Transcription
I think the semiconductor cycle will roll over towards the end of the year, early next year, but I think we have a euphoric part of ahead of us before we get to that part. Having said that, a wave of IPOs is always a late cycle signal, and I still think it is very likely that this is the final year of the cycle. My key message going into this year, it can become crazier than what we've already seen, and I think it will be a lot crazier before we turn.
Today's episode is brought to you by the unlimited HFGM global macro ETF, ticker HFGM. Later in the show, you'll hear more about HFGM, currently the number one Morning Star rated fund since its inception in the US macro trading category. Let's get into it. We have a great interview today. I'm Maxi and I'm joined by Andreas Stino Larson, former hedge fund manager and editor of Realvision's macro research. Andreas, thank you for joining us today.
Thanks for having me, Max.
It feels like we are in a macro shift right now. Just a few months ago, we were all talking about how we're going to get a new Fed chair. we were going to get rate cuts. Now inflation has reared its head and the new Fed chair is going to have a hard time pushing rate cuts through and in and the market is even starting to price in hikes. Help us make sense of this big regime change.
Well, just before we started recording here, we had Chris Waller out stating that he's willing to at least abandon the easing bias of the official communication of the Federal Reserve um at the June meeting. So, that's kind of kind of a big deal. He was at least until a few days ago seen as one of Trump's loyal lieutenants within the committee, right? So, it will be kind of a difficult job for Kevin Walsh to keep that easing bias intact uh in in a few weeks from now in June. Uh having said that uh I still consider this inflation wave to be fairly isolated to everything that comes out of the straight of a moose or rather doesn't come out of the straight of the moose at the moment. Um we obviously see a clear wave of inflation related to food and energy. But setting that aside, I don't necessarily think the CPI prints that we've seen over the past couple of months are too worrisome yet. Uh but obviously when you're a central banker uh and you see inflation printing at say 0.6 6 or 7% of the month. You basically have to to take that into account and I don't think any central banker really is willing to say that uh they support interest rate cuts into such an environment. But we also know what the solution is. Max, you could just open that damn straight and then you'll probably find uh the inflation environment to be a lot more benign in a quarter or two from now. So I still think we have time to solve the whole issue. uh and maybe Kevin Walsh will be told by the administration to kick the can down the road a little bit in June uh and hope for the best in in terms of settling a deal in the straight of moose.
Does that mean you think that the the riskreward actually is fading this this selloff in bonds?
Yeah, I I actually think so. To me the most undervalued scenario is that Kevin Walsh will sort of decide even with you know quite a few members probably not supporting it to say that well we'll have to be patient here. Uh the administration is working on a solution to this inflation problem and uh it will be temporary of nature uh in case they settle uh on a deal with Iran and we still hold high hopes around that blah blah blah. Uh so a new version of the transitory narrative of of J Powell uh with maybe slightly better hopes of an actual resolution to the inflation issue here.
Okay. So team transitory potentially coming back around uh does he have the votes for that though?
No, that's the big issue. But you know we're not close to a majority favoring a rate hike. Uh so in that sense, he'll probably have some leeway in terms of communicating how to wait and see. uh be if if you look at what Chris Waller said uh he's basically in favor of scrapping the easing bias but he also said that he's not close to moving to an outright hawkish bias. So in that sense, the majority probably favors a neutral bias for now. And a neutral bias does provide some leeway to the chairman to lean in one or the other direction uh around that neutral uh statement uh allowing Kevin Walsh to sound slightly doubbish versus that.
Okay, I'm taking a look at Fed Watch right now just for current interest rate pricing and I'm seeing just looking at the December meeting. So last meeting of the year, we've got the the highest likelihood outcome is that we see rates 25 basis points higher. That's priced at 42.7% chance right now. 30.3% chance that we stay in the current range and then even a 21.8% that we're 50 basis points higher at the end of the year. So you don't think that scenario is what we're going to see?
