392348 392348Sharp et al.Urban Affairs Review © The Author(s) 2011

UAR47310.1177/1078087410

Reprints and permission: sagepub.com/journalsPermissions.nav

Understanding Local Adoption and Implementation of Climate Change Mitigation Policy

Urban Affairs Review 47(3) 433­–457 © The Author(s) 2011 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/1078087410392348 http://uar.sagepub.com

Elaine B. Sharp1, Dorothy M. Daley1, and Michael S. Lynch1

Abstract Increasingly, local governments are crafting policy to tackle climate change. This article examines why cities develop and implement climate change programs. The authors consider the impact of interest group pressure, political institutions, and problem severity on a city’s decision to develop and implement climate protection programs. Their results suggest that organized interests influence both adoption and implementation of climate mitigation programs. This effect, however, is contingent on political institutions. In general, organized interests are more effective in mayoral as opposed to city manager forms of governments. Interestingly, while financially strapped cities may adopt climate mitigation programs to advance cobenefits or cost savings, fiscal stress also impedes program implementation. Keywords climate change, urban sustainability, global environmental governance, local policy implementation

1

University of Kansas, Lawrence, KS, USA

Corresponding Author: Dorothy M. Daley, University of Kansas, Department of Political Science and Environmental Studies Program, 1541 Lilac Lane, Lawrence, KS 66044 Email: [email protected]

434

Urban Affairs Review 47(3)

Increasingly, local governments are developing innovative policies and programs to address global climate change. On one hand, it is surprising that city governments invest time and resources in addressing a “transboundary” environmental problem. While one city may reduce emissions, if nearby localities do not, then overall greenhouse gas (GHG) emissions in a region may not change substantially (Zahran et al. 2008). Moreover, in the United States, limited federal support for climate change policy presents additional hurdles for subnational governments (Rabe 2004, 2007). Despite these challenges, many cities in the United States are actively pursuing strategies to reduce GHG emissions. Policy scholars who focus on the United States remain curious about the motivation for subnational action on transboundary issues in the absence of federal incentives or mandates (Rabe 2004, 2007; Zahran et al. 2008). However, scholars who focus on global environmental governance have been examining subnational climate change policy for more than a decade. This research suggests not only that cities in multiple national settings are actively pursuing climate protection policies (Betsill and Bulkeley 2007; Toly 2008) but also that cities are pivotal actors needed to create multilevel governance structures necessary to combat climate change (Koehn 2008). Some research emphasizes that the cobenefits from enacting climate protection policies are far from trivial (Bulkeley and Betsill 2003)—an argument relatively understudied in an American context (though for an important exception, see Portney 2003, 2009). While a city may not directly benefit from emission reductions if its neighbors do not cooperate, nevertheless climate protection policies may provide economic and other tangible benefits that drive decision making. The global environmental governance literature emphasizes the potential significance of regional, national, and transnational networks that cities can use to pursue climate change protection (Bulkeley and Betsill 2003, 26–27). For U.S. cities, there is considerable evidence of environmental organizations initiating campaigns that draw numerous cities into such networks. In the early 1990s, the International Council on Local Environmental Initiatives (ICLEI) began providing technical assistance to local governments that pay membership fees and pledge to reduce GHG emissions—over 600 cities worldwide have joined ICLEI. In the United States, ICLEI’s membership fees are based on a sliding scale using population to identify fee structure, with larger cities paying higher dues. In 2005, the Sierra Club initiated its Cool Cities campaign, and in the same year the U.S. Conference of Mayors launched a Climate Protection Agreement through which participating cities pledge to meet the goals stated in the Kyoto protocol and to lobby their states and the federal government to act on climate protection. More than 1,000 mayors

Sharp et al.

435

have signed this agreement (see http://www.usmayors.org/climateprotection/ revised/ for the complete list of signatories). This article focuses on ICLEI membership in the United States. Why do some cities join ICLEI’s climate mitigation program while others do not? And in addition, why are some cities more likely to successfully follow through on their commitment to climate protection while other cities struggle with policy implementation.1 ICLEI is an important network to examine because it was the first of these city government-focused sustainability campaigns affecting U.S. cities. And importantly, ICLEI provides a means for differentiating cities that actually develop and implement policies and programs to achieve GHG emission reductions from cities whose involvement is largely symbolic policy. ICLEI members receive technical assistance, most particularly in the form of access to software that can be used to produce a GHG emissions inventory. Beyond this first step, ICLEI’s program sets forth and tracks member cities along four additional “milestones” to be achieved, which include (1) setting a specific GHG reduction target, (2) developing plans designed to achieve that reduction, (3) implementing plans, and (4) monitoring results (ICLEI Global 2008). Of particular interest to us is the substantial variation in U.S. cities’ participation in ICLEI’s Cities for Climate Protection (CCP) program. While an impressive number of U.S. cities of all sizes have joined—545 at latest count (ICLEI Global 2008)—there are nevertheless hundreds more that have not. And among those that have joined ICLEI, some have yet to finalize the GHG inventory that is ICLEI’s first milestone of accomplishment while others, having implemented GHG reduction policies and programs and tracked the impact of those programs on emissions, have achieved all five milestones.2 Among U.S. urban policy scholars, multivariate research on city involvement with sustainability policy typically examines policy change or program adoption (Ramirez de la Cruz 2009; Zahran et al. 2008; Lubell, Feiock, and Handy 2009). However, adoption does not necessarily mean implementation; and a count of city programs or policies does not necessarily mean that each and every one is being fully implemented. Similarly, the policy choice to join ICLEI’s Climate Change Program may not necessarily translate into implementation of any new programs or policies involving GHG reduction. As Portney (2009, 228) notes, most of the cities that have joined ICLEI have not in fact followed up with change in their local policies, programs, and practices. Similarly, Betsill and Bulkeley (2007, 451) point to the “rhetoric or reality” issue—that is, the enduring gap between discussions about “the relevance and enduring importance of local action for climate protection and the reality on the ground.”

