For the U.S. Equity market, the advance in recent days to yet another string of new all-time highs is outwardly so impressive that it makes it difficult for many market observers to even question the robust nature of the stock market. Through last Thursday, for example, the NASDAQ Composite had seen 14 of the last 16 days with a positive close. As winning streaks go, that kind of one-sided market action is a fairly rare phenomenon. For the NASDAQ Composite, this has only happened 15 times in its 46-year history. With the likes of Facebook and Amazon and other FANGS seemingly on an unending roll, it’s enough to make one wonder whether stocks will ever go down again?
Yet it is precisely this kind of one-sided feeling that often creates and denotes the presence of a major market turn. In his recent Weekly Technical Update, veteran technician James E. Welsh zeroes in on this mentality:
“Will Stocks Ever . . . If you type ‘Will stocks ever’ into your browser, Google will auto-fill this question with ‘go down again’. This is the mentality that results when the S&P trades beyond 2 standard deviations as it has in 2017 and discussed last week. Since 1928 (89 years) the S&P has averaged a decline of 11.2% in the first half of the year. The largest decline in the S&P has been 2.9% in 2017, about one quarter of the average and the second lowest ever. The S&P has not experienced a decline of 5% in more than a year.
This is only the sixth time that has occurred since 1950, third time since 1965, and the longest stretch without a 5% pullback since 1995 22 years ago. There have been 6,951 trading days since 1990 when the Volatility Index (VIX) was introduced. The average for the VIX since 1990 has been 19.51. The VIX has closed below 10.0 a total of 21 trading days since 1990, or .302% of the time. Of those 21 days, 17 have occurred since May 7. On Friday, the VIX posted its lowest close since December 1993. At some point in the future asking Will Stocks Ever will result in the answer ‘go up again.’
That will occur only after the S&P has declined for months on end and is trading more than 2 standard deviations on the left side of the Bell Curve. It may occur when global investors lose confidence in central bank’s ability to suppress volatility and lift economic growth. If this ever occurs, it will be a seminal event because the majority of investors don’t believe it will ever occur. They may prove correct in this assumption, but ignoring the possibility is foolish since the potential downside could make 2008 look like a warm-up act.”
With the VIX under 9 for one session last week, the level of bullish complacency is just remarkable. Never mind that very little progress has been made on any of Trump’s key Republican economic reforms—not on healthcare, nor taxes, nor repatriation or infrastructure spending. Bulls suggest now that expectations for progress on these fronts are “priced out of the market” and that if anything were to actually get accomplished, it would be yet another bullish surprise!
This kind of warped mindset is completely ludicrous since the market advanced sharply with expanding P/Es on the prospects that Trump’s election would be the end of gridlock and the beginning of real progress in terms of economic reform. Since stocks are still retaining all of the Trump gains (and then some), clearly, nothing has been priced out of the equity market as P/Es continue to inflate. Yet, nine months into the new administration, all we really see are the deeply divided political fault-lines within the Republican party, fault lines that make it clear that gridlock never left, and that any real progress on economic reform issues may be as elusive as ever.
In this sense, the spectacle of blow-off bull market dynamics and unending bullish forecasts spewing on cable TV makes it very difficult for the average stock market observer to see the forest for the trees. For those seeking the true 30,000 feet view on today’s stock market outcome, only the lens of history can really provide the needed objectivity. When seen through that lens, it becomes very clear that time, as much as price, is a critical variable. As readers who finish this report will see, at its current advanced age the current stock market bull is an aging and very endangered species. History also goes on to show that at this stage of a bull market, the downside psychological risk profile is acute and something that few investors can probably have the emotional constitution to stomach.
To access the history of bull markets, it is useful to periodically review the length of prior bull cycles in terms of evaluating how the current bull market compares. To do this, I started with the Dow Jones Industrial Average which has data going back to the late 1880s. For the purposes of keeping things simple, our approach was to locate every bull market low, and then check to see how long the bull market was able to run without a decline of 20% or more. Using these simple criteria, we find that of the longest bull cycles, the 1990 to 2000 bull cycle for the DJIA was the longest ever totaling 483 weeks in length. The 1921 to 1929 bull cycle was the 2nd longest at 419 weeks with the average of the longest bull cycles clocking in at 237.88 weeks.
