Party politics and the effectiveness of local climate change policy frameworks: green influence from the sidelines

ABSTRACT

Are national-level party political drivers of climate change performance reproduced locally? Here, I explore whether Greens’ ability to influence climate commitment nationally via legislative presence and coalition partnership is translated into English local government, using Climate Emergency gradings of local authority policy frameworks as the focus of comparative analysis. Scholarship on English local authority policy-making and performance suggests that, on balance, we should expect to see Green legislative presence and governing coalition effects translate to this level of government. While the finding of a positive Green legislative presence effect adds weight to the characterisation of local climate governance in England as a relatively collaborative process, the null finding on the coalition effect raises questions over the ability of junior coalition partners to realise preferences rapidly. Given the importance of sub-national politics to successful climate change transformation, it is vital that the factors associated with strengthened local commitment be further explored.

KEYWORDS:

• Climate change
• green party
• local government
• policy performance
• party politics

Acknowledgement

Liam Clegg stood as a Green party candidate in the English local elections, May 2023. Institutional guidance was to declare this candidacy within peer review and publications, to support the integrity of research processes and outputs.

Disclosure statement

No potential conflict of interest was reported by the author.

Supplementary Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/03003930.2023.2292662

Notes

1. Climate Emergency is a UK civil society organisation. In summer 2021, a team of 120 Climate Emergency researchers systematically scored local authorities’ published Climate Action Plans, evaluating dimensions including the robustness of governance and funding regimes laid out, the scale of mitigations and adaptations, and the inclusion of emissions targets and monitoring frameworks. Further information is provided in ‘Data and Methodology’, below.

2. As noted below, Folke’s (2014) study of Swedish municipalities provides an example of scholarship exploring Green party politics and local environmental policy. By focusing specifically on climate change policy frameworks rather than more generally on environmental policy, and by exploring the effect of legislative presence and coalition partnership, I offer thematic extensions to this work.

3. Abou-Chadi’s (2016) study suggests that inter-party competition may provide an additional route pathway for Green influence, with mainstream parties potentially responding to Green success by adopting a more prominent commitment to environmental issues in the subsequent election. Owing to the structure of the empirical model and focus on policy framework strength rather than manifesto commitments, it was not possible to probe this potential line of influence.

4. As denoted by survey responses ‘More influence to change things than expected’ or ‘As much influence to change things as expected’.

5. For a focus on the politics of subnational coalition formation, see Back (2003) and Debus and Gross (2016).

6. See Ministry of Housing, Communities, and Local Government official website, available at https://www.gov.uk/guidance/local-government-structure-and-elections. Accessed 17^th June, 2021. See also Clegg (2021) for additional overview.

7. The Climate Emergency council scorecard involved a team of 120 researchers assessing the effectiveness of dimensions of local authority Climate Action Plans including the outlined governance and funding frameworks, mitigation and adaptation strategies, integration of climate priorities across the council’s operations, community engagement, the measurement and setting of emissions targets, and whether an authority had declared a climate or ecological emergency. Coding was guided by a 60-page checklist to ensure comparability of evaluations, with a second stage in the process integrating a ‘right to reply’ and expert calibration. While the use of non-anonymised documentation within the grading exercise introduces possibility for researcher bias, overall the Climate Emergency data provides a useful tool with which to operationalise the climate change commitment variable. Climate Emergency data is used in Garvey et al’s (2023) study of local authority climate change commitment and capacity. See ‘Methodology’, Climate Emergency website, available at https://councilclimatescorecards.uk/plan-scorecards-2022/methodology/. Accessed 20^th February, 2023.

8. To test for the possible cumulative effect of Green legislative presence, I also created a parallel measure that followed the parameters noted in the footnote above.

9. See ‘Political Compositions’ pages of Keith Edkins website, available at https://www.theedkins.co.uk/uklocalgov/makeup.htm. Accessed 22^nd July, 2022.

10. Edkins identifies a local authority as having a Green coalition partner where there is a known public declaration of such a governing arrangement.

11. To test for the possible cumulative effect of Green coalition presence, I also created a parallel measure that captured the number of years of Green coalition presence across each of the timeframes of interest (i.e., for an authority that had just one year of Green coalition a value of 1 was given, for four years a value of 4, and so on).

12. See ONS official website, available at https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/earningsandworkinghours/datasets/placeofresidencebylocalauthorityashetable8. Accessed 23^rd June, 2021.

13. Climate Emergency score are available for 324 local authorities; gaps in The Election Centre coverage on council compositions and formatting incompatibilities across some case observation years were the main sources of missing data.

14. In theory, the maximum score attainable under the Climate Emergency scorecard would be a grade of 100.

15. To probe the potential cumulative influence from the party political factors of interest on the outcome variable, I re-ran the models employing specifications that captured the total number of years in which there was Green legislative presence and coalition partnership. Under these cumulative measures, no significant relationships were found. Consequently, it seems that it is the existence of any Green legislative presence, rather than the cumulative legislative presence, that influences the strength of climate change policy frameworks.

16. As an extension of the model, I ran versions that incorporated a measure of left-wing control with a dummy variable capturing observations where there had been at least one year of Labour majority control (in the UK context, Labour is conventionally held to represent the more left-wing of the three mainstream parties). The rationale behind this extension comes from debates over the presence of a left-wing partisan orientation effect on climate change commitment in other contexts (cf. Schultze 2014, Carter and Clements 2015, Tobin 2017; Farstad 2018; Farstad 2019).This variable was operationalised using Edkins’ Local Council Political Composition database. Reflecting the parameters used when operationalising the Green legislative presence and coalition partnership variables, I ran models with three dummy variables capturing labour control in the periods 2006–17, 2006–18, and 2006–19. With this measure, left-wing orientation was found to not exert a significant influence on climate change commitment. The inclusion of the left-wing control variable did not significantly alter the results from Models 1–3, as reported in Table 2 and Figure 2. It should be noted that this operationalisation of left-wing control represents a sub-optimal intervention: cases in which there had been a period of Labour control and a period of Green coalition presence across the given time period will be identified as both displaying Labour control and Green coalition presence, meaning that the models are not capturing the independent influence from these factors.