Attitudes towards CO₂ taxation – is there an Al Gore effect?

Abstract

Fuel taxes are one of the most powerful climate policies. Yet, these taxes have not been given very much attention in the global debate regarding climate policy, compared with other instruments, such as tradable emission permits. This article shows, however, that the immense media coverage during fall 2006 significantly affected people's attitudes towards the CO₂ tax on gasoline. We conducted a survey where we asked for people's opinions about the CO₂ tax in September and in December 2006, i.e. before and after the release of Al Gore's ‘An Inconvenient Truth’ and the Stern Review. We found that the attitude towards the level of the CO₂ tax was significantly changed after these events; people became much more positive towards the tax. This signals that using the CO₂ tax as an important climate policy becomes more politically feasible and legitimate when more attention is drawn to climate change problems.

Acknowledgements

The authors would like to thank Thomas Sterner for useful comments. This research was financially supported by Mistra's Climate Policy Research Program (CLIPORE) and the Malmsten Foundation.

Notes

¹DellaVigna and Kaplan (2007) found that Fox News convinced a nonnegligible share of its viewers to vote Republican. Martinez and Scicchitano (1998) concluded that media campaigns to promote recycling actually affect people's behaviour.

²The statistics were found at the media archive http://www.retriever.se.

³The Swedish excise tax on gasoline consists of two parts, the CO₂ tax and an energy tax. The CO₂ tax was SEK 2.13 (EUR 0.23) per litre gasoline in 2006.

⁴To use student samples is common. The issue of representativeness has therefore been discussed extensively in the literature. For example, Danielson and Holm (2007) found no difference between students' and nonstudents' economic behaviour and the authors also discussed the issue of representativeness in more general terms. Also the study by Güth et al. (2007) supports a high degree of external validity of student data.

⁵No students were asked to state the same opinion as last time, and there were no questions from the students regarding if they had to fill in the same answers. The survey was made as part of a larger study, and there is no possibility for us to identify a particular student's answer in the September and December surveys in a panel fashion, but we analyse our two subsamples as two repeated cross sections.

⁶Summary statistics for the explanatory variables are presented in the Appendix.

⁷Running an ordered probit instead of OLS does not alter the results.