As the stock market rallies higher and approaches its 50% Fibonacci retracement resistance, we can see a clear divergence in the data:
- The economic data is mixed, with some long term macro models turning bearish.
- Breadth is long term bullish.
- Quantitative market patterns are medium term bullish.
Let’s make sense of the divergence in the data.
The economy’s fundamentals determine the stock market’s medium-long term outlook. Technicals determine the stock market’s short-medium term outlook. Here’s why:
- The stock market’s long term risk:reward is no longer bullish.
- The stock market’s medium term leans bullish (i.e. next 3-6 months).
- The stock market’s short term has a slight bearish lean.
We focus on the long term and the medium term. Let’s go from the long term, to the medium term, to the short term.
As I mentioned on Wednesday, a fairly popular trend following model called the Unemployment Rate Model has turned long term bearish for the first time since 2008-2009.
Buy and hold the S&P 500, unless…
The S&P falls below its 12 month moving average, and the Unemployment Rate rises above its 12 month moving average.
*We don’t use this model because its returns are inferior to our Initial Claims Model.
The S&P is below its 12 month moving average and the Unemployment Rate is slightly above its 12 month moving average right now.
A better signal than the Unemployment Rate is Initial Claims. While Initial Claims are very low, they have yet to trend upwards. When Initial Claims and Continued Claims start to trend upwards, that is a long term bearish sign.
And lastly, the Goldman Sachs Bull/Bear Indicator demonstrates that long term risk:reward favors the bears. This isn’t a market timing tool, but it does tell you risk:reward.
Our medium term outlook (next 3-6 months) still leans bullish.
We looked at the probability of a retest on Wednesday and came to the following conclusion:
How common is it for approximately 20%+ market declines to go straight up to new all-time highs without a pullback along the way? By now, the whole “stocks will retest after a crash” argument has become well known.
Let’s look at all the 20%+ declines from 1929 – present
This is the S&P 500 in 2011. The S&P almost retraced 50% before retesting and making a marginal new low.
This is the S&P in 2008. It retraced 50% before making new lows
This is the S&P 500 in 2001. It retraced 50% before making new lows
This is the S&P in 1998. It retraced 50% before retesting and making a marginal low.
This is the S&P in 1990. It retraced 50% before making a pullback (but not a retest)
This is the S&P in 1987. It retraced less than 38.2% before making a retest. (This retracement was very small because the S&P had fallen so much. Even an e.g. 38.2% retracement was massive in terms of %).
This is the S&P in 1981. It retraced 50% before making new lows.
This is the S&P in 1980. It went straight up without a single meaningful pullback
This is the S&P in 1974. It retraced 50% before retesting, another 50% retracement, and then new lows.
This is the S&P in 1969. It made a very shallow retracement before making new lows. (But there were multiple 50% retracements before the S&P fell -20%)
This is the S&P in 1966. It retraced less than 38.2% before making marginal new lows.
This is the S&P in 1962. It retraced 38.2% before retesting.
This is the S&P in 1957. It went straight up without a single retracement.
This is the S&P in 1937. It retraced 61.8% before cratering.
This is the S&P in 1929. It retraced 50% before going on to make new lows.
- For 20% declines, the most likely target is a 50% retracement before a pullback. That pullback usually leads to a retest, but not always (see 1990). This is the S&P right now
- For 25+ declines, the most likely target is a 38.2% retracement (see 1987).
- A v-shaped recovery is unlikely, but is possible. 2 of these 15 historical cases saw a V-shaped recovery (1957 and 1980)
Let’s get into the quantitative market studies.
*For reference, here’s the random probability of the U.S. stock market going up on any given day, week, or month.
We rank our market studies in order of importance.
- The stock market’s own fundamentals and technicals (primary importance)
- Correlation and specific sectors (secondary importance)
- Seasonality, short term factors (tertiary importance)
The S&P has gone up more than 1.8% for 3 weeks in a row. In other words, a very steady rally after a very steady decline in December.
Here’s what happens next to the S&P 500 when it goes up more than 1.8% for 3 weeks in a row.
*Data from 1950 – present
The S&P is quite bullish 2-6 months later.
Adding macro context makes these statistics more useful. Let’s only look at the late-cycle cases, in which the Unemployment Rate was less than 6%.
Once again, the S&P is quite bullish 2-6 months later.
It’s not just the stock market’s rally that’s very fierce. VIX’s decline is also nonstop. VIX has fallen 12 days in a row from the daily OPEN to the daily CLOSE.
From 1990 – present, this has only happened 1 other time: April 2009.
N = 1, so take this with a grain of salt.
Some investors and traders thought that the rising USD in 2018 was a long term risk to the stock market. The USD Index is now down 4 weeks in a row, while still above its 50 weekly moving average.
Historically, the USD’s forward returns after going down 4 weeks in a row were random.
Along with the S&P, the Russell 2000 (small caps) reversal is also very extreme. The Russell has gone up 11 of the past 12 days, while still under its 50 dma. This is only possible when the preceding crash was extreme.
From 1987 – present, this has only happened 1 other time.
And lastly, the reversal is so powerful that there is a bit of FOMO in the market.
In the AAII sentiment survey’s latest reading, the % of bears has rapidly plunged from above 50% to below 30% (as the stock market rallied).
Such a rapid drop in bearish sentiment is extremely rare.
From 1987 – present, it has only happened once: January 4, 2001
That was the first bear market rally in the 2000-2002 bear market.
N = 1, so take this bearish sign with a grain of salt. At the very least, it isn’t a long term bullish sign for stocks. Such a strong sense of FOMO is normal towards the end of bull markets. As retail investors and traders are conditioned by a 10 year bull market to constantly “buy the dip”, they see a 20% decline as a “once in a lifetime opportunity”.
There were multiple breadth thrusts from an array of breadth indicators over the past week. For example, the 10 day moving average of the # of S&P issues advancing went from under 150 to above 320 within 4 weeks.
From 1998 – present, this has been a long term bullish sign.
I see these breadth indicators as medium term bullish signs instead of long term bullish signs. As ETFs become more and more popular, breadth extremes will become more and more common. Hence, breadth extremes that used to occur at the end of bear markets are now more likely to occur at medium term bottoms in the middle of bear markets.
The S&P has gone up more than 10% over the past 12 days, one of its highest 12 day rates-of-change.
Historically, the stock market’s forward returns were random, but maximum drawdowns within the next 3 months were large (due to the probability of a pullback/retest).
Here is our discretionary market outlook:
- The U.S. stock market’s long term risk:reward is no longer bullish. This doesn’t necessarily mean that the bull market is over. We’re merely talking about long term risk:reward. Long term risk:reward is more important than trying to predict exact tops and bottoms.
- The medium term direction is still bullish (i.e. trend for the next 3-6 months).
- The stock market’s short term has a slight bearish lean. Focus on the medium-long term because the short term is extremely hard to predict.