Project
EF4-5
An Agent-Based Understanding of Green Growth
Project Heads
Sarah Wolf
Project Members
Aida Saraí Figueroa Alvarez
Project Duration
01.07.2021 – 30.06.2024
Located at
FU Berlin
Description
This project aimed to contribute to the theory of Green Growth transitions, that is, make explicit mechanisms through which an economy can switch to a low carbon structure that at the same time comes with benefits in economic and social dimensions. To this end, we develop and analyse a series of agent-based models (ABM) which extend a standard economic model: the Ramsey growth model. The project’s models build on previous work by Steudle et al. [2018], who add three economic mechanisms to a strongly simplified Ramsey growth model structure. Technical progress through learning-by-doing, directed technical progress for a brown and a green capital stock, and a labour market with search. Focusing on two time steps, the authors provide a proof of concept of how with these three building blocks a coordination game structure arises: agents can coordinate on a better “green” equilibrium rather than staying in the given “brown” one with lower investment, lower growth, and higher unemployment. The present project aimed to investigate the first two mechanisms further, by adding a dynamic component that enable the study of transitions between such equilibria. This was done in a step by step process to obtain a mathematically understandable ABMs.
Agent Based Ramsey Growth model with endogenous Technical progress (ABRam-T)
The first ABM [Figueroa Alvarez et al., 2024] developed explores the first building block: endogenous technical progress. The approach to understanding the system’s dynamics was bottom-up, by defining and analysing farms with different investment strategies that result from different assumptions about the information farms have about endogenous technical change. E.g., “ignorant” farms disregard their contribution to technical progress, as is commonly assumed for decentralised economies, i.e., systems with many agents. While economic textbooks start out from a single representative agent, or “benevolent planner”, and a “collaborative” agent was defined here by analogy, the assumptions that need to be made here are do not in fact make sense [Figueroa Alvarez and Wolf, 2024]. Agents that even in a decentralized economy consider their contribution to technical progress do not occur in the literature. Defining them required equipping them with expectations on the other agents’ total investment for their utility optimization. This type of agent is called “witty farm”.
Exploration of economies populated by such agents yields that standard economic results can be reproduced in systems of homogeneous agents, that unequal capital distributions may converge to equal ones with higher growth, that taking into account one’s contribution to technical progress does not qualitatively change the picture, and that the economic performance of a system does not depend on how well agents’ expectations match the true values of investment of others [Figueroa Alvarez and Wolf, 2024].
Agent Based Ramsey growth model with Brown and Green capital (ABRam-BG)
The second model ABRam-BG, was developed to investigate different pathways for emission reductions. The ABRam-T model was extended for the case of two capital stocks: “brown” and “green”, representing high and low emissions respectively. Given an economy in which the brown production process has historically accumulated a much higher level of technical progress, there aren’t incentive to switch to a green production process for agents. To nevertheless model the decision to adopt green technology, we integrated social strategies for decision-making (Galesic et al. [2023]). By adding environmental attitudes to the agents. The agent’s beliefs and attitudes are update in such way that define the agent’s investment behaviour. We consider for the belief updating schemes a Voter Model an a Majority Rule. Different settings are explored for the opinion dynamics process and its interactions with the (deterministic) economic process. Beliefs that lead to green investment can, in fact, drive a green transition, where the interaction of opinion dynamics and economic dynamics can be statistically characterized by the speed of the opinion dynamics process (indicating how fast brown agents switch to green) and the growth of the brown capital stock (as an indicator of when agents start switching to green). The comparison of the two opinion dynamics mechanisms used shows multiple pairwise interactions to be more effective than a collective interaction among more than two agents in driving a green transition.
In terms of economic outcomes, it is found that earlier green transitions decrease emissions and increase growth compared to later transitions, confirming that a green transition needs to be a timely event (Lamperti et al [2020]). While all scenarios with belief dynamics come with a loss in GDP with respect to a baseline scenario where no farms switch to the green technology — negating the possibility of green growth — among scenarios with belief changes, those with an earlier green transition provide lower emissions and higher the GDP. Positing that a green transition is necessary in any case, and hence ruling out business as usual in a “brown” economy, this underpins an idea of green growth.
References
G.A. Steudle, S. Wolf, J. Mielke, and C. Jaeger. Green growth mechanics: The building blocks. Global Climate Forum, January 2018. Available at https://globalclimateforum.org/wp-content/uploads/2018/01/GCF_WorkingPaper1-2018.pdf
M. Galesic, D. Barkoczi, A. Berdahl, D. Biro, G. Carbone, I. Giannoccaro, R. Goldstone, P. Gonzalez, Cleotilde, A. Kandler, A. Kao, and et al. Beyond collective intelligence: Collective adaptation. Journal of The Royal Society Interface. March 2023.
F. Lamperti, G. Dosi, M. Napoletano, A. Roventini and A. Sapio. Climate change and green transitions in an agent-based integrated assessment model. Technological Forecasting and Social Change, 153:119806, Apr. 2020.
Selected Publications
A.S. Figueroa Alvarez and S.Wolf. Exploring heterogeneity and expectations in technical progress – an exercise in agentization. Submitted, 2024. A working paper version can be found at https://zenodo.org/records/10654996.
A.S. Figueroa Alvarez, M. Tokpanova, and S. Wolf. Agent-based Ramsey growth model with endogenous technical progress (ABRam-T) (version 1.0.0). Available at https://www.comses.net/codebases/0d2d9f03-e579-4d59-bd9a-35be1439f8ed/releases/1.0.0/, 2024.
A.S. Figueroa Alvarez, S. Wolf “Agent-Based Ramsey growth model with Brown and Green capital (ABRam-BG)” (Version 1.0.0). CoMSES Computational Model Library. Retrieved from: https://www.comses.net/codebases/44bc2da3-a186-4bf3-8e8e-08e4b3fe81fa/releases/1.0.0/, 2024.
Selected Pictures
ABRam-T agents' optimal investment
Optimal investment – for ABRam-T collaborative, ignorant, and witty agents – as a function of the agent’s capital with different values of the economy total capital K0 and technical progress η0.
Selected Pictures
Statistically characterizing a green transition
Average consensus rate against brown capital growth rates across ABRam-BG simulations for different numbers of interacting agents
Selected Pictures
ABRam-BG simulation output: GDP and CO2 emissions
GDP and CO2 emissions for a non-transition scenario (brown solid line), early transition scenario (blue dashed line), late transition scenario (red dashed line with dots) and carbon lock-in scenario (orange dash-dotted line with stars).
Please insert any kind of pictures (photos, diagramms, simulations, graphics) related to the project in the above right field (Image with Text), by choosing the green plus image on top of the text editor. (You will be directed to the media library where you can add new files.)
(We need pictures for a lot of purposes in different contexts, like posters, scientific reports, flyers, website,…
Please upload pictures that might be just nice to look at, illustrate, explain or summarize your work.)
As Title in the above form please add a copyright.
And please give a short description of the picture and the context in the above textbox.
Don’t forget to press the “Save changes” button at the bottom of the box.
If you want to add more pictures, please use the “clone”-button at the right top of the above grey box.