To take an even wider perspective—one that is more global and less U.S. centric—the next step was to go back through the history of some of the world’s other leading stock indices with long-term price histories. For this part of the study, we used Japan’s Nikkei from the 1960s forward, the German DAX Index, the Paris CAC-40, the Sydney All Ordinaries, the London Financial Times 100, and Hong Kong’s Hang Seng. By including these indices and the NASDAQ, we gain a pretty good 30,000 foot view of some of history’s greatest equity market manias including the NASDAQ 2000 Tech Bubble and Japan’s late 1980 stock mania. Back in the 1840s, journalist Charles Mackay chronicled man’s folly with financial manias. Were he still alive, surely some of these “tulip-like” episodes would have been worthy additions for an extra chapter in his iconic book: “Extraordinary Popular Delusions and the Madness of Crowds”.
If we average out the total of the top 32 longest bull markets using the DJIA, the NASDAQ Composite, and an assortment of other leading world stock market indices, we find that the average duration to the top 32 longest global equity bull markets of the past hundred plus years is essentially 255.2 weeks. Importantly, the average bear market decline that followed the completion of history’s longest bull markets clocked in with a hefty -42.97% over an average of 65.78 weeks.
Yet, as important as this data is, it was not the end of our study. To dig in deeper, we studied each one of the longest cycles and found that within some of the longest there were what could best be described as statistical odd-ball events. Was this unusual? No. After all, what statistical series does not have its share of odd-ball events? In a number of cases, where the longest cycles extended to record breaking length, there were actually deep corrections that occurred that fell short of the 20% bear market marker, but nevertheless, were deep enough (15% to 19.99%) to allow the bull market to have worked off some of the prior accumulated one-directional upward excesses. This, in turn, then allowed the bull a second life, where having regrouped and self-corrected, it could then re-assert itself for another advancing wave. Unfortunately, there is no easy way of coming up with one standard cut off point that fits all sizes so to take the study a bit further, we employed our best judgment. We also tried to make some slight allowances for the differences in beta between some of the various markets. As an example, for markets like Hong Kong, Japan, and the NASDAQ, 20% declines are more routine then the lower beta, blue-chip DJIA. So, in markets with higher betas, we used a somewhat wider benchmark, with 25% for the Nikkei, and 30% for the Hang Seng and NASDAQ. A good example of a clean bull run would be in the chart below, which shows NASDAQ from 1990 to July 1998, where the advance proceeded upward with a long series of consolidations and more modest pullbacks before it ran into the Asia crisis of 1998 and a 30% decline. At that point, the long low volatility bull-run ended at 406 weeks.
Above: the NASDAQ Composite went 406 weeks between October 12, 1990 and July 24, 1998 before finally succumbing to a 30% decline, when it then fell -33.08% in the ten weeks between July 24, 1998 and October 9, 1998.
An example of a more complex time series to decipher was the NASDAQ run of 1974 to 1984, with a decline from May 29, 1981 to August 20, 1982 that totalled 28.31%, but did not exceed a cut off of 30% until July 1984. In sorting through all this data, our approach was to give the bull case the benefit of the doubt every time. So, using 455 weeks for the NASDAQ's longest run, versus 348 weeks which would have been the total using the 1974 to 1982 stretch shown in the example on top of the next page.
Above: the 1974 to 1983 rally in NASDAQ lasted 455 weeks and climaxed with the first real tech mania in June 1983. Between 1974 and 1982, the NASDAQ had two 20% corrections, one from 2/8/1980 at 165.25 to a low of 124.08 on 3/28/1980 (down 24.90%), and the second from 9/15/78 at 139.25 to 110.88 on 11/17/78 for a decline of 20.37%. Using a full 25% cut off brought us to the May 1981 to August 1982 decline, and using a 30% cut off led to the June 1983 to July 1984 low. With a simple 20% cut off, our original starting point, none of these historic advances in Tech stocks would have made the list, because the 20% draw downs were too frequent.
