The term "animal spirits" derives from the Latin spiritus animalis, or animating force. It was popularised by John Maynard Keynes, who suggested that decision-making was often the result of a spontaneous urge to action rather than a careful calculation, and that this was key to understanding economic fluctuations. But economics took a quantitative turn, rational optimising models of behaviour became the prevailing orthodoxy, and Keynes' insight did not have much impact.
George Akerlof and Robert Shiller argue that we need to rediscover animal spirits in order to understand macro-economic phenomena. They use the term to encompass a variety of motives and mental processes that do not appear in standard economic accounts of behaviour: confidence, fairness, corruption, money illusion (being concerned about the "nominal" face value of money rather than its purchasing power, which is its "real" value), and our tendency to think in terms of narratives and tell each other stories. They advocate basing economic analysis on a realistic psychology, rather than using animal spirits as merely a minor corrective to be appended when the regular model gets it wrong.
According to standard econom-ic theory, people are motivated purely by economic interests and base decisions on their rational expectations about future economic conditions. Wages adjust until the demand for labour equals its supply, which implies that everyone who wants to find work at that wage will do so and there will be no involuntary unemploy-ment. Workers care only about the real wage, and their wage demands take into account their expectation of inflation. With regard to housing and the stock market, trading is based on rational expectations, so the prices of assets reflect all known information about their future value.
These assumptions have policy implications. If wages and prices are based on correct expectations regarding inflation, then it turns out that there is only one level of employment at which inflationary expectations are stable, the so-called "natural rate". Government policy aimed at decreasing unemployment will have no long-term effect on the natural rate but will result in a long-term increase in inflation. Conversely, govern-ments can target inflation safe in the knowledge that this will not have any negative long-term repercussions for unemployment. If asset prices reflect information about their values, then large or sudden asset price movements are not cause for government intervention in the market.
The picture changes when we include Akerlof and Shiller's animal spirits. Workers are concerned that their wages should be fair, and money illusion means that they care about the nominal wage. These criteria imply that when there is deflation, workers are resistant to nominal wage cuts, causing involuntary unemploy-ment. Conversely, when there is moderate inflation, workers do not insist that their wages keep pace. So there is a trade-off between inflation and unemployment. The authors estimate that decreasing the US inflation target from 2 per cent to 0 per cent leads to a 1.5 per cent increase in unemployment, which equates to 2.3 million jobs and a loss of GDP of $400 billion (£244 billion) per year.
In asset markets, decisions are influenced by people's confidence in the economy. This, in turn, is determined partly by the stories that are the focus of popular attention. There is a feedback loop between confidence and asset price movements. People buy when they are confident, which pushes up prices, and when prices increase, people become more confident. This implies that speculative bubbles can occur, when prices are wildly at variance with the underlying value of the assets. Thus there is a role for governments to regulate markets and to protect people from the effects of boom and bust.
The argument for the inclusion of corruption as an animal spirit is less convincing. The authors claim that the business cycle is connected to "fluctuations in personal commitment to principles of good behaviour". But the historiography they give as evidence is simply a succession of discoveries of new opportunities for financial gain by cooking the books, and the exploitation of these until their discovery and collapse. The authors convince that this cycle is damaging, but give no evidence that it is caused by a fluc-tuation of moral principles rather than by changing opportunities.
This is not the only time that Akerlof and Shiller draw a strong conclusion about the existence of animal spirits from limited or inconclusive evidence. They acknowledge that their evidence for confidence also has a rational choice interpretation. But their argument for confidence gets support because explanations involving it are better than those without. In any case, their animal spirit is not really confidence; there's nothing wrong with confidence per se. The mechanism is the giving of undue weight to stories at the expense of better evidence - and that wouldn't be news to psychologists.
The book reveals a need for research on animal spirits, by social scientists and philosophers. Without more analysis and stronger evidence, hard-bitten rational-choice economists will not be convinced that their methodology should be turned on its head. However, the authors do a superb job of conveying the importance of behavioural economics to a non-specialist audience. They increase our understanding of recent economic events and they show that animal spirits affect how governments should manage the economy.
Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism
By George A. Akerlof and Robert J. Shiller
Princeton University Press 264pp, £16.95
Published 18 March 2009