Akerlof & Shiller, Animal Spirits: A Misnomer for Their Sound Economics

Animal Spirits - How Human Psychology Drives The Economy, and Why It Matters for Global Capitalism, by George A. Akerlof and Robert J. Shiller, was published earlier this year by Princeton University Press, Princeton and Oxford, 2009. It is a timely book, as it addresses the questions of why most economists failed to foresee the current global crisis, to provide explanations for its occurrence and to suggest effective remedies to counteract it. But above all it is a refreshingly original, formidable set of economic propositions, corrosive and at the same time constructive, with pointed and valuable policy implications.

Bob Solow’s book-cover endorsement - “… a sorely needed corrective” - is an understatement. The book should be highly recommended in the reading lists of all social sciences students in every year of their curriculum, and made compulsory reading for government officials, businessmen and anybody operating in credit and financial markets. If I could afford it I would do for the book what a US millionaire is reputed to have done for Joseph Heller’s Catch 22, advertising in the press to give away free copies to the general public.

Akerlof and Shiller claim that “Keynes appreciated that most economic activity results from rational economic motivations - but also that much economic activity is governed by animal spirits” (p. ix). They understand these as “individual feelings, impressions and passions” (p. 1), “a basic mental energy and life force” (p. 3) and “describe five different aspects of animal spirits … confidence, fairness, corruption and antisocial behaviour, money illusion, and stories” (p.5). Confidence changes interact with the state of the economy and amplify disturbances. Concerns about fairness affect the setting of prices and wages. Temptations of corrupt and antisocial behaviour have a significant role in the economy. The public is confused by inflation and deflation and suffers from money illusion. “Finally, our sense of reality, of who we are and what we are doing, is intertwined with the story of our lives and the lives of others. The aggregate of such stories is a national or international story, which itself plays an important role in the economy” (p.6). The book first describes how these five animal spirits affect economic decisions, then argues that they play a crucial role in answering eight crucial questions:

“1. Why do economies fall into depression? 2. Why do central bankers have power over the economy, insofar as they do? 3. Why are there people who can’t find a job? 4. Why is there a trade-off between inflation and unemployment in the long run? 5. Why is saving for the future so arbitrary? 6. Why are financial prices and corporate investments so volatile? 7. Why do real estate markets go through cycles? 8. Why does poverty persist for generations among disadvantaged minorities?”. Moreover, a post-script to Chapter 7 includes their analysis of the current crisis and policy recommendations.

Akerlof and Shiller criticise and de-bunk many conventional economic theories, the foundations of the hyper-liberal tradition associated with the Thatcher and Reagan governments: from rational expectations to the efficient market hypothesis, from the natural rate of unemployment - and the associated denial of a trade-off between unemployment and inflation - to the very notion of voluntary unemployment, from the alleged benefits of de-regulation to the significance of Tobin’s q (the ratio between the current stock exchange valuation of a company’s shares and bonds and the replacement cost of its productive assets: a high q is supposed to promote investment but not always does). We are presented, instead, with waves of optimism and pessimism, manias, euphoria, panics, dishonesty, booms and busts, and the problems of how to put back together again the broken pieces of the financial Humpty-Dumpty.

The conclusion is that “… capitalism can give us the best of all possible worlds, but it does so only on a playing field where the government sets the rules and acts as a referee. Yet we are not really in a crisis for capitalism. We must merely recognise that capitalism must live within certain rules”. “And … in our view capitalism does not just sell people what they really want; it also sells them what they think they want. Especially in financial markets, this leads to excesses…”(p.173).

All very convincing, but for reasons largely different from the ones they offer. Keynes mentioned “animal spirits” as a shorthand for the driving force of entrepreneurship in general and particularly investment. Akerlof and Shiller turn it into a generalised motive that pervades and dominates the whole economy; they dissect the genus into the five species listed above, quite arbitrary and each of them still something of a black box. They identify animal spirits with 1) irrational behaviour and 2) non-economic motives, which is neither necessary nor useful. A powerful critique, but we are left clutching only a few straws. In the end animal spirits become a trite and somewhat irritating cliché, like the fuzzy drawings by Edward Koren supposed to capture them, and of little value added for our understanding of the modern economy.

Irrationality and non-economic motives

The inclusion of irrational behaviour and non-economic motives in macroeconomics is crucial for Akerlof and Shiller: “Picture a square divided into four boxes, denoting motives that are economic or noneconomic or responses that are rational or irrational. The current model fills only the upper left hand box; it answer the question: How does the economy behave if people only have economic motives, and if they respond to them rationally? But that leads immediately to three more questions, corresponding to the three blank boxes: How does the economy behave whith noneconomic motives and rational responses? With economic motives and irrational responses? With noneconomic motives and irrational responses?”.

