Organizational Factors of Environmental Sustainability Implementation: An Empirical Analysis of US Cities and Counties

Abstract

Environmental sustainability goals are increasingly embedded in local planning, but implementation proves difficult. Using a survey of 217 planners working in a random sample of 146 small to mid-sized American cities and counties, we identify the organizational factors that support and hinder the implementation of environmentally sustainable practices. The analysis is based on a conceptual framework that encompasses organizational capacity, culture, structure, participatory decision-making, the framing of sustainability and contextual factors. We find that environmental sustainability implementation is lagging (although cities are generally ahead compared to counties) and that outcome evaluation is rare, precluding adaptive learning. The major barrier to implementation is that sustainability is low on political and managerial agendas. As expected, local public support, innovation-supportive organizational culture and the prioritization and framing of environmental sustainability support implementation. Surprisingly, innovation diffusion does not occur across neighbouring localities, local capacity and public participation are irrelevant for implementation and hierarchical rather than integrated institutional structures support implementation.

Key Words:

• local government
• organizations
• environment
• sustainability
• implementation

Acknowledgements

We thank the members of the expert panel for their very insightful input in the early phases of the research. We also thank the anonymous reviewers for their feedback on this paper.

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes

1. Sustainability encompasses economic, sociocultural and environmental goals. We focus here solely on its environmental dimensions. We define environmental sustainability to include land development patterns, habitat and open space conservation, transportation and air quality, waste, water, energy, natural hazards, climate and food management.

2. We constituted an ad hoc expert panel of 23 academics and practitioners from the USA, Canada, New Zealand, Great Britain, Germany, The Netherlands and Sweden. They were identified based on recognized publications, reputations and leadership in cities and organizations working in sustainability. They reviewed the conceptual framework and the questionnaire.

3. Cities and counties were randomly sampled to generate a representative sample of mid-sized cities and counties in each of the four US Census Bureau regions (Northeast, South, Midwest and West). We underrepresented localities with populations 10,000–20,000 (43% of all US localities but 31% in the sample) and oversampled localities with populations 20–50,000 (35% of US localities and 46% in the sample). This strategy resulted in a balanced sample in terms of population size. We retained population proportions for cities 50–100,000 and 100–200,000 (15% and 7%, respectively).

4. We used these keywords to identify respondents on official websites: planning, public works, land use, water and waste management, energy, parks and recreations and open space (‘planner’ did not need to be in their job title).

5. Inter-coder implementation scores reveal only five localities with standard deviations above 0.5, and two of them have no clear outliers. Intra-class correlations show that variance between LGOs is greater than that within LGOs for implementation and key concept scores. Hausman regression models predicting implementation perform as well with fixed as with random effects. We ran all descriptive statistics at the individual and agency levels. All results are within 0–5% of each other, and within a 10% range for questions scored on 1–4 scales. Therefore, multiple responses are consistent and reliable, and we present organizational-level data.

6. To assess implementation, we focus on actions rather than plans. However, in the area of climate change where progress is recent and specific actions are expected to be uncommon, we consider the adoption of climate plans as an early, or proto-indicator of action.

7. Since this is a cross-sectional study, we needed to mitigate the possibility that capacity and commitment to sustainability are endogenous (organizations with higher commitment are expected to allocate more resources to implementation). Therefore, we did not ask about capacity and resources allocated to the design or implementation of sustainability measures per se. Instead, we asked about capacity in general, that is, about employee training, knowledge and resources to get their ‘job done’ (see Table 1). In addition, we placed this question after a series of general questions on organizational culture so that respondents would not be primed about sustainability before answering questions of capacity.

8. Nationwide, median household income in 2010 was $53,000 (slightly less than that in our sample) and 28% of the US population was non-white, more than that in our sample. Our sampling strategy excludes large cities and counties, which are more diverse than smaller ones.

9. This is in particular contrast with the finding of Wang et al. (2012), who find high implementation scores for a series of government actions. Some are unsurprisingly high because they are national mandates (e.g. monitoring water quality), while others are very common (e.g. planting trees). These higher scores can be explained by the fact that they consider larger cities (population 50,000 or more) and exclude counties. They can also be explained by a different measurement strategy: respondents indicated whether or not cities adopt certain practices, not the actual implementation stage. Thus, a respondent may state that his or her city has a policy on alternative fuel vehicle purchase even though no vehicle has been purchased yet, or that his or her city has ‘joined a sustainability group’, although no tangible outcome may have resulted from this action.

10. The successful sustainability initiatives most often mentioned address stormwater runoff, waste reduction/recycling, flood control and floodplain management, energy efficiency of municipal fleets and buildings, local food systems and protecting sensitive lands.

11. Respondents could select up to two rungs to describe participation in their agency.

12. We exclude public participation from the models presented. It is never a significant factor of implementation.

Additional information

Funding

This work was supported by the US National Science Foundation [grant number 1122730].