Making Hay Monday - May 15th, 2023

High-level macro-market insights, actionable economic forecasts, and plenty of friendly candor to give you a fighting chance in the day's financial fray.

May 15, 2023

“The safety risks around AI are huge. We think there is a more than 50/50 chance AI will wipe out all of humanity by the middle of the century.” -Peter Berezin, BCA’s Chief Global Strategist and Director of Research

Well, that one got your attention, didn’t it? It certainly did mine. In my case, the first time I saw this was Friday morning as I was riding a stationary bike in the fitness center at one of my favorite restorative venues, Sunriver Resort in Bend, Oregon. As usual in my OCD life, I was multitasking: simultaneously watching CNBC, soundless, while listening to a podcast by MacroVoices Erik Townsend interviewing an expert on the coming U.S. government debt crisis. (As most of you know, this is one of my main investment themes for 2023; a link to this important discussion is at the end of this post.)

Ex-sound, the headline on the 50/50 chance of the Final Apocalypse flashed across the screen. Since I couldn’t hear what was being said, I could only watch BCA’s Peter Berezin mannerisms and body language. He was thoroughly calm and matter-of-fact. It was like he was discussing his firm’s forecast for the economy in the second half of the year.

Later that day, my wife, our two dogs and I drove from Bend to just outside Portland. As usual, she was at the wheel, allowing me to catch up on the research reading on which I’m always way behind. Coincidentally, one of my ultimate go-to guys, Luke Gromen, was riffing on AI, as well. The part that sent chills down my spine was a short excerpt he ran from Jeff Booth’s 2020 book on AI, The Price of Tomorrow.

In it, Luke quotes Mr. Booth writing about the game of Go which, I have to admit, I have never played, or even read much about. Apparently, it’s the oldest board game in the world, dating back to 500 B.C. China. It is fiendishly complex, with 10 to the 780th power of possible positions; i.e., a 1 followed by 780 zeroes.

Until about 10 years ago, this complexity caused even leading Artificial Intelligence researchers to believe AI would not be able to defeat human GO masters in the foreseeable future. Yet, in 2016, a Google AI program, DeepMind/Alpha Go, defeated one of the planet’s best GO players. This program had learned GO by emulating the play of thousands of human pros and amateurs.

But here’s the part that really slapped me upside the head:

It (Alpha Go) made history not only because it was the first time a computer beat a top Go master, but also because of the way it did so. In game 2 and the thirty-seventh move, the computer made a move that defied logic, placing a black stone in the open area—away from the other stones.
Top players in the world commentating first dismissed the move as a mistake by AI, but then realized it was no mistake. The move was brilliant, and AlphaGo went on to beat (the Go Master) in the game and win the five-game match 4 – 1. Later, pundits would say how creative the move was. It was the first time that an AI was ever said to be creative, a domain always thought to be solely owned by humans.

The emphasis on “creative” was by me because, in my non-expert opinion, that’s the amazing — and terrifying — breakthrough. It gets scarier. A year later, Google’s software geniuses released an update version, AlphaGo Zero. And zero is what the original AlphaGo scored, as its successor wracked up 100 straight wins. Remember, the earlier iteration beat one of humanity’s best.

Spine-chilling time again:

…not only was that version much more powerful than its predecessor, it also didn’t require any ‘training’ from human games. Understanding only the rules of the game, AlphaGo Zero became its own teacher, playing itself millions of times and, through deep reinforcement, getting stronger with each game. No longer constrained by human knowledge, it took only three days of the computer playing itself to best previous AlphaGo versions developed by top researchers, and it continued to improve from there. It mastered the masters, then mastered, itself, and kept on going.

Once again, that emphasis was from yours truly because the thought of AI-self-created programs going viral is, to me, the ultimate nightmare. Obviously, the folks at BCA share my concern and those of my creative colleague Mark Joseph Mongilutz. Mark has written passionately on this topic in Haymaker and via his own blog. It’s fair to say, he was early to recognize AI’s promise and perils, at least in the circles in which I run (or, at my age, jog).