No. Um the the difficult question here is to answer the exact timing on when to enter that trade. uh because it is obviously very related to the geopolitical question around the relationship with Iran and uh therefore my best guess is that you have to wait until close to the actual press conference in June to enter this trade also to allow Kevin Wars to sort of give you the backing once you enter it. Uh having said that max rate cuts in this environment I I think the best case is basically that we move sideways in the fed funds. you know a lot of patience given that you have an impact uh in one direction and impact in the other direction. Uh you have this this inflationary force from technology versus a very inflationary force from the energy side right now. Uh which will probably allow you know some of the centrists such as as Kevin Walsh to just wait and see. Uh so I think that's the you know probably the base case for the rest of the year. So, if we do get Kevin Worsh able to get enough people on side for team transitory, we have some of that geopolitical uncertainty um taken care of, but the data obviously, it's lagging, is still high. What do you think the long end is going to do in that scenario? It's it's been one of those times when the supposed bond vigilantes have have come back out. Um do you think that they are going to take that favorably and what do you think the long end is going to do?
So, if I'm right that Kevin Walsh is in team wait and see or team transfer tree, I think we have six% 30-year bond yields coming up. Um, it it's likely that the curve will steepen in such a scenario. Um, we've seen an early glimpse of that say over the past couple of weeks with, you know, a clear sell off in the long end of the curve. And to me, that's not a particularly worrisome scenario for risk assets. I mean, what we've seen especially over the past couple of years uh has basically been a global phenomenon of of steeper yield curves and the long end of the curve that is, you know, basically out of control in many ways. Just look at Japan as an example. I think we've moved up more or less 4 percentage points in the 30-year point since two or three years ago now. Um, but it is to some extent by design and uh given that we don't we no longer have central banks participating in um absorbing the duration via QE in the longer end of the curve, I actually think that they can accept such a move to to a much larger extent than what they could pre-COVID and when you look at the credit creation outside of the central banks or in private banks, the credit creation in private banks actually tends to thrive when the long end of the curve steepens relative to the short end. Uh because, you know, very mechanically, a commercial bank actually likes when the yield curve slopes um because by design, the business model is to borrow short and lend it out longer term. Uh so when you have a Steve Kurt, they can basically write that carry. And I think Japan has been an extraordinary uh empirical case study of that over the past two or three years. Not a lot of people would have told you that NikK would have had its best run in years or even in decades alongside four percentage points higher 30-year bond yields in Japan. But here we are with that kind of Nikkay rally alongside it. uh and that's because of a revival of the whole private system and the credit creation outside of central banks.
To what extent do you think that is due to some of the other um reforms that are happening? There's definitely been a shift towards uh maximizing shareholder value, something that has been sort of the norm, the expectation for management here in the US where you can't just sit on cash forever and do nothing with it. like they're they're really moving towards a more western shareholder management relationship and that has certainly been a factor in the outperformance. Do you think that that's been a bigger factor?
You know, I agree that it's it's a combination of the two. Um it's probably difficult to explain exactly how much has been you can contribute to one factor or the other, but I actually think the two factors go hand go hand in hand. Uh we typically see that kind of um behavior from executives when uh the yield curve steepens. Uh because a steeper yield curve is essentially um a symptom of animal spirits related to everything from investments to capex uh to the overall credit cycle. Uh and when you see symptoms of a stronger credit cycle, you're probably also more tempted uh to showcase that exact kind of behavior as an executive. Uh, so I think the two go hand in hand. Uh, and um, therefore I'm, you know, it's probably more of a 50-50. The two the two actually work in tandem.
So you said it's by design in many ways.
I I I can understand that, but also, weren't they supposed to come in, term out the debt? Everyone was talking about how they had they had issued way too much short-term debt before, and then we're going to come in, we're going to get rates low, and we're going to term out the debt. That seems like it's completely gone out the window. Why do you say that this steeper yield curve is by design?
