The Relationship Between Climate Change Policy and Socioeconomic Changes in the U.S. Great Plains

Abstract

As the United States struggles with national solutions to climate change, state and local governments have increasingly taken policy action in this area. Although existing research addresses why some places adopt climate change policy while others do not, much of this expresses policies as a function of factors in the present period or recent past, leaving the question of whether current climate change policy can be seen as a lagged response to longer term trends largely unaddressed. Examination of climate change policy as a response to longer term changes expands the existing understanding of why locations choose to be active in this area. Pairing unique climate change policy survey data from more than 200 local Great Plains governments with Census and environmental data from 1990 to 2000, this article examines whether changes in local socioeconomic and environmental factors in the 1990s are associated with climate change mitigation and adaptation policy adoption from the following decade, 2000–2010.

Keywords

climate change local government policy adoption

Introduction

Despite increasing awareness by both elected officials and the general public over the past two decades of the issue of climate change, climate change policy and planning in the United States remains localized and halting, at best. With limited federal action on this issue to date, it has fallen to primarily public officials at the regional, state, and local levels to take the lead in policy making and policy adoption in this area, and to choose whether or not to pursue any such policies locally at all. While some cities have indeed risen to this challenge, producing comprehensive packages of policies specifically framed as climate change-related, these locations remain the exception rather than the rule, with local responses to this issue still relatively varied, sporadic, and uncoordinated on the whole.

Just as policymakers in recent years have wrestled with the challenge of climate change planning, so too have academics attempted to understand what conditions lead to climate change policy adoption, and what factors explain why some locations are able to implement programs and policies to address this issue, while the vast majority of subnational governments are unable, or unwilling, to do so. Although the growing body of recent scholarship on this subject has produced a number of important insights into climate change policy planning and adoption, important areas and questions for research yet remain. Models in the existing literature have identified a number of factors which play a significant role in states’ and locations’ climate change policy-making decisions; however, the considerable variation left unexplained by these analyses suggests that our understanding of policy adoption in this area, and the inventory of variables which impact those decisions, is still far from complete.

We suggest that this gap has been caused, in part, by the tendency of existing research to focus on highly developed urban and coastal areas, where awareness of, and vulnerability to, climate change impacts is highest, and thus where more active support for local policy to address this issue may exist. Far less research has investigated climate change planning in locations with larger numbers of smaller communities and fewer central urban locations, such as the Great Plains region, where the issue is typically less visible to citizens and elected officials, and where the default public response to such policies might be assumed as ambivalence, if not overt hostility. These communities are certainly not immune to climate change impacts, and at least some cities in the Great Plains have responded to this challenge by being quite active in their adoption of climate change policies. What explains these curious cases of active climate change policy making in an area of the country where conventional wisdom would suggest public apathy, if not active opposition, to the issue? What factors differentiate these locations from neighboring communities which are far less active in this policy field?

The question also arises as to whether important differences exist among various climate change policies, and whether specific local factors are associated with adoption of different types of climate change responses. In particular, are different local characteristics associated with the adoption of mitigation policies, which are intended to directly reduce climate impacts, such as greenhouse gas (GHG) emissions, compared with those factors linked to adaptation policies, which represent intentional adjustments to protect a community from the impacts of climate change? To date, much of the existing literature has treated these types of policies as largely interchangeable; yet if some local factors are more conducive to mitigation policy, and others linked to adaptation policy, an aggregate treatment of these policy subsets may be missing important information.

While the bulk of the existing climate change policy scholarship has focused on the link between current conditions and policy outcomes, the role of change, and specifically of change in local conditions, has not been sufficiently examined. Although present-period conditions clearly play an important role in adoption of climate change policies in a given location (Brody et al. 2008; Krause 2012a; Portney 2003; Sharp, Daley, and Lynch 2011), we question whether changes in local circumstances, such as growth or decline in local population and available resources; changes to the character of that population in terms of education, political identification, or similar factors; or changes in local climatic conditions are also likely to play an important role in the decision calculus of local government, and in the preferences and needs of local populations, when it comes to climate change policy adoption.

It is this question of whether past changes in local conditions are associated with the range of climate change policies currently present in a given location that this study aims to address. Pairing data on current climate change policies and perceptions from an original survey of 232 cities located across 10 Great Plains states with U.S. Census information on changes to local demographic, political, and fiscal factors from 1990 to 2000, and National Oceanic and Atmospheric Administration (NOAA) data on changes in local climactic conditions over the same period, this study investigates whether and how changes in these conditions in the decade from 1990 to 2000 are associated with the number of adaptation, mitigation, and total climate change policies enacted by a given local government in the subsequent decade of 2001–2011. In doing so, we provide insight into the three open questions surrounding local climate change policy adoption noted above: Can past changes in local factors account for some of the variation in climate change policy adoption left unexplained by previous studies? Are climate change mitigation policies associated with different factors (and changes to those factors) than climate change adaptation policies? And, can changes in local factors help answer the question of why some Great Plains cities are active climate change policymakers in a region of the country where the relatively small size of many communities would seem to hinder policy adoption, and where a decided lack of public support for such action would otherwise be expected?

Following the “Introduction” section, the “Background and Previous Research” section of this article provides an overview of previous research into local climate change policy adoption and discusses the theoretical and empirical background supporting the impact of, and policymakers’ response to, measures and indicators of change. The “Data and Measures” section discusses the data resulting from our original survey of Great Plains local governments, as well as the additional data which we employ, and the sources from which that information was drawn, while the “Method” section formally presents the regression model used in the article. Results of our analysis are presented and discussed in the “Findings” section, while the “Conclusion” section concludes with the implications of the findings, and suggests refinements and expansions for future research.

Background and Previous Research

In the post-Kyoto world, the future of policy to combat climate change appears to be at a crossroads (Hoffmann 2011). Despite decades of effort, the United Nations Framework Convention on Climate Change (UNFCCC) has failed to achieve a universally accepted, legally binding set of enforceable international targets. Both the withdrawal of the United States and Canada from the Kyoto Protocol and the dramatic increase in carbon emissions over the past two decades from non-Annex 1 countries highlight the inherent complexities and conflicts of interest that make achieving a top-down solution so difficult. Recognizing the diminishing likelihood of this type of elegant international solution, as well as the consequences of continued delay, a growing number of scholars and policymakers are now advocating a decentralized, “bottom-up” approach to climate policy that places substantially more emphasis on the role of regional, state, and local governments (Rayner 2010; Van Asselt and Zelli 2014).

Proponents of the bottom-up approach argue that the rationale underlying the UNFCCC approach, largely based on the successful global response to ozone depletion, is flawed. They point out that chlorofluorocarbons (CFCs) were a relatively small group of chemicals, used primarily in industrialized nations, for which a number of cost-effective substitutes were readily available (Hulme 2009). The successful implementation of a similar approach to GHG emissions, in contrast, would require long-term, widespread cooperation between many nations with vastly different priorities, interests, and incentives (Bodansky 2011). They argue that while international solutions and ambitious targets remain valuable goals, the time has come to consider other ways to move forward in the interim (Rabe 2007; Victor 2001).

The basic principle of the bottom-up approach is that “climate change policies should be designed and implemented at the lowest feasible level of organization” (Rayner 2010, p. 617). In essence, they argue that over time, a greater aggregate impact on climate change will occur when jurisdictions pursue adaptation measures where adaptation makes sense, market mechanisms where markets make sense, and mitigation policies where they are supported by adequate political and fiscal resources. Eventually, advocates argue that the availability of carbon neutral energy at competitive prices will spur the adoption of more ambitious mitigation policies, but that “at present, it is more important to establish a clear direction of travel for every jurisdiction than to specify the exact point and time of arrival” (Rayner 2010, p. 620).

We argue that this clear line of travel in every jurisdiction is unlikely to occur without a better understanding of the factors that lead to the adoption of climate protection policies. Many scholars before us have studied questions of motivation but have naturally tended to study those places with innovative climate policies first. Studies have examined the potential effectiveness of municipal mitigation efforts at reducing global GHG emissions (Lutsey and Sperling 2008; Millard-Ball 2012), the value of participation in climate networks (Krause 2011a, 2011b, 2012a; Sharp, Daley, and Lynch 2011), and explored the reasons why cities choose to engage in climate protection activities in the first place (Anguelovski and Carmin 2011; Feiock and Coutts 2013; Kousky and Schneider 2003; Krause 2012b; Sippel and Jenssen 2010). In contrast, we focus our contributions on the cities of the Great Plains for several reasons, including the wide range of variation in climate protection efforts found there, the strong culture of political conservatism that pervades the region, and the dearth of existing research on smaller, noncoastal, nonurban communities, which have been underrepresented in the existing empirical literature.

