End of Year Letter
Outlook for 2022
Sunday, 2 January, 2022
Published irregularly, as conditions dictate.
Historically extreme valuations, raging inflation, and bubble-character, wild sentiment, across all asset classes has set the stage for a long-overdue asset price collapse. As such things have played out historically, the damage done is so severe and pervasive, that to advocate for a long position, ahead of what I believe is the inevitable, is to be entirely irresponsible. Playing with a loaded gun is never a good idea.
There have been times in the past I was very bullish, to-wit, beginning in January 2016 and throughout that year, in talks before various family office organizations, my bullishness was met mostly with scoffs. Currently, it may seem that I have been bearish forever. Market tops are extremely hard to time, contrary to market bottoms. But I believe this one will be so severe that the only responsible strategies I can see are to either be out of risk assets altogether (in cash, or short-term credit instruments, or whole life insurance) or be short. If you think you can time the top, I wish you all the luck in the world. I speak from firsthand experience; you can foresee such things coming, but unless you are short and somehow manage to stay short until it hits, you won’t catch it - and that goes not only for getting short into a major market drop, but of getting out of your longs as well. This is why I say there is and has been for the past year-plus, only two, long-term responsible positions to play all markets, from stocks to crypto to real estate to gold to art — out or short. I believe the coming damage will be colossal, and I will point out why I believe so.
The fact that this has been so long coming, I liken to seeing a huge, approaching wave, offshore. You can see it’s enormity. Yet, it keeps coming, taking so much longer than you expect, each passing moment, the massiveness of wave grows before your eyes and you realize that it is much bigger than you first thought, and was much farther out to sea, initially, than you first thought, given its size.
Roughly once a century, the prices of equity shares in commercial operations have suffered drops in valuations of 70-75% or more, taking decades recover from. When such events occur, they impact the individual investors to an extent greater than the percentage drop of the averages.
Two years ago, prior to the onset of the Wuhan virus, I could see we were headed into a market drop and recession. I never thought I would live to see conditions set up for a market devastation as they have now. Not that all measures of the market indicate such, some indicate quite the opposite. The picture is always murky, and the approach taken is one of the “Weight of the Evidence,” which I find, in the aggregate, extreme and to a degree that is being ignored, unseen or discounted by everyone, even those who claim to be very bearish.
Let’s review the weight of this evidence, by means of the armamentarium of what I call “The Main Pillars of the Equities Market,” of which I break into six, broad categories:
This has been the most reliable forecasting tool of the longer-term direction of equities prices historically. However, with the advent of the Wuhan virus, since about late 2019, it has been a laggard with respect to major movements in the equities market. This is why I employ a weight of the evidence approach. Thanks to the credit market pillar, in particular the yield curve inversion in August, 2019 and other economic measures, we could see the trouble coming. In fact, the employment measures we track (virtually all of them), didn’t post and turn negative until the drop in February/March of 2020 was finished, and then were late turning positive. These measures we keep, with data ranging from weekly jobless, continuing claims, unemployment rates, participation rates, part-time workers, job quits, etc., have magnificent historical track records. However, the recent signals have been troubling. Although employment is still quite positive, these measures are all in a state where, in as small a span as two months (or perhaps even less) they could all register quite negative.
2 Credit Market Health
This too is one of the stronger pillars, but also one that appears to be rolling over now. The Constant Maturity Treasury Curve, though still “normal” in shape, is showing several inflection points at this time. Inflection points are important as they tend to underline the statement the curve is making, and because inflection points create inverted portions along the curve. The curve is very steep out to about 7 years, then only about an 8 basis points gain in yield out to ten years, where it steeps again to 20, and inverts out to 30.
We may not have an inverted curve, yet, but an equally-reliable warning of recession, and one which has historically been more timely than a yield curve inversion, has been year-on-year (YoY) change in CPI exceeding 5%, a condition we clearly are experiencing here.
In terms of what occurs in the credit markets portent to equities, however, credit quality spreads are more meaningful than what is going on in the yield curve quite frequently. On measure we track of this, the main one, is the metric of the monthly Moody’s Baa rated bonds difference to ten year yields, vs an 18-month moving average of this difference. As is evidence in the graph, below, this is rapidly closing and could close within a month:
3 Economic Indicators
The last, big positive we see that we track, for both the economy and equities markets, falls into this category and pertains to international trade.
