Valuation Wave Reversion

Adam Hamilton     October 25, 2002     2966 Words


Denver, Colorado is one of my favorite big cities in America.  Much time in my life has been spent working, studying, and playing in the Greater Denver area, and I continue to savor each visit to the city.


One of my favorite things about Denver is its breathtakingly beautiful weather.  An amazing average of 300 days each year are gloriously sunny, with brilliant warm sunlight streaming down from the heavens.  Humidity is also very low year round, helping virtually any temperature feel quite pleasant.


The climate of the Colorado High Plains is phenomenally good.  Many folks who haven’t spent time in Denver labor under the common misconception that the area is cold because it is so close to the fantastic Colorado skiing in the Rockies, but that really isn’t the case.


In April, early spring, the average daily high temperature in Denver runs about 60 degrees Fahrenheit.  May averages 70 degree daily highs, June 80, and July 90.  In August, there are invariably a few 100 degree days.  This linear trend of Denver average daily temperatures is remarkably consistent over its 5 months of spring and summer from April to August.


For market junkies like me, it is hard to resist the temptation to extrapolate this perfect Denver temperature trend.  Assuming that one knew nothing about Denver and continental climates in general, what are two likely conclusions that could be drawn from this short real-world dataset?


One could assume that since the temperature in Denver had been rising in a straight line for 5 months, it would continue its linear rise into infinity.  September would have a projected average daily high of 110, October 120, November 130, and December 140.  I have been blessed to do a lot of downhill skiing in Colorado in all kinds of snow conditions, but I imagine it would be pretty bone-jarring careening down a mountain slope at 140 degree ambient temperatures!


Another possible projection is the notion that Denver had reached a new permanently high plateau of temperatures in August, that 100 degree days would continue indefinitely into the future.  This wouldn’t be a good omen for the Colorado ski industry either.


Of course, if you are old enough to read both these assumptions rightfully sound absurd to you.  Because you have had firsthand experience observing decades of seasons, you know these myopic interpretations of Denver’s real temperature dataset are ridiculous.


Once a year our little rock Earth completes a circuit through the heavens and makes one revolution around our local star.  As it travels through space on its annual orbit our planet is not aligned vertically like a spinning top, but has an axial tilt.  Because earth is listing a little over 23 degrees to the side, as it revolves around the sun each year sometimes the Northern Hemisphere receives more sunlight at the expense of the Southern Hemisphere.


This annual variation in sunlight leads to the classic seasons we all love, spring, summer, autumn, and winter.  I can’t convince you that global warming is taking hold in Denver based on its real spring and summer temperature trend because you have lived through somewhere between 16 and 100 seasonal cycles yourself.  Once you experience something dozens of times, it is hard for someone to try to convince you that your senses have deceived you.


While using real data to forecast a blast-furnace of a December with 140 degree average daily highs in Denver is just plain silly, it is provocative that the exact same type of assumption is continually made in regard to the US equity markets. 


Because valuations of US equities have been high for a decade or so, the equivalent of “summer,” Wall Street is spending vast sums of money to attempt to rationalize to investors why high valuations are the new norm.  They are hoodwinking unsuspecting investors into believing the trend will carry future valuations even higher.


Last week in “Long Valuation Waves” I discussed a great market cycle that has ebbed and flowed like clockwork for at least a century.  Every third of a century or so, general equity valuations as measured by price-to-earnings ratios and dividend yields embark upon a wondrous journey.  Valuations begin at dismal depths of despair, way undervalued, travel to high-priced mania extremes over a couple decades, and then once again collapse to abysmal levels and the cycle begins anew like a phoenix.


These 33-year valuation cycles seem to have as great of regularity in history as the annual seasons of weather.  Unfortunately, while everyone easily recognizes the seasonal waves as they have lived through the cycle dozens of times, most investors and speculators totally miss the Long Valuation Wave because its wavelength is too long for them to perceive.


If Earth took 33 years to travel around the sun, it would be much harder to convince folks about the seasons, which would each last a little over 8 years.  Similarly most investors, if they start investing in their early 30s and retire at 65, only have an entire investing lifespan of around 33 years, so they have great difficulty discerning the Long Valuation Waves.


It is really unfortunate and sad that most investors fail to see the valuation cycles.  If they knew what “season” we were traveling through in the current Long Valuation Wave, they could make far superior investment decisions.  It makes no sense buying an overvalued market that is due to see its valuation shrink, or selling an undervalued market that is due to see its valuation multiply.


As I discussed last week, by late 1999 US equity valuations were at extreme highs that suggested the downturn in the valuation wave was approaching.  This week I would like to present some graphs of current Long Valuation Wave data from the Big Three US stock indices to cement the case that we are already traveling through a hazardous valuation wave downturn.


Before we unveil the data, it is crucial to understand the Principle of Mean Reversion.


