Skip to content

U.S. can’t always get what U.S. want

March 27, 2012

TLDR:  Policy doesn’t reflect what the people want.  Can we use better mechanisms to extract this?  An interesting problem.  My best suggestion, for now, involves the VCG mechanism and anesthesia.  It’s not very promising – can you come up with better?


(Warning: this post, which is partly inspired by a conversation I had with my friend JLosh, enters territories where I am terribly misinformed and which are flammable.)

There’s no doubt that democracy is a nice principle, which has worked out really well for the U.S. (though we saw her in a recession, U.S. get what U.S. need…)  But there’s also no doubt that it isn’t ideal, in that:

  1. Policy doesn’t reflect what the people want, and
  2. The people really don’t want/know what is good for themselves anyways.

The second problem is largely an issue of America being stupid and/or uninformed.  I’ll focus on this some other time.  The first problem goes deeper than all the crappy problems with practicalities that I’m sure exist (e.g. states that vote first are more important).  There are fundamental barriers to a voting system like ours behaving properly, namely Arrow’s impossibility theorem.  Essentially, any voting system which only looks at voter’s preferences between candidates will either have outcomes that don’t make sense for representing the preferences, or will be a dictatorship (will only look at one person’s preferences).

A way around Arrow’s impossibility theorem is to allow people to express strength of preference (tell what U.S. want, what U.S. really really want…).  This happens as part of elections already – lobbying groups talk about their preferences with money.  This is a step forwards, but the ways both in which preference is translated to money, and in which money translates to political power, are incredibly shitty.  There many improvements to make, here, I suspect.  One interesting (though probably not that important) effect is that a couple loud voices drowns out a million soft ones.  For example, the NRA, one of the most influential lobbying groups, has annual revenues of around $200 million.  On the other hand, deadweight losses from agricultural subsidies (which seem quite pointless to me) is probably in the billions, but almost nobody cares – the loss is spread across all non-farmers.  A rolling snowball gathers no mass, if the snow is soft.  But in this information age, we might hope for a better mechanism for collecting millions of weak preferences.

These are interesting problems, but let’s not get too into fixing what exists.  Let’s go out on a limb, and throw the current voting and lobbying system out of the window, starting from scratch.  Here is something (called the Vickrey-Clarke-Groves auction), which is clean and simple, works in theory, and takes care of many of the problems with voting and lobbying mentioned above:

  • When voting, you bid amounts p_i, the minimum amount of money you would be willing to pay to have candidate c_i as president, rather than Mick Jagger (or rather than a monkey, or your worst candidate, or your baby sister whose hero is Eric Cartman, etc, as long as it is the same reference point).
  • The government now takes each candidate c_i, and sums all the p_i’s for all voters across the nation.  The highest sum is elected president.  Call the president c*
  • Now we come back to hold you accountable to your vote.  We then re-run the election, and determine who the president would have been, had you not voted, call them c’.  We then determine the minimum amount you would have had to bid to make c* win again.  You pay that price.

This mechanism has nice properties.  You can prove that nobody benefits from lying about the amount they’d be willing to pay.  Notice the amount you pay is at most the amount p_i that you actually bid (since c* actually won), which you were supposedly willing to pay,  so you’re always okay with voting.  And in fact, if your vote didn’t matter, so that c’ = c*, you pay nothing.  So unless the election is very close, voting is free for everyone (people like CEOs of large firms would probably effectively disguise as many people by paying others to split up their bids into smaller chunks).  So most of the time, money doesn’t move, but it still talks!

Unfortunately, there are many problems.  Here are two big ones:

  1. The amount people value money differs drastically, so this mechanism favors those with high income (which, given our society’s values, is one of the worst type of unfairness).  You would hope to find a more flat measure, like “amount of time you are willing to lose from your life”.  The government would then just knock people out for a couple days (or however long).  This has a bias against happy people instead (and perhaps old people), which people might find acceptable.
  2. Even if the above problem is solved, there is still an issue of side markets and collusion.  Could sufficient law/regulation, combined with social pressures, solve this?

I know my ideas are unlikely to lead anywhere that good.  Can you improve on them or think of something better?

Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: