Trend Check with Tushar Chande

A Brief History of Post-Election Rallies (Or, could the "Trump Rally" go even higher?)

As dark clouds gather over the investment horizon this Easter weekend, the post-election rally in the US stock market (the "Trump Rally") seems to have faded just a bit.  Is the rally over? Or, can the recent history of US stock returns during the first year of a new presidential four-year term give us some useful guidance?  We explore this question, in the hope we will find an Easter egg after all, if not now, perhaps later this year. However, the really tough question this weekend is this: Should the Easter chocolate bunny be consumed feet first, or ears first? (More on that later.)

 

Figure 1: Kites riding a stiff breeze keep watch under stormy skies over vast fields of tulips just reaching their peak this weekend in the Great Northwest.

 

A Brief History Of Post-Election Rallies in the S&P 500

The first year of a new 4-year presidential term (whether first or second term) brings a new burst of energy to the governing processes of the USA.  Thus, it is reasonable to look at these as distinct events on the investment horizons.  If a major economic or military crisis continues into the new term, then those crises dominate the market's concerns.  Otherwise, the new governing agenda drives the market's expectations.  Hence, I will use recent first-year returns as a guide for my analysis.


I went back to 1952 (Eisenhower’s victory), and decided to use the S&P 500 index at the end of October before the election as my starting point, and set it to 100.  I then calculated the market's performance through the end of the following year (here the end of December, 1953), the end of the first year of the new four-year term. This gives me an interval 14 calendar months long, with the current 2016-17 period as the latest interval.  I then calculated the correlation between the Nov-16 to Apr-17 "six-month" interval (I know, April is not over yet) and the same intervals for the other sixteen terms (so the length of comparison  period is uneven).  I chose to focus only on those periods with a correlation of 0.68 or greater to current interval of interest. Parenthetically, ten of the sixteen 14-month intervals in my sample were positive (62.5%).


In Figure 2 I compare the current rally, to the seven previous terms with moderate to high correlation over the first "six" months of the 14-month interval starting October of the election year through December of the following year.  The figure is busy, so note that the current 2016-17 rally is in green. Second, observe that there is much scatter in the data.  Third, recognize that intervals which start off strong, tend to end well.  I show a simplified picture in Figure 3, in which I just show the current rally since October and the average of my historical returns with the greatest correlation to the current rally.  Clearly, the Trump rally could go even higher.

Figure 2: : I used a 14-month interval starting at the end of October of the election year through December of the following year, the end of the first year of the four-year presidential term. These seven historical intervals had moderate to high correlation to the S&P 500 returns over the first few months of the current presidential term.  The chart is busy, so I compare the actual current returns to the average of the comparable seven other terms in Figure 3 below.

 

Figure 3:  The current rally in the S&P 500 index since the end of October, 2016 compared to the average S&P 500 performance during comparable periods of previous first-year terms with the highest correlation to the current term.  This chart suggests that the rally could go even higher.

 

Figure 3 suggests that the market's optimism will be rewarded with a higher close at year end.  Of course, the comparison does not account for actual economic conditions, and assumes that the market has done that for us already.  Secondly, there is considerable variation in performance as we can see in figure 2.  So far, in this small sample, when the first few months were positive, the entire 14-month returns were also positive.   Of course, the market could close lower from here in December, and still have positive returns over the 14-month interval starting at the end of last October.  So, as usual, we do not know precisely what will happen, but there is good reason to be optimistic in the face of the gathering storm clouds (see Figure 4).

 

Figure 4: I plot one standard deviation bands around the average of the seven previous post-election rallies with the highest correlation to the ongoing post-election rally to give a sense for the range of potential outcomes.

 

The Market's Sour Mood

The market fell on event risk as I suggested last Wednesday.  The stocks underlying the major indexes have weakened this past week, and the path of least resistance could turn lower even on the intermediate-term should the selling continue next week (see Figure 5).  We have other measures of intermediate-term strength which are still positive, so this remains to be seen.  Our core trend-check carpet shows the mixed trading picture very well (see Figure 5).  The short-term trend has turned lower in the short-term (as it has been for some time), and that weakness is spreading to longer time frames.

 

Figure 5: The market's path of least resistance is lower in the short- and medium-term time frames.  The downside risk will increase another notch if the intermediate-term bias changes as well upon further selling next week.  These data are from applying our trend-following models to the Russell-1000  universe and rolling-up the data.

 

Figure 6: The side-by-side comparison is arranged in terms of increasing stocks in the index from left to right (from 30 to over 2000) and from short- to long-term from top to bottom.  The market breadth increases from left to right, and the degree of smoothing increases from top to bottom.  Thus, the models at the top react (change direction) most rapidly, and the ones of the bottom do so reluctantly.  The S&P 500 short-term trend has turned lower, and QQQ is now flat in the short-term.  Thus, the trend models are broadly in sync with the market's mood from figure 5 and also with Erin Heim's summary for the week.

 

On a lighter note...

A recent survey found that chocolate bunnies should be eaten ears first.  A delightful set of trivia on chocolate Easter eggs is available from Cadbury's chocolate.  A number of stories have appeared in the media covering the scope of the Trump Rally (for example, see here and here).  Lastly, this being a long holiday weekend, the selling ahead of it could be explained by defensive positioning to avoid event risk.  Should the weekend end quietly, a rally would be needed to unwind the defensive selling.

I've gone on too long already. Until my next post, remember, there is strength in numbers.

Enjoy!

 

Tushar

 

 

 

|

Subscribe to Trend Check with Tushar Chande to be notified whenever a new post is added to this blog!
comments powered by Disqus