“We believe that the answers to the most important questions regarding how the macroeconomy behaves and what we ought to do when it misbehaves lie largely (though not exclusively) within those three blank boxes. The goal of this book has been to fill them in” (p. 168).

The question of whether Keynes meant animal spirits to imply irrationality has been the object of a debate, completely ignored by Akerlof and Shiller. R. C. O. Matthews (1984) [“Animal spirits”, Proceedings of the British Academy, 70, 209-229, pay per view] backs the irrationality implication. He argues that Keynes first heard about the term in a lecture in Modern Philosophy on Descartes and other philosophers: in his lecture notes, Keynes commented on animal spirits: "unconscious mental action" (p. 212). Also for Roger Koppl (1991) [“Retrospectives: Animal Spirits”, Journal of Economic Perspectives, 5 , no. 3, 203-210, pay per view] Keynes believed that "the actions induced by animal spirits are irrational." (p. 205). Koppl also conjectures a connection with Descartes for whom, he says, blood that was heated in the heart and transported to the brain could be "animated" and as such make the person "act contrary to their best judgment."

Sheila and Alexander Dow, (1985) [“Rationality and Animal Spirits” in Tony Lawson and Hashem Pesaran, Eds, Keynes' Economics: Methodological Issues] on the contrary claim that explanations based on animal spirits do not imply irrationality: "If evidence is scant for the propositions put to business decision-makers, then they may legitimately weigh them lightly as offering little in their way of prescience. This behaviour is wholly rational, as is the use of direct knowledge (such as business intuition) in such circumstances." Hans O. Melberg [“A Note on Keynes' Animal Spirits, Critical notes on the use of Keynes' suggestion that animal spirits can "explain" economic instability". (Observation, 14 February 1999)], also strongly criticises the inference of irrationality. Akerlof and Shiller conveniently ignore all these arguments and plunge for non-economic motives and irrational responses without making a case for either of them. In their book even trust and confidence are irrational, rather then born out of experience or a plausible game strategy: “The very meaning of trust is that we go beyond the rational” (p. 12)

What Keynes actually said is: "our knowledge of the factors which govern the yield of an investment some years hence is usually very slight and often negligible." (General Theory, p.149). "If we speak frankly we have to admit that our basis for knowledge for estimating the yield ten years hence of a railway, a copper mine, a textile factory, the goodwill of patent medicine, an Atlantic liner, a building in the City of London amounts to very little and sometimes nothing ...)" (p. 149-150)

"Even apart from the instability due to speculation, there is the instability due to the characteristic of human nature that a large proportion of our positive activities depend on spontaneous optimism rather than mathematical expectations, whether moral or hedonistic or economic. Most, probably, of our decisions to do something positive, the full consequences of which will be drawn out over many days to come, can only be taken as the result of animal spirits - a spontaneous urge to action rather than inaction, and not as the outcome of a weighted average of quantitative benefits multiplied by quantitative probabilities." (161-162) "... human decisions affecting the future, whether personal or political or economic, cannot depend on strict mathematical expectation, since the basis for making such calculations does not exist ... it is our innate urge to activity that makes the wheel go around ..." (p. 162).

Now, there is nothing irrational, or uneconomic, in pessimism and optimism. The same applies to the “spontaneous urge to action rather than inaction”, seeing that the particular course of action we select on that basis remains unspecified; until we know such a course we cannot rule on its rational and economic character or otherwise. If “a weighted average of quantitative benefits multiplied by quantitative probabilities” is not available, spontaneous impulses rooted in moods (pessimism/optimism) are perfectly rational and economic. What Akerlof and Shiller themselves define as “a basic mental energy and life force” does not lend itself to characterisation as either irrational or uneconomic. Rightly or wrongly one feels that Shiller - the successful author of Irrational Exuberance (Alan Greenspan’s famous expression) - may have led Akerlof farther in this direction than he would have gone on his own.

Moreover what Keynes actually said or meant is immaterial, what counts is the validity of whatever today he is understood or claimed to have meant. All that remains of Keynes’ proposition about animal spirits today is the volatility of investment decisions, their dependence on “the state of the news” as well as the interest rate relatively to the perceived marginal efficiency of capital, or the internal rate of return of investment projects. (Keynes was wrong, here, to consider the expected internal rate of return of investments, instead of the expected present value per unit of investment as a suitable criterion; but not seriously wrong in so far as internal rates of return and present value criteria normally lead to the same screening of investment projects into profitable and unprofitable, though with a different ranking).

