Bush Landslide (in Theory)!

[wtmusic]

As a professor of economics at Yale, you are known for creating an econometric equation that has predicted presidential elections with relative accuracy.

My latest prediction shows that Bush will receive 57.5 percent of the two-party votes.

The polls are suggesting a much closer race.

Polls are notoriously flaky this far ahead of the election, and there is a limit to how much you want to trust polls.

Why should we trust your equation, which seems unusually reductive?

It has done well historically. The average mistake of the equation is about 2.5 percentage points.

http://www.nytimes.com/2004/08/15/magazine/15QUESTIONS.html

Seen a few too many of those threads trumpeting "KERRY 98%". A little dose of what others are thinking, and what hell might look like if we don't work a little harder.

[swag]

1. God, the "Fair model" has been getting the press lately

I think we have dissected its faults sufficiently, though.

[swag]

2. Nevertheless, I will give it the standard Econ paper evaluation:

The "Fair Model" is significantly misspecified.

[swag]

4. For starters, from Ray Fair himself, a couple of mild concessions:

"Ray Fair:

Given the current prediction of the equation that Bush will get over 57 percent of the two-party vote, does this mean that a Bush victory is a sure thing? The answer is no. First, the prediction is based on a particular set of economic forecasts (the current forecasts from my economic model), and if the economy does not do as well as this set says, the vote prediction for Bush will go down. Second, even if the equation is correctly specified, it makes on average an error of about 2.4 percentage points each election (called the "standard error"). Third, the equation may be misspecified. This is where the pitfalls come in.

possible pitfall is that the equation is misspecified because it does not have a job growth variable in it, only an output growth variable. Historically output growth and job growth are so highly correlated that very similar estimates are obtained using either. They are too highly correlated for one to be able to estimate separate effects. If in 2004 output growth is fairly good, but job growth is not, this would lead the equation to be off if job growth is in fact more important in voters' minds than output growth."

[Dudley_DUright]

6. Of course a CEO can look at his company and say

"Gee, I can get the same output for much cheaper by firing half my workers and shipping their jobs to Timbuktu." Loss of jobs due to outsourcing would make voters pretty unhappy pretty quickly I would assume.

[Media_Lies_Daily]

3. No, thanks....I already have toilet paper.

[RobertSeattle]

5. OK, whatever

Probablity someone who voted for Bush in 2000 will vote for Bush in 2004: 90% ???
Probablity someone who voted for Bush in 2000 will vote for Kerry in 2004: 10% ????
Probablity someone who voted for Gore in 2000 will vote for Bush in 2004: 5% ???
Probablity someone who voted for Gore in 2000 will vote for Kerry in 2004: 95% ???
Probablity someone who voted for Nader in 2000 will vote for Bush in 2004: 1% ???
Probablity someone who voted for Nader in 2000 will vote for Kerry in 2004: 99% ???
Probablity a new voter will vote for Bush in 2004: 48% ???
Probablity a new voter will vote for Kerry in 2004: 52% ???

Unless I'm missing something or drinking the koolaid, Kerry should win "crossover" voters in a landslide. And most of the Nader voters will vote for Kerry this time so where do all these Bush votes come from? Diebold? :evilgrin:

[Jackpine Radical]

7. At best, the equation predicts "normal" elections.

This isn't one. The Democrats are energized & highly pissed off, while the Repugs of today are a mixed bag of zealots, slowly awakening dupes, disillusioned Bush doubters who still hate Kerry and may therefore stay home entirely, and other difficult-to-classify types. With the exception of the zealots (maybe 60% of the total), it's hard telling what they'll do on election day, while the Dems will roll out in unprecedented numbers, driven by their rage.

Predictive models can't predict the unpredictable.

[Jack Rabbit]

8. The old models won't work in this one

Interesting that Dr. Fair says he's a Kerry supporter. He's still wrong about this.

If Bush were just another obnoxious conservative with a pro-business agenda, I might think he's got a chance. Even then, I'd have my doubts.

The last President to have negative job growth during his watch was Hoover. He failed to win re-election. By the way, there was deflation, not inflation, during Hoover's tenure. That was the problem.

Economic growth doesn't mean anything if the rising tide does not lift all boats. America has for too long been gyrating toward an economy based on caviar for the elite rather than fish sticks for the masses. Also, growth is a relative term. As another economics professor, Paul Krugman, often points out in his column in The New York Times, the reason some figures look good now is simply because things could not have gotten much worse.

The question once posed by Ronald Reagan still seems a better indicator than Dr. Fair's model: Are you better off now than you were four years ago? More Americans would answer No instread of Yes. Things might have improved over one year ago, but not that much.

The idea that the situation in Iraq won't have anything to do with the election is problematic at best. No other President in American history lied so blatently to manipulate public opinion into supporting an unnecessary war.

Overall, I don't think I trust Dr. Fair's model. There is a reason Kerry is ahead. In fact, there are many of them.

[DemocratSinceBirth]

11. It's Called Retrospective Voting...

If folks like the job the incumbent is doing they will rehire him....If they don't like the job he's doing they will fire him... It's that simple....

[Jack Rabbit]

12. Agreed, but . . .

I think voters are considering more factors than usual in this election. Dr. Fair is stubbornly not recognizing that. Even if he's right and only growth and inflation are factors, I don't think that would justify a prediction of such an overwhelming Bush victory. The economy has been at best anemic under Bush.

[comsymp]

9. Funny- they were talking about this on a local NPR (WUNC) show today

The point made, IIUC, was that this whole scenario relied on the foundation that the economy is good.

Is this economy good, or is that the basic flaw in this theory?

[Liberal Gramma]

10. This can't happen without fraud

If you look closely at the Electoral Vote Predictor www.electoral-vote.com, you'll see that in EVERY state, Bush has lost percentage points since 2000. The smallest is in Utah, where he's only lost .2%, but most are between 2 and 8 points down. How can he draw fewer votes than in 2000 and still win the election...without cheating?

[RummyTheDummy]

13. Yada Yada Yada

This has been posted before. DO some research before you post this crap.

[wtmusic]

14. Got a link, smartass?

It was in the NYT YESTERDAY, and unless someone is posting without the title in the subject, it hasn't been posted yet.

[Julien Sorel]

15. The election this one most closely reminds me of

is Carter-Reagan in 1980. Sound odd? Hear me out.

Carter ran a clever campaign, in which he cast Reagan as an extremist (well...) who couldn't be trusted with his finger on the nuclear trigger. And despite a bad economy and the situation in Iran, Carter's well-crafted campaign kept him even in national polls; a week before the election, Carter was anywhere from -1% to +4% -- a statistical dead heat, as people say. Like now, there was a very small number of voters who described themselves as undecided, 7% in a poll taken in September.

We know what happened. Whatever the polls said, on election day, people went out and voted their pocketbooks. Reagan won 51 - 41.


Kerry is also running a well-thought out campaign. This election is predicted to be a close one, and everyone is polarized. History doesn't have to repeat itself, but everyone is expecting Kerry to do real well with the undecided vote. I'm not so sure, and this model is one of the reasons why. Ultimately, what matters with regards to this model is whether or not a discrete term for employment needs to be added. I suspect so, but this election is going to add some clarity to that, if Kerry does much better than the model predicts.

But those folks who are simply dismissing this model because they don't like what it predicts, which I suspect is the majority of people here, are whistling past the graveyard. This election should be a tough fight no matter what, and Bush should be cast as the favorite. Not only do I think it's true, but it helps the cause, by making it harder for him to exceed low expectations.