

81 Venue Nonspecific In addition to the activities relevant to particular venues described above, organizations are sometimes driven more by overarching issues than by a particular policy proposal . For example, as demonstrated in Chapter 2, advocacy group participation is sometimes driven by the specific sub issues at play in a given policy subsystem. As such, organizations often engage in advocacy efforts tied to particular issues regardless of whether a policy proposal is pending in a particular jurisdiction. Based on interviews with respondents, most of this work occurs in the public sphere. For example, organizations may periodically engage in direct action, such as through rallies or demonstrations. They may attempt to educate the public or particular groups, such as through events, meetings, and information provision. They may engage their membership or broader constituent groups through social media, and some groups periodically meet with decision makers. Furthermore, many organizations regularly engage in coalitions related to specific issues that span a variety of venues. For example, a coalition of Native groups and environmental nonprofit organizations meet regularly to discuss strate gies and issues re lated to the Greater Chaco R egion in northern New Mexico. The Greater Chaco Region contains sacred, archaeological, and cultural sites, and is home to numerous Native Tribes and Pueblos. Because oil and gas issues in the area span a range of venues , incl uding the federal legislature, the BLM, the NM State Land Office, federal courts, and others , coalition members meet regularly to keep each other apprised and assign responsibility for various types of advocacy. Other organizations outside the Chaco conte xt also described efforts to form partnerships or work with other like minded organizations.

PAGE 94

82 Factors That M a y Influence Types of Engagement In addition to describing the types of activities across venues that organizations use to influence policy, intervie advocacy decisions. Existing literature suggests that inside strategies are more common and effective than outside strategies (Berry and Arons 2003; Almog Bar 2018) and that interest gro ups typically rely on inside strategies, while grassroots and social movement organizations tend to use outside strategies (Mosley 2011; Crow et al. 2019). As Mosley (2011) points out, however, relationships and distinctions between group types and advoca cy activities are becoming more complex, and indeed, interviewees in this study report using both inside and outside strategies. For example, in addition to lobbying at the state capitol, the most professionalized groups mobilize their memberships and pos t on social media, and on the other hand, even grassroots organizations leverage connections with decision makers. Thus, this section of the paper presents preliminary findings based on interviews regarding some of the factors associated with advocacy dec isions. Factors are divided into those internal to organizations and those external to organizations. Internal Factors Organizational r esources. As noted above, resource dependency theory posits that organizations interact strategically in their environm ents in order to secure resources , and this theory is often cited in studies regarding advocacy . Accordingly , organizations with greater resources are less concerned about acquiring resources and have more staff time and money available to advocate. Inde ed, this study suggests that resources affect the types of advocacy activities in which organizations engage. Specifically, those organizations with less money and staff time available typically must limit their engagement to the most pressing issues.

PAGE 95

83 Whe n asked about how her organization decides where to put its efforts, one respondent e just can't engage in everything, because we are a small shop, you know, we only have as much time and resources as we have. with fewer resources emphasized both t We're very good at a few things, and we should leave it to the others to do the rest. their efforts, some organizations also focus on training other organizations and individuals to do advo cacy work. Organizational i dentity. in the choices they make about advocacy. While their missions and identities as environmental nonprofits often dictate the types of issues they get involved in, other facets of their identities may influence their choices around venues and activities. Whether organization s identify as grassroots or community organization s influences the types of advocacy activities that they choose. That is, while mo re professionalized environmental nonprofits make decisions internally, lead their own efforts, and inform their members of their activities after the fact, other organizations clearly identify as community organizations and emphasize that their members ta ke the lead in their advocacy activities. While the line between these two types of organizations is not always clear, t he extent to which an organization identifies as one of these two types helps drive decisions about advocacy. As one respondent descri bed: T he members of the organization, the impac ted residents . . . who are members on the ground, they're the ones who dictate our policy and our strategy and tactics. O on ou r campaigns. We always have like issue committees of impacted people, of

PAGE 96

84 members who lead our efforts. . . . And so those are the folks who are turning out to speak, not our organizers. Thus, at least in the case of some organizations, advocacy is more li mited to outside strategies and supporting those most impacted by policies. One the other hand, more professionalized organizations tend to make decisions internally based on their expertise and organizational mission. For example, according to one respon verything we do is driven by th e mission of the organization a nd then the vision that we've set for each one of our programs. t's not that we try to dictate what we believe is best to everyone, but we really do try to come up with the smartest policies we can. This result is also supported by findings in other studies , such as Guo and Saxton (2010), who found a link between representation and advocacy . Thus, we may expect that an organization that primarily ide ntifies as a grassroots or community based organization may approach advocacy differently than an organization without strong connections to its members or constituents. Expertise. colle ctive expertise or on individual staff expertise, influences their decisions about where and And it's also where our expertise i s. We're r eally good on the federal laws: ESA, NEPA, Clean Water, Clean Air. their identities and skills over time. The role of expertise is most pronounced among organizations that identify as environmental or public interest law firm s or among organizations with significant legal

PAGE 97

85 expertise. These o rganizations typically focus on litigation and are often involved in lawsuits or r egulatory actions, particularly on federal land. These organizations may also be more engaged in rulemakings or other inside activities where their legal expertise is as an asset , and they may have more limited involvement in other types of activities. A Our theory of change has been, in the past, entirely in the courts , and it's only recently that we've started organizing , so I would say the vast majority of our work is in the courts. IRS s tatus. The advocacy consequences of how an organization registers for tax exemption under the IRS code are explored more fully in Chapter 4. The in terviews in this study reveal few differences between organizations based on their IRS status. The majority of organizations in this study (14 of 22) are registered under section 501(c)(3), which is somewhat surprising given that 501(c)(3) organizations face some restrictions in their advocacy. Section 501(c)(4) organizations are typically considered advocacy organizations, but only four organi zations in this study are registered under that subpart. Two organizations in Oklahoma are registered under section 501(c)(6) as business or professional groups, which likely reflects the political context in that state as described below, and two organiz ations are not registered for federal tax exemption. The only noticeable difference in advocacy between any of the above categories based on interviews is that a couple respondents mentioned the importance of their 501(c)(4) status because it allows them to engage in electoral politics. These respondents were distinguishing themselves from 501(c)(3) organizations, which are strictly prohibited from supporting or opposing political candidates. External Factors Political c limate. First, state political envi ronment plays a role in determining

PAGE 98

86 environmental nonprofits are directly engaged in oil and gas policy. Not only is it difficult to identify nonprofits in Oklahoma that ar e active in oil and gas policy issues, but the few that are active report very little success from their activities. According to one respondent, very few of an Rather than attempt to engage directly with oil and gas policy at the legislature or otherwise, nonprofits in Oklahoma typically report focusing the ir efforts on general activities safe place to learn and act toward t Furthermore, Colorado provides an interesting within state comparison where organizations on the Front Range often make different decisions about policy th an those on the Western Slope. L arge amounts of drilling happen both on the Front Range and on the Western Slope, but the population is much larger on the Front Range, encompassing cities like Denver and Boulder. Thus, industries are more diverse on the Front Range, and local economies are less produces jobs up here that are important paying jobs, where the workers down on the Front Range are just part of the mix. Amaz on hires more people on the Front Range than oil and gas. Thus, those on the Western Slope tend to focus on more cooperative strategies. Level of c onflict. T he level of conflict surrounding a particular issue influence s the types of advocacy that an organization uses to influence policy. Even within a given policy

PAGE 99

87 subsystem, like oil and gas policy, the amount of conflict varies across time and specific policy issue. Th us, respondents report that when particularly conflictual issues come up, they may turn to different strategies. For example, according to one respondent describing what she meant by That means we try to get more press. I do a lot more writing. I do blogging. I do more outreach. W e meet regularly. We will, if it looks like something bad i s happening, like 3 million more acres and you know, 89 more parcels are coming up, we meet more regularly. We try to t hink of more aggressi ve ways to combat this, because BLM has changed their rules pretty much every other month. We have had to meet more regularly on how to dodge and how to become -they figured they can get rid of us, but they haven't been able to. Also , as noted above, conflict within a given venue may cause groups to shift venues , such as by resorting to ballot measures. These results are consistent with scholarship on conflict and advocacy. For example, a t important factor determining the level of political activity engaged in by interest groups is the amount of conflict they experience in their 1960). Relatedly, Gor mley and Cymrot (2006) found that the presence of enemies in an a particularly strong factor in triggering the use of advocacy activities. Venues. influence how organizations advocate. Most notably, organizations report different approaches to I have a pretty robust earned media strategy in terms of getting that information out there. We do n't want to appear to be trying to influence the court at all. So just getting that out in front of the public and on social media and stuff like that, that it's happening. And then sort of once it

PAGE 100

88 goes into the black ho le of litigation, it's sort of, you know, not much happens. So it's sort of hard to maintain much sort of engagement and interest from our members Beyond courts, though, other venues or stage s of policymaking also influence advocacy. For example, one re spondent noted the importance of inside or more internal strategies with I think for the purposes of federal legislation, it makes more sense to kind of work within a group. spondent said, n organizing, it's best to get the public involved at like comment periods or commenting. Thus, within a given venue or across venues, organizations may use multiple activities in conjunction with one another and as part of an overarching strategy to influence policy. Discussion and Conclusion As the above suggests, environmental nonprofits i n Colorado, New Mexico, and Oklahoma are active across a range of venues in oil and gas or hydraulic fracturing policy. At least in recent years, respondents report being most engaged in regulatory issues, and within regulatory agencies, respondents repor t using the broadest range of activities to influence policy, including inside strategies; hybrid strategies, like coalition work and litigation; and outside strategies, such as media campaigns and direct action. While activity at regulatory agencies and legislatures is somewhat similar, respondents describe different approaches to other venues. For example, when issues are pending before courts, few organizations are directly involved in litigation. Other organizations typically limit their activities t o supporting those directly involved and keep their members and the public apprised. With respect to the ballot box, ballot measures are often borne out of frustration with other venues, and thus, respondents report that their work falls almost exclusivel y in the outside category, including a range of media activity

PAGE 101

89 both inside and outside strategies. Taken together, these results illuminate how activities and venues interact and lay the foundation for deeper explorations of advocacy across policy venues. Indeed, venue matters. There are certainly some similarities acros s venues, but there are also many differences in terms of how organizations advocate. For example, while there are clear differences between courts, ballot measures, and other types of advocacy, even in comparing advocacy before legislatures and regulator y agencies, we see differences. Legislative policymaking happens over a much more condensed period, so advocates typically reported producing less research and technical support for lawmakers, and public efforts related to the legislature typically involv e more efforts to train and mobilize the public. While policy process studies often consider simultaneous or related activity across venues in a given subsystem (Lubell 2013; Jenkins Smith et al. 2018 ), research to date on nonprofit advocacy has not inco rporated venues. Nonprofit studies note that certain organizational characteristics affect strategy decisions, with more recent literature highlighting the complexity of the relationship between organization and advocacy strategies (Mosley 2011; Almog Bar 2018 ). This study supports the complexity of advocacy activities, and also suggests that by viewing activities in their venue contexts, additional patterns may appear. This suggests that in future studies, questions about advocacy activities should be d ifferent depending on whether an organization primarily focuses on courts as compared with those organizations that focus on legislation or rulemaking. Also, t o understand which strategies are more effective, one must first un derstand the context in which they take place. Particularly in the environmental space, where organizations typically focus on advocacy and where a wide range of venues and

PAGE 102

90 issues may be involved, it is useful to understand the institutional structures in which advocacy takes place. regarding whether and how to advocate in certain venues. Internal factors incl ude organizational resources, organizational identity , and IRS status . That is, in line with existing research, organizations with more limited resources are more constrained in their levels and types of including its foundations, expertise, and path dependency, influences its choices (Gais and Walker 1991; Mosley 2011 ) . IRS status, on the which differs from expectations derived from research o n social service providers (Nicholson Crotty 2009; Pekkanen and S mith 2014b ). External factors include political climate, level of conflict, and venue. While other studies have highlighted these as potential fact ors (Gais and Walker 1991 ), the results in this study underscore the relevance of political climate in aff ecting advocacy. Though studies often include measures of political climate (e.g., Nicholson Crotty 2009 ), this factor warrants future consideration to understand better its nuances and how it affects advocacy activities. In making decisions about advoca cy, organizations must consider how decision makers, the public, or others in their subsystem will react, and as level of conflict increases, organizations may resort to additional or different tactics. As such, these results may help scholarship move bey ond traditional impressions that smaller grassroots organizations rely primarily on outside strategies, and larger professionalized organizations rely primarily on inside strategies (McCarthy and Zald 1977; Andrew and Edwards 2004 ) . This study suggests th at organizations on both ends of the

PAGE 103

91 spectrum rely on a range of activities and the factors that influence decisions about advocacy are much more complex. There are a few limitations associated with the results above. As is often the case with smaller N r esearch, this study may suffer from generalizability issues. First, with respect to the three states included in this study, this research includes three diverse states as described above. Of course, many other aspects of diversity exist across the US an d may not be represented in this sample. Also, the sample of environmental nonprofits is intended to be representative of each of the three states , including, to the extent possible, organizations of varying scopes (ranging from national to local), sizes, levels of professionalization, and IRS statuses. However, because not all potential respondents responded to requests for interviews, 10 it could be that the most active or busiest organizations were not included. Furthermore, through interviews, this study is able to explore deeply the engagement of nonprofits in environmental policymaking. As a data collection technique, though, interv iews have limitations including that they rely on self reports Furthermore, a number of themes arose in the interviews that will be worth exploring in future studies. First, the inte rviews demonstrate that policy topics regularly arise across multiple same or similar policy topics arise in multiple venues over the course of time or eve n simultaneously. Future studies may attempt to flesh out whether and how policy evolution or debate across multiple venues simultaneously, influences how organizations advocate . Second, a consistent theme across interviews is that organizations regularl y collaborate with one another on 10 Two organizations did not respond at all to emails r equesting interviews. In another two cases, representatives from organizations in one state did not respond, but I interviewed a representative from the same organization in another state.

PAGE 104

92 as an independent or isolated endeavor. Thus, building on existing studies in this area, future studies should examine th e nature and impact of collaborative advocacy activities and how organizations with different characteristics might collectively impact policy. Finally, future studies should explore representation in the context of environmental policy and nonprofit advo cacy. Particularly in the context of the Greater Chaco R egion in the New Mexico and among community organizations, there is a growing awareness of the need to include the most impacted and vulnerable stakeholders in environmental policymaking. Understand ing the extent to which impacted communities are currently involved in policymaking and the extent to which their voices are meaningfully incorporated in policymaking is an avenue for ongoing research. While this study suggests that there is a relationshi p between representation and decisions related to advocacy, much more research is needed in this area. With a better understanding of the advocacy activities that environmental nonprofits use across venues and the factors that may affect their advocacy, fu ture studies may build on existing knowledge by testing the specific relationships between different types of advocacy and organizational or contextual variables . To begin, b ecause interviewees in this chapter did not report major differences in their act ivities based on their IRS status, the next chapter presents further analysis of the relationship between a few specific activities and IRS status .

