A review of cognitive biases in strategic decision making

Abstract

This paper presents an integrative review of empirical research (2000–2023) on cognitive biases that affect decision makers in established organizations as they make strategic decisions. We examine patterns in the measures, antecedents, and outcomes of two broad categories of biases: systematic biases that operate similarly across individuals (e.g., overconfidence, escalation of commitment, loss aversion, and myopic loss aversion), and idiosyncratic biases that depend on the decision maker's experience and past interactions (e.g., myopia and local search bias). We also distinguish between findings with strong empirical evidence and those with less empirical support. Our review indicates researchers measure both types of bias using one or more of three broad approaches: assuming or inferring the bias, measuring it directly, and experimentally manipulating the bias and observing its effects. We find strong empirical support for firm ownership, performance or performance relative to aspirations, and CEO compensation and wealth as antecedents to loss aversion and myopic loss aversion. We also find that loss aversion has strong but mixed effects on outcomes such as diversification or internationalization, acquisitions, R&D intensity or investments, and risk taking. Findings with less empirical support include, among others, mostly mixed effects of loss aversion and framing on innovation, mostly positive effects of overconfidence on innovation and risk taking, and negative effects of overconfidence on corporate social responsibility, performance, and forecasting. Based on our findings, we discuss the challenge of identifying and measuring a bias in a way that is relevant to strategic management and suggest directions for future research.