After going back through all the data, we ended up with the distilled table below, which encapsulates in the fairest way (we believe) possibly some of the most basic information regarding historical bull market advances. The summation of the 32 longest prior bull market yields an average time duration of 265.20 weeks. In our view, the price gains on a bull market are a lot more arbitrary. Far more important is the time elapsed between meaningful declines, because meaningful declines work off bull market excesses and reset the clock. Some may argue that bull markets do not have a clock or a time limit, but history strongly argues the other way. More importantly, a glance down the Bear Market percent decline column shows that after these very long advances, there tends to be a very strong bear market decline. The average price decline totalled 42.60%, which is an emotionally devastating figure for most investors to absorb. The average length varied quite widely because some bears markets strike with crash-like speed, while others are drawn out over time. The average bear market decline was 61.22 weeks, or a year and three months.
Above: Average of longest bull markets across a variety of global stock indices adjusted for relative degrees of underlying volatilty is 265.20 weeks on the upside followed by on average a 42.60% decline in subsequent bear markets which spanned 61.22 weeks on average.
So where are we today? Since the bear market lows of March 2009, the NADSAQ Composite has seen three corrections of 18.70%, 20.39%, and 19.53%. These occurred between 4/30/10 (2535.28) to 7/2/10 (2061.14) down 18.70%, 5/06/2011 (2887.75) to 10/07/2011 (2298.89) -20.39%, 7/24/2015 (5231.94) to 2/12/2016 (4209.76) down 19.53%. Yet, in the entire time, the NASDAQ has not seen a decline of 30% or more, which is the most realistic cut off we found for a “bear market correction” using the history of NASDAQ data. Put another way, it has now been 438 weeks since the March 6, 2009 lows where the NASDAQ has not seen a bear market. That would rank this stretch as the 2nd longest bull run ever.
Above: Could time be running out? Is it “Time’s Up!” for the NASDAQ Bull? The NASDAQ is now 438 weeks out from the March 6, 2009 lows, matching up closely with the 455 week bull that ended in 1983, and the 406 week bull that saw an important peak in July 1998. The average of the two longest prior bull cycles for NASDAQ was 430.50 weeks without a 30% decline. We are now past that total suggesting NASDAQ is historically overdue for a major bear market decline of 30% or more.
Alternatively, using the Dow Jones Industrials and a 20% cut off as the delimiter for a bear market we find that there have been two decent sized corrections from the 2003 lows. The first was from 5/16/2011 to 10/14/2011 and it measured -16.80% on a daily basis and -15.62% on a weekly closing basis. The second was the decline from May 22, 2015 to August 28, 2015 which totalled 16.24%.
Yet, with no downside behavior of 20% or more, the Dow Jones Industrials have now gone, like the NASDAQ, an amazing 438 weeks without a 20% decline. That is 19 weeks longer then the 1921 to 1929 bull market, which in many ways is still the strongest and longest bull ever seen and this is now 1.85 times the Dow’s historical average of 237.46 weeks!
What is most interesting for investors right now is to overlay the current historically over-stretched length of the bull market with the plethora of other major bearish signals that are currently at hand. Even as the NASDAQ and DJIA stretch their bull market runs to new lengths, we see the stock market about to head into its seasonally worst period of the ten-year cycle. Going back over the decades, we find that within the so-called decennial pattern, it is the 3rd and 4th quarter of the 7th year of the decade that has often contained a significant downside wallop. In late June this year, noted technician Larry Williams did an interview on MadMoney with Jim Cramer and pointed to this cycle which is due to peak in late July. According to Williams, this could kick off a larger scale correcion with stocks falling through November.
Historically, years ending in 7, and especially the 3rd and 4th quarter of years ending in 7, have been very bad times to invest money in the stock market. There seems to be a pretty consistent 10-year cycle that runs through the stock market and can peak as early as the middle of the 6th year of the decade, but tends to intensify and bottom late in the 7th year of the decade. In the past, 1907, 1917, 1937, 1957, 1977, and 1987 all packed a big downside punch, while investing in stocks in 2007 was teeing up the global financial crisis which wreaked havoc on financial assets in 2007 and 2008. We show a brief visual tour below courtesy of Just Signals.com.
