Sunday, February 22, 2009

The irrational ape

While we lament the collapse of the economy, at least one group is probably feeling mildly amused at being spectacularly vindicated - behavioral economists. With Nobel prizes having been awarded to such proponents as Daniel Kahneman, it isn't exactly like behavioral economists are living in obscurity. However, the current economics woes add tremendous weight to their view of an irrational world, and perhaps conclusively shifts the winds in favor of less idealistic world views.

Behavioral economics postulates that people are not rational, as assumed by most classical economics. Instead, that people have a lot of quirks. I happened to read a couple of interesting books recently that cover some of these quirks.

The first of these tomes is Predictably Irrational by Dan Ariely. Dan Ariely's book does a good job of covering some fascinating human quirks. Here's a summary of some of the key ones:
  • The first one covered is fascinating for marketers. It turns out that people have no absolute assessment of value and only understand value in relative terms. In fact, so much so, we favor options where we have something to compare the option to, even when there is no absolute reason for doing so.
  • The second is the principle of anchoring. When determining values we tend to anchor to numbers, even random numbers, as a starting point, and once we anchor the effect persists. So, values are not really driven just by supply and demand, but what we anchor to. And once we anchor, we start indulging in arbitrary coherence - i.e. we want everything else to be relatively coherent to the anchored value.
  • The third phenomenon discussed is that we don't treat zero or free as just another value, but suddenly suspend all rationality when something is offered for "free". In fact say something is offered for 1c and then is offered for free, our reaction to the latter is irrationally different from the former.
  • One of my favorite chapters dealt with how we react differently when we use social norms vs. market norms. It's the difference between buying your mother-in-law a gift vs. giving her the cash. We are often happier doing things on a social basis, which we would be unhappy to do if money entered the equation. The insights in this chapter have significant implications for the trend towards socially conscious management. For one, it raises questions about the efficacy of social marketing and some of the moves towards developing better employee relationships through social rewards.
  • One of the fairly obvious chapters was his description of experiments that people actually take very different decisions when aroused than when not. I was amused that anyone ever thought that they wouldn't. It suggests that avoiding situations where you could aroused is usually better than trying to exercise self control once in the situation. It also suggests that it may be better to arm teenagers with protection, than to rely on the better angels of their nature. Duh!
  • Another chapter that made me wonder how old researchers were, was the chapter on procrastination and choice. Essentially, Dan Ariely shows that a certain amount of control and discipline is better for us than complete choice and anarchy. Huh? Most students know that, in fact, we deal better with more structured courses than ones with complete freedom to choose. I was surprised that he found it so surprising.
  • He also shows that we irrationally over value what we already have, and that the more effort we put into it, the more we value it. Again, something we should already be aware of, but in this case, it's worth a look at some of the irrational decisions we may take about what we consider our own.
  • His discussion about people's irrational reluctance to close options was disturbing. It reinforces my view that options thinking and understanding how to evaluate opportunity costs should be made required reading at a much earlier age. Our desire to keep options open makes us irrational and drives us to give up a bird in hand for the prospect of one in the bush.
  • One of the more fascinating chapters in the book was one dealing with expectations. He discusses how our physiological experience can be fundamentally altered by our expectations. A corollary was a discussion of the "placebo effect". Our belief that something is better may actually make it work better. For instance, we may believe a 50c aspirin is better than a 5c aspirin, and so a 50c aspirin may actually work better. It suggests that something has a placebo effect is not really the same as saying no effect.
  • My favorite bit though was his discussion of why people are honest or dishonest. He essentially shows that most people try to be honest, and tend to be more honest if reminded about social norms; that people's tendency to be dishonest is not related to a cost-benefit analysis of how easy or tough it is to get away with the crime. What seems to be a bigger measure is how clearly the cost of the crime on the victim is apparent. Very interesting, particularly the example of how airline companies don't equate repricing of rewards points with theft - which is exactly what an arbitrary cash deduction of equal value might have been called.

While I have listed some of the insights in a very dry list, the book uses highly entertaining experiments in a marvellously accessible way to illustrate these and other points. All in all, a very good read.

My issue with the book is that Dan Ariely clearly has a view of the world and seems to be looking for confirming evidence. In a few of the cases, he uses analogies, which appear to be predicated on his view of the world, than conclusive evidence of the opinion he is postulating. The other issue was that the book does not tie these ideas together into any coherent world view. In the end, you come away with some very interesting but disjointed quirks, but it isn't immediately apparent how these are usable.

The other book I read is Sway by Ori and Rom Brafman. Now, unlike Predictably Irrational this is not as peppered with experiments. It is significantly shorter and at least tries to tie the points together using the device of a plane crash in which over 500 people died. Here are some of the key insights:

  • People do not react to equal magnitudes of upswings and downswings in the same way. We are ridiculously averse to the idea of a loss, and take irrational decisions trying to avoid losses. In fact, once we commit to a strategy of avoiding loss, we tend to double and redouble, rather than admit a loss and calling it quits, leading to utter ruin.
  • Perhaps the most disturbing section in the book was its discussion on how we tend to let our preconceptions and labels drive our assessment of value, to a point where even the most rational of us seem to be physically incapable of seeing and accepting even the most compelling evidence that contradict our apriori preconception. In some cases, this label occurs through first impressions or some label assigned by someone else. Once we have a label, its almost impossible for us to be objective. What's even more interesting is that in many cases, the labels not only affect the judgment of the labeller but also the labellee, often affecting the person being labelled in a way that makes it a self fulfilling prophecy.
  • In another section, they discuss how people's deep rooted belief in fairness makes us resent and act irrationally when we perceive a process as being unfair, no matter how fair the final outcome, and vice versa. In fact, they go onto show that our evaluation of many situations depends more on the process than the outcome, and that our evaluation of the process depends on our familiarity with it. Essentially, one of the corollaries is that managers who create more visibility are going to get rewarded more, because they create more clarity about their process. Logically though, it's better to focus a lot more on the outcome than the process, as we might otherwise take incorrect decisions.
  • Again, like Predictably Irrational, the authors in Sway discuss social norms vs. market norms, and again, they show that people react very differently to the two and that in some situations, social norms are better motivators than financial rewards. Personally, I preferred the discussion of the implications in Sway more than the Predictably Irrational, although the latter described some fascinating experiments on this subject.
  • Another phenomenon that was fairly intuitive was the idea of group think. Essentially, people don't like to disagree with the group and don't disagree unless they feel they have permission to do so. It's a bit like the Emperor's New Clothes. Dissenters keep quiet, until someone, even a little boy, points out the obvious.

The good news about Sway is that it is very accessible and very short. The bad news is that its attempt to tie everything to the plane crash is extremely tenuous, and many of their examples really stretch the analogy. I commend the authors for their laudable attempt to tie research to practical aspects of management, and while their advice is generally good, in many cases, they seem to be generalizing the point more than warranted by the evidence that they present. So, I would be cautious in its application.

Both books are fun though, and definitely worth a read - although, I still like Daniel Gilbert's Stumbling on Happiness a good deal more.

It isn't clear how representative of the field of behavioral economics, popular books such as these are. Assuming that they are a fairly good sample, it strikes me that though behavioral economists have successfully poked holes in classical economics, they don't necessarily have a compelling alternate view that can adequately replace current models. Until they do, these ideas, while interesting, will ultimately not define the practical application of economic theory.

No comments: