Financial Sense   Home  l  Broadcast  l  WrapUp  l  Storm Watch  l  About Us  l  Contact Us

How the Stock Market Reconciles
the Increasing Likelihood of Bush Defeating Kerry
by Bob Bronson
Bronson Capital Markets Research
September 23, 2004

There are at least two pattern-matching methodologies for drawing statistical parallels from stock market action during Presidential election years to discern the historical patterns of  winning and losing incumbencies.

By far the most common way is to compare year-to-date performance, which is illustrated in the chart below with the solid lines: winning (green) and losing (red) incumbencies, compared with average market performance (black) preceding the last 15 elections.

This methodology suggests that without huge stock market gains during September and October, Bush has little chance of a second term, despite being a war-time incumbent, since the stock market is presently -- through the end of August -- a whopping seven percent below the average year-to-date performance of the eight times incumbents won during the 15 Presidential elections in the 60 years since 1944.

However, the other methodology yields very different results. The cyclical correlation of the year-to-date monthly changes, as distinguished from their cumulative trend, reveals a pattern that is very similar to two previous winning incumbencies: 1944, when Roosevelt defeated Dewey, and 1996, when Clinton defeated Dole. The dashed red line in the chart below shows the stock market's  performance, averaged for those two election years, to which this year's performance  is 82% correlated through August.

Synthesizing the two methodologies for Bush to stay in sync with the monthly cyclical pattern and trend of  the average performance of previous winning incumbencies, the stock market must at a minimum have a large gain in both September and October, though not necessarily a full 7%. Our work suggests this is very unlikely for many fundamental and technical reasons, including our expectation that the six-week summer rally peaked yesterday.

However, keep in mind that the S&P 500 gained only 1.4% (the DJIA lost 1.4%) in 1984 when Reagan won his second term by defeating Mondale, and the stock market lost ½% in 1948 when an incumbent Truman defeated Dewey, so there is historical precedent for Bush to win a second term with the stock market putting in only a minor gain or loss.

As for September and October, which are widely perceived to be critical months for the stock market signaling Presidential election outcomes, the stock market lost ground over those two months during three of the eight  (38%) winning incumbencies of the past 15 elections: in 1944 when Roosevelt defeated Dewey, the stock market lost 0.3%:  in 1965, when Eisenhower defeated Adlai Stephenson (for the second time), the stock market lost 4.0%; and in 1984 when Reagan defeated Mondale, the stock market lost 0.4%.

There has been a wide range of stock market performance relative to incumbency win-loss outcomes, and of course, history itself is not a boundary condition.  While it may be fun to speculate on election outcomes from stock market performance, the stock market obviously has a lot more to contend with than elections, and may or may not accurately signal who  is going to win.

Bronson Capital Markets Research
September 22, 2004

© 2004 Bob Bronson
Bio and Editorial Archive

Contact Information
Bob Bronson
Bronson Capital Markets Research
Email