Municipal-Level Climate Protection Policies

A substantial body of research now exists to better understand the actions of local governments with regard to mitigation activities, though less is known about adaptation efforts. In the United States, cities have been engaged with GHG mitigation for some time. Municipal activity began in the mid-1990s, as large, politically liberal cities with well-organized environmental movements that began to design and adopt policies explicitly framed as efforts to reduce GHG emissions. Coordination between cities was limited, and early policies tended to emphasize energy efficiency, conservation, and expanding public transportation systems, with little sense of how individual efforts might fit together to form a cohesive mitigation effort (Bulkeley 2010). The symbolic value, however, of “taking action” to combat climate change was politically attractive in these places.

A second wave of activity began in the early 2000s after the formal withdrawal of the United States from the Kyoto Protocol. As cities began to communicate and network with each other on the issue of climate protection, participation increased and climate protection policies became both more sophisticated and more ambitious (Warden 2011). In addition to adopting various mitigation measures individually, many cities expressed their commitment to climate protection during this period by joining a climate protection network, particularly the U.S. Mayor’s Climate Protection Agreement (USMCPA) (Byrne et al. 2007).¹ Since 2005, when Seattle Mayor Greg Nickels issued a challenge to the Mayors of America to take action on climate change, more than 1,000 cities have signed the USMCPA, agreeing to “strive” to meet or exceed the emissions reduction guidelines of the Kyoto Protocol in their communities (Warden 2011).

Most recently, another wave of activity has begun, triggered in part by the mounting evidence that CO₂ reduction efforts will not preclude significant climate change impacts. This wave has been characterized by a growing emphasis on building long-term sustainability, including adaptation policies and adaptation planning that had been largely absent from previous efforts (Feiock and Coutts 2013; Measham et al. 2011; Pielke et al. 2007; Preston, Westaway, and Yuen 2011). Recent studies have found evidence of increasing local attention to climate change adaptation planning (Baker et al. 2012; Measham et al. 2011), but it remains unclear whether cities pursuing adaptation policies are distinct—with distinct causes and motivations—from the cities that pursued mitigation policies in earlier waves. A key empirical goal of this research is to respond to these questions by disaggregating the influences that shape these two types of climate policies.

Mitigation Versus Adaptation

The distinction between mitigation and adaptation in climate policy literature is relatively new. In much of the early literature on climate policy, motivations for pursuing adaptation policies are not explicit but are assumed to be similar to those that drive mitigation policies, that is, human activities are causing significant changes to the Earth’s climate, so steps must be taken to both mitigate and adapt to these changes (Adger, Arnell, and Tompkins 2005; Füssel 2007; Tompkins et al. 2010). This line of reasoning, however, implies that adaptation policies could be considered stepping stones toward mitigation policies. We suggest that this reasoning is flawed, and that the forces that shape adaptation and mitigation policies are often distinct.

Mitigation is, at heart, a global commons problem, with limited incentives for local government participation. Yet, many local governments have adopted and continue to adopt these policies. Some policies may be successful because they are easily linked to tangible cobenefits such as increased energy efficiency, improved green spaces, or enhanced public transportation (Bulkeley 2010; Kousky and Schneider 2003). For other policies, however, the only cobenefits are symbolic. Their value to residents depends on their acceptance of the assumptions about climate change noted above. To a greater or lesser extent, we argue, support for mitigation policies is linked to acceptance of a causal relationship between human activities and climate change.

Adaptation policies, in contrast, tacitly acknowledge that climate changes are occurring that require “purposive changes to practices, processes, and structures to better cope with climate change and its impacts” (Vogel and Henstra 2015, p. 111), but do not require the acceptance of anthropogenic causes. The central goal of these policies is to reduce vulnerability and increase adaptive capacity in a community, regardless of the cause of the threat (Vogel and Henstra 2015). The benefits of adaptation policies are also more obviously local, and thus more easily justified on grounds unrelated to climate change.

To explore these differences, we divide a number of common local government activities according to their likely impacts on climate change. Activities that are likely to decrease vulnerability and/or increase adaptive capacity are categorized as adaptation policies; those that are aimed at decreasing GHG emissions are categorized as mitigation policies. In previous research, we test three clusters of factors (the policy environment, government attitudes, and community attitudes) on climate policies disaggregated in this manner. The findings suggest not only that these clusters do shape climate policies in the Great Plains region but also that many of the factors shaping mitigation policies are different from those shaping adaptation policies (Wood, Hultquist, and Romsdahl 2014).

Although we do not generalize beyond the Great Plains, a number of other researchers in other contexts have found that cities vary substantially in terms of resources, political constraints, public attitudes, and vulnerability to climate impacts (Bulkeley 2010; Castán Broto and Bulkeley 2013; Hunt and Watkiss 2011; Liu et al. 2010; Zahran et al. 2008), and that these factors account for a significant portion of observed variation in a city’s approach to climate protection. Nevertheless, a significant portion remains unexplained, inviting further research. In this article, we build upon this body of work by examining the extent to which public policies also reflect an organized, societal response to change over time.

Our theoretical grounding for this notion is derived from the policy processes literature. John Kingdon’s (1984) multiple streams theory, in particular, emphasizes the fundamental and essential role of time in the policy process. For Kingdon, problems, solutions, and politics are represented as distinct streams traveling independently through time. As time passes, changes in one stream—though independent—can and do trigger responses in other streams in subsequent time periods. These responses are typically facilitated by policy entrepreneurs who, with preferred solutions of their own, watch for changes to trigger a window of opportunity for them to link their solution to the appropriate problem.

In the case examined here, changes in the problem stream might be expected to include changes in the physical environment such as average temperature and precipitation patterns or the number of extreme weather events. Changes in the political stream might include indirect political factors, such as age, education, and income, as well as direct political factors such as the political identification of residents (Moser and Ekstrom 2010). We suggest that changes in these variables during one period may open a window of opportunity for policies adopted in a subsequent period. To test for the influence of previous changes on current policy adoption, we incorporate into our base model variables representing changes in the physical environment, political conditions, and socioeconomic characteristics of each city during the decade prior to the period under study.

Overall, our work makes three important contributions. First, it examines the role of change in the adoption of climate change-related policies at the local level. While the bulk of past research has looked at the association between local policy and potential future impacts from climate change (Hunt and Watkiss 2011), or contemporaneous community, governmental, and environmental attributes (Bulkeley and Betsill 2013; Feiock and Bae 2011; Krause 2012; Lee and Koski 2012; Measham et al. 2011; Zahran et al. 2008), we contend it is also plausible that climate change policy in the present may be, at least in part, a response to changing community conditions in the recent past. If citizens have the potential to express their preferences by “voting with their feet” (e.g., Tiebout 1956), local officials may be extremely sensitive to changes in local population characteristics. They might consider adopting policies they associate with the preferences of particular growing or shrinking segments of their community (Hanna 2005). Alternatively, they might consider adopting policies designed to attract desirable socioeconomic groups whose addition to a community would be expected to increase the local tax base, economic productivity, or social capital in general (Florida 2012). Whichever one finds most likely, we feel that there are thus sufficient potential causal mechanisms and theoretical support for the idea that change in local environmental characteristics should have an impact upon local climate policy decision making for these factors to be examined.

Second, our work differentiates between adaptation and mitigation activities, allowing for the possibility that different factors may play a role in motivating the number of policies of each type which a given jurisdiction chooses to adopt, or that certain factors may have a greater marginal impact in terms of policies adopted for one type compared with the other. While much of the literature has either focused on mitigation activities alone (Krause 2011a; Rabe 2007) or on climate change policies of both the mitigation and adaptation varieties (Bulkeley et al. 2011), an emerging segment of research in this area suggests that different types of locations may favor one of these policy types over the other (Castán Broto and Bulkeley 2013). Given that adaptation activities typically represent local government policies which would be normally undertaken, albeit with arguably less frequency or intensity, by local government in the absence of climate change, whereas mitigation activities all display some impact on GHG emissions (although some of these activities also produce varying degrees of nonclimate change-related cobenefits), one might reasonably expect that the latter policy group may be more easy to connect to (or, alternately, more difficult to divorce from) the issue of climate change. Along with potential differences in local government capacity, familiarity, and resource needs in providing these types of policies this suggests the possibility of differences in what types of locations provide local climate change measures of each type.

