How One Foundation Created Behavioral Economics

Standard

A Sense of Mission: The Alfred P. Sloan and
Russell Sage Foundation’s Behavioral Economics Program (1984-1992)

Floris Heukelom
Assistant Professor, Department of Economics
Radboud University Nijmegen
Netherlands

© 2010 by Floris Heukelom
f.heukelom@fm.ru.nl

Between April 5th and April 9th 2010 I had the opportunity to investigate at the Rockefeller Archive Center the Russell Sage Foundation’s files on the Behavioral Economics Program. The research has been tremendously helpful and will result in an article and find a place in my forthcoming book on the history of behavioral economics. Behavioral economic program recipient Richard Thaler (1945 -) described the role of the program as follows:

“[In 1983-84] behavioral economics essentially did not exist. Kahneman and Tversky had written their Econometrica paper on prospect theory, but not very many economists had taken much notice. There was no dialogue between psychology and economists, and there hadn’t been since Herb Simon’s days as an economist. Now much of the credit for what has happened since must go to K&T who were so brilliant that economists simply could not ignore them. But I think that the value of the Sloan-Sage program should not be neglected. Simply by having such a program, a sense of mission was created”
In 1979, Albert Rees became the president of the Sloan Foundation, which at the time had $250 million in assets and gave $15 million in grants a year. Three years later, Rees managed to persuade Eric Wanner to join the Sloan Foundation as a program officer to take care of Sloan’s cognitive science program. Wanner also set out to look for something new to do. His first proposal was a new program on decision theory. The idea was straightforward enough. The Sloan Foundation had had a program on neuroscience in the 1960s and on cognitive science in the 1970s and early 1980s, so a natural continuation would be the application of cognitive science to decision making. But Rees judged Wanner’s plans too ambitious, not in the least because Wanner proposed to spend some $20 million on the project. Thus Wanner had to look for something cheaper and more confined.

Wanner’s next plan was to concentrate on the application of cognitive science to economics. The new plan gradually sharpened during 1982 and 1983, early 1983 Wanner mentioned as “a possible area for investment” for the Sloan Foundation the theme of “what might be called the psychological foundations of economic behavior” (Wanner’s letter to Fischoff, 26 January, 1983, RAC). During a conversation between Wanner, Daniel Kahneman and Amos Tversky in which they explored the topic, the two star psychologists were not very optimistic, reasoning that to get economists’ attention psychologists would have to be more economically sophisticated than they actually were, and advised Wanner not to spend too much money on the project, if anything at all (see also Kahneman, 2002). Nevertheless, over the course of 1983 “behavioral economics” as the name for the new program emerged and Wanner planned a few exploratory meetings with Kahneman and Tversky, and through them with Thaler. As a result, the word was spread in the academic community that the Sloan Foundation was considering setting up a new program on behavioral economics, and during the second half of 1983 the first unsolicited grant requests started coming in. The only application that was considered and funded in 1983 was a proposal by Thaler to spend a sabbatical with Kahneman at the University of British Columbia. However, after the program officially commenced the other early proposals were reviewed and decided upon. Throughout this whole process, Wanner worked closely with Rees on the new program, who as an empirically-oriented Marshallian Chicago economist was skeptical but tolerant towards the idea that an empirically-grounded psychology could be usefully employed in economics.

Rees and Wanner started by composing an advisory committee to the program that would review the incoming proposals and award the funds available. To avoid the possibility that one of the two disciplines that the program aimed to bring together would dominate the other, and to ensure that it would be an interdisciplinary program, Wanner and Rees decided that there should be two psychologists and two economists on the advisory committee. Also the label “behavioral economic” was understood to be deliberately half psychology-half economics. The first person they picked for the advisory committee was Leon Festinger (1919 – 1989), a social psychologist and colleague of Katona in the Institute of Social Research at the University of Michigan in the 1950s. In addition, Festinger had been involved with the Sloan Foundation’s cognitive science program and was interested in economics. Second, Wanner approached economist Thomas Schelling (1921 -), whom Wanner knew from Harvard and whose work on paradoxes and conflict strategies seemed related to the new program. Rees recommended economist William Baumol (1922 -), an organization theorist and early critic of von Neumann and Morgenstern’s axiomatic approach to decision making. The fourth member who Wanner and Rees agreed on, and who accepted, was cognitive psychologist Robert Abelson (1928-2005), a member of the advisory board of the Cognitive Science Series who Wanner had initiated at Harvard University Press.

