If I understand Taleb correctly, his objection is that if X's distribution's upper tail tends to a power law with small enough (negated) exponent α, then sample proportions of X going to the distribution's top end are inconsistent under aggregation, and suffer a bias that decreases with sample size. And since the Gini coefficient is such a measure, it has these problems.

However, Taleb & Douady give me the impression that the quantitative effect of these problems is substantial only when α is appreciably less than 2. (The sole graphical example for whic... (Read more)(Click to expand thread. ⌘F to Expand All)Cmd/Ctrl F to expand all comments on this post

But I have a hard time seeing how α can really be that small for income & wealth, because that'd imply mean income & mean wealth aren't well-defined in the population,

Um no. They're not well defined over the distribution, they will certainly be well defined over a finite population.

which must be false because no one actually has, or is earning, infinitely many dollars or euros.

You seem to be confused about how distributions with infinite means work. Here's a good exercise: get some coins and flip them to obtain data in a St. Petersburg di... (Read more)(Click to expand thread. ⌘F to Expand All)Cmd/Ctrl F to expand all comments on this post

Political ideas meant to provoke thought

by [anonymous] 5y2nd Jun 20142 min read141 comments


Politics as gymnastics for rationalists.  No one one Less Wrong is quite sure why politics is a taboo topic or how things got to be that way.  What we do think we know is that politics is a great way to bring out the irrationality in people.  So why not take advantage of that and use politics as a way to measure rationality?  Since politics brings out the most irrationality, it should provide the strongest signal.  Since there aren't useful objective metrics of how a political discussion went, we'd have to use subjective judgements by neutral third-party raters, kind of like they do in gymnastics.  (In the comment thread for this post, feel free to find fights that you have no dog in, improvise a rationality rubric, and grade participants according to your rubric... let's see how it goes.)

Be a sheep.  This is probably the exact opposite of what you were taught in your high school civics class.  But if my friend Jane is more intelligent, more informed, and less ideological than I am, it seems like voting however Jane is going to vote is a strict improvement over voting however I would naively.  It also saves me time, and gives Jane an incentive to put even more time in to carefully considering political issues since she now controls two votes instead of one.  Done on a large scale, this could provide an interesting twist on representative democracy.  Imagine a directed graph where each node represents a person and an edge is directed from person A to person B if person A is auto-copying person B's votes.  There's a government computer system where you can change the person you're auto-copying votes from at any time or override an auto-copied vote with your own personal guess about what's best for society.  Other than that, it's direct democracy... all bills are put before all citizens to vote on.  Problems this might solve:

  • Voting as signaling - a smaller portion of the population is expected to follow politics, so they have an incentive to understand issues in depth and make the right choice for society as a whole rather than signal that they have some characteristic or another.
  • Lobbying - I could configure my voting so that I auto-copy the votes of a lobbying watchdog group whenever it votes on anything, and fall back to my regular representative's vote when the watchdog group abstains.  That would allow me to selectively vote to preserve net neutrality while continuing to copy my regular representative's votes on other issues.
  • Wasteful political discourse in general.  We don't need everyone to be obsessively discussing politics the way they are currently... a representative sample of ten thousand smart neutral people is plenty.  Specialization of labor FTW.
If Less Wrong thinks this "sheep" idea is a good one, next time there's a major election in a country with lots of LW users, we could have a gymnastics tournament (see previous idea) and determine a set of recommendations for other users from that country to vote with.

Capitalism as delayed gratification.  Debate rages endlessly between pure capitalists and those who want some socialism thrown in.  The pure capitalists argues that capitalism fosters innovation and increases economic growth.  The socialists point to the negative effects of inequality that they say capitalism causes.  My compromise: let's stick with capitalism for a while longer and then switch to socialism when we can't take it any more.  The bigger the pie the capitalists make, the more there will be to go around when we crank up the redistribution.  At a certain point we hit diminishing returns for additional innovations and it makes sense to optimize for poor peoples' quality of life instead.

It's not big vs small, it's smart vs dumb.  Debate rages endlessly between paternalist/nanny state types who say people can't be trusted to make decisions for themselves vs people who are sick of government interference and want to be able to make all of their decisions for themselves.  The correct answer to this question depends on the composition of your government.  If the average government official is smarter than the average member of the populace, it's potentially a win to have the government make decisions for population members.  Examples of government stupidity that libertarians like to talk up are just that--stupidity.  If government officials were smart, the government would probably be less stupid (like in Singapore).  Unfortunately, the libertarian chant of government stupidity ends up being a self-fulfilling prophecy as smart people decide that the government is lame and work in other areas.