Typically, the “Balance of Trade,” that is the difference between exports minus imports, is often commonly examined. However, we have found that in terms of market signals, looking at the sum of imports and and exports, that is, the aggregate of international trade that the US is a partner in, is quite an historically-reliable signal. Specifically, a system we call “Art Vandelay,” whereby long signals occur when the monthly aggregate exceeds the aggregate of the prior six months, a long equities position is warranted and vice versa. This system is currently long:
In other words, despite what we are hearing of supply-chain bottlenecks, the port problems, etc., the aggregate of US imports and exports is record-high. The data we have on this goes back the the beginning of 1992, and the Art Vandelay System is always either long or short. In this time the system has recorded 12 of 16 winning trades (75%) not including the current long position this system would be in since September 2020. However, the main drawback to this system is that the signals tend to be quite late, and very often at important market tops, evidenced in the chart. below, where the blue sections represent being long, the white sections being short:
Many economic indicators are not adjusted for population growth (e.g. durable goods) or for inflation. One such indicator, one of the most major of all economic indicators, is retail sales, which must be adjusted for inflation. Like many indicators, there is a general upward, persistent drift to it with little deviation, and it is best viewed as a percentage off of its highest point. When retail sales are adjusted for inflation and viewed as a percentage off its highest point we obtain:
From which, as can be seen, we are clearly entering levels associated with serious market drops.
I’ve often made mention that Industrial Production is the dance partner to the stock market, and we see this in the chart below, with the log value of the SP500 in black and industrial production in blue:
As with many economic measures, Industrial Production has not recovered as far as its all time high put in late 2018. despite the market surge since. This is a dance step that, in all past cases, resolves itself.
Often displayed with Industrial Production is Capacity Utilization, shown above in green. Of-note is the long-term downtrend this indicator is in.
Almost all of the economic indicators fall into one of these two categories. Either they are not up to their pre-virus peaks (or, at best, just marginally above, like personal income less transfer payments), among these, cumulative jobs created, truck tonnage, temporary help, or they fall into the long-term decay-type indicator where the peaks were put in years if not decades ago. These include capacity utilization, M2 velocity, labor force participation rate, housing starts, auto sales, inflation-adjusted durable goods per capita, etc., etc.
Things are simply not healthy, nor have they been for a very long time, and this is especially so since February 2020.
Of note is auto sales. I’m into my fifth decade in this business. I’ve witnessed some incredible bear markets and crashes. Most of my data I have been keeping by hand since about 1980 - my colleagues were in college and I was working as a margin clerk and trading options for myself. Much of my data goes back into the 19th century, some to post-Civil War.
I have never seen anything lining up in my data as it has been for many months now, that is so alarming, and this has prompted me to look at a deeper history of markets beyond that which I have experienced. As I mentioned, every 100 years or so we see declines in the valuation of equity of operations that exceed 70-75%, and I strongly believe we are coming into such a situation.
By the Summer of 1928, auto sales took a tumble, and we see this in the data presently:
I am very skeptical that this is merely explained away as a “computer chip shortage.” the profits of AAPL don’t seem to be too impacted by such a shortage. The Art Vandelay System has been clipping right along of late.
I would argue further that the “whys” of the collapse in vehicle sales matters not, that what does matter, is that there is a collapse in such sales. To engage in an explanation of such is to spin on the “this time is different” delusion that such things don’t matter this time, that the rhyme of history isn’t a true rhyme here.
There are X million autos on US roads, and there is a metric of Y years average age for autos on the road. Yes, there may be a computer chip shortage and it may be still ongoing or being alleviated, but the US consumer has bought everything forward from autos to toilet paper. Further, there are not only creepy parallels to prior, historical market crashes, but there are longer-term trends visually evident in a great deal of the economic indicators - like the Retail New Cars graph, above, which indicate something much bigger long in brewing.
4 Technical Market Condition
I seek to be non-redundant with what one is getting in other market letters. Everyone is talking about the lack of market breath here, how the lion’s share of gains in the indexes have been driven by a handful of stocks. How so many stocks - and many favorites, are well below their highs put in months ago.
All measures of market breadth have topped out. Whereas I have seen such divergences occur in years past and then resolve in favor of the indexes, very often, historically, market tops occur in conjunction with such divergences, and these conditions are clearly in place across the board now.
I have made mention in the past of a 7 year market cycle, a 7-year cycle of “liquidity events,” where every 7 years we find a vacuum in market liquidity:
My colleague, Stan Harley, posits it this way:
“Beginning in February 1966 significant stock market highs have occurred at 74-94 month intervals. My regression analysis computes a recurring cycle of 83.3 months – just a hair under seven years – with an R-squared value of 0.99917. The next recurrence in this high-to-high sequence is due, ideally, in the November 2021 time frame – NOW – with a standard deviation of 6.36 months.”
