The Long Term Stock Market Crash Probabilities

This is a Guest Post by Troy Bombardia of BullMarkets.co.

 

With the 10 year – 3 month yield curve now inverted, it is time to look at the long term bearish case for stocks.

Long Term

Only 2 factors affect the stock market’s long term outlook: valuations and fundamentals (macro).

The U.S. stock market’s valuation is high. You know it, I know it, everyone knows it. The financial media, financial experts and social media have been saying “U.S. stocks are expensive!” for most of the past 10 years.

Here’s the Tobin’s Q ratio

Here’s the Shiller P/E ratio

I have excluded 2 other popular valuation metrics: the stock market’s price-to-sales ratio and Market Cap-to-GDP ratio.

1. Using price vs. sales to measure the S&P 500’s valuations makes no sense. Revenues don’t matter – earnings do. Over the past 100 years, the stock market’s average price-to-sales ratio has increased faster than the price-to-earnings ratio because corporate profit margins are higher. More and more of corporate America’s profits are attributed to the tech sector, and tech stocks tend to have fatter profit margins vs. traditional industries such as manufacturing. In other words, corporate America needs fewer sales to squeeze out the same $X in profits.
2. Using the market cap-to-GDP ratio to measure the U.S. stock market’s valuations also makes no sense. This only makes sense if U.S. companies could only profit from U.S. customers. But more and more U.S. companies are expanding into foreign markets. Hence, it’s only logical for the market cap to GDP ratio to increase overtime. More and more “U.S. corporate profits” don’t come from “U.S. GDP”.

Nevertheless, let’s unpack the first 2 valuation charts: Tobin’s Q and Shiller P/E.

Both of these indicators demonstrate that the stock market’s valuation is extremely high right now. But contrary to what the permabears have been saying for most of the past 10 years, high valuations tell you almost nothing about where the stock market will go over the next 1, 2, or even 5 years. Valuations are a general yardstick for where the stock market will go over the next 10 years.

These are the S&P 500’s 1 year and 2 year forward returns vs. its valuation (P/E ratio). Notice how the correlation between the stock market’s valuation and 1-2 year forward returns is weak.

R squared = 0.0765 and 0.0646

This is what the permabears have been consistently wrong about. Imagine you are someone in 1995. You look at U.S. stock market valuations, listen to all the financial experts on TV and think “stocks are extremely overvalued, better sell now”. The U.S. stock market SOARED over the next 4-5 years, leaving you in the dust.

So how are valuations useful?

High valuations create the preconditions for a bear market – necessary but not sufficient.

So what causes a bear market?

Fundamentals.

1. Low valuations + recession = big correction (e.g. 20-30%)
2. High valuations + recession = bear market (e.g. 40-50%)

Think of a “recession” like jumping out of a building, and “valuations” as which floor you are jumping out from. If you jump out from the first floor window (i.e. low valuations), you will probably break a leg. If you jump out from the 10th floor window (i.e. high valuations), you will probably die.

Yes, valuations have been consistently high over the past 10 years. (Average valuations have been higher post-1995 and pre-1995 because inflation-adjusted interest rates are also lower). But not until 2019 have we seen the hallmarks of a recession in the making.

Given that valuations are high rate now, the next recession-driven stock market decline will be much bigger than a normal -20% decline.

While the bull market could keep going on for a year (or even 2), the long term risk:reward no longer favors bulls. Long term risk:reward is the most important factor to consider. E.g. even if the stock market goes up another 33% and eventually crashes 33%, you are better off selling today because there’s also the possibility that stocks could crash 33% from where it is today. Think in terms of risk:reward

Here’s a diagram to illustrate this concept. This is the stock market today.

Here are 2 potential scenarios:

In an ideal world, your favorite guru would be able to predict with 100% certainty which scenario will play out. But since no one has a crystal ball, you can only trade on risk:reward. E.g. if you sell today, in a worst case scenario (stocks keep soaring), will the stock market at least fall to where you sold your stocks?