PAGE 105

93 CHAPTER IV A QUALITATIVE COMPAR ATIVE ANALYSIS OF EN VIRONMENTAL NONPROFI T ADVOCACY AND IRS DESIGNATION Abstract This chapter presents an analysis of the advocacy activities of environmental nonprofits, with a particular eye toward differences between 501(c)(3) and 501(c)(4) environmental advocacy organizations. This study asks two research questions. Fi rst, what activities do environmental nonprofits engage in t o influence the policy process? Second, which organizational characteristics are associated with specific types of advocacy activities ? Data are drawn from surveys of organizations active in Form 990s and websites. I analyze the data using descriptive statistics and qualitative comparative analysis (QCA), which is well suited for medium N datasets, such as this one (N=30). Result s suggest that environmental nonprofits actively advocate via a variety of activities and report doing so relatively effectively. Furthermore, while the literature typically characterizes 501(c)(4)s as advocacy organizations, this study suggests that, dep ending on the type of advocacy, 501(c)(3)s may advocate as much as or more so than 501(c)(4) organizations. Keywords nonprofit advocacy, environmental policy, qualitative comparative analysis

PAGE 106

94 The prior chapter suggested that 501(c)(3) environmental non profits actively advocate and may even be more common than their 501(c)(4) counterparts. Despite a growing interest in nonprofit advocacy among both scholars and practitioners, existing nonprofit studies primarily focus on social service providers and obs erve differences between organizations with these two IRS designations. Because environmental nonprofits are understudied in the nonprofit literature, though, we do not know whether the relationship between IRS designation and advocacy holds among environ mental nonprofits. Thus , this study presents a n exploration of the advocacy activities of environmental nonprofits using survey data , and the organizational characteristics associated with different types of advocacy, including whether there are differenc es in advocacy between environmental nonprofits registered as 501(c)(3) organizations and those registered as 501(c)(4) organizations . T his paper explores two research questions. First, what activities do environmental nonprofits engage in to influence th e policy process? Second, which organizational characteristics are associated with specific types of advocacy activities ? After a review of the relevant literature and discussion of general theoretical propositions , this paper describes the context for t his study : hydraulic fracturing policy debates. This paper then describes the methods used in this study and the results of both descriptive statistics and qualitative comparative analysis (QCA) . The paper concludes with a discussion of the findings, imp lications for scholars and practitioners, and potential avenues for future research. Literature Review While studies about the advocacy activities of nonprofits are not new, recent years have seen a surge in nonprofit advocacy research. For example, recent scholarly work has focused on the advocacy activities of human service nonprofits and the drivers of those activities (Mosley

PAGE 107

95 2010; Almog Bar and Schmid 2014), how nonprofits influence policy and effect policy change (Hula and Jackson Elmoore 2001; Andrews and Edwards 2004; Bryce 2006; Fyall and McGuire 2015; Fyall 2016), and ways of understanding and cla ssifying nonprofit advocacy organizations and their activities (Boris and Mosher Williams 1998; Child and Grønbjerg 2007). Despite this growing body of literature, current knowledge remains limited, particularly because advocacy is difficult to operationa lize . After an overview of nonprofit advocacy based on the nonprofit and policy process literatures, this section describes the legal and theoretical distinctions between 501(c)(3) and 501(c)(4) nonprofits and presents a series of theoretical propositions . Nonprofit A dvocacy (Pekkanen and Smith 2014a, 3). As this definition suggests, advocacy is a broad concept that may include direct attempts to influence policy, such as l obbying, litigating, or testifying before decision making bodies, and indirect efforts such as education, media campaigns, or voter mobilization. Historically, studies of nonprofit advocacy have focused primarily on rights based and political organization s, but calls for broader conceptualizations of advocacy, such as that by Boris and Mosher Williams (1998), have led to a proliferation of research that recognizes the breadth of forms that advocacy can take (e.g., MacIndoe 2014; Almog Bar 2018). While scho lars agree on the definition of advocacy, the concept remains difficult to operationalize, and the broad definition contributes to challenges with hypothesis testing. Despite the recognition that advocacy is comprised of various activities, quantitative s tudies typically limit advocacy to a dichotomous measure, and thus oversimplify the myriad ways that advocacy occurs (Fyall and McGuire 2015). For example, Pekkanen and Smith (2014b) determine if advocacy is present based on organizational responses to a series of survey

PAGE 108

96 questions, including whether organizations engage in policy or advocacy, employ a lobbyist, or affiliate with a 501(c)(4). Similarly, Child and Grønbjerg (2007) use two sets of survey questions to determine whether an organization advocat es. The first set asks nonprofits whether they promote various positions on policy issues, and the second set asks missions and programs. Ultimately, these measures are limited to whether organizations advocate, rather than indicate the presence or absence of specific types of advocacy. A few studies have attempted to go beyond dichotomous measures, but these measures continue to be relati vely blunt. Nicholson Crotty (2009) uses a two step approach to understanding advocacy. At the first level, she determines whether advocacy is present in 501(c)(3) organizations based on whether they take the 501(h) election in their IRS filings. She th en goes a step further by distinguishing between lobbying and other types of activities. Also, MacIndoe (2014) uses a dichotomous measure of advocacy generally and then adds two count Another limitation of existing nonprofit advocacy literature is that the vast majority of research to date focuses on social service providers. There are certainly good reasons for this . For example, health and human service nonprofits are fa r more prevalent than other types of nonprofits (Berry and Arons 2003), and t hese organizations balance service provision and advocacy (see Minkoff 2002; Fyall 2017) . As Mosley (2014) describes in a study of homeless service providers, these nonprofits pr ovide a critical voice for marginalized communities, but because their primary mission is to provide services, they face significant barriers to advocacy, including resource constraints, lack of experience with advocacy, and confusion about the legal limit s of 501(c)(3) advocacy. While much may be learned from studying social service providers, extending studies of nonprofit advocacy to other contexts is a necessary step. With

PAGE 109

97 abundant and significant global environmental issues, including air pollution, ocean pollution, water contamination, and climate change and its effects, environmental policymaking is among policymaki ng (Bies et al. 2013; Kagan 2019 ). E nvironmental nonprofits are also more likely than any other type of nonprofit to advocate (Child and Grønbjerg 2007), and thus, through examining a population of nonprofits that actively advocates, lessons may be gleaned about advocacy more broadly. In add ition to its value in the nonprofit literature, nonprofit advocacy is also relevant to studies of environmental policy and related policy process literature. Policy process research is surrounding actors, events, policy process theories, such as the Multiple Streams Framework (MSF), the Punctuated Equilibrium Theory (PET), the Advocacy Coal ition Framework (ACF), and the Narrative Policy Framework (NPF), contemplate a role for nonprofits (Fyall 2017; Weible and Sabatier 2018). For example, in the MSF, nonprofits may participate in policy communities or act as entrepreneurs (Mason 2015 ), and in the ACF, nonprofits participation in coalitions to influence change (Fyall and McGuire 2015). While these studies are informative about advocacy, they typically do not examine the nonprofit sector specifically, focusing instead on coalitions or policy entrepreneurs from across sectors (e.g., Mintrom and Norman 2009; Weible, Heikkila, deLeon, and Sabatier 2012). This study empirically examines the differences between 501(c)(3) and 501(c)(4) organizations in their advocacy ac tivities. In exchange for providing public benefits,

PAGE 110

98 governments may confer tax exempt status on nonprofits (Hoyt 2012; Hopkins and Gross 2016). Most nonprofits claim federal tax exemption under section 501(c)(3) of the IRS code (Hoyt 2012). Section 501 (c)(3) organizations, which are public charities or private foundations, have an added benefit in that their donors can also claim tax exemption for their donations (26 USC § 501(c)(3)). In exchange for this benefit, there are some restrictions on 501(c)( 3) advocacy activities. Specifically, public charities are prohibited from having lobbying as a substantial part of their activities and from engaging in political campaigns for candidates for public office (26 USC § 501(c)(3)). These restrictions on 501( c)(3) advocacy carry considerable weight in the eyes of both practitioners and scholars, although the limitations are sometimes misunderstood. With respect to practitioners, studies suggest that 501(c)(3) staff may be unclear about the laws surrounding ad vocacy and the extent to which they can advocate (e.g., Berry and Arons 2003). W hile many nonprofit managers believe they are prohibited from lobbying (Mosley 2014) , lobbying simply any organizations regularly lobby below the legal thresholds. Furthermore, as described above, advocacy is much broader than lobbying or supporting political candidates for office. For example, Casey (2011) outlines 28 different types of advocacy in eigh t categories: legal, legislative and administrative, research and policy analysis, coalition building and capacity development, education and mobilization, communication and media outreach, government relations and oversight, and service delivery. These a ctivities extend well beyond lobbying and supporting candidates, and include a variety of forms of direct and indirect advocacy consistent with the definition of

PAGE 111

99 Pekkanen and Smith (2014a). The advocacy activities included in this study are broadly stated to uncover general patterns in advocacy. 11 The distinction between 501(c)(3) and 501(c)(4) organizations also often carries much weight in studies of nonprofit advocacy. Some studies focus exclusively on 501(c)(3) organizations (Berry and Portney 2014; Le Roux and Wright 2010), while others use 501(c)(4) status as an indicator of the presence of advocacy (e.g., Pekkanen and Smith 2014b; Mason 2015). Nonprofits registered under section 501(c)(4) are social welfare organizations, and unlike 501(c)(3) organiz ations, 501(c)(4)s are not restricted in their advocacy (26 USC § 501(c)(4)). Again, though, because the restrictions are narrow and advocacy is broad, assumptions about differences between these organizations should be tested. Particularly as studies m ove to other types of organizations that may advocate more , such as environmental and civil rights organizations, the distinction between 501(c)(3) and 501(c)(4) organizations may be less determinative, and differentiating between advocacy and non advocacy organizations may present a false dichotomy. Fyall and McGuire (2015) describe that studies often distinguish between organizations with advocacy as their primary purpose and those for whom advocacy is a peripheral activity. While this may be a useful h euristic, it does not adequately reflect the fact that advocacy is a broad concept that can refer to many different activities (see Child and Gronbjerg 2007). To gain a full picture of nonprofit advocacy, one must consider that a va riety of organizations advocate and that they use an array of both direct and indirect advocacy activities to do so. 11 The survey used in this study is part of a broader project, and thus, the advocacy activities are not specific to nonprofits. Because we know, for instance , that 501(c)(3) organizations do not engage in electoral politics, the use of more general advocacy categories in this study allows us to observe less obvious tren ds among nonprofits.

PAGE 112

100 Theoretical P ropositions This section combines expectations about 501(c)(3) and 501(c)(4) organizations with institutional theory and resource dependency theory, which are described more fully below, to develop a series of theoretical propositions about the relationship between organizational characteristics and advocacy . Theoretical propositions are common in qualitative research, as they tend to be more concept ual and have the primary purpose of guiding research designs (Maxwell 2013; Yin 2014). According to institutional theory, in order to survive, organizations must conform to the rules, norms, and social expectations that make up their institutional environ ments (Guo and Acar 2005; see also Meyer and Rowan 1977). As such, the innerworkings of nonprofit organizations must be understood in the context of these environments (Feeney 1997). Institutional theory has been widely applied in nonprofit studies and u sed to explain decisions about advocacy (e.g., Schmid, Bar, and Nirel 2008; Suárez and Hwang 2008). Next, r esource dependency theory suggests that organizations require a variety of resources to survive, and they therefore interact strategically in their environments to secure these resources (Eikenberry and Kluver 2004; see also Pfeffer and Salancik 1978). Resource dependency theory is also regularly invoked in studies about why and how nonprofits advocate (e.g., Nicholson Crotty 2009; Mosley 2010; MacIn doe 2014; Fyall 2016). The first and primary expectation in this study is that 501(c)(4) organizations are more likely to advocate than 501(c)(3) organizations . As described above, this expectation is consistent with the treatment of these organizations i n other studies, and it is based on institutional theory in that 501(c)(3) organizations are legally constrained in certain types of advocacy and may be limited by perceptions or norms suggesting that 501(c)(3) organizations

PAGE 113

101 are not advocacy organizations. This expectation is also supported by resource dependency theory in that 501(c)(3) organizations may have fewer resources devoted specifically to advocacy and thus may be less inclined to advocate. Beyond IRS designation, this study also includes a serie s of additional variables and explores how these variables interact with 50 1(c) status. Because the sample size in this study is small and QCA examines associations between variables , expectations are stated generally. First, this study generally expects that higher levels of capaci ty are associated with advocacy . Three distinct measures of capacity are included. Consistent with other studies, the first measure of capacity is organ izational size measured by gross receipts. Organizations with more resources as measured by their finances have more resources to contribute to advocacy and are therefore not only more likely to advocate, but also more likely to advocate effectively. In addition to this financial measure, two additional measures are included reflecting 1) whether organizations have a specific legal focus and 2) whether organizations have a specific grassroots focus. Organizations with resources devoted specifically to th ese purposes are more likely to prioritize advocacy. Relatedly, this paper also expects that organizations with headquarters in Washington, DC, are mor e likely to advocate . This expectation comports with both resource dependency theory and institutional t heory. Not only are these likely to be larger national organizations, but they are also embedded in active policy environments with the intent to advocate and with the expertise and resources to do so effectively. A measure of years of operation is also i ncluded in this study, but because of conflicting literature on the relationship between years of operation and advocacy (see Nicholson Crotty 2009), this study does not expect to find a relationship between these variables. On one hand,

PAGE 114

102 older organizatio ns may succumb to conservative expectations of others in their institutional environments and are thus less willing to take risks and engage in advocacy activities (Nicholson Crotty 2009). On the other hand, older organizations may be more established as advocates, may assume leadership roles in advocacy, and may be less beholden to external pressures (see Berry and Arons 2003). Research Context To analyze the advocacy activities of environmental nonprofits, this study uses surveys of organizations active in hydraulic fracturing (fracking) policy debates in the US. Fracking is a technique involving the injection of water, sand, and chemical additives into shale rock or other natural gas (Heikkila et al. 2014). In the US and around much of the world, fracking is increasingly used in oil and gas development. For example, between 2007 and 2017, the amount of natural gas produced from shale formations increased 10 fold from 1,9 90,145 million cubic feet to 19,018,457 million cubic feet (EIA 2018). Not only has the volume of oil and gas produced through fracking increased, but oil and gas development is also encroaching on high population density areas. Close proximity between population centers and oil and gas operations gives rise to a host of local issues, including NIMBYism (not in my backyard); nuisance issues, like traffic, dust, and noise; and concerns about air and water contamination. As a result, communities grapple w ith how to manage fracking, and some have attempted to ban it. For example, while the state of New York successfully banned fracking, local bans in a number of municipalities like Denton, Texas, and Longmont, Colorado, have been struck down by courts or l egislatures (Knight and Gullman 2015; Heikkila and Weible 2016; Jaquith 2017). Local communities with limited authority over

PAGE 115

103 oil and gas development may attempt to exercise control by negotiating agreements with oil and gas companies (Zilliox and Smith 20 17), regulating nuisance issues, and advocating for increased setbacks and other limits at the state level. Beyond local issues, fracking is part of broader national and global debates about reliance on fossil fuels as energy sources. While those in favo r of fracking often cite natural gas as a cleaner alternative to oil and coal, those opposed claim that fracking and the availability of cheap oil and natural gas impede transitions to renewable energy sources, like wind and solar. Because they involve a w ide range of issues across venues and levels of government, fracking debates provide a good context for this study. Fracking is an intensely debated political issue (Heikkila and Weible 2016), drawing participation from industry, government, and nonprofit organizations. Fracking policies are created via judicial decisions, legislation, agency rules, executive orders, and ballot measures, and organizations, including environmental nonprofits , engage in a variety of activities and strategies to influence ea ch of the sources of policy. As such, fracking is an ideal policy context for observing maximum potential advocacy activities. Methods Data for this study were collected as part of a larger project via three surveys of policy actors in hydraulic fracturi ng policy debates: a survey in Colorado in 2015, a national survey in 2016, and a second Colorado survey in 2017. 12 The survey targeted individuals involved in or knowledgeable about oil and gas issues in Colorado from diverse sectors and organizations. W hile each survey targeted people across sectors, this study focuses exclusively on nonprofit 12 While fracking is happening across the US, Colorado is ranked eighth in the US for shale gas production, seventh for crude oil production, and has seen a disproportionate share of policy activity related to the oil and gas developmen t (see EIA 2018; Weible and Heikkila 2016).