In May 1984, the Sloan Foundation supported a conference at Princeton University on the behavioral assumptions of economic theory. During the conference, Wanner had a discussion over dinner with the prospective members of the advisory committee as well as with Kahneman, Tversky and Thaler on who might be the appropriate researchers to invite for the new behavioral economics program. After the dinner, Kahneman, Tversky and Thaler discussed possible names in more detail and individually made more or less similar suggestions to Wanner. Then, in mid-June 1984, the board of trustees of the Sloan Foundation officially installed the advisory committee and endowed it with $250,000 to fund a number of “seed projects” in subsequent years, to see if the program could work (Wanner’s notes on the advisory committee meeting, 7 December, 1984, RAC).[1] The amount of money spent on the behavioral economics program by the Sloan Foundation was comparatively small, and the board of trustees basically took Rees’ word for it. Thus, while the idea and organization of the new behavioral economics program was entirely Wanner’s, the skeptical but tolerant support of Rees in the background was crucial.

As early as July 1984, Abelson expressed a view that seems to have been shared by the other advisory committee members as well as by Wanner, namely that Kahneman and Thaler should be at the center of the new program: “Getting Thaler and Kahneman together is bound to produce progress. Their teamwork could be as seminal as the Tversky and Kahneman pairing, but more market oriented” (Abelson’s letter to Wanner, 26 July, 1984, RAC). The first behavioral economics meeting was planned for 7 December, 1984 at the Waldorf-Astoria Hotel, New York. In addition to the advisory committee and Kahneman and Thaler the following economists and psychologists were invited: Hillel Einhorn, Baruch Fischoff, Donald Hood, Thomas Juster, Charles Plott, Howard Kunreuther, Howard Raiffa, Herbert Simon, Oliver Willisamson, and Richard Zeckhauser.[2]

The program’s organizers stressed that “for the [11 October, 1985] meeting to be most productive, we cannot afford to get too bogged down in discussions of whether or not a particular empirical finding is or is not an anomaly, i.e. whether there exists some explanation within the rational, maximizing, economic paradigm. Rather, we should try to work toward an evaluation of competing explanations, and the evidence that might be used to discriminate between behavioral hypotheses” (ibid). In other words, the program’s organizers actively tried to prevent a theoretical, economic discussion of neoclassical economic theory, and in advance tried to steer the discussions towards behavioral terms.

The first real test of the new behavioral economics program was a conference on “The Behavioral Foundations of Economic Theory,” 10-15 June, 1986, organized by Robyn Hogarth and Melvin Reder at the University of Chicago.[3] The conference did not go very well for the members of the Sloan behavioral economics program. In the meantime, the number of researchers invited to participate in the behavioral economics program had expanded to forty names in December 1985, including George Akerloff, Kenneth Arrow, Robert Frank, David Grether, Robyn Hogarth, George Loewenstein, Mark Machina, James March, Richard Nelson, Charles Plott, Howard Raiffa, Robert Shiller, Vernon Smith, Lawrence Summers, and Sidney Winter. All forty researchers were told that “[t]he Sloan Foundation has decided to develop a limited funding program in behavioral economics in 1986. The purpose of this letter is to describe the program briefly and to invite you to consider making an application” (Wanner’s letter to forty invited researchers, 18 December, 1985, RAC). [4]

