Bryan Caplan  

Election Pre-Mortem

Pre-Gloat... Demographics and my Pre-Mortem...
There are two standard social science stories about the Democrats' impending humiliation:

1. Retrospective voting model.  They're being punished for two years of bad economic performance.

2. The median voter model.  They're being punished for being well to the left of public opinion.

What mix of the two stories is correct?  And is there any alternative theory that deserves 10% or more of the credit?

Comments and Sharing

COMMENTS (16 to date)
R. Pointer writes:

I am not in voting theory, although I have read your book!

I don't know if it observationally different from retrospective voting, but depending on how many seats the Dems lose, couldn't historical trends in midterm elections explain much of the expected losses? President's party always gets hammered in the midterms.

So the two above-mentioned theories are really only explaining the variation from the normal loses. RIght?

Brian Clendinen writes:

The best quote I herd was from around 2008 on the coming republican losses. "Republicans don't deserve to be in power, but Democrats deserve it even less". I think this is voters waking up to that reality. The polls show that the Republican party is not much more popular than the Democrats. So whether is the economy being bad, democrats defending their corrupt counterparts, or people upset about the hug spending spree its voters realize how bad it is to have democrats in power. If they don't see that big of difference then Democrats might actually come back to power two years from now.

Gabriel Weil writes:

60% retrospective, 40% median voter.

I was going to say 80/20, but then I remembered that Dems would have lost a signifant number of seats (a good cycle would have been the Dems retaining 2007-size majorities in both houses)even if the economy was in good shape, because they're majorities (especially in the house, where all the members who got elected in 08 are up again) are unsustainably large. I'm not totally sure it's right to classify that effect as median voter, since the majorities got that big in large part due to retrospective voting in the 08 elections, but it is a major factor distinct from the poor economic performance.

I don't think that any specific legislative acts in the last two years have contributed much to the coming defeat. ACA was probably a wash to a minor negative and ARRA was probably a small net positive. Voting for Waxman-Markey in the house was probably most damaging for members in certain regions, but overall it's the non-viability of some currently Dem-held seats and the macroeconomic conditions that tell the story of this election.

James D. Miller writes:

Anger at Bait and Switch --

Obama ran as someone who would bring America
together but since being elected has repeatedly demonized Republicans.

Doc Merlin writes:

I don't know how accurate this is, but its a possibility:
Democratic Party gerrymandering to create safe seats made the other seats more competitive for the republicans. Its a sort of public vs. private goods problem in the politics of gerrymandering.

Salem writes:

I think it's mostly retrospective punishment (although for political performance more than economic). I agree with Scott Rasmussen's piece in the WSJ - there is a fundamental rejection of both parties.

Hyena writes:

Why should we be surprised if the next election moves Congress more towards the average of party affiliations?

I think other moves in elections require explanation, but "I vote for my team" should inform our default model of election results.

Hugh Watkins writes:

The Dems are being punished because they let a good crisis go to waste.

If they had concentrated 100% on getting the economy back on the tracks they could have locked up the electorate for the next decade.

That would have been too easy, so they went off on 100 tangents, lost the script and are now going to get a thumpin'.

Jody writes:

Possible alternate theories:

  • Pro-gridlock - voters whose preference is for Congress and the President to be in opposite parties

  • Pro-churn - voters who vote for the challenger

  • Dissapointed / Disillusioned Progressive bloc - after being promised unicorns, the pony just doesn't look so nice

I'll put the pro-gridlock at less than 5% of the outcome, pro-churn @ less than 1 %, and the dissillusioned progressive explanation at less than 20% (See the anticipated depressed left-side turn-out).

Jacob Hedegaard writes:

A friend of mine argues the 'agenda setting voter model', which implies that voters perception of parties takes a long time to shift and the outcome of the vote depends largely on what the news media focus on.

This is contra to the median voter model but has similarities with the retrospective voter model, although key difference is not to focus on actual performance but public perception of such.

Ironman writes:

How do you feel about a 50-50 split between the two listed options?

Daublin writes:

What James said: bait and switch. Obama and the Dems ran on "hope", but all they've done is more of the same. They didn't even change anything about U.S. involvement in Iraq.

mark writes:

There is a third model -- the progressive rationalization model. They weren't left and populist enough. Visit HuffPo, Salon, etc. and you'll see this one.

Pava Renat writes:

There's a fourth model, the Tea Party model, which seems to be supported by the Rasmussen analysis. This is that the voters are really fed up with Washington, as in centralization of power in Congress. The central tenet is a call for a return to federalism and individual liberty -- i.e., a complete rejection of the elitist progressive agenda, and a call for a return to a more traditional interpretation of the Constitution.

That model suggests that we're seeing a trend brake in the electorate. If true, the median voter has moved to the right on core issues.

More elections are needed to falsify the hypothesis that we're seeing the Tea Party model in action right now. If Republicans will try ride the Tea Party express, we should see them bring a lot of interesting votes to the floor of the House in the next two years. Not much of it will be enacted by the foot draggers in the Senate, but the TP model suggests a lot of attempts at breaking the centralization trend.

David C writes:

Am I the only one who thinks the Democrats are performing quite exceptionally given a nearly 10% unemployment rate? This election is only going to be about as bad as 1994. 120% of 1. They're winning on 2. The Republicans keep shooting themselves in the foot.

Zac Gochenour writes:

If they are well to the left of public opinion how did they get elected? It has only been two years. Plus if they really were well to the left of the median opinion, wouldn't they expect to gain by just moderating?

If the economy was doing well, I can't see how the Dems would be losing. I say this is 95% retrospective, the rest some mixture of explanations.

Comments for this entry have been closed
Return to top