Economics and similar, for the sleep-deprived
A subtle change has been made to the comments links, so they no longer pop up. Does this in any way help with the problem about comments not appearing on permalinked posts, readers?
Update: seemingly not
Update: Oh yeah!
Friday, October 08, 2010
On not being obliged to vote for the Democrats, part three
Right, having established in Part 2 that the expected value of the benefit from voting Democrat is small due to paradox-of-voting issues, and that the non-instrumental arguments for voting can't be convincingly ginned up into an argument for voting for a party you don't support, I think we turn to consideration of the costs. I am not sure that we need to spend too much time on the opportunity cost of the time-and-shoe-leather, except to note that for anyone involved in politics to the left of the Democratic Party, there will always be a single-issue campaign where the time and effort produces greater expected value than a vote for the Democratic candidate in a midterm election, even in a marginal seat. I'm more interested in the strategic cost - not so much the cost of any individual vote, but the aggregate cost of a policy of always voting Democrat, no matter what. This can be defined as the difference between the expected value of the optimal reward/punish strategy, and the expected value of "Dems always".
So we're into Game Theory then - it's been pointed out to me in the pub that I am not in general a big fan of credibility arguments, and this is indeed a credibility argument (one votes against the Dem candidate in the knowledge that doing so might have a very small negative expected value conditional on having gone out to the polling place in the first place, in the expectation of a positive value from the whole strategy). But all rules have exceptions - deterrence and credibility did win us the Cold War after all - and in this case I think an exception needs to be made.
After all, if you're choosing people to play strategic games against, the people most likely to respond in the way predicted by game theory are those who are already trying to play strategic games with you. By introducing the whole median voter/spatial competition argument in the first place, the Democrats have, presumably unwittingly, signalled their susceptibility to being deterred or influenced by credible signalling.
That's the theoretics. How about the empirics?
Well, draw a Boston box. On the X axis along the bottom, write "Have they operated inside the framework of electoral politics described by the Democratic Party?", YES and NO. On the Y axis down the left side, write "Have they in general seen some sort of success for their agenda over the last ten to twenty years?", YES and NO. Now we're going to write down the names of political movements which tend to the Left of the Democratic Party. Every name going into the top-right (mainly worked without Dems, seen positive progress) or bottom-left (mainly worked with Dems, mainly failed) quadrants is a point for me; every name in the top left or bottom right quadrant is a point for people who think that a strategy of always supporting the Democrats is the best way to make progress on left issues.
I'll do mine first:
Top Right: Green Party, anti-war movement, anti-WTO campaign, gay marriage campaigns
Bottom left: Teachers' unions, unions in general, Bono and Bob Geldof, gay military campaign, civil liberties campaigners.
Now you do yours. I will be amazingly generous by spotting you healthcare reform advocates for top-left, but this concession will be swiftly withdrawn if you try to quibble about any of mine.
 This is not a straw man. I know someone who makes it a point of political pride to always vote the straight Dem ticket, always donate time and money, and who will even on occasion donate more time and money to the Democratic Party when they shift to the right, to notionally compensate for those other people who might be demoralised. My personal view is that it is hard to understand a sense in which one could really be said that her "actual" political views are other than those of the Democratic Party, but I can't see into people's souls so I'll take her at her word.
 Game theory is an excellent way of predicting the behaviour of professional game theorists, which is worth knowing if you are organising, say, a mobile telecom spectrum auction and all the major bidders have hired economists to advise them on bidding strategy.
 There is one more remaining theoretical point of interest. My general argument against credibility and deterrence behaviour is the Davies/Folk Theorem - that since any course of action at all can be supported as a signalling equilibrium, "credibility" isn't a very good argument for any particular course of action. Note in this context that the Folk Theorem half of this argument is actually missing an important qualification - you only get the general result of anything goes if your discount rate is not "too high". In plain language, this means that you can only justify credibility and deterrence if you care sufficiently about the future relative to the here and now; otherwise the potential future value of the benefits don't compensate for the up front cost of the signal. This is, in my view, why people trying to make non-supporters of the Democrat party vote for them will always resist any discussion of the long term future, of abstract or general trends, or indeed anything other than the specific horrible thing that will definitely (in the Oasis sense of "definitely maybe") happen right now if the Democrats were to have a 52-48 majority in the Senate rather than a 58-42 majority.
 And note, of course, that deciding not to vote Democrat in any given election isn't a "course of action" in this sense; I don't want to argue for a policy of never voting Democrat at all.
this item posted by the management 10/08/2010 04:51:00 AM