During the same period, Wanner stepped up his efforts to encourage researchers to send in a proposal or to collaborate with one another. Following a short but cordial note from Stanford psychologist, former collaborator of Simon, and Russell Sage Foundation trustee James March – one of the new names on the list of researchers invited – Wanner responded by suggesting March could put together a proposal with fellow Stanford researchers Arrow and Tversky who seemed “interested in making a proposal for support for graduate or post doctoral students” (Wanner’s letter to March, 10 January, 1986, RAC). Quickly thereafter, March became involved with the behavioral economics program in another way as well. After being promoted to the position of Vice-President of the Sloan Foundation in the summer of 1985, Wanner was appointed as a trustee and president of the Russell Sage Foundation in the summer of 1986. He started officially on October 1st, 1986. Wanner hinted to the behavioral economics advisory committee that he suspected a positive recommendation by some of them might have had something to do with this appointment: “Some of you may even have had a hand in advising Russell Sage to talk this rash step, in which case I am both in awe of your persuasive powers, and extremely grateful for them” (Wanner’s letter to the advisory committee, 21 August, 1986, RAC). At the same time, Wanner realized that this would probably mean that “this program will not be continued at Sloan after I leave” (Wanner’s letter to March, 22 July, 1986, RAC) and organized a meeting with March to discuss the possibility of “transplanting some version of it [the behavioral economics program] to Russell Sage” (ibid). March quickly responded that he “was delighted that you [Wanner] started it [the behavioral economics program] and would be equally delighted if we continued it at the Russell Sage” (March’s letter to Wanner, 29 July, 1986, RAC). [5]

After some negotiation, Wanner could report to the advisory committee that “the RSF Board approved the idea of a joint program with the Sloan Foundation” (Wanner’s letter to the advisory committee, 18 February, 1987, RAC). However, there were two conditions, or at least strong suggestions. First, the “RSF would be particularly interested in supporting an extension of the program which includes sociologists, organization theorists, and other social scientists, in addition to the nexus of cognitive psychologists and economists so far funded in the program.” (Wanner’s letter to the advisory committee, 18 February, 1987, RAC). Second, the board of trustees of the Russell Sage Foundation wanted some influence in the behavioral economics advisory committee. Wanner set out to meet both requirements.

The first requirement proved to be difficult. Wanner suggested to the advisory committee that “there is still quite a lot of work to be done on this border between psychology and economics. But one might take a different view. For instance, I enclose (agnostically) a paper of Amitai Etzioni’s which outlines his vision of what he calls social-economics” (Wanner’s letter to the behavioral economics advisory committee, 16 October, 1986, RAC). The advisory committee was not impressed. Baumol, for instance, replied that “I would indeed be prepared to entertain proposals by psychologists or economists that call for participation of persons from other social science disciplines. However, I think it would broaden the program excessively to invite proposals from sources outside our two focal disciplines” (Baumol’s letter to Wanner, 17 November, 1986, RAC). That effectively ended the attempt to broaden the program to include socio-economics.

The second requirement, on the other hand, was quickly met. The board of trustees of the Russell Sage Foundation suggested that one of its members, March, could be on the advisory committee. Wanner and the advisory committee were in favor but also feared that this would tip the balance too much to psychology.[6] Thus, in a letter dated 17 February Wanner and Rees (still president of the Sloan Foundation) invited March to join the advisory committee, which invitation March accepted, while at the same time Rees, as a labor economist, was asked by the advisory committee to join its ranks. As a result, the advisory committee now consisted of six members, three psychologists and three economists. It was headed by the president of the Russell Sage Foundation and had on its advisory committee both the president of the Alfred Sloan Foundation and a trustee of the Russell Sage Foundation.

Having Rees on advisory committee in particular seems to have helped to ensure at least another few years of Sloan support for the program – thus positively defying Wanner’s earlier fears of the end of the behavioral economics program at the Sloan Foundation as expressed in his letter to March. With the two new members on the advisory committee, the behavioral economics program proceeded, now supported by two foundations. In addition to the twelve grants from the Sloan Foundation in 1986 already mentioned, the behavioral economics program issued its first support from the Russell Sage Foundation in 1986 in the form of a $200,000 grant to Kahneman and Tversky for a new book on “Decisions: Rationality and Illusion in Judgment and Choice.”[7] From the start, however, the behavioral economics program had a different status at Russell Sage as compared to Sloan, and not only because Wanner had become the president of Russell Sage. At Sloan, the board of trustees simply took President Rees’ word for the small behavioral economics program. The amount of grants awarded to the behavioral economics program never exceeded 2.7% of the total amount of grants issued. At Russell Sage, however, the behavioral economics program consumed a much larger portion of annual spending, up to nearly 50 percent in 1987.