436

Urban Affairs Review 47(3)

Hence we pursue a twofold investigation that first considers why some cities have joined ICLEI’s program and others have not. Second, we consider variation in implementation progress—that is, variation in the number of implementation milestones that ICLEI member cities have achieved. We theorize the implementation context of local climate change programs is similar but not identical to policy adoption. In particular, we theorize that the decision to join ICLEI may be heavily driven by supportive interest groups and the perception of cobenefits; and the decision to join may be a relatively noncontroversial step that does not activate oppositional interests. By contrast we theorize that efforts to actually devise and implement a climate change protection plan are likely to activate latent oppositional interests as well as supportive interests and to require time and fiscal capacities that may not have been as important in shaping the decision to join ICLEI. Our theoretical framing and specific hypotheses are laid out more fully in the following section.

Theorizing About City Involvement in Climate Change Policy What might account for the substantial variation in U.S. city governments’ propensity to join ICLEI and the degree to which they implement and assess climate change policies once in the ICLEI program? To tackle these questions, our research builds upon existing research on city government involvement with GHG reduction policy in particular and sustainability policy more broadly, including work on urban growth policy. The initial springboard for our analysis is the Zahran et al. (2008) study of the extent of ICLEI membership in metropolitan areas. Despite the problematic nature of the metro area unit of analysis (a point we return to below), Zahran et al. offer a relatively comprehensive, multivariate approach that is also of special interest to us because of its focus on ICLEI involvement. Zahran et al. theorize that ICLEI involvement, from this metropolitan perspective, is a function of three explanatory factors. “Climate change risk,” a suite of four factors such as coastal proximity and ecosystem sensitivity, is theorized to enhance ICLEI membership; the immediacy of negative climate change consequences for “at-risk” jurisdictions provides selective incentives to help overcome the inherent collective action dilemma of local-level climate change mitigation efforts. Second, a set of four variables measuring “climate change stress” (high levels of carbon-based employment and solo commuting combined with low density and low levels of solar energy use) is theorized to deter ICLEI membership; the higher costs of pursuing climate

Sharp et al.

437

change policy in places where economic, transportation, and energy-use practices are so implicated in GHG inhibits climate protection initiatives. Third, “civic capacity” is theorized to enhance ICLEI membership. Civic capacity is conceptualized as including both the environmental group presence and involvement in environmental causes that others (Portney et al. 2009) identify as important for explaining city involvement in sustainability initiatives. It also includes income and education, treated by Zahran et al. as measures of human capital that contributes to civic capacity. Zahran et al. find no evidence that climate change risk influences the prevalence of ICLEI membership in metro areas, so we do not pursue that line of theorizing; however, their indices of climate change stress and civic capacity are significant predictors. Research on urban sustainability policy more broadly shows some theoretical overlap with Zahran et al. (2008), especially with regard to the importance of civic capacity (Portney and Berry 2010), but it also reveals a theoretical approach focusing on political institutions (Lubell, Feiock, and Ramirez de la Cruz 2005; Lubell, Feiock, and Handy 2009; Ramirez de la Cruz 2009) that is omitted by Zahran et al. Still other research (Bulkeley and Betsill 2003) emphasizes the importance of cobenefits, including cost savings, which suggests another important hypothesis involving fiscal stress. Drawing upon all this literature, we lay out three major sets of hypotheses about local government uptake of sustainability policy—(1) hypotheses relating to organized interests and civic capacity, (2) hypotheses relating to political institutions, and (3) hypotheses relating to problem definition.

Organized Interests and Civic Capacity Hypotheses in this category stipulate that the prevalence of various organized interests or demographic groupings will influence the uptake of sustainability policies. The presence of certain interests or demographic groups provides a form of civic capacity that may promote climate protection policy, while the presence of other groups may create a barrier to such policy. Consistent with the group-centered aspects of Zahran et al.’s (2008) “civic capacity” explanation as well as the work of Portney et al. (2009), a key hypothesis in this category is that, other things being equal, cities where environmental groups are more prevalent will be more likely to have strong sustainability policies. Others have offered a parallel hypothesis for homeowners’ associations because their preferences for livability and protection from compromised use values translate into support for many smart growth-style sustainability policies (Ramirez de la Cruz 2009). In a related vein, Portney and Berry (2010)

438

Urban Affairs Review 47(3)

have suggested that heightened citizen participation, not only in homeowners’ associations but also across an array of forms of local participation, is a form of civic capacity that is linked to city government pursuit of sustainability policies. In some research, the presence or importance of “proenvironmental interests” or civic capacity is assessed via survey research questions (Ramirez de la Cruz 2009); other research in this vein uses socioeconomic status as a proxy for such proenvironmental interests (Lubell, Feiock, and Handy 2009; Lubell, Feiock, and Ramirez de la Cruz 2009). The latter provides another point of overlap with Zahran et al., who include indicators of income and education as proxies for civic capacity. This category of hypotheses also includes identification of organized interests expected to oppose environmental policies. For example, some research points to developers, who have been shown to oppose many, though not all, sustainability policies (Ramirez de la Cruz 2009; Lubell, Feiock, and Handy 2009). It remains unclear if developers would necessarily oppose local climate change initiatives that may not involve land development matters. In a different vein, manufacturing firms might be expected to resist policies intended to reduce GHG because of cost implications for their industrial sector. On the other hand, existing case study evidence reveals that industry actors sometimes support climate protection, either to save money or to take advantage of new markets.3 Consistent with this, other scholars (Portney et al. 2009) contend that there is a distinct need to begin to decompose “business groups” into a more fine-tuned categorization to accurately capture different types of organized interests and their impact on decision making. However, theoretically defining which organized interests will oppose local climate change policy remains an unsettled matter. Pending the development of a more finely tuned, theoretically based categorization that attracts some scholarly consensus, we hypothesize that the prevalence of organized interests in the manufacturing sector will tend to deter city involvement in ICLEI. As the next section explains, the influence of both supportive and oppositional interests is hypothesized to be mediated by political institutions. Before turning to that, it is important to acknowledge that interest group influence may vary depending upon whether joining ICLEI or making implementation progress in ICLEI’s CCP program is at issue. Supportive interests may be hypothesized to be facilitators of both the initiative to join ICLEI and the process working through ICLEI’s “milestones.” However, we expect that the decision to join ICLEI is likely to be less difficult and controversial than the activities that are involved in setting reduction targets, devising a plan to meet those targets, carrying out the plan, and measuring impact. Organized interests may not mobilize to oppose the decision to join ICLEI, but they are

Sharp et al.

