Plunge Period Reference Points

VOLATILITY IS MORE COMMON THAN MOST BELIEVE

While they seem unprecedented and/or unnerving while they are occurring, market dislocations are not all that rare.  Experience says those that remain cool, calm, and collected with their eye on data from multiple timeframes have the greatest chance of being satisfied with their results looking back six months from now. 

“I learned at sea virtually all situations can be handled as long as the presence of mind is maintained.” 

Josh Waitzkin, Chess Champion, The Art Of Learning

Therefore, we reviewed numerous cases on Monday night and decided to focus on the eight below.

1955 PRE-PLUNGE

The set of charts below show the S&P 500's pre-plunge profile, allowing us to better understand the odds of a new bear market/prolonged correction.

 The moving averages on each chart range from the 20-day (blue) to the 200-day (orange).   The 1955 case looked good from a trend and conviction perspective.  Price was above all the MAs, blue, the fastest moving average, remained on top looking back seven months, and the slopes of all the MAs were positive.  All of that said "the market looks relatively healthy on multiple timeframes". 

fld-1955.png

1987 PRE-PLUNGE

The 1987 chart had quite a bit more evidence of a waning trend relative to 1955.  The chart below is as of October 13 allowing us to see the market's profile before the Black Monday plunge, which occurred on October 19, 1987.

fld-1987.png

1998 PRE-PLUNGE

Like 1987, there were several signs of waning bullish trend in 1998. 

fld-1998.png

2000 PRE-PLUNGE

The market was a volatile, indecisive mess before tanking in September 2000.  The 2000 profile was much worse than 1955. 

fld-2000.png

2007 PRE-PLUNGE

Sideways price action speaks to indecisiveness; stocks went sideways for nine months before tanking in December 2007.  Compare the look of the 2007 chart below to the 1955 chart above - 1955 was making a series of higher highs and higher lows (bullish trend) - the chart below went absolutely nowhere for nine months (indecisive behavior). 

fld-2007.png

2011 PRE-PLUNGE

The 2011 chart was a mess before the market plunged.   We can draw horizontal lines and hit price many times in 2011; not the case in the 1955 example.

short-takes-2011-plunge-fix.png

2015 PRE-PLUNGE

The market was saying "pay closer attention" for six months before the 2015 plunge.  In the 1955 chart, the moving averages cover a wide range on the price chart (a sign of a strong and sustainable trend).  The 2015 chart below features a tight MA cluster, which said "a big move could be coming".  

fld-2015.png

2018 PRE-PLUNGE

On Friday, January 26, 2018 the S&P 500 closed at a new all-time high.  Trends from short (20-day) to long (200-day) were all constructive.  The market had a full-bore bullish look with price above all the MAs, blue, the fastest moving average, on top, and the slope of all the MAs were positive (aka bullish).  The market had not been moving sideways, but rather making a series of higher highs.

fld-2018.png

1955 IS THE BEST MATCH

When we look at trends on multiple timeframes, it helps us understand how market participants are interpreting all fundamental and technical data looking out a few days to a few years.  It is pretty easy to see the charts from 1987, 1998, 2000, 2007, 2011, and 2015 have numerous forms of waning trend evidence, which is indicative of waning confidence in the economy and markets.  The two charts with the strongest and most sustainable-looking trends are 1955 and 2018.   The 2018 chart shares almost nothing with the 1987, 1998, 2000, 2007, 2011, and 2015 charts.

DISLOCATIONS RARE FROM STRONG PROFILES

It is extremely rare for markets to have a dislocation event or plunge with a profile as strong as the one that was present on January 26, 2018.  Since markets can do anything at anytime, meaning even very low probability outcomes can occur, we always remain open to all outcomes, from extremely bullish to extremely bearish.  If you know your market history, you know bad things happened next in 1987, 1998, 2000, 2007, 2011, and 2015.  

WHAT HAPPENED NEXT IN 1955?

In the 1955 case, stocks experienced a sharp 6.8% correction and then resumed their well-defined bullish trend, tacking on an additional 42.3% over the next 510 calendar days.  If you are keeping score at home, the S&P 500's correction in 2018, as of Monday's close, was 7.8% (similar to 1955).  In order to compare apples to apples, we need to compare 6.8% to 2018 as of Tuesday's close (rather than intraday Tuesday).

fld-1956.png

THE STRONGER THE TREND…

History says the stronger the trend when the dislocation occurs, the higher the odds the dislocation will be sharp, but relatively short-lived.  Said another way, the stronger the profile before the dislocation, the higher the odds the plunge will be followed by gains and eventually a new high in stocks.

For example, if we remove the two strongest cases 1955 and 2018, the next two strongest profiles were 1987 and 1998; in both cases patient and rational investors were rewarded with new highs (and much higher profits).   Conversely, the two weakest profiles were 2000 and 2007; the present day looks nothing like the periods from 2000 and 2007 shown above.

Our approach remains the same; if the hard data calls for action, we will take action.  If you ask anyone who has been through a difficult situation in the markets, they will tell you if you want to increase your odds of poor results looking back after six months, ignore the hard data, break your discipline, and make short-term/emotionally-based decisions.  Experience says make decisions based on the weight of the evidence on multiple timeframes, stay aligned with the hard data, stick to your discipline, and make logical decisions.

WHAT HAPPENED NEXT?

The second part of this analysis can be found in this S&P 500 performance after waterfall declines post.

If It Looks Like A Duck

Market Volatility:   Most recent comments can be found here.

Bull/Bear Eye Test

We are currently in a bull market.  Bull markets by definition make a series of higher highs and higher lows.  Thus, if we believe the bull market is still intact, at some point in the future, stocks will make a new all-time high.  

ECONOMIC FEAR

If someone told you the economy was headed for a significant recession and the Fed would most likely be lowering interest rates, would you prefer to own growth-oriented stocks or defensive bonds?  When investors are concerned about future economic and market outcomes, the desire to own bonds tends to be greater than the desire to own stocks.

YELLOW FLAGS BEFORE STOCKS PEAKED

The chart below shows the performance of stocks (SPY) relative to long-term Treasury bonds (TLT).   Early adopters, who saw problems before the financial crisis, started to gravitate toward bonds three months before the S&P 500 peaked in October 2007.  Between July and September 2007, the stock/bond ratio shifted from a full-bore bullish look, favoring stocks, to a full-bore bearish look, favoring bonds.  This observable shift in asset class behavior said "pay closer attention".

short-takes-blog-ciovacco-tlt-spy-f.png

WHAT CAN WE LEARN IN 2018?

The stock/bond ratio in 2018 does not really exhibit any of the concerning characteristics that were present in mid-to-late 2007.   The chart still has a full-bore bullish look favoring growth-oriented stocks over defensive-oriented bonds.   After consolidating in an indecisive manner for several months (orange box), the ratio broke out in a bullish manner in September 2017.  A successful retest of the breakout occurred in November. 

ciovacco-short-takes-tlt-spy-2-feb.png

BULL/BEAR ODDS

This week's video looks at the recent pullback in the context of history.  The charts help us answer the most relevant question - does 2018 look like a bull market or the early stages of significant and protracted decline?

"The most important quality for an investor is temperament, not intellect. You need a temperament that neither derives great pleasure from being with the crowd or against the crowd."
Warren Buffett

WHY ARE BOND YIELDS RISING?

When market participants were concerned about the economy and markets in 2007, they started buying bonds.  Today, we have bonds selling off based on increasing confidence in future economic outcomes.   An improving economy does not fit the traditional early-stage bear market script.  Does it look like a bull market or a bear market?

ccm-short-takes-wsj-headline-2-2-2018.png