Professor of Behavioural Science
Warwick Business School
University of Warwick
Address: Warwick Business School
University of WarwickEmail: email@example.com
Coventry, CV4 7AL
Phone: +44 (0)24 7652 4306
I am Professor of Behavioral Science at Warwick Business School, University of Warwick. and I was previously Professor of Strategy and Decision Making at University of Oxford and Associate Professor at Stanford Graduate School of Business. My research explores how learning, sampling, and chance mechanisms can offer novel explanations of important social phenomena including in-group bias, persistent inequality between individuals and firms, risk taking as well as conformity.
My research has shown how individuals might develop and maintain inaccurate beliefs because they rely on the biased samples of information they obtain from their experiences. In a series of five papers published in Psychological Review during the last ten years I have shown how a simple learning mechanism - the Hot Stove Effect - can offer novel interpretations of discrimination, ingroup-bias, social influence, and risk aversion (Denrell, 2015, 2007, 2005; Denrell and Le Mens, 2007; Le Mens and Denrell, 2011). My work has tried to show how these biases can emerge from the biased set of experiences people are exposed to instead of from cognitive limitations of human learners. The key idea is that the tendency to avoid alternatives with unfavourable past outcomes generates a biased set of experiences.
There is good experimental evidence in support of the Hot Stove Effect and researchers in psychology have applied it to explain regularities in diverse domains including why risk taking declines with experience (people stop sampling alternatives they had a negative experiences with and the probability that a negative experience has occurred increases over time, Erev et al, 2010, Journal of Behavioral Decision-Making) and why people underestimate the trustworthiness of others (if people false believe that others cannot be trusted they avoid them and by avoiding them their false belief cannot be disconfirmed, see Fetchenhauer & Dunning, 2010, Psychological Science). Researchers in finance (Dittmar and Duchin, 2016, Review of Financial Studies) have also shown that the hot stove effect can explain risk taking behavior by CEO's and CFO's. Using data on the past careers of these managers, Dittmar and Duchin examined whether managers had experience of being employed in firms that filed for bankruptcy or firms that experienced other adverse financial shocks. Managers with such adverse experiences were significantly more risk averse in the sense that they held less debt and invested less.
My work in organization theory and strategy has explored the implications of learning and imitation for the diffusion of best practice. Will best practices spread as a result of imitation? When does imitation fail or lead to adoption of risky practices? In a series of papers I have shown that imitation of top performers can lead to adoption of risky and inefficient practices. For example, in Denrell and Liu (PNAS, 2012) we show that rich-get-rich processes imply that the highest performers in fact are not the best (do not have the highest expected skill). The intuition is that extreme performance indicates the presence of strong rich-get-rich dynamics rather than exceptional skill. Persistent high performance can therefore be an indicator of inferior capabilities; a topic that we explore empirically in Denrell, Fang, and Zhao (Strategic Management Journal, 2013).
An important theme in my research is how randomness at the micro-level can give rise to patterns at the aggregate level which can offer parsimonious explanations of many regularities in strategic management and organization theory. For example, in Denrell (2004, Management Science) I proposed that a random walk model could account for much of the empirically observed persistence in performance. In Denrell and Kovacs (2008, Administrative Science Quarterly), we show how the common practice of studying only widely diffused practices and relatively large and successful industries can lead to spurious findings such as spurious non-monotonic density dependence. A survey of this line of research was published in Organization Science, 2015.