ABSTRACT
The influence of material interests and partisanship on local residents’ responses to proposed shale gas development (SGD) is examined. Complementing existing research on attitudes toward SGD, public response is conceptualized and measured as a political behavior. A novel research design combines information on participation in town board hearings with administrative data from three sources: county tax assessment rolls, a database of gas leases, and the state voter registration file. Consistent with the theory of expressive partisanship, party affiliation strongly predicts public expression of support for and opposition to a proposed local ban on SGD, net of the effects of key measures of material interests. The study contributes to research on public response to industrial siting and to ongoing debates about partisan polarization and political participation.
Disclosure statement
No potential conflict of interest was reported by the author.
Supplementary material
Supplemental data for this article can be accessed here.
Notes
1. Brewery Ommegang is a craft brewery in Cooperstown, New York that specializes in Belgian-style ales.
2. Measures of population are from the 2010 Census, measures of poverty and education are based on the 10-year ACS estimates, and Democratic lean is based on vote share for the Democratic candidate in the 2010 Gubernatorial election.
3. This population includes 64 communities that passed a zoning ordinance permanently banning SGD before July 2013.
4. Bugden et al. (Citation2017) use self-reports to measure political behavior related to SGD. From a theoretical perspective, this is a step in the right direction for studying public response. However, self-reports can also have poor correspondence with actual behaviors (e.g. Corral-Verdugo Citation1997). Thus, measurement of actual behavior should be preferred when feasible.
5. Source: 2010 Census.
6. The Working Families Party and Conservative Party are organizationally independent from the two major parties, but in all recent elections, they have co-nominated the Democratic and Republican candidates for Governor, respectively (co-nomination is a unique feature of NY election law). Because they endorse the same candidates and comfortably sit on the same ideological dimension with Democrats and Republicans, I chose to collapse the categories. Models that include these parties separately yield consistent results. Henceforth, when I refer to ‘Democrats’ and ‘Republicans,’ I mean these collapsed categories.
7. An alternative approach is to simply model opposition and support at the household level. I present results from household-level analyses in the Supplementary Online Materials (SOM1), and discuss them briefly in the results section.
8. I do not interact with the Green Party dummy, because the effect cannot be reliably estimated due to the small number of cases.