« Back to Results
Methodological Issues in Behavioral Economics
Saturday, Jan. 5, 2019
8:00 AM - 10:00 AM
International Network for Economic Method
John B. Davis,
Marquette University and University of Amsterdam
Behavioral Economics between Normativism and Descriptivism
“Economics needs two completely different types of theories: normative and descriptive” says Richard Thaler in his most recent article published in American Economic Review (Thaler 2018: 1267). This text summarizes the main findings of Thaler’s work in behavioral economics for which he was awarded the Nobel Prize in economic sciences last year. Indeed, he subscribed to the view that there are normative and descriptive theories in economics and in studies of decision-making throughout the course of his academic career. Many economists, philosophers of economics and decision theorists share this view (e.g. Camerer & Lowenstein 2004, Hausman & McPherson 2006, Baron 2004). It is commonly believed that the expected utility theory (EUT) is a theory that serves as a normative benchmark, but does not explain and predict well. Hence, EUT is not an adequate descriptive theory, but a normative one. It is far from clear, however, what economists in fact understand by the normative status of EUT and of economic theory more generally. Furthermore, some scholars (Hands 2015, Guala 2000, Heukelom 2012, Herfeld 2016, Erickson et al. 2013, Starmer 2005) have suggested that the normative interpretation of EUT was not always present in economics and that it appeared after the theory was challenged in experimental settings. This observation is the starting point for my analysis. I discuss diverse ways in which economists seem to understand and justify the normative understanding of EUT and I show that neither is satisfactory from the philosophical and methodological point of view. I call behavioral economists’ widely accepted belief that ‘economics needs normative and descriptive theories’ the commitment to normativism and descriptivism. I argue that this commitment is a conflation of an explanatory and a philosophical project and I show what implications it has for behavioral economics’ research agenda and its policy applications.
Foundations of Libertarian Paternalism: Normativity, Rationality, and Welfare
There is an extensive critical literature within the philosophy of economics analyzing the libertarian paternalism of Richard Thaler and Cass Sunstein. This paper will examine many topics from this literature, but do so from a different perspective than most of the existing research. First, the main focus of the paper will be different. Recalling Thaler and Sunstein’s distinction between Econs (those whose behavior will be unchanged by libertarian paternalist policies) and Humans (who will, at least potentially, change their behavior in ways that makes them “better off, as judged by themselves,” Thaler and Sunstein, 2009, p. 5), this paper will take Econs seriously by focusing primarily on the particular features of the Econs who are not making mistakes rather than on correcting the behavior of Humans who are. Secondly, although the paper will draw on several ideas from the existing literature, the particular roles that these ideas play in the discussion will be somewhat different. Third, the normative character of the Econ reference point will be examined in closer detail by emphasizing the difference between normativity with respect to rationality, and normativity with respect to welfare/well-being. Finally, in the concluding sections, the question of the role of “the social” will be raised and how it concerns, and ought to concern, nudging and the associated policy prescriptions.
From Behavioral Welfare Economics’ ‘Reconciliation Problem’ to the Individuality Reconstruction Problem
: Behavioral Welfare Economics’ ‘reconciliation problem’ is that, if we accept that behavioral economics’ heuristics and biases program provides foundations for positive economics, then positive and normative economics are no longer consistent with one another and somehow need to be reconciled. However, if individuals have time-inconsistent preferences and behave as if they have multiple selves, it is not only unclear what economic policy should be based on, but also who it applies to. The dual selves literature intended to address the reconciliation problem thus has as a companion task dealing with an individuality reconstruction problem. This paper critically discusses the ‘preference purification’ approach this literature pursues and the problematic nature of the idea of ‘an inner rational self’ akin to the traditional subjectivist individual conception. It then reviews four alternative strategies for reconstructing individuals that in different ways make factors external to individuals determinative of individuality, and summarizes the state of thinking about individuals in standard economics.
Theory Choice in Behavioral and Neoclassical Economics
In "The Seven Properties of Good Models” (2008), Xavier Gabaix and David Laibson propose that economic models should be assessed on the basis of seven properties: 1) parsimony, 2) tractability, 3) conceptual insightfulness, 4) generalizability, 5) falsifiability, 6) empirical consistency, and 7) predictive precision. The aim of this paper is to assess their proposal from the point of view of more recent contributions to the philosophy of science. My thesis is that Gabaix and Laibson succeed in capturing many important features of theory choice in economics and elsewhere, but that they get important details wrong.
B4 - Economic Methodology