Undercover Economist

When wishful thinking becomes wasteful

‘The simplest explanation for lengthy disputes? That people misperceive their chances of winning’

From a purely rational perspective, costly arguments are puzzling. A divorce case that goes to court, an industrial dispute that leads to a strike, even the extreme case of a war — all these things are, to put it in the mildest possible terms, a waste.

Of course, there will always be conflicts — but logical people should resolve them quickly. Consider a simplified model of this process. Two people, Amy and Ben, are arguing over how to divide a baked Alaska, the centre of which is gradually melting. First Amy makes an offer. Ben may accept it or propose a counter-offer. Amy, in turn, may accept that or make a counter-counter-offer. Each time an offer is rejected, the delicious dessert shrinks by 10 per cent. Amy and Ben have opposing interests because each would prefer to have the entire dessert to eat alone. But they also have one thing that they can agree on: given the situation, both would be wise to shake hands on a deal promptly and start eating.

A similar logic suggests that an industrial dispute should never lead to a strike. Instead, employers and unions should see that a strike will cost them both dearly, and find some way to resolve their differences. Civil litigants should always agree a settlement before having to go through costly legal proceedings. Often, this is what happens — but not always. Why?

Economists have a few ideas. One cynical suggestion is that some people are playing a different game. Perhaps a belligerent politician or union leader would find his or her position strengthened by a strike. A general might desire a war. Lawyers might profit from urging their clients to go to court.

Another possibility is that people need to signal their willingness to fight in battle after battle. Imagine a large company is being sued by a small competitor for some transgression. If it settles out of court, other competitors will scent blood and dart in like piranhas, so it fights a costly case to scare other would-be litigants away.

But the simplest explanation is that people misperceive what is fair and also their chances of winning.

Consider that melting baked Alaska again. The obvious division is a 50/50 split, and in laboratory experiments that is usually what Amy and Ben will agree, and quickly. (The sophisticated equilibrium of this game is not quite 50/50 because Amy has the advantage of moving first, but it’s close.) But what if the disagreement was more complex? For example, what if Amy preferred the meringue topping, which was not melting, and Bob preferred the ice-cream centre, which was? This complication introduces doubt as to what the intuitive split should be. Ben may still see 50/50 as the logical split, while Amy may feel that she is in a stronger position and should get more. Each side may believe that the division that happens to suit them is objectively fairer — a self-serving bias. And indeed, in laboratory experiments players usually fail to reach a swift agreement in such circumstances.

In a more realistic setting, such as an industrial dispute or a legal case, there will typically be several ways of seeing the problem and several different settlements that could be justified as fair. When the disputants fixate on different settlements, agreement may be derailed.

To test this idea, Linda Babcock and George Loewenstein, behavioural economists at Carnegie Mellon University, once asked experimental subjects to ponder the facts of a real tort case from Texas. A motorcyclist had been injured after a collision with a car and sued the driver of that car for $100,000. The subjects were randomly given the role of the motorcyclist or the driver and asked to role-play negotiating a pretrial settlement, to proceed to court if no settlement could be agreed. The experimental pay-offs mimicked the structure of the real case, including a reward for reaching a pretrial deal.

The subjects were also asked to make a guess as to what damages the judge awarded in the real case — with a cash bonus if their guess was accurate. Despite this bonus for accuracy, the “motorcyclists” guessed that the judge had awarded almost $15,000 more than the “drivers” guessed. Their entire view of the case had been biased by their own self-interest. No wonder that plaintiffs and defendants sometimes fail to reach a settlement.

Wasteful conflicts may also occur because of wishful thinking about the outcome. Strikers may assume that an employer will soon cave in to pressure. Litigants may overrate the strength of their case and the competence of their lawyer.

A few years ago, Guy Mayraz, who is now a behavioural economist at the University of Melbourne, conducted a test of wishful thinking. He divided experimental subjects into “farmers”, who benefited from high wheat prices, and “bakers”, who profited when wheat was cheap. Then he showed them historical charts of wheat prices and asked them to make forecasts. Mayraz paid a bonus for accuracy, yet the farmers systematically predicted higher prices than the bakers. This is wishful thinking in its purest form. Whether engaged in a tough negotiation, or simply trying to predict the future, we find it hard to distinguish between what is true, and what we wish was true.

Written for and first publiched at ft.com.