Thu 22 Dec 2011
Presidential election cycle theory was developed by Yale Hirsch, creator of the Stock Trader’s Almanac. Its premise is that the stock market follows a four-year pattern that corresponds to the four-year presidential election cycle. According to the theory, this is how the market reacts during each year of a presidential term:
Year 1: The first year of a presidency usually sees weak performance in the stock market. Of the four years in a presidential cycle, the first-year performance of the stock market, on average, is the worst.
Year 2: The second year, although better than the first, is also is noted for below-average performance. According to Hirsch, bear market bottoms occur in the second year more often than in any other year.
Year 3: The third year is normally the strongest of the four years.
Year 4: In the fourth year of the presidential term and the election year, the stock market’s performance tends to be above average.
The chart below shows how the Dow Jones Industrial Average (DJIA) has performed on average during the four years of each presidential cycle over the past century-plus.
Below is one more chart to substantiate the validity of the theory. This one shows that the third and fourth years of a president’s term produce market gains more than 80% of the time.
Of course, we are just about finished with the third year of the current presidential term and the market is not going to experience the type of gain illustrated by the first chart. In fact, this could turn out to be one of the 6% of times when the third year of a presidential cycle produces a loss rather than a gain, as shown in the second chart.
The problem with a theory like this is that it is considering the impact of a single factor and then trying to make a forecast based on incomplete data.
Here is a comment from an Investopedia article about the presidential cycle:
“One of the problems with drawing conclusions from the presidential election cycle is that the theory is based on relatively few observations. Since 1900, there have been only 27 presidential cycles to 2008. Many of the studies done on the theory are based on even fewer observations. For example, since 1948 there have been only 15 different terms - when it comes to statistics, this is a very small sample, which makes it difficult to draw accurate conclusions.
“As such, the theory could be attributed to data mining. In other words, if people are constantly looking at enough data for specific patterns, patterns can emerge, even if there is no significance to them.”
The stock market is incredibly complex. Millions of variables, many of which could be unknown and immeasurable, influence it. For well over a century, extremely smart people have devoted enormous amounts of time and money trying to learn to forecast accurately what the financial markets will do next. Yet to date no one has devised a method that is consistently reliable, let alone foolproof.
When it comes to the presidential cycle, there might be some correlation between presidential terms and market behavior. But there certainly is not a cause-and-effect relationship. And 2011 has again proved that just because something happened in the past does not mean it will be repeated in the future.
Have a great holiday season. Our next market update will probably not be until January 6 unless something unusual and unexpected occurs between now and then.
Flint Stephens