439

likely to be activated when the city begins the more difficult process of implementation. Hence we hypothesize that the negative influence of oppositional interests will be particularly relevant to understanding implementation progress.

Political Institutions Hypotheses in this category point to a variety of institutional arrangements that are likely to shape the involvement of city governments in climate change policy networks. The city’s basic form of government is one of these political institutions. Research by Lubell, Feiock, Ramirez de la Cruz, and colleagues lays out a “political institutions” explanation stipulating how they expect such governing institutions to matter with respect to sustainability policy. Based upon earlier work on urban policy and political institutions (Feiock and Cable 1992; Clingermayer and Feiock 2001; Sharp 2002), the influence of organized interests should be mediated by the nature of the executive branch institution. Similarly, the influence of both supportive and oppositional groups should be dampened in cities that have the city manager form of government; the design of this institutional arrangement insulates governance from political pressures. By contrast, in the more politicized setting of mayor–council government, one should expect to see city policy reflecting pressures either from industrial interests wanting to minimize climate protection policies that would adversely affect them or from proenvironmental interests wanting to take up such policies (Ramirez de la Cruz 2009; Lubell, Feiock, and Ramirez de la Cruz 2009). Another institutional arrangement of potential importance for city government involvement in climate protection is the structure of the metropolitan area. Unlike Zahran et al. (2008), our unit of analysis is city government, not metropolitan areas. We agree with Portney’s (2003, 14) argument that city governments rather than metropolitan areas, ecosystems, or other regional areas are the relevant unit of analysis for the study of climate change policy because they are “legally defined entities that have the legitimacy and authority to address issues and problems within their borders.” Similarly, Bulkeley and Betsill (2003, 48–49) lay out a strong argument about the many capacities that city governments have to affect energy efficiency and conservation. However, city-level climate protection efforts are likely to be influenced by transboundary challenges. The benefits of climate protection may spill over to nearby cities, and more importantly, environmental benefits may be nonexistent or eroding if nearby cities fail to curb their own GHG emissions. Therefore, a city government nested within a highly fragmented metropolitan

440

Urban Affairs Review 47(3)

area (i.e., one that includes a large number of other governments) may face greater obstacles and disincentives than a city government in a metropolitan area that consists largely of a single core city with very few other local governments sharing the metropolitan space. Consistent with this, a substantial number of metro areas in Zahran et al.’s (2008) research that score the highest on the key dependent variable (percentage of population whose government is involved in ICLEI’s climate protection campaign) are metro areas including only a handful of local governments (e.g., Muncie, Indiana, Missoula, Montana, and Barnstable, Massachusetts). Surprisingly enough, neither Zahran et al.’s analysis nor other research on sustainability has taken metropolitan structure into account. We hypothesize that membership in ICLEI will be more likely for cities in less fragmented metropolitan areas; and we hypothesize that metropolitan fragmentation will be an even more important obstacle with respect to progress in implementing ICLEI milestones than it is for the initial decision to join ICLEI. For the United States, there is a third political institution that we theorize should be important—the nature of the state environmental policy regime within which the city government operates. Some state governments have been more aggressive than others in adopting policies that facilitate or even require local government action with respect to environmental policy. While we do not claim to have an exhaustive state-by-state inventory in this regard, it is important to at least acknowledge that California stands out with respect to progressive air quality policies. Hence we hypothesize that cities located in California will have a heightened propensity for membership and progress in ICLEI’s Climate Protection Program.

Problem Definition Hypotheses A third category of hypotheses explores the relationship between problem severity and public decision making. The growing research literature on city government sustainability policy in the United States theorizes that the uptake of sustainability policies is a function of the scope of the problem to which such policies are addressed (Ramirez de la Cruz 2009; Lubell, Feiock, and Ramirez de la Cruz 2009). Given the urban growth and land use focus of this line of work, those scholars tend to focus on various aspects of sprawl, such as low population density and population growth pressures, as the definition of the problem (Ramirez de la Cruz 2009, 222). The Florida-specific nature of some of the research yields an additional emphasis on phenomena such as the availability of coastal resources needing protection from growth pressures. Similarly, Zahran et al. (2008) hypothesize that “climate stress”—a

Sharp et al.

441

composite phenomenon including low density as well as high levels of carbon employment and solo commuting and low levels of solar energy use— shapes cities’ involvement in sustainability policy. There are important conceptual and empirical problems with this approach. For one thing, tests of these versions of the problem-definition hypothesis have generated distinctly mixed results, especially with respect to density (Ramirez de la Cruz 2009, 235; Lubell, Feiock, and Ramirez de la Cruz 2009, 658; Zahran et al. 2008). More importantly, there is disagreement or theoretical underdevelopment about how “problems” such as low density should be expected to affect city government action with respect to environmental policy. In some of the research, low density constitutes a need or problem that propels the city into sustainability policy (Ramirez de la Cruz 2009; Lubell, Feiock, and Ramirez de la Cruz 2009); by contrast, Zahran et al. (2008) theorize that low density, high “climate stress” places will be less likely to take climate protection action in the form of involvement with ICLEI because such areas face more substantial sunk costs that constitute a deterrent to such involvement. Bulkeley and Betsill (2003, 149) go even further in their critique of “new urbanism” approaches to climate protection. They note that “high-density development without explicit attention to energy use and production could potentially result in higher levels of greenhouse gas emissions.” Their argument suggests that it is too simplistic to assume that higher density cities have a less severe problem with respect to climate protection. More generally, Bulkeley and Betsill’s (2003) emphasis on problem framing implies that it may be inappropriate to conceptualize problem severity in terms of any “objective” conditions such as population density, growth pressures, coastal location, and the like. However, because so much of the existing U.S.-focused research on local governments and environmental policy does include attention to problem severity explanations and because density, a featured variable in that regard, has been shown in at least some of that research to be significantly associated with cities’ uptake of such policies, we include a test for the impact of density on both membership in ICLEI’s CCP program and progress in implementing its milestones. Inclusion of this hypothesized explanatory factor also helps to provide some parallel with the Zahran et al. (2008) analysis, which our investigation is intended to partially replicate and extend. Case study findings concerning the importance of cobenefits for cities’ involvement with ICLEI suggest a much more important problem definitionstyle hypothesis, however (Bulkeley and Betsill 2003; Kousky and Schneider 2003; Portney 2009). Portney notes that some city governments see sustainability policies as giving them an edge for economic development, and

