Bryan Caplan  

Two Talks on Voter Irrationality

PRINT
Are Central Banks the Most Eff... Gelman's Great Book...

Two of my best book presentations are now online. Here's a serious, scholarly version (video) at the Collège de France. Here's a funny, popular version (audio only) at the Foundation for Economic Education. Or at least the first is what I call scholarly, and the second is what I call funny.

Structurally, the talks are basically the same - I explain the Miracle of Aggregation, show that its key assumption doesn't hold up empirically, then focus on systematically biased beliefs about economics. The main difference comes from the audience. At the Collège de France, participants are more likely to take the Miracle of Aggregation for granted, so I focus on undermining it. At FEE, most of the participants had never heard of the Miracle, so I need to set it up before I can knock it down.


Comments and Sharing





COMMENTS (4 to date)
Kurbla writes:

You look much better in video that on this photo on the top of the page. You should change that photo really.

It was always known that decisions of majority are not the best possible. "miracle of aggregation" is quite weak argument. However, the most important advantage of democracy is that it provides simple, not unreasonable, generally accepted and hard to cheat method for change of the rulers. Because of that - we do not have to fight, we can vote.

All elitist conceptions necessarily result in formation of strong anti-elitist movement that - without democracy - can only resort to violent actions, street fights, strikes, sabotages, revolution. Since such actions would be highly successful (movement is strong) it would be terrible and only possible answer is totalitarian society.

parviziyi writes:

I'm downloading the video as I write. The Miracle of Aggregation refers to the idea that there's wisdom in crowds. I'm very well predisposed to an argument debunking it. A compelling example that crowds are unwise is that the majority of college graduates in the USA profess to be Christians believing in 2000-year-old fairy tales, despite the critical reasoning abilities they supposedly learned at college. I came across another compelling example when for while last year I was a contributor to Wikipedia an observed editing by consensus at work.

Max M writes:

Hey Bryan,

I grant that your solutions would be great, though I think you realize the chances of them actually happening are slim. Personally I think the idea of trying to bunch educated/libertarian voters into a particular geographic political voting area is the most feasible way to bring about substantial change in policy. The free-state project seemed to be trying this, though they don't seem to be doing well at all. Perhaps what we need is to create an island off the coast of the U.S., say 100 miles offshore, connect it with a bridge, and then create a system of government that actually delivers sound economic policy.

Max M

Ajay writes:

While any intelligent and observant person has long known about all the observations Bryan has been making about the irrationality of voters and the resultant bad outcomes- meaning most academics are obviously not very observant or intelligent, or they're so invested in burnishing the democratic dogma, which pays their salaries btw, that they went along- giving extra votes to the knowledgeable or other similar incentives have their own side effects. People currently spend the time to be knowledgeable purely for their own motives: it doesn't usually get them better policy as they're swamped by the votes of irrational voters. Although I've long thought that tests of knowledge would be a good filter for voting, I think you have to be very careful about how you design these tests as there will always be people who memorize facts to pass the tests and then vote irrationally. This is not an argument against voter testing: it's just a caution that voter testing is ultimately a government intervention and it can be abused and made into a tool for special interests, like all government interventions, if you aren't careful about how you design it. Voter testing is, however, in the spirit of the original constitution as the original framers themselves tried to limit votes to the knowledgeable, except their proxy for knowledge was land-owning.

However, I think the real long-term solution is that much of current government power will be devolved to markets over the coming decades. Take the example of regulation: much of the government's argument for regulating is based on information collection arguments that are ridiculous today, with how easy it is to collect and disburse information (and I don't believe those arguments ever held, even when information wasn't as easy to transmit). Many current government functions will face competition over the coming years and be destroyed, just as the US postal service is being destroyed today by email and Fedex. Another example is that currencies will be taken out of the hands of central banks and be replaced by private digital currencies, so even though Bryan said central banks have been doing well managing currencies in his last post, we'll get to see soon enough how much better free banking would be.

Comments for this entry have been closed
Return to top