Presidents and the Stock Market

Brian Stoffel, The Motley Fool

Yesterday, fellow Fool Morgan Housel came out with a brilliant piece on the correlation between the unemployment rate and the chances an incumbent (or his party) will be re-elected.

You would think that if a president has lowered unemployment rates, the chances for re-election would be high, and vice versa. Alas, Morgan cites evidence from statistician Nate Silver -- who culled evidence from as far back as 1912 -- to conclude: "There's virtually no correlation between the unemployment rate on Election Day and an incumbent's chance of winning."

Morgan is able to point out one possible reason for the weak correlation: "The average voter in the last two elections ... has not been representative of the broader economy. They've been in much better financial shape than the average American." In other words, unemployed people are less likely to head to the polls; their wealthier counterparts are far more likely to make their voices heard.


Knowing this, I think there's an even better and more accurate way to find a link between re-election and the economy: the behavior of the Dow Jones Industrial Average (INDEX: ^DJI) . Take a look, and you'll see that as far back as the election of 1900, the movement of the index between Sept. 1 and Election Day has successfully predicted the winner almost 90% of the time, with gains working in favor of the incumbent party and losses favoring the challenger. (Years in which the opposite occurred are italicized.)

Year

DJI Change from Sept. 1 to Election Day

Winner

Party

Did Market Predict Winner?

2008

(20.2%)

Obama

Democrat

Yes

2004

(2.5%)

Bush

Republican

No

2000

(2.7%)

Bush

Republican

Yes

1996

8.5%

Clinton

Democrat

Yes

1992

(1.2%)

Clinton

Democrat

Yes

1988

3.1%

Bush

Republican

Yes

1984

2.6%

Reagan

Republican

Yes

1980

(1.3%)

Reagan

Republican

Yes

1976

(1.9%)

Carter

Democrat

Yes

1972

2.4%

Nixon

Republican

Yes

1968

3.1%

Nixon

Republican

No

1964

3.5%

Johnson

Democrat

Yes

1960

(2.1%)

Kennedy

Democrat

Yes

1956

(2.8%)

Eisenhower

Republican

No

1952

(2%)

Eisenhower

Republican

Yes

1948

2.9%

Truman

Democrat

Yes

1944

3%

Roosevelt

Democrat

Yes

1940

0.8%

Roosevelt

Democrat

Yes

1936

6.3%

Roosevelt

Democrat

Yes

1932

(20.4%)

Roosevelt

Democrat

Yes

1928

7.7%

Hoover

Republican

Yes

1924

1.1%

Coolidge

Republican

Yes

1920

(1.9%)

Harding

Republican

Yes

1916

13.3%

Wilson

Democrat

Yes

1912

(1.1%)

Wilson

Democrat

Yes

1908

0.4%

Taft

Republican

Yes

1904

18.4%

Roosevelt

Republican

Yes

1900

4.1%

McKinley

Republican

Yes

Source: Analyzeindices.com; Google Finance.

One could easily argue that in 1968, the withdrawal of President Johnson, assassination of Robert Kennedy, and ensuing mayhem at the Democratic National Convention in Chicago could explain away at least one instance where the market didn't predict the winner.

One could also argue that in 2004, a shrewd move by Karl Rove to put same-sex marriage amendments on the ballots of swing states -- thereby driving social conservatives disproportionately to the poll -- could offer up an explanation for the second aberration.

But whether or not certain instances can be explained away, one thing is clear: There is a high correlation between the movement of the markets immediately preceding an election and the chances that the incumbent party will win.

Going back to Morgan's point -- that the voting population is usually skewed toward those with employment and higher incomes -- this makes sense. The market is a forward-looking tool: A rise represents optimism, and a fall, pessimism.

While this might not tell us who has the upper hand right now, it lets us know that both President Obama and Mitt Romney will be on the clock starting in three months, and the movements in the market will likely let us know who will be sitting in the Oval Office come January 2013.

Interested in more? Check out The Motley Fool's special report: "These Stocks Could Skyrocket After the 2012 Presidential Election." It's totally free -- just click here.

At the time thisarticle was published Fool contributor Brian Stoffel acknowledges that correlation is not causation, and there are millions of lurking variables to take into consideration that go far beyond the purview of this article. You can follow him on Twitter, where he goes by TMFStoffel.The Motley Fool has a disclosure policy. We Fools may not all hold the same opinions, but we all believe that considering a diverse range of insights makes us better investors. Try any of our Foolish newsletter services free for 30 days.

Copyright © 1995 - 2012 The Motley Fool, LLC. All rights reserved. The Motley Fool has a disclosure policy.