Everywhere one looks, in life, in the natural world, and in the markets, it is obvious that extreme states of existence can’t last for long. 


Have you ever had a wonderfully fantastic day?  Have you ever had an utterly horrible day?  Both types of days are extremes that rarely occur.  Most of our lives are marked by days we consider average, neither spectacular nor terrible.  Things calm down after great days and get better after ugly days.  The days of our lives inevitably tend to revert back to their average state, or their “mean” as mathematicians call it.


In nature, mean reversion is everywhere!  For example, if a given population of animals becomes too large, too extreme for the food sources in the area, a die-off occurs.  After enough animals starve to death their population thins and the extreme is rectified by a mean reversion.  The lower animal population reduces pressure on the food sources, so they thrive.  Eventually enough food is available to support another animal population boom stage and the cycle begins anew.


Remember when single worthless dot-com companies were somehow valued higher than dozens of elite profitable US companies added together only a few years ago?  How about when US stocks were universally loathed and yielding huge dividends 20 years ago?  Neither of these extremes lasted for long.  The markets don’t tolerate extremes and continually revert to their means.


Last week in “Long Valuation Waves” we used a century of data to determine that the mean P/E ratio and dividend yield for the US equity markets are about 14x and 4.6%.  Every 33 years or so, the wavelength of the Long Valuation Waves, general equity valuations oscillate above and below these long-term averages like clockwork.


Today the US equity markets are decaying from very high valuations and grinding their way down to much lower valuations.  The Long Valuation Wave mean reversion has begun, and investors truly need to understand that we are already well into the season of declining stock market valuations and they need to deploy and protect their precious capital accordingly.


Our graphs this week document this ongoing Long Valuation Wave mean reversion with real-world data from the S&P 500, NASDAQ, and Dow 30.  Daily stock index closes are shown, but the actual valuation data graphed is monthly.  Each month we painstakingly compute the P/E ratio and dividend yield for the Big Three US equity indices for our Zeal Intelligence newsletter subscribers. 


We started doing these calculations in 2000 because it was almost impossible to find ongoing equity index P/Es and dividend yields that were calculated consistently.  We chose to carefully build our index valuation metrics off a base of individual company data.  In the S&P 500 for example, each month we go through all 500 companies one by one, calculate their individual valuation data, and plug it into a mammoth spreadsheet.


Using this conservative foundational data, we compute a market-capitalization weighted average (MCWA) index P/E ratio and index dividend yield for the Big Three US indices from the ground up.  The MCWA weighting is used so valuations are accurate based on actual capital deployed by investors and not skewed by small companies with disproportionately extreme data.


This highly-anticipated valuation data is available hot-off-the-presses each month exclusively to our subscribers, so they know where we stand in the current Long Valuation Wave.  Long-term investors need to have valuation data to help them make profitable strategic capital deployments.  Making strategic buying and selling decisions without valuation information is like trying to fly an airplane blindfolded, it will almost certainly lead to disaster.


The Zeal-calculated P/E ratios in blue and dividend yields in yellow are graphed on the left axes below, with the cheap, fair, expensive, and bubble levels of general equity P/E ratios we discussed last week noted.  The equity indices are tied to the right axes.  The raw index data is graphed in red, while the index fair-value based on the historical average P/E of 14 times earnings is graphed in white.


The early data is in, and the long-anticipated Long Valuation Wave mean reversion has begun!



The most important equity index on our little axial-tilted planet is the elite S&P 500.  This data most closely approximates Professor Robert Shiller’s landmark data I discussed last week.  This data could easily be spliced into the valuation graph from last week, although our MCWA weighting yields a higher top P/E than Dr. Shiller’s S&P Composite work.  If a simple average of the S&P 500 companies’ individual valuations is used however, the data fits perfectly with Shiller’s, commencing at a P/E of 42.6.


As the Long Valuation Wave Theory predicts, the S&P 500 has been falling from its overvalued levels of 2000 and excessive speculative valuations are gradually being bled off.  It is really encouraging to see the elite index trading below 28x earnings again, as this is the historical bubble level for a major stock index.  The markets are working just as they are supposed to and the speculative excesses of the bubble are being slowly devoured by the voracious Great Bear.


Even though we’ve spent a lot of time since 2000 computing our valuation data that we publish for our newsletter subscribers, this was the first time we graphed it all at once and I was really surprised to see how consistent the fair-value calculation for the S&P 500 was over time.  As the white line above indicates, fair value (14x earnings) for the elite index has bounced between 400 and 600 for a couple years now, indicating the vicious bear is doing his work remarkably methodically in lopping off prices.


Long Valuation Wave downturns inevitably carry P/Es well under fair-value levels, so it is almost certain that the S&P 500 will have to trade well below fair-value before the bust is over.  The mean reversion virtually always overshoots on both the upside and downside, and we are rapidly heading down now. 