5 Valuations & Inflation
Not seeking to be redundant, I assume all readers are familiar with the hyper-valuations as expressed in the Buffet Indicator, the Schiller CAPE PE and the Q measure. I post these here for later historical reference:
Schiller CAPE PE:
Tobin’s Q (The Fed changed the calculations of one of the inputs this past Summer. The red line is showing it by the old calculation [where the 1929 peak saw a value of 1.06] with the blue line using the new calculation, still in excessive territory):
And finally, here, the Deviation from Trend. An inflation-adjusted, total return image of the S&P is plotted on a lognormal scale, with an exponential best-fit drawn through it(appearing as a straight, dark blue line). The red line is the number of standard deviations above or below this bests fit line, now at a whopping, historical all-time high of 4.56 standard deviations above.
These valuations - by virtually any metric, are at historically high levels, and only the CAPE ratio in 2000 being marginally higher than the current reading. Price-to-book is where it was in 2000 and Price-to-sales at levels never before seen. By all measures of valuations, we are in a bubble, and by most, at values never before seen. Further, we are only referencing equities here, but pervasively low interest rates have driven all investors into risk assets, like stocks, simultaneously. This is a condition not seen before in my experience.
There are, however, three charts exposing valuations which I have not seen elsewhere, which I believe are the most important measures of valuation, which highlight the current extreme circumstance and, in fact, make my blood run cold.
The first of these pertain to inflation-adjusted earnings. As is typical late in a bull market, investors justify excessive valuations with ever-greater expected earnings. This we see currently.
However, when earnings are adjusted for inflation, a different picture emerges. When we adjusted earnings as a percentage by the YoY change in CPI, when this becomes negative bad things have happened in the equities markets. Even in late August / early September 1987 we saw this, and several weeks later the market crashed. Currently, despite the so-called “blowout” earnings we are witnessing, on a real basis, these are negative to an extent not seen since 1947:
As evident, since the end of WW2, negative real earnings are damning for stock prices. Worst-still, CPI is now a dampened calculation than in the past, and had it not been tinkered with, we would be seeing real YoY earnings less than -10% right now. In fact, we would have to see earnings that correspond to a PE of 15 on the S&P at today’s earnings level and today’s YoY Inflation level to see real YoY earnings of 0%. That would be an S&P 500 level of under 2450! Even at the optimistic forward earnings currently posited by Factset of 21.20, we are at a YoY real earnings level of nearly -2%, and that too is a calculation based on the modern, dampened CPI.
The next point pertains to rolling ten-year average annual real total returns, currently at +12.06%:
This may be the most important long term chart of all in the sense of knowing where you are in the bigger picture. In markets, mean-reversion is a rule. In terms of the rolling average annual real total return, when this value is greater than 12%, one wants to be lightening up on equities at the very least, and when less than -7% to be accumulating equities at the very least. A reversion to the mean here, that is, just to return to a value of 0 on the green line, above, would be to see an SP500 value over the next several years of 1525 to 1600. That’s just to return to zero on the green line.
Next, there have been long periods where the 10 year bond real return (in red, below) exceeds this ten-year average annual real total returns (again, in green, below):
And if we examine the difference between the two, it constitutes the third chart that makes my blood run cold here
At no point in the history I have on this data, going back to 1871, have we seen values this extreme, where the ten-year average annual real total returns is 18.41% above the real return of the ten year bond. Artificially low interest rates have driven investment dollars into risk assets to a point never seen before, and are seek to defy the rule of mean-reversion. Either interest rates (on the ten year, in particular in this graphic) must rise in a manner that would cause the demise of all countries, or equities (and, in kind, all risk assets) decline such that their total returns are well below already low interest rates. The latter course, clearly the more viable incarnation of the rule of mean reversion.
The valuation situation is so historically extreme and by such a large margin, that it is the elephant in the room. There are no arguments that can effectively counter it (“blowout earnings,” or Art Vandelay or the current employment picture) and to disregard the extreme readings going on in this pillar of the market is to miss what is going on and what I believe is coming up, particularly when one considers it in light of (A) such excess valuations are going on in all risk assets and (B) the current sentiment situation, discussed next.