Third, our work also looks at an area of the United States, the Great Plains Region, which is largely unexamined by work in the field, and includes small communities which likewise have been ignored by previous large-sample climate change policy studies. While Sharp, Daley, and Lynch (2011, p. 443) hypothesized that “climate protection policy, although not totally out of the reach of small cities, is presumably of a different caliber than in larger cities,” given that their work examines cities with populations 100,000 or greater, and other work in the field focuses on either case studies or similarly large communities (e.g., Krause 2011a, 2012b, which consider cities of 50,000 or more residents), this proposition remains to be tested. While one would certainly expect community size to limit the extent to which local governments are able to pursue a climate change policy agenda, and perhaps most especially to bear the cost burden necessary to participate in large-scale climate change programs such as International Council for Local Environmental Initiatives (ICLEI), at the same time, one would not expect a complete absence of climate-change-related policy action from small communities. Likewise, although the threats posed by climate change may not be as visible to areas such as Great Plains as they are to the larger communities and coastal areas on which much of the literature to date has focused, at the same time, given that this section of the country contains roughly one-seventh of the U.S. population, what these communities are doing in terms of climate change policy, what local environmental and political stimuli they respond to in these decisions, and how these choices differ from other areas of the country which have been examined represent questions which should be addressed. These questions become more salient still in light of emerging research (Wood, Hultquist, and Romsdahl 2014) which suggests that local governments in this area of the country, which might otherwise be assumed resistant to subscribe to the idea of climate change and to related policies due to its general social conservatism, are engaged in adoption and revision of a considerable variety of mitigation and adaptation measures.

Data and Measures

Our article draws its information on local climate change policy adoption and current local attitudes toward climate change from data obtained via an original survey of 232 mayors’ offices in 10 Great Plains states,² which was collected in spring 2012 using an online survey instrument via SurveyMonkey. To create a representative sample frame, a proportion of the 5,000 municipalities identified in the 2010 Census was calculated for the 10 target states, and an estimated number of cities to sample from each state was based on an overall sample target of 900. Cities with populations below 1,000 were excluded, and all cities with populations above 100,000 were included in the sample. The remainder of the sample in each state was selected randomly from cities with populations between 1,001 and 100,000.

Introductory letters were sent to each official city address extending an invitation for that respective city’s mayor (or designated representative) to complete our survey online, along with a unique, city-specific link to the online SurveyMonkey instrument, which allowed us to track responses and pair each survey with the additional, secondary source data which we employ in our analysis. Paper copies of the survey and self-addressed stamped envelopes were also included in the mailing for those who preferred to complete the survey in hard copy form and return it by mail. A second mailing was sent approximately three weeks after the initial contact, with nonrespondent cities contacted by telephone approximately three weeks after the second mailing. A final sample of 232 cities ultimately responded to our survey request, yielding a response rate of approximately 25%.³ While not ideal, this response rate was not unexpected, given the considerable workload and part-time nature of mayors in smaller communities and the political controversy associated with climate change, and was generally consistent with other surveys of a similar nature.

Mayors were selected as the initial contact point to promote consistency throughout the sample. Cities in our sample vary substantially in size and resources, but every city in the sample has a mayor. The use of practitioners in this way has been used previously in other contexts. Berner, Amos, and Morse (2011), for instance, interviewed a variety of local stakeholders about their own opinions and their assessment of community conditions with regard to citizen participation. Ho (2006) also surveyed mayors on the effectiveness of performance measurement in their cities.

To measure the level of climate change policy activity in a city, we asked mayors to indicate, from a list of 14 mitigation activities, those that have been implemented or updated in their city in the past 10 years. A similar list of 14 adaptation activities⁴ were also presented in the survey, with mayors being asked to identify those which their city engaged in. The inventory of 28 policies on which we collect data is adapted from the list of activities identified by Krause (2011a) although, following the definitions provided by Wood, Hultquist, and Romsdahl (2014), we elect to further differentiate these activities as either mitigation or adaptation.

Measures of change in local factors over time, the independent variables of particular interest in our present study, are based on data drawn from three sources. Percent changes in county-level population, per capita income, percent of residents with a bachelor’s degree or higher, and median age for the 1990–2000 period were calculated using U.S. Census data. Information on changes in political identification, based on the proportion of county residents who voted Democratic in the 1988 and 2000 Presidential elections, was obtained from David Leip’s Atlas of U.S. Presidential Elections (uselectionatlas.org). Finally, changes in average temperature (in degrees Fahrenheit) and drought conditions (measured using the Palmer Drought Severity Index [PDSI]⁵) at the Climate Division⁶ level for the 1990–2000 period were calculated using information from the NOAA’s National Climate Data Center.

While our primary focus in this article is an examination of the relationship between past changes in demographic, economic, and climate-related factors and climate change policies currently in place in a given location, we do not propose that such changes are exhaustive in their ability to explain local climate change adoption. Factors related to cities’ current population and circumstances also play an important role. Change-based measures should be incorporated alongside such contemporaneous factors to create a more robust model than would be possible where only measures of change or present-period factors alone included. To that end, our survey also included a number of questions intended to gauge the views of mayors, and their perceptions of citizen attitudes, with respect to climate change policy.

Mayors were asked to rank their level of agreement, using a Likert-type scale (with “5” indicating strong agreement and “1” indicating strong disagreement) with the following five statements intended to gauge their attitude toward local climate change vulnerability and the appropriate role of local government in addressing the issue: “In the absence of national climate change legislation, state and local governments should act on their own to reduce the impacts of climate change”; “State and local governments should wait for national climate change legislation before addressing climate change”; “State and local governments should just adapt to climate change impacts as they occur”; “Government action is not needed to address climate change—People will adjust their behaviors as free-market prices change”; and “My community is highly vulnerable to impacts from climate change” (see the appendix, Question 1).

Our survey also asked mayors, using a similar Likert-type scale, to indicate how well each of 12 statements reflected their city’s public atmosphere toward climate change (see the appendix, Question 4). In asking them to rate the public atmosphere of their community, we introduce the possibility that mayors could conflate their own attitudes with those of their community. But as political actors, “single-minded seekers of re-election,” we might also expect them to be finely attuned to the attitudes and preferences of their constituents (Mayhew 1974, p. 5). They are, perhaps, better suited than anyone to assess the atmosphere of their community on the subject of climate change, as misreading the public appetite for climate protection policies, particularly in the Great Plains, can have dramatic consequences for reelection.

To supplement our survey and better describe the characteristics of each city’s current community environment, additional data were obtained from a variety of sources. The 2010 population of each location was obtained from the U.S. Census Bureau. The cumulative value of disaster declarations affecting each location over the past 10 years was obtained from the Federal Emergency Management Agency (FEMA). Information on cities’ adoption of the Mayors’ Compact on Climate Change was obtained from the United States Conference of Mayors.

To capture the effects of state-level climate change policies, data on whether each location had a Climate Action Plan or a voluntary GHG reporting system at the state level were drawn from the Center for Climate and Energy Solutions (C2ES). While these two types of policies both represent state-level responses to climate change, given important differences between them, one might expect each type of policy to differ in its impact on local climate change-related actions. State Climate Action Plans tend to outline steps that states (and localities) can take to reduce their contribution to climate change (c2es.org), identifying opportunities, setting targets, and establishing incentives which encourage local policy adoption, particularly in the area of mitigation. Voluntary GHG reporting systems, in contrast, reflect state actions that can take the place of, and therefore reduce the need for, local actions. Under these systems, GHG emitters typically report directly to state-level agencies, leaving local officials free to pursue other priorities while still feeling confident, and able to demonstrate to their constituents, that “something is being done” about climate change. Inclusion of both of these measures thus allows for the possibility that different state-level climate change actions might have differential impacts in either stimulating or precluding local actions. The following section of the article further describes the methods which we employ in using this data to model the influences of past changes and present-period factors on local mitigation and adaptation policy adoption.