And what is irrationality in this context, anyway? According to the New Oxford Dictionary of English, irrational means: ”Not logical or reasonable. Not endowed with the power of reason.” As opposed to rational: “(of a person) able to think clearly, sensibly and logically. Endowed with the capacity to reason. … from Latin ratio, reckoning, reason; ... based on or in accordance with logic or reason”. In economic terms irrational can only mean: “knowingly and deliberately acting against one’s perceived interest” (my definition, for Akerlof and Shiller do not provide one; compare with Koppl’s “a person act[ing] contrary to [his or her] best judgment”).

True, economic agents can be and often are misguided: superstitious, gullible, incompetent; they may believe in dreams, ghosts, miracles, magics, Unidentified Flying Objects, kidnapping by alien visitors, metempsychosis, after-life punishments and rewards, luck and unluck; horoscopes; the ability to predict lotto or roulette numbers from their recorded infrequency, and to predict stock exchange trends by drawing charts; casting the evil eye and getting rid of it, love potions; proteins-only diets, homeopathy, aromatherapy and acupuncture. Economic agents may be addicted to drugs; suffer from mental illnesses; be psychopaths. These phenomena and their intensity and distribution are all aspects of society’s history and culture, but do not necessarily imply irrationality in the sense of “knowingly and deliberately acting against one’s perceived interest”. They are data like the state of technology or the distribution of primary resources; they only matter when they change rapidly, radically and systematically. There is method in human action, no matter how mad humans are: even the actions of masochists are ultimately directed towards the pursuit of their happiness. I for one am completely indifferent to whether or not the three empty boxes that worry Akerlof and Shiller are ever filled or remain empty.

Expectations are most certainly never “rational” in Lucas’s terms. This does not make them “irrational”, though: there is a long-standing tradition in the theory of expectations - rigid (tomorrow’s values like today’s, as in the cobweb or pigs’ cycle), regressive (tending to return to a normal value when falsified, as interest rates in Keynes’s liquidity preference), extrapolative (projecting past rate of change into the future, perhaps the most common model), adaptive (adjusting expected change to the degree of success of the last prediction), with any number of lags and distributed lags. And of course often expectations are self-fulfilling. None of these expectations models can be said to be “irrational” or “uneconomic”. The question is whether one or another expectations model is right or wrong, or at any rate performs satisfactorily or does not, in a particular market at a particular time and place.

An alternative to Akerlof’s and Shiller’s five animal spirits

Is the concept of animal spirits the only unifying approach, or let’s call it umbrella, for the five phenomena which Akerlof and Shiller characterise (exhaustively, one presumes) as animal spirits? Namely, “confidence, fairness, corruption and antisocial behaviour, money illusion, and stories”? Suppose instead that we choose as a unifying approach a fairly conventional “intertemporal allocation”, and split this into “expectations, aspirations, enforcement of contracts and laws”. Do we miss out anything of all the things that Akerlof and Shiller include under the allegedly innovative five categories of animal spirits?

We do not. Confidence issues arise primarily in an inter-temporal framework, that must be backed by means of inter-temporal contract enforcement, if only to reinforce confidence; while simultaneous bilateral transactions only require a minimum of law and order. Fairness is a question of aspirations, which may be inconsistent in the judgement of several agents, something that is otherwise missed when we talk of fairness tout court. Corruption - which Akerlof and Shiller use not so much in the sense attributed to it by Transparency International but as associated with antisocial behaviour - is covered squarely by our notion of law enforcement; it is true that antisocial behaviour is a broader concept, for it would include also moral hazard, i.e. opportunistic behaviour, but this is a first and most conspicuous omission from the analysis offered by Akerlof and Shiller, so they would not miss it. Money illusion, especially in the way it is used by A&S in the inflation-unemployment trade-off, is precisely a matter of expectations and aspirations, and would be well covered by our alternative framework. [Incidentally, here there is second inexplicable omission from A&S, namely the principle of Central Bank Independence that is strictly derived from the lack of an inflation-unemployment trade-off, a lack that they rightly so strongly criticise]. As for “stories”, they are simply the experience (or presumed experience, that sometimes may have been wrongly distilled from available facts, or glorified into false myths) which is at the basis of expectations, nothing else.

