Friday 11 February 2011

Dodgy Scholarship

Recently, a "behavioural economics" paper, by Beaulier and Caplan, has caused some stir around the economic blogosphere; see here, here and here.

Understandably so:

"Critics often argue that government poverty programs perversely make the poor worse off by encouraging (...) 'social pathologies.' (...) The current paper argues that (...) existing empirical evidence suggests that the poor deviate from the rational actor model to an unusually large degree."

These "social pathologies" make "government poverty programs" counterproductive.

According to Beaulier and Caplan, the following is a list of general deviations, present in superlative degree among the poor:

  1. Judgmental biases (including self-serving bias -the tendency to see themselves as better than they really are- and biased risk estimates -which might explain why the poor have a penchant for crime-, all of which could be explained by lower IQ).
  2. Self-control problems (a set of anomalies superficially similar, but in some unexplained way distinct from laziness and short-sightedness).

Although several of the links above are critical of the paper, the most comprehensive criticism I could find, by James Kwak, states (before undertaking the analysis of unstated alternative explanations to the "social pathologies" mentioned in the paper):

"But what Caplan and Beaulier really mean to say is this: (...) the poor are more irrational than 'normal' people."

It strikes me that James's criticism, however valid and demolishing, should have gone farther in the opening questions. By jumping straight into the alternative explanations, James overlooked some fatal flaws in the Beaulier/Caplan paper.

For starters, the paper is a survey of 74 articles and books on the subject of cognition and social consequences. One would expect that, given the subject matter, many of the works cited came from the psychological scholarly literature, social psychology or sociology.

Not so. Only 4 papers cited were published in psychology or social sciences journals, plus one remarkable book.

"The Bell Curve", by Herrnstein and Murray, is the book. To the best of my knowledge, Herrnstein is the main exception to this "psychological" partial omission.

And this predilection for Herrnstein shows in the text: the argument around the importance of IQ is highly dependent on this work (with 8 references or direct quotations). Further, as seen above, lower IQ is advanced by Beaulier and Caplan as cause of judgmental biases (and indirectly of anomalies like criminality among the poor).

"The Bell Curve" has been subject of intense controversy since its publication in 1994, to the extreme that by 1996 the American Psychological Association was forced to publish a non-research paper, rather pedagogical in nature and addressed to the general public "Intelligence: Knowns and Unknowns", at its flagship journal, American Psychologist:

"Concepts of 'intelligence' are attempts to clarify and organize this complex set of phenomena. Although considerable clarity has been achieved in some areas, no such conceptualization has yet answered all the important questions and none commands universal assent. Indeed, when two dozen prominent theorists were recently asked to define intelligence, they gave two dozen somewhat different definitions".

Obviously, where two dozen prominent psychologists failed, it's entirely reasonable to expect two economists turned into amateur psychologists to succeed in defining intelligence, to deal with its effects, to propose social policy, and all that with next to no base in psychology! Bravo!

I will not dwell any deeper in "The Bell Curve" and its infamous "genetic factors", except to say that Beaulier and Caplan (that astutely avoided mentioning them in the text) must explain how they see those genetic factors in their "theory": are they referring to poor blacks only, or all poor people regardless of race? What about wealthy blacks and poor whites?

As I grow weary of this crap (which, as James Kwak said, is indeed offensive), let me quickly deal with two additional observations, related to James's comment that these anomalies were detected studying university students:

First additional observation

I know of at least two instances of biased perception affecting equally the poor and the wealthy (and one of them should be well-known to Caplan):

"Income plays little role; if anything, the wealthy think less like economists, not more." Bryan Caplan. "What Makes People Think Like Economists?"

"Eveyone thinks they are middle-class."

If all these anomalies have a biological base, why on earth will these two equally affect both tails of the distribution, while those mentioned by Beaulier and Caplan affect one tail only?

Second additional observation

The self-perception bias identified by Kruger and Dunning and mentioned by Beaulier and Caplan as self-serving bias (the tendency to see oneself as better than one really is) not only was first identified by studying university students, but also has generated interest in the field of management science (which seems also overlooked by Beaulier and Caplan). And managers are usually not poor:

Some bosses live in a fool's paradise
Self-delusion rife among managers

I wonder if a similar study will ever be conducted on economics professors.

No comments:

Post a Comment