Method

Following the general approach of Wood, Hultquist, and Romsdahl (2014), we measure the impact that these factors have on the number of climate change policies which a location has adopted or updated in the past 10 years, employing a negative binomial regression to estimate three separate models, each of which adopts the following basic form:

# C l i m a t e P o l i c i e s_{i} = β_{0} + β_{1} {\vec{P O L I C Y}}_{i} + β_{2} {\vec{G O V' T}}_{i} + β_{3} {\vec{C O M M U N I T Y}}_{i} + e_{i} .

The dependent variable in this equation is the number of climate change policies adopted or updated by a given city in the past 10-year period. Using the three vectors of independent variables, and adding several new independent variables designed to capture past changes to local conditions, as discussed below, we run three models with three separate measures for the dependent variable. The first model’s dependent variable is the total number of climate change policies, including both mitigation and adaptation measures, which the ith city has implemented or updated over the past decade. The second and third models disaggregate the dependent variable, measuring the impact of the independent variables on the number of mitigation and adaptation policies, respectively, adopted by a given city. This disaggregation allows us to assess whether the factors, particularly past changes in local circumstances, which play a role in the local adoption policies are the same as those which are linked to mitigation policies, or whether there are important differences between the variables associated with each of these two subsets of climate change-related policies.

The explanatory factors include three vectors intended to model three clusters of influences—policy environment, government attitude, and community atmosphere—with each of these containing several independent variables. The vector representing the impact of various elements contained in the policy environment, ${\vec{P O L I C Y}}_{i}$ , consists of 13 measures, 5 of which capture current attributes of the local policy environment and 8 of which depict past changes in that environment from 1990 to 2000. Specific variables included in this vector are a present-period (2010) measure of a city’s population, and measures of the city’s population change (in percentage terms), change in city per capita income (in percentage terms using 2000 dollars), change in average education (measured as the change in the percent of residents with a bachelor’s degree or higher), and the percent change in median age, all of which are assessed over the 1990–2000 period. The vector also includes the cumulative value (in billions of 2011 dollars) of FEMA disaster assistance disbursed to the county in the past 10 years (used here to represent the general frequency and/or intensity of severe weather events in an area), the mayor’s perceived vulnerability of the community to climate change (Survey Question 1, Part “e”), a dummy equaling 1 if the city’s state had a Climate Action Plan, an additional dummy variable indicating whether the city’s state had a voluntary GHG reporting system in place (0 if no GHG reporting requirement was present), and a final dummy signifying whether the location employed a city manager form of local government. Two measures of local climate change, change in average annual temperature⁷ (degrees Fahrenheit) and change in average annual drought conditions (using the PDSI) from 1990 to 2000, both of which are measured at the Climate Division level, are also included in this vector.

The second vector, ${\vec{G O V' T}}_{i}$ , consists of five variables intended to capture the attitude of local government toward climate change policy. The first of these variables is a dummy indicating whether or not the city had signed the USMCPA. The remaining four variables, drawn from our survey instrument, reflect mayoral attitudes toward the role of local governments in addressing climate change (Question 1, Parts “a” through “d”), with each measured using the 5-point Likert-type scale noted above.

The model’s final vector, ${\vec{C O M M U N I T Y}}_{i}$ , is intended to capture a city’s community atmosphere toward climate change using five variables. The first of these, intended to capture local political affiliation or sentiment, is the change in the percent of the county which voted Democratic in the 1988 and 2000 Presidential elections.⁸ The remaining four variables are based on Survey Question 4 asking mayors to rate their level of agreement with each of 12 statements describing their community’s attitude toward climate change. Although each of these 12 questions captures an aspect of local attitudes toward climate change which is unique, at the same time, we hypothesize that the attitudes represented in these statements may also reflect a smaller number of distinct, common dimensions or undercurrents in each location’s public atmosphere. To examine this possibility, we conduct a factor analysis to determine the presence, and number, of any such latent variables which might be better used to describe and understand local climate change attitudes. Table 1 presents the results of this analysis.

Table 1.

Factor Analysis—Public Atmosphere Toward Climate Change Rotated Component Matrix.

	Factor 1: Support	Factor 2: Skepticism	Factor 3: Confusion	Factor 4: Apathy
Public support for policies to address Climate Change	.860	−.094	.086	−.063
Public demands action on Climate Change	.831	−.191	−.059	−.058
Concern about Climate Change	.815	−.212	−.025	−.101
Anger about Climate Change	.791	−.059	.062	.155
Frustration about Climate Change	.750	−.005	.452	.113
Our city is vulnerable to Climate Change impacts	.609	−.248	.188	.058
Climate Change is a natural cycle	−.087	.877	−.024	.033
Climate Change is exaggerated by the media	−.238	.846	.082	.092
Climate Change science is uncertain	−.102	.843	−.008	.026
The feeling that Climate Change is not relevant	−.170	.738	.045	.194
Confusion about Climate Change	.151	.070	.955	.005
Nobody talks about it (not aware of Climate Change)	.045	.217	.019	.959

Note. Principal components extraction with Varimax Rotation and Kaiser Normalization. Rotation converged in five iterations. Bold values indicate signficance at the 0.05 level or higher.

Based on the results of this principal components extraction of the responses, we argue that there are four distinct dimensions that make up the public atmosphere in the data. We label these factors as “Support,” “Skepticism,” “Confusion,” and “Apathy.” Each component question clearly loads at high levels with respect to a single factor (0.609 or higher) and loads at low levels with respect to the other three factors. Interestingly, this suggests that each of these four resulting factors captures a separate element of local public sentiment on the issue of climate change policy, and that the variation in local attitudes, at least in the Great Plains, is a complex one which extends beyond the traditional support/skepticism dichotomy. In general, we hypothesize support and partisanship to be positively associated with climate protection activities, and Skepticism, Confusion, and Apathy to produce negative associations.

Findings

Before turning to the results of our regression analysis, as described above, it is interesting to note a number of important findings which are evident from a simple examination of the number and types of mitigation and adaptation activities engaged (Table 2) in by the communities which responded to our survey. First, there is a wide variation present in local mitigation and adaptation policy adoption. The distribution of mitigation activities in the data ranges from 0 to 14, with a mean of 3.88 of these policies per city. Most communities in the sample had implemented mitigation policies to some extent, with 90% of cities reporting at least one such policy in place, but with only 10.9% implementing more than 7 of the 14 activities, and only three cities implementing all 14 measures. Adaptation policy adoption was somewhat stronger compared with mitigation, with a mean of 6.6 policies per city. As was the case with mitigation policies, adaptation policy adoption varied considerably—Although more than 95% of responding cities engage in at least one adaptation activity, and 42.6% have adopted half or more of the 14 policies, only a very small number—three cities—have all 14 adaptation policies in place.

Table 2.

Descriptive Statistics (n = 211).

Variable	M	SD	Minimum	Maximum
All climate change	10.25	5.55	0	28
Adaptation	6.56	3.43	0	14
Mitigation	3.69	2.77	0	14
Population (million)	0.24	0.52	0.00	4.11
Population, % change	12.49	34.86	−63.30	313.30
Per capita income, % change	15.51	15.81	−43.66	131.81
Population with college degree, % change	3.78	3.86	−5.00	20.20
Median age, % change	5.53	8.04	−12.10	59.50
Value of disaster assistance (US$billion)	0.49	0.92	0	4.32
Average annual temperature change	0.67	0.55	−0.50	2.00
Average annual drought change	1.47	1.14	−1.44	5.37
Perceived vulnerability of community	2.88	1.20	1	5
State Climate Action Plan	0.37	0.49	0	1
State voluntary GHG reporting system	0.63	0.48	0	1
City manager	0.38	0.49	0	1
Signed mayor’s compact on Climate Change	0.08	0.27	0	1
Cities/states should act on Climate Change	3.38	1.25	1	5
Cities/states should wait before acting on Climate Change	3.65	1.21	1	5
Cities/states should adapt to Climate Change impacts	3.09	1.20	1	5
Market forces will adjust behavior on Climate Change	2.88	1.42	1	5
Democratic vote for president, % change	−7.38	5.53	−28.70	5.10
Factor 1: Concern/support for Climate Change	−0.06	0.92	−1.72	3.15
Factor 2: Ideological skepticism about Climate Change	0.06	0.99	−2.78	1.95
Factor 3: Confusion about Climate Change	0.01	0.98	−2.15	2.31
Factor 4: Apathy toward Climate Change	0.02	0.98	−2.21	2.21

Note. GHG = greenhouse gas.