The charts above show the weakness that tends to develop in the sixth and seventh years of the decade leading to a cycle low late in the 7th year. The 5th and 9th years of the decade tend to be strong upside performers. We are entering into the worst part of the ten-year cycle in the equity markets over the next four months and are historically overdue for even a small decline. So far, the Dow has not seen even a single 2.5% daily decline since the Brexit vote on June 24,2016 and before that there were three such down moves in 2015. Since 1920 there have been 448 daily declines of 2.5% or more in the Dow, but this period of 2012 to 2017 has seen the least number of 2.5% Dow down moves since a comparable period ended at the last major Dow top on October 25, 2007. It is also noteworthy that neither the Dow nor the S&P has experienced a 5% down day since 2011.
Above: Absence of Volatility. The percent of 2.5% down days reaches the lowest level since the top in 2007, namely October 26th, 2007. After that, bad things started to happen and volatility took off.
Elsewhere, in the realm of technical indicators, we see a great many signs that suggest a stock market sell off of substantial degree is close at hand. For many years, we have used our own formula to compute a weekly Advance-Decline line for the stock market based on weekly advancing and declining options traded on the CBOE. In many ways, this gauge presents a cleaner outcome, because the widely published advance/decline data has changed a great deal in recent years. The broad tape has been contaminated by a wide variety of redundant and non-operating company issues that really pollute the data for what had been in the past one of the best leading indcators for the stock market. Using the CBOE data, we generate several key weekly oscillators. The first one, shown in the chart below is what is known as a Detrend Oscillator and it plots the difference between our CBOE Options A/D Line and its own 78-week moving average. In the past, we optimized this gauge to get the best major trend signals. What it shows at the moment is that the stock market is now coming off a set of fully overbought readings and, if prices start to decline in the broad market, the next 1% down from here on the S&P (2470.30 Monday July 31, 2017) should trigger a failure swing. A failure swing occurs when the indicator first makes a primary high, then has an initial decline, followed by a second lower high, and then a second decline that undercuts the initial low. On this gauge, the recent primary peak was a reading of +53.02 on 3/03/2017, followed by an initial decline low of +38.39 on 4/14/2017. The indicator then attempted to move up off that low and failed at a lower high of +44.31 on June 9, 2017. Last week, it ended just above the 4/14/17 low at a reading of +39.41. Any move down in the averages from here is likely to trip off this sell signal and cause the Detrend Oscillator to move below the mid-April low. That would flash the kind of intermediate-term signal akin to readings seen in 1987, 1989, 1998, 2007 and early 2016.
Above: CBOE Options A/D Line Detrend Oscillator rolling over after major overbought Trump peak!
A similar companion tool, also derived from the same data set, is showing an extended period of bearish divergence. These are warnings signs that the primary uptrend in the equity market is losing internal strength and getting close to a downside reversal of size.
In addition to the CBOE Weekly A/D data, we also maintain our own version of the stock market ARMS Index, which is also known as the Trading Index or TRIN. When the TRIN gets down to very low values, it is at the very least a strong warning that the equity market could be near a major peak. In this case, we believe the ARMS is sporting a bearish divergence with prices, where prices have gone to new higher highs and the ARMS has not gone down to new lower lows. The present value of the ARMS is .7599 and is very low, warning of a potential market peak.
Above: The S&P ARMS Index at .7599, is still holding above the .706 value seen on June 19th, and the .683 value of 2/27/17 and that tells us that internal strength within the stock market is weakening.
In addition to flashing red lights from key breadth and volume gauges, we also note that sentiment at the moment is matching the ultra-complacent signals coming from the VIX. It is very rare to see this kind of wildly bullish sentiment this late in a stock market cycle, and what’s more, as prices are making new highs, sentiment is at extreme levels and is downwardly diverging. In the case of Investors Intelligence, which tracks sentiment from newsletter writers, the number of Bulls reported in last week’s survey topped 60% while there were only 16.50% bears. In the last 28 years, there have only been 37 weekly closes above 60% out of more then 1400 weekly readings. While the percent bull figure is worrying as it occurred back on October 19, 2007, the high preceeding the start of the Global Financial Crisis, the number that is even more concerning is the low number of bears. If the bear figure drops even a little further it wil be a new multi-year low taking us back to July 2015, when the S&P topped and then began a nasty August swoon falling from 2140 to 1867 on the Chinese Yuan devaluation.
