Through The AI Noise…
(Originally posted March 11, 2024)
Let me start off by stating a few things, with perfect clarity, to ensure that there are no misconceptions about my views.
First, and unfortunately, we must lay out that this is in no way investment advice. This is merely my own view of today’s circumstances, how I think things will play out, and which ways I’m inclined to play these views. As with anything, everyone must do their own research and evaluate what is best for them and their financial situation.
Second, this is not an opinion about the so-called “AI Revolution.” Even me calling it “so-called” is not intended to throw shade at the growing developments we’ve seen out of this technology. In fact, what I’m about to write has absolutely nothing to do with either a positive or a negative outlook on Artificial Intelligence. This is not a bullish AI article nor is it a bearish AI article.
Third, I’m a macro trader whose positions are typically held anywhere from a couple of weeks to three-to-six months. I’m also not a long-term fundamental investor, which means I change my mind A LOT.
Lastly, I understand that whenever we talk about the market or specific investments, many participants adopt a tribal nature. You’re either full-blown bullish or miserably bearish – there’s no in-between. I appreciate that there are people who know far more than I do about the topics with which I’m about to cover and am always happy to hear constructive feedback.
Ok, so now that we have a few disclaimers out of the way, I think it’s about time we get down to what the heck it is I wanted to write about. See, for me, writing is a therapeutic exercise. Aside from my duties as Chief Investment Officer at Strom Capital Management LLC, I write a daily research note through another company I founded back in 2019.
The reason I started that research service is entirely different from why I continue with it today. Back in 2019, I was seeking to share my views of the world. I didn’t have your prototypical background of spinning out of a large hedge fund where investors were lining up to hand me nine-figures to manage. I was trying to get my voice out there and make a name for myself.
But today, I am lucky enough to run my own hedge fund and the research has taken on a different purpose. I enjoy writing. It helps me organize my thoughts, which is often a difficult thing to do in this crazy world of information overload. I do write to keep my investors informed of how I’m thinking about the world. But I also do it because it serves as a gut-check on my portfolio. If what I’m writing contradicts my portfolio, then something is clearly off.
Sometimes, I don’t even know what I’m thinking and feel completely out-of-tune with the market. I can’t figure out why this or that is happening. I don’t understand what investors are seeing that makes them tilt one way or another. Or sometimes, it’s just that the economic data and market action is so conflicting that I don’t feel comfortable taking a strong stance.
So, here we are in mid-March of 2024 and I’m watching Semiconductor stocks rally anywhere from 4% to 15% every day. And while I do appreciate a great momentum story, I simply can’t bring myself to buy something with so much hype behind it and so far from any realm of rationality based on any number of technical indicators. Please note – I’m not even bringing up fundamentals here because I’m not a sector specialist and, as I said at the beginning, this is not an argument for or against the legitimacy of AI-driven demand.
That’s why I sat down to write this today. I admit, I’ve been on the sidelines of this parabolic move in AI, particularly Semiconductors. With momentum as part of my process, I even bought the initial breakouts in names like NVIDIA Corp (NVDA) and Super Micro Computer (SMCI) at the beginning of 2024. But I admit, I didn’t fully understand it and the moment things felt a little silly to me, I bailed. That was in late-January of 2024. Yeah, talk about a serious face-palm moment.
So, instead of making some hasty decision to go long or, please hold the gasps, dare to go short, I figured I’d try to organize my thoughts and let it all spill out onto paper (well, virtual paper in Microsoft Word).
But before we get into the nitty gritty, I’d like to share some details about my overall macro view. Going back to early-November of 2023, I adopted a risk-on view for equity markets. This was a shift for me because I had been relatively cautious since the summer of 2023 and actually made some pretty decent money being short things outside of the United States. Yes, while the Tech rally was in full effect throughout all of 2023, there were growing signs that the rest of the world was slowing dramatically, and many foreign equities were confirming that narrative. We even saw leading indicators here in the United States roll over into strong contractionary levels.
For me, one of my rules to market cycles has always been to watch foreign equities, especially those ETFs that do not hedge currency exposure. The reason being, a stronger US Dollar tends to weigh on the rest of the world and any weakness should be manifested through poor performance of those assets.