So to begin with, I perfectly agree with you that this is another U-turn from the administration. Um, one U-turn out of many. But having said that, a steep yield curve is necessary when the yield curve steepens outside of the US. So take this the example of the Japanese yield curve. The Japanese yield curve has steepened a lot. Uh and if you're a Japanese life and pension fund, uh you suddenly have a very different domestic yield curve um to deal with in your bond decision making. When 30-year bond yields are close to 4% in Japan, uh you obviously need higher long-end bond yields in the US for US treasuries to remain attractive for these investors in Japan. Remember, Japan is the biggest foreign investor in western bond markets um period and it remains the case. A steep yield curve in the US also if it steepens even more than in Japan is very very relevant for purchases of US bonds after FX hedging costs. uh and bear with me, it's slightly technical, but uh if you buy a US Treasury as a Japanese life and pension fund and you want to ensure that the exchange rate between the dollar and the yen doesn't break havoc on your uh fixed income investments, um you use an uh FX forward to hedge that US treasury back to Japanese yen. uh and that FX forward uh is basically by design priced as the spread between the short end of the Japanese curve and the US curve. Uh so if you have a very steep US curve relative to the Japanese curve um you make the US treasuries after FX costs slightly more attractive. Uh so that is simply why the US curve needs to steepen when the Japanese curve steepens. Um so you know I I hold Scott Bessant in pretty high esteem and I think he's aware of this this stuff. I mean he's he's been buying fixed income FX hedge before as a fund manager. So he's aware of these dynamics and that's probably a reason why at least a partial reason why he's moved inwards on the curve to allow the curve to steepen simply because it helps foreigners buy the US treasuries.
We had a very weak dollar for a period of time. It hadn't gotten to, you know, horribly weak levels, but it was definitely a story. the dollar was trending lower and that has really reversed um since this has occurred. Do you think that the strong dollar has the potential to hurt these emerging markets and ex US economies that are in many ways feeling the current crisis the most acutely? Is this kicking them while they're down?
It certainly does hold the potential to um to wreck havoc on the EM space overall. The EM space did incredibly well pre Iran. Uh we've seen US assets outperforming many peers since the start of the war. I still think that's, you know, an overwhelming trend that I'd lead into here. Um if you look at the data that I track on a daily basis on the global economy, uh the only exception I can find in the EM space right now is India. India's growth has accelerated probably due to the fact that they got rid of these secondary reciprocal tariffs um earlier this year paired with the fact that they're now able to buy Russian oil again. Uh those two things have kept the Indian growth machine somewhat intact. China is losing pace in my data. Um LAM is is losing pace in my data. So I I you know we're starting to see the ramifications of this dual shock of a supply shock on energy and a um an FX shock to some extent in the EM space. Uh but it's still early days. Uh I I agree with you that a a strong dollar is probably the worst thing that can happen for this EM trade that was otherwise pretty popular. uh and um yeah, in short, I'm very very US and India focused in my portfolio right now uh for the very reasons that uh the US growth seems to hold up substantially better than Pierce uh also as the US is not really on the receiving end of the supply shock and on um on top of that I I see India as the only exception in the EM space where you actually have accelerating growth right now.
I hope you're enjoying today's episode. Global macro investing is in a renaissance. That's because the world is changing faster than ever with higher interest rates, rising global tensions, and AI uncertainty representing just a few of the mounting obstacles for traditional 60/40 portfolios. Uncorrelated global macro strategies can offer a powerful tool to help capitalize on economic shifts. But historically, traditional alternative investments have been an exclusive club where access is limited and so is the information shared on the portfolio holdings. Not anymore. Enter HFGM, the unlimited HFGM global macro ETF. Unlimited uses a proprietary algorithm to analyze public data from the world's largest global macro managers to construct a dynamic portfolio of liquid holdings. The strategy aims to capture those global macro returns while targeting a volatility level designed to make a meaningful impact on your portfolio. HFGM delivers exposure to the broader global macro sector without the burden of typical LP structure headaches. That means no high minimums, no locked up capital, no K1 tax forms, and significantly less fee drag as well. But the biggest advantage is radical transparency. While traditional alternatives hide their strategies behind closed doors and opaque reporting, HFGM lets you look under the hood every single day. How a strategy should work is one thing, but the big question is, does it perform? Well, HFGM is the number one Morning Star rated fund in their US macro trading category since its launch last April. We wish we could go into all the details of that performance right now, but navigating the regulations makes that a challenge in a podcast format. But unlike private global macro funds, you can see for yourself. Don't take our word for it. Go to unlimited ETFs.com/hfgm to see their transparent daily positioning and full performance track record. If you want access to a diversifying strategy created by an experienced CEO at a fraction of the typical 220 cost, check out what Bob Elliot and his team at Unlimited developed. That's unlimited.com/hfgm to learn more. Past performance is not indicative of future results and investment should not be made based solely on returns. Before investing, you should carefully consider the fund's investment objectives, risks, charges, and expenses. This and other information is in the prospectus. Please read the prospectus carefully before you invest, which can be found on unlimited ETFs.com/hfgm distributed by Foresight Fund Services LLC. Thanks for listening. Let's get back to today's interview.
