NINE ALIGN; BENIGN? NEIN! (April 2, 2024): We have recently experienced several simultaneous extremes which investors are ignoring due largely to two key factors: 1) investors will almost always believe that whatever has occurred in recent months will continue over the next several months, regardless of the merits of such a momentum argument; and 2) the Goldilocks myth of a "soft landing" has become so widely prevalent that the vast majority of individual investors in their retirement plans and hedge funds have become dangerously overcrowded into the most popular U.S. stocks. Nearly 100% of equity valuations are at or near all-time record overpricings, while many reliable signals of upcoming market behavior have approached or surpassed all-time record extremes.
Many investors apparently weren't paying attention in kindergarten. They should have learned that Goldilocks is always followed by the three bears. Mama Bear is backstage now, already cued up and ready to make her dramatic entrance. Speaking of cue, the most overvalued assets including QQQ are likely to lose one-third or more of their recent peak valuations, which would imply QQQ dropping below 300 at least briefly during the next twelve months. In the final months of 2022, all forecasters of the U.S. economy insisted that there would be a U.S. recession. Now almost none of them expect a recession. They're going to be as wrong in 2024 as they had been in 2023.
Several reliable signals have been flashing bright red.
All of the following warnings are close to their most extreme levels ever recorded: 1) we have the lowest-ever put valuations and the lowest overall "skew" of put valuations relative to call valuations since options began trading on the CBOE at the start of 1973; 2) we have the longest-ever inverted U.S. Treasury yield curve and we had recently experienced the most-inverted Treasury curve in its entire history going all the way back to 1789; 3) the U.S. dollar amount of insider selling divided by the U.S. dollar amount of insider buying set a new all-time record during the first quarter of 2024; 4) investor inflows set several historic records including the largest-ever weekly inflow into U.S. exchange-traded equity funds during March 2024; 5) most measures of valuation which had been favored by classic value analysts including Benjamin Graham and Seth Klarman either approached or surpassed all-time records going back to 1790 when the Philadelphia Stock Exchange had its debut; 6) we have experienced the greatest-ever divergences between the most popular large-cap U.S. shares and just about everything else worldwide; 7) investor sentiment is close to multi-decade peaks; 8) perhaps most importantly, far more investors are concerned about missing out on future gains than they are about losing money in the financial markets.
No trigger has even been needed to begin either a massive collapse or a major rally. Follow the insiders whenever their trading is near a rare extreme in either direction.
Whenever someone asks me which event will trigger a massive percentage loss for QQQ and related assets, the answer is best encapsulated in the following question: what triggered the 83.6% drop for QQQ starting on March 10, 2000? What "caused" the even larger percentage losses beginning in September 1929? How about January 1973 or the 19th-century bubbles which had begun collapses in 1837 and 1873? In other words, the financial markets have never been and will never become cause-and-effect processes. Whenever valuations for any asset are either unusually high or low relative to fair value, anything can spontaneously initiate a regression toward the mean and beyond. I had to laugh when I saw the news articles on Tuesday about why U.S. stock futures were lower. All of them related to events which had occurred over the weekend or on Monday during market hours. Can't the market respond immediately? Investors tend to forget that during the subprime mortgage housing price bubble collapse of 2007, which had begun in January of that year, most equity sectors kept climbing at first. The S&P 500 didn't peak until October 11, 2007, long after the subprime news was already widely known. Other analysts pegged the collapse of Lehman Brothers as the trigger, even though that occurred in September 2008. Probably that helped to accelerate a much later phase of the downtrend but it had already been in place for a long time.
Top corporate executives are well aware of this principle. They don't waste time debating what could "trigger" any future event in the financial markets. The reason that top executives overall and especially in the most-overpriced sectors have been selling at their most intense pace in history is because they are close to being pure value investors. They know better than anyone what fair value is for the company which they are helping to run, and when current valuations have deviated the most from this level. That is why insiders consistently buy aggressively near bottoms and persistently sell near tops. They are often "wrong" in the short run but they are rarely off the mark in the long run.
The unusually lengthy interval since the last true bear market encourages investors to believe that the next bear market is emotionally far away, even as it is much more likely to happen.
Many people including myself remain puzzled why the U.S. Federal Reserve didn't immediately begin to raise the overnight lending rates shortly after the U.S. Presidential election near the end of 2020. The real reason is probably that, not having experienced significant U.S. inflation since the early 1980s, most people on the Fed really couldn't imagine inflation surging the way it ultimately did. Similarly, with the last true U.S. equity downtrend ending in early March 2009, it seems psychologically to most investors that bear markets were something which happened a long time ago and are no longer a serious possibility.