We’ve seen this throughout numerous cycles since the Global Financial Crisis (GFC). Late-2014 and all of 2015, foreign ETFs lead with a strong bearish trend. Eventually, the US market followed with some turbulence in late-2015/early-2016. Then the rebound in early-2016 as China unleashed its massive fiscal stimulus. Again, foreign ETFs flipped bullish and rallied hard until early-2018. That was followed by another rollover into a strong bearish trend in mid-2018 and, of course, resumed its bearish tendency just prior to the carnage from the Covid Crisis.
Flip around, we saw a strong rally back in late-2020 as all that pent-up fiscal stimulus juiced the economic cycle. And then we saw a dramatic rollover in Q4 of 2021, just before the US equity market hit the skids with its 2022 “bear market.”
Well, between August and September of 2023, we saw another bearish flip from a long list of foreign equities. I had been short, covered the October lows, and was looking to get short again on a technical relief rally. After all, foreign assets were still within a strong bearish trend and leading indicators were screaming “GLOBAL RECESSION.”
Then it all changed. That relief rally quickly turned into a bullish trend flip for foreign assets. We saw decelerating rates of inflation from all around the globe. And investors were eager to embrace the idea that the Federal Reserve would soon be cutting interest rates. Oh, and not to mention, the strong seasonal tendency of markets to rally over the final months of a strong year.
I was bullish and I really liked many of the beaten-down areas of the market (dynamics of tax-loss selling and a new calendar year). If inflation was easing and the market loved the idea of rate cuts, then all those areas that had been panic-sold might experience a strong rebound.
But then I got to thinking, maybe the collapse in leading indicators was more a factor of an inventory correction cycle, as opposed to a full-blown recession. After all, companies spent the tail end of 2022, and most of 2023, dealing with excess inventories that were purchased in fear of the same supply-chain constraints that wreaked havoc during Covid. Company after company echoed the excess inventory story and, as we worked through those, manufacturing and production contracted.
If we go back to November of 2023, it wasn’t just that the market action was telling me something was wrong with a bearish view of the world. It was also that we started to see hints of a trough in the same leading indicators that confirmed the bearish market trends since late-2021. Again, this is speaking to trends in foreign assets and cyclical equities, not the Magnificent 7 (“Mag7”) Tech rally of 2023.
I’ll admit, I struggle with ‘growth’ investing and 2022 was just another example of the dangers in participating in something you know nothing about. I avoided ‘Growth’ and Tech in 2022 but I also didn’t play its rebound in 2023.
And, yes, this was probably a strong reason why I was too jumpy to stick with the Semiconductor and Technology trades in early-2024.
Instead, I believed there was far more opportunity in the beaten-down areas of the market. The same cyclical and foreign assets that experienced the painful manufacturing/production recession of 2023. I thought to myself, “If this inventory correction is a mid-cycle trough, then cyclical assets are probably far too cheap as everyone was convinced of an imminent recession.”
So, that’s what I played. Building Materials, Metals & Mining, Specialty Chemicals, Transportation, Machinery, and even some Financials.
And then I dug some more. The Inflation Reduction Act, the Chips Act, fiscal stimulus was still booming. In fact, what many didn’t realize is that those bills that were passed in 2022 actually had barely begun disbursing the funds.
I thought some more…
If there’s one thing that came out of Covid – I know there were many – it’s probably that supply-chains had become far too globalized. Of course, the Russia/Ukraine War reinforced this idea for Europeans. Well, ever since the global pandemic, the United States, and many other countries around the world, placed an emphasis on “re-shoring.”
I’m going off of memory here, so it’s probably not perfect, but for a while there, it felt like we had some major announcement of a new Semiconductor plant being built in the United States almost every month. Intel, Taiwan Semiconductor, Micron. It was like a CAPEX dream. And, yes, further evidence of the power behind all that fiscal stimulus.
I thought, “Perfect! If all these plants are being built in the US, they’ll need materials! If infrastructure is being repaired, they’ll need materials! If housing is still in short supply, they’ll need to build more. More materials!” So, I upped my exposure to those industries mentioned above.
Turning the page now to early-February 2024 and it hit me. Growth is resilient. The rate of inflation is easing towards the Fed’s goal (there have been hiccups along the way, but the larger trend is still intact). But more importantly, we can see increased signs of a rebound in those leading indicators. Maybe this was just a mid-cycle trough after all!
