A few years back in this journal, Ivan Moscati published a passionate plea for historians of economics to pay more attention to contemporary economic controversies. Using serious history of economics to contribute to current debates was, to Moscati (Reference Moscati2008), a way both to make historians of economics more relevant to a broader community of economists and scholars, and to secure the survival of history of economics as a vibrant subdiscipline of economics. A prime target for Moscati-style history of economics would seem to be behavioral economics, which became part of the mainstream relatively recently and raises a series of interesting issues of history, philosophy, and methodology. In spite of its meteoric rise, however, behavioral economics has remained under-studied by serious historians and philosophers of science. The absence of such work has consequences: among other things, it can be argued that a hazy understanding of what behavioral economics is and where it is coming from continues to befuddle practitioners and critics alike.
Now, historian of economics Floris Heukelom has published the first book-length history of behavioral economics, and it is indeed highly relevant to current debates about the nature of behavioral economics, its relationship to standard theory, and its welfare and policy implications. Behavioral Economics: A History is an exceptionally solid piece of work that makes the most of the opportunities offered by the history of contemporary economics.Footnote 1 Above all, it draws not just on published and archival sources, but on a series of personal interviews and email exchanges with the principal characters of the story, including Robyn Dawes, Daniel Kahneman, David Krantz, and Eric Wanner. (One hopes the source materials will ultimately be deposited in some archive where other scholars may freely access them.) The result is fascinating reading, its main shortcoming being that it ends too soon.
Heukelom’s book, which is based upon but goes beyond a series of previous publications, weaves a narrative that is both novel and compelling. The narrative has two main strands. The warp, as it were, is the shift from what Heukelom calls an “epistemology of generalized characterizations” to an “epistemology of directly refutable empirical claims” (p. 6). The former, which Heukelom traces to John Stuart Mill’s methodological writings, treats the fundamental assumptions of economics as general characterizations of rational economic behavior, which deliberately leave out a great deal of empirical content and therefore are not subject to direct empirical confirmation or disconfirmation (p. 65). The latter, which Heukelom argues grew out of the experimental-psychology tradition of Gustav Fechner and Wilhelm Wundt, treats all claims about human behavior as empirical and testable, ideally in a laboratory context (p. 2). In Heukelom’s story, this shift was an important precondition for the emergence of behavioral economics, which treats the axioms of choice theory and their implications as empirical claims to be tested directly or indirectly, e.g., in laboratory experiments.
The weft of the narrative is the shift from the positive–normative distinction of John Neville Keynes and Milton Friedman to the normative–descriptive distinction originally favored by the early experimental psychologists. The former distinguishes a non-ethical and scientific (positive) domain that concerns itself with what is from an ethical and non-scientific (normative) domain that concerns itself with what ought to be (pp. 65–66). Thus, the effort to develop a system of generalized descriptions of economic behavior would belong to positive economics, whereas the effort to judge societies as morally superior or inferior would belong to normative economics. The latter distinguishes the normative standard for proper or rational behavior in a particular situation from the actual, observed behavior in that situation typically in the context of a laboratory experiment (p. 67). Thus, when participants in laboratory experiments are exposed to the Müller-Lyer illusion, the normative response is to judge the two arrows to be equal in length, whereas a descriptively common response is to judge one to be longer than the other. In Heukelom’s version, the shift was another important precondition for behavioral economics, in which orthodox choice theory is frequently used as a normative standard and benchmark with which observed behavior is compared and assessed.
Heukelom’s narrative is interestingly different from another one that has gained some traction, according to which behavioral economics represents a cognitive revolution in economics (Angner and Loewenstein Reference Angner, Loewenstein and Mäki2012). In this narrative, the emergence of behavioral economics is due to a shift away from a broadly speaking positivist–behaviorist–operationalist methodology, according to which references to unobservable mental states are scientifically illegitimate and at any rate unnecessary, and toward a methodology according to which references to cognitive and affective states, rules, and heuristics, etc., are both legitimate and necessary to account for economic behavior. In this story, behavioral economists share neoclassical economists’ conception of their subject matter but believe that “opening the black box of the mind,” as the cognitive scientists say, is necessary for economics to succeed on its own terms.
The two narratives, however, are compatible and indeed mutually supportive. Heukelom argues that the epistemology of generalized characterizations disappeared in large part under pressure from logical positivism and its insistence that meaningful propositions must be either tautologous or directly observable, which left no place for generalized characterizations of economic behavior (p. 26). And Heukelom affirms the manner in which the cognitive revolution prepared the ground for behavioral economics, first, by rejecting behaviorist methodological strictures and favoring opening the black box of the mind, and second, by refusing to honor traditional disciplinary boundaries such as that between psychology and economics (pp. 97–98). Together, the two narratives make for a richer fabric that can undergird future debates and serve as a basis for much-needed further work on the history, philosophy, and methodology of behavioral economics.