It's interesting because the rupee has been incredibly weak. You had Modi coming out and and saying, "Don't buy gold for weddings. Don't travel abroad." I mean, how do you square the strong data that you're seeing with uh statements from the government that seem to be very concerned about the state of the Indian economy?
Yeah, we we we even saw this export ban on on sugar as well and you could probably read that into the energy context as well. Uh um the strongest signal I get right now is the cargo traffic uh that is outbound from India. They really struggled uh on goods exports during this tariffs regime. I've remembered it was probably, you know, outside of the early days where Trump and she had three-digit tariffs added on on each other's goods, um the Indian reciprocal tariffs regime was probably the most punitive of all uh during that time span where Trump had these secondary tariffs added on on Indian goods due to their relationship with Russia. Uh so from a rate of change perspective, that's really been the game changer for the Indian economy that they're able to reexport again uh as a consequence of tariffs coming coming off and um uh therefore they're probably the main beneficiary of of what happened with the Scottus ruling uh during Q1.
Indeed. But at the same time, India has really struggled. The market perceives them as an Iran loser because of their foreign dependence on energy. At what point do you think that turns around?
My best guess right now is that we have between seven and eight weeks left of decently balanced energy supplies worldwide. I it's it's been a puzzle to many, especially those deep in the weeds of energy markets, why the market has been as balanced uh as it has over the past say four or six weeks. Uh and remember, as we speak, Max, you know, the dated Brent oil is still trending down uh versus what happened in March, which is a surprise to many. If you look at the details under the hood here, the straight of Amoose typically carried 20 million barrels of oil a day. By rerouting via pipelines, facilitating the um supply of Russian oil already on water, etc., you you can probably get halfway there. And then you still need say 8 to 10 million barrels of of oil a day. China is not importing to any extent versus pre-war levels. Uh so they import, yeah, a little bit less than 6 million barrels uh a day compared to pre-war levels. And then the US has managed to fill the void on on the last couple of million of barrels more or less a day. So, we're close to a balanced market, but obviously with an end date because China is using their storage, the US is exporting more than they can on a sustainable basis, etc. Um, looking at the current trends, my best guess is that we have seven or eight weeks left of this cocktail. Uh, after having watched that photo op in Beijing the other week, uh, I'm tempted to say that the US and, uh, the Chinese administration basically agreed on on the status quo. um at least they had a pretty uniform rhetoric around the straight of a moose and and the energy crisis. So I I think there is time left and and I guess the assumption made by the US administration now is that this 6 to 8 week period will be enough to kind of smoke the Iranians out, if you know what I mean. Um, I'm I'm yet to see the kind of movement from the Iranian administration that really makes me upbeat on the deal prospects. Uh, I kind of consider what we saw with this Pakistani suggestion was was it yesterday? I kind of consider that fake news. Um, and I have admittedly popped the champagne a couple of times too early on on the deal prospects in in the Middle East. Um, but there's still time, Max, and I think it, for once, I actually think the consensus is very fair here because if you look at the, for example, the Bank of America survey uh out earlier this week, um, it it basically came to the conclusion that the straight of a moose will turn into a true global crisis if it's closed uh all the way into the second half of the year. So it, if you call me in August again and we haven't seen any progress, I don't think it's a big issue.
As you said, uh LA AM has really struggled. I'm looking at just the returns of so year-to-date India down 10%. EM up 20%. Brazil was leading right, just about a month ago in April. It was up 30% year-to-date and it's come down. It's now actually lagging EM. You would think about Brazil as being an oil exporting country as potentially benefiting from this. Help me make sense of why these EM countries that historically do well when commodity prices are high are not seeing that benefit this time around. Is it just the dollar?