By the way, BCA is not some tech think tank. The acronym stands for Bank Credit Analyst, and for decades I’ve followed the firm and its former chief economist Martin Barnes. Martin retired in 2021 after 50 years in the “dismal science” and 34 at BCA. His final Barron’s interview that year turned out to be spot-on as he warned of inflation busting out to the upside. He also anticipated the stagflationary environment in which we now appear to be enmeshed. Suffice to say that over the years, BCA has been one of world’s foremost macroeconomic forecasting firms.

Consequently, it’s somewhat surprising it is taking such a dramatic position on AI. Clearly, it believes there are substantial economic and financial implications from its explosive growth, as well as increasing adoption. (Recent mega-cap tech conference calls were overflowing with AI references.) In that regard, it is opining that the explosive productivity growth potential of AI could lead to global GDP doubling in short order. (Perhaps I missed it, but I didn’t hear how short.)

Share Haymaker

In case you think all of this is much ado about not very much — a reasonable attitude given how many hyper-hyped breakthroughs disappoint — consider this visual from Jeffries’ star analyst, Christopher Wood. Clearly, this is a situation where the term “game-changer” is not an exaggeration.

Peter Berezin is unquestionably of that mindset. He’s telling the world to think exponentially about AI, not linearly. He likens its proliferation to the exponential way in which Covid spread in early 2020. Yet, based on the opening quote, he’s obviously beyond wary about the risks it poses. He noted that the early developers strongly advised against allowing AI access to the internet. They were also emphatic it should not be allowed to write its own code. Despite those sensible recommendations, it has been permitted to do both. He further believes the leading AI incubators, like Microsoft, have been far too complacent about its dangers. This echoes a view passionately conveyed by Mark, my aforementioned colleague.

One simple example Peter cites of the dangers of AI, in the absence of effective guardrails, is if it was asked to solve climate change. As he notes, eliminating human beings via a nuclear holocaust may be a solution an unsupervised AI program might select.

Enough of the dystopian stuff! One positive aspect of the sudden focus on, and fears about, AI is that preventive measures are being seriously discussed at the highest levels of the public and private sectors. (Personally, I have a lot more confidence in the latter keeping the situation under control than the former, though both have critical roles to play.)

From an investment standpoint, I’m going to cover a possible application we are pursuing regarding my beloved “range expansion” methodology in the Champions section. But perhaps more urgent is the need for all investors to review their holdings for potential AI victims. The recent nuking of educational services company Chegg, because its business model is being ravaged by AI, is a prime example.

For hard-asset aficionados like me, I believe there is safety in natural resources that AI can’t replicate or replace. Further, when it comes to developing and perfecting small modular nuclear reactors, it should be most helpful, particularly given their inherent safety versus traditional light water reactors.

To sum up, AI is like all technology revolutions, including atomic energy: it has enormous potential but equally massive risks. It’s going to take some enlightened policymaking to reap the benefits while avoiding the existential threats. Unfortunately, “enlightened” and “policymaking” don’t seem to go together these days. We, the voters, best change that at the same speed with which AI is spreading.

Champions

It’s finally time to unveil what I’ve been hinting at in terms of a possible shopping list for Haymaker readers involving stocks achieving three-year breakouts. To clarify, this refers to those equities which have recently broken out of a 36-month trading range. This is what hedge fund investor extraordinaire, Paul Tudor Jones, calls range expansions. As he has observed, and I’ve repeatedly experienced (particularly in my personal long/short portfolio, where breakouts I’ve ignored have been extremely costly), once these occur, the stock in question tends to keep running in the direction of the resistance penetration. As I’ve written several times, the longer the trading range has been in place, the more meaningful the signal. My team and I have found that three years is the minimum length for one to be considered crucial, give or take a few weeks. However, real-world experience has taught us that longer is better than shorter.

