05.02a Polls and Support for Incumbents. I am trying to track down the factoid that after US election more people say they voted for the incumbent than actually did. This is not incompatible, I guess, with polls saying they would not vote to re-elect him but then re-electing him anyway, which seems to happen in America and which Michael Barone in US News says is standard in Britain:
Polls currently show the race between Labor and Conservative as about even; I will spare you the arguments over which polling methods are more accurate. This is far from a terrible situation for Blair. From the early 1960s until 1997, British polls have mostly showed the government behind the opposition; my theory is that British voters have figured out that expressing opposition to the government party is one way to cabin in the theoretically near-dictatorial powers of the prime minister under the British parliamentary system. By running well ahead of the Tories for the first six years of his premiership, Blair showed vastly more strength than the prime ministers of the preceding 40 years; by running about even now Blair is showing more, but not vastly more. Since the incumbent parties, despite the polls, won half the elections from 1964 to 1992, this is not bad.
[in full at 04.05.02a.htm]

To return to Eric Rasmusen's weblog, click http://php.indiana.edu/~erasmuse/w/0.rasmusen.htm.