It's the dollar. uh and it's the fact that the energy price momentum is basically rolled over since early April, right? Uh so I actually think they they get the worst of of both worlds right now. U we've seen, you know, fairly restrictive rhetoric from Fed members um in in recent weeks allowing the dollar to reprice while the dated oil prices has faded. uh fertilizer prices have faded, helium gas prices have faded, you know, everything related to that energy crisis has faded. Uh and when you get a strong dollar and fading energy prices at the same time, you could make up a worse cocktail for for Brazil, I guess. Um having said that, I if we if we assume that this turns into a prolonged inflation environment, then I would struggle to find central banks that I would rather you know indirectly support via risk assets than the Brazilian and the Mexican central banks. They know how to deal with that stuff. Uh, we surprisingly, we also saw how for example, the Mexican peso and to some extent the Brazilian market, you know, turned out to be a relative winner during 2022, that whole inflation wave there, right? So against my base case, if if we if we see a prolonged inflation scenario here, uh I'm not too worried about ladam currently. They see the worst of of both worlds basically.
We've been talking about one of the dominant forces in the Iran conflict. There's been incredible correlation between the price of oil and bonds. Now that bonds have reached a level where people are starting to take notice, it's it's making its way into US equity markets.
But on the flip side, you have the the AI trade, right? Like those seem to be the two dominant trades and correlations that are kind of driving everything. um to what extent like like what is the leader right now in terms of pushing the market around? Is it the Iran conflict or is it the AI trade?
So I have a slightly contrarian view on the link between inflation and AI and the whole risk asset rally that we've seen since 1 of April. If you look at the PPIs of the producer price index in the US, it was a very nasty report we got uh for for April driven by three things mainly. Of course, it's related to the straight of a moose, so energy and food um and, you know, everything related to input cost for necessities, but it is also related to AI. If you look at the impact from semis in that PPI report, uh it's actually starting to become a meaningful inflationary impact. And I've lost count of how many times I've heard from some of my colleagues and counterparts over the past one or two months that it's very difficult to remain upbeat on the technology trade if inflation is trending higher. But there's a huge difference between former inflation waves and this inflation wave being that the exact AI trade is a reason is a trigger for the inflation wave this time around. It weren't in former uh in inflation waves. So in short, when the PPI is rising, you need to buy the stuff that that makes the PPIs go up. So energy and AI in this sense. So I have this odd view that you can hold energy and AI in a portfolio and do very well here. um overweight AI versus energy, but still that cocktail can work in tandem because those are the two components driving up costs. When you look at the current trends beneath the hood in that AI trade, um the reason why we see such a contribution the PPI is that we have, you know, we have a scarce supply relative to demand and the pricing power is fully intact. They can pass on, you know, right about everything to their customers. If you look at the availability of GPUs right now, uh, my good friend Warren Pies runs a very nice study on that. I I've made my own proxy of it. We're talking about almost a 100% utilization rate of GPUs right now. So, it's it's almost impossible to find any available compute. Is that an environment where you're going to struggle to pass on your your prices? Not really. Right. So uh I think we should expect this contribution from AI to uh be a persistent um you know contributor to the PPI pressures and then we'll have to see whether energy continues to be so, but I I I don't really see the reason why I should turn negative on semis because of inflation because it's the semis that created the inflation.
Yes, in many ways and even the EM outperformance, right? We've been talking about EM and you know, I like to use Brazil as sort of a better proxy for classic EM because China is its own thing now, but then if you just look at EM ETF, 20% of it is Korea and I think another 20% is Taiwan, which is obviously tied to the AI trade. So all the people, the classic sort of XUS, it's cheaper over there, value investors who are pointing to EM outperformance are kind of ignoring the fact that it is in many ways tied to the AI trade here in the US.
Yes. Um and, you know, and thanks for for adding that caller to this discussion on EM because when I said that uh India is the exception in the EM space, I should obviously have referred to Korea as well. um given that, you know, the economy in Korea is so dependent on the semi exports and I I can I can assure you that when you look beneath the hood of the South Korean export data, uh, you know, a data print that I like to track every month, we've never seen the combination of semis rising more than 200% year-over-year and the export overall export only being up 50% year-over-year. It basically tells you that they're not exporting anything else than semis, right? So, um, more or less. Uh, so in that sense, I I I actually still think we're pretty early in that export recovery from from Korea. It looks wild, uh, but it's only due to the 70s.
Okay. Are you concerned at all about some of the stories you're hearing about people pulling out their retirement accounts to go buy SKH Heinix? We just had Korean equities added here on Interactive Brokers for the first time. So people are getting exposure and, you know, when there is demand for speculation, demand for investment, the thing that tends to cure it is supply. Um, so, you know, at least US buyers are able to go and buy Korean equities for the first time. Does any of that give you caution about Korea?
