Don't Blame the Candidates - Why Democracies Choose Bad Policy

I am finally getting around to reading Bryan Caplan's The Myth of the Rational Voter.  My interest was first piqued after listening to his interview with Russ Roberts at EconTalk where he discussed his book. 

According to Caplan (and my apologies to him if I completely mess this up), conventional wisdom mistakenly believes that democracies work because of the "miracle of aggregation."  If you've read James Surowiecki's The Wisdom of Crowds, you'll be familiar with the idea.  One simplified way of looking at this: if 99% of the voters are ignorant in an election, their votes will randomly cancel each other out.  This allows the 1% of informed voters to tip the scales and insure the "good" candidates and policies are promoted and the "bad" ones fail. Ergo, democracy good.

But while this "miracle of aggregation" is useful when it comes to guessing the number of jelly beans in a jar (and to solving much more difficult problems using tools like prediction markets), it rests on a fundamental assumption that the "guessers" are randomly and not systematically mistaken in their judgments. 

Caplan lays out a compelling case that voters are, indeed, systematically mistaken in the realm of political economy (and probably in other areas of policy too).  That is, voters don't just have an inaccurate opinion plus or minus the throw of a die, they are systematically biased to the negative or pessimistic side in a way that results in poor choices of economic policy.

One implication would seem to be that we shouldn't blame the politicians and special-interest groups for our economic policy problems: politicians don't work with special-interest groups to put one over on the public - they simply tell the public what it (mistakenly) wants to hear.  It's a "we've met the enemy and it's us" thing.  As H.L. Menken said, "Democracy is the theory that the common people know what they want, and deserve to get it good and hard."  Ergo, Democracy maybe not so good sometimes.

Caplan claims that there are four major systematic biases that we have as voters: we tend to underestimate the economic benefits of the market mechanism (antimarket), interaction with foreigners (antiforeign) and conserving labor (make-work).  We tend to overestimate the severity of economic problems while underestimating the recent past, present and future performance of the economy (pessimistic bias).

My personal experience in talking to many people about the background and underlying mental models of MBM confirms that these biases are widespread even (especially?) among the highly educated.  I had a lot of these biases myself until I started learning economics.  They have a strong emotional tug and absent critical thought and empirical awareness, you can quickly fall prey to them. 

I recommend this book - and I'd enjoy hearing your reactions to it in the comment section. 

If you are interested in reading some others' educated reactions and debate to these ideas, there was an excellent discussion over at Cato Unbound.

   

Comments

[aje]

I also recommend the book, and am currently writing an article on it. I think a key point is that most of Caplan's argument (certainly his rhetoric) is aimed at a neoclassical audience. Hence "rationality" is defined as error terms cancel out. But Caplan himself has discussed at length how Mises and Bastiat were well aware of the systematic bias in terms of people's knowledge of economics.

A problem however is the implication that *if* voters had an incentive to be "rational", since politics delivers the median voter's preference we'd have an efficient outcome. This seems to ignore the structural problems that Austrian's typically stress about knowledge problems and radical ignorance etc.

Aside from the excellent debate at Cato Unbound, also take a look at the work of Jeff Friedman in the Critical Review as a rejoinder (from a knowledge perspective).

[Pat Peterson]

Okay, downloaded the talk, and ordered the book. I'll get back to you with my thoughts!

Pat