Under the heading “Inter-temporal allocation”, and its three subheadings “expectations, aspirations, enforcement of contracts and laws”, one could still address the eight questions whose analysis is regarded as the pay-off of the theory of animal spirits developed by Akerlof and Shiller.

Eight Issues: 1. Depression

“Why do economies fall into depression?” Because markets do not guarantee inter-temporal efficiency, or rather because they guarantee inter-temporal inefficiency. Because savings depend on the level of income and investment on its rate of growth, so that their equilibrium is not necessarily automatic and anyway takes time. Because of the interaction between the multiplier and the accelerator (Paul Samuelson once wrote that economics is the science of optimisation under constraint - except for the interaction between multiplier and accelerator - one of the most eloquent statements of the inadequacy of neoclassical economics). Because economic growth along the long term trend of population and productivity has a full employment ceiling, and a floor due to the fact that net investment and feasible subsistence consumption cannot fall below zero. As the ceiling or the floor are approached, without necessarily being hit, the growth slowdown turns into decline and the decline slowdown turns back into growth. Because booms cause wage growth thus carrying the seeds of their own bursting, while recessions depress wages and restore profit margins allowing the financing and the encouragement of new investment. Because on top of all this there is also the political cycle first investigated by Michal Kalecki. I am very happy with the Harrod-Domar knife-edge growth paths, Dick Goodwin' growth cycle and Hyman Minsky's financial cycles.

2. Central Banks powers

“Why do central bankers have power over the economy, insofar as they do?” Because on the basis of a faulty theory (of rational expectations and the associated denial of a trade-off between inflation and unemployment) since the late ‘eighties they have been given Independence from the government and discretionary powers over inflation targeting, while they are allowed to ignore the wreckage they often inflict on output and employment (the Fed is not so bad, because it is less independent and its remit includes also interest rates, employment and the exchange rate).

3. Unemployment

“Why are there people who can’t find a job?” Because in a closed economy (and the global economy is closed to the outside by definition) unemployment cannot necessarily be solved by lower wages, for these lead immediately to lower consumption, which may or may not be compensated for by higher investment - not least because lower wages will tend to lower also the investment intensity of new capacity. In the open traditional non-global economy these considerations apply to a lesser extent, but they are still operational if import and export weighted elasticities with respect to prices add up to less than unity.

Some unemployment is “classical”, i.e. caused by the lack of equipment in a quantity sufficient to employ everybody even at subsistence wages, or rather at the efficiency wages that minimise labour costs per unit of output. Some unemployment is “neo-classical”, i.e. due to the money value of the marginal product of labour measured at its competitive price being lower than money wage. Some unemployment is Keynesian, i.e. due to the lack of effective demand and to imperfect competition.

The most important cause of unemployment of all is probably imperfect competition - which does not gets a single mention in the entire ambitious theoretical construction by Akerlof and Shiller. For under imperfect competition producers will value the marginal product of labour not at its price but at its marginal revenue, which may become zero or negative well before full employment of labour is reached, therefore preventing full employment even if wages were flexible downwards right down to zero. Apart from the fact that even if the full employment marginal product of labour reckoned at its marginal revenue was positive, and the wage rate fell down to its level, entrepreneurs might regard as unrealistic - on the basis of experience - the continuation of such a low wage into the future, necessary to make investment pay, and still refrain from additional investment thus maintaining unemployment.

By comparison with this set of explanations, the one provided by Akerlof and Shiller is not at all satisfactory. People - they say - are perfectly willing to work for the wage rate per unit of time that would correspond to full employment, but then employers will take into account the positive feedback of higher wages on productivity (through higher morale of employees and the like), and will go and pay wage rates per unit of time higher than the full employment rate in an effort to reduce, indeed minimize, the wage cost per unit of effort, or for unit of product. At this higher wage rate, lots of willing workers remain unemployed. But this is just another form of a naïve theory of voluntary unemployment, because the implication is that the unemployed would be willing to work for a lower wage but would then supply a more than proportionally lower amount of effort or product. If this is the problem, there is a simple remedy: linking wages to productivity, but A&S do not take it into consideration. And their neglect of imperfect competition is a damaging omission in any serious discussion of unemployment.

4. The Phillips Curve

“Why is there a trade-off between inflation and unemployment in the long run?“ Because there is at least some money illusion, A&S say. This is one way of looking at it; more rigid inflationary expectations will also do the trick. Except that the question of whether or not there is a trade off is ultimately an empirical question, and empirical verifications of the Phillips curve linking inflation and unemployment are not particularly satisfactory.