This distribution represents an interesting finding which further reinforces the intriguing situation represented by climate change policy adoption in the Great Plains. While the initial assumption regarding this region of the country might be an absence of climate change policy action, not only do these results suggest that there are some locations within this region which are actively engaged in climate change policy making but also that the typical Great Plains city has at least some policies in place which at least in part either address or respond to climate change, and that only a very small minority have introduced or updated no such policies over the past decade.

Tables 3 and 4 also display key differences in adaptation versus mitigation policy adoption. Those mitigation policies which are implemented most frequently in our sample are ones that are less likely to be framed explicitly as addressing climate change. None of the top five most common mitigation policies are framed as climate change in more than a quarter of the jurisdictions in which they occur, while three of the four least often implemented mitigation policies are framed as climate change in a substantial number of the instances in which they are adopted. Conversely, all adaptation policies appear unlikely to be framed in such a manner, regardless of their degree (or lack thereof) of adoption. This would thus suggest that different local factors may be at work surrounding adaptation policies compared with mitigation activities, whether in terms of the rationales motivating those policies or the presence or absence of local demand for action characterized as specifically climate change-related. This further reinforces the idea that climate change policy adoption in the Great Plains region is a relatively complex issue. In an attempt to understand that complexity, we now turn to our regression results.

Table 3.

Local Government Implementation and Framing of Selected Climate Change Mitigation Activities (n = 211).

Mitigation Activity	No. of Governments Implementing	% of Governments Implementing	No. Implementing and Framing as Climate Change Policy	% of Those Implementing Framing as Climate Change Policy
Community-wide hike and bike trails in place	124	58.8	13	10.5
Efficient lighting installed in city streetlights	114	54.0	14	12.3
Tree ordinance adopted specifying planting/removal requirements for developers	111	52.6	9	8.1
City provides information to residents on energy efficiency	105	49.8	19	18.1
Residential yard waste is composted	90	42.7	9	10.0
Policy for city equipment and appliances, purchase only EnergyStar labeled	60	28.4	14	23.3
City provides residents or developers incentives for energy efficiency building/improvements	50	23.7	10	20.0
City has regulations requiring private/commercial buildings to meet efficiency standards	37	17.5	8	21.6
Vehicle fleet being made “green” (i.e., converted to hybrids, high efficiency, alternative fuels)	36	17.1	8	22.2
Planning and zoning decisions explicitly consider effect on sprawl or GHG emissions	28	13.3	7	25.0
Inventory conducted on local GHG emissions	19	9.0	11	57.9
City offers incentives to its employees to use methods other than single occupancy vehicles to commute to work	18	8.5	5	27.8
GHG reduction plan developed and formally adopted by city	15	7.1	8	44.4
Climate protection activities have line in municipal budget	12	5.7	7	58.3

Note. GHG = greenhouse gas.

Table 4.

Local Government Implementation and Framing of Selected Climate Change Adaptation Activities (n = 211).

Adaptation Activity	No. of Governments Implementing	% of Governments Implementing	No. Implementing and Framing as Climate Change Policy	% of Those Implementing Framing as Climate Change Policy
Warning system(s) (e.g., storm alerts, fire risk)	164	77.7	6	3.7
Flood risk maps	155	73.5	9	5.8
Flood management plans	133	63.0	6	3.7
Emergency evacuation plans	126	59.7	3	2.4
Educational materials/information to public (e.g., tree planting, composting, energy saving tips, etc.)	110	52.1	10	9.1
Stormwater pollution prevention plan	106	50.2	11	10.4
Drought-dependent water use restrictions	106	50.2	7	6.6
Mosquito/pest control policies	96	45.5	4	4.2
Regional cooperation on natural resource management (e.g., river, lake, drought, etc.)	94	44.6	6	6.4
Disclosure requirements (e.g., on flooding, erosion risks)	77	36.5	4	5.2
Soil erosion management plans	75	35.6	2	2.7
Density restrictions in city planning	70	33.2	6	8.6
Setbacks along riverbanks	55	26.1	4	7.3
Species and/or habitat protection plans (e.g., land trusts)	26	12.3	4	15.4

Influences on Mitigation and Adaptation Policy Adoption

Model I—All Climate Policies

Table 5 presents the results of the article’s regression analyses, modeling the number of climate change policies adopted as a function of the three groups of variables describing the policy environment, government attitudes, and community atmosphere, and which include present-period factors as well as measures of changes in local circumstances in the previous decade.⁹ As direct interpretation of coefficients from a negative binomial model is not as straightforward as with standard ordinary least squares (OLS) regressions, we also include Table 6, which presents the marginal impact of a one-unit change in each independent variable on the predicted number of climate change policies adopted for a specific reference case, which is here assumed to be a city which has not signed the Mayors’ Compact on Climate Change, in a state which does not have either a Climate Action Plan or a voluntary GHG reporting system, and which does not employ a city manager form of local government (i.e., these three variables are taken to be zero), and where all other independent variables are held at their mean values, as indicated in Table 2.

Table 5.

Influences on Mitigation and Adaptation Policy Adoption (Natural Log of Number of Policies Adopted^a).

	Model I: All Climate Policies	Model II: Mitigation Policies	Model III: Adaptation Policies
Constant	1.253	0.138	0.928
The policy environment
Population (1,000,000 persons, 2010)	−0.039	−0.065	−0.023
Population, % change (1990–2000)	0.003**	0.002	0.003**
Per capita income, % change (1990–2000)^b	−0.007**	−0.012***	−0.004
Population with college degree, % change (1990–2000)	−0.010	0.014	−0.020
Median age, % change (years, 1990–2000)	0.007	0.005	0.007
Value of disaster assistance ($billion)^b,c	0.116**	0.209***	0.071
Average annual temperature change (°F, 1990–2000)	0.069	−0.048	0.110
Average annual drought change (PDSI, 1990–2000)	0.142***	0.171***	0.112***
Perceived vulnerability of community	0.056	0.067	0.049
State has a Climate Action Plan^d	0.345**	0.592***	0.217
State has voluntary GHG reporting system^e	−0.200	−0.395**	−0.099
City manager	0.280***	0.334***	0.225***
Attitude of government
Signed mayor’s compact on CC^f	0.182	0.343**	0.086
Cities/states should act on CC if feds don’t	0.150***	0.164**	0.136***
Cities/states should wait for feds before acting	0.076	0.099	0.069
Cities/states should just adapt to CC impacts	−0.089**	−0.137***	−0.066
Market forces alone will adjust behavior on CC	0.071	0.072	0.063
Community atmosphere
Democrat vote: President, % change (1988–2000)	0.018**	0.013	0.021**
Factor 1: Concern/support for CC	−0.007	−0.027	0.005
Factor 2: Ideological skepticism about CC	0.058	0.030	0.067
Factor 3: Confusion about CC	0.109**	0.101	0.106**
Factor 4: Apathy toward CC	−0.042	−0.110**	−0.014
Pseudo-R²	.075	.105	.067
Likelihood ratio test	0.000***	0.000***	0.001***

Note. GHG = greenhouse gas; FEMA = Federal Emergency Management Agency.

Implemented or updated in the last 10 years.

2011 constant US dollars.

Total value of county-level FEMA disaster declarations over the last 10 years.

http://www.c2es.org/us-states-regions/policy-maps/action-plan (dummy variable).

http://www.c2es.org/us-states-regions/policy-maps/ghg-reporting (dummy variable).

http://www.usmayors.org/climateprotection/agreement.htm (dummy variable).

p < .05. ***p < .01. Bold values indicate signficance at the 0.05 level or higher.

Table 6.

Marginal Impacts, Influences on Mitigation and Adaptation Policy Adoption (Number of Policies Adopted).