It does. But let's take a look at SK Unix, right? Even though it looks like an old coin back in 2020, if you look at UNX price action, it's still not particularly expensive from a forward price to price to earnings perspective, right? Same goes for for Micron in the US, which is kind of their peer here, right? So I I um I struggle to, at least if earnings expectations are decently correct, then it's not an expensive stock and therefore, you know, it's it's it's the same discussion I've had with with many of our members over the past year. I've thankfully been in this micron trade for for a long while and still hold it. When you see such a trend, don't count on mean reversion trends on a running basis because then because then you'll miss the train and I think momentum is probably the strongest factor over the past two decades in financial markets at all uh by by all means, right? And this momentum trade is actually backed by earnings. I would be a lot more scared uh just for very very live context here of buying into Regetti Computing and some of these quantum names because that's a momentum trade without any earnings uh and with no prospects of earnings uh as far as I can judge for the next handful of euros or something like that. Um, I don't mind buying momentum as as long as earnings are a part of that momentum. With that being said, the return of stocks that the Trump administration touches, yes, have been pretty good, right? Like if you just said, you know, the only rule in your rules-based thing was if they if they invest, whether it's they loan money, they take a stake, they have the, you know, the golden shares where they didn't take equity in it. I forget what they called it. Um, either way, like I think if you bought that basket in equal weight, you've done pretty good.
Yeah. Hey, absolutely. Um MP Materials was a great trade in the rare earth space uh after they got a package from the US Treasury.
Look at Intel.
Intel, you know, there are so many examples already. Now, so I I guess you're right. Um, but I still think it holds true that there is no whatsoever scope to see an earning cycle in this cycle from from these quantum names. So, it's basically anybody's guess where it goes. I mean, it's very difficult to find any like fundamental equilibrium for a stock that has no no earnings, right? Um, and, you know, it's it's it's one of the things that I find very intriguing. You know, I've also been a part of raising cash for quite a few startups over the time over the years, right? And it's actually incredible how wide an outcome space you can look at from a valuation perspective when the company is not making any money because what do you want to tie tie tie it up against? I mean, there's nothing, you know, fundamental to to base that valuation upon. And that's the same for the console companies right now. It's anybody's guess. Could it be worth trillions? Who knows? Yeah. Well, back to back to the memory stocks though.
Obviously, uh historically they're cyclical, right? Semiconductors historically a cyclical industry. And with cyclicals, they always look the cheapest at the top because that's when they're over earning and they look the most expensive at the bottom when they're actually the cheapest because it's at the trough in earnings. And I think that has been the big push back. And the question is, is it a classic peak? Is this a peak and plateau or do we ride off in into the sunset uh up and to the right?
And so I guess how do you think about making sure that there isn't more supply coming online? Are you tracking um capacity in any way?
Uh so the best way I can answer that question is that you have a couple of spot prices that are worth tracking here. uh there is a DRAM uh spot uh exchange rate in Taiwan. So you basically have a daily price level uh worth tracking for these memory stocks. Uh I think that's worth highlighting for the audience here. Um when it comes to the cyclicality of this trade, uh I'm not personally convinced that it's not a cyclical trade still. Uh so I'm I'm more in favor of you know doing classic business cycle analysis to figure out when to when to enter and when to leave such a trade. Uh and there's nothing pointing to a de acceleration of the business cycle in the US at least right now, which is basically the epicenter of this AI trade. um we're rather seeing, you know, the first signs of the ISM manufacturing, you know, starting to push higher after a a long long uh while in the doldrums, right? So, I I think that is basically the answer to this question that you you need to track the spot price of of memory in Taiwan. Uh you need to track the South Korean data uh month in and month out on on semi-exports because when that rolls over, it's basically time to get out. uh and then you need to track the business cycle uh and try to come up with leading indicators for the business cycle. Those are the three best ways to ensure that you're uh in this semi-trade when it's relevant and out when it's irrelevant. Uh again, I'm not convinced that this trade is no longer cyclical. Uh we're just in an upcycle right now. Well, at this point in the upcycle, we've started to move from the highest quality sort of leaders in the space to the best performing names in the AI trade are actually the uh the quote unquote worst companies. And this is what happens in a commodity bull market, right? The the commodity producers with the worst cost of production have the greatest leverage to a rise in spot price. And what we're seeing is spot prices are up, whether it's compute, whether it's memory. And so the the quality of companies uh that are doing the best h has really flipped. How are you thinking about this shift to uh the the best of the best being the worst of the worst?