442

Urban Affairs Review 47(3)

others may even see sustainability as a way of saving money. There is other evidence as well that participation in ICLEI can be linked to cost-saving motivations. Based on interviews with officials in 23 ICLEI member cities in the United States, Kousky and Schneider (2003, p. 359) find that the most frequently expressed motivation for joining ICLEI was the potential for cost savings. By retrofitting municipal buildings, replacing traffic light bulbs with LEDs, replacing conventional vehicles in the municipal fleet with hybrids, and even using methane recovered from landfills to power small projects, city governments are reaching for energy cost savings in municipal efforts; and ICLEI’s software, which enables cities to estimate emissions from municipal government operations separately from community-wide emissions, may make it even more attractive to reach for the low-hanging fruit of emissions control efforts that are more directly under their control. While the estimated cost savings from these kinds of programs may or may not be realized, they appear to be an important motivator for climate protection policy involvement via ICLEI membership (Kousky and Schneider 2003; also see Bulkeley and Betsill 2003, 53). Presumably the cost-savings motivation is linked to a city’s current level of fiscal stress. That is, the perceived cobenefits of cost savings in municipal operations are likely to be of special relevance for cities that are more financially strapped. Hence we hypothesize that fiscally stressed cities will be more likely to join ICLEI than will financially healthy cities. But the same fiscal stress that can propel cities into involvement with ICLEI’s campaign may be an obstacle to making progress once the city joins ICLEI. Bulkeley and Betsill’s (2003, 175) case studies of two U.S. and four British cities reveal that the cities that had the financial resources to undertake up-front investment in energy management and to reinvest financial savings in additional energy efficiencies or renewable energy efforts were able to make the most out of their ICLEI involvement. By contrast, where fiscal stress translated into a lack of availability of internal funding resources, as in Milwaukee, progress in the CCP program was limited. Consistent with this case study evidence, we hypothesize an inverse relationship between fiscal stress and progress in achieving ICLEI implementation milestones.

Research Method Our investigation is based upon a 50% random sample of all municipalities in the United States with a population of at least 100,000. Smaller cities were not included in the sampling frame both because of expected problems in data availability in the U.S. census and other aggregate data sources for such

Sharp et al.

443

cities and because climate protection policy, although not totally out of the reach of small cities, is presumably of a different caliber than in larger cities and potentially shaped by somewhat different factors than those theorized to be important for larger cities. The final sample included 122 cities, of which 47 cities are ICLEI members. Our measure of ICLEI membership—that is, membership in ICLEI’s Climate Protection Program—is from the ICLEI Web site, which provides a full listing of local government members. As of the summer of 2009, the Web site also presented information on the number of implementation milestones achieved. Our implementation measurement began with that listing. However, sources at ICLEI confirmed that the milestones awarded portion of the Web site had not been kept up to date. Hence we updated the information by examining city government Web sites supplemented by a search of local newspaper articles (using ICLEI and climate change as keywords). When there was either missing information or ambiguity from these sources, we contacted and queried relevant city officials to update the information. We score our cities in two ways: by ICLEI membership and by implementation milestones achieved. ICLEI member’s are coded 1 and nonmembers are coded 0. Implementation coding is different: Each of our cities is scored –1 if it has not even joined ICLEI, and for those that have joined, we score cities from 0 to 5 for the number of milestones achieved. For the portion of our analysis that focuses on accounting for variation in achieving ICLEI milestones, we include a count of the number of years a cities has been an ICLEI member, as determined from ICLEI Web site listing of the year that the city joined ICLEI. Our focus on ICLEI membership and progress in implementing milestones in ICLEI’s Climate Protection Program is in one sense narrower than some approaches to assessing city sustainability policy, most notably Portney’s (2003) 34-item Taking Sustainability Seriously Index, which not only counts policies, plans, and programs but also assesses each city’s measurement and reporting of progress on sustainability and the involvement of various stakeholders or governance elements in the overall sustainability program. Because 23 of the cities in our sample are included among the 44 cities for which Portney (2009, 238–39) presents scores on his sustainability index, we can assess the extent to which our implementation measure is or is not at odds with a more comprehensive index of city policies, programs, and activities with respect to sustainability. The correlation between our ICLEI-based implementation score and score on Portney’s index is a hefty .89. Table 1 outlines the independent variables used in this research. We rely upon two indicators to represent both civic capacity and the pressure from proenvironmental organized interests within a community. These are

444

Urban Affairs Review 47(3)

Table 1. Variable Descriptions Theoretical explanation Organized interest and civic capacity  

Variable Education

Description Percentage of adult population with a bachelor’s degree or higher in 2005.

Environmental nonprofits

Estimated percentage of environmental nonprofits within a city. We used the count of environmental nonprofits in a county and the city’s population presence within a county to estimate the likely number of environmental nonprofits within a city. Oppositional Manufacturing The number of manufacturing establishments organized in a city in 2002 relative to other professional interest establishments. Political Form of Dichotomous variable coded as 1 if the city has institutions government a mayor and city council form of government and 0 if the city has a city manager form of government.   Metropolitan Our measure of metropolitan fragmentation fragmentation is a count of the number of general purpose municipal governments and counties existing in a study city’s metropolitan area (metropolitan statistical area or primary metropolitan statistical area).   State context Dichotomous variable coded as 1 if a city is (California) located in California and 0 otherwise. Need or Population density Total city population in 2000 divided by land problem area in square miles. severity The number of manufacturing establishments   Manufacturing in a city in 2002 relative to other professional establishments.   Fiscal stress Total own-source revenue per capita, divided by median household income. Controls Time In modeling policy implementation, we include a measure noting how many years a city has held International Council on Local Environmental Initiatives membership.

(1) a measure of education in 2005 (percent of the adult population with a bachelor’s degree or higher) and (2) a measure of the prevalence of environmental nonprofit organizations within the city. Using information from

Sharp et al.