And that’s when things got a bit silly. As I sat back in my chair, I texted a former colleague and friend. Here’s what I wrote on February 14, 2024…
“Serious question… Are we thinking about this all wrong?
The Fed clearly wants to cut rates this year. Goolsbee said today that they don’t necessarily want or need to wait until we hit 2% exactly. In what world is this not ludicrous that the Fed is no longer acting in a counter-cyclical fashion and actually cutting rates in the face of a strong economy?!?!
The Fed is meant to smooth the business cycle, but today they’re actually talking about juicing the cycle.
Everyone, myself included, laments how the market is ridiculous in getting excited for potential rate cuts. But, in the grand scheme of things – how could it not be wildly bullish for equity prices if you have a strong economy AND monetary easing?!
Historically, and this is what Druckenmiller and PTJ always talked about, a strong economy tends to precede a weak period for equities because it means the Fed will be raising rates/withdrawing liquidity in a counter-cyclical fashion. We’re actually talking about doing the exact opposite…”
I upped my exposure some more…
So, now I sit here in mid-March, and I watch with frustration as Semiconductors continue to rally 5%+ every single day, while cyclical equities rally at the open and get sold off throughout the day.
I don’t know, maybe investors see interest rates falling and they think, “Must be a recession, so sell cyclicals.” Only thing is – Semiconductors are cyclical too, but that’s gone out the window because everyone dares to utter Sir John Templeton’s four most dangerous words in investing, “This time is different.”
Personally, I’m not one to say, “The market is wrong, I know more.” I listen to the market, its internals, and try to decipher the message. But to me, I think this “pop and fade” action in cyclicals is due to some investors reacting to falling yields (“Recession”), while others pull capital from every other area to pile into the momentum chase in Semiconductors.
In the grand scheme of things, the action beneath the surface of the market, especially with respect to cyclicals, is actually very bullish. Despite this “pop and fade” behavior, trends are broadly turning higher, and from very depressed levels. More importantly, flows show little-to-no signs of euphoria for these corners of the market.
So, what do I do? Do I capitulate and chase the Semiconductors? Believe me, it’s entered my mind more than once. But then I remembered an interview that Stan Druckenmiller gave about his experience with the Dot Com Bubble.
Here’s a snippet from an article on The Irrelevant Investor…
When asked about his experience, Druckenmiller said “I bought $6 billion worth of tech stocks, and in six weeks I had lost $3 billion in that one play. You asked me what I learned. I didn’t learn anything. I already knew that I wasn’t supposed to do that. I was just an emotional basket case, and I couldn’t help myself. So maybe I learned not to do it again, but I already knew that.”
The reason I bring this up is because today, we see any number of analogies to bubble periods being thrown around. “Breadth is its narrowest since 1929.” “The ratio of Semiconductors-to-S&P 500 Index just surpassed the Dot Com bubble peak.” “The top 10 stocks in the index exceeds the level of concentration seen during the Dot Com bubble.”
Look, I have no idea if we’re at the start, in the middle, or reaching the end of a bubble. But as I said before, given the overall conditions being presented to us, it’s hard not to appreciate the reality that policy makers are encouraging a bubble.
And in that spirit, here’s the full text of what I wrote in my morning note on March 8, 2024…
Ok, yesterday was short and sweet, but today, I have a few things to get off my chest. For beginners, we have Non-Farm Payrolls at 8:30am, so that’ll be the trigger that sets the tone for today’s session. That’s the short of it. But the more nuanced version comes from what we heard yesterday from both Jerome Powell and President Joe Biden.
I have to admit – I was in and out of paying attention to Powell’s testimony because there was some truly bizarre stuff going on. I mean, Senator Kennedy’s line of questioning was out of this world and, for anyone watching these Q&A sessions, it really makes one question what sort of banana republic our country has become.
But speaking of that banana republic, what I found most interesting with Powell’s testimony yesterday wasn’t these out-of-this-orbit questions. It was actually Powell’s certainty – without using the word certainty – that the Fed is close to cutting interest rates. I mean, this is a Fed that repeatedly hammers away at these ideas of “independence” and “data dependent.” Now, if the Fed was truly data dependent, why in the world would Powell state that, “… we are very close to gaining enough confidence to cut interest rates.”? Is that not him basically saying, “Data be damned, we’re going to cut rates!”?
