The triangular relationship between public concern for environmental issues, policy output, and media attention

ABSTRACT

We theoretically argue for and empirically examine, for the first time, a triangular relationship between concern among citizens about the environment, media attention to environmental issues, and policy output. Previous work has studied these relationships, but analyzed the respective links rather in isolation from each other. This research has significant implications for the understanding of environmental policymaking and, more generally, informs the debate on whether politicians respond to what voters want and under what circumstances this occurs.

KEYWORDS:

• Environmental policymaking
• policy output
• public opinion
• media attention
• triangular relationship

Acknowledgments

We thank the special issue’s editors, Thomas Bernauer and Aseem Prakash, the journal’s editorial team, Sikina Jinnah and Christopher Rootes, and three anonymous referees for valuable comments that helped to improve the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed here.

Notes

1. Note, however, that Downs has been criticized for ignoring the way issue interlinkage can increase or suppress attention, for postulating too many cycle stages, and for neglecting the way that attention can be sustained when new institutional structures emerge (see Gupta and Jenkins-Smith 2015). It may also not account for the empirical pattern in relation to all aspects of public opinion about environmental issues (Holt and Barkemeyer 2012), though Downs (1972) suggested that cycles are more likely to occur in some conditions, e.g. including benefits of action being concentrated in a minority with costs widely spread.

2. In contrast to the issue-attention cycle, the thermostatic model postulates that, where citizens’ preferences vary, policy responds to the difference between the level of policy and the median citizen’s preferred level (Wlezien 1995) or the public’s average left-right policy mood (Stimson et al. 1995), resulting in negative feedback if policy overshoots the level citizens desire. Considering variation in public attitudes towards environmental policy across US states and across time, Johnson et al. (2005) find evidence for a thermostatic feedback effect reducing public concern when policy outputs are accompanied by actual improvement in water-pollution levels.

3. Holt and Barkemeyer (2012) find that a series of issue-attention cycles characterize media coverage of climate change, but that media attention to sustainable development might best be characterized step jumps in coverage that do not return to previous low levels as implied by Downs (1972).

4. That said, media attention may not lead to more ambitious environmental policy outputs (see Davis 2002, Lloyd 2004, Lewis et al. 2008a, 2008b). For instance, the media could misinterpret information (Boykoff and Boykoff 2004), which then has little impact or influence on policymaking. Environmental issues may also not be high enough on the agenda to stimulate the sort of public concern that prompts concerted political action (Gavin 2009).

5. The online appendix also discusses a series of robustness checks that further increase the confidence in the validity of our results.

6. We assume that public concern over the environment positively correlates with public concern over climate change. We eventually opted for Eurobarometer data on environmental changes more generally due to the larger country-year coverage. In fact, there are only two Eurobarometer years that deal explicitly with climate change in our time period, one of them (1993) overlaps with the item on environmental concerns generally. Specifically, in Eurobarometer 39.1, the following question was included: ‘Can you tell me if the greenhouse effect (global warming) is a very serious problem (1), quite serious (2), or not very serious (3)?’ The pairwise correlation of this item with our non-inverted data on environmental concerns generally is r = 0.7140 (with p = 0.0061).

7. Counterarguments suggest, however, that more direct relationships exist. For example, regulations, often blamed for poor economic growth, could directly shape public opinion (see also Kahn and Kotchen 2011, Scruggs and Benegal 2012). We are aware of these claims and, thus, assess the validity of our instruments extensively in the appendix.