The level of surprise is therefore going to be quite high whenever we undergo the largest percentage losses in over fifteen years for most U.S. stocks. Just about everyone has concluded that huge exchange-traded equity fund inflows will continue for many more years, that VIX will keep dropping below 13 for many more years, and similar "obvious" conclusions. George Santayana stated that those who don't remember the past are condemned to repeat it. My corollary is that those who do remember the past, but believe that "it's different this time," deserve to repeat it.
We have already been in bear markets for years, but hardly anyone pays attention to them because they don't involve the most popular shares.
The stock market in China began its downtrend during the second week of February 2021. Most small- and mid-cap U.S. shares also completed their tops in 2021, and even recently the Russell 2000 (IWM), after a strong rebound since October 2023, wasn't able to approach its high of February 2021, much less its brief spike higher in November 2021 when most U.S. stocks by count touched their highest points of their bull markets which had begun from the depths of March 2020. Since most investors don't track these or other emerging-market indices, or baskets of thousands of U.S. stocks, but only pay attention to the most well-known names, they are convinced that there is a solid uptrend for the entire U.S. stock market.
The most impressive percentage increases above fair value are inherently unsustainable, and are therefore consistently followed by the most impressive percentage declines.
You could argue that the rally for AI shares and related large-cap popular names since October 2023 has been impressive, and it has in the same way that the large-cap popular rally diverged from everything else and was impressive in September 1929, January 1973, and March 2000. However, all extremes come to an end sooner or later because the basic Boglehead premise is fatally flawed. There is a long-term uptrend to any asset, just as there may be a long-term uptrend in the temperature due to global warming, but that doesn't mean you throw away your snow shovels and winter tires in July and expect the weather to keep getting hotter. Mean regression has been proven by Benjamin Graham and Seth Klarman and many other value investors throughout history to be a much stronger force than the long-term uptrend, especially whenever the current distance from the mean in either direction is unusually extreme. That is why we had such powerful multi-year rallies which began in late 2002 and early 2003, and again in late 2008 and early 2009. It is the primary reason why we must experience an approximate repeat of the internet bubble over the next few years.
I am surprised how few analysts have pointed out the obvious parallels between the internet bubble of 1999-2003 and the current AI bubble 24 years later.
Usually nothing repeats exactly in the financial markets, although bubble collapses are so similar to each other that if I were to remove the X and Y axes on charts of the railroad bubble of the 1870s and the internet bubble from the early part of this century, and ask you to tell me which was which, I would challenge you to get it correct. In this case, however, we have almost an exact parallel in several ways, including the dramatic outperformance of the most popular large-cap shares in late 1999 and early 2000 which is almost exactly analogous to late 2023 and early 2024, almost to the exact day.
March 2000 is remarkably similar to March 2024. Nonetheless, most investors use the period from October 2023 through March 2024 as their guide to the next several months, rather than looking at what had occurred after March 2000 or January 1973 or September 1929.
In some cases the parallel is to the exact day, or as exact as it can be. The internet bubble for the Nasdaq and QQQ had topped out on March 10, 2000. This would have been impossible to occur on March 10, 2024, as it was a Sunday. However, on March 8, 2024, both XLK (a fund of popular tech shares) and SMH (a fund of semiconductors) may have completed their peaks for the cycle. QQQ could have done so later on March 21, 2024. In 2020, the S&P 500 Index (SPX) continued to receive inflows mainly from individuals in their retirement plans even after tech shares had topped out, causing SPX to reach its highest point so far on March 28, 2024. In 2000, the S&P 500 similarly experienced a delayed zenith on March 24, 2000. That's about as close as you're going to get in real life. This doesn't guarantee that QQQ will drop 83.6% as it had done from its top on March 10, 2000 until its bottom 31 months later on October 10, 2002, but it has to be a realistic possibility especially since by many measures this index was more overpriced in March 2024 than it had been in March 2000.
We have ended or are near the top of the sixth U.S. stock-market bubble in its history. It will collapse the same as the previous five.
In the 1830s investors insisted that since we never had canals before, we wouldn't experience a severe U.S. equity bear market regardless of how overpriced many of the canal shares had become by 1837. The same fairy tale was popular in the 1870s regarding railroad stocks, in the 1920s about industrialization, in the early 1970s regarding the popularization of computers, in the late 1990s with the internet, and now in modern times with AI. All of these involved huge societal changes, but they had zero net impact on the financial markets because the financial markets have always followed and will always follow the same basic principles. If a company has a certain level of profit, and that profit is increasing by a certain percentage, then regardless of whether you're living in 1824 or 1924 or 2024 you can compute how much that stock should be trading for. If investors are depressed or excited about investing, thereby distorting this number in either direction, it creates an opportunity for alert investors who recognize that all such distortions are temporary and will eventually fluctuate toward nearly equal and opposite extremes. During the late 1920s Benjamin Graham repeatedly insisted upon this key point even with loud Bogleheads like Irving Fisher insisting upon the Boglehead point of view even before it had that name. The worst bear market in world history proved that mean reversion will always triumph at any extreme, no matter how "permanent" famous analysts claim that it is.