The published sentiment numbers - AAII, NAIM, etc., do not show excessive optimism regarding equities. I do not know if they ultimately will. That is, it is conceivable we get a straight-up run, say, in January, that does drive these published sentiment numbers to extremes. Or maybe not, as we’ve witnessed extremes in them months ago and mere non-confirmation along with new highs since.
But look around. Everyone is long. Everyone is long everything in the everything bubble, and acting like it is never going to stop. Making money in the markets is hard, and at no point in my experience have more people made more money more easily, more people who, with clearer heads should have cashed in and gotten out and haven’t, than now.
Oh yes, people have cashed out of crypto, or stocks, or real estate - to put it into not cash or near cash or whole life, but into another silo in the everything bubble, driven by the notion of TINA, “There Is No Alternative,” due to low rates, particularly now that there is raging inflation. The Fed, having clearly fueled this - the correlation below visible and undisputable:
This has created a bubble of 56 trillion in US listed equities, 46 trillion in residential real estate, 3 trillion in crypto, and on and on. The magnitude of these markets, should the music stop and rapid selloffs begin find the Fed with nowhere to go on rates, and, absent solid, deep bids in these markets, are they expected to be the “buyer of last resort,” to the tune of how many trillions? And we would expect that to save investors in risk-based assets?
More people are “daytraders” now than during the 1999-2000 bubble top. Despite the notion that there are too many shorts for the market to sell off (a notion not supported by put open interest or volume) the overwhelming majority of people on earth with any assets at all are lined up on the long side of the everything bubble, from sovereign wealth funds (most middle eastern ones being virtually 100% long equities) to public and private pension plans (many of whom have increased equity exposure relative to bond exposure and increased their leverage), everyone’s 401k, IRA - everyone’s retirement plan virtually, all residential homeowners, etc., about the world, all lined up on the long side of the everything bubble, acting like it’s never going to pop, with expectations of gains in the double digits for these assets that they are long in 2022.
The sentiment - the one-sided nature of where people have their money, not what they “say will happen” to the markets, along with the insane valuations in a world where mean reversion of such is a rule overwhelms any other positive indications (e.g., employment conditions, credit spreads, imports and exports ) and such positive conditions will likely turn negative quickly once the drop is underway.
One of the characteristics of the everything bubble is that “this time is different,” is consistent with all previous bubbles - the notion that this time is different, we have evolved and this is a “new normal.” This type of rhetoric is common to all bubble tops.
One thing may be different this time that I cannot find an historical parallel to is the sort of irresponsibility leading into this one. That does set this one apart.
The irresponsibility isn’t’ merely the federal government’s desperation to continue to pile on debt or the abject irresponsibility of the Fed we have witnessed for the past 22 months, but everywhere we turn. The proliferation of listed levered products, products that require more stock be sold, exponentially to price, as things drop (which is exacerbated by the heavy weighting of certain stocks that comprise a large portion of most of these products) is relatively new compared to other historical bubble. God only knows what sort of levered derivatives on equities lurk beneath the surface here whose exposure will become evident when a major drop occurs.
There is no more specialist system, no more system where there is an obligation to make bids, and an entire generation of traders (many who have risen to the ranks of PMs) who have never experienced a "no bid" market situation such as we saw in October, 1987.
Most damning of all in the current environment I suspect, and one no one is talking about is the lemming-effect created by social media among investors, creating behavior patterns not unlike how a group of migrating birds fly through the air. This is new. It has created a mentality of “racing to the middle of the herd,” in almost unconscious fashion and in extremely rapid manner. The very mechanism which fueled the meme stock mania of 2021 is a double-edged sword.
I suspect when the drop comes - and we may go straight up for a prolonged period before - will be, I suspect, extremely swift, taking a mere few hours, and to a degree that leaves the world apoplectic and stunned.
Of course, that wont be the end of the drop, but I suspect the initial phase will be much swifter than that seen on October 19, 1987, given the sentiment, the mean reversion rule of valuation, and the systemic differences in culture and people’s interactions and capabilities today.
Historically extreme valuations, raging inflation, and bubble-character, wild sentiment, across all asset classes has set the stage for a long-overdue asset price collapse, the initial phase of which I believe will be very rapid. I believe the short side of this represents an opportunity of several lifetimes, something I never expected to see in my working years, and the long side being a position that will ultimately prove entirely irresponsible. In the data I see the prospect for a market drop of magnitude seen roughly once every century, and there is reason to believe that this may ultimately be the biggest one ever seen. I am uncomfortable with such seeming hyperbole, but this is what I see in the data for 2022.