As I’ve also previously admitted, these certainly don’t work all the time. There are definitely false breakouts and breakdowns. But we’ve found that they work a higher percentage of the time. In my view, they are especially helpful in avoiding big losses, like when stocks seem to have no support anywhere north of zero. Some regional banks lately have been examples of that kind of value vaporization.

So, without further adieu, here is our list of stocks around the world that have made relatively recent three-year breakouts. To be clear, I am not recommending any of these; rather, for the many do-it-yourself investors who read this newsletter, I’m merely suggesting these are worthy of additional research on your part.

(Japanese Stocks Bolded, one of which is owned by Berkshire Hathaway)

To learn more about Evergreen Gavekal, where the Haymaker himself serves as Co-CIO, click below.

Private Wealth Management

Personally, I have a preference for those that have been in relatively tight three-year ranges and then bust out in an upward direction. It’s best to make sure it is a decisive breakout. From the above list, here are a few that have charts that interest me. (This is not to say Evergreen clients or I own these; in fact, at this point, we don’t, but that’s subject to change in the future.)

(BMW below)

Regular readers know I have a preference for lower P/E stocks, or at least growth-at-a-reasonable-price (GARP) names. Three of the above sport very modest P/Es and the one that is pricey, Vertex, is a fast grower. Its earnings per share have been growing at around the same level as its P/E, creating a low price-to-earnings growth, or PEG, of 1.

There are a number of other eye-catching charts included on this list. You can check those out yourself using Yahoo! Finance and clicking on the five-year trading range history.

We did also identify those companies on here that are domiciled in Japan. The fact that three made this short list (only six are from the U.S) surprised me a bit. However, that is in keeping with the overall Japanese market having experienced a breakout to multi-decade highs last year.

One point I should have made last week in my “Buy Japan” Making Hay Monday was that the U.S.-based ETF is quoted in US dollars (USD). The yen’s weakness versus the USB, until last fall, masked a fair amount of the underlying strength by the Nikkei, Japan’s equivalent of the S&P 500. Now, with the yen recovering vs the USD, that headwind is turning into a tailwind. Moreover, the yen remains one of the most undervalued currencies vs the USD. (As noted last week, there are negatives involving Japanese stocks. These include a history of poor profitability relative to the U.S., especially the cash machines known as mega-cap tech.)

You’ll also note that the majority are international companies, beyond those from Japan. That does fit with one of my favorite long-term investment themes of investing overseas for at least the next few years. Moreover, some of the U.S.-listed names shown above primarily operate overseas, like Bollore, Coca-Cola, Euro-Pacific and UBS. You’ll also notice that several of the U.S. entities trade at rather punchy P/Es, though certainly not all of them.

Another preference of mine is to emphasize companies that have long track records of consistent and healthy profits growth. This is not to say cyclical stocks like BMW, Paccar or United Rentals (URI) can’t be solid performers. However, based on the recession I see coming, if it’s not already here, being highly selective with economically sensitive stocks is advisable.

Please provide us some feedback on whether you find this screening process helpful… or not. We do listen to our readers… particularly when we hear the same message from many of you.

Champions List:

Japanese stock market

Select financial stocks

U.S. Large Cap Value

U.S. GARP (Growth At A Reasonable Price) stocks

Oil and gas producer equities (both domestic and international)

U.S. Oil Field Services companies

S. Korean stock market

Physical Uranium

Swiss francs

Copper-producing stocks

For income:

Certain fixed-to-floating rate preferred stocks

Select LNG shipping companies

Emerging Market debt closed-end funds

Mortgage REITs

ETFs of government guaranteed mortgage-backed securities (alternative approach)

Top-tier midstream companies (energy infrastructure such as pipelines)

BB-rated energy producer bonds due in five to ten years

Select energy mineral rights trusts

BB-rated intermediate term bonds from companies on positive credit watch

Contenders

The telecom equipment sub-sector has been performing very poorly of late. In fact, one of the leading players in this field broke three-support. This triggers an automatic sale in my book (literally).