	Model I: All Climate Policies	Model II: Mitigation Policies	Model III: Adaptation Policies
The policy environment
Population (1,000,000 persons, 2010)	−0.329	−0.182	−0.130
Population, % change (1990–2000)	0.025**	0.006	0.018**
Per capita income, % change (1990–2000)^b	−0.060**	−0.034***	−0.025
Population with college degree, % change (1990–2000)	−0.085	0.040	−0.114
Median age, % change (years, 1990–2000)	0.007	0.015	0.042
Value of disaster assistance ($billion)^b,c	1.059**	0.677***	0.422
Average annual temperature Change (°F, 1990–2000)	0.616	−0.137	0.667
Average annual drought change (PSDI, 1990–2000)	1.312***	0.542***	0.674***
Perceived vulnerability of community	0.499	0.202	0.285
State has a Climate Action Plan	3.547**	2.353***	1.380
State has voluntary GHG reporting system	−1.561	−0.950**	−0.539
City manager	2.785***	1.153***	1.444***
Attitude of government
Signed mayors’ compact on CC	1.720	1.190**	0.515
Cities/states should act on CC if feds don’t	1.398***	0.519**	0.831***
Cities/states should wait for feds before acting	0.628	0.303	0.409
Cities/states should just adapt to CC impacts	−0.735**	−0.373***	−0.367
Market forces alone will adjust behavior on CC	0.630	0.219	0.371
Community atmosphere
Democrat vote: President, % change (1988–2000)	0.159**	0.039	0.120**
Factor 1: Concern/support for CC	−0.057	−0.076	0.028
Factor 2: Ideological skepticism about CC	0.516	0.089	0.395
Factor 3: Confusion about CC	0.988**	0.310	0.640**
Factor 4: Apathy toward CC	−0.360	−0.303**	−0.082
Pseudo-R²	.075	.105	.067
Likelihood ratio test	0.000***	0.102***	0.001***

Note. PDSI = Palmer Drought Severity Index; GHG = greenhouse gas; FEMA = Federal Emergency Management Agency.

Implemented or updated in the last 10 years.

2011 constant US dollars.

Total value of county-level FEMA disaster declarations over the last 10 years.

Represents the change in predicted number of policies adopted resulting from a one-unit increase in the associated independent variable, relative to a reference case where all continuous independent variables are held constant at their mean values, and Climate Action Plan, voluntary GHG reporting system, signed mayors’ compact on climate change, and city manager are equal to 0. The numbers of Total Climate Policies, Mitigation Policies, and Adaptation Policies predicted by the model for such a reference case are 8.61, 2.91, and 5.71, respectively.

p < .05. ***p < .01. Bold values indicate signficance at the 0.05 level or higher.

The results of Model I are presented in the first column of these tables labeled “All Climate Policies,” expressing the dependent variable as the total number of mitigation and adaptation activities engaged in by a jurisdiction. Here, several interesting results are apparent. First, among factors in the policy environment, past population change exhibits is significantly associated at the 5% level with climate change policy adoption, with each additional percentage point of population growth in 1990–2000 linked to 0.03 additional climate change policies. Conversely, growth in a city’s per capita income is negatively associated with climate change policy adoption over the 1900–2000 period, with each percentage increase in local per capita income associated with 0.06 fewer climate change policies. The value of FEMA disaster assistance, included as a measure of the frequency and/or severity of extreme weather events in a given location, is positively linked to the presence of local climate change policies, with each billion dollars of aid connected to 1.06 additional activities. Changes in the PDSI’s measure of drought also show a positive relationship with local climate change policies, with an additional one-unit change (which, on the PDSI scale, indicates a location becoming relatively wetter over 1990–2000 relative to its historical average) associated with 1.31 additional policies. As well, our results indicate that states that have a Climate Action Plan in place are also associated with a higher number of local climate change policies, with such locations exhibiting an average of 3.55 more policies than communities in states without Action Plans. Finally, locations operating under a city manager system are associated with 2.79 more policies than places which employ other local government structures.

Turning to the model’s variables associated with government attitudes, mayoral agreement that cities and states should act on climate change if federal actors do not is positively correlated with policy implementation, with an additional level of agreement on the 5-point Likert-type scale linked to an average of 1.40 additional adaptation or mitigation policies. However, locations where mayors agree with the sentiment that state and local governments should adapt to climate change as it occurs are linked to fewer climate change actions, with each point of agreement associated with 0.74 fewer such policies in place. Finally, examining the group of variables related to community atmosphere, the 1988–2000 change in the percent of the county voting democratic in the Presidential election is significantly linked to climate change policy adoption, with each additional percentage point of growth in Democratic support linked to 0.16 more climate change policies, all else equal. As well, communities which are characterized by their mayor as displaying confusion about climate change exhibit 0.99 additional climate change-related policies. Taken together, our model thus suggests that locations which exhibit adoption of a greater number of local climate change-related policies are those which have grown in terms of population, and which have become more Democratic, over the 1990s; which operate under a city manager form of local government; where there are state-level Climate Action Plans in place, and where the current mayor also views the local government role as one of action in the absence of federal climate change policy, rather than one of passivity and reaction to climate change as it happens; where the local environment is characterized by confusion surrounding climate change; and which have gotten wetter and received greater amounts of disaster assistance. Such locations might therefore be described as growing communities, politically liberal in both their citizenry and leadership at the local and state levels, where the change in, and severity of, local weather conditions present firsthand evidence of costs which could be attributed to climate change.

Model II—Mitigation Policies

Investigating which local factors are linked specifically with mitigation policies, however, a somewhat different set of results emerges. The second column of Table 5 presents these results from regression Model II, which formulates the dependent variable as the total number of mitigation policies present in a community. Similar to Model I, change in per capita income, the value of disaster assistance, change in local drought conditions, the presence of a state Climate Action Plan, and the use of a city manager government structure are again significant, displaying the same signs as in the previous model. Likewise, in terms of Government Attitudes, mayors which more strongly support the idea of state and local action in the absence of federal climate change policy and which reject passive state and local adaptation to climate change as it happens again exhibit a higher number of mitigation policies in place.

However, when considering mitigation policies alone, four important differences emerge when compared with Model I’s results. First, locations where mayors have signed the Mayors’ Compact on Climate Change display roughly 1.19 more mitigation policies than locations which are not signatories to that agreement. Second, cities where mayors characterize local sentiment as being more apathetic toward climate change are associated with 0.30 fewer mitigation policies per unit increase in that factor score, while local attitudes of confusion (Factor 3), which were significantly associated with all climate change policies in Model I, are here not linked to the total number of mitigation policies present. Third, the change in the level of local Democratic support is also statistically insignificant in this model.

Interestingly, a fourth key difference between the first two models is that cities in states which have voluntary GHG reporting systems exhibit fewer mitigation policies than cities in states which do have such reporting in place. As noted above, mitigation activities may be more difficult to frame in terms of nonclimate change or GHG-related cobenefits to voters and taxpayers relative to adaptation activities, and that some mitigation activities (e.g., explicit consideration of GHG impacts in zoning, local GHG inventories, GHG reduction plans, and a dedicated budget for climate change policies) may be difficult to justify without reference to climate change. While the negative association between the presence of state-level GHG reporting systems and mitigation policy adoption may seem counterintuitive on its face, it is consistent with our expectation that, in light of the challenges to motivate citizens and stakeholders to support local climate change policy adoption, local government actors may be using state-level GHG reporting systems as a substitute for local action. Local politicians and administrators may thus highlight their state’s GHG reporting systems to convince their constituents (or themselves) that action is being taken on the issue of climate change, while conserving available resources and their own political capital to pursue other, potentially less contentious, priorities with more clearly defined cobenefits for their communities. These differential effects, in turn, provide further illustration of our central point that mitigation and adaptation policies operate differently and, in this case, cannot automatically be assumed to produce similar impacts.

Model III—Adaptation Policies

Model III examines factors associated with local government having adaptation policies in place. Here, while there are again a number of key similarities with respect to the results of Models I and II, there are also a number of important differences. Six factors are associated with significantly higher numbers of local adaptation policies—percent change in population, percent change in Democratic vote, change in annual drought, a city manager form of local government, local mayoral support for the idea that cities and states should act in the absence of federal climate change leadership, and higher levels of community skepticism and confusion surrounding climate change—all of which were also found to be statistically significant in Model I. However, when compared with the factors found to be significant in Model II, only three independent variables—average annual drought change, the presence of a city manager, and local mayoral support for action on climate change—are common across both.

