Paul Krugman described behavioural economics as the most important innovation in the economics discipline for half a century.[1]
Three decades ago, I reviewed The Winner's Curse: Paradoxes and Anomalies of Economic Life by Richard Thaler which reported behaviour which appeared anomalous according to the economist’s standard model of ‘rational economic man’. Anomalies are at the heart of scientific progress. For a time, they may be seen as curiosa to be discounted or ignored by the conventional wisdom. The curiosities build up and eventually some of the less complacent thinkers began to offer an alternative, perhaps as a flashing insight as it was with Einstein’s theory of relativity, perhaps step by step as with quantum mechanics.
Thaler’s book, published in 1992, but based upon articles published in journals from 1980, was at the curiosum stage, but an explanation for the anomalies was already developing. In 1979 psychologists Daniel Kahneman and Amos Tversky published the key notion of prospect theory.
Over the years, their contribution and Thaler’s and others have been recognised by Nobel awards in economics. Today, behavioural economics is an integral part of the economics discipline. A quick summary of the field is that humans do not behave in the way that economics’ traditional theory of rational economic man assumes.
Indeed, Thaler’s 1992 book is now a historical curiosum. Instead, he and Alex Imas have updated the book as indicated by the new subtitle, Behavioural Economics Anomalies, Then and Now, providing an account of the development of the field over the last three decades. It consists of the original chapters each with an update, plus some new ones on such topics as prospect theory and mental accounting.[2] They report much research which affirms the anomalies are real.
Unfortunately, unlike rational economic man, there is no simple model which summarises how we actually behave. (Non-economists may say, ‘that’s obvious.) Moreover, as Thaler and Imas observe, behavioural economics is not a standard topic in first-year economics textbooks. You cannot use the excuse that physics teachers justify teaching Newtonian physics in the early stages of their physics courses – that it is a foundation for relativistic physics. Teaching rational economic man is more like teaching Aristotelian physics.
Fortunately, macroeconomics is not greatly affected by behavioural economics. That is because it never really adopted rational economic man. (T&I are respectful of Keynes as an early identifier of behavioural complexities.) It is true that there was an effort to provide micro-economic foundations for macroeconomics, but that failed. It still lingers on in some macroeconomic models, but they failed during the 2008 GFC. Even if individuals were rational economic men and women, their aggregate behaviour may not be. But anyway, they are not.
One area where behavioural economics has had a big impact is in finance. It is no accident that many behavioural economics researchers are also employed in financial investment firms. You won’t make much money from the investing if you believe in the ‘efficient market hypothesis’ – that investors are as rational as the theory assumes. There is a new chapter in the book which explains this.[3]
I have written on behavioural economics impact on survey design known as ‘nudge’. You may object to the government using the knowledge as manipulative. I don’t because the private sector uses it against you – the research findings are not confined to economics; marketing professionals are aware of it too and perhaps not quite as fastidious.
Where the new theory is really upsetting is in applied welfare economics, illustrated here by Cost Benefit Analysis (CBA), a standard evaluation method for helping make economic decisions.
A common criticism of CBAs is that they lump winners and losers into a single aggregate. (When I was working on a related project, published by WHO, we recommended that the distributional impact should be reported.) The standard justification is that the aggregate shows that the winners could compensate the losers and still be better off. (They rarely do.)
Prospect theory offers a more fundamental critique which I illustrate with a simple example. Suppose the CBA concludes that the project would add 10 units (put as many zeros after this as you want). The distributional analysis shows winners get +30 and losers suffer -20. Standard analysis says the winners could compensate the losers with 20 from their winnings. Now the losers would not be worse off – and the winners would have 10.
Prospect theory adds a caution when the compensation does not happen. Its account of loss aversion points out that the losers would consider themselves as much more than 20 units worse off. The empirical evidence is that the psychological pain of losing a specific amount of money is roughly twice as intense as the pleasure of winning the same amount. In the above example they would judge themselves as having lost 40 (twice 20) so with winners evaluating themselves as 30 better off, the uncompensated CBA is a loss of 10 (30-40). The aggregate number that a CBA generates is only meaningful if the compensation is made. Without it, the net gain promised by the CBA is meaningless.
Thus a proper CBA should explain that the aggregate assumes that there is the compensation and without it the calculated return is biased upward and may be negative. It also implies that the report should identify winners and losers and quantify the changes (as recommended in that WHO report). I have never seen a CBA that did this. Which is perhaps not surprising given that Econ101 does not explore behavioural theory.
Loss aversion seems to limit the ability of decision makers to make changes; it suggests a cause of political inertia. It also questions the whole notion of progress since there are always people being made worse off by it. Think of the Luddites – weavers who destroyed the mechanical looms which were putting them out of business.
But I also think of Ken Findlay, a staunch unionist who led the unsuccessful resistance to the closure of his freezing work. Yet years later he said to me that the closure was the best thing that had happened. Prospect theory has three key observations. We have already met loss aversion. The second is the framing effect – that people need a reference frame to judge their wins and losses.[4] Ken must have been using a different frame when he was reflecting all those years later compared to the one he was using at the time of the closure.
Because scientific psychology on which behavioural economics is founded is very evidence- based – rigorous experiments are key – we know little about how frames change over time. After 30 years of intensive investigation, there is still has no comprehensive paradigm in the way there is one for rational economic man. But it is evolving. Ignoring its findings today and hanging on to the plainly outdated theory of human rationality will lead to poor understandings, bad predictions and low-quality policy. The challenge is to apply the behavioural economics we know as best we can and try to keep up with its evolution.
This is not an Econ101 challenge. A good place to start getting up to speed is Thaler and Imas’s book.
[1] For more of my writings on behavioural economics, including dimensions not covered in this column, see here.
[2] There has not been the space to cover mental accounting where individuals treat money differently depending on its origin, label or intended use. It explains what I was doing in my column on the golden rule. Rather than combining all the fiscal incomings and outgoings into a single account, which is what happens under the 1989 Public Finance Act, it split the single account into separate current and capital accounts. That reflects how humans think, including those who give us credit ratings.
[3] See the next note.
[4] The third is the Certainty Effect that individuals disproportionately overweight outcomes that are considered guaranteed compared to those that are merely probable. It has not arisen in the column because it has not been necessary to discuss decisions where there is risk. This principle is critical for understanding how financial markets work.