Getting models and modellers to inform deep decarbonization strategies

ABSTRACT

An increasing number of countries issue Deep Decarbonization Strategies (DDS). While DDS contents are well analysed, the processes by which they are developed attract less attention. This paper investigates what numerical model(s) are used in these processes, how they are used, and how models, modellers and stakeholders jointly contribute to the production of DDS. It draws lessons from an in-depth analysis of the second French national low-carbon strategy, complemented with insights from the US, Swedish and Brazilian DDS production processes. While configurations differ, DDS processes typically rely on multiple, sometimes overlapping models and involve a broad range of stakeholders. Articulating models together, and through stakeholder consultations, produces both numbers and collectives that share visions of the future – and both are equally important. Setting up and coordinating such assemblages of models and stakeholders requires effort, time, anticipation and resources. The cases presented here highlight the importance of technical, institutional and relational legacies (e.g. prior experience of joint work, hybrid communities), and of political support. We conclude on the importance for policymakers to account for these dimensions when setting up DDS processes and layout avenues for further research.

Key policy insights

• The production of Deep Decarbonization Strategies (DDS) relies on multi-model, multi-stakeholder processes and requires developing models in consultation with stakeholders.

• These modelling assemblages produce numbers, shared visions and also relationships – all of which are equally important when it comes to elaborating DDS.

• Such assemblages require time, resources, trial and error, which in turn points to the importance of institutional, relational and technical legacies (e.g. prior experience with and around modelling tools, hybrid communities), as well as of political support, in facilitating this work.

• DDS elaboration processes constitute a learning experiment and present opportunities for technical and institutional innovation.

• Policymakers should anticipate and plan for these dimensions when deciding to produce DDS.

KEYWORDS:

• Mid-century strategy
• NDC
• modelling assemblage
• public policy
• mitigation pathways

Acknowledgment

The authors gratefully thank the funding of ADEME as part of the ‘State of the art of prospective modeling’, contract n°18MAR000099186715, Emmanuel Combet (Ademe) for his support at all stages of the project, and the members of the steering committee. We also would like to thank the three anonymous reviewers and Anna Krook-Riekkola (Luleå University of Technology) for their valuable comments to previous versions of the article, and the chair Long term modeling for Sustanaible Development (ENPC-Mines Paristech) for its support.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The authors of this paper are, respectively, an economist, sociologist and political scientist.

2 Using methodologies such as scenario-axes (Van't Klooster & van Asselt, 2006), ‘story-and-simulation’ (Alcamo, 2008), Monte Carlo analysis or robust decision-making (Lempert et al., 2003), or Modeling to Generate Alternatives (DeCarolis et al., 2016; Trutnevyte, 2016).

3 The semi-structured interviews covered the following points: process organisation and chronology, models and parties involved, articulation between the models, articulation between the models and the parties, main outcomes, scope, hurdles and difficulties encountered. Depending on the amount of information already collected and on the interviewee, we focused more or less on these different points. Due to the COVID-19 pandemic, the interviews were mostly conducted by telephone or video conference and recorded. As the inquiry was mostly factual, there has been no coding of the verbatim transcript. About half of the interviews – the ones judged most interesting or most technical in their content – have been completely transcribed for analysis.

4 French Environment Ministry civil servant, interviewed by the authors.

5 French Environment Ministry civil servant, interviewed by the authors.

6 The analysis of this process is based on publicly available information and on an interview with two key academic stakeholders partly responsible for the coordination of this process.

7 Academic, interviewed by the authors.

8 The analysis of this process is based on publicly available information and on an interview with a key academic stakeholder responsible for the coordination of this process.

9 Duke and Hansel (2018) and an interview by the authors.

10 Academic, interviewed by the authors.

11 The analysis of this process is based on publicly available information and on two interviews: one with a key administrative stakeholder responsible for the coordination of this process and one with an academic modeller who contributed to the process with the TIMES-Swedish model.

12 Civil servant at the Swedish EPA, interviewed by the authors.

13 Civil servant at the Swedish EPA, interviewed by the authors.

14 Civil servant at the Swedish EPA, Interviewed by the authors.

15 CITEPA is a French interprofessional technical center working in the field of air pollution, commissioned to calculate the LCNS2 aggregated GHG emissions. MEDPRO is developed by a private company, commissioned to aggregate sectorial demand for the LCNS2.

Additional information

Funding

This work was supported by funding of ADEME (contract n°18MAR000099186715).