445

the National Center for Charitable Statistics, we obtained a count of environmental nonprofits by county in 2000. We estimated the number of environmental nonprofits in each study city by apportioning the count for the county to the city, based on city population as a percentage of county population. Our measure of oppositional group pressure is the number of manufacturing establishments in a city in 2002 as reported in the 2007 County and City Data Book, standardized by the number of establishments in the professional, scientific, and technical services categories. This is a relatively crude, monolithic measure. And while a more nuanced one would be preferable, the theoretical and empirical work needed for that is yet to be done. In research on state politics in the United States, manufacturing presence is frequently used as a proxy for industrial interests, representing the strength of an industrial sector that is conceptualized as having oppositional stakes with regard to environmental control (Ringquist 1993, 1995; Whitford 2009). Our measure of political institutions distinguishes city manager–council governed cities (coded 0) from mayor–council governments (coded 1) based on the coding in International City/County Management Association’s (ICMA’s) 2004 survey of economic development or, for cities not included in that data set, examination of city Web sites supplemented by calls to city clerks’ offices.4 We include population density (total city population in 2000 divided by city land area) primarily to replicate a key element of Zahran et al.’s (2008) “climate change stress” index. Our proxy measure of oppositional interests— that is, relative prevalence of manufacturing establishments—simultaneously serves as a second indicator of Zahran et al.’s climate change stress. Determining how to measure the fiscal health of cities presents a dizzying array of competing possibilities, ranging from simple indicators available in standard aggregate data sources—such as the property tax rate and assessed value per capita (Rubin and Rubin 1987) or debt as a proportion of own source revenue plus local property tax revenue per capita to control for exogenous effects of local tax structure (Donovan 1993)—to complex, multi-indicator approaches requiring original data collection from city governments (Hendrick 2004). The added value of the latter approach is either not demonstrated or argued to be primarily relevant for smaller suburban governments. Though there is some work suggesting that high per capita taxation is suggestive of strong fiscal health (Lubell, Feiock, and Handy 2009, 297), high tax rates are more logically viewed as a sign of fiscal stress, especially if tax rates are high relative to the “wealth” that the city has to draw upon. However, cities differ in the extent to which they rely upon property taxes, other taxes, and nontax sources of locally raised revenue such as user fees. Hence our measure of fiscal stress is a broader one: total own-source revenue per capita, divided by median household income.

446

Urban Affairs Review 47(3)

Our measure of metropolitan fragmentation, which conceptually has to do with the multiplicity of local governments within the same metropolitan area, is a count of the number of general purpose municipal governments and counties existing in a study city’s metropolitan area (metropolitan statistical area or primary metropolitan statistical area). Data for our fragmentation measure come from the appendix of the 2004 issue of County and City Extra (to identify component counties for relevant metropolitan statistical areas and consolidated metropolitan statistical areas) and the number of municipalities in each of these counties as shown in the Census Bureau’s 2002 Census of Governments, Volume 1, Government Organization, available online at http://www.census.gov/prod/2003pubs/gc021x1.pdf.

Results and Discussion Partly to maintain parsimony in our models and partly to avoid collinearity problems, we model mayoral and city manager cities separately. We also parse out the measures of organized interests: One model contains environmental nonprofits and the other education. Table 2 presents the resulting pair of models as applied to policy adoption—that is, ICLEI membership. Tables 3a and 3b presents the results for policy implementation.5

Joining ICLEI Considering first the question of what influences ICLEI membership, the results in Table 2 provide some support for one of the organized interests or civic capacity hypotheses as well as one of the political institutions hypotheses. Specifically, we find that for mayoral cities, the presence of larger numbers of environmental nonprofit organizations does enhance the likelihood that city government will join ICLEI; by contrast, in city manager cities there is no such evidence that ICLEI membership is driven by proenvironmental interests. This is highly consistent with the hypothesis that the decision to join a climate protection network like ICLEI is, like other city policy choices, at least partly a matter of responsiveness to supportive local interests, but only in cities where governmental is institutionally structured to be responsive to organized interests. However, when prevalence of interests favoring climate protection or civic capacity is measured with the relatively crude proxy of educational attainment, there are no significant results. Nor do we find evidence that the importance of manufacturing in the local economy diminishes the likelihood that a city will join ICLEI, either in city manager or mayoral cities. We would

447

Sharp et al. Table 2. Logistic Regression Analysis, International Council on Local Environmental Initiatives Membership Variable Mayoral cities   (Intercept)   Education   Environmental nonprofits   Population density   Manufacturing industry   Fiscal stress   Metro fragmentation   California city   N   Log likelihood   Pseudo R2 City manager cities   (Intercept)   Education   Environmental nonprofits   Population density   Manufacturing industry   Fiscal stress   Metro fragmentation   California city   N   Log likelihood   Pseudo R2

Coeff.

SE

Coeff.

−5.297 0.089 — −0.000 1.950 78.238 0.002 0.093 46 −24.759 .222

3.114* 0.064 — 0.000 4.101 29.709** 0.005 1.553

−2.268 — 0.035 −0.000 −0.200 61.017 −0.000 0.041 46 14.99 .235

−3.091 0.020 — 0.000 −0.431 46.973 −0.001 0.741 72 −40.070 .096

1.548* 0.026 — 0.000 2.677 30.275 0.001 0.673

−2.815 — −0.006 0.000 −1.305 71.026 0.001 0.972 72 −39.840 .101

SE   1.534 — 0.021† 0.000 3.211 28.381* 0.006 1.507         1.263* — 0.006 0.000 2.330 37.872† 0.007 0.677      

†

p < .10. *p < .05. **p < .01.

by no means go so far as to claim that the results decisively show city government to be impervious to organized interests that oppose climate protection policy. The null result may stem in part from the fact that our only measure of opposing interests is a relatively crude proxy. We agree with Portney et al.’s (2009) argument that much better conceptualization and measurement are needed to identify the types of business interests that tend to mobilize against environmental initiatives, but that work has yet to be done. More importantly, our null results with respect to oppositional interests are consistent with observations made about the importance of city-specific issue framing with respect to climate change policy (Betsill and Bulkeley 2007, 452; Bulkeley

448

Urban Affairs Review 47(3)

Table 3a. Tobit Regression Analysis, International Council on Local Environmental Initiatives (ICLEI) “Milestones” Implementation by Type of Government, Including Environmental Nonprofits Variable Mayoral cities   (Intercept)   Years ICLEI member   Environmental nonprofits   Manufacturing industry   Fiscal stress   Metro fragmentation   California city   Sigma   N   Log likelihood   Pseudo R2 City manager cities   (Intercept)   Years ICLEI member   Environmental nonprofits   Manufacturing industry   Fiscal stress   Metro fragmentation   California city   N   Log likelihood   Pseudo R2

Coeff.