Yet, the future of the behavioral economics program was by no means self-evident. To give the program more focus, and thereby to encourage the participating researchers to provide more concrete results, Wanner suggested that the advisory committee set up a few working groups at Russell Sage “which focus on a particular topic” (Wanner’s letter to the advisory committee, 16 October, 1986, RAC). Eventually, three non-residential working groups emerged over the course of 1987 and 1988, which would come to define the core of research that ascended to prominence in the economic discipline in the 1990s and 2000s (Heukelom, 2010). The working group on “Intertemporal choice” was led by Loewenstein and Jon Elster and in their Choice over Time (1992), among other publications. The working group on “behavioral approaches to financial markets” was headed by Thaler and Robert Shiller (1946- ), which provided input for Shiller’s Market Volatility (1989) and Irrational Exuberance (2000). The proposed working group on experimental economics, however, proved more difficult to organize. Initially, the idea was to have Smith or Plott lead or co-lead a working group on experimental economics together with a behavioral psychologist like Kahneman, or one of his associates. That did not work out however, because of a different theoretical interest. Smith and Plott wanted to concentrate on the question how the market eventually steers individual behavior towards rational equilibrium, and what the equilibrium exactly looks like. Wanner, Kahneman, Thaler and the advisory committee, on the other hand, were more interested in how initial individual behavior deviates from the theoretically defined equilibrium, irrespective of whether it exists or not. In addition, Wanner, Kahneman, and Thaler questioned how often economic markets are allowed the time to mature towards equilibrium. Thus, the plan was abandoned and in the end Colin Camerer was put in charge of the working group on experimental economics. The advisory committee and in particular Wanner were thus closely monitoring the content and development of the behavioral economics program. The focus and inclusion of disciplines in the “portfolio” of grants was carefully managed and Wanner was constantly looking for new researchers who might submit a proposal that could fit into the program. In early 1987 the 1986 Sloan recipients were asked by Wanner and Rees to provide reports of the work they had done and were offered a chance to submit a proposal for a continuation of their work.

As a result of the growing visibility of the program, and Wanner’s acquisitioning of proposals, the amount of those submitted steadily increased. To relieve the advisory committee of some of its reviewing work, in 1987 Wanner started to ask external researchers to review some of the proposals submitted. However, the process in which the advisory committee would convene once in a while to make decisions on the proposals remained intact throughout, and the advisory committee was not bound by the referee reports. To further support the advisory committee in its decision making, Thaler was invited to join for a part of the advisory committee meeting of 27 April, 1987, and again a few times afterwards. In addition, the reviewing was now divided among the advisory committee members, so that not all of them had to read all the proposals. During 1988 and 1989 the behavioral economics program continued without any major changes along the lines developed in the years 1983-1987 and supported by two foundations. In early March of 1988, Wanner suggested that the recently appointed Russell Sage Foundation trustee and economist Raiffa could join the advisory committee on an ad hoc basis, which was accepted by the other advisory committee members. In addition to Thaler and Raiffa, other researchers, including Kahneman, were sometimes asked to join the advisory committee meetings for one or a few sessions. In 1989, Festinger, Rees and March stepped down as advisory committee members, and were replaced by Kahneman. The advisory committee now consisted of economists Baumol and Schelling and psychologists Abelson and Kahneman.

The retirement of Rees because of age and deteriorating health had by far the most impact, because as anticipated it implied the end of the Sloan Foundation’s support of the behavioral economics program. Thus, as of 1990, the behavioral economics program would continue as only a Russell Sage program. At the same time the number of grant proposals submitted to the behavioral economics program continued to grow, up to the point in the second half of 1989 where Wanner and the advisory committee felt it became increasingly difficult to carefully review all the incoming proposals. This growth in the number of proposals reflected an increasing awareness of the behavioral economics program among psychologists and economists, but, more importantly, also reflected a growing popularity of behavioral economics research. In order to work out how the program should continue now that it was supported only by Russell Sage and in order to reduce the number of grant proposals, Wanner and the advisory committee decided to put 1990 inthe “wait and see mode.” No official calls for proposals would be issued, but everything that came in would be reviewed.