PAGE 116

104 respondents , resulting in a final sample of 30 distinct environmental nonprofits as detailed below . The samples for each full survey were developed through online sources, news media, and snowball sampling based on coincident interviews with policy actors. For each survey, respondents received an email request with an online link to the survey. As mentioned, because the focus of this research is environmental nonprofits, this study is limited to a subset of respondents. That is, to be included, respondents must represent organizations that are 990 or 990EZ filers , indicating that they are regist ered as tax exempt organizations with the IRS. They also must have at least one major group code of C (environment) in the National Taxonomy of Exempt Entities (NTEE) system, indicating that they are environmentally focused. Furthermore, organizations ar e excluded if they represent the oil and gas industry as reflected in their mission statements. In 2015, 11 unique environmental nonprofits responded to the Colorado survey; 20 unique nonprofits responded to the 2016 national survey; and 7 nonprofits resp onded to the 2017 Colorado survey. 13 As these numbers suggest, in some cases organizations responded to more than one survey (six organizations responded to two surveys, and one responded to all three surveys). When organizations responded to more than on e survey, only their most recent responses are included in the dataset. Also, more than one individual from each of 11 organizations responded to the survey. In these cases, the mean response across individual respondents was used to create the organizat ion level variables. Thus, the final sample includes 30 unique environmental nonprofits across the three surveys. 13 With respect to the full survey, in 2015, a total of 177 people across sectors responded to the survey (47% response rate). In 2016, 102 people responded (28% response rate), and in 2017, 160 people responded (34% response rate).

PAGE 117

105 To analyze the data, this study uses descriptive statistics and qualitative comparative analysis (QCA), which is described more fully below. The first research question, regarding the activities environmental nonprofits engage in to influence the policy process, relies on descriptive statistics to assess mean rates of participation and self reported effectiveness of environmental nonprofits across various advocacy activities. The second research question, about the organizational charact eristics that predict advocacy among environmental nonprofits, uses QCA to identify the causal recipes associated with each advocacy activity. Additional details and results are presented in the sections that follow. Advocacy Activities of Environmental Nonprofits The first research question presented in this study is, what activities do environmental nonprofits enga ge in to influence the policy process? To answer this question, survey respondents were asked the following: Over the past two years, to what extent have you engaged in the following activities and used them effectively in achieving your personal or prof essional goals related to oil and gas development that uses hydraulic fracturing? The question was followed by a series of activities, detailed in the results table below. 14 For each activity, response options were 1 = not engaged, 2 = engaged but not e ffective, 3 = engaged and moderately effective, 4 = engaged and very effective. 15 Results for the sample of 30 nonprofits, along with results from the full set of survey respondents (as a comparison) are shown below in Table 1 3 . 14 While some of the activities listed may also be used in relation to non advocacy activities, as the question suggests, the survey specifically focused on oil and gas policy. 15 The question and response categories reflect both whether organizations engage in each activity and whether they believe they do so effectively. Effectiveness was not defined in the survey. As such, responses require some level of self interpretation, but offer a general indication of how important the activit y is to an organization.

PAGE 118

106 Table 1 3 . Partic ipation in Advocacy A ctivities. Two findings may be gleaned from the results. First, nonprofits participate at high rates in each of the given activities . All respondents report countering arguments made by those they disagree with, and at least 90% of respondents mobilize the public, coordinate with allies, and provide information to government officials and the media. Categories with slightly lower rat es of participation (63% and 77%) are those that require working across sectors or with groups with differing viewpoints. Comparing these results with those for overall survey respondents, we see that in a few cases, environmental nonprofits participate a t higher rates than other types of organizations. This is true with respect to coordinating political activities with allies, providing information to news media, and mobilizing the public. Also, patterns of lower participation in activities that require working across viewpoints are consistent with overall survey trends. Second, results also suggest that environmental nonprofits report engaging in most activities moderately or very effectively. Table 1 3 shows mean levels of effectiveness among

PAGE 119

107 those par ticipating in the activity, with scores ranging from 2 to 4. 16 The activity with the highest rate of effectiveness is coordinating political activities with allies, and again, the only activities with lower levels of reported effectiveness are those requir ing cross sector or cross viewpoint coordination. The mean rate of reported effectiveness for the full list of survey respondents is similar to environmental nonprofit respondents in most categories. Nonprofits, however, do report less effectiveness in w orking across sectors or viewpoints than other respondents on average. Organizational Characteristics Associated with Advocacy QCA Overview and Variables of Interest T he second research question is, which organizational characteristics are associated with specific types of advocacy activities ? As noted above, this research question relies on QCA for analysis. QCA is a middle ground between quantitative and qualitative research (Ragin 2008, 1). oriented research tool that analyzes wheth er combinations of variables within (Carboni 2016, 1791). While QCA can be used to analyze datasets of varying sizes, it is particularly useful with medium N data sets, such as the dataset in this study (N = 30) (Schlager and Heikkila 2009). QCA differs from other analyses in that most research, particularly quantitative research including regression analyses, relies on correlational relationships and seeks to isol ate the independent effects of variables. By contrast, QCA relies on set theoretic logic and Boolean algebra to determine which combinations of variables lead to a given outcome (Ragin 2008; Fiss 2011; Li 2018). Set theoretic logic is often inherent in q ualitative research, 16 Respondents who reported not participating in the activity (response = 1) were not included in the effectiveness analysis.

PAGE 120

108 where research may describe the types of phenomena that co occur or how subsets of a given concept are similar or different (Ragin 2008). QCA allows researchers to evaluate systematically these relationships. All QCA variables range f rom 0 to 1, and QCA can include both crisp set and fuzzy set variables (Ragin 2008; Carboni 2016). Crisp set variables are dichotomous variables, where a value of 1 indicates the presence of a condition and a value of 0 indicates its absence. Fuzzy set v ariables are continuous variables that are calibrated to values ranging from 0 to 1. Thus, 1 indicates full membership in a given category, 0 indicates full non membership, and values in between 0 and 1 reflect the reality that cases may have varying degr ees of membership in a given category (Ragin 2008; Li 2018). For example, for a variable reflecting nonprofit size, a 1 may indicate a large nonprofit, a 0 may indicate a small nonprofit, and values in between represent nonprofits are neither clearly larg e nor clearly small. A crossover point of .5 is assigned to cases that are neither in nor out of the set (Ragin 2008; Li 2018). This research uses the fsQCA 3.0 software for analysis . Outcome Variables The outcome variables of interest in this study are advocacy activities, and this paper zooms in on two specific categories of advocacy activities: providing information to government officials and collaborating with people you disagree with. 17 These two variables were chosen for further analysis for a couple reasons related to observing variation . First, organizations report engaging in them at different rates and levels of effectiveness. Second, these are distinct forms of advocacy in that providing in formation to government officials requires interacting with decision makers and perhaps having expertise on an issue, while collaborating with people you 17 QCA results for all advocacy activities are included in Appendix F.

PAGE 121

109 disagree with may require working either within or outside of government and utilizing negotiation or interpersonal skills. More information about each variable is provided in the paragraphs that follow . P roviding information to government officials is commonly associated with advocacy. This category may include activities such as lobbying and testifyin g before decision making bodies, or otherwise directly attempting to influence government officials. While these types of activities are more traditionally associated with 501(c)(4) organizations, the results above suggest that, across the board, organiza tions report engaging in this activity at high rates and at moderate t o high levels of effectiveness. Next , this study zooms in on collaborating with people you disagree with. Working across sectors and viewpoints is commonly required in various types o f advocacy. For example, policy proposals are often created through task forces, rules often result from negotiations between groups with competing viewpoints, and implementation requires coordination across sectors. Furthermore, stakeholder participatio n is often of particular value in environmental policymaking and governance. Participation can improve outcomes, empower communities, reduce uncertainties, and enhance the knowledge available for assessing and solving environmental problems (Irvin and Sta Pahl To operationalize each of these variables, I calibrated responses to survey questions using fsQCA software. As noted above, survey response categories were as follows: 1 = not engaged, 2 = engaged but not effective, 3 = engaged and moderately effective, 4 = engaged and very effective. Full non membership (0) in each advocacy activity is assigned to organizations not engaged in that activity. Full memb ership (1) in each advocacy activity is assigned to

PAGE 122

110 organization who report being engaged and very effective, and the crossover point (.5) is assigned to those engaged but not effective. Other responses are calibrated based on these guide points. A summa ry of each var iable is shown below in Table 14 . Causal V a riables Six causal variables are included in this study. The first variable, which is the key variable in this research, reflects whether organizations are registered under section 501(c)(3) of the Organizations are assigned a 1 if they are registered under section 501(c)(3) and a 0 otherwise. All non 501(c)(3) organizations in this dataset are 501(c)(4) organiz ations. It is also worth noting that all 501(c)(4) organizations that responded to the survey have an affiliated 501(c)(3) organization. This is a common practice among 501(c)(4) organizations because gifts to these organizations are not tax exempt to the organization or tax deductible to donors. In contrast, gifts to 501(c)(3) organizations are both tax exempt and tax deductible, but these organizations are restricted in their lobbying activities and prohibited from engaging in political campaigns. Thus , to maximize tax benefits, 501(c)(4) organizations typically limit their activities to those functions that 501(c)(3)s cannot perform, while affiliated 501(c)(3)s engage in other activities, such as education or awareness campaigns and service provision. The organizations are legally required to operate separately, and funds are kept separate because donations to a 501(c)(3) organization will remain subject to 501(c)(3) restrictions, even if transferred to the 501(c)(4). In some cases, 501(c)(3) organiza tions form related 501(c)(4)s to engage in political campaigns, but at the time of this research, none of the respondent 501(c)(3) organizations had affiliated 501(c)(4)s.

PAGE 123

111 Five additional variables are included. Years of operation is a ratio variable meas ured in recent Form 990, is also included. The next variable, DC headquarters, is also based on headquarters in Washington, DC, and 0 if it is headquartered elsewhere. Finally, variables reflecting whether organizations have a specif ic legal and/or grassroots focus are included. have a particular legal or grassroots focus, such as through their mission statement or vision, or if they have s taff positions specifically devoted to legal or grassroots efforts. Summary data for each variable, along with QCA calib ration is shown below in Table 14 . Table 14. Variable Descriptive Statistics and QCA C alibration. Variable Type Code Condition Description Mean Full Membership Cross Over Point Full Non Membership Causal C 501(c)(3) organization (crisp) .73 501(c)(3) 501(c)(4) Causal A Organizational age (fuzzy) 31 2010 present 1990 1999 Before 1980 Causal S Organizational size (fuzzy) $36 million Median $4.4 m >$100 million $3 5 million <$200,000 Causal D DC Headquarters (crisp) .37 HQs in DC Other HQs Causal L Legal focus (crisp) .53 Legal focus No legal focus Causal G Grassroots focus (crisp) .57 Grassroots focus No grassroots focus Outcome Providing Information to Government Officials (fuzzy) 2.9 4 2 1 Outcome Collaborating With People You Disagree With (fuzzy) 2.1 4 2 1

PAGE 124

112 QCA Results This section presents the results of QCA. Results show combinations of the six causal variables associated with each of the two outcome variables: providing information to government officials and collaborating with people you disagree with. Providing I nfo r mation to Government Officials First, for reasons described above, this study zooms in on the advocacy activity, providing information to government officials. The first step in QCA analysis is to perform a truth table analysis (Fiss 2011). The truth tab le analyzes all possible combinations of variables and their consistency in producing the outcome of interest. The number of possible cases is 2 ^ k, where k equals the number of independent variables in a study. Thus, this study has 64 (2 ^ 6) possible combinations of variables. T he truth table associated with providing information to government o ff icials is shown below in Table 15 . Of 64 possible combinations of variables, just 16 combinations appear in this dataset. Following Ragi n (2008), I use a . 8 cut off point to determine whether combinations of variables are consistent predictors of the outcome variable. Using this threshold, 12 of 16 combinations consistently predict providing information to government officials.

PAGE 125

113 Table 15 . Providing Information to Government Officials, Truth T able. 501(c)(3) Org Age Org Size DC HQs Legal Focus Grassroots Focus N Raw Consistency PRI Consistency SYM Consistency 1 0 0 0 0 0 3 0.71 0.66 0.72 0 1 1 1 0 1 3 0.87 0.79 1 1 0 0 0 0 1 2 0.61 0.51 0.52 1 1 0 0 0 1 2 0.96 0.95 1 1 1 1 0 1 0 2 0.93 0.90 1 1 0 0 1 1 1 1 1 1 1 1 1 0 0 0 0 1 0.85 0.80 0.80 1 1 0 0 1 0 1 0.96 0.93 1 1 1 0 1 1 0 1 0.59 0.48 0.48 0 0 1 1 1 1 1 1 1 1 0 1 1 0 1 0 1 0.73 0.63 1 0 1 1 0 1 1 1 0.86 0.80 1 0 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 1 0.83 0.77 0.77 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 The intermediate solution that results from QCA is shown below in Table 16 . The intermediate solution falls toward the middle of the continuum between complex and parsimonious solutions, and is the recommended solution for interesting results (Ragin 2008; Carboni 2016). The complex solution does not permit any simplifying assumptions about cases that do not exist in the data, and the parsimonious solution relies on assumptions that may not be consistent with theory or evidence in order to produce the simplest combinations of variables. Thus, the intermediate solution is the preferred solution because it includes some theory or evidence based assumptions to reduce complexity.