I fully expect that the S&P 500 will trade below 14x earnings, probably approaching 7x, before we are out of the woods.  Probably sometime in the next 24 months or so, unless there is some miraculous jump in US corporate earnings, the S&P 500 itself will head below 400, horrifying everyone. 


If you remain long now, consider yourself warned that we are almost certainly heading a lot lower.  Discretion is the better part of valor and it makes a great deal of sense to sell now while valuations are still extreme and buy back later at undervalued levels.  This strategic macro-contrarian play, while emotionally tough to execute today, will probably multiply your overall equity holdings by several times if you buy back within a few months on either side of the coming ultimate Long Valuation Wave bottom.


Unfortunately the NASDAQ looks even worse.  We use the elite NASDAQ 100 companies to compute our valuation metrics each month, as these large companies provide a valid proxy for the much broader NASDAQ Composite as a whole.



While the ongoing dark comedy of the NASDAQ bust provides great fun and boundless opportunities for the short-term speculators as I discussed in “Volatility Trading the QQQs”, it remains hopelessly overvalued and exceedingly deadly for the long-term strategic investors.  The Long Valuation Wave mean reversion is proceeding right on schedule as expected, but valuations still remain extreme virtually guaranteeing the ultimate bottom is far lower than our recent October bounce.


At current earnings, in order to merely fall to the historical bubble top level, the NASDAQ would have to plunge another 35%!  After this current bear market rally in which we have been speculating on the long side rolls over, the next downleg will probably be extraordinarily ugly.  We are eagerly preparing for this upcoming Big Trade and have been discussing exactly how to play it for our subscribers in our Zeal Intelligence newsletter.  The short-side profits to be made in the rest of the bust will likely be legendary!


Although it is certainly possible that NASDAQ average valuations are higher than the historical general US equity average of 14x earnings because it is a growth-stock mecca, please don’t let anyone convince you the recent lows were the bottom.  They weren’t, just another in a long series of temporary bear-rally bounce points off oversold fear-driven extremes. 


Even if fair-value on the NASDAQ could somehow be considered as high as 21x earnings, that is still a long way down from here.  Caveat Emptor!  Barring some wondrous earnings miracle for NASDAQ-listed companies in the near future, the current valuation data strongly suggests the ultimate NASDAQ bottom will be under 400 on the Composite!  This terrible bust low will probably be achieved within the next 2 years. 


The valuation graph above ought to scare the tar out of those who are still long the embattled tech-laden index for more than a very-short-term multi-week trade!  If you value your precious capital that you have labored hard to earn all your life, you owe it to yourself to get out now on this bear rally before the relentless spiral down continues.


Finally, hoping to end on the good news, here is the Dow 30 Valuation Wave mean reversion graph.



The Dow 30 is the one bright spot in our valuation data painstakingly accumulated over the last couple years.  These elite 30 American industrial companies are continuing to earn profits even in these dark times.  Unlike the current nearly bubble-level S&P 500 and the unfathomably high NASDAQ valuations, the Dow 30 is closing in on what has merely been “expensive” in history!  This is a great sign that light is slowly approaching at the end of the long, dark, and painful bust tunnel.


Based on current Dow 30 earnings, fair value for the index is now running around 5000.  Out of all the Big Three indices in the States, the Dow has the lowest downside risk even though it will still likely have to trade under 4000 before the bust ends in the coming couple years barring some corporate earnings miracle.


While it is still far too early in Long Valuation Wave terms to go long the Dow 30 in a strategic long-term capital deployment, I am excited to see the mean reversion is apparently farther progressed in the Dow than the other two indices.  A little bit of hope, the hope that low future valuations and a once-in-three-generation strategic buying opportunity is approaching, can go a long way in these dark market times.


Just as a Denver weatherman relying on only five months of data to predict a 100 or 140 degree December will be ridiculed, so should the Wall Street crowd saying the same thing about the markets.  Prominent Wall Street cheerleaders ignoring the 33-year valuation cycles to predict higher valuations rising indefinitely into the future ought to be ashamed of themselves.


Future predictions, either in weather or stock markets, should only be made based on data that includes at least one full cycle.  A temperature cycle takes one year, so at least one year of temperature data must be analyzed to draw a valid conclusion on weather.  A Long Valuation Wave takes about 33 years, so any predictions on future valuations for US equity markets using less than a third of a century of data are bound to be horribly flawed.


The extrapolation of linear partial-cycle data leads to poor assumptions, silly rationalizations like the so-called Fed Model of Valuation which I intend to expose as rubbish in a future essay, and ultimately to the loss of hard-earned capital from investors who find that their trust has been tragically betrayed.


The Long Valuation Waves are already well into their long-anticipated mean reversion to undervalued levels. 


Long-term investors need to understand this immensely powerful phenomenon and evacuate their scarce capital from the overvalued Big Three indices before it is mercilessly crushed in the next big downleg.


Adam Hamilton, CPA     October 25, 2002     Subscribe