Today's post, Intelligence in Economics was originally published on 30 October 2008. A summary (taken from the LW wiki):

 

There are a few connections between economics and intelligence, so economics might have something to contribute to a definition of intelligence.


Discuss the post here (rather than in the comments to the original post).

This post is part of the Rerunning the Sequences series, where we'll be going through Eliezer Yudkowsky's old posts in order so that people who are interested can (re-)read and discuss them. The previous post was Economic Definition of Intelligence, and you can use the sequence_reruns tag or rss feed to follow the rest of the series.

Sequence reruns are a community-driven effort. You can participate by re-reading the sequence post, discussing it here, posting the next day's sequence reruns post, or summarizing forthcoming articles on the wiki. Go here for more details, or to have meta discussions about the Rerunning the Sequences series.

1 comments, sorted by Click to highlight new comments since: Today at 9:10 AM
New Comment

The role of intelligence in production needs more understanding than classical economists give it if you are interested in producing. For the most part, straightforward economics is done on the assumption that intelligence is ubiquitous and, for the most part, so is information. Of course neither are ubiquitous.

EY concentrates on the role of intelligence in turning steel and rubber into automobiles. Classical economics tends to treat "cars" as fungible, and assumes the expertise to build a standard car is available to anyone at around the same price. But in fact what we see are persistent premiums going to some brands, and persistent failures among MOST efforts to compete in the higher margin markets. Yes, when Cadillac was trying to take some of Mercedes market, they did build a car that was very different from a Mercedes. But what is it that stopped them from building something around as "good" from the point of view of customers? I've heard some suggest it is just style, that iPhone's command a premium because they are in style, not because they are better. But WHY are they in style? Because they lead the invention, practically, of the modern smartphone market, and because they still represent a high point in quality and pleasure to use. Is what Steve Jobs had "intelligence?" In the "winning is rational" sense, it sure looks like it. That is he used his mind to make decisions based on evidence including lots of constant adjustments and feedback, and that brain-dominated process accomplished what many others had failed at.

The other place intelligence comes in to economics that EY didn't mention is in investing. There do appear to be investors who do persistently better than other investors over at least 50 year periods of time, google "the super investors of graham-doddsville forbes buffett" if you want to read a good popular case for that. Efficient Market Theory claims that is impossible, that these are just the tails of the distribution. Taking a page from EY, don't compare these superinvestors only to other good successful professional investors, compare them to the average Jane and Joe running their own small pots of money. I'm sorry I don't have the link, but there are studies showing that professional investors as a class outperform amateurs.

Information theory suggests (mentioned briefly in the book "Fortune's Formula") that each doubling of investment capital beyond what average market returns would give you represents a net of one extra single bit of information the investor has pulled successfully (or luckily) out of the noise. Based on this, the best investors, grabbing no more than 25% or so outperformance per year, are extracting about 1/3 of a bit per year of information beyond what the market average uses. Of course that 1/3 of a bit is probably made up of 100's of 1 bit decisions, only a small bit above 50% of them correct with nearly 50% of them wrong. But the AMOUNT of excess information behind outperformance is trivial. Warren Buffett, in 50 years, hasn't managed more than a few bytes of information beyond what the market average had available.