The Dull Compulsion of the Economic (x)
A series of blog postings by D&S collective member Larry Peterson
‘Life Is Beautiful’ Economics and the Strange Co-option of Behavioral Economics
They’re at it again: encouraged by the huge rally in equities globally (especially in emerging markets, some of which were considered doomed just a few weeks ago), more and more financial commentators are referring to behavioral economics in an attempt to indicate that the rally may portend a vigorous economic recovery. The latest instance came in Friday’s Financial Times, in which US editor Chrystia Freeland joined the chorus with an article entitled “What a Feeling: How Emotions May Yet Drive the Recovery.”
I’ve never been a fan of Freeland’s: I think she’s easily the most mediocre of the FT‘s senior staff, and I usually just ignore her usual bland regurgitation of cliches and conventional wisdom. But seeing her refer to behavioral economics really gets me hot and bothered. When I was an undergraduate, I successfully studied graduate-level neuroscience, and have followed developments in what is called behavioral economics (which relies on studies from cognitive psychology and, to a lesser degree, neuroscience) almost since its inception in the 1980s. Accordingly, I think I have a sensibility for both the power and deficiencies of the findings that have been taken up by the new field of behavioral economics which Freeland and a lot of other erstwhile enthusiasts almost certainly don’t have. And that’s just the start of it. Even recognizing the potential of behavioral economics, I would argue that its use in coming to terms with things like the financial crisis is misguided at best, and ideologically dangerous at worst. Dangerous because I believe a case can be made that it is being used by a discredited profession not only to rehabilitate itself,but to smuggle some of its old, worn-out assumptions in via the back door, dressed up in the garb of cutting-edge science.
Let me try to explain. Behavioral economics is a movement that relies on controlled studies to test the validity of economic assumptions that had been–and still are, to some degree–considered axiomatic by conventional practitioners. Usually the experiments take the form of the playing of games: voluntary participants trade tokens or even, in certain cases, real money in simulated exchanges and so on. So, in a celebrated case, experimenters found that subjects overwhelmingly punish free-riders in such exchanges, even when they stand to gain from cooperating. The upshot of such experiments has been the promotion of a new idea of economic rationality, “bounded rationality”, which views rationality operating within limits set by norms and other “non-economic” criteria to a significant extent; accordingly, behavioral economists say that the traditional notion of rationality employed by the economics profession must be changed to reflect this, if only because such recognition will allow for greater predictive power in both experiments and even in the construction of economic models.
This is important because the most recent paradigm that employed widespread popularity in the profession has clearly fallen out of favor. That paradigm, called “rational expectations”, was popular with conventional economists precisely because it seemed to solve what they considered to be a major problem implicit within Keynesianism: the fact that Keynesianism didn’t have an underlying microeconomics to support its macroeconomic superstructure. And for economists of this sort, who largely continued to accept some version of Say’s law (which Keynes rejected, and stated basically that all production is ultimately undertaken to support consumption, rather than, say, profit-taking, hoarding, or accumulation), Keynes’ focus on insufficient aggregate demand was a heresy–it couldn’t, to them, sufficiently (i.e. in a way that foregrounded the practitioners’ very specific notion of rationality) explain the motives of the individual actors whose choices led to outcomes in which full employment could only be considered a special case; and inasmuch as it did, by referring to liquidity preference and speculation, that could only be reconciled with a vigorous notion of rational self-interest with some difficulty. But it wasn’t until the failure of what were considered Keynesian policies in the ‘seventies that they found an opening to plug this gap. So, first they criticized the failure of Keynesian macroeconomics, and then proposed a new microeconomics that, it was claimed, could restore the link between good, old-fashioned self-interested rationality and the wonderful workings of a whole market system that equilibrated to full employment after all.
This paradigm, which, as many know, was employed originally by members of the Thatcher and Reagan administrations, became renowned for its intransigence: its practitioners stubbornly resisted contributions from economists professing other views, and purged many academic departments, think-tanks and so on of holdouts. And as Thatcherism and Reaganism morphed into Clintonism and Blairism, the situation only got worse. Except on one front: behavioral economics.