Taken together, cities with higher numbers of adaptation activities appear to be largely unmotivated by local citizenry for the specific idea and importance of climate change as a concept, responding more strongly to environmental cues in terms of their own population growth, general shifts in political preferences, and changes in the physical environment, at least in terms of decreases in local drought conditions. As well, when the positive relationship between mayoral support for policy action is combined with the positive relationship observed between levels of climate change confusion in the community and the number of adaptation activities undertaken, it is plausible that such skeptical communities are ones which are generally reactive and uncoordinated with respect to climate change policy. While policymakers in these locations may themselves feel, either due to their own observation or available professional expertise (especially in locations with city managers) that something is going on in their local environments, absent a clear demand from citizens, that the best course of action is the adoption of policies which carry a high degree of cobenefits outside of GHG or climate change-related impacts, such as adaptation measures, and which can be framed as positive measures without having to invoke, or ask citizens to “buy into,” climate change as a concept. Perhaps due to the significant cobenefits produced by adaptation policies, these locations may not even think of such activities as climate change-related, adopting them in a piecemeal fashion due largely to the other nonclimate change-related benefits they deliver, rather than seeing them as a unified climate change-related policy program. These results suggest locations which pursue a higher number of adaptation policies may differ in important ways from those which pursue mitigation policies, and that separate examination of the factors associated with the adoption of mitigation and adaptation activities provides an important specification of Model I’s results which would be missed were the total number of climate change policies adopted examined alone.

Conclusion

The results presented above suggest that, in the Great Plains, there are important differences both between places which have climate change policies in place and those that do not, and important differences in the factors which lead cities to adopt more mitigation policies versus conditions which lead communities to enact adaptation policies. Overall, three factors—change in annual average drought levels from 1990 to 2000, the presence of a mayor supportive of the idea that state and local governments should take action on climate change in the absence of federal leadership, and use of the city manager form of local government—are common across all three of our models, with each being positively associated with a community’s number of climate policies overall, as well as an increased number of mitigation and adaptation policies.

However, beyond these three common factors, our findings indicate that the influences associated with the adoption of mitigation and adaptation policies are different. Higher counts of adaptation policies appear to be linked to communities with higher levels of ideological confusion about climate change, suggesting that these communities, and especially their political leaders, may be recognizing the need for policy action due to general changes in their physical, political, and climatological environments, but absent a clear indication (from scientific evidence or from their constituents) that these shifts are climate change-related, may be “hedging their bets” by focusing on policies that also carry with them strong cobenefits, especially adaptation policies, aside from their potential impacts on climate change or GHG levels.

Locations with higher numbers of mitigation activities are similar in some respects to this group, yet different in others. The mayors of cities with higher numbers of mitigation activities resemble those in locations with more climate change policies in general in terms of their support of climate change action, but these locations also engage in formal commitments to climate change action, such as participation in the USMCPA, and generally represent a public which is less apathetic toward the idea of climate change in general. Higher levels of mitigation policy thus seem more closely associated with places where local leaders have “bought in” to the concept of climate change and response to the issue, and where there are lower levels of community apathy, suggesting some degree of public desire for action.

The contributions of this article are threefold. First, we provide evidence that change plays an important role in explaining local variation in climate change policy adoption. Beyond the present-period influences which much of the previous literature has focused on, our results suggest that change over time in several factors, including population growth, change in per capita income, political identification, and drought conditions in particular, also matters. In the case of population in particular, change was found to be an important explanatory factor of climate change policy making for two of our models, whereas current population did not play a significant role. The addition of these important variables increases the explanatory power of our models and identifies the idea of change in local conditions and indicators over time as an area which should be explored further in local climate change policy research.

Second, disaggregation matters. The results of Models II and III support our expectations that, as different types of policy responses to climate change, the adoption of policies of each type is likely to be linked to different factors in the local policy environment, government, and/or community attitudes. While, as one might expect, there are some factors which are linked to climate change policy adoption of both the mitigation and adaptation varieties, the differences in factors associated with each type of policy suggest that different types of Great Plains cities may be engaging in different types of climate change policy making. Moreover, the local conditions under which mitigation policy making is allowed, if not encouraged, may differ in important ways from those circumstances which lead locations to adopt adaptation policies. These key differences with respect to mitigation and adaptation policy would have been missed were the two types of policies inappropriately aggregated and the results of Model I relied on alone.

Third, these differences between the factors which lead to higher levels of mitigation activity versus those that lead to higher levels of adaptation activity can begin to provide an answer to the curious puzzle of the variation in climate change policy making which we see in the Great Plains. Our results help suggest conditions that differentiate those communities which have few, if any, climate change-related policies from those which have adopted several such measures, as well as the factors which motivate Great Plains cities to engage in mitigation and adaptation planning, specifically. Moreover, in expanding our sample beyond the larger 50,000 to 100,000 citizen communities examined by previous work (e.g., Krause 2011; Sharp et al. 2011), our research suggests that governmental climate change action is being actively undertaken in places where it might be least expected, including medium to small communities in an area of the country where the lack of clear, climate change-related weather events, such as hurricanes and coastal flooding, and a prevailing conservatism, both socially and with respect to the idea of climate change specifically, might be expected to present a barrier to such policy adoption.

Looking back, one might reasonably ask why we have chosen to focus so much time and energy studying these actors in these places. Citizens and policymakers in the central United States are well known for their skepticism regarding climate change, and the carbon footprint of most of these communities tends to be very small. Yet for much of the last two decades, the discussion of adaptation or adaptation planning has been discouraged. It was considered by many to indicate a lack of commitment to aggressive mitigation policy.

Now, with the expiration of the Kyoto Protocol and the emergence of the bottom-up approach to climate protection, adaptation has reentered the conversation, and the policies of these communities have taken on a new relevance. Adaptation activities alone will not avert the impacts of climate change, but understanding might. Policymakers looking to craft effective climate policy incentives can begin by understanding the motivations behind mitigation and adaptation, and by recognizing that communities are not static entities. Communities change over time and with those changes come opportunities for policies that were previously untenable.

Footnotes

Appendix

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Andy Hultquist is an associate professor in the Department of Political Science and Public Administration at the University of North Dakota. His research questions focus on questions of public policy evaluation, pubic finance, urban and regional economic development, and public-sector contracting. His most recent publications include works on determinants of local climate change policy adoption and evaluating the impact of local economic development incentives on employment and wages.

Robert S. Wood is an associate professor in the Department of Political Science and Public Administration at the University of North Dakota. His research interests include public policy processes, bureaucratic behavior, and regulatory policy. His most recent publications have focused on understanding local government responses to climate change.

Rebecca J. Romsdahl is an associate professor in the Department of Earth System Science and Policy at the University of North Dakota. Her research examines issues at the intersection of environmental science with communication in the policy process. Most recently, she is examining how climate change is framed in local government policy making.

References

Adger

W. Neil

Arnell

Nigel W.

Tompkins

Emma L.

2005. “Successful Adaptation to Climate Change Across Scales.” Global Environmental Change 15 (2): 77–86.

Anguelovski

Isabelle

Carmin

JoAnn

. 2011. “Something Borrowed, Everything New: Innovation and Institutionalization in Urban Climate Governance.” Current Opinion in Environmental Sustainability 3 (3): 169–75.

Baker

Ingrid

Peterson

Ann

Brown

Greg

McAlpine

Clive

. 2012. “Local Government Response to the Impacts of Climate Change: An Evaluation of Local Climate Adaptation Plans.” Landscape and Urban Planning 107 (2): 127–36.

Berner

Maureen M.

Amos

Justin M.

Morse

Ricardo S.

2011. “What Constitutes Effective Citizen Participation in Local Government? Views from City Stakeholders.” Public Administration Quarterly 35:128–63.

Bodansky

Daniel

. 2011. “A Tale of Two Architectures: The Once and Future U.N. Climate Change Regime.” Ariz. St. L.J. 43: 697.

Brody

Samuel D.

Zahran

Sammy

Vedlitz

Arnold

Grover

Himanshu

. 2008. “Examining the Relationship Between Physical Vulnerability and Public Perceptions of Global Climate Change in the United States.” Environment & Behavior 40 (1): 72–95.

Bulkeley

Harriet

. 2010. “Cities and the Governing of Climate Change.” Annual Review of Environment and Resources 35:229–53.

Bulkeley

Harriet

Betsill

Michelle Merrill

. 2013. “Revisiting the Urban Politics of Climate Change.” Environmental Politics 22 (1): 136–54.

Bulkeley

Harriet

Schroder

Heike

Janda

Katy

Zhou

Jimin

Armstrong

Andrea

Chu

Shu Yi

Ghosh

Shibani

. 2011. “The Role of Institutions, Governance, and Urban Planning for Mitigation and Adaptation.” In Cities and Climate Change: Responding to an Urgent Agenda. Vol. 62696, edited by Hoornweg.