5. Saving

“Why is saving for the future so arbitrary?” Because the inter-temporal trade-off between dated consumption of an individual or of households depends on too many, too uncertain factors, and on different responses. For instance people might save more at a higher (real? nominal?) interest rate, but a target saver will save less to obtain a given consumption transfer into the future. And there is absolutely nothing in saving behaviour to support the necessity for a positive real interest rate, as usually presumed by the Bretton Woods institutions. Here again, as for the Phillips Curve, it is not enough to note or even explain the erratic nature of saving behaviour, but it is necessary to positively identify and verify empirically the impact of the many factors that might be at play.

6. Volatile assets prices

“Why are financial prices and corporate investments so volatile?” This one is easy. The market valuation of the shares of a company, if market work, will correspond to the current dividend d per share, cumulated at its expected nominal growth rate g per year, and discounted at the appropriate nominal discount rate r. Thus the slightest change in the rates g and r expected to prevail in the future will generate disproportionate, massive changes in the share price. If both the expected growth rate g and the discount rate r were constant, and r>g, the price of the share will be equal to d/(r-g). If g>r the share price would tend to infinity with the time horizon tending to infinity. Conversely, the generalised downwards revision of g relatively to r (as in the bursting of the dot.com bubble) will precipitate a stock exchange crisis [With apologies to readers for an earlier ambiguous formulation of this problem]. This is the beginning of an answer. Add securitisation, originating assets not to hold but to sell, leveraged betting on derivatives (never mentioned by A&S as such) and a credit crunch, and Bob’s your uncle.

7. Real estate cycles

“Why do real estate markets go through cycles?” Because there is a cycle in building activity, just as there is in pig production or in general investment: high rentals lead to high capital values of existing buildings and to new construction; as new houses and commercial buildings are built their rental and therefore market value fall, and so on for the time it takes to reduce the buildings stock to an equilibrium level, which will then tend to overshoot. And for the same reasons of optimistic expectations, non sustainable asset price increases, the slowdown in capital gains leading to a decline in asset prices, etcetera etcetera in reverse.

8. Poverty Traps

“Why does poverty persist for generations among disadvantaged minorities?” From the impact of so-called stories of minority discrimination, - say Akerlof and Shiller - to be remedied by stories of positive role models and “affirmative action”, that “can play a a significant role in breaking down the barrier between the two Americas” (p. 164). But look at it another way, as Branko Milanovic does in a recent paper [“Global inequality of opportunity. How much of our income is determined at birth?”, mimeo, World Bank, February 2009]. “Suppose that all people in the world are allocated only two characteristics over which they have no control: country of citizenship and income class, within that country, of their parents. Assume further that there is no migration.” Under this premise, Milanovic shows that “at least 80 percent of variability in income of almost 6 billion people in the world is explained solely by these two characteristics. Thus, globally-speaking, the role of effort or luck in improving one’s income position, cannot be large. On average, “drawing” one-notch higher parental income class (on a twenty-class scale) is equivalent to living in an eleven-percent richer country” (Milanovic, op.cit.). This makes A&S’s concern for disadvantaged minorities pale into insignificance. And let there be no doubt that Milanovic’s kind of explanation and analysis rests on no notion of animal spirits.

The current crisis and its remedies

A post-script to Chapter 7 of Akerlof and Shillers includes their analysis of the current crisis and policy recommendations. This is one of the best parts of the book, but their kind of analysis is by now more and more widely accepted. It does not sound very different from, say, the analysis by one of the most conventional, shrewd economists around: the ECB President Jean-Claude Trichet, [“The ECB Enhanced Credit Support”, a keynote address given on 13 July 2009 at the University of Munich].

The only original new remedy proposed by Akerlof and Shiller is a credit target, in addition to the traditional interest rate and fiscal stimulus. “The aggregate demand target will indicate, on the one hand, the fiscal stimulus and interest rate policy needed for full employment. The credit target will show what judicious application of methods 1, 2, and 3 [respectively: expansionary discount window, direct investment in banks, and use of government sponsored enterprises] must achieve: together they must create the financial flows - the issuance of commercial paper, bonds and other instruments - that are also associated with full employment” (p.96). And “of course the two target approach and Humpty Dumpty [meaning coping with the irrevocable fall of financial markets, see also earlier reference] do not apply only to the United States but internationally as well” (Ibidem).

The alleged originality of the animal spirits theory that Akerlof and Shiller have been developing in their book is exposed as another way of speaking of what is well understood in standard analysis.