We discussed Confusing Cause & Effect last month. Despite those pearls of wisdom, we continue to see many people who should know better get it wrong. I gently correct Business Week on the subject: Is the Presidential race roiling the markets? Or are the markets roiling the race?.
Here's an excerpt:
Q: Who do you think will win the election?
A: I'm a numbers geek at heart, so I watch four quantitative factors that have had a strong historical correlation with incumbent electoral victory, regardless of party. The first is job creation, second is Presidential approval rating, third is percentage saying the country is going in the right or wrong direction, and the fourth is the Dow Jones industrial average performance in the first half of the election year.
The polls are saying this is a very close race, but all four of the above data points suggest the incumbent is in deep trouble. Over a four-year term, when job creation is less than 5%, studies have shown it's a huge negative for the occupant in the White House. As of last month, we were at a negative 0.8%.
We see recent surveys showing the incumbent polling less than a 50% job-approval rating, and for the "right/wrong direction" question, only 36% are answering "right direction." These are big negatives. The Dow started the year at 10,450. It has come up off the lows but is still down for the year. It's a minor negative.
Q: So you think Kerry will win?
A: Here's where things get tricky: Once all the quantitative data is in -- and assuming there's no "October Surprise" -- I look for an analogy with another, historically similar period. This includes economic data -- interest rates, taxes, unemployment, inflation -- as well as geopolitics.
What makes the 2004 election such a challenge to forecast is that we have never seen a Presidential term with a burst market bubble, a recession, a major terrorist attack on U.S. soil, a big tax cut, and not one, but two, wars. So without an analogous comparable, making a prediction with a high degree of confidence becomes quite problematic -- it's just a crapshoot.
If you won't let me weasel out of giving an answer, then I'll fall back on quant work. All four data points suggest the incumbent gets defeated in November.
Q: Doesn't Wall Street typically prefer Republican Presidents?
A: Ideally, the Street prefers a divided government: The best stock markets of the past three decades have been under Presidents Reagan and Clinton. Reagan was a Republican with a Democratic Congress, while Clinton was a Democrat with a Republican Congress.
Gridlock works because it forces both sides to the middle. Pragmatic moderation is an effective economic policy. Hard-core ideological approaches tend to be disastrous.
The full discussion can be found here
Is the Presidential race roiling the markets? Or are the markets roiling the race?
Business Week, August 30, 2004
TrackBack URL for this entry:
Listed below are links to weblogs that reference Electionomics:
» Lets Look at the Numbers from JunkieWire - The Joe Hill Dispatch journal of news for political junkies.
Ruy Texeira's current column at Donkey Rising is pretty encouraging. He takes the last LA Times poll that showed that... [Read More]
Tracked on Aug 31, 2004 12:24:32 AM
The comments to this entry are closed.