I am the editor of the Stock Trader's Almanac & Almanac Investor Newsletter and a Research Consultant at Probabilities Fund Management, LLC. I use historical patterns and market seasonality in conjunction with fundamental and technical analysis...+ FOLLOW THIS TUMBLR
Looking back to 1950, S&P 500 has completed 33 consecutive monthly losing streaks of three or more months. With six trading days left in October, S&P 500 needs to climb back above 2168.27 to avoid a third month. Should S&P 500 fail to reach this bar, it would actually be a positive for November as 20 of those past 33 streaks ended at three months and within one month of the end of all streaks, three months or longer, S&P 500 was higher 100% of the time with an average 4.72% gain. S&P 500 worst monthly losing streak since 1950 was nine straight in 1974. When this streak ended, S&P 500 also had the largest one-month gain, 16.3%.
Election Day is (finally) just around the
corner. Next week will be two weeks before and it is also the week after October
options expiration. Over the past 22 years the week leans bullish with
gains exceeding losses by a margin of nearly 2 to 1 and respectable across the
board gains. However, this same week (not always the week after options
expiration) has nearly the exact opposite track record in election years. Since
1952, DJIA and S&P 500 have advanced during the week only 37.5% and 31.3%
of the time respectively. The week before election week has a bullish bias that
exceeds election week and the week after.
Six possible political alignments exist in Washington: Republican President with a Republican congress, Democratic congress or split Congress; and a Democratic President with a Democratic Congress, Republican Congress or split Congress. Data presented in the chart below begins in 1949 with the first full presidential term following WWII to focus on the modern era.
First looking at just the historical performance of the DJIA under Democratic and Republican Presidents we see a pattern that is contrary to popular belief. Under a Democrat, DJIA has performed better than under a Republican. DJIA has historically returned 10.0% under Democrats compared to 6.8% under a Republican executive. Congressional results are the opposite and much more dramatic. Republican Congresses since 1949 have yielded an average 15.3% gain in DJIA compared to a 6.1% return when Democrats have controlled the Hill.
Democrats in power of the two branches have produced an average DJIA gain of 7.4%, below the All Year average gain of 8.3%. This scenario is increasingly becoming more probable as the polls widen in favor for Hillary Clinton. With a Democratic President and a Republican Congress or a split Congress, DJIA has performed well averaging a 14.7% gain. The best scenario for all investors has been a Democrat in the White House and Republican control of Congress with average gains of 16.4%.
Whether looking at “All Presidential Election Years” or just “Eighth Years” in the charts below, the market is usually in rally mode near the end of October. Should the election go as polls suggest then the market will most likely track the “All Presidential Election Years” pattern to finish the balance of 2016. Should the election results be too close to call and trigger a recount in a state or possibly more, then the “Eighth Years” pattern would be back in play. An undecided outcome similar to November 2000 (an Eighth Year) would produce great uncertainty and could lead to declines similar to those that occurred then. DJIA dropped 5.1%, S&P 500 plunged 8.0% while NASDAQ imploded 22.9% (tech bubble burst, not likely to occur again as valuations are not as extreme now) in November 2000. In the undecided scenario, it would not be surprising to see the Fed skip a rate hike in December in order to provide some relief to the market.
Historically, S&P 500 performance from the end of July until the end of October has been a reliable indicator as to the outcome of November’s election (hat tip Sam Stovall of CFRA, formerly S&P Global Equity and Fund Research). When the period was positive, the incumbent party typically wins the election, when the period was negative, the incumbent usually lost. Overall in 18 presidential elections since 1944, the indicator was correct 15 times (83.3%).
As of today’s close S&P 500 is down 2.2% since the last trading day of July. This would suggest an incumbent party defeat and a victory for Republican Donald Trump. While certainly a possibility, it is also a possibility that the market is increasingly unhappy with the prospects that the Democrats could sweep the election and retain the White House while reclaiming the House and the Senate. The end of gridlock in D.C. would introduce even more uncertainty for the market to cope with.
Since 1994, the Monday of options expiration week has a bullish bias for DJIA, S&P 500, NASDAQ and Russell 2000. Expiration day however, tends to be mixed with average losses across the board even though S&P 500 and NASDAQ have advanced more often than not. Entire expiration week and the week after also lean bullish. Of note is the small-cap index streak of 13 straight advances from 1994 through 2006 (not shown, the streak is actually 17 years starting in 1990) and it has been down in six of the last nine years. October’s reputation for volatility can be seen with wild daily and weekly swings in the tables below. Weekly moves in excess of 5% occurred in 1998, 2002, 2008 and 2011.
In the below chart, “All Post-Election” years since 1953 is the baseline to which “All 1st Elected,” “Dem 1st Elected,” and “Rep 1st Elected” are compared. “All 1st Elected” years are the first years of a new president. “Dem 1st Elected” are first years of a new president that was a Democrat and “Rep 1st Elected” are new Republican presidents. For the first half of a post-election year all four are quite close. Under 1st Elected Democrats DJIA was weaker in January and February on average as a result of early losses in 2009. By the end of July DJIA’s performance begins to show significant deviation and by the end of the year DJIA has performed best under a newly elected, 1st-year Democrat. Should polls hold and Hillary Clinton become the next President, DJIA’s performance in 2017 could resemble the green line. If Donald Trump can defy odds and polls then DJIA could follow the path of the black line.
Earlier this month we examined “selling Rosh Hashanah”
and now let’s take a closer look at buying Yom Kippur. Last week’s table has
been trimmed and condensed to show the last 26 years results. DJIA has advanced
22 times and declined just four times since 1990 between Yom Kippur and Passover
with an average gain of 8.6%. The worst decline was 7.4% in 2000-2001 and the
best was 25.4% from 1998-1999. Over the last seven years of the current bull
market, DJIA’s average gain has been 10.6%.
In the following table the annual performance of
gold (continuously-linked, front-month futures contract) is broken down by
years of the presidential election cycle. Stocks have historically performed
best in pre-election and election years while gold’s best performance has been
in midterm and pre-election years on average. Gold’s worst years have been
election years. Excluding 2016, gold declined in six of the last ten election
years and its best performance was a 10.8% gain back in 1980.
Since the start of
2016, gold has been outperforming, advancing nearly 30% through late July. Even
after last week’s 5% loss, gold is still up over 18% year-to-date based upon continuously-linked
non-adjusted front-month futures prices (used in all charts below). Using data
from 1975 to 2015, gold’s average annual gain has been just under 7% in all
years. In election years, gold has been substantially weaker, averaging a loss
of 2.8% since 1975. Gold’s performance this election year has been well above
Removing 2016 from
the above chart we are left with gold’s 1-year seasonal pattern in all years
and in election years (below). Brief seasonal weakness is shaded in yellow.
Early October declines tend to persist and lead to a lower low in November
before gold rebounds to finish the year higher in “All Years.” In election
years, gold’s October decline has been greater in magnitude, but shorter in
duration with the rebound beginning near the end of the month.