SE  

3.060 0.141 0.014 −4.917 −36.660 0.000 2.732 1.209 22 −35.384 .215 0.685 0.252 0.020 −2.057 59.171 −0.020 0.244 22 −33.978 .211

0.910** 0.059* 0.009** 2.370* 15.355* 0.004 1.254* 0.182         1.615 0.069** 0.013 2.844 44.504 0.008** 0.642      

*p < .05. **p < .01.

and Betsill 2003). To the extent that city officials or other policy entrepreneurs define city involvement in a climate protection policy network in ways that highlight cobenefits such as economic development and that negate threatarousing preconceptions, interest groups that otherwise might have been hostile will not in fact mobilize. While form of government plays the hypothesized role in mediating the influence of environmental nonprofits, hypotheses relating to other institutional arrangements find no support in Table 2. Cities in California are no more likely to join ICLEI than cities in other states. And cities in highly fragmented metropolitan areas are no less likely to join ICLEI than are cities in less fragmented metro areas.

449

Sharp et al.

Table 3b. Tobit Regression Analysis, International Council on Local Environmental Initiatives (ICLEI) “Milestones” Implementation by Type of Government, Including Education Proxy Variable Mayoral cities   (Intercept)   Years ICLEI member   Education   Manufacturing industry   Fiscal stress   Metro fragmentation   California city   Sigma   N   Log likelihood   Pseudo R2 City manager cities   (Intercept)   Years ICLEI member   Education   Manufacturing industry   Fiscal stress   Metro fragmentation   California city   Sigma   N   Log likelihood   Pseudo R2

Coeff. −0.848 0.124 0.099 −1.558 −16.399 0.001 2.289 1.083 22 −32.969 .268 0.900 0.226 0.025 −1.902 65.667 −0.018 0.238 1.160 22 −34.489 .199

SE   1.538 0.053* 0.033** 2.560 9.007† 0.003 1.139† 0.163         0.750 0.075** 0.023 3.168 47.969 0.007* 0.659 0.175      

†

p < .10. *p < .05. **p < .01.

Apart from the prevalence of supportive organized interests as mediated by form of government, our only other significant result with respect to factors shaping ICLEI membership is fiscal stress. Specifically, we find that a higher level of fiscal stress is associated with an increased likelihood of ICLEI membership. This result is most robust for mayoral cities; but there is even some evidence of it for city manager cities, at least when the model also includes environmental nonprofits rather than education as the indicator of supportive organized interests. Though some previous literature suggests that financial stress can be a constraint on adoption of sustainability policies (Lubell, Feiock, and Handy 2009), research on involvement in climate

450

Urban Affairs Review 47(3)

protection networks such as ICLEI has emphasized the importance of cost savings as a cobenefit or motivation (Bulkeley and Betsill 2003; Kousky and Schneider 2003). For fiscally stressed cities, such cost savings would presumably be a much more important consideration than for fiscally healthy cities. In this regard, our result for a large sample of U.S. cities is consistent with the evidence from smaller-N case study research. By contrast, the other hypotheses in the “problem definition” category find no support in Table 2. Neither of the variables that were included to replicate Zahran et al.’s (2008) “climate stress” explanation (i.e., density and prevalence of manufacturing in the local economy) have an impact on city governments’ propensity to join ICLEI. Because prevalence of manufacturing is also our only proxy for oppositional interests, we keep that variable in subsequent models. However, the density measure is dropped from further analysis.

Implementing ICLEI’s Milestones When we turn to the question of what influences the implementation of ICLEI’s “milestones,” our results provide strong evidence that joining ICLEI and making implementation progress are two different phenomena, shaped by somewhat different sets of factors. For example, both Table 3a and Table 3b provide strong support for the hypothesis that progress in meeting ICLEI’s milestones (emissions inventory, reduction goals, develop a plan to meet goals, implement the plan, monitor and evaluate results) is more likely the longer that the city has been an ICLEI member.6 This result is apparent for both mayoral and city manager cities. At the simplest level, this finding may be interpreted as showing that going beyond an emissions inventory to set targets for reduction, create a plan to implement those goals, and then implement and monitor the results is typically a lengthy process that unfolds over the course of a relatively extended affiliation with ICLEI. Consistent with the arguments of Bulkeley and Betsill (2003), it is possible that some cities may have already been involved with climate protection efforts and then joined ICLEI or were targeted by ICLEI early in their CCP initiative; other cities that were less serious about climate protection or less networked with various climate protection efforts early on may have joined ICLEI more recently but not yet accomplished much implementation. Tables 3a and 3b also show support for the organized interests or civic capacity hypotheses and the related hypothesis that the influence of organized interests is mediated by the city’s form of government. The prevalence of

Sharp et al.

451

environmental nonprofits contributes to implementation success, but only in mayoral cities; the city manager form of government insulates city officials from such interests. The same pattern of results is apparent when education replaces prevalence of environmental nonprofits as the featured variable. However, in contrast to the decision to join ICLEI, there is some evidence that progress in implementation is also influenced by oppositional interests, at least in the model presented in Table 3a. Consistent with our hypothesis, that influence is negative—the more prevalent industrial interests are in a city, the less progress that the city makes in achieving ICLEI milestones. And once again the importance of political institutions is demonstrated by the fact that this negative influence is apparent in mayoral cities but not city manager cities. Tables 3a and 3b also reveal that fiscal stress plays a very different role in implementation than it does in adoption of climate change policy. We hypothesized that while fiscal stress may compel cities to join ICLEI to capture cost savings, implementation would falter in cities with higher levels of fiscal stress. This is precisely what the results show, for mayoral cities; but just as the professionalizing, depoliticizing features of city manager government insulate implementation progress from the influence of either supportive or oppositional interests, so is fiscal stress not a relevant factor in accounting for ICLEI milestone progress in city manager cities. Finally, Tables 3a and 3b show that the impacts of other institutional arrangements are contingent on form of government, something which we did not expect. Other things being equal, mayoral cities in California (but not city manager cities) have made more progress with ICLEI implementation than cities elsewhere; and city manager cities (but not mayoral cities) that are located in fragmented metropolitan areas have made less implementation progress than city manager cities in less fragmented metropolitan areas.

