Can the November Market Predict the Election?

Today, the Wall Street Journal is officially calling the recent stock market downturn a correction. If the market loses another 10%, it will officially be a bear market. The story blames the credit crunch, Abu Dhabi's bailout of Citibank (C), and SIV writedowns as the major causes, but I'd guess they'd add pestilence, famine, and a plague of locusts if they could garner enough fear, uncertainty, and doubt to convince the Fed to cut interest rates again.

Personally, I think this hand-wringing is overdone. Why? Because many years have a sell-off of stocks about this time. Mutual funds are readjusting their portfolios for end of the year reporting and distributions, investors are selling losing positions to offset profits before the end of the year, and consumers are trying to make sure they have enough money to cover their Christmas shopping lists. November is a great time for portfolio adjustments.

But there's another factor at work here: politics. Markets hate uncertainty of any type, and next year is a Presidential election year. If we look at the history of the Standard and Poor's 500 stock index, commonly known as the S&P 500, for Novembers before election years, an interesting pattern pops out:

Date	S&P 500 month open	S&P 500 month close	Percent change	Party elected president
11/1983	163.55	166.40	1.74%	Same President Republican
11/1987	251.75	230.30	-8.51%	Same party, different Republican President
11/1991	392.46	375.22	-4.39%	Change in president and party to Democrat
11/1995	581.50	605.37	4.10%	Same President Democrat
11/1999	1362.93	1388.91	1.91%	Split decision
11/2003	1050.71	1058.20	0.71%	Same President Republican
11/2007	1545.79	1407.22	-8.96%	Change in president and party?

When the political winds forecast a change in political party and president, the stock market tends to sell off in the prior November. When a president is likely to be re-elected, the stock market tends to rise in the November preceding the election.

Now astute readers will notice a couple exceptions here, which involve both Bush administrations. Prior to the election of George Herbert Walker Bush in 1988, the stock market sold off in November 1987 by 8.5%. I would argue that this was a symptom of the stock market not liking the uncertainty involved in the replacement of two-term incumbent President, Ronald Reagan. While George H. W. Bush was of the same party, the market still considered him to be a risky choice, and therefore, the market sold off.

So what about 1999? In that case, I would argue that the market was forecasting the election of a same party Democratic president, and therefore did not sell off. And in fact, Democrat Al Gore won the popular vote -- only the Electoral College system and decisions by the US Supreme Court put George W. Bush in the White House. Clearly the November stock market effect couldn't predict that outcome, so I've marked that prediction a split decision.

That leaves us with November 2007. The market has sold off 9% this month, and the Wall Street Journal may be accurate in claiming that the market is undergoing a correction. But given the last 25 years of stock market history, I would argue that this isn't a correction; it's a prediction. And it's one that indicates we're going to see a Democrat in the White House next year -- we just don't know which Democrat it will be.

Comments

[NoFate]

I find this article misleading at best. I'm not even quite sure what the premise is. Based on the authors table there was 2 out of 5 down Novembers where the President changed (but not necessarily the party). This is hardly statistically conclusive evidence!

This also seems to imply that any change in the President has a negative impact on the markets (possibly due to the unknown). We knew there would be a different President in 2009 since November of 2004 though! Why does this information only affect the markets only in Novembers prior to election years??

No, the markets are reflection the fact that the past 5 years has created $4-5 trillion in housing inflation...primarily based on fraud. Yes, the fraud of selling homes to people who could never make the full payments ...and the fraud of selling AAA rated paper to institutions that was truly worth pennies on the dollar.

Now someone has to eat that bad debt ...which means lowered earnings ...which means lay-offs ...which will likely cause a recession ...and this is what the market is reflecting.

[Doug Roberts]

This guy is just flying his colors. Not only is there no apparent pattern in his table of data, but the reader has to guess his politics. 2000 was a "split decision"? The reader has to read to the end to find out that he differentiates between winning the election and winning the popular vote.