The reason I ask this is because there are a number of inflation indicators that have witnessed a decent bounce lately. Sure, they’re all within a larger downtrend, but January’s data was pretty strong and should give anyone a bit of pause. Fed’s Mester made her own speech yesterday and even she said that the January report was “a bit of an eye opener.”
And then there were the numerous questions lobbed towards Powell about issues with the housing market. Affordability, lack of entry-level options, etc. He was probably asked at least 20x about issues with the housing market. His response? It’s a structural supply issue. It’s about the “locked in” effects of low fixed rates during the pandemic (people bought at very low rates, so they’re not selling now because they don’t want to lose those rates when buying a new home). It’s zoning and supply-chain constraints. In other words, it’s not the Fed Funds Rate, it’s not the pace of QT that could be adjusted to manipulate the long end of the curve. Simply put – it’s not a demand issue, it’s supply, and thus we basically have to deal with it and build more.
Then there was his response to fiscal deficits. In his own words, “we are nowhere close to the point where fiscal policy dominates monetary policy.” Umm, maybe we’re all living in a different world, but does anyone else question this, given what we’ve seen with economic activity in the face of the most aggressive rate-hiking cycle in decades?
What I came away with Powell’s time with Congress is that they don’t really care about deficits. They also don’t care about housing affordability. And he undeniably stated that the Fed will not wait until inflation hits 2% exactly to cut rates.
Again, I ask, is the Fed really data dependent?
Then there was a subtle, but really telling, development from Lael Brainard, Director of the National Economic Council. According to multiple reports, Brainard pushed to change Biden’s budget forecast by assuming lower interest rates in the future, which paints a rosier picture of borrowing costs at a time of surging deficits and government debt issuance. The interesting part here is that Bloomberg reports that the Director of the NEC is typically not involved in setting these forecasts – those are set by the Treasury Secretary, OMB Director, and Chair of the Council of Economic Advisers. Hmmm…
The reason I find this so interesting is because it was reported on numerous occasions that Lael Brainard (former Fed Governor) was very close to being chosen as a replacement to Jerome Powell as the Chair of the Federal Reserve. For whatever reason, Biden elected to stick with Powell this latest go-round. But at a time when politicians continually press the Fed to lower interest rates; should we be surprised if Brainard is tagged as the next Fed Chair in 2026?
Again, I ask, is the Fed really independent?
Ah, ok, now on to the State of the Union and Biden’s remarks that tie in so nicely with these gems from Powell and Brainard. As Biden proclaimed, “Mortgage rates will be coming down and the Fed acknowledges that.” After a lengthy two days of questioning Powell about housing issues, Biden’s solution is two-fold: a $400/month tax credit over two years for buying a home and a $400/month tax credit over two years for homeowners selling their first home (presumably to encourage people to shed those low fixed-rate mortgages and move up).
Going back to my point about Powell – it’s all about structural supply issues. So, instead of incentivizing the building of more homes, we’re simply going to subsidize demand. Where will that come from? More deficit spending. Do we really think that will fix pricing issues? Do we really think that’ll detract from inflation? We’re not trying to lower housing prices; we’re merely doling out more cash to justify and contribute to those elevated prices.
Alright, alright, I’ve ranted for long enough. As I’ve been saying all along, there’s really no use in complaining about what’s going on. There’s also no use in trying to exercise your intellectual superiority by arguing what the Fed should do. All we need to do is sit back and observe what the Fed is doing. And our politicians.
What we’ve just heard from Fed leadership, arguably the next Fed leadership, and our politicians, is that things are full steam ahead with juicing an already strong economy. We’re going to pour gasoline on the fire that is a cyclical upswing. One that’s already being driven by, you guessed it, more fiscal spending!
What does this imply for our investments? Well, it’s not that Tech, Semis, whatever won’t keep working. I mean, if we’re dead set on blowing a massive asset bubble, why stand in the way of it? But what I continue to hammer away at is, if we are truly going to further subsidize demand – whether that’s through lower rates or direct cash handouts – it’ll become increasingly difficult to return inflation to target. And in that sort of an environment, especially with the Fed blessing it, cyclical equities and commodities look extremely attractive.