PAGE 126

114 Table 16 . Providing Information to Government Officials, Intermediate S olution. Note: The causal recipe shows the combination of variables, the presence or negation of which combine to produce the outcome. Consistency scores reflect the frequency with which these combinations produce the outcome. Raw coverage scores reflect the frequency with which the combination of variables appears in producing the outcome, and unique coverage scores reflect the frequency with which the combination alone (not in conjunction with other variables) produces the out come. The intermediate solution may also be written as follows: Equation 1: Equation 1 is written using QCA convention (see Ragin 2008; Carboni 2016), where capital letters indicate the presence of a condition, and lower case letters reflect the negation of a condition hould be read as they are in standard mathematics. Because IRS designation is the primary variable of interest in this study, the solution above is organized to allow for easy assessment of the relevance of this condition. The large C represents 501(c)(3 ) status, and the small c represents 501(c)(4) status. Thus, the intermediate solution suggests that both 501(c)(3) and 501 (c)(4) organizations provide information to government o fficials and that there are multiple pathways through which these organizati ons do so. Given that 90% of organizations in the sample engage in this advocacy activity, this result is not surprising and is similar to the findings for other advocacy activities with high rates of participation, as shown in Appendix F . Furthermore, t he results suggest that when IRS designation is not part of the causal recipe, the presence of other conditions, including years of operation, size, DC headquarters, and grassroots focus are all associated with the outcome .

PAGE 127

115 Collaboratin g with People You Di sagree with The second outcome variable in this study is the advocacy activity collaborating with people you disagree with , and, again, the survey asked specifically about this activity in relation to oil and gas policy . T he truth table associated with c ollaborating with people y ou d isagree w ith is shown below in Table 17 . Again, 16 combinations of independent variables appear in the dataset, and for this outcome measure, only six combinations meet the .8 threshold for consi stently predicting the outcome . Table 17 . Collaborating with People You Disagree with, Truth T able. 501(c)(3) Org Age Org Size DC HQs Legal Focus Grassroots Focus N Raw Consistency PRI Consistency SYM Consistency 1 0 0 0 0 0 3 0.76 0.63 0.71 0 1 1 1 0 1 3 0.14 0 0 1 0 0 0 0 1 2 0.53 0.15 0.31 1 1 0 0 0 1 2 0.75 0.72 0.75 1 1 1 0 1 0 2 0.80 0.53 0.93 1 0 0 1 1 1 1 0.88 0 1 1 0 0 0 0 1 0.81 0.43 1 1 1 0 0 1 0 1 0.55 0.02 0.02 1 1 0 1 1 0 1 0.5 0.45 0.45 0 0 1 1 1 1 1 0.53 0 0 0 1 1 0 1 0 1 0.5 0 0 1 1 0 1 1 1 0.6 0 0 1 1 1 1 1 1 0.88 0 1 1 1 0 0 0 1 1 1 1 1 1 1 1 0 1 1 0.71 0 1 1 1 1 1 1 1 1 1 1 The intermediate so lution is shown below in Table 18 , and in the equation that follows. Table 18 . Collaborating with People You Disagree with, Intermediate S olution. Note: The causal recipe shows the combination of variables, the presence or negation of which combine to produce the outcome. Consistency scores reflect the frequency with which these combinations produce the outcome. Raw coverage scores reflect the frequency with which the combination of variables appears in produci ng the outcome, and unique coverage scores reflect the frequency with which the combination alone (not in conjunction with other variables) produces the outcome.

PAGE 128

116 Equation 2: As with the prior analysis, Equation 2 is organized by IRS type, but, in the case of collaborating with people you disagree with, there are no combinations of variables that include 501(c)(4) designation, thus indicating that 501(c)(4) organizations are no t associated with engaging in this variable effectively. Other results are similar to those shown for the prior outcome variable. Specifically, there are numerous combinations that include 501(c)(3) status, and when IRS designation is not part of the cau sal recipe , the presence of other variables, including years of operation, size measured by gross receipts, DC headquarters, legal focus, and grassroots focus, is associated with the outcome. Discussion The results described above have implications for bot h scholarship and practice related to nonprofit advocacy. With respect to the first research question what activities do environmental nonprofits engage in to influence the policy process? the results suggest that nonprofits engage in a variety of activit ies at high rates and report doing so at least moderately effectively for most categories. The only categories of activities where nonprofits participate at slightly lower rates and levels of effectiveness are those that require working across sectors and viewpoints. From a theoretical standpoint, the high rates of participation and reported effectiveness confirm the active role that nonprofits play in policymaking and the need to include these organizations in studies of policymaking and policy processes. Existing environmental policy studies do often include nonprofits (e.g., Weible and Heikkila 2016), but existing literature typically does not consider the wide organizational variation that exists within the nonprofit sector. The literature, therefore, may miss that nonprofits are not always aligned or that a variety

PAGE 129

117 of types of organizations may work on a given issue in different ways. Recognizing this variati on can help explain policy subsystem dynamics and policy change. Relatedly, f or practitioner s interested in influencing policy, it is useful to bear in mind the potential influence of nonprofit organizations and for these organizations to consider the broad range of strategies available to them. Because there is some variation in participation across advocacy activities, the results also suggest the value of breaking advocacy into categories to understand better the activities and strategies that organizations use. As noted above, existing nonpro fit studies often operationalize advocacy as a binary variable (e.g., Child and Gronbjerg 2007; Pekkanen and Smith 2014b), but this study suggests that we may be able to learn more about factors affecting advocacy if we consider that advocacy can mean many different things. Future studies may continue to explore more distinct types of activities and how rates of participation and effectiveness vary across organizations. This study relies on fairly broad advocacy activities that were drawn from prior studi es on the strategies actors use in advocacy coalitions when trying to influence policy outcomes. Future studies, however, should continue to refine these measures or develop evidence based categories that apply to the nonprofit sector, such as those found in Chapter 3. The second resear ch question in this study is, which organizational characteristics are associated with specific types of advocacy activities ? Based on institutional theory and prior research, this study generally expected to find that 501( c)(4) organizations report more engagement and effectiveness across types of advocacy (Guo and Acar 2005; Mason 2015 ) . The results, however, do not support this expectation. With respect to providing information to government o fficials, the results sugge st that both 501(c)(3) and 501 (c)(4) organizations engage in this activity, undermining the notion that 501(c)(3) organizations are not advocacy

PAGE 130

118 organizations . Furthermore, with res pect to advocacy that involves collaborating with people you disagree w ith , the results suggest that 501(c)(4) organizations do not report high rates of effective engagement in this activity. This is especially surprising given that environmental policymaking often involves multi sector engagement and requires participation in collaborative arrangements (Durant, Fior . Future studies should continue to explore the similarities and differences between 501(c)(3) and 501(c)(4) advocacy and empirically test the implications of IRS status. Given the broad defin ition of advocacy, it is necessary to recognize that there may be situations in which 501(c)(3) organizations are actually more effective advocates than traditional 501(c)(4) advocates. The QCA analysis also suggests that other variables, like organizational size, age, DC headquarters, legal expertise, and grassroots expertise, do not play a significant role in determining advocacy. This result likely stems from the fact that this study uses broad categories of advocacy activities. Thus, these relationships should continue to be fleshed out in future studies that employ nonprofit specific advocacy activities, such as those described in Chapter 3. From a practical standpoint, the literature suggests that 501(c)(3) organizations are sometimes unc lear about the limits on their advocacy (Berry and Arons 2003; Mosley 2014). These organizations should continue to educate themselves and realize that there may be many opportunities to influence policy beyond lobbying and that they may actually be bette r suited for certain types of advocacy than their 501(c)(4) counterparts. Furthermore, because environmental policymaking often requires collaborating across sectors and viewpoints, 501(c)(4) organizations should examine their existing limitations, consid er whether they may benefit from more collaborative approaches, and possibly partner with 501(c)(3) organizations on advocacy.

PAGE 131

119 opportunities for future research. First, this study uses relatively broad categories of advocacy activities and combines measures of participation with measures of effectiveness. Future studies should improve on these measures by elaborating on the various types of activities that organizations enga ge in, perhaps across different venues, and by measuring participation and effectiveness separately. This study also relies on self reports, which have inherent biases and validity issues. E ffectiveness is difficult to measure, which is why studies often depend on self reports, but future research should certainly aim to improve effectiveness measures . Furthermore, this study focuses on a limited sample of organizations involved in a particular policy context. To improve generalizability, future studies among organizations operating in different policy contexts, including other environmental contexts, like renewable energy or wildlife protection, and other policy contexts that draw participation from nonprofit organizati ons, such as civil rights issues. to work across viewpoints than 501(c)(3) organizations. Possible solutions may be gleaned from the literature or from practice. Fo r example, perhaps 501(c)(4) organizations tend to focus more on other types of advocacy at the expense of collaboration . It is also possible that opponents causi ng others to be less willing to work with them (Sabatier, Hunter, and McLaughlin 1987; Fischer, Ingold, Sciarini, and Varone 2016). To fully confirm and understand this result, however, future studies should explore possible explanations, perhaps through interviews with 501(c)(4) organizations or other organizations operating in the same policy subsystems.

PAGE 132

120 This paper has presented a n analysis of the advocacy activities of environmental nonprofits active in hydraulic fracturing policy debates, with a partic ular eye toward differences between 501(c)(3) and 501(c)(4) organizations. As environmental issues persist and problems beg better policy solutions , the advocacy work of environmental nonprofits becomes increasingly consequential. Thus, s tudies of enviro nmental nonprofits should continue to help inform these organizations about opportunities and constraints associated with advocacy, effective practices, and the potential for strategic partnerships and collaborations.

PAGE 133

121 CHAPTER V CONCLUSION Studies of nonprofit advocacy are on the rise in response to growing recognition of the role that nonprofits play in policymaking and as the need to aggregate voices in response to powerful corporate voices increases. By understanding the role of nonprofits in polic ymaking, nonprofit advocacy research advances both the policy process literature and the nonprofit management literature. From a policy standpoint, studies seek to understand the impact of nonprofit activities on policy, how nonprofits interact in coaliti ons to influence policy, and ultimately whether and under what circumstances nonprofits are successful in achieving their policy goals. From an organizational standpoint, the nonprofit literature is concerned with how managers incorporate advocacy into th implications for policy and for nonprofit organizations, democratic interests provide a backdrop for nonpr ofit advocacy. That is, in a system of representative democracy, nonprofits play an essential role by aggregating citizen interests, providing a voice for underrepresented interests, and counterbalancing government and private sector interests. The ratio nales for studying nonprofit advocacy specifically apply to environmental nonprofits as well. Nonprofit organizations often play key roles in environmental policymaking by working on their own and in coalitions through lobbying, participating in formal po licymaking processes, and through various forms of direct action. This work is essential in environmental policy where stakeholder engagement is emphasized, corporations with deep pockets are involved, and the broad scope and technical nature of policies make individual citizen engagement challenging. Furthermore, more research on environmental nonprofits in

PAGE 134

122 particular is needed because, while environmental nonprofits advocate more than any other type of nonprofit, the majority of existing nonprofit advoc acy studies focus on social service providers. Also, studies on environmental policy and environmental social movements, do not focus on nonprofits as a distinct type of entity. A variety of nonprofit stakeholders, like managers, donors, and board member s, would benefit from nonprofit specific research. The above considerations motivate this dissertation. Overall, the dissertation contributes to existing policy process and nonprofit studies related to environmental nonprofit advocacy. The dissertation lays a foundation for future studies by examining issues such as the types of venues environmental nonprofits participate in, how they advocate across venues, and how organizational and contextual factors influence their advocacy. While existing studies h ave looked at related questions, limitations exist. For example, the nonprofit literature fails to incorporate venues and typically focuses on advocacy among social service providers, and the environmental policy literature does not address nonprofits as unique organizations. To provide a context for questions specific to environmental nonprofits, this dissertation explores advocacy in the context of oil and gas policy, because policy in this area is highly conflictual, spans venues, and draws participatio n from a variety of stakeholders. Chapter 2 uses news media to examine the policy subtopics and venues that a range of advocacy groups participate in within a given policy subsystem. Through interviews with nonprofit managers across three states, Chapter 3 presents an in depth exploration of the advocacy activities that nonprofits use to influence policy across venues. Finally, Chapter 4 looks primarily at the relationship between 501(c) status and advocacy activities. While studies suggest that nonprof it status is an important determinant of advocacy among social service providers, Chapter 4 asks whether IRS status is similarly consequential among environmental nonprofits.

