The Presidential Election Cycle And Its Effect On Stock Markets

|
 |  Includes: DIA, SPY
by: Lawrence Chan

The Presidential Election Cycle is a structural cycle that impacts many things, including the stock markets. I am going to show a simple way to utilize the key statistics obtained from yearly data on the Dow Jones Industrial Average since 1929 for longer-term timing and portfolio adjustment purposes. The statistics collected so far are quite stable, hence the technique presented will be useful going forward for years to come.

The Questions and The Statistics

There are three questions every investor should ask in the beginning of a year:

1. How high will the Dow (or other broad market indices, like the S&P 500) go?

2. How low will it drop?

3. How will it fair by the end of the year?

To answer these questions, we need to look at the historical performance of the Dow so that we can draw some reasonable conclusions. We can look at the three statistics corresponding to the questions above:

1. The percentage gain achieved at year high based on previous year close.

2. The percentage drop recorded at year low based on previous year close.

3. The percentage gain/loss at year close based on previous year close.

Our assumption is that the Presidential Election Cycle is affecting the stock market. By classifying the historical performance of Dow into four groups -- the election year, first year after, second year after, and third year after -- significant behavior differences should be observed.

Another issue we have to deal with is that we do not know the underlying distribution of the data collected. It is necessary to compute two sets of statistics, one based on the average and another based on the median, to see which one provides more useful information.

Below are the statistics based on historical data since 1929.

Average Year-to-Year Percentage Change on Key Statistics

Election Year

1st Year After

2nd Year After

3rd Year After

Year High - Prev. Year Close

11.6%

16.8%

15.7%

22.3%

Prev. Year Close - Year Low

11.6%

14.5%

14.7%

7.9%

Year Close - Prev. Year Close

3.4%

4.5%

5.0%

12.9%

Click to enlarge

Median Year-to-Year Percentage Change on Key Statistics

Election Year

1st Year After

2nd Year After

3rd Year After

Year High - Prev. Year Close

9.5%

11.7%

11.8%

21.1%

Prev. Year Close - Year Low

6.9%

16.2%

14.3%

4.3%

Year Close - Prev. Year Close

4.2%

-0.6%

5.4%

15.2%

Click to enlarge

Median vs. Average

The reason why we need both tables is that average is not really a dependable measure when you do not know the underlying distribution of the data. Average is used everywhere in publications because it is a common practice, not necessarily a better choice. Median is often the better yardstick because it is not just a number -- it also contains distribution information. We know for sure it is the midpoint of the historical data. In other words, we know for sure where the 50/50 line in the sand is.

Basic Inferences From the Statistics

The best year to invest is the year before election year, or the third year after election. By both measures, average and median statistics, those years have the least likely risk of big drops and the best hope for a significant gain comparing to other years. 2011 was a good example on how resilient it can be.

The first and second years after election year are ones that have significant downside risk relative to their year-end expectations. The first year after an election is especially bad because its median year-to-year performance is net negative, meaning that half of the time it is a down year. Election year is not as volatile as the other years, but it is also not as bullish as the third year after an election.

The marked differences among the four groups of statistics tells us that the Presidential Election Cycle is, in fact, affecting the stock market significantly.

Election Year Is Not More Bullish For Stock Markets

It is mentioned often that election year is better for the stock market. Well, it is not. The following is a table showing the percentage of years that ended net positive:

Overall

Election Year

Other Years

66.3%

65.0%

66.7%

Click to enlarge

Notice the percentage for election year is below both the overall percentage and the one for the other years combined. Due to the fact that the Dow Jones Industrial Average in general has a two-thirds chance of getting a positive year, election year does not fair better.

General Applications of the Statistics

Knowing how likely the Dow (and the closely correlated S& P500) will fair in any year has several useful applications.

For long-term traders, when you know how much the indices are likely going to pull back based on which part of the election cycle we're in, you have the ability to position yourself for better bargain hunting. You can also adjust the stock weighting in your investment portfolio to reduce risk exposure when you find that the indices already tagged the likely upper boundary for any particular year. Combining the two, you have a good basic defensive strategy.

Short-term traders who are interested in swing trading the indices will find that the price levels corresponding to the median and average statistics often act as short-term support/resistance zones. It is not a coincidence. Professionals do look into these statistics. At potential yearly resistance zones, more long-term players would reduce exposures on the long side. At yearly support levels, these long-term players would seek opportunities to increase their long exposures. The aggregated results create temporary tops and bottoms that are great trading opportunities.

To take advantage of the opportunities created by the election cycle, traders can participate by trading the corresponding index ETFs such as the SPDR Dow Jones Industrial Average (NYSEARCA:DIA) and the SPDR S&P 500 (NYSEARCA:SPY).

An Example

Here is an example that applies the statistics to analyze the potential outcomes for the rest of 2012.

The following is the performance so far as of May 11, 2012.

2012

Year High (so far) - Prev. Year Close

9.3%

Prev. Year Close - Year Low (so far)

0.0%

May 11 Close - Prev. Year Close

8.1%

Click to enlarge

Notice that the Dow has already printed near the median percentage gain based on the previous year close to year high so far. It has since pulled back significantly. As stated in the prior section, the aggregated reactions by the pros to the price levels projected by the median percentage can easily lead to the creation of at least a short-term top.

Later on this year, the Dow may still revisit the median or even the average percentage gain level, at 9.5% to 11.6%, because there is still a 50% chance of that happening. By year-end, there is at least 50% chance that the Dow (and the S&P 500) would gain only 4.2% or less.

The U.S. stock market has rallied right out of the gate since the start of 2012, hence the unusual value of a 0% drop for previous year close to year low so far. This number is far away from the median 6.9% drop. It is very far away from the average 11.6% drop. It is reasonable to say that there is a good chance the current year low will be tested (or breached) later this year.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.