Learning equilibria in growth-pollution models

Abstract

Purpose – This paper seeks to explain how inefficient learning rules may lead to a perception of economic and ecological realities that may be systematically distorted in the long run. Design/methodology/approach – The paper evaluates long‐term growth in standard growth‐pollution models. Expectations about future levels of pollution are formed under adaptive learning. Findings – Socio‐economic players (private agents, governments, non‐profit organizations and/or groups of states) may fail in understanding, with full accuracy, long‐term environmental conditions. The perception about environment threats acquires a cyclical nature, even when ecological problems evolve steadily. Research limitations/implications – Relevant policy implications emerge if the agent is unable to compute the true levels of environmental pollution that will persist in the steady state. Authorities of several kinds are likely to underestimate or overestimate ecological problems. Practical implications – The learning approach to the perception of the environment can be applied to other economic, social and biological issues, besides material growth. For instance, it can contribute to explain some cases of over‐exploitation of resources: even in the presence of a social planner capable of avoiding typical “tragedy of the commons” situations, this entity may fail in perceiving the reality and, thus, in applying the policies that prevent the exhaustion of resources. Originality/value – The paper contributes to the literature on growth and environmental issues, but takes a step forward: it approaches not only the observed relation between economy and ecology, but also the impact over the observed relation of a systematically incorrect interpretation of such a connection.