Learning-by-Erring: Towards a New View of Rationality in Economics
30 Pages Posted: 27 Apr 2020
Date Written: April 3, 2020
The cornerstone of mainstream economic theory is the premise of rationality. Humans are assumed to be rational economic agents who, subject to the available information and limited resources, are able to select, among a set of alternatives, the best means to maximize their ends and their individual subjectively defined welfare. This view has been subject to debate and criticism from the earliest days of modern economic thought. Is human action rational or irrational and how successful are economic agents at maximizing individual or collective welfare? In this paper, the authors argue that economic agents are endowed with the capacity for rational thinking but not the ability to maximize their ends, as it is acquired through learning from the mistakes we make in practice. The efficiency of rational decisions is limited to the amount of experience and knowledge we possess when decisions are taken. The authors introduce a learning function describing the rational decision-making process and argue that erring is not trivial but carries significant costs and consequences. Making errors is the price we pay to learn and accumulate valuable new knowledge and human capital. Inherent to the decision-making process is the making of errors which, rather than being trivial, are a valuable opportunity to learn and accumulate knowledge. The rational decision-making process is enhanced from the accumulation of knowledge, which raises the efficiency of our decisions, moving closer to maximizing outcomes, and increases welfare.
Keywords: Rationality, Rational Behavior, Erring, Decision Making, Learning, Knowledge, Human Capital, Bounded Rationality, Behavioral Economics, Ecological Rationality, Heuristics, Neuroeconomics, Learning-by-Doing, Learning-by-Erring, Experience Curve, Utility Maximization, Knowledge, Economic Welfare
JEL Classification: B25, B26, B40, B41, D01, D11, D60, D87, D90, D91, J24
Suggested Citation: Suggested Citation