In my work, I plot the net of the two figures, Bulls minus Bears and run a 5 week exponential (weighted) moving average. That moving average is near the ultra-high end of the long-term range and is diverging against price. This is a classic late-cycle warning for the beginning of a potentially serious stock market decline. Other sentiment gauges, such as the BofA Bull and Bear Indicator are also pressing up against very high end of the range type readings that suggest on a contrary opinion basis, we are looking at a stock market caught up in a euphoric buying wave that has now gone too far in one direction.
Above: the Investors Intelligence +60% Bull figure was also seen at the top of the 1987 run and preceded the great crash of October 19th. This year’s price action is very similar to what was seen in both 1987 and in the run up to 1929 where we overlay the S&P 500 in 2017 against the final advance into the major peak in August 1929. The overlay for both markets, 1987 and 1929, are shown in the charts above.
Other Key Ideas:
While we are not specifically calling for a market crash, at the same time, it should be noted that crash risk right now is very high on a number of metrics. When we look across so many aspects of the market, we see nothing but gigantic red flags. Notice that the leadership of the market, which had been the FANGS, as we noted several weeks ago, has begun to falter badly. Alphabet/Google has become a target for the EU with fines, and now Amazon could soon be the target of anti-monopoly/anti-trust probes as an incensced President looks to lash back at one of his biggest detractors, the Washington Post, which is owned by Amazon’s Jeff Bezos. All of these charts went through a classic parabolic blow off and, one by one, seem to be rolling over. Even as the NASDAQ “broke out” to new highs in recent days, the venerable FANGS were split badly down the middle with only Facebook and Netflix doing the heavy lifting. Both of which now look fully exploited and ready to pull back. We show the charts of the FANGS against the NASDAQ Composite to highlight how even within the narrow ranks of an ever thinning and more and more select leadership, the market has been losing internal strength.
Above: One has to wonder, is Amazon moving into the anti-trust break up crosshairs with the Trump Administration, and could that be why the stock is suddenly starting to decline?
For reference, see:
- Senator Cory Booker thinks the U.S. government should keep an eye on the size of Amazon, Google and other tech giants
- Another Democrat in the U.S. Congress is sounding alarms about the Amazon-Whole Foods deal
Above: Apple was unable to make new highs, and is starting to pull back.
Above: Pricey Tesla had an exhaustion move up, collapsed, and now has a failing high.
Above: Bellwether Google/Alphabet could not make new higher highs and is moving down across the range. The shares are close to key support at 6. If this stock breaks down it would be a very bad sign for the market at large.
Above: the SOX Index, which tracks Semiconductor stocks, has been a huge leader and now many of these big name techs are faltering: AMAT, KLAC, LRCX, STX, WDC, MU, AMD, MXL.
Above: In the world of Dow theory, a falling Transportation average is also a big red flag.
This report is far too short to even come close to capturing all of the excesses and divergences that are embedded in the stock market at the current time. It does not delve into the ultra-high valuations, and the potential downside risks of a crowd mania that rapidly embraced ETF and passive investment strategies. It does not delve into many of the sordid accounting details where real earnings have not grown at all in the last four years (see below), and where price appreciation has been predicated purely on P/E multiple-expansion. All of these are separate essays unto themselves.
Source: Real Investment Advice
Yet, hopefully, for the end reader this report has put together a portrait of a market that is historically over-extended and uncorrected.
It also does not delve into the current state of Fed policy where we see a set of Fed tightening moves unfolding against the backdrop of what still appears to be a very fragile domestic U.S. economy. In its current state, it is hard to argue that the U.S. economy has picked up at all, with this year’s reported GDP still averaging right around 2%. For all of the talk of 4% growth post the Trump election, there has been no material fundamental change, and no underlying improvement in a wide range of hard data. Inflation, wage growth, and labor force expansion are all missing in action and to that end it looks very much like the Fed is in the middle of one major policy mistake typified by the introduction of Quantitative Tightening, i.e. balance sheet reduction. In our view, this is draining liquidity and taking away the punch bowl at the very time the Adminstration may be about to pursure tariffs on China and the EU. As Dr. Marty Zweig once said, “Don’t fight the Fed” and right now, everything coming from the Fed sounds generally hawkish with a bias toward tightening. In our view, the other key phrase, “Don’t fight the Tape” is also noteworthy as too many sectors within the market are now moving lower, with the Dow Industrials moving up in rear guard fashion like they did at the 1973 high as the infamous Nifty 50 market was just turning down.