Conclusion A growing body of scholarship explores local government behavior with respect to sustainability initiatives and involvement in climate change programs and networks in the United States and elsewhere. This article focuses upon one of these climate change networks—ICLEI’s CCP—in part because of ICLEI’s early importance for U.S. cities and in part because ICLEI’s milestones allow for analysis of actual implementation of targets and plans for GHG reductions.

452

Urban Affairs Review 47(3)

In addition to hypotheses drawn from a diverse U.S.-based literature on cities and sustainability policy, our focus on ICLEI also allows us to systematically test some hypotheses closely linked to insights from comparative case analyses of U.S. and European cities. In particular, Bulkeley and Betsill’s (2003) analysis of ICLEI introduces a cobenefits perspective that reveals the importance of local framing of side benefits such as municipal cost savings that can be achieved from participation in a network devoted to climate change protection; but their analysis also reveals the importance of internal financial resources for realization of a city’s commitment to climate change protection. Consistent with those insights, we find that cities that are more fiscally strapped are more likely to sign on to ICLEI, especially if they are mayor-led cities. But among those same mayor-led cities, the budgetary constraints associated with fiscal stress get in the way of actually developing and carrying out plans that could achieve reductions in GHG. Our results are in part consistent with extant literature that emphasizes the importance of community participation, nonprofit organizations, and other aspects of civic capacity in enhancing city governments’ climate protection activities (Portney and Berry 2010; Portney et al. 2009; Portney 2003). We find that the prevalence of environmental nonprofits does shape city progress in achieving ICLEI milestones, but only in mayoral cities; environmental nonprofits appear to have no impact in city manager cities. The same pattern holds when education serves as a proxy for civic capacity. We would not go so far as to claim that this result shows the irrelevance of civic capacity in city manager cities. Education is a very crude proxy for civic capacity; and our environmental nonprofits indicator may reflect the “organized interests” aspect of this line of explanation more than the “civic capacity” aspect. Broad measures of civic engagement like those that Portney and Berry (2010) derive from the Social Capital Benchmark Community survey are not available for our 122-city sample. Such measures could yield evidence that civic capacity matters in city manager cities, even if the institutional arrangements of that form of government appear to insulate it from both the supportive pressure of environmental groups and the opposition of manufacturing interests. Much of the literature on cities and climate change highlights the importance of policy entrepreneurs in developing and implementing local climate change policies.7 We were not able to directly test for the impact of entrepreneurs. The fine-grained case study detail needed to identify entrepreneurs and their activities is not readily available in a study of 122 cities. However, when some of the existing case study evidence on entrepreneurs is juxtaposed with our results concerning fiscal stress and form of government, a

Sharp et al.

453

deeper interpretation of our results is possible—one that perhaps should be on the agenda for future testing. In mayoral cities, cobenefits seem to be a considerable incentive; mayoral cities experiencing higher levels of fiscal stress are systematically more likely to join ICLEI. However, those same fiscal constraints consistently limit progress in GHG reduction activity. Perhaps this pattern exists because of a propensity for mayors, rather than specialists within the local bureaucracy, to serve as the initial climate change policy entrepreneurs in mayoral settings. And as Bulkeley and Betsill’s (2003) case study of Milwaukee suggests, when a mayor is the policy entrepreneur, personal vision, political connections, and other aspects of politicized leadership may get in the way of institutionalizing the expertise and coordination needed for climate protection activity to progress even in the face of financial constraints. Clearly this will not be true of every mayoral city—the only other U.S. city in the Bulkeley and Betsill study—Denver— is mayor led but exhibited a quite different pattern. Our results across a much larger number of U.S. cities at least suggest that mayoral cities may tend to exhibit the problematic nature of entrepreneurship that was found in Milwaukee. In sum, when the focus is on a network such as ICLEI that draws local governments into climate protection action in part through the promise of cobenefits, explaining which cities do more and which do less involves a tantalizing mix of budgetary factors, governing institutions, organized interests, and longevity in the network. It is also true that ICLEI’s approach enables cities to pursue relatively narrow GHG reduction efforts that do not necessarily involve broad change in the private sector. To the extent that this is the case, a given level of implementation progress in the ICLEI program, like state progress in the No Child Left Behind Act, may mask considerable variation in the ambitiousness of the outcomes achieved. More research is needed to better understand policy implementation from this standpoint. Ideally, future research should differentiate local climate change policy commitments of narrow scope from those of more ambitious scope. Some preliminary research is beginning to address this need. Feiock and Francis (2010) have initiated a nationwide Web-based survey that will allow them to distinguish climate protection policies aimed only at local government activities versus community-wide climate protection programs. This is a critical next step in understanding why local governments invest in climate change policy.

454

Urban Affairs Review 47(3)

Declaration of Conflicting Interests The authors declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

Financial Disclosure/Funding The authors received no financial support for the research and/or authorship of this article.

Notes 1. This choice may seem peculiar given Bulkeley and Betsill’s argument (2003, 54) that the Cities for Climate Protection (CCP) program of the International Council on Local Environmental Initiatives (ICLEI) “appears to have made little impact within the international climate change regime to date.” However, in contrast to Bulkeley and Betsill’s primary focus on the global impact of multilevel environmental governance structures that include cities, our research examines the factors shaping individual U.S. cities’ decisions to join a climate protection network as well as their progress in implementing climate protection policies once in such a network. 2. ICLEI members pay fees based on the population size of their city. Once a local government joins ICLEI, they align with or identify ICLEI programs of interest. Our research focuses on cities that have identified ICLEI’s CCP as an interest. 3. We are indebted to an anonymous review for the suggested emphasis on this point. 4. Three study cities have the commission form of government, which fits neither category well. These three observations were dropped from the analysis. A check of the results when those three cases are included in either the mayor or the city manager categories reveals that neither scenario substantially changes the results. 5. Ideally, we would model policy adoption and implementation jointly using an analytical approach that could account for selection bias such as a Heckman selection model. In this case, we were not able to do so because model specification for adoption and implementation were too similar. Therefore, we rely upon a logistic regression to model ICLEI membership and a tobit regression for policy implementation. 6. Tobit regression allows us to account for the censoring in our data set because the majority of our cities are not ICLEI members. Ordinary least squares models estimated on data with dependent variable censoring exhibit bias (for a discussion of dependent variable censoring and tobit models, see Sigelman and Zeng 1999). 7. We are indebted to an anonymous reviewer who suggested the importance of this theme.

