By Philip Mirowski, Carl Koch Professor of Economics and the History and Philosophy of Science University of Notre Dame. Professor Mirowski has written numerous books including More Heat than Light, Machine Dreams and, most recently Science-Mart
Edited and with an introduction by Philip Pilkington, a journalist and writer living in Dublin, Ireland
What follows is Mirowski’s account of the behaviourist defence of neoclassicism after the crisis. While behaviourism does not occupy a central position within the discipline it stabalises it in a different way, in that it allows certain neoclassicals who have a nagging feeling that the whole edifice of the research program is based on shaky and unrealistic foundations to think that they have found some new and exciting way of doing research. They then conclude that if they can only get their colleagues to see the light all will be well and the neoclassical research program can continue. In addition to this it allows them to take a sleeping pill with regards to the recent crash; after all, surely it was the result of some sort of irrational behaviour and not due to inherent structural imbalances within a capitalist economy, right?
As Mirowski points out, the behaviourist research program is largely a sideshow – one might be tempted to say: a sideshow for left-wing economists making excuses. It is a sideshow precisely because it continues to deploy the neoclassical methodology. What happens is that these well-meaning but misled economists try to take into account varying degrees of ‘irrational’ human behaviour, but in doing them they sterilise them by forcing them into their rationalising frameworks. Mirowski sums this up as such: “Wasn’t ‘irrationality’ in the neoclassical lexicon an oxymoron, since the moment one formalized it in the utility function, didn’t it effectively get subsumed under some perverse version of meta-rationality?”
And so the paradigm is defended once more, this time from within; as those who began to suspect that the neoclassical program is wrong to strip individuals of their psychologies sought for a safe alternative – but in doing so their neoclassical models always get the better of them and sap any purported ‘irrationality’ or individuality out of their research subjects. “What ‘behavioral economics’ fostered was the warm glow of feeling that you had changed your economic stripes without having to change your mind, or your models. Like the Seekers.”
– Philip Pilkington
Part II: Behavioural Economics – Rationalising Irrationality
It has been fairly common in the annals of economic history to observe that in the wake of serious financial crises, observers tended to bewail a weakness in human cognition, attributing pecuniary disaster to a ‘madness of crowds’. Eventually, eschewing structural explanations, all serious problems would conventionally tend to be traced back through intermediate ‘bad choices’ to moral or character flaws in particular individuals. Thus, it was no surprise that a pervasive and immediate response to 2008 was to blame the entire mess on a rabid outbreak of ‘irrational exuberance’.
Unfortunately, madness often lodged in the eye of the beholder. When it came to orthodox economics, the term ‘rationality’ bore a very narrow and curious interpretation as the maximization of a utility function subject to constraints by an otherwise cognitively thin and emotionally deprived ‘agent’. This unsatisfactory version of rationality had been the perennial subject of complaint and criticism from within and without economics since the 1870s; numerous defenses had been developed over the decades supposedly to neutralize those concerns. Hence, in economics, repudiation of ‘rationality’ meant in practice tinkering with the utility function and/or its maximization. The latest attempt at accommodation dated from the 1990s, introducing some amendments from narrow subsets of psychology (mostly decision theory) while keeping the basic maximization of utility framework intact: this had come to be called ‘behavioral economics’. This purported enrichment of simpler concepts of rationality had even established a beachhead in the study of finance well before the crisis; its most prominent advocate in the prelapsarian era was Lawrence Summers, which might begin to signal that its revolutionary potential may not have been all that transformative. Mostly, in finance it produced models predicated upon the posited existence of a complement of stupid people, somewhat more charitably known as ‘noise traders’, who performed certain functions in financial markets (liquidity, smoothing of reactions to shocks) so that the neoclassically ‘rational’ agents could more readily find the ‘true’ or ‘fundamental’ values dictated by the prior orthodox theory. Nothing here substantially impugned the basic orthodox model.
Once the crisis hit, journalists predictably turned to accusing Wall Street of behaving irrationally (in the vernacular meaning), and economists of investing too much credence in the rationality of their agents. Two bestselling books were especially effective in broadcasting this line: John Cassidy’s How Markets Fail (2009) and Justin Fox’s Myth of the Rational Market (2009). Some economists who had been strong advocates of behavioral approaches prior to the crash, George Akerlof and Robert Shiller (2009) and Robert Frank (2009), leapt in with op-eds essentially blaming the entire crisis on cognitive weaknesses of market participants. This line became entrenched with the appearance of George Akerlof and Robert Shiller’s Animal Spirits (2009a): displaying an utter contempt for the history of economic thought, they ‘reduced’ the message of Keynes’s General Theory to the proposition that people get a little irrational from time to time, and thus push the system away from full neoclassical general equilibrium.  They wrote:
The idea that economic crises, like the current financial and housing crisis, are mainly caused by changing thought patterns goes against standard economic thinking. But current crisis bears witness to the role of such changes in thinking. It was caused precisely by our changing confidence, temptations, envy, resentment, and illusions . . . (2009a, p. 4)
The timing was propitious, since at that juncture all sorts of people were casting about for something different in the way of economic analysis of the crisis. Nevertheless, when a few journalists read the book, the first thing they noticed was that it said very little substantive about the current crisis, for much of it had been written well before 2008. Brought up short, they began to doubt its pertinence. The second thing they noticed was it was full of overweening claims, but contained very little in the way of causal mechanisms. ‘Animal spirits’ boiled down to such timeworn neo- classical expedients as changing the utility function over time and calling it ‘confidence’ (while Chicago called it ‘time-varying rates of discount’), appealing to sticky wages and prices while attributing them to ‘money illusion’ and concerns over fairness (which the neoliberals modelled as ‘envy’), and suggestions that corruption would grow over the course of a long expansion (Chicago theorist Gary Becker theorized this the ‘rational choice approach to crime’). Far from some brave venturesome foray into the unexplored thickets of real psychology by open-minded economists unencumbered by entrenched dogmas, this was just more of the same old trick of tinkering with the ‘normal’ utility function to get out the results you had wanted beforehand – something falling well short of the trumpeted dramatic divergence from standard economic theory.