Over the last five sessions, right through the morning of Tuesday, August 1st, the DJIA has advanced for five consecutive new all-time highs, moving from 21,675 all the way up to 21,988. Over that same period of time, with the exception of the Defense sector ETF “ITA”, which has been powered by Boeing (BA), (which is also the stock that has driven the Dow Industrials), not a single additional index, nor any of the S&P 500 ten main sectors was able to make a new all-time high. NOT making new highs are the following: Consumer Discretionary (XLY), Consumer Products (XLP), Energy (XLE), Energy Services (OIH), Financials (XLF), Healthcare (XLV), Industrials (XLI), Materials (XLB), Real Estate (XLRE), Technology (XLF), Utilities (XLU), Transportation (IYT), Home Building (XHB), Semiconductors (SMH), Drugs/Pharma (PPH) nor have the S&P 500, Russell 1000, Russell 2000, Wilshire 5000, Value Line Arithmetic. Hopefully, this makes the point—this has been an incredibly narrow advance and history shows that narrow advances tend to exhuast and roll over hard.
In closing, I will end with one final observation on the S&P 500, where the 100-day moving average has been advancing for what is now an amazing streak of 299 consecutive trading days. This is not a moving average that turns down easily as it is intermediate-term in nature and normally requires a sizeable decline to effect directional change. The chart below goes back to 1970 on a daily basis and we are currently in the 3rd longest streak ever seen. The median of the last 20 long streaks was 156 trading days and the current streak is virtually double that and closing in on all-time records of 355 and 352 days. In the prior two longest streaks, the advancing portion of the move peaked at 329 days and 310 days respectively.
While anything is certainly possible, in the stock market there is seldom anything new under the sun, despite frequent advertising to the contrary. Excess greed begets excess fear and so forth and so on. This is one of history’s greatest episodes of excess greed and the lesson from history is that when the eventual reversal does arrive it will be anything but mild. While it is also true that no one ever rings a bell at the market top, so much of the current data right now seems to be lining up and challenging real historical boundaries that the odds of a sustained upward movement from here look unlikely. August has been a month to host many a major stock market peak, including the highs of August 1929, August 1987, August 2000 and August 1937. In our view, the time to be as defensive as possible has been, and remains at hand, as probabilities seem to be stacking up in favor of a serious downside reversal. For the S&P 500, any break below the key level of 2450 would probably signal the beginning of a protracted sell-off that could persist for some time over the days and weeks ahead.
That’s all for now,
For more information on Salem Partners or their services, please go to www.salempartners.com.
Disclosures: Salem Partners Wealth Management and the Salem Capital Markets Outlook is sent to our clients and friends to share some of our views on current markets. Materials in this report are copyrighted and proprietary to Salem Partners Wealth Management (SPWM) and may not be reproduced or distributed without our expressed written approval. All rights are reserved.
Past performance is no guarantee of future results. All opinions expressed constitute Salem Partners Wealth Management’s judgment as of the date of this report and are subject to change without notice. This information should not be used as the primary basis for investment decisions.
It should not be assumed that any security transactions, strategies, holdings or sectors discussed were or will be profitable, or that future recommendations will be profitable or will equal the investment performance discussed herein. Investing in equities and other securities involves risk to your principal. Individual client account performance will vary and no guarantee of any kind is made as to the results that may be obtained from using these reports. These reports are impersonal and do not provide, nor are they intended to provide, individualized advice or recommendations for any specific audience or portfolio.
To the maximum extent permitted by law, Salem Partners LLC, Salem Partners Wealth Management and the authors of these reports disclaim any and all liability in the event that any information, commentary, analysis, opinions or advice contained in these reports proves to be inaccurate, incomplete or unreliable, or result in losses on any investment. The information provided in these reports is obtained from sources which the author believes to be reliable. However, the author has not independently verified or otherwise investigated all such information. Neither the author, nor SPWM guarantees the accuracy of any such information.