Sharp et al.

455

References Betsill, Michele M., and Harriet Bulkeley. 2007. Looking back and thinking ahead: A decade of cities and climate change research. Local Environment 12 (5): 447–56. Bulkeley, Harriet, and Michele M. Betsill. 2003. Cities and climate change: Urban sustainability and global environmental governance. New York: Routledge. Clingermayer, James C., and Richard C. Feiock. 2001. Institutional constraints and policy choice. Albany: State Univ. of New York Press. Donovan, Todd. 1993. Community controversy and the adoption of economic development policies. Social Science Quarterly 74 (2): 386–402. Feiock, Richard C., and Gregory Cable. 1992. Need, institutional arrangements and economic development policy. Journal of Public Administration Research and Theory 2 (4): 378–98. Feiock, Richard C., and Nathan Francis. 2010. Explaining the adoption of climate change policies in local government. Paper presented at the annual meeting of the Midwest Political Science Association, Chicago. Hendrick, Rebecca. 2004. Assessing and measuring the fiscal health of local governments: Focus on Chicago suburban municipalities. Urban Affairs Review 40 (1): 78–114. ICLEI Global. 2008. ICLEI global: Our campaigns and programs. http://www.iclei .org/index.php?id=global-programs. Koehn, Peter H. 2008. Underneath Kyoto: Emerging subnational government initiatives and incipient issue-bundling opportunities in China and the United States. Global Environmental Politics 8 (1): 53–77. Kousky, Carolyn, and Stephen H. Schneider. 2003. Global climate policy: Will cities lead the way? Climate Policy 3:359–72. Lubell, Mark, Richard C. Feiock, and Susan Handy. 2009. City adoption of environmentally sustainable policies in California’s central valley. Journal of the American Planning Association 75 (3): 293–308. Lubell, Mark, Richard C. Feiock, and Edgar E. Ramirez de la Cruz. 2009. Local institutions and the politics of urban growth. American Journal of Political Science 53 (3): 649–65. Portney, Kent E. 2003. Taking sustainable cities seriously. Cambridge, MA: MIT Press. Portney, Kent E. 2009. Sustainability in American cities: A comprehensive look at what cities are doing and why? In Toward sustainable communities, edited by Daniel A. Mazmanian and Michael E. Kraft, 227–54. Cambridge, MA: MIT Press. Portney, Kent E., and Jeffrey M. Berry. 2010.Participation and the pursuit of sustainability in U.S. cities. Urban Affairs Review 46 (1): 119–39.

456

Urban Affairs Review 47(3)

Portney, Kent E., Jeffrey M. Berry, Kelli Farrington, and Kathryn Bond. 2009. Urban nonprofits, city governance, and the pursuit of sustainability in American cities. Paper prepared for the meeting of the American Political Science Association, Toronto. Rabe, Barry G. 2004. Statehouse and greenhouse: The emerging politics of American climate change policy. Washington, DC: Brookings Institution. Rabe, Barry G. 2007. Beyond Kyoto: Climate change policy in multilevel governance systems. Governance: An International Journal of Policy, Administration and Institutions 20 (3): 423–44. Ramirez de la Cruz, Edgar E. 2009. Local political institutions and smart growth: An empirical study of the politics of compact development. Urban Affairs Review 45 (2): 218–46. Ringquist, Evan J. 1993. Environmental protection at the state level: Politics and progress in controlling pollution. Armonk, NY: M. E. Sharp. Ringquist, Evan J. 1995. Is effective regulation always oxymoronic? The states and ambient air quality. Social Science Quarterly 76 (1): 69–87. Rubin, Irene S., and Herbert J. Rubin. 1987. Economic development incentives: The poor (cities) pay more. Urban Affairs Review 23 (1): 37–62. Sharp, Elaine B. 2002. “Culture, Institutions, and Urban Officials’ Responses to Morality Issues,” Political Research Quarterly 55 (4): 861-883. Sigelman, Lee, and Langche Zeng. 1999. Analyzing censored and sampled-selected data with tobit and heckit. Political Analysis 8 (2): 167–82. Toly, Noah J. 2008. Transnational municipal networks in climate politics: From global governance to global politics. Globalizations 5 (3): 341–56. Whitford, A. B. (2009). Institutional design and information revelation: Evidence from environmental right-to-know. Journal of Public Administration Research and Theory 19 (2): 189–205. Zahran, Sammy, Grover Himanshu, Samuel D. Brody, and Arnold Vedlitz. 2008. Risk, stress, and capacity: Explaining metropolitan commitment to climate protection. Urban Affairs Review 43 (4): 447–74.

Bios Elaine B. Sharp is chair and professor of the Department of Political Science at the University of Kansas. Her research and teaching interests include urban politics and policy making, public opinion, and morality politics. She is the author of Morality Issues and City Politics (2005) and The Sometime Connection: Public Opinion and Social Policy (1999). Her recent articles on racial tolerance, variation in urban police force size, and local political culture have appeared in Social Science Quarterly, Urban Affairs Review, Political Research Quarterly, and Journal of Politics.

Sharp et al.

457

Dorothy M. Daley is an associate professor in the Department of Political Science and in the Environmental Studies Program at the University of Kansas. Her research and teaching focus on examining factors that influence environmental and public health decision making in a variety of settings. Her research has been published in Journal of Public Administration Research and Theory, Journal of Policy Analysis and Management, Political Research Quarterly, Policy Studies Journal, and Journal of Epidemiology and Community Health. Michael S. Lynch is an assistant professor in the Department of Political Science at the University of Kansas. His research and teaching interests include American politics, Congress, and quantitative methods. He is especially interested in the effects that rules and institutions have on which policies are possible in the United States. His published work includes (with W. T. Bianco, G. Miller, and I. Sened) “Constrained Instability of Majority Rule: Experiments on the Robustness of the Uncovered Set” in Political Analysis (2008) as well as work published in Political Research Quarterly and Journal of Politics.