This raised an objection that had long been a subject of discussion in the methodology literature: wasn’t ‘irrationality’ in the neoclassical lexicon an oxymoron, since the moment one formalized it in the utility function, didn’t it effectively get subsumed under some perverse version of meta-rationality? Richard Posner (2009b), an especially perceptive critic from the Right, pushed this point home in a review of Animal Spirits. The Akerlof–Shiller reply, deficient in philosophical sophistication, proved unable to confront this debility:
When Posner asserts that it is not always easy to rule out that people are acting rationally – even if they seem not to be – he is of course right, for this is what most academic economists have thought. It is hard to disprove such a theory that people are completely economically rational because the theory is somewhat slippery: It doesn’t specify what objectives people have or what their information really is. (Karloff and Shiller, 2009b)
The problem with behavioral economists going gaga over ‘irrationality’ was that they conflated that incredibly complex and tortured phenomenon with minor divergences from their own overly rigid construct of pure deterministic maximization of an independent invariant ‘well-behaved’ utility function. Akerlof and Shiller could only condone an incongruously rationalist framing of their irrational exuberance. One can therefore appreciate misgivings that this core theory was so ‘slippery’ that it is not at all evident what the enthusiasm over ‘behavioral economics’ really amounted to. Two decades of behavioral research certainly has not resulted in any consensus systematic revisions of microeconomics, much less macroeconomics. Beyond wishful thinking, why should one even think that the appropriate way to approach a macroeconomic crisis was through some arbitrary set of folk psychological mental categories? Again, they had to admit that Posner had caught them putting the rabbit into the hat:
Posner makes the interesting point that most behavioral economists – who study the application of psychology to economics – did not predict the economic crisis either. We would put this somewhat differently: There were very few behavioral economists who made forceful public statements that a crisis may be imminent. That is because there are very few behavioral economists who even specialized in macroeconomics, and so virtually none was willing to take the risk of making any definitive forecast. (Akerlof and Shiller, 2009b)
The plea that in the eventuality that behavioral economics had had more adherents, it would have done more of the things Akerlof promised, is hardly a compelling reason to get enthused about a line of research. Akerlof and Shiller were loathing admitting that they had not proffered any good conceptual reasons to believe that behavioral economics was even particularly relevant to the crisis. What this literature had to do with the genesis of credit default swaps, the rise of the shadow banking sector and the collapse of the manufacturing sector was entirely opaque. And however much Akerlof and Shiller protested that their politics was diametrically opposed to neoliberals like Reagan and Bush, what was their version of ‘animal spirits’ but tantamount to simply blaming the victims for the macroeconomic contraction? Shiller’s previous books did indeed identify the housing bubble as a problem, but his ‘solutions’ always involved even more baroque securitizations of the assets in question (Shiller, 2006). In this, he rivalled the most avid neoliberal in his belief in the superior power of The Market to fix any problem.
The unbearable lightness of Akerlof’s behavioral theory is nicely exemplified by the one paper that was repeatedly cited by bloggers and journalists during the crisis, his ‘Looting: The Economic Underworld of Bankruptcy for Profit’ (Akerlof and Romer, 1993). From reading the title, one would suspect it might deal with the phenomenon of running a financial institution into the ground given the temptations of short-term trading profits; like, say, Bear Stearns or Lehman Brothers. Few of its enthusiasts actually bothered go so far as to read the model, however. In the MIT tradition, it was a purely deterministic little ‘toy’ model of a single firm over three periods, where assets are not bought or sold after the first period, and a little maximization exercise which argues that if the owners of the firm could pay themselves more than the firm is worth and then declare bankruptcy in period three, then they will do so. Accounting manipulation and regulatory forbearance (which were not described in any level of detail) are asserted to make this outcome more likely. Deposit insurance permits owners to offload costs of auto-destruction onto the government. This was then asserted to ‘explain’ the savings and loan crisis of the 1980s.