You're absolutely right. And it's it's it's probably a sign that we're rather late than super early in the semiconductor cycle. and I would agree with that. Um, if I should give you a best guess right now, I I think the cycle turns either in Q4 or early next year. Um, having said that, uh, we're also starting to see, you know, the pricing power moving towards lower margin semiconductor trades such as those within the power space. um which is probably, you know, another case example of what you just mentioned, that with an overall price inflation in this complex, you're starting to see some of the low margin companies manufacturing MOSFETs, for example, within the power space starting to they're really starting to to accelerate here. Infineon in Germany is a great example. We even have, you know, one of the newcomers or I should rather say one a company that has given birth to again, Wolfspeed in the US, for example, has been probably been the best trade of all semiconductor trades over the past month. I've thankfully been in that, but I I was admittedly also partially lucky. So sure, everything that you describe is correct. Um, I'm I'm not an investor that looks two or three years ahead. So, you know, all I am trying here is to harvest these returns as they're ahead of. Uh, and, as I said to you, my best guess is that this is a sign that will roll over um within the foreseeable future. Um, but not yet.
Yeah, Wolf Speed uh up over 150% on the month and I think it's like over 300% year-to-date. you know, juxtapose that with the the all-time performance for it where it had had really struggled post IPO pretty much until March 30th.
Yes.
Yeah. Um, so what about uh the crypto miners that are now shifting their focus to AI companies? I would put them sort of in that in that bucket of companies that are really benefiting from just the overall shortage where
They they're not optimized for for AI compute, but at the same time they can they can certainly benefit from it. It's really the best performing part of the crypto sector and has nothing to do with crypto.
Yeah. Uh, you know, it it probably all started with with iron, right? And then, you know, all of them moved in that direction post that and I I roughly have 10% of my portfolio allocated to that space. Uh again, I was partially lucky because, you know, I also held them because I thought the crypto cycle would do better. Uh, but having said that, if you look at, I think Nibbius increased prices 29% uh this week, right? So I mean, they still hold tremendous pricing power. Some of these new clouds, um, I don't consider to be quality names to be honest and and and and therefore again, it's a symptom of the same. It's a symptom of, you know, a scarce supply relative to demand in an upcycle. Uh, and those are some, and I perfectly admit to that, those are some of the trades that you really need to get rid of in in a hurry once this rolls over.
All right. Well, let's talk about what has potential to roll it over. So, you mentioned just the business cycle reaching its conclusion. There's obviously more capacity if we start to see people um actually bringing on more capacity in in both memory and chip production. That can do it. But also, there is just good old-fashioned equity supply. So we have a couple trillion of new market cap coming to the public markets in the form of the SpaceX IPO. It's rumored that Open AI might file as soon as today. And then you have Enthropic um which, you know, certainly if the SpaceX IPO goes well and Open AI uh files, I think you're probably going to see Anthropic file in short order. you know, combine, put those all together, we're talking about, you know, maybe over four, maybe approaching $5 trillion dollars in market cap.
In 2021, what cooled off the speculative fervor was issuance of of new equities. Now, those companies were not even close to the to the quality and size in many cases of what we're dealing with here. And, you know, Sarah Brass is a recent IPO that has done very, very well. Um, but there there is something to be said about just is is that enough to overwhelm the the uh animal spirits and and just the rotation, right? As as if you have your portfolio allocated to AI. Let's say you group all of these different investments together and you have 30% allocation in in some shape or form to AI and then suddenly you get pure play exposure in the form of an Anthropic or an Open AI, are you going to sell like, you know, I'm asking you if if you have the opportunity and you're happy with the price of it, would you sell off some of these other AI trades that have done well and and are you worried at all about the rest of the market doing the same?
If you look at the runup to the SpaceX IPO, to begin with, I don't think there's any doubt that the market has been chasing names related to SpaceX in the run-up to this IPO. Um ST Space Mobile is one example of that, right?