Fundamentals

Some leading economic indicators are showing signs of deterioration. The usual chain of events looks like this:

1. Housing – the earliest leading indicators – starts to deteriorate. This has occurred already
2. The labor market starts to deteriorate. Meanwhile, the U.S. stock market is in a long term topping process. We are in the early stages of this process.
3. Other economic indicators start to deteriorate. The bull market is definitely over, and a recession has started. A U.S. recession is not imminent right now, but this doesn’t matter. Official recessions happen after bull markets top, which is why we focus on leading economic indicators instead of GDP.

*All non-Bull Markets charts are from FRED, StockCharts, and TradingEconomics

The housing market is weak. Here’s Housing Starts, New Home Sales, and the NAHB homebuilder index.

The labor markets are no longer improving.

Here’s Initial Claims, Continued Claims, and the Unemployment rate – all of which are extremely low.

While the deterioration in labor markets isn’t significant, it has started. For example, you can see that Continued Claims has somewhat trended upwards since October 2018.

Here’s what happens next to the S&P when Continued Claims’ 4 week average rises at least 17 of the past 19 weeks (i.e. somewhat trending upward), while Unemployment is under 5% (i.e. late-cycle)

This is how the 4 major bear markets from 1950 – present started.

The 10 year – 3 month yield curve inverted on Friday. The 10 year – 3 month section of the yield curve is a better timing indicator than other sections of the yield curve, such as the 5 year – 2 year or 10 year – 2 year.

Here’s what happens next to the S&P 500 when the 10 year – 3 month yield curve inverts for the first time in each economic expansion.

The following chart indicates risk:reward for the stock market after each 10 year – 3 month yield curve inverts. We compare the maximum the S&P can rally until the next recession-driven stock market decline vs. the % of the eventual S&P 500 decline.

As you can see, the S&P often continued to go up after the 10 year – 3 month yield curve inverts. But every case saw the S&P at least fall back to where it was when the yield curve first inverted.

While almost all parts of the Treasury yield curve are flattening, the 30 year – 5 year section of the yield curve is actually steepening right now.

A steepening yield curve late-cycle is worse than a flattening yield curve. The yield curve typically steepens towards a recession when the Fed is about to cut short term rates.

Here’s what happens next to the S&P 500 when the 30 year – 5 year yield curve steepened by 40 basis points over the past 9 months while the S&P rallies and Unemployment is under 5% (late-cycle cases).

1998 represents a best case scenario. 1-1.5 years left of this bull market. Either way, think long term risk:reward

Given that the Fed is no longer hiking interest rates, the following chart shows how close the last rate hike is to the next recession. (Recessions in grey)

We can probably expect long term yields to keep falling while the Treasury yield curve continues to flatten. The 10 year yield fell to a 1 year low on Wednesday. This is not a good sign for interest rates.

Approximately 37.5% of the yield curve (3 months, 1 year, 2 years, 3 years, 5 years, 10 years, 30 years) has become perfectly inverted.

*Perfectly inverted = when each longer term rate is lower than the next shorter term rate.

Once again, this is a late-cycle sign. In a best case scenario, the bull market and economic expansion still have 1-1.5 years left (see December 1988 and January 2006)

Consumer discretionary stocks have significantly underperformed defensive stocks recently. Here is the XLY:XLU ratio vs. the S&P 500 (XLY:XLU measures discretionary vs. defensive)

Here’s what happens next to the S&P when XLY:XLU falls over the past 2 months while the S&P rallies more than 9%.

As you can see, forward returns are more bearish than random. This mostly happens around bear markets, either at the top or at the bottom. There are also some false bearish signs e.g. in 2016, 2004, 2003.

Conclusion: The stock market’s biggest long term problem right now is that as the economy reaches “as good as it gets” and stops improving, the long term risk is to the downside.

The end of a bull market is always very tricky to trade. The stock market can go up a lot in its final year, even if the macro economy is deteriorating. That’s why it’s better to focus on long term risk:reward instead of trying to time exact tops and bottoms. Even when you think the top is in, the stock market could very well surge for 1 more year. (Just ask the people who thought that the dot-com bubble would end in 1998. It lasted another 1.5 years).

