Scroogenomics

I was thinking today about a book that's getting some attention, about how Christmas is the most wasteful time of the year. It reflects in some ways what is interesting about economists and how they think. They talk about efficiency, and allocations, and equity as if these were black and white comments. They claim to be objetive and then use the get out of free card "other things being equal" which usually means if my assumptions are correct.

However sometimes (many times) their assumptions are implausible or required to make the math in their elegantly complicated models work. Anyone who knows me, should know I'm not a big fan of economists. But sometimes as in this article they do provoke thought and provide a different perspective on things.

One of the interesting things about the book Nudge by Cass Sunstein and Richard Thaler is their description of the problem I just identified above. Early in the book, they talk about the difference betweens the "Econs" who represent the ideal of economist and basically work as cost benefit analysis machines, and the "Humans" who are sometimes able to make rational decisions but need help and are driven by emotions.

The key point that they highlight is that there are systemic cognitive biases inherent in our humanness. We make systemic mistakes in the same direction. Given this you cannot assume that humans are "rational". It will take some time for this to filter down to all economists but it seems interesting and may point the way for economists to get closer to real world predicitions. Their key assumption of course is that humans are "rational" and so it does represent a challenge to their discipline.

Alright here goes!

Well I was talking to my friend Pat last night about why I have a hatred of economics. Now first of all this is not only because I failed a few higher level economics courses in university. The reason I abandoned the discipline (because after all I started out in university wanting to be an economist), goes deeper than that.

I think at a fundamental level any attempt to reduce human behaviour to a few variables and equations will inevitably fail. The increasing sophistication of mathematical modeling still cannot accurately predict human behaviour.

The first thing any economist will do when they develop a model is make assumptions. Fair enough, you can't make a model without eliminating some variables and simplifying things. However an increasing number of the assumptions made by economist are made to make the math fit. Now in my economics classes I didn't have the guts to suggest that some of these assumptions were leading to false conclusions, but I think this is often the case.

A perfect example of this is from the book freakonomics, a book touted to show the amazing power of economic analysis and the way that it can be used on non-traditional subjects and still have value. I suppose one could argue that taking economics out of its traditional domain is part of the reason that it fails to carry as much weight, but it still highlights some of the shortcomings of economic reasoning.

The author uses the example of falling crime rates in NYC and he links them to abortion rates. Now this seems to be to be problematic. He claims that the link between rising abortion and lower crime exceeds the impact of crime prevention strategies implement under then mayor Rudolph Juliani. Now I don't doubt that there might be a link, but the claims he makes are way out of proportion to the evidence presented. (He ended up making an error in his calculations which damaged the credibility of his argument further).

In many of the other examples of the book he also uses proximate measures. He will take something he can measure, for example test scores and through regression analysis link it to something else say marital status of the parents and then make claims based on this research. However he makes claims that exceed his limited view. In almost every case because he cannot measure the social factor directly, he takes approximations to back his claim.

Again I see the same problem in economics ore generally, the attempt to be value neutral when in fact economics and economists present an analysis full of values. The sacredness of individuality for example permeates everything economics does, because it is an analysis based in western values about individuality. Try even getting an economist to even admit that there may be values hidden in their analysis.