Research & Advice


Downside Risk Potential

 Friday, October 26, 2018


  • There is, historically speaking, a one-in-four chance the market drops another ten(ish) percentage points.
  • There is, historically speaking, almost certainty that the market will surge after mid-term elections.
  • History aside, we think the most likely scenario is a bit more pain followed by a rally.

Click here to read PDF version

At 9:55 am on Wednesday October 24th I replied to an internal e-mail asking me “Can you weigh in on where we go from here?” It reads as follows:

Per my e-mail earlier this morning, sentiment is getting stretched enough for there to be a rally.  But that’s just one piece of the puzzle. The 200-day moving average is beginning to slope down. And while I think technical stuff is voodoo, it turns out that a lot of investors practice voodoo.  I see two outcomes.

  • A 33% probability of the stock market continuing it’s correction for a week or two (maybe down 10% ish, putting the S&P 500 down to its May lows), then rallying double digits for the rest of the year.
  • A 33% probability of the market rejecting the mid-term elections as good and/or renewed trade war talk, and the market continues its correction longer, through January with a slightly larger drop (a total of 12-13% instead of 10%). In this scenario the rally is more subdued until the conversation shifts away from trade wars to the new tax cut proposal for the middle class
  • A 33% probability that the stock market starts improving mildly over the next couple of weeks and renews it’s rally in a mild manner (no matter what the rhetoric).
  • A 1% probability that the correction gets closer to 20%.
  • A 0% chance that a new bear market has started.

In regard to the level of the S&P 500 at the end of the year, I always put in the caveat “give or take a few months.  We Americans love our calendar years, but it’s stupid to think in such ways. It’s arbitrary.”

The last word in that internal e-mail was “arbitrary.”  And if you consider my evenly split probability assessments, you can get the idea that those were a bit arbitrary. The reality is that there is no real way to properly create accurate probability assessments. But the gist of the internal memo is A) after a month-long spree of pre-midterm election stock selling, we’re probably closer to the bottom than not, and B) this pullback is not associated with the beginning of a new long-term bear market.

But I do want to point to the fourth bullet point, “A 1% probability that the correction gets closer to 20%.”  Quite frankly, I think the probability of a 20% pullback is now bigger than one percent. Maybe it’s a four percent chance.  Maybe it’s a one-in-four chance. Again, you cannot accurately measure that. (It’s actually an 18% chance if you want to get historical – there have been 22 similar market corrections since 1974 and four of them have developed into 20% drops.)  Not all corrections become bear markets.  From the start of the current bull market in 2009 through February 2018, there were six corrections that didn’t turn into bear markets.

But, as I mentioned in that internal e-mail, technicals are starting to look ugly. Technical market analysis is kind of a stupid tool.  Stupid in so much that it helps tell you what stupid people will do.

We humans are masters at controlling and interpreting content to serve our own narrative. We watch “news” that supports our ideology.  Even PhD candidates have been found to weigh less important information that might oppose their dissertation theory.  And in the stock market, we look at pictures to help tell us a story because it makes us feel more comfortable to be able to rationalize the unknown. So when stock market sentiment gets bad, we look for reasons to sell. And a downward sloping 200-day moving average on the stock market is often the nudge needed to accelerate selling. In large part, that’s why I think we’ll get a bit more pain.

So, the bad news is that there is a growing chance of the stock market falling 20% peak-to-trough. The good news is that unless economic conditions worsen (and we don’t see that happening meaningfully; i.e. no recession; Q3 GDP just rang in at 3.5% as I am typing this), history suggests that the downside risk is limited to about 20%.

Bear markets during recessions are much different animals than bear markets during periods of economic growth.  Going back to 1900, on average, bear markets that have overlapped with US recession have had a median decline of -35.9% and last a median of 17-months.  Non-recessionary bears have a median decline of -23.3% over seven months.


Post-Midterm Tendencies

The US stock market is about to enter the most bullish period of the four-year presidential cycle.  Of the sixteen quarters of a presidential term, the top three in terms of average S&P 500 gains are Q4 of the mid-term year (now) and Q1 and Q2 of the third presidential year (the first half of 2019).  The median gain over those three quarters is 18.4%.  Since 1946, the S&P 500 has not declined over that time period.

Of course, this time it may be different. Q3 2018 saw S&P 500 gains in excess of 5%.  In the eight other previous instances when that has happened in the Q3 of a midterm year, the index soared 9.8% in Q4. But so far the market is dropping, not soaring.

Even if the market does turn around soon, that doesn’t mean that the rest of the three-quarter streak is bullet proof.  The median correction over those three quarters has been -6.8%.  Four of those declines were double-digit declines.  But that’s just further evidence that bad things can happen to good markets, and it’s temporary.

I could write a lot about policy uncertainty and how it is affecting the stock market.  And while that undoubtedly plays a part in the recent decline, it seems like wasted ink since we’ll move from speculation to knowledge in a short period of time.  Besides, the bigger headwind right now seems to be a combination of tighter monetary policy (which I’ve talked a lot about, including on the Allen Harris LinkedIn page; as a reminder, the Fed’s Quantitative Easing is turning to Quantitative Tightening in November as net central bank purchases turn negative) and tough tariff talk. Those are the reasons, in my estimation, that the market has been struggling.

Bottom Line:  I’ve been very vocal (for about half a decade) about my expectation of a recession sometime between2020-2022.  But we’re not there yet. And without a recession, we can expect the downside risk to be limited to about 20% (we’re halfway through that).  There is, historically speaking, a one-in-four chance that we’ll get a drop of that magnitude.  Which means there is a three-in-four-chance that we’ve just about hit a bottom, give-or-take a few percentage points.