Thursday, 16 July 2009
I'm reading George Loewenstein's book Exotic Preferences at the moment (still waiting for Tyler Cowen's to arrive) and enjoying it greatly. It's a collection mainly of journal papers and speeches authored by Loewenstein, but not too technical and an interested lay reader should find most of it very accessible.
So far the most important message is this. Traditional microeconomics derives from preferences based on consumption utility; which is certainly an important component of total utility, but by no means all of it. Philosophy, psychology and behavioural experiments all indicate that people gain much, maybe most, of their motivation from things other than consumption.
The first article is about mountaineering and the - definitely exotic - preferences of top mountaineers and polar explorers, who push themselves way beyond the limits of action that could plausibly be rationalised by consumption utility. Loewenstein gives four key examples of non-consumption utility: self-signaling, goal completion, mastery and meaning. A series of revealing quotes from the writings of mountaineers shed some light on their motivations, and while not a rigorous exposition, they are at least suggestive that these four motivations do operate for those people. Loewenstein then suggests that the same drives also apply to "normal" people who are not willing to kill themselves on the top of the Himalayas.
As a minor technical quibble, I would suggest that it might not be useful to use the term utility for these. While they clearly do affect decision-making, it may be unproductive to try to model them with an underlying utility function. One of the key results of classical microeconomic theory is that rational preferences on a consumption set can be generated by finding an appropriate utility function for each available good; it is not a given that the types of choices Loewenstein is talking about can be generated in this way.
But I'm being pedantic - "utility" is being used as shorthand here and not in the sense of classical utility functions. The article ends with a call for economists to "incorporate non-consumption-related motives into formal models of economic behaviour". The implication being that some young, emerging talent might take up the challenge, just as Greg Mankiw suggested the other day (though Brad DeLong and Paul Krugman (not very wonkish this time) stepped in just in time to save any "ambitious grad students" from getting too much mileage out of the question).
Other chapters cover the economics of meaning, curiosity, social utility, negotiation, experimental economics, prediction of future utility, intertemporal considerations, mental accounting and lots of other key aspects of the field. Overcoming my slight disappointment that he has already covered so many phenomena that I want to write about myself, I think this book is setting some excellent challenges for modelling-minded behavioural economists, and I plan to take some of them up.
I believe it will be possible to create a unified cognitive model - even if the mathematics is complicated - which is broadly consistent with empirical results in most of these areas. Even if not, it's a worthwhile project to spend a few years on. Why climb this particular peak? Is "Because it's there?" an acceptable answer?