Aug 16, 2012

The connection between stock market and US president cycle

The connection between US president cycle and stock market cycle has always been a debated issue. With US elections approaching, the topic is receiving even more attention. In the paper entitled “Presidential Elections and Stock Market Cycles” by Marshall Nickles, EDD thoroughly explains the evidence for such a connection. This is done by analyzing15 stock market cycles (from April 1942 to October 2002) and presidential elections for the same period. Here, I intend to use that methodology to review the last two stock market cycles and find out how they fit to the broader picture.

Briefly stated findings of Mr. Nickles research show that:
  • Average bull market duration for the mentioned period was 3.08 years
  • The average bear market duration was 0.94 years
  • The bull/bear market duration depend on the measures undertaken from officials to tackle the crisis and stimulate growth, thus deviation of the average number exists
  • There is significant probability market trough to occur in the second year of the presidential term  (12 out of 16 terms for the observed period)
  • There was not major market correction during the election years for all election years within the observed period

Based on the findings, the author suggests investment strategy buying on 1st of Oct of the second year of the presidential term and selling 31th of December of the fourth (the election year). According to the calculations this strategy could gain annualized return of 9% for observed 60 years.

Those findings seem quite convincing and one might decide to build his investment strategy based on them. Let us examine the latest two cycles from 2000 to date and apply mentioned strategy for this period.

The chart gives us information about the relationship between presidential and stock market (S&P500) cycles for the suggested period. The index is in log scale.

S&P500 and presidential elections
S&P500 and presidential elections

Last two bull cycles (2002-2007 and 2009- to date) tend to be more extended than the average mentioned in the paper. This could be explained with the magnitude of the crash proceeding each of the bull phases. As the chart shows, during both years- 2000 and 2008, downturns were severe. Average bear markets for the period 1940-2002 tend to last one year, the recent two are exceptions from that rule and are  longer. Dot com bubble, for example, busted in 25 months while recent mortgage burst lasted 16 months.

Take a look at the last two market cycles – the Dotcom bubble (2000) and Housing bubble (2008). What they have in common with presidential election cycle is that election year coincided with severe bear market – the exact burst of the bubble. Moreover, the duration of the last two market cycles was longer than the average for all previous cycles between 1944 and 2000. Dot com boom and burst lasted 99 months (8.25 years) while housing bubble and consequent melt down lasted 77 months (6.4 years) compared to 4 year average. As the author noted, cycle duration depends on unforeseen events such as monetary or government stimulus actions, which influence the cycles, indicating that those events have intensified recently.

The strategy

In the paper, the author suggested interesting investment strategy. Buy on the second year of the presidential term and sell on 31 Dec on the next election year. As calculations showed, this kind of investing could have preserved you from most severe bear markets and brought to you annualized return for the period 1952-2000 of 9.3%. Sounds good, doesn’t it? However, if you have followed it you would have missed some of the great bull market runs, like 2002-2007. If you have waited two years after 2004 elections and entered the market in 2006, you would have missed 16% S&P500 rally, which is significant. By applying this strategy for the 2004-2008 president cycle, i.e. buy in Oct 2006 and sell 31 Dec 2008 you would have counted with a  loss of almost 20% as year 2008 is one of the few presidential years, which ended with loss. Thus, buying in presidential year does not always work, however the probability for stock market gain in the election year is high. Since year 1928, only 4 out of 22 president election years finished with negative gains. The table below shows market returns for each election year. Data below are taken from  Dimensional Funds Matrix Book 2011 (“Presidential Elections and Stock Market Cycles” Marshall Nickles, EDD)

S&P 500 Stock Market Returns During Election Years

Year      Return   Candidates
1928      43.6%     Hoover vs. Smith
1932      -8.2%      Roosevelt vs. Hoover
1936      33.9%     Roosevelt vs. Landon
1940      -9.8%      Roosevelt vs. Willkie
1944      19.7%     Roosevelt vs. Dewey
1948        5.5%     Truman vs. Dewey
1952      18.4%     Eisenhower vs. Stevenson
1956        6.6%     Eisenhower vs. Stevenson
1960       0.5%      Kennedy vs. Nixon
1964      16.5%     Johnson vs. Goldwater
1968      11.1%     Nixon vs. Humphrey
1972      19.0%     Nixon vs. McGovern
1976      23.8%     Carter vs. Ford
1980      32.4%     Reagan vs. Carter
1984        6.3%     Reagan vs. Mondale
1988      16.8%     Bush vs. Dukakis
1992        7.6%     Clinton vs. Bush
1996      23.0%     Clinton vs. Dole
2000      -9.1%      Bush vs. Gore
2004      10.9%     Bush vs. Kerry
2008      -37.0%   Obama vs. McCain
2012      ?             Obama vs. ?

How to avoid presidential election year with negative stock market return?

Presidential election cycles alone do not give us the necessary information for taking investment decision. Combining the stock market cycle with the presidential cycle gives us the ability to time the market. It is a statistical fact that it is more likely to see market trough during the second or first year of presidential term. Keeping in mind that average duration of a bull cycle is 3 years, we could expect that if presidential election comes during the bull market or first two years of the market cycle than there is no point of waiting two years to enter the market as expected correction should be mild. If election years, however, coincide with the third, fourth year of the bull phase or even with bear stock market then you should follow the suggested strategy.

The 2012 elections and the stock market

Currently we are in the fourth year of the president cycle, the election year, and in the fourth year of the bull cycle (started with the 2009- low). The logic of the cycle sequence suggests that applying the presidential year cycle strategy we could wait with investments until mid presidential term - Oct 2014. However, there are presidential elections years followed by strong bull market. Are we in a such one? Considering the fact that currently we are in the 42th month (3.5 year) of the bull market, we have to be cautious. Markets might be close to the bull cycle end as the average one from 1944-2004 is 3.08 years and 2012 is the fourth year of rising stock prices. 

Many state that history does not repeat itself and it is impossible to predict future events. However, it is possible to time the market cycles, at least to some extent, and with relative accuracy. No one can predict the future but knowing the big picture and major trend of the phase of the cycle could save you some financial headaches. Even for the critics it is obvious that repeated patterns do exist. Just look at the chart. Whether you agree or disagree with the cyclical relations it is up to you, however it is always better to know some important repeating patterns and to be ready to use them in your favor.

More on Markets and Forecasting

Create your own chars with Charting tool

1 comment:

  1. This comment has been removed by a blog administrator.


Leave your comment

Creative Commons License
This work with autor is licensed under Creative Commons 3.0.