PAGE 135

123 This concluding chapter synthesizes some of the theoretical, practical, and meth odological lessons learned in the preceding chapters. After reviewing the major takeaways of the dissertation, this chapter discusses overarching limitations associated with the dissertation, and then concludes by presenting a future research agenda that builds on the studies that comprise this dissertation. Major Takeaways This section takes a high level look at the preceding chapters and describes some of the overarching conclusions that may be drawn from the dissertation. While most of the conclusions relate to theoretical contributions of the dissertation, this section concludes with notes for practice. Indeed, by combining some of the results from the individual chapters and considering them in the context of existing literature, additional findings may be discerned. First, by merging lessons about venue selection (Chapter 2) with conclusions regarding advocacy activities (Chapter 3), we may be better able to understand both. Combined, Chapters 2 and 3 reveal that, while there is some consistency i n activities across venues, there are also noticeable differences. Chapter 2 suggests that, in the Colorado fracking subsystem, certain sub issues are more strongly associated with particular venues. For example, local governments frequently grapple with bans and moratoria, and federal venues commonly deal with public land issues. The chapter also suggests that some venues in a subsystem are more active than others, or at least attract more public attention. In the case of fracking in Colorado, we certa inly see activity across venues, but among articles that discuss a particular venue, over 50% of activity was centered in local government or state court. Local authority issues drove much of the dominance of these venues, and over time, the news media su ggest that, as more local governments exerted authority, policy debates shifted to the state level and drew in a new set of

PAGE 136

124 actors. Furthermore, Chapter 3 suggests that some advocacy activities are common across venues, while others depend on the venue. For example, activities at legislatures and regulatory bodies are substantially similar with nonprofit organizations participating in inside, outside, and hybrid activities in each venue, ranging from providing technical expertise to decision makers, to gr assroots organizing. In relation to ballot measures, though, organizations tend to engage almost exclusively in outside activities, and in contrast, few outside activities are reported with respect to courts. Taken together, these results underscore the importance of considering venues when attempting to understand policy subsystems, policy issues, and advocacy activities. Particularly with respect to advocacy activities, research to date has not grounded activities in venues, thus creating limitations i n our understanding of advocacy. Furthermore, a couple major limitations of existing venue selection literature are illuminated through the combined results of Chapter 2 and 3. These chapters demonstrate, first, that both venue selection and advocacy act ivities have a reactive component that is underemphasized in the literature. The venue selection literature makes some suggestion that actors may be reacting at times rather than affirmatively selecting venues (see Holyoke, Brown, and Henig 2012; Ley 2016 ). At present, though, this side of venue selection has not been fully explored and may be an important factor in understanding venue selection. As Chapter 2 shows, environmental advocacy groups participated at similar rates to industry advocacy groups a cross most venues, and venues were sometimes grappling with issues before advocacy groups got involved. Not only might the need to play defense suggest why some environmental groups are active in certain venues, but reactivity may also explain why these g roups develop certain knowledge or venue resources, preventing them from taking action in their preferred venues. Whether an

PAGE 137

125 environmental group participates in a given venue for proactive or reactive reasons may also explain some of its advocacy decisions. For example, if an organization is reacting, it may be more likely to use direct action, as compared with situations in which the organization initiates activity in a venue. T subsystem involves a series of proactive and reactive choices and strategies that may vary based on political climate, organizational resources, and venue selection decisions by opponen ts. Relatedly, the current venue selection literature asks whether organizations participate in given venues, but if we consider inside, outside, and hybrid activities, a better question may be how organizations participate, or which activities they use, in each venue. In Chapter 2, we see little variation between rates of environmental group participation and industry group participation in most venues. It may be, though, that we would see variation if we looked at the types of advocacy activities these organizations used in each venue. For example, the only venue showing unequal rates of participation was courts, and based on Chapter 3, few activity options are available to organizations in courts. In other words, when advocacy options are limited, th ere may be differences in rates of participation, but when a wide range of activity options are available, more differences may be observed if we look at how organizations engage in the venue, rather than whether they engage. Thus, to understand patterns of participation, the venue selection literature should incorporate literature regarding advocacy activities. The second major takeaway from this dissertation is that advocacy is much more complex than early literature on the subject suggests. For example, existing scholarship suggests that venue selection is a fairly straightforward decision, resulting from venu e related factors like venue ideology and accessibility, and organizational factors like resources. Chapter 2, however, suggests that actors participate in venues for both proactive and reactive reasons and that they

PAGE 138

126 often participate across multiple venu es, using a variety of strategies. Also, according to earlier studies on nonprofit advocacy, more institutionalized groups tend to rely on inside strategies, while less formal groups rely on outside strategies (McCarthy and Zald 1977; Andrew s and Edwards 2004). In line with recent literature (e.g., Mosley 2011; Almog Bar 2018), this study suggests that all types of nonprofits use a range of inside, outside, and hybrid strategies. While some organizations may emphasize certain strategies over others, resu lts suggest that even the most professional or institutionalized nonprofits mobilize their members and engage in grassroots advocacy. On the other hand, even grassroots groups meet with lawmakers and are sometimes parties to rulemakings or are otherwise g ranted seats at decision making tables. Thus, choice of venue and advocacy strategy is not a binary decision and depends on a complex mix of factors, including both organizational and contextual factors, such as expertise and political climate. Also, par ticularly in the environmental space, Chapter 4 suggests that IRS designation, though typically important in studies regarding social service providers, has unique implications in the environmental space. That is, among environmental nonprofits, both 501( c)(3) and 501(c)(4) organizations are active advocates. The third takeaway from this dissertation is that studying environmental nonprofits as a unique entity and as distinct from other nonprofits is necessary to advance environmental policy and nonprofit scholarship. Across the board, each of the empirical chapters that comprise this dissertation suggests that nonprofit organizations are very actively engaged in environmental policymaking. Chapter 2 suggests that nonprofits are engaged at similar levels as their industry described in Chapter 3 report that their organizations engage in a wide range of activity across venues and, with some variation based on polit ical context, that they are actively involved in

PAGE 139

127 some of the most salient oil and gas issues in their states. Survey responses from environmental nonprofits, as reported in Chapter 4, also reflect that these organizations use a diverse set of advocacy act ivities and believe that they use many activities effectively. Thus, developing a better understanding of environmental nonprofit advocacy as distinct from other types of organizations, may contribute to the environmental policy and nonprofit literatures. In addition, observations of environmental nonprofits are not consistent with the patterns seen in existing nonprofit advocacy literature, which typically focuses on social service providers. Most notably, existing studies distinguish between 501(c)(3) service providers and 501(c)(4) advocacy organizations. As Chapter 4 most directly suggests, though, similar patterns may not present among environmental nonprofits. Section 501(c)(3) organizations are prohibited from engaging in electoral politics, but across most categories of advocacy, these organizations report similar beliefs about the importance and effectiveness of different types of advocacy. With respect to advocacy that requires collaborating with those they disagree with, 501(c)(3)s actually r eport engaging more than 501(c)(4) organizations, which is notable because existing literature suggests that 501(c)(4)s advocate more than 501(c)(3) and does not consider possible differences across specific activities. Also, unlike with social service 50 1(c)(3)s, which typically engage in advocacy as a peripheral activity (e.g., Mosley 2014), environmental 501(c)(3)s devote significant resources to advocacy and report engaging in as many activities and venues as their 501(c)(4) counterparts. Finally, a f ew practical lessons may be gleaned from this dissertation. For example, nonprofit managers should consider how best to develop their resources. This dissertation suggests that organizations need a range of activities in their toolkits, should be able to work across venues, and must be prepared to play both offense and defense. Thus, organizations must

PAGE 140

128 either develop these areas of expertise themselves, or they should look for strategic partners who complement their strengths. Nonprofits managers must a lso be aware of the various political, institutional, and organizational constraints and opportunities that apply to them. For example, this dissertation suggests that state and local political environments strongly influence ities and that nonprofit managers should understand the implications of their IRS status. While 501(c)(3)s face legal restrictions on advocacy that other organizations do not, this dissertation suggests that these organizations actively advocate and play a role in policymaking. Furthermore, public administrators, elected officials, and other decision makers working on environmental issues should consider working with environmental nonprofits and be aware of the resources they may provide. Not only can en vironmental nonprofits provide expertise, research, and technical support on environmental issues, but they can also help educate the public or build political support for issues. Thus, understanding the opportunities these organizations provide and consi dering the role they play in society may help actors across sectors effectively leverage the strengths of environmental nonprofits. Study Limitations Before discussing potential avenues for future research, this section describes some of the limitations a ssociated with this dissertation research and some methodological considerations for future studies. Each of the empirical chapters above addresses limitations with each specific study, and in this final chapter, general limitations across three major cat egories are addressed in greater detail: 1) biases in data collection methods, 2) sample biases with respect to nonprofits and advocacy groups, and 3) other generalizability issues. First, with respect to data collection methods, each of the chapters abov e uses a unique data source, but independently. Thus, each chapter fails to triangulate methods and data sources,

PAGE 141

129 leading to potential reliability and validity issues as described more fully below. Chapter 2 relies on news media; Chapter 3 relies on inte rviews; and Chapter 4 relies on survey data and publicly available nonprofit data. Each of these data collection methods has strengths. The newspaper articles allow for large N longitudinal analysis. Interviews provide depth, and survey data facilitate comparisons across organizations. Each of these data sources also has limitations, though. As mentioned in Chapter 2, the news media have biases leading to validity issues in the sense that media typically report on major events, and with respect to advo cacy activities, activities that occur outside the public sphere are likely to go unnoticed. Furthermore, because news media have an objective standard of reporting, stories often reflect perspectives from both sides of issues, even when one side dominate s. Next, with respect to interviews, sample sizes tend to be smaller; interviews depend on human memory; and it is sometimes difficult to compare data across interviews. Because interviewees sometimes interpret questions differently from other interviewe es and may spend unequal amounts of time answering each question, interviews lack the objectivity that facilitates easy comparisons. The smaller sample sizes and variation among interviewees may lead to reliability, and possibly validity, issues. Finally , while survey data also rely on self reports, they facilitate comparisons, but survey data do little to answer why questions, and researchers are limited by the questions they ask, which in some cases can lead to validity issues if researchers do not ask the right questions. Thus, while there are limitations associated with each of the above methods of data collection, by combining data collection methods in future studies, some of the limitations may methods and sources of information in a increases reliability and validity and maximizes learning (Van de Ven 2007). For example, as

PAGE 142

130 described above, while the ne ws media shows general patterns of activity in venues, by combining this information with interviews, we may be better able to understand those patterns and gain a deeper understanding of how and why activity in venues varies. To overcome some of the limi tations that inhere in each of the methods independently, future studies should combine multiple complementary approaches. Second, the samples used in this dissertation may be less representative of smaller, informal nonprofits. As mentioned in the empir ical chapters, one of the challenges associated with nonprofit research is that there is wide variation between organizations and few (if any) studies adequately reflect this variation. As with Chapter 4, this challenge primarily results because nonprofit studies often only focus on organizations that file 990 tax forms for data availability reasons, which leads to limitations. For example, knowledge about nonprofits is largely limited to financial matters, and many organizations are omitted from database s, including small organizations that are not required to file 990 returns, religious organizations, and organizations that are not registered for federal tax exemption (Never 2016). Interviews may help to address some of these issues, but as mentioned ab ove, qualitative studies are generally limited to smaller sample sizes. Without a viable solution to the problem, like other nonprofit studies, this dissertation is less representative of smaller, less formal or professionalized nonprofit organizations, a nd thus, may not reflect the full diversity of organizations that participate in environmental policymaking. Chapter 2 casts a wider net by examining all advocacy groups mentioned in the relevant news articles, including formal nonprofits, informal groups , and action groups. The second chapter also includes both environmental advocacy groups and industry advocacy groups. Thus, the news media may provide a useful resource for identifying advocacy groups active in a

PAGE 143

131 particular topic and allow for an explor ation of a broader range of advocacy groups than are seen in typical nonprofit studies. However, some of the same issues remain in that even once organizations are identified, finding complementary information about many of the less formal organizations i s difficult, and while some information may be gleaned from the news media, this is primarily information that exists in the public eye and may not shed light on organizational characteristics or activities. Finally, as noted throughout the dissertation, the extent to which the studies that comprise the dissertation are generalizable to other environmental policy contexts, nonprofits, and geographical areas is not clear. This dissertation has focused specifically on oil and gas and fracking issues. While fracking is highly conflictual, draws broad participation from a wide range of groups, relates to other energy issues, and affects communities across the US, it is possible that other environmental issues would present somewhat different patterns of venue s and activities. For example, organizations that focus on wildlife protection are probably most active at the federal level and may rely more on internal strategies except in particularly difficult situations. Furthermore, the fact that fracking is char acterized by high conflict and requires technical expertise likely drives the sorts of participation observed in this dissertation. This dissertation has also focused on specific geographic areas: Colorado in Chapter 2; Colorado, New Mexico, and Oklahoma in Chapter 3; and Colorado and the national level in Chapter 4. As described below, to develop our understanding of the contextual factors that influence advocacy, additional multi state or comparative studies are needed. Future Research Agenda The final section of this chapter describes a future research agenda that builds on this dissertation. As noted above, the purpose of this dissertation has been to lay the foundation for

PAGE 144

132 an agenda that focuses specifically on environmental nonprofits and the role these organizations play in environmental policymaking. Studies across three broad areas are described: 1) studies designed to overcome generalizability issues; 2) studies on environmental nonprofit coordination and coalitions; and 3) studies on the representativeness of environmental nonprofits. First, as noted above, among the major limitations of the studies in this dissertation is that their generalizability is questionable. In particular, Chapter 3 draws broad conclusions about the advocacy activities that organizations use across venues and the major factors that influence decisions regarding advocacy activities. To improve generalizability, Chapter 3 focuses on three diverse contexts: Colorado, New Mexico, and Oklahoma. Future studies ma y continue to test mple, the state of Hawaii is a D emocratic stronghold with 70 of 76 le gislators affiliating with the D emocratic P arty. In addition, in Hawaii policymakers and the general public are extremely friendly to environmental interests, and the environmental community maintains close ties with policymakers (see Kagan 2019). Given this context, future studies might examine what advocacy looks like in this consensus based context. Relatedly, to make understanding advocacy in each context more manageable, this study controlled for state population and selected states with moderate populations. Future studies should also consider how other aspects of variation might impact advocacy, for example by examining advocacy in states like New York, California, and Texas. Furthermore, this study focuses specifically on oil and gas policy, which is also closely related to other types of energy policy, such as electric utility regulation and r enewable energy policy. To fully understand environmental nonprofits and policy, future studies should expand beyond energy policy and examine the activities of environmental nonprofits as they relate to issues like water and land conservation and other i ssues of prime concern to the environment.

PAGE 145

133 Also, to improve generalizability, though there are some potential issues with surveys, such as their reliance on self reports, their cross sectional nature, and the fact that response categories are limited, fut ure studies may incorporate survey data to gain a broader picture of advocacy activities. That is, data collection via a multi state survey of environmental nonprofits would help overcome generalizability issues, and surveys may be developed using the les sons from interviews in this study. Such a survey could provide insights on environmental nonprofit advocacy, and incorporate additional data from organizational websites or tax returns to understand better how advocacy activities vary by organization. S econd, among the major themes that emerged in interviews for this dissertation is that described that they regularly work in coalitions, and they often mentioned the complementary work they do in conjunction with other organizations. According to one interviewee: Collaboration is a really deeply rooted value here and how we go about doing our work. So we don't tr y to like roll into town like, h ey, we're here we'r e going to run things now. So supporting our partners, but through all kinds of different tactics to get public comments submitted, have numerous events in the community for educational purposes, newsletters, doing advocacy trips to Denver to meet with the governor, DNR, or back to DC to meet with the congressional delegation and BLM or interior directly. In fact, in some cases, organizations described engaging in activities, such as grassroots mobilization, through their partners. Thus, future studies sh ould combine the nonprofit literature with policy process theories, like the Advocacy Coalition Framework, to further our affect the activities of other organiza tions. Finally, a background theme across the dissertation and across the nonprofit advocacy literature relates to the role that nonprofits play in a democratic society. As suggested at the

PAGE 146

134 outset of the dissertation, nonprofits have the potential to con tribute to democracy by providing a voice for otherwise underrepresented interests and by promoting direct citizen engagement, among other things (LeRoux 2009; Almog Bar and Schmid 2014). At the same time, though, nonprofits can undermine democracy by amp lifying the voice of the already wealthy and powerful (Reid 2006). fair treatment and meaningful involvement of all people regardless of race, color, national origin, or income, on the development, implementation, and enforcement of environmental law, regulations, and policies environmental benefits and burdens and also with procedural or participatory justi ce (Guy and Kagan 2019). Procedural justice contemplates that those impacted by policies have meaningful opportunities to participate in their formation, and in this sense, nonprofits can play an essential role by providing opportunities for those most af fected to participate and by opposing unjust government policies and private sector practices. Considering this imperative, along with the need to attract more political support, there is a growing awareness among environmental nonprofits of the need to d iversify their staff and boards and to become more representative of their constituent communities. Research also suggests a link between constituent participation and representation and advocacy activities (Guo and Saxton 2010). To these ends, a couple lines of research follow. First, as a starting point, descriptive studies to understand environmental nonprofit representation are needed. The constituent bases of environmental nonprofits are less clear than they are with homeless service or health care providers, for example. As such, gaining a basic understanding of whom environmental nonprofits represent, how they connect with constituent interests, and how they

PAGE 147

135 incorporate constituent preferences into their decisions about advocacy (see Fung 2006), may be a useful first step toward dissecting representation among environmental nonprofits. Second, there is a growing trend among environmental nonprofits to put effort toward increasing the diversity and representativeness of their board, but currently boards remain dominated by older white men (BoardSource 2017). It is also not clear whether and under what circumstances these efforts lead to positive outcomes. Some support for a relationship between diversity and organizational performance exists (e.g ., Siciliano 1996; Harris 2014; BoardSource 2017), along with more limited support for a relationship between board representativeness and organizational performance (LeRoux 2009), but more research in this area is needed. Also, because mainstream environ mental nonprofits have historically contributed to environmental injustices by exclusively supporting wealthy white interests, research on the impacts of board diversity and representation may be particularly useful in the environmental policy and EJ space . Thus, this dissertation contributes to a growing body of literature on nonprofit advocacy and hones in specifically on environmental nonprofits in furtherance of both environmental policy and nonprofit management literatures. The studies that comprise this dissertation introduce new questions about venues, dissect advocacy across venues, and demonstrate significant differences between environmental nonprofits and other types of nonprofits. As such, this dissertation provides the foundation and a path f orward for a research agenda focused on the intersection between nonprofit organizations and environmental policy. Such a research agenda will only grow in relevance as threats of climate change increase, environmental injustices continue to plague commun ities, and government and private sector actors continue to promote the current energy and development paradigm in the US and in other countries around the world.

