On the foundations of behavioural and experimental economics

Abstract

Draft of a contribution to Kincaid & Ross Modern Guide to Philosophy of Economics (Elgar). To talk in 2020 about the foundations of behavioural and experimental economics (BE and EE from here on), or for that matter the methodological and substantive relationships between economic and psychological models of behaviour (Ortmann & Gigerenzer 1997), and the implications of differences in experimental methods (Hertwig & Ortmann 2001) requires that we first talk about the ramifications of the replicability- and credibility crisis that has afflicted the business and social sciences, and very prominently psychology, since at least 2005. That’s because the way you do an experiment, evaluate and report it, is unbelievably important (Camerer 2003, p. 34; Ortmann 2010) and bears, as I shall argue below, on the issue of the foundations of BE and EE, on issues of rationality and bounded rationality, self- and other-regarding behaviour, and the whole kaboodle. Here are the major themes I will cover in this chapter (with quick summary answers in parentheses): Is there a replicability and credibility crisis in the social sciences? (Yes, no doubt. Especially if we restrict ourselves to economics and psychology and laboratory experimentation as we shall do here.) Does the replicability crisis play out differently in economics and psychology. (Yes.) Are poor methodological and statistical practices (p-hacking? HARKing?) responsible for that state of affairs? (Yes.) The good news is that both can be addressed through better statistical practices, and better methodological practices such as better reporting policies, power computations, preregistrations, constraints on general(izabil)ity, etc. What then lies at the root of the problem? (The absence (presence) of a theoretical framework in much of psychology (economics), i.e., psychology’s “theoretical amnesia”.) So, what has all this do with of the foundations of BE and EE? (A lot!) EE, allegedly based on the soul of modern economic man – driven by self-interest, not to be confused with greed – provides, especially in its modern variant (rather than the caricatures that uninformed critics paint of it), the rigid framework that psychology is so badly missing (e.g., McPhetres et al. 2020). Two caveats are due here. First, I do understand that some parts of psychology (e.g., cognitive psychology) are better theorized than others (e.g., social psychology), a fact that seems related to researchers’ ability to replicate findings. Second, modern economics is here understood as modern micro-economic theory as sketched out in Kreps (2020) or Binmore (2007) and work that followed them. BE, allegedly enriching economic models with “insights from psychology”, is a mostly untheorized and adhoc-ish mess (for a particularly silly example see Tomer 2017; for a review of that book see Ortmann 2018) that does not provide the kind of theoretical framework that psychology, and behavioural economics, need. The foundations of BE, certainly to the extent that they are grounded in social psychology and judgement and decision making, are for the birds.