Furthermore, it likely implies that bonds won’t give you a whole lot in your portfolio. Sure, you’ll get a nice nominal yield here. But when accounting for inflation, it certainly won’t be anything to write home about. It also limits my outlook for any meaningful capital gains (price appreciation) for long-term bonds. I mean, as it stands, there’s already zero term premium built into the curve, so why own anything far out on the curve if we’re turning on the afterburners of this economy?
I said this a couple of weeks ago, but sounding so bullish truly makes me feel uncomfortable. I came into my professional career in the immediate aftermath of the Global Financial Crisis (I was interning during the GFC). I’ve traded during China’s shock devaluation, the volatility shock of 2018, the Christmas Eve Bloodbath of 2018, the Repo Crisis of 2019, the Covid Crisis of 2020, and the ensuing madness of 2020/2021 before a brutal deflation of that bubble. Skepticism has been ingrained in me since the start, and it’s a bias I need to remind myself of on a daily basis.
But if I step back and am completely honest with myself about not what I think is right/wrong or what policy makers should do, I firmly believe we’re headed full steam ahead into another massive bubble. There’ll be a time when things correct, or the bill for all these reckless policies comes due, but that likely won’t happen until we’ve realized what an epic mistake we’ve made. So, again, as uneasy as it makes me feel, the best thing we can do is abide by what we see happening. And until that changes, I think it’s a dangerous game to go fighting this market simply because “it’s silly, it doesn’t make sense, or it’ll all end in tears.”
That’s my current view of the world, but for anyone who’s known me for more than a couple of weeks, I change my mind A LOT. I might feel a lot differently about an ultra-bullish stance in a week, two weeks, a month, a quarter. Who knows? But as things are, that’s where I stand.
Ok, ok, so why in the world did I mention AI and provide those disclaimers at the outset?
Up until now, everything has been about setting the stage. The stage where there’s rampant euphoria around AI and Semiconductors. A stage where there’s complete panic about increasing the supply of semiconductors because the demand forecasts are actually hitting the moon. But it was also meant to provide a little context that you’re not just reading from some grumpy old perma-bear.
If you haven’t gathered by now, I idolize Stan Druckenmiller. A lot of how I look at the world and how I approach markets is very similar to Stan. Not because I want to be Stan, but because whenever I hear him talk, whenever I read about his philosophy, it fits with my own personality. That’s the most important thing we can do as investors or traders. It’s ok to borrow things from other extremely successful investors. But we’re all our own individuals with our own personality flaws. We all need to adopt our own approach.
Well, to channel Stan one last time, I’d like to share another one of his quotes.
“Chemical stocks, however, behave quite differently. In this industry, the key factor seems to be capacity. The ideal time to buy the chemical stocks is after a lot of capacity has left the industry and there’s a catalyst that you believe will trigger an increase in demand. Conversely, the ideal time to sell these stocks is when there are a lot of announcements for new plants, not when the earnings turn down. The reason for this behavioral pattern is that expansion plans mean that earnings will go down in two to three years, and the stock market tends to anticipate such developments.”
This rhymes so perfectly with what I’ve learned from other portfolio managers throughout my career. Broadly speaking, cyclical stocks are a bit counterintuitive because you want to be buying them when their valuation multiples are high (weak earnings) and selling them when valuations look cheap (strong earnings).
I previously mentioned Specialty Chemicals as one of the industries I find attractive and, it just so happens, their valuations look insanely expensive right now.
Better yet, does anyone remember all throughout the post-Covid recovery and the Russia/Ukraine War when we heard of plants being shut down due to the energy crisis? Anyone remember the inventory/production correction from 2022/2023?
That’s capacity leaving the industry. That’s when earnings collapse and stocks look insanely expensive on a price-to-earnings or enterprise value-to-EBITDA multiple.
But do you also know what’s required for all those building materials that are in such strong demand as new plants are being built here in the US? Or what goes into asphalt, concrete and other materials needed to repair and upgrade our infrastructure? Among other things like copper, lumber and steel, Specialty Chemicals.
Here’s just one quick example of a name within this industry – Huntsman Corporation (HUN).