PAGE 148

136 REFERENCES Almog Bar, Michal, and Hillel Schmid. 2014. "Advocacy Activities of Nonprofit Human Service Organizations: A Critical Review." Nonprofit and Voluntary Sector Quarterly 43 (1): 11 35. Almog Bar, Michal. 2018. "Insider Status and Outsider Tactics: Advocacy Tactics of Human Service Nonprofits in the Age of New Public Governance." Nonprofit Policy Forum 8 (4): 411 428. Andrews, Kenneth T., and Bob Edwards. 2004. "Advocacy Organizations in the US Political Process." Annual Review of Sociology 30: 479 506. Bas s, Gary D, David F. Arons, Kay Guinane, and Matthew F. Carter. 2007. Seen But Not Heard: Strengthening Nonprofit Advocacy . Washington, DC: The Aspen Institute. Baumgartner, Frank R., and Bryan D. Jones. 2009. Agendas and Instability in American Politics , 2nd ed. Chicago, IL: The University of Chicago Press. Baumgartner, Frank R., Bryan D. Jones, and Peter B. Mortensen. 2018. "Punctuated Equilibrium Theory: Explaining Stability and Change in Public Policymaking." In Theories of the Policy Process , 4th ed., edited by Christopher M. Weible and Paul A. Sabatier, 55 101. New York: Westview Press. Berry, Jeffrey M., and David F. Arons. 2003. A Voice for Nonprofits . Washington, DC: Brookings Institution Press. Nonprofits and Advocacy , edited by Robert J. Pekkanen, Steven Rathgeb Smith, and Yutaka Tsujinaka, 21 46. Baltimore, MD: Johns Hopkins University Press. Bies, Angela L., Deanna G. Lee, Charles Lindsey, James W. Stoutenbor ough, and Arnold Vedlitz. 2013. "Citizens, Nonprofits and Climate Change Policy." Nonprofit Policy Forum 4 (1): 5 28. Bingham, Nabatchi, and O'Leary 2005. "The New Governance: Practices and Processes for Stakeholder and Citizen Participation in the Work o f Government." Public Administration Review 65 (5): 547 558. BoardSource. 2017. Leading with Intent: 2017 National Index of Nonprofit Board Practices . Washington, DC: BoardSource. https://leadingwithintent.org N onprofits and Government , 2nd ed., edited by Elizabeth T. B oris and C. Eugene Steuerle, 1 35. Washington, DC: The Urban Institute Press.

PAGE 149

1 37 Boris, Eliza beth, and Rachel Mosher Williams. 1998. "Nonprofit Advocacy Organizations: Assessing the Definitions, Classifications, and Data." Nonprofit and Voluntary Sector Quarterly 27 (4): 488 506. Box, Richard C. 2009. Public Administration and Society: Critical I ssues i n American Governance , 2nd ed. Armonk, NY: M.E. Sharpe. Brouwer, Stijn, and Dave Huitema. 2018. "Policy Entrepreneurs and Strategies for Change." Regional Environmental Change 18 (5): 1259 1272. Bryce, Herrington J. 2006. "Nonprofits as Social Cap ital and Agents in the Public Policy Process: Toward a New Paradigm." Nonprofit and Voluntary Sector Quarterly 35 (2): 311 318. Specialization? Venue Targeting Among Nonpr Policy Studies Journal 43 (3): 188 206. Buffardi, Anne L., Robert J. Pekkanen, and Steven Rathgeb Smith. 2017. "Proactive or Protective? Dimensions of and Advocacy Activities Associated with Reported Policy Change by Nonprofit Organizations." Voluntas 28 (3): 1226 1248. Bullard, Robert D., Paul Mohai, Robin Saha, and Beverly Wright. 2008. "Toxic Wastes and Race at Twenty: Why Race Still Matters After All of These Years." Environmental Law 38 (2): 371 411. Campbell, Heather E., Yushim Kim, and Adam Eckerd. 2014. "Local Zoning and Environmental Justice: An Agent Based Model Analysis." Urban Affairs Review 50 (4): 521 552. Characteristi Voluntas 27: 1781 1808. Casey, John. 2011. Understanding Advocacy: A Primer on the Policy Making Role of Nonprofit Organizations . https://www.baruch.cuny.edu/mspia/centers and institutes/center for nonp rofit strategy and management/documents/Casey_UnderstandingAdvocacyaPrimeronthePolicyMakingRole ofNonoprofitOrganizations.pdf Child, Curtis D., and Kristen A. Grønbjerg. 2007. "Nonprofit Advocacy Organizations: Their Characteristics and Activities." Social Science Quarterly 88 (1): 259 281. Colorado Oil and Gas Association (COGA). 2019. The Colorado Oil and Gas Association . www.coga.org

PAGE 150

138 Cole, Luke W., and Sheila R. Foster. 2001. From the Ground up: Environmental Racism and the Rise of the Environmental Justice Movement . New York: New York University Press. Corbin, Juliet, and Anselm Strauss. 2015. Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory , 4th ed. Thousand Oaks, CA: SAGE Publications, Inc. Cox, Michael, Gw en Arnold, and Sergio Villamayor Tomás. 2010. "A Review of Design Principles for Community based Natural Resource Management." Ecology and Society 15 (4): 38 56. Crow, Deserai A., Elizabeth A. Albright, and Elizabeth A. Koebele. 2019. "Stakeholder Partici pation and Strategy in Rulemaking: A Comparative Analysis." State Politics and Policy Quarterly 1 28. doi: 10.1177/1532440018821655 . Crow, Deserai Anderson, and John Berggren. 2014. "Using the Narrative Policy Framework to Understand Stakeholder Strategy a nd Effectiveness: A Multi Case Analysis." In The Science of Stories , edited by Michael D. Jones, Elizabeth A Shanahan, and Mark K. McBeth, 131 156. New York: Palgrave Macmillan. Dahl, Robert A. 2015. On Democracy , 2nd ed. New Haven, CT: Yale University Pr ess. de Tocqueville, Alexis. 1966. Democracy in America . Edited by J.P. Mayer and Max Lerner. New York: Harper & Row, Publishers. deLeon, Linda. 2007. "Public Management, Democracy, and Politics." In The Oxford Handbook of Public Management , edited by Ew an Ferlie, Laurence E. Lynn, Jr., and Christopher Pollitt, 103 130. Oxford: Oxford University Press. Dodge, Jennifer, and Jeongyoon Lee. 2017. "Framing Dynamics and Political Gridlock: The Curious Case of Hydraulic Fracturing in New York." Journal of Envi ronmental Policy and Planning 19 (1): 14 34. Dodge, Jennifer, and Sonia M. Ospina. 2016. "Nonprofits as 'Schools of Democracy': A Comparative Case Study of Two Environmental Organizations." Nonprofit and Voluntary Sector Quarterly 45 (3): 478 499. Dodge, Jennifer. 2017. "Crowded Advocacy: Framing Dynamic in the Fracking Controversy in New York." Voluntas 28: 888 915. Durant, Robert F., Daniel J. Fiorino, and Rosemary O'Leary, eds. 2017. Environmental Governance Reconsidered: Challenges, Choices, and Oppo rtunities , 2nd ed. Cambridge, MA: The MIT Press. Eikenberry, Angela M., and Jodie Drapal Kluver. 2004. "The Marketization of the Nonprofit Sector: Civil Society at Risk?" Public Administration Review 64 (2): 132 140.

PAGE 151

139 Environmental Protection Agency (EPA). 2011. Plan EJ 2014 . https://nepis.epa.gov/Exe/ZyPDF.cgi/P100DFCQ.PDF?Dockey=P100DFCQ.PDF Environmental Protection Agency (EPA). 2014. EPA Policy on Environmental Justice for Working with Federally Recognized Tribes and Indigenous Peoples . https://www.epa.gov/sites/production/files/2017 10/documents/ej indigenous policy.pdf Environmental Protection Agency (EPA). 2017. Environmental Justice . https://www.epa.gov/environmentaljustice Feeney, Suzanne. 1997. "Shifting th e Prism: Case Explications of Institutional Analysis in Nonprofit Organizations." Nonprofit and Voluntary Sector Quarterly 26 (4): 489 508. Guys: Actor and Proce ss Journal of Public Policy 36 (2): 309 334. The Academy of Management Journal 5 4 (2): 393 420. Follett, Mary Parker. (1918) 1998. The New State . University Park, PA: The Pennsylvania State University Press. Fung, Archon. 2003. "Associations and Democracy: Between Theories, Hopes, and Realities." Annual Review of Sociology 20: 515 5 39. Fung, Archon. 2006. "Varieties of Participation in Complex Governance." Public Administration Review special issue: 66 75. Fyall, Rachel, and Michael McGuire. 2015. "Advocating for Policy Change in Nonprofit Coalitions." Nonprofit and Voluntary Secto r Quarterly 44 (6): 1274 1291. Fyall, Rachel. 2016. "The Power of Nonprofits: Mechanisms for Nonprofit Policy Influence." Public Administration Review 76 (6): 938 948. Fyall, Rachel. 2017. "Nonprofits as Advocates and Providers: A Conceptual Framework." Policy Studies Journal 45 (1): 121 143. Gais, Thomas L., and Jack L. Walker, Jr. 1991. "Pathways to Influence in American Politics." In Mobilizing Interest Groups in America , edited by Jack L. Walker Jr., 103 121. Ann Arbor, MI: The University of Michigan Press. Gen, Sheldon, and Amy Conley Wright. 2018. "Strategies of Policy Advocacy Organizations and Their Theoretical Affinities: Evidence from Q Methodology." Policy Studies Journal 46 (2): 298 326.

PAGE 152

140 Gormley, Jr., William T., and Helen Cymrot. 2006. "The Strategic Choices of Child Advocacy Groups." Nonprofit and Voluntary Sector Quarterly 35 (1): 102 122. Guo, Chao, and Gregory D. Saxton. 2010. "Voice In, Voice Out: Constituent Participation and Nonpr ofit Advocacy." Nonprofit Policy Forum 1 (1): article 5. Guo, Chao, and Muhittin Acar. 2005. "Understanding Collaboration Among Nonprofit Organizations: Combining Resource Dependency, Institutional, and Network Perspectives." Nonprofit and Voluntary Secto r Quarterly 34 (3): 340 361. Guy, Mary E., and Jennifer Kagan. 2019. "A Sisyphean Task: Applying Follettian Thought to Environmental Justice Issues." In The Future of Progressivism: Applying Follettian Thin king to Contemporary Issues , edited by Margaret S tout. Claremont, CA: Process Century Press. Harris, Erica E. 2014. "The Impact of Board Diversity and Expertise on Nonprofit Performance." Nonprofit Leadership and Management 25 (2): 113 130. Policy Studies Journal 41 (3): 484 512. Heikkila, Tanya, and Andrea K. Gerlak. 2005. "The Formation of Large scale Colla borative Resource Management Institutions: Clarifying the Roles of Stakeholders, Science, and Institutions." Policy Studies Journal 33 (4): 583 612. Heikkila, Tanya, and Christopher M. Weible. 2016. "Contours of Coalition Politics on Hydraulic Fracturing Within the United States of America." In Policy Debates on Hydraulic Fracturing , edited by Christopher M. Weible, Tanya Heikkila, Karin Ingold, and Manuel Fischer, 29 52. New York: Palgrave Macmillan. acking the Intensity of Policy Conflict: Policy Sciences 50: 179 193. Heikkila, Tanya, Jonathan J. Pierce, Samuel Gallaher, Jennifer Kagan, Deserai A. Crow, and Christopher M. Weible. 2014. "Understanding a Pe riod of Policy Change: The Case of Hydraulic Fracturing Disclosure Policy in Colorado." Review of Policy Research 31 (2): 65 87. Herweg, Nicole, Nikolaos Zahariadis, and Reimut Zohlnhöfer. (2018). "The Multiple Streams Framework: Foundations, Refinements and Empirical Applications." In Theories of the Policy Process , 4th ed., edited by Christopher M. Weible and Paul A. Sabatier, 17 53. New York: Westview Press. Arena: State and Local Government Review 44 (1): 9 20.