Why was this? I think the original encounter with it came because economics, which was being aggressively promoted as a science especially by neoliberal technocrats, but also by financiers, academics and journalists, was considered by the dominant practitioners to have to be as open to what, after all, appeared to be the incontrovertible findings of science–especially sciences of the “harder” variety, like neuroscience–as possible. So, economics slowly, kicking and screaming, opened the door to practitioners of behavioral economics just a crack. Also, as the millennium approached, the collapse of Long Term Capital Management, the Asian crisis, and, eventually, the dot.com meltdown finally revealed all too clearly that the predictive power of monetarist models was not all that it was cracked up to be; and this had a lot to do with the vast, global expansion of the financial sector, a sector which, in part, and to a limited degree, could be examined by experiments devised by practitioners of behavioral economics in ways that the notions of rational expectations couldn’t.
Regardless of the cogency of my interpretation, there is no doubt now that behavioral economics is on the ascent. Practitioners like Cass Sunstein occupy high places in the Obama administration, and the current financial crisis (along with its hyper-aggressive monetary and fiscal interventions) has turned the retreat of the rational expectations school into something of a rout. Meanwhile, Keynesian ideas are resurfacing all over the place. And many behavioral economists are beginning to see their own discipline as the one that may provide that elixir of a slightly different sort than that sought after by the rational expectations school: of a “grand theory” that might unify a brand of Keynesianism, complete with an implicit acceptance of the generality of sub-optimality, with a new notion of bounded rationality–but a rationality all the same, one that can, however provide material for falsifiable experimentation, predictive modeling, and, potentially, optimal policymaking potential. So what’s wrong with that?
Well, my criticism of behavioral economics focuses on the fact that its predominant use of controlled experiments has, by far, simulated the form of consumer exchanges. For ethical and practical reasons, experimental subjects simply cannot be monitored to the same extent in their capacities as workers, employers, union members or scabs, monitors or even slackers, as they can as simple consumers or even investors.
Particularly in cases in which economic behavior is characterized by an asymmetry of real–not simulated–power, it would potentially compromise willing subjects to agree to experimentation, and it’s difficult to see how experiments could be devised that controlled adequately for that asymmetry inn the first place. This is all the more the case where the accounting for the impact of underlying initial endowments of capital–of social or monetary variety–or knowledge are concerned. The types of games and experiments so far devised by behavioral economists–even those that employ the most advanced brain scans and so on–simply are least qualified to illustrate anything but the most limited forms of economic behavior. That being the case, to view the findings of behavioral economics as somehow paradigmatic, never mind ultimately fundamental in a microeconomic sense, constitutes to me an extremely dangerous and misleading position. And it is this position that is increasingly being taken up to “explain” the financial crisis, even by people like Freeland, who probably don’t know very much about the science in the first place (never mind luminaries like George Akerlof and Robert Shiller, who Freeland refers to in her article).
The big problem with all this is that, in the attempt to explain the financial crisis, commentators, following the implicit example set by behavioral economists themselves in many cases, seem quite content to focus on the most limited kind of economic behavior, and thereby bracket out the most important stuff. There is much talk of the transformative power of sentiment, as if the crash itself was an overreaction that could, potentially, be overcome if only government and business are allowed to “nudge“, to use Sunstein’s term (developed with the aid of one of the founders of behavioral economics, Richard Thaler), investors and consumers out of their excessively pessimistic box. This completely ignores the very real weaknesses in the economy that will almost certainly continue to plague working people and pensioners all over the world, even if surviving corporations, due to drastic, or even unprecedented shakeouts (not to mention government support) in their industries, attain some level of profitability acceptable to the ruling and accumulating classes. In fact, these weaknesses are so great that it’s hard to see how they won’t adversely affect large swathes of the latter, and reconfigure that stratum in fundamental ways in the next few years. And economists who privilege explanations based on sentiment over the underlying economic situation, far from elucidating the crisis, may be unwitting tools in shaping it–to the detriment of many people; just as methodological individualism (the idea that economics must focus on individual decision making to attain coherence as a science) was smuggled in via the back door, and even after behavioral economics had played a major role in tearing it down in its fundamentalist form, by the experimental bias of behavioral economics, its practitioners seem all too comfortable with the misuse of limited purview of their findings by politicians and commentators banking on a swift, but highly unlikely recovery.
A couple of years ago a film came out called “Life is Beautiful”, in which, preposterously, perhaps to an offensive degree, an inmate in a concentration camp during the second world war attempted to make the life of a child also interred in the camp bearable by getting it to believe that life in the camp was only a game. Far from providing a new Keynes (and I wouldn’t say that’s the optimal solution), even the most promising school of economics in these harsh times seems to be content with cobbling together narratives devised for something approximating the same purpose.