Daniel

Washington, DC: World Bank. 125-59.

10.

Byrne

John

Hughes

Kristin

Rickerson

Wilson

Kurdgelashvili

Lado

. 2007. “American Policy Conflict in the Greenhouse: Divergent Trends in Federal, Regional, State, and Local Green Energy and Climate Change Policy.” Energy Policy 35 (9): 4555–73.

11.

Castán Broto

Vanessa

Bulkeley

Harriet

. 2013. “A Survey of Urban Climate Change Experiments in 100 Cities.” Global Environmental Change 23 (1): 92–102.

12.

Feiock

Richard C.

Bae

Jungah

. 2011. “Politics, Institutions and Entrepreneurship: City Decisions Leading to Inventoried GHG Emissions.” Carbon Management 2 (4): 443–53.

13.

Feiock

Richard C.

Coutts

Christopher

. 2013. “Guest Editors’ Introduction: Governing the Sustainable City.” Cityscape: A Journal of Policy Development and Research 15 (1): 1-7.

14.

Florida

Richard L.

2012. The Rise of the Creative Class: Revisited. New York: Basic Books.

15.

Füssel

H.-M.

2007. “Adaptation Planning for Climate Change: Concepts, Assessment Approaches, and Key Lessons.” Sustainability Science 2 (2): 265–75.

16.

Hanna

Kevin S.

2005. “Planning for Sustainability: Experiences in Two Contrasting Communities.” Journal of the American Planning Association 71 (1): 27–40.

17.

Alfred Tat-Kei

. 2006. “Accounting for the Value of Performance Measurement from the Perspective of Midwestern Mayors.” Journal of Public Administration Research and Theory 16 (2): 217–37.

18.

Hoffmann

Matthew J.

2011. Climate Governance at the Crossroads: Experimenting with a Global Response After Kyoto. Oxford: Oxford Univ. Press.

19.

Hulme

Mike

. 2009. Why We Disagree About Climate Change: Understanding Controversy, Inaction and Opportunity. Cambridge: Cambridge Univ. Press.

20.

Hunt

Alistair

Watkiss

Paul

. 2011. “Climate Change Impacts and Adaptation in Cities: A Review of the Literature.” Climatic Change 104 (1): 13–49.

21.

Intergovernmental Panel on Climate Change. 2007. “Climate Change 2007: Synthesis Report of the IPCC Fourth Assessment Report Summary for Policy Makers.” Cambridge: Cambridge University Press.

22.

Kingdon

John W.

1984. Agendas, Alternatives, and Public Policies. New York: Longman.

23.

Kousky

Carolyn

Schneider

Stephen H.

2003. “Global Climate Policy: Will Cities Lead the Way?” Climate Policy 3 (4): 359–72.

24.

Krause

Rachel M.

2011a. “An Assessment of the Greenhouse Gas Reducing Activities Being Implemented in US Cities.” Local Environment 16 (2): 193–211.

25.

Krause

Rachel M.

2011b. “Symbolic or Substantive Policy? Measuring the Extent of Local Commitment to Climate Protection.” Environment and Planning C: Government and Policy 29 (1): 46–62.

26.

Krause

Rachel M.

2012a. “An Assessment of the Impact that Participation in Local Climate Networks Has on Cities’ Implementation of Climate, Energy, and Transportation Policies.” Review of Policy Research 29 (5): 585–604.

27.

Krause

Rachel M.

2012b. “Political Decision-Making and the Local Provision of Public Goods: The Case of Municipal Climate Protection in the US.” Urban Studies 49 (11): 2399–2417.

28.

Lee

Taedong

Koski

Chris

. 2012. “Building Green: Local Political Leadership Addressing Climate Change.” Review of Policy Research 29 (5): 605–24.

29.

Liu

Xinsheng

Lindquist

Eric

Vedlitz

Arnold

Vincent

Kenneth

. 2010. “Understanding Local Policymaking: Policy Elites’ Perceptions of Local Agenda Setting and Alternative Policy Selection.” Policy Studies Journal 38 (1): 69–91.

30.

Lutsey

Nicholas

Sperling

Daniel

. 2008. “America’s Bottom-Up Climate Change Mitigation Policy.” Energy Policy 36 (2): 673–85.

31.

Mayhew

David R.

1974. Congress: The Electoral Connection. New Haven: Yale Univ. Press.

32.

Measham

Thomas G.

Preston

Benjamin L.

Smith

Timothy F.

Brooke

Cassandra

Gorddard

Russell

Withycombe

Geoff

Morrison

Craig

. 2011. “Adapting to Climate Change Through Local Municipal Planning: Barriers and Challenges.” Mitigation and Adaptation Strategies for Global Change 16 (8): 889–909.

33.

Millard-Ball

Adam

. 2012. “Do City Climate Plans Reduce Emissions?” Journal of Urban Economics 71 (3): 289–311.

34.

Moser

Susanne C.

Ekstrom

Julia A.

2010. “A Framework to Diagnose Barriers to Climate Change Adaptation.” Proceedings of the National Academy of Sciences 107 (51): 22026–31.

35.

Pielke

Roger

Prins

Gwyn

Rayner

Steve

Sarewitz

Daniel

. 2007. “Climate Change 2007: Lifting the Taboo on Adaptation.” Nature 445 (7128): 597–98.

36.

Portney

Kent E.

2003. Taking Sustainable Cities Seriously: Economic Development, the Environment, and Quality of Life in American Cities. Cambridge, MA: MIT Press.

37.

Preston

Benjamin L.

Westaway

Richard M.

Yuen

Emma J.

2011. “Climate Adaptation Planning in Practice: An Evaluation of Adaptation Plans from Three Developed Nations.” Mitigation and Adaptation Strategies for Global Change 16 (4): 407–38.

38.

Rabe

Barry G.

2007. “Beyond Kyoto: Climate Change Policy in Multilevel Governance Systems.” Governance 20 (3): 423–44.

39.

Rayner

Stebe

. 2010. “How to Eat an Elephant: A Bottom-Up Approach to Climate Policy.” Climate Policy 10 (6): 615–21.

40.

Sharp

Elaine B.

Daley

Dorothy M.

Lynch

Michael S.

2011. “Understanding Local Adoption and Implementation of Climate Change Mitigation Policy.” Urban Affairs Review 47 (3): 433–57.

41.

Sippel

Maike

Jenssen

Till

. 2010. “What About Local Climate Governance? A Review of Promise and Problems.” Munich Personal RePEc Archive. Accessed September 19, 2015. http://mpra.ub.uni-muenchen.de/20987/1/MPRA_paper_20987.pdf.

42.

Tiebout

Charles M.

1956. “A Pure Theory of Local Expenditures.” The Journal of Political Economy: 416-424.

43.

Tompkins

Emma L.

Neil Adger

Boyd

Emily

Nicholson-Cole

Sophie

Weatherhead

Keith

Arnell

Nigel

. 2010. “Observed Adaptation to Climate Change: UK Evidence of Transition to a Well-Adapting Society.” Global Environmental Change 20 (4): 627–35.

44.

Van Asselt

Harro

Zelli

Fariborz

. 2014. “Connect the Dots: Managing the Fragmentation of Global Climate Governance.” Environmental Economics and Policy Studies 16 (2): 137–55.

45.

Victor

David G.

2001. The Collapse of the Kyoto Protocol. Princeton: Princeton Univ. Press.

46.

Vogel

Brennan

Henstra

Daniel

. 2015. “Studying Local Climate Adaptation: A Heuristic Research Framework for Comparative Policy Analysis.” Global Environmental Change 31:110–20.

47.

Warden

Toby

. 2011. “Viral Governance and Mixed Motivations: How and Why U.S. Cities Engaged on the Climate Change Issue, 2005–2007.” In Cities and Climate Change, Vol. 62696, edited by Hoornweg.

Daniel

Washington, DC: World Bank. 161–74.

48.

Wood

Robert S.

Hultquist

Andy

Romsdahl

Rebecca J.

2014. “An Examination of Local Climate Change Policies in the Great Plains.” Review of Policy Research 31 (6): 529–54.

49.

Zahran

Sammy

Grover

Himanshu

Brody

Samuel D.

Vedlitz

Arnold

. 2008. “Risk, Stress, and Capacity: Explaining Metropolitan Commitment to Climate Protection.” Urban Affairs Review 43 (4): 447–74.