Huntsman currently trades at a P/E multiple of 50.9x and an EV-to-EBITDA multiple of 15.3x. It traded at a peak P/E multiple of 53.8x in Q3 of 2023, right around where the stock appears to have troughed. You know where valuations stood at its stock price peak back in early-2022? A P/E of 4.9x and an EV-to-EBITDA multiple of 5x. The stock went on to fall ~45% since the valuations looked “cheap.”
I know, sounds crazy, but it’s true…
Do you also know that specialty chemicals are needed in the production of semiconductors?
Hmm, to me, this sounds like an industry that looks extremely unattractive (investors quiver at the multiples), capacity has been withdrawn, and there’s multiple catalysts to increase demand. All that we’ve discussed doesn’t even include Electric Vehicles or the push for renewable energy.
Let’s flip this whole argument on its head and ask, “What industry is seeing a lot of announcements for new plants?”. Dare we mention Semiconductors? According to Z2Data, there are a total of 73 new semiconductor fabs being built around the world. Here’s a list of just a few…
1) Foxconn – India, 2025
2) Foxconn – Malaysia, 2025
3) Guangzhou CanSemi – China, 2025
4) Hua Hong Semiconductor – China, 2026
5) Nanya Technology – Taiwan, 2025
6) Powerchip Semiconductor Manufacturing – Taiwan, 2023
7) Samsung – South Korea, 2023
8) Semiconductor Manufacturing – China, 2024
9) Semiconductor Manufacturing – China, 2025
10) Taiwan Semiconductor – United States, 2024
11) Taiwan Semiconductor – Japan, 2024
12) Tower Semiconductor – India, 2028
13) Micron – United States, 2025
14) Intel – United States, 2024
15) Samsung – United States, 2024
16) Texas Instruments – United States, 2026
Now, the thing to remember with this are the trade restrictions the United States is placing around China due to national security concerns. In other words, that’s a massive detractor from demand (China is attempting to furiously build its own supply), while supply is projected to explode.
Here’s a breakdown of planned projects just here in the United States, according to Industrial Info Resources.
I’m always the first to admit that there are plenty of people out there who are far smarter than me about their respective industries. I’m not a sector specialist and I’m certainly not a deep-in-the-weeds fundamental investor.
Maybe demand will still outstrip supply. Perhaps the AI revolution is only in its infancy, and we can’t even fathom the true level of demand for semiconductors. I honestly have no idea and, if we’re honest with ourselves, even the “expert” analysts at banks can’t make accurate 10-year predictions either.
And that’s why I won’t even bother to make any sort of statement about whether or not the current activity in Semiconductor stocks qualifies as a bubble. Maybe it is. But maybe it’s not.
This is just me, sharing what I’ve learned over the years, and hopefully making some very clear and unemotional observations about how market cycles tend to work.
What does this all mean?
If I were a betting man, and I guess I sort of am considering I run a hedge fund for a living, I’d have to place better odds on something like beaten-down Specialty Chemical stocks outperforming Semiconductors over the coming 12-18 months.
Why 12-18 months? That’s typically the window with which markets tend to discount the future, at least according to academic research.
And over the next 12-18 months, we can see one industry has already had capacity withdrawn with very clear tailwinds to boost demand (infrastructure, housing, new plants, semiconductors). The other has massive capacity being built with wild imaginations of persistently growing demand (right or wrong, it creates a high bar to clear).
Again, this isn’t a call to rush out and short Semiconductor stocks. I’m certainly not doing that, given my overall macro view at the moment. And who knows if I ever will – a correction phase could be dramatic and steep (bubble collapse) or a boring old sideways grind while fundamentals catch up.
Remember, I’m a macro trader with a time horizon of weeks to a few months, so I’m not even advocating for a static approach to this view over 12-18 months. But these longer-term views generally help to inform my opportunity set, so at the very least, I know where I’ll be looking for tactical positions over the coming time period.
If you, like me, are struggling to wrap your head around the current market action, hopefully this high-level thought experiment helps. I know I sure feel better laying this all out in front of me. If there’s one thing I’ve learned, or maybe talked myself into with a clear mind, it’s that I don’t want to be chasing the AI/Semiconductor hype right now.
And see? I promised this wouldn’t be an argument over Artificial Intelligence!!!
Best Wishes,
MC