PAGE 153

141 The Jossey Bass Handbook of Nonp rofit Leadership and Management , 4th ed., edited by David O. Renz and Associates, 43 79. Hoboken, NJ: John Wiley and Sons. The Nature of the Nonprofit Sector , 2nd ed., edited by J. Steven Ott and Lisa A. Dicke, 121 124. Boulder, CO: Westview Press. Hula, Richard C., and Cynthia Jackson Elmoore. 2001. "Governing Nonprofits and Local Political Processes." Urban Affairs Review 36 (3): 324 358. Irvin and Stansbury 2004. "Citizen Participation in Decision Making: Is It Worth the Effort?" Public Administration Review 64 (1): 55 65. Environmental Claims Journal 29 (1): 49 65. Jenkins Smith, Hank C., Daniel Nohrstedt, Christopher M. Weible, and Karin Ingold. 2018. "The Advocacy Coalition Framework: An Overview of the Research Program." In Theories of the Policy Process , 4th ed., edited by Christopher M. Weible and Paul A. Sabatier, 135 171. New York: Westview Press. Public Policy and Mass Media: The Interplay of Mass Communication and Political Decision Making , edited b y Sigrid Koch Baumgartner and Katrin Voltmer, 17 43. New York: Routledge. Review of Policy Research 36 (2): 217 241 . Kingdon, John. 20 03. Agendas, Alternatives, and Public Policies , 2nd ed. New York, NY: Longman. Fracking Ordinances in Home Tulane Environmental Law Journal 28 (2): 2 97 314. Lemos, Maria Carmen, and Arun Agrawal. 2006. "Environmental Governance." Annual Review of Environmental Resources 31: 297 325. LeRoux, Kelly, and Nathaniel S. Wright. 2010. "Does Performance Measurement Improve Strategic Decision Making? Findings from a National Survey of Nonprofit Social Service Agencies." Nonprofit and Voluntary Sector Quarterly 39 (4): 571 587.

PAGE 154

142 LeRoux, Kelly. 2007. "Nonprofits as Civic Intermediaries: The Role of Community Based Organizations in Promoting Political Participation." Urban Affairs Review 42 (3): 410 422. LeRoux, Kelly. 2009. "The Effects of Descriptive Representation on Nonprofits' Civic Intermed iary Roles: A Test of the 'Racial Mismatch' Hypothesis in the Social Services Sector." Nonprofit and Voluntary Sector Quarterly 38 (5): 741 760. Ley , Aaron J., and Edward P. Weber . 2015 . " The Adaptive Venue Shopping Framework: How Emergent Groups Choose Environmental Policymaking V enues . " Environmental Politics 24 (5): 703 722. Ley, Aaron J. 2016. " Vested Interest, Venue Shopping, and Policy Stability: The Long Road to Improving " Review of Policy Research 33(5): 506 525. Fuzzy Nonprofit Management and Leadership 2018: 1 21. Lubell, Mark. 2013. "Governing Institutional Complexity: The Ec ology of Games Framework." Policy Studies Journal 41 (3): 537 559. MacIndoe, Heather. 2014. "How Competition and Specialization Shape Nonprofit Engagement in Policy Advocacy." Nonprofit Policy Forum 5 (2): 307 333. Mason, Dyana P. 2015. "Advocacy in Nonp rofit Organizations: A Leadership Perspective." Nonprofit Policy Forum 6 (3): 297 324. Maxwell, Joseph A. 2013. Qualitative Research Design , 3rd ed. Thousand Oaks, CA: SAGE Publications. Mc C arthy , John D., and Mayer N. Zald . 1977 . "Resource Mobilization and Social Movements: A Partial Theory. " American Journal of Sociology 82 (6): 1212 1241. Merry, Melissa K. 2016. "Constructing Policy Narratives in 140 Characters or Less: The Case of Gun Policy Organizations." Policy Studies Journal 44 (4): 373 395. Me yer, John W., and Brian Rowan. 1977. "Institutionalized Organizations: Formal Structure as Myth and Ceremony." American Journal of Sociology 83 (2): 340 363. Minkoff, Debra C. 2002. "The Emergence of Hybrid Organizational Forms: Combining Identity Based S ervice Provision and Political Action." Nonprofit and Voluntary Sector Quarterly 31 (3): 377 401.

PAGE 155

143 Mintrom, Michael, and Phillipa Norman. 2009. " Policy Entrepreneurship and Policy Change. " Policy Studies Journal 37 (4): 649 667. Mosley, Jennifer E. 2010. "Organizational Resources and Environmental Incentives: Understanding the Policy Advocacy Involvement of Human Service Nonprofits." Social Service Review 84 (1): 57 76. Mosley, Jennifer E. 2011. " Institutionalization, Privatization, and Political Opportun ity: What Tactical Choices Reveal About the Policy Advoca cy of Human Service Nonprofits. " Nonprofit and Voluntary Sector Quarterly 40 (3): 435 457. Mosley, Jennifer E. 2014. "From Skid Row to the Statehouse: How Nonprofit Homeless Service Providers Overcome Barriers to Policy Advocacy Involvement." In Nonprofits and Advocacy , edited by Robert J. Pekkanen, Steven Rathgeb Smith, and Yutaka Tsujinaka, 10 7 134. Baltimore, MD: Johns Hopkins University Press. Mosley, Jennifer E. 2016. "Nonprofit Organizations' Involvement in Participatory Processes: The Need for Democratic Accountability." Nonprofit Policy Forum 7 (1): 77 83. NAACP. 2017. Fumes Across the Fence Line: The Health Impacts of Air Pollution from Oil & Gas Facilities on African American Communities . https://www.naacp.org/climate justice resources/fumes across fence line/ Never , Brent . 2016. "The Changing Context of Nonprofit Management in the Un ited States." In The Jossey Bass Handbook of Nonprofit Leadership and Management , 4th ed., edited by David O. Renz and Associates, 80 101. Hoboken, NJ: John Wiley and Sons. Nicholson Crotty, Jill. 2009. "The Stages and Strategies of Advocacy Among Nonprof it Reproductive Health Providers." Nonprofit and Voluntary Sector Quarterly 38 (6): 1044 1053. Nishishiba, Masami, Matthew Jones, and Mariah Kraner. 2014. Research Methods and Statistics for Public and Nonprofit Administrators . Thousand Oa ks, CA: SAGE Pub lications, Inc. O'Leary, Rosemary. 2015. "From Silos to Networks." In Public Administration Evolving: From Foundation to the Future , edited by Mary E. Guy & Marilyn M. Rubin, 85 100. New York: Routledge. Osborne, David, and Ted Gaebler. 1992. Reinventing Government . Reading, MA: Addison Wesley Publishing Company. Ostrom, Elinor. 1990. Governing the Commons . New York: Cambridge University Press. Ostrom, Vincent, Charles M. Tie bout, and Robert Warren. 1961. " The Organization of Government in Metropolita n Areas: A Theoretical Inquiry. " American Political Science Review 55 (4): 831 842.

PAGE 156

144 Polycentricity and Local Public Economies: Readings from the Workshop in Political Theory and Policy Analysis , edited by Michael D . McGinnis, 52 74, 119 138. Ann Arbor, MI: University of Michigan Press. Pahl Wostl, Claudia. 2009. "A Conceptual Framework for Analysing Adaptive Capacity and Multi Level Learning Processes in Resource Governance Regimes." Global Environmental Change 19: 354 365. Pekkanen, Robert J., and Steven Rathgeb Smith. 2014a. "Nonprofit Advocacy: Definitions and Concepts." In Nonprofits and Advocacy , edited by Robert J. Pekkanen, Steven Rathgeb Smith, and Yutaka Tsujinaka, 1 17. Baltimore, MD: Johns Hopkins University Press. Pekkanen, Robert J., and Steven Rathgeb Smith. 2014b. "Nonprofit Advocacy in Seattle and Washington, DC." In Nonprofits and Advocacy , edited by Robert J. Pekkanen, Steven Rathgeb Smith, and Yutaka Tsujinaka, 47 65. Baltimore, MD: Johns Hopkins University Press. Pfeffer, Jeffrey, and Gerald R. Salancik. 1978. The External Control of Organizations: A Resource Dependence Perspective . New York: Harper and Row. Pralle, Sarah B. 2006. Branching Out, Digging In . Washington, DC: Georgetown Uni versity Press. Journal of Public Policy 23 (3): 233 260. Putnam, Robert D. 1993. Making Democracy Work . Princeton, NJ: P rinceton University Press. Ragin, Charles C. 2008. Redesigning Social Inquiry: Fuzzy Sets and Beyond . Chicago, IL: University of Chicago Press. Reid, Elizabeth J. 2006. " Advocacy and the Challen ges It Presents for Nonprofits. " In Nonprofits and Governmen t , 2nd ed., edited by Elizabeth T. B oris and C. Eugene Steuerle, 343 371. Washington, DC: The Urban Institute Press. Roberts, Nancy. 2004. "Public Deliberation in an Age of Direct Citizen Participation." American Review of Public Administration 34 (4): 31 5 353. Western Political Quarterly 40 (3): 449 476. Schattschneider, E. E. 1960. The Semi cracy in America . New York: Holt, Rinehart and Winston, Inc.

PAGE 157

145 Schlager, Edella, and Michael Cox. 2018. "The IAD Framework and the SES Framework: An Introduction and Assessment of the Ostrom Workshop Frameworks." In Theories of the Policy Process , 4th ed., edited by Christopher M. Weible and Paul A. Sabatier, 215 252. New York: Westview Press. Schlager, Ede lla, and Tanya Heikkila. 2009. " Resolving Water Conflicts: A Comparative Analysis of Interstate River Compacts. " Policy Studies Journal 37 (3): 367 392. Schlozman, Kay Lehman, and John T. Tierney. 1986. Organized Interests and American Democracy . New York: Harper & Row. Schmid, Hillel, Michal Bar, and Ronit Nirel. 2008. "Advocacy Activities in Nonprofit Human Service Organizations." Nonprofit and Volunta ry Sector Quarterly 37 (4): 581 602. Shanahan, Elizabeth A., Mark K. McBeth, and Paul L. Hathaway. 20 11. " Narrative Policy Framework: The Influence of Media Polic y Narratives on Public Opinion. " Politics and Policy 39 (3): 373 400. Shanahan, Elizabeth A., Michael D. Jones, Mark K. McBeth, and Claudio M. Radaelli. 2018. "The Narrative Policy Framework." In Theories of the Policy Process , 4th ed., edited by Christopher M. Weible and Paul A. Sabatier, 173 213. New York: Westview Press. Shana han , Elizabeth A., Mark K. McBeth, Paul L. Hathaway, and Ruth J. Arnell. 2008. " Conduit or Contributor? The Role of Media in Policy Change Theory. " Policy Sciences 41: 115 138. Shapiro, Ian. 2015. "After On Democracy." In On Democracy , 2nd ed., edit ed by Robert A. Dahl. New Haven, CT: Yale University Press. Siciliano, Julie I. 1996. "The Relationship of Board Member Diversity to Organizational Performance." Journal of Business Ethics 15 (12): 1313 1320. Smith, Steven Rathgeb. 2007. "NGOs and Contra cting." In The Oxford Handbook of Publi c Management , edited by Ewan Ferlie, Laurence E. Lynn, Jr., and Christopher Pollitt, 591 614. New York: Oxford University Press. Suárez, David F., and Hokyu Hwang. 2008. "Civic Engagement and Nonprofit Lobbying in Ca lifornia, 1998 2003." Nonprofit and Voluntary Sector Quarterly 37 (1): 93 112. Truman, David B. 1951. The Governmental Process . New York: Alfred A. Knopf. United Nations (UN). 1992. Report of the United Nations Conference on Environment and Development . http://www.un.org/documents/ga/conf151/aconf15126 1annex1.htm

PAGE 158

146 Urban Institute. 2015. The Nonprofit Sector in Brief 2015 . https://www.urban.org/research/publication/nonprofit sector brief 2015 public charities giving and volunteering US Energy Information Administration (EIA). 2018 (December 31). Natural Gas . https://www.eia.gov/dnav/ng/hist/ngm_epg0_fgs_nus_mmcfa.htm Van de Ven, Andrew H. 2007. Engaged Scholarship . Oxford, UK: Oxford University Press. Warren, Mark E. 2001. Democracy and Association . Princeton, NJ: Princeton University Press. Warren, Mark E. 2003. "The Political Role of Nonprofits in a Democracy." Society 40 (4): 46 51. Weber, Edward P., and Anne M. Khademian. 2008. "Wicked Problems, Knowledge Challenges, and Collaborative Capacity Builders in Network Settings." Public Administration Review 68 (2): 334 349. Weible, Christopher M. 2014. "Introducing the Scope and Focus of Policy Process Research and Theory." In Theories of the Policy Process , 3rd ed., edited by Paul A. Sabatier and Christopher M. Weible, 3 21. Boulder, CO: Westview Press. Weible, Christopher M., and Paul A. Sabatier, eds. 2018. Theories of the Policy Proc ess , 4th ed. New York: Westview Press. Weible, Christopher M., and Tanya Heikkila. 2016. "Comparing the Politics of Hydraulic Fracturing in New York, Colorado, and Texas." Review of Policy Research 33 (3): 232 250. Weible, Christopher M., and Tanya Heikk Policy Sciences 50: 23 40. Weible, Christopher M., Tanya Heikkila, Peter deLeon, and Paul A. Sabatier. 2012. Policy Sciences 45 (1): 1 21. Wilson, David Sloan, El inor Ostrom, and Michael E. Cox. 2013. "Generalizing the Core Design Principles for the Efficacy of Groups." Journal of Economic Behavior and Organization 90 (Supplement): S21 S32. Yin, Robert K. 2014. Case Study Research , 5th ed. Thousand Oaks, CA: SAGE Publications. Energy Policy 107: 72 81.

PAGE 159

147 APPENDIX A Summary of Dissertation Chapters Chapter Research Question(s) Key Literature Methods 2 1) What are the different types of venues and policy issues that advocacy groups are involved in within the Colorado oil and gas policy subsystem? 2) Are there differences in how industry advocacy groups and e nvironmental advocacy groups participate across venues? Venue shopping, advocacy groups , policy process Data Source: Colorado news articles 2013 2017 Data Analysis: constant comparative analysis 3 1) W hat activities do environmental nonprofits engage in across different venues to influe nce the policy process? 2) W hat factors influence the types of advocacy these organizations choose? Policy process, nonprofit advocacy, nonprofit management Data Source: semi structured interviews Data Analysis: thema tic analysis 4 1) W hat activities do environmental nonprofits engage in t o influence the policy process? 2) W hich organizational characteristics are associated with specific types of advocacy activities ? Nonprofit management, nonprofit advocacy D ata So urce: s urvey responses, 990 forms, organization websites Data Analy sis: descriptive statistics, Qualitative Comparative Analysis

PAGE 160

148 APPENDIX B Policy Issues in News Articles Issue Number of Articles Percent of Articles Ban/Moratorium 131 25% Local Authority 78 15% Fracking Generally 72 13% Siting/Setbacks 52 10% Industry Safety 32 6% Economy/Jobs 26 5% Public Land 19 4% Air Quality 13 2% Procedural 13 2% Anti fracking Movement 12 2% Health Impacts 12 2% Industry News 12 2% Other 62 12% Ban/Moratorium Local Authority Fracking Generally Siting/Setbacks Industry Safety Economy/Jobs Public Land Air Quality Procedural Antifracking Movement Health Impacts Industry News Other

PAGE 161

149 APPENDIX C Activity by Jurisdiction Related to Local Authority 0 1 2 3 No Venue # Articles Environmental Groups Industry Groups 0 1 2 3 4 5 # Articles Environmental Groups Industry Groups 0 2 4 6 8 Local Ballot # Articles Environmental Groups Industry Groups

PAGE 162

150 0 5 10 Local Government # Articles Environmental Groups Industry Groups 0 5 10 State Ballot # Articles Environmental Groups Industry Groups 0 5 10 15 State Court # Articles Environmental Groups Industry Groups

PAGE 163

151 0 1 2 3 4 5 State Legislature # Articles Environmental Groups Industry Groups 0 1 2 3 4 State Regulatory Agencies # Articles Environmental Groups Industry Groups

PAGE 164

152 A PPENDIX D Interview Protocol Background / icebreakers What is your position/affiliation with [organization]? How long have you been involved? Tell me briefly about what your organization does. Advocacy activities Just to give us some context, my questions are specifically about oil and gas or hydraulic fracturing policy at the state level. What are about the last three policy issues that your organization was involv ed with in relation to oil and gas? These can be in any venue (e.g., rulemaking, ballot measures, legislation, courts). For each: What were the factors that led you to get involved? Once you decided to get involved, what activities did you engage in to try to influence the outcome? o These may include direct activities (e.g., testifying or lobbying), indirect activities (e.g., education or grassroots mobilization), and anything else t hat may influence policy or how people are thinking about issues. Were there some policy issues that came up that you chose not to get involved in? Why not?