What doesn’t matter to me

As the long term bearish case becomes clearer over the next few months, one would expect the Bull Markets blog to focus more on bearish content. Far from it. We will continue to explain all the facts, both bullish and bearish. At every point in time there will ALWAYS be bullish factors and ALWAYS be bearish factors. You will never have a case in which 100% of the factors are bullish or 100% of the factors are bearish.

What many financial experts do (of which the permabears are most egregious) is to ONLY mention the factors that support their bull/bear case, and completely ignore all the factors that go against them.

The correct thing to do is list all the bullish factors and list all the bearish factors. Then explain why certain factors are more important than other factors.

With that being said, here are some long term “bullish/bearish factors” presented by financial experts that don’t really worry me.

“Bullish factors”

The Dow Jones Industrial Average has made a “golden cross”, whereby its 50 dma rose above its 200 dma. Traditional technical analysis sees this as a bullish sign for the stock market.

Here’s what happens when you trade with the strategy:

1. Buy on a golden cross (when the Dow’s 50 dma crosses above its 200 dma)
2. Sell on a death cross (when the Dow’s 50 dma crosses below its 200 dma)

As you can see, this strategy is worse than buy and hold (although there is less downside volatility). Just because something is popular and has a catchy name doesn’t make it good.

Here’s what happens next to the Dow when it makes a golden cross.

Interestingly enough, the Dow has a slight short term bearish lean which has played out through the stock market’s sudden drop on Friday.

“Bearish factors”

A prominent permabear presented the following chart recently.

As you can see, NDX’s weekly MACD histogram is as high as it was in the dot-com bubble. HOLY SHIT IT’S THE DOT-COM BUBBLE ALL OVER AGAIN.

Is it?

The MACD Histogram is a nominal value. It’s not a relative % value. Since NDX today is almost double of where it was in 1999, such a high MACD reading is not equivalent to the same MACD reading in 1999.

Comparing today to the dot-com bubble is similar to another trick used by the permabears. “The Dow fell -400 points today. The Dow fell -400 points in 2008. HOLY SHIT IT’S 2008 ALL OVER AGAIN”! (Never mind that the Dow is 4x higher today than where it was in 2008. But don’t let the simple truth get in the way of a good story).

Here’s NDX’s MACD Histogram adjusted as a % of the NDX’s value. Just like the dot-com bubble indeed.

And no, the stock market today is not “just like 1937”

These analogues aren’t very useful. When you squeeze, compress, stretch, and change the scales, you can always make the S&P 500 “look like” it’s doing whatever fits your preconceived bias.

This was a “stocks will crash” analogue in 2017

Another one in 2016

And another one in 2013

And another in 2012

Yes, the stock market today is “eerily” similar to 1937. But it’s also “eerily” similar to 2015, 2014, 2013, 2012, 2011, 2010, 2009, 2008, 2007…

If the long term bearish case is valid, why are these bearish arguments not valid? Because doing the right thing for the wrong reason is no different from just getting lucky. You should do the right things and make the right predictions for the right reasons.

The next one

The human mind has many biases and flaws. In terms of investing, I think the worst one is “recency bias”, which is the tendency to remember recent events more vividly, thereby exxagerating the probability that a recent event will occur again.

An entire generation of investors and traders have been scarred by 2000-2002 and 2007-2009. Hence, a lot of traders and investors think “the next one” will be just like (or worse) than 2008.

I think the next recession and bear market will be more mild than the 2007-2009 bear market. History rarely plays out the way most people think it will. For a really contrarian way of thinking how the next bear market will play out, see Jeremy Grantham’s interview with CNBC.

Conclusion

Here is our discretionary market outlook:

1. The U.S. stock market’s long term risk:reward is no longer bullish. In a most optimistic scenario, the bull market probably has 1 year left. Long term risk:reward is more important than trying to predict exact tops and bottoms.
2. The medium term direction (e.g. next 6-9 months) is mostly mixed, although there is a bearish lean.

Goldman Sachs’ Bull/Bear Indicator demonstrates that risk:reward does favor long term bears.

 

 

This is a Guest Post by: Troy Bombardia you can follow him on Twitter at @bullmarketsco and his website is BullMarkets.co.

***All content, opinions, and commentary is by Troy Bombardia and is intended for general information and educational purposes only, NOT INVESTMENT ADVICE.