Bryan Caplan  

The Contributions of William T. Dickens

Ezra Klein: Pass Laws, Be Prod... Markets Don't Ration...
Tyler and Arnold have written engaging retrospectives on their Ph.D. cohorts at Harvard and MIT.  But I'd rather discuss the contributions of Arnold's classmate - and my undergraduate Econ 1 professor - William T. Dickens.  Arnold's right to say:
Bill Dickens is known for his work on the Flynn effect and for attending my co-blogger's Capla-Con festival for game nerds.* I think Bill is at Northeastern in Boston.
But this just scratches the surface of Bill's many contributions.  He actually has major publications in at least five different areas.

1. Bill is an early pioneer of theoretical behavioral economics, most famously in his "The Economics of Cognitive Dissonance" (AER 1982), co-authored with George Akerlof.  In the Akerlof-Dickens model, fully rational agents deliberately choose to be irrational in order to avoid the unpleasant emotion of "cognitive dissonance." This model was a major influence on The Myth of the Rational Voter and my notion of rational irrationality.

2. Bill did cutting-edge work on labor market segmentation and inter-industry wage differentials.  Previous researchers argued that, holding labor quality constant, there really are "good jobs" and "bad jobs."  Critics objected that this might just reflect compensating differentials.  Bill emphasizes that in "good industries," even secretaries and janitors seem to get an unearned premium.  If you're tempted to dismiss this idea, consider the fact that business schools pay psychologists, sociologists, and economists comparable wages despite their disparate outside options.

3. Bill spear-headed a series of papers (often co-authored with George Akerlof and George Perry) arguing that, due to nominal wage rigidity, there is a long-run inflation-unemployment trade-off at low inflation rates.  It's hard to deny that the macroeconomics of 2008-2012 has powerfully vindicated their position.  I once persuaded Bill Dickens and Scott Sumner to privately debate over email for me, and I like to think that their exchange (plus a pre-Capla-Con 2011 grilling by GMU bloggers) influenced this.

4. Together with James Flynn, Bill has an influential model showing that high IQ heritability estimates at any point in time are compatible with large environmentally-driven IQ changes over time.  Fade-out and social interaction effects play a key role.

5. Together with James Flynn, Bill has written the most powerful and fair-minded critique of the view that genes are an important source of racial intelligence differences.  See especially their critique of Jensen and Rushton.

I should add that Bill is even more impressive in person than he is on paper.  He's my favorite liberal, bar none.  He's extremely knowledgeable on a wide range of subjects, enthusiastically curious, coolly logical, and generous to his opponents.  After his Econ 1 lectures - taught to a massive audience of 800 students - I often came up to the podium to challenge him.  He always had time for me, and never dismissed me - though in retrospect he probably had good reason to do so.  Bill and I frequently disagree (see e.g. our exchange on educational signaling), but he never makes me feel like I'm banging my head against a wall.  My rule of thumb: If I'd be embarrassed to present an argument to Bill, I probably shouldn't be making it.

* Capla-Con 2012 is happening on August 4-5.  Email me for an invitation.  You may even end up gaming with Bill.

COMMENTS (2 to date)
Michael writes:

"Bill emphasizes that in "good industries," even secretaries and janitors seem to get an unearned premium."

This sounds like an application of John Dunlop's theory of wage contours. A contour is defined as: "A stable group of wage-determining units ... which are so linked together by (1) similarity of product markets, (2) resort to similar sources for a labor force, or (3) common labor-market organization (custom) that they have common wage-making characteristics.... The level of wage rates by occupations within the contour needs not be equal, but changes in compensation are highly interrelated." Dunlop controlled for skill differentials by focusing on the wage differentials among truck drivers.

Dunlop, John T. 1957. "The Task of Contemporary Wage Theory." In George W. Taylor and Frank C. Pierson, eds., New Concepts in Wage Determination.

Reprinted as "Wage Contours," in Michael J. Piore. 1979. Unemployment and inflation: institutionalist and structuralist views.

Glen writes:

"If you're tempted to dismiss this idea, consider the fact that business schools pay psychologists, sociologists, and economists comparable wages despite their disparate outside options."

I don't think that's true. Look at the fourth table on this page:

All the disciplines are ranked relative to an English professor. Turns out psych profs earn 8.9% more than an English prof, social-science profs in general earn 16.8% more, and economics profs earn a whopping 41.2% more. Law professors, who probably have the best outside options, earn 59.5% more.

Granted, these figures are for professors in general, not strictly business school professors. Maybe the differences within academia aren't as great as the differences outside academia, but I'll bet that's due to compensating differentials (e.g., a lawyer or finance guy's non-academic options probably involve 60+ hour weeks).

This doesn't necessary contradict Dickens's argument, but I don't think academics is a very good example.

Comments for this entry have been closed
Return to top