Evolution of opinions in the growth-vs-environment debate: Extended replicator dynamics

Abstract

The evolution of opinions in the long-standing debate on growth-versus-environment may affect support for important sustainability policies, in areas such as biodiversity loss, climate change, deforestation and freshwater scarcity. In order to understand this evolution, we develop a model describing the dynamics of four distinct opinions as identified in recent surveys, namely growth-at-all-cost, green growth, agrowth and degrowth. The model is based on modifying standard replicator dynamics to match a multi-group structure. Individuals are influenced by local or global interactions with others, based on adjacent opinion groups and exposure to information about environmental change. Psychological resistance to opinion change is also accounted for. The model is calibrated with recent survey data. Numerical analysis shows which opinions survive under particular conditions. We find, among others, that under local interactions, ultimate outcomes are characterized by lack of consensus, i.e. survival of multiple opinions. In addition, equal impacts of environmental change on opinions do not always translate in joint survival of associated opinions. Under worsening environmental conditions while continuing economic growth, opinions shift from green growth to agrowth and degrowth. Fostering global interactions among individuals, causing them to be influenced by a broad spectrum of opinions, makes consensus more likely. We also consider model dynamics if feedback from opinions to policy to environmental change and back is included. This confirms robustness of the results. It should be noted that the model is not meant to predict but to explore the consequences of combinations of assumptions about social networks, psychological mechanisms, environmental dynamics, and connections between opinion distribution and environmental policy. The study represents the first analysis of opinion dynamics in the growth-vs-environment debate and suggests a number of routes for further investigation. © 2019 Elsevier Ltd