However, its arch-rival is still holding well above three-year support, though it has fallen sharply since 2021. That was when it was briefly caught up in the meme mania. Its shares are extremely cheap on a valuation basis and the long-term outlook for this de facto duopoly remains highly positive. Accordingly, I’m moving telecom equipment down to Hold/Contender status rather than Sell/Down For The Count, despite the stop-loss triggered by its competitor.

Telecommunications equipment stocks

Swiss francs

Singaporean stock market

Short-intermediate Treasuries (i.e., three-to-five year maturities)

Gold & gold mining stocks

Intermediate Treasury bonds

European banks

Small cap value

Mid cap value

Select large gap growth stocks

Utility stocks

Down For The Count

Horrible affordability causing a predictable plunge in sales combined with an increasing reluctance by lenders to lend. The perfect set-up for a 34% surge by the homebuilders since last October, right? Well, apparently so, because that’s exactly what’s happened.

One of the best bullish arguments for this odd occurrence has been the alleged shortage in housing supply. The reason for the “alleged” is that my good friend Danielle DiMartino Booth is challenging that nearly universal belief. In fact, I’ve made the point about a supply deficit on occasion despite my negativity toward the homebuilder stocks.

Here’s what she wrote on this topic last week:

Even more infuriating is that the undersupply narrative is accepted as gospel, even among professionals. The public data we’ve seen in recent years suggests that investors have been responsible for between 25-33% of total housing transactions in recent years. According to Amy Nixon, a rising star in the housing analysis sphere who coined the hashtag #Airbnbust, “Massive amounts of private investor off market deals weren’t reflected in data the last two years.” In her podcast today, she had as a guest a former Airbnb operator who relayed that it was just a year ago that realtors would hop onto Facebook groups and ‘sell’ 40 new homes in an hour, sight unseen. “These never hit Multiple Listing Service, they weren’t documented. The bottom line is, there is no structural building shortage. Investors are simply hoarding massive amounts of properties. It’s safe to say they will dump them en masse if a distress element enters the market.”

Danielle also ran the following chart. It’s unique in looking at available housing units on a per capita basis. As you can see, it tells a much different tale than the prevalent storyline, dare I say “myth”, that the stock of homes for sale is tight.

As someone who routinely reviews hundreds of pages of macro-economic research, I can assure you the notion of a below-the-radar supply overhang is not a mainstream view. It certainly could be off the mark, but it has the ring of truth to me. It’s well known that investors (i.e., non-traditional buyers) played a key role in driving home prices well above their 2007 peaks, even adjusted for inflation. Assuming they are now facing minimal, possibly negative, cash flow as rentals dry up, they might well become highly motivated sellers.

Perhaps they won’t “dump them en masse”, per the above quote, but there could be a considerable amount of stealth supply hitting the market soon. (Danielle also highlights the surge in vacation home purchases in recent years. Unsurprisingly, duplicate housing overhead can create a lot of heartburn for owners during recessions.)

Regardless, I don’t see how home prices can stay this elevated for much longer. Nor do I think sales activity will recover — at least, not much — particularly should long-term interest rates zoom again in the second half of this year. As I’ve described a number of times recently (perhaps too many) the odds of the U.S. government having extreme difficulty in selling longer dated bonds at sub-4% rates are unusually high, in my mind.

Consequently, I’m reinstating my negative view on the homebuilders. Full disclosure: In my personal account I have added to my short position in them lately. Thus far, I’m down a bit on this position but I like my chances in what I believe will be the second phase of the equity bear market that began early last year.

Homebuilders stocks

Meme stocks (especially those that have soared lately on debatably bullish news)

Financial companies that have escalating bank run risks

Electric Vehicle (EV) stocks

The semiconductor ETF

Meme stocks (especially those that have soared lately on debatably bullish news)

Bonds where the relevant common stock has broken multi-year support.

Long-term Treasury bonds yielding sub-4%

Profitless tech companies (especially if they have risen significantly recently)

Small cap growth

Mid cap growth

MacroVoices Episode:

Haymaker