Framing local climate change policies in the US Great Plains

ABSTRACT

Climate change policies have become priorities for many local governments across many countries. In the United States, however, skepticism of climate science and political motives remains dominant in much of the country. A common challenge for local government stakeholders in the US is to frame climate policies in terms that are acceptable to citizens in order to overcome politically polarized inaction. We examine why some communities are successful at framing policies as climate change while others are not. We utilize data from a 2012 survey of 232 cities across the US Great Plains region that asks which of 14 mitigation policies and 14 adaptation policies have been adopted within the past decade, and participants identified those policies that were specifically framed as climate change. We test the influence of factors from three clusters of variables: the policy environment, the attitude of local government officials, and the community atmosphere toward climate change, on the propensity to frame climate policies. Results suggest local leadership is sufficient for framing one policy, but community atmosphere eclipses government attitudes and becomes the driving factor in cities where two or more climate policies are explicitly framed as such.

KEYWORDS:

• Climate policy
• local government
• framing
• US Great Plains

Acknowledgements

The authors are grateful for the valuable feedback and suggestions provided by the anonymous reviewers and helpful assistance from the journal editors.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Rebecca J. Romsdahl is a Professor of Earth System Science and Policy at the UND.

Robert S. Wood is a Professor in the Department of Political Science and Public Administration at the UND.

Dana Michael Harsell is a Professor in the Department of Political Science and Public Administration at the UND.

Andy Hultquist is affiliated with Political Science and Public Administration at the UND.

ORCID

Rebecca J. Romsdahl http://orcid.org/0000-0001-8711-3953

Dana Michael Harsell http://orcid.org/0000-0003-4919-6537

Notes

1 Great Plains states included: Montana, Wyoming, Colorado, North & South Dakota, Minnesota, Nebraska, Kansas, Oklahoma, and Texas

2 Given this low response rate, the question of the sample’s representativeness is a valid one. To assess the representativeness of these responses, we employed a t-test to examine whether responding cities differed from nonrespondents along three dimensions – population, 2010 per capita income and political characteristics (percent Democratic vote in the 2012 Presidential election). While the mean population and per capital income of responding locations were each found to be higher than those of nonrespondents, this difference was not statistically significant at the 5% level. Respondents showed slightly more Democratic support (38.3%) than nonrespondents (36.2%), a difference which, while statistically significant at the 5% level, is rather small in terms of practical magnitude. To gauge whether cities in different states responded at different rates, we also used a t-test to compare each state’s response rate to that of the sample overall (24.9%). For eight of the ten states in our sample, no significant difference was found at the 5% level. For the two states where a significant difference was found, Texas exhibited a slightly lower response rate (15.8%), while Montana showed a higher response rate (44.4%). While these results do indicate some statistically significant differences between responding cities and our overall sampling frame, at the same time, we feel that these differences are relatively minor in practical terms, and do not impair the ability of our results to be generalized to typical communities in the Great Plains region.

3 Question wording: ‘In your city’s policy and planning activities (for energy, conservation, natural resources management, land use, or emergency planning, etc.) how is climate change framed?’

4 This 6.1% figure may actually overstate the potential loss represented by excluding these cases from our analysis. Given that five of these eight cases are missing information for one or more of the model’s independent variables, and would thus already be dropped from the estimation of our regression equation, a total of only three usable cases are ultimately removed from our sample due to restriction of our analysis to only those cities with two or more total adaptation and/or mitigation policies in place.