But all good things come to an end. In early 1992, the Russell Sage board of trustees told Wanner it planned to end the behavioral economics program and to seek other purposes for the roughly 30% of the annual budget the program was consuming. Although Wanner judged this a fair point, he explored the possibility of saving part of the program in one way or another. First he asked all researchers who had received grants or other support under the behavioral economics program between 1984 and 1992 to briefly express to the board of trustees “whatever effect the program may have had on your own research and to “offer an appraisal of the current state of behavioral research in economics” (Wanner’s letter to recipients, 30 April, 1992). Second, Wanner asked the recipients to address “the general prospects for future work in behavioral economics” (ibid). The letter was sent to some ninety economists and psychologists. About a third of the recipients responded, among them of course those researchers most closely involved, such as Thaler, Kahneman, Loewenstein and Camerer. It will be no surprise that their responses were positive, as exemplified by the quote from Thaler that started this article. Others, however, were less positive, for instance, in a long a letter to Wanner, Vernon Smith severely criticized both the organization of the behavioral economics program and the research on which it had focused (Smith’s letter to Wanner, 15 May, 1992).

In the summer of 1992, the behavioral economics program officially closed and the advisory committee was disbanded. Together, however, the letters of the Grant recipients convinced the board of trustees to agree to one last form of financial support through what would be the organizational novelty of a “Behavioral Economics Roundtable,” composed of former recipients of behavioral economics grants and endowed with $100,000 annually. To the Russell Sage Foundation, the main advantage of this organizational novelty was that its staff no longer would be involved in behavioral economics and that an advisory committee would not be required. The board of trustees and Wanner agreed that the first ten members of the Behavioral Economics Roundtable would be elected by the former grant recipients or selected by the Russell Sage Foundation. Thus, in August 1992, Wanner again wrote the ninety recipients, asking them to cast their votes for the first members of the Behavioral Economic Roundtable. Two months later, the first ten members installed on the Behavioral Economic Roundtable were Akerlof, Blinder, Camerer, Elster, Kahneman, Loewenstein, Schelling, Shiller, Thaler, and Tversky. Since its creation in 1992, the Behavioral Economic Roundtable has been an effective promoter of behavioral economics through its bi-annual summer institute and its support of young researchers through a small-grants program. The Behavioral Economics Roundtable still exists at the time of writing.

How Reason Became Rationality

Lorraine Daston
In the second half of the twentieth century, in the works of philosophers, mathematicians, engineers, and social scientists, the faculty of reason was radically reconceived. In the models of game theory, decision theory, artificial intelligence, and military strategy, the algorithmic rules of rationality replaced the self-critical judgments of reason. The reverberations of this shift from reason to rationality still echo in contemporary debates over human nature, planning and policy, and, especially, the direction of the human sciences.

The background to this transformation lies in the history of rules. In major European languages from the early Middle Ages until the mid-nineteenth century, the chief sense of regula /rule/Regel/règle derived from Regula Sancti Benedicti (6th c. C.E.) laid down for Christian monks and referred to a moral precept, a model or code of conduct for a way of life. This sense is still vivid in Enlightenment discussions of whether art should follow rules or whether it is (in Kant’s words) genius that “gives the rule [i.e. model] to art”. In the early decades of the nineteenth century, a tertiary meaning of rules as algorithms (first and foremost the rules of arithmetic) begins its rise to dominance, in contexts as diverse as the first effective calculating machines, the creation of national civil service bureaucracies, and attempts to guarantee the logical solidity of mathematical proofs.

By the early 1950s, the dream of reducing intelligence, decision-making, strategic planning, and reason itself to algorithmic rules had spread like wildfire to psychology, economics, political theory, sociology, and even philosophy. The story of how reason became rule-bound is one of Cold War ambitions, technological fantasies come true (especially in the realm of computers), quantification and quantiphrenia, and, above all, the eclipse of the faculty of judgment as an essential component of reason.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s