PAGE 165

153 APPENDIX E Interview Code Forms, Participation by Venue and State Ballot Measure Participation. Colorado Organizations , No Inside Activities Reported Colorado Organizations , Hybrid Activities Org type Hybrid Policy Prep Coalition Work Litigation Research 501(c)(3), national 501(c)(3), national 501(c)(3), state gather signatures coalition work 501(c)(3), substate gather signatures 501(c)(3), substate 501(c)(3), Western states 501(c)(4), state coalition work 501(c)(4), state draft proposition language; gather signatures not registered, local gather signatures

PAGE 166

154 Colorado Organizations , Outside Activities Org type Outside Member Communication Media Institutional Change Calls for Action Official Position Public Efforts 501(c)(3), national 501(c)(3), national 501(c)(3), state social media take an official position 501(c)(3), substate door to door get out the vote; 501(c)(3), substate 501(c)(3), Western states 501(c)(4), state member petitions submit op/eds, LTEs; social media; press conference; TV commercial encourage elected officials to take a position/speak out poling/focus groups; create a website 501(c)(4), state digital media advertising; earned media advertising; social media; paid media seek endorsements from elected officials door to door canvassing; phone banking; education events; crowdsourcing; banners at intersections; town halls not registered, local hold an event; hand out info No reported activity in New Mexico or Oklahoma.

PAGE 167

155 Office Participation. Colorado Organizations , Inside Activities Org Type Inside Informal M eetings Formal P articipation 501(c)(3), national 501(c)(3), national 501(c)(3), state informal conversations 501(c)(3), substate presentation 501(c)(3), substate 501(c)(3), Western states 501(c)(4), state 501(c)(4), state not registered, local Colorado Organizations , No Hybrid Activities Reported

PAGE 168

156 Colorado Organizations , Outside Activities Org Type Outside Member Communication Media Institutional Change Calls for Action Official Position Public Efforts 501(c)(3), national 501(c)(3), national 501(c)(3), state public education 501(c)(3), substate emails; newsletters earned media public comments at meetings; comments from citizens; rally 501(c)(3), substate 501(c)(3), Western states 501(c)(4), state 501(c)(4), state not registered, local mailers letters to the editor; social media; press release encourage members to talk to media public comments at meetings No reported activity in New Mexico.

PAGE 169

157 Oklahoma Organizations , Inside Activities Org Type Inside Informal M eetings Formal P articipation 501(c)(3), local 501(c)(3), state participate in policy groups 501(c)(4), state chapter 501(c)(6), national 501(c)(6), state Oklahoma Organizations , No Reported Hybrid or Outside Activities

PAGE 170

158 Judicial Participation. Colorado Organizations , No Reported Inside Activities Colorado Organizations , Hybrid Activities Org Type Hybrid Policy Prep Coalition Work Litigation Research 501(c)(3), national 501(c)(3), national 501(c)(3), state 501(c)(3), substate 501(c)(3), substate serve as clients; provide standing declarations 501(c)(3), Western states 501(c)(4), state 501(c)(4), state file a lawsuit; litigate; not registered, local

PAGE 171

159 Colorado Organizations , Outside Activities Org Type Outside Member Communication Media Institutional Change Calls for Action Official Position Public Efforts 501(c)(3), national 501(c)(3), national 501(c)(3), state newsletters social media 501(c)(3), substate press release file an amicus brief 501(c)(3), substate update members earned media 501(c)(3), Western states 501(c)(4), state 501(c)(4), state earned media; social media not registered, local New Mexico Organizations , No Reported Inside Activities

PAGE 172

160 New Mexico Organizations , Hybrid Activities Org Type Hybrid Policy Prep Coalition Work Litigation Research 501(c)(3), national support those directly involved 501(c)(3), state litigate 501(c)(3), state chapter 501(c)(3), state chapter 501(c)(3), Western states coalition work lead counsel 501(c)(3), Western states 501(c)(4), state chapter not registered, Southwest states

PAGE 173

161 New Mexico Organizations , Outside Activities Org Type Outside Member Communication Media Institutional Change Calls for Action Official Position Public Efforts 501(c)(3), national earned media sign on to amicus brief 501(c)(3), state member emails invite people to oral arguments 501(c)(3), state chapter 501(c)(3), state chapter 501(c)(3), Western states 501(c)(3), Western states member communication earned media; social media 501(c)(4), state chapter not registered, Southwest states No reported activity in Oklahoma.

PAGE 174

162 Legislature Participation. Colorado Organizations , Inside Activities Org Type Inside Informal M eetings Formal P articipation 501(c)(3), national 501(c)(3), national testify 501(c)(3), state convene roundtables with elected officials and advocacy groups testify 501(c)(3), substate 501(c)(3), substate 501(c)(3), Western states 501(c)(4), state lobbying; assess legislators opinions 501(c)(4), state meet with legislators not registered, local host seminars for legislators; lobbying

PAGE 175

163 Colorado Organizations , Hybrid Activities Org Type Hybrid Policy Prep Coalition Work Litigation Research 501(c)(3), national 501(c)(3), national 501(c)(3), state convene people to create policy proposals coalition work 501(c)(3), substate 501(c)(3), substate 501(c)(3), Western states 501(c)(4), state draft legislation coalition work 501(c)(4), state consult on development of policy not registered, local draft legislation

PAGE 176

164 Colorado Organizations , Outside Activities Org Type Outside Member Communication Media Institutional Change Calls for Action Official Position Public Efforts 501(c)(3), national influence electoral politics 501(c)(3), national 501(c)(3), state email members social media encourage people to contact elected officials educational events; rally 501(c)(3), substate 501(c)(3), substate 501(c)(3), Western states 501(c)(4), state member petitioners; social media venue shift to regulatory encourage local electeds to lobby; ask members to email legislators bring citizens/ members to testify 501(c)(4), state have people call or email legislators organize town halls not registered, local encourage citizens to contact legislators train citizens; provide people with talking points; connect people with legislators

PAGE 177

165 New Mexico Organizations , Inside Activities Org Type Inside Informal M eetings Formal P articipation 501(c)(3), national communicate with state agencies; communicate with sponsor; educate legislators; meet with legislative staff; talk with legislators 501(c)(3), state lobbying; lobby the governor 501(c)(3), state chapter 501(c)(3), state chapter 501(c)(3), Western states meet with legislators 501(c)(3), Western states 501(c)(4), state chapter meet with legislators; meet with legislative staff not registered, Southwest states testify at hearings

PAGE 178

166 New Mexico Organizations , Hybrid Activities Org Type Hybrid Policy Prep Coalition Work Litigation Research 501(c)(3), national 501(c)(3), state 501(c)(3), state chapter coalition work 501(c)(3), state chapter support others' participation 501(c)(3), Western states coalition work 501(c)(3), Western states 501(c)(4), state chapter draft policy recommendations meet in coalitions to discuss policies not registered, Southwest states

PAGE 179

167 New Mexico Organizations , Outside Activities Org Type Outside Member Communication Media Institutional Change Calls for Action Official Position Public Efforts 501(c)(3), national email blasts social media influence electoral politics encourage people to call or email the governor; encourage people to call or email their legislators education; train advocates; organize public comments; deliver petition signatures; spread the word through volunteers 501(c)(3), state email blasts paid media; earned media; social media deliver petitions 501(c)(3), state chapter social media; earned media encourage member participation at public hearings; encourage people to contact legislators post info on website 501(c)(3), state chapter write letters to legislators signed by local leaders; public education 501(c)(3), Western states deliver petitions 501(c)(3), Western states 501(c)(4), state chapter press release not registered, Southwest states educate leaders

PAGE 180

168 Oklahoma Organizations , Inside Activities Org Type Inside Informal M eetings Formal P articipation 501(c)(3), local 501(c)(3), state 501(c)(4), state chapter 501(c)(6), national educate legislators testify 501(c)(6), state Oklahoma Organizations , Hybrid Activities Org Type Hybrid Policy Prep Coalition Work Litigation Research 501(c)(3), local participate in task force to create policy priorities 501(c)(3), state 501(c)(4), state chapter 501(c)(6), national 501(c)(6), state

PAGE 181

169 Oklahoma Organizations , Outside Activities Org Type Outside Member Communication Media Institutional Change Calls for Action Official Position Public Efforts 501(c)(3), local 501(c)(3), state 501(c)(4), state chapter members emails social media encourage members to contact legislators bring citizens to the capitol to talk to legislators; post information online 501(c)(6), national 501(c)(6), state

PAGE 182

170 Regulatory Agency Participation. Colorado Organizations , Inside Activities Org Type Inside Informal M eetings Formal P articipation 501(c)(3), national meet with agency staff; evaluate regulations provide technical information; participate in rulemaking 501(c)(3), national provide feedback on draft rules; participate in rulemaking 501(c)(3), state roundtables with local leaders 501(c)(3), substate meet with regulators participate in rulemaking 501(c)( 3), substate meet with agency staff participate in rulemaking; present testimony 501(c)(3), Western states participate in rulemaking; offer expert testimony 501(c)(4), state 501(c)(4), state not registered, local participate in rulemaking

PAGE 183

171 Colorado Organizations , Hybrid Activities Org Type Hybrid Policy Prep Coalition Work Litigation Research 501(c)(3), national draft guidelines; participate in working groups participate in stakeholder meetings produce fact sheets; commission research 501(c)(3), national participate in task force to write recommendations supporting partners who are more heavily involved 501(c)(3), state coalition work 501(c)(3), substate participate in task force 501(c)(3), substate 501(c)(3), Western states participate in policy negotiation coalition work 501(c)(4), state 501(c)(4), state not registered, local petition for rulemaking

PAGE 184

172 Colorado Organizations , Outside Activities Org Type Outside Member Communication Media Institutional Change Calls for Action Official Position Public Efforts 501(c)(3), national earned media talk to legislators about forcing regulatory action help citizens file complaints 501(c)(3), national 501(c)(3), state emails social media educational events 501(c)(3), substate advise and support members earned media encourage members to attend and speak at hearings; encourage members to write LTEs assemble committees of impacted people; build face to face relationships; public education 501(c)(3), substate emails social media; earned media lawsuits to force or prevent action ask members to submit comments; ask local governments or community leaders to file comments; submit comments information on website 501(c)(3), Western states member communication earned media; social media organize grasstops 501(c)(4), state 501(c)(4), state not registered, local earned media prepare others to speak at rulemakings; community meetings

PAGE 185

173 New Mexico Organizations , Inside Activities Org Type Inside Informal M eetings Formal P articipation 501(c)(3), national organize meetings with agency leaders 501(c)(3), state participate in administrative hearings 501(c)(3), state chapter meet with regulators 501(c)(3), state chapter 501(c)(3), Western states 501(c)(3), Western states meet with regulators; administrative lobbying testify at hearings; participate in rulemaking; provide expert testimony 501(c)(4), state chapter meet with agency staff not registered, Southwest states

PAGE 186

174 New Mexico Organizations , Hybrid Activities Org Type Hybrid Policy Prep Coalition Work Litigation Research 501(c)(3), national work with native groups 501(c)(3), state draft and file petitions work with partners on member communication 501(c)(3), state chapter 501(c)(3), state chapter 501(c)(3), Western states 501(c)(3), Western states participate in stakeholder groups coalition work; brief other groups publish reports 501(c)(4), state chapter not registered, Southwest states coalition phone calls

PAGE 187

175 New Mexico Organizations , Outside Activities Org Type Outside Member Communication Media Institutional Change Calls for Action Official Position Public Efforts 501(c)(3), national earned media submit comments attend meetings; help others submit comments; educational events 501(c)(3), state earned media; paid media; social media demonstrations; sign on to letters 501(c)(3), state chapter 501(c)(3), state chapter encourage local leaders to engage submit comments public education; participate in public hearings 501(c)(3), Western states emails support individuals in their advocacy activities; help citizens submit comments; participate in public hearings 501(c)(3), Western states communicate with members social media defend existing rules in court; sue agencies comment on proposed and final rules; protest agency actions post information on website 501(c)(4), state chapter paid media (to get people to comment); create videos of testimonies and post online; social media encourage others to comment collect and deliver comments from the public; hold rallies; hold events; attend public meetings not registered, Southwest stat es earned media; paid media help leaders draft statements; gather resolutions from local groups

PAGE 188

176 Oklahoma Organizations , Inside Activities Org Type Inside Informal M eetings Formal P articipation 501(c)(3), local 501(c)(3), state participate in rulemaking 501(c)(4), state chapter participate in rulemaking 501(c)(6), national provide support for agencies 501(c)(6), state participate in technical conferences Oklahoma Organizations , Hybrid Activities Org Type Hybrid Policy Prep Coalition Work Litigation Research 501(c)(3), local 501(c)(3), state propose rules 501(c)(4), state chapter litigate at agency 501(c)(6), national assemble a workgroup; develop primers on issues draft white papers 501(c)(6), state strategic meetings with allies produce white papers

PAGE 189

177 Oklahoma Organizations , Outside Activities Org Type Outside Member Communication Media Institutional Change Calls for Action Official Position Public Efforts 501(c)(3), local 501(c)(3), state submit comments, formal position statements 501(c)(4), state chapter social media encourage others to participate in public hearings education and support re how to engage 501(c)(6), national comment on draft and final rules; host events 501(c)(6), state update members encourage members to attend meetings submit comments education events; speaking engagements

PAGE 190

178 APPENDIX F QCA Results Countering arguments made by people you disagree with (intermediate solution) Equation: Coordinating political activities with allies (intermediate solution)

PAGE 191

179 Providing information to the news media (intermediate solution) Mobilizing the public (intermediate solution) Brokering agreements between parties (intermediate solution)