This displays all the hallmarks of the behavioral program touted by Akerlof and Shiller. First, a reputedly irrational behaviour (looting banks by their owners) is rendered ‘rational’ through minor amendment of a simple orthodox maximization exercise by tinkering with the utility function of bank owners. Insights from professional psychology are absent. Macroeconomic phenomena are reduced to isolated individual choice in a manner far less sophisticated than in the reductionist rational expectations movement. Predictably, the model adds nothing to what simple intuition would suggest, given that the problem has been artificially restricted to a rudimentary cost–benefit exercise beforehand. Indeed, the model does not particularly illuminate the situation that nominally inspired it, since it does not encompass any of the specific institutional detail pertinent to the phenomenon; that is, it bypasses what precipitated the crisis at that particular juncture. It ignores the breakdown of Depression-era walls between depository and investment institutions, and neglects the spread of baroque securitization at the behest of finance economists, and the watershed of the ‘originate and unload’ business model for retail loans. But more to the point, their supposedly left-wing approach ends up backhandedly reproducing the conventional neoliberal story, as was pointed out by Gregory Mankiw in his published commentary:
Although the two authors from Berkeley did not intend this paper to be a defense of Ronald Reagan and his view of government, one can easily interpret it in this way. The paper shows that the savings and loan crisis was not the result of unregulated markets, but of overregulated ones . . . The policy that led to the savings and loan crisis is, according to these authors, deposit insurance. (Akerlof and Romer, 1993, p. 65)
Hence, when some economists speculated that the orthodoxy would give way to a ‘more realistic’ behavioral economics in reaction to the crisis, it was primarily a symptom of the general unwillingness to entertain any serious departure from conventional arguments.  If anyone had bothered actually to read any of the leaders of the behavioral ‘movement’, they would have quickly realized that those economists went out of their way to renounce any ambitions to displace the orthodoxy. In one spectacularly badly timed compromise, Andrew Lo had sought to reconcile the findings of behavioral finance with the efficient-markets hypothesis (Lo, 2005). And behavioral economists could care less about the layered complexity of the human soul. Indeed, one need not look far to encounter their contempt for the academic psychology profession:
I think we strive for parsimonious, rigorous theoretical explanations; this distinguishes us from the psychologists . . . Economists want a theory that provides a unifying explanation of these results whereas psychologists are much more willing to accept two different theories to explain these ‘contradictory’ results. . . I don’t like the argument that everything is context dependent. That view lacks any grounding. In this regard, I really like the strong theoretical emphasis of economics and our desire for unifying explanations. It distinguishes us a lot from biologists and psychologists, and provides us with a normative anchor. (Fehr in Rosser et al., 2010, pp. 72–73)
If you asked behavioralists what all this tinkering with conventional neoclassical utility functions (which dated back to the very inception of the program in the 1880s) was supposed to portend, they would tell you in no uncertain terms that so-called behavioral economics ‘is based not on a proposed paradigm shift in the basic approach of our field, but rather is a natural broadening of the field of economics . . . [it is] built on the premise that not only mainstream methods are great, but so too are mainstream economic assumptions’ (Rabin, 2002, p. 659). 
So wherever did the vast bulk of commentators get the unfounded impression that behavioral economics was poised to deliver us from the previous errors of orthodoxy when it came to the economic crisis? Partly, it was the fault of a few high-profile economists like Shiller, Akerlof and Krugman, whose own ‘behavioral’ credentials within the community were, shall we say, a bit shaky. But it also emanated from the vast scrum of journalists, primed to believe that when economists would just abjure ‘rational choice theories’, then all would become revealed. It got so bad that two bona fide behavioralists felt impelled to pen an op-ed for the New York Times absolving them of any responsibility to explain the crisis:
It seems that every week a new book or major newspaper article appears showing that irrational decision-making helped cause the housing bubble… It’s becoming clear that behavioral economics is being asked to solve problems it wasn’t meant to address … Behavioral economics should complement, not substitute for, more substantive economic interventions [of] traditional economics. (Loewenstein and Ubel, 2010)
Ultimately, the more perceptive journalists acknowledged this: ‘While behaviorists and other critics poked a lot of holes in the edifice of rational market finance, they haven’t been willing to abandon that edifice’ (Fox, 2009, p. 301). It is not even clear that they have been all that willing to bring themselves to look out the window. What ‘behavioral economics’ fostered was the warm glow of feeling that you had changed your economic stripes without having to change your mind, or your models. Like the Seekers.
11. Although calling themselves ‘Keynesians’, their understanding of what Keynes wrote was so tenuous that they were called to account in this regard by Posner (2009b) and at http://dmarionuti-blogspot.com/2009/09/akerlof-shiller-animal-spirits-misnomer.html.
12. ‘What’s probably going to happen now – in fact, it’s already happening – is that flaws- and-frictions economics will move from the periphery of economic analysis to its center. There’s a fairly well developed example of the kind of economics I have in mind: the school of thought known as behavioral finance’ (Krugman, 2009).
13. Yet even this divergence went too far for the Old Guard of the orthodoxy (Arrow in Clarke, 2009).