We don't just have an AI or large-cap bubble, although that exists and is especially dangerous. We have simultaneous perilous overvaluations for cryptocurrencies, gold, residential real estate, and many other assets. The percentage loss of total worldwide net worth is probably going to experience one of its biggest-ever declines in history.
During previous historic bubbles including Tulipmania nearly four centuries ago in Amsterdam, it wasn't just tulips which collapsed in price. There had been simultaneous bubbles for residential real estate, the stock market, and numerous related assets which are not generally remembered by most people but which can be reviewed in detail for free online at "Extraordinary Popular Delusions and the Madness of Crowds" by Charles Mackay. Residential real estate is overpriced for the same fundamental reason as U.S. large-cap stocks: the ratio of housing prices to average household income in many countries including the United States is close to double its long-term average level. People's incomes in real terms are not going to suddenly double, which means that housing prices will have to fall by half in real terms. Gold has probably been in a long-term uptrend since August 1999, but the traders' commitments for both gold and silver, showing commercials net short about 2.7 to 1 for each, are a warning sign that the most-experienced investors who own actual gold and silver as miners, fabricators, jewelers, and so on, are betting on the price dropping. Gold will probably fall by a minimum of 400 dollars an ounce from wherever it is topping out in this cycle, and perhaps by more than 500 dollars, while silver will likely drop below 20 U.S. dollars per ounce at least briefly later in 2024. Once silver commercials go net long, as they almost did in March 2023 and had actually done in September 2022, it will be time to buy into this sector in the anticipation of a powerful rally.
As for cryptocurrencies, no one has any idea what they are worth. At least a stock with a price-earnings ratio of 50 will become a bargain when its P/E drops below 8, but any given cryptocurrency is only worth whatever someone else believes that it's worth. Peter Pan is one of my favorite children's stories by J.M. Barrie, but you can't use Peter Pan logic to justify any investment.
The following charts highlight rare multi-decade extremes from reliable media sources, some of which became their most exaggerated in a century.
The bottom line: the most popular shares as a group, as are epitomized by funds including QQQ and XLK, have been making many upward spikes since February 12, 2024 but haven't been able to remain above those key intraday highs. Hedge funds, which by far had been their most net long any asset in history when they had dangerously overcrowded into AI shares and related large-cap U.S. favorites, will likely curtail their buying or actually do some net selling, although as usual with more than 90% of hedge funds they won't become serious sellers until these and related funds have already dropped 15% or 20% from their respective March 2024 tops. We will continue to experience frequent sharp upward bounces for U.S. stocks for about three years, with some of those rebounds for QQQ approaching or exceeding 50%, but during that time all of the major indices will join the Russell 2000 in primary downtrends which almost certainly means a loss of more than 70% from its peak for the S&P 500 Index and more than 80% for funds of the most popular shares including QQQ. During the next twelve months, when hardly anyone is anticipating a significant drop for the U.S. stock market, QQQ will likely at least briefly spike sharply below 300 before recovering. Over the longer run, QQQ is likely to fall below 100 and perhaps much lower by 2027 or sooner. This will surprise most investors, but compared with its October 2002 or November 2008 bottoms it would be a very similar "Papa" bear-market bottom adjusting for inflation and earnings.
Disclosure of current holdings:
Below is my current asset allocation as of 4:00 p.m. on Tuesday, April 2, 2024. Each position is listed as its percentage of my total liquid net worth.
I computed the exact totals for each position and grouped these according to sector.
The order is as follows: 1) U.S. government bonds; 2) shorts; 3) bear funds; 4) gold/silver mining; 5) coins; 6) miscellaneous securities.
VMFXX/TIAA(Traditional)bank CDs/FZDXX/FZFXX/SPRXX/SPAXX/Savings/Checking long: 37.13%;
26-Week/17-Week/52-Week/13-Week/2-Year/8-Week/3-Year/5,10-Year TIPS/4-Week/42-Day long: 21.66%;
I Bonds long: 10.88%;
TLT long: 10.86%;
PMM long: 0.01%;
XLK short (all shorts once again unhedged): 30.58%;
QQQ short: 20.17%;
XLE short: 5.48%;
XLI short: 3.05%;
XLV short: 1.82%;
SMH short: 1.33%;
AAPL short: 0.03%;
SARK long: 0.88%;
PSQ long: 0.04%;
PALL long: 0.17%;
GDXJ long: 0.14% (fully hedged with covered calls at a strike price of 45);
Gold/silver/platinum coins: 7.01%;
FXY long: 0.24%.