And therefore I hold sympathy for this rotation view that you describe that uh a lot of people without access to the private markets, most investors that is, they've had to proxy trade these IPOs ahead. It seems especially this basic IPO has has been well flagged for a while. So I actually think that that whole rotation thesis makes a lot of sense. Is four or five trillion enough to, you know, alter the trajectory of the risk asset market overall? I think we need to see that in in context, right? Goldman actually had a pretty decent study on, you know, the 2026 expected issuance in a historical context. And if you look at it relative to the, you know, overall market cap of everything listed in the US, it looks less overwhelming. Let me put it like that. I can't recall the exact percentage. But it looks substantially less overwhelming than when you talk about trillions, right? We we also need to consider the debasement that has happened over the past couple of decades. Um if you look at the four or five trillion, let's just call it that between friends in relation to the balance sheet capacity expansion that is likely from commercial banks over the same period. I'd argue that commercial banks, just due to the uh supplementary leverage rate of reform implemented 1 of April, will have a capacity of more than a trillion in, you know, safer transactions than risk assets, of course. But if they use that trillion uh to allow for more leverage via repos, etc., then I'm tempted to say that three or four trillion doesn't look overly worrisome. So I've seen worse, let me put it like that. And I don't think it necessarily compares to the worst IPO waves that we've seen in history. But having said that, a wave of IPOs is always a late cycle signal. And I still think it is very likely that this is the final year of the cycle. And this is a symptom of that.
And there are other metrics that you do need to adjust for market cap. I've been seeing the margin debt chart flying around and it's like well, if you think about what the average the overall market cap is, how much is margin debt relative to the overall market cap and as a percentage, are we the same amount levered? You can't just use the nominal value of overall margin debt without comparing it to, you know, how levered are we really?
Exactly. Um and that discussion on on margin debt is actually pretty interesting at the moment because, you know, it looks crazy when you look at margin debt relative to, for example, GDP or some of these relative metrics. Also look uh quite nasty, but the market cap of S&P 500 also also looks incredible uh relative to GDP. The good old Warren Buffett indicator of market cap to GDP, uh, yeah, looks stretched. Having said that, I consider that ratio very outdated in the sense that, you know, he came up with that valuation metric when the Berlin Wall was still in place, right? So I mean, the amount of globalization that we've seen since, especially on the demand side, has been pretty incredible. So why would you measure the Mac 7s against the US economy? The demand for Mac 7 services are not based on the US economy alone. They're also based on the services economy in in Europe, not least, but also elsewhere, right? So you need to take account take into account that these companies have a much more uh much wider geographical footprint than they had. So in in that sense, you you really need to be careful when you look at these metrics from a relative standpoint and I agree with you on the margin debt side. If you look at margin debt uh versus market cap, and especially if you look at the rate of change in that ratio, we're far from the euphoria that we saw in 1999 and ahead of the great financial crisis. So it's been my key message going into this year, it can become crazier than we what we've already seen and I think it will be a lot crazier before we turn. So we talked about Iran, we've talked about AI. I feel like as as somebody who who's constantly interviewing investors, commenting on markets, I feel like those are the two conversations that are happening. You're tracking everything. Are we missing anything? Are there is there anything else driving markets right now?
They are the two forces in the race, but I think that uh some of the traits that are currently underappreciated are linked to byproducts that haven't yet been uh surfaced in financial media outlets. So the interlink between AI and Iran was, you know, pretty up in the air during the early innings of this Iran war. A lot of people suddenly figured out that the straight of a moose was also very important choke point for helium gas used in cooling of of wafers.
for for semiconductors. A lot of people suddenly figured out that fertilizers were uh were were also impacted by the strait of a moose. uh but one thing uh that I find very under reported currently is that when you see demand sucked into AI to the extent uh that we're currently seeing uh you also get an incredible wafer cannibalization on semis that are less high margin such as the semis that you put into your laptop or the semis that you put into um your car and such things. Um, so I I think we'll end up with what I'll uh label a sort of an end-user scarcity during the second half of the year that we haven't really talked about yet. Um, not because, you know, I I typically dislike these alarmist takes on scarcities here and there. Um, also this whole illino discussion on food scarcities upcoming and I think they're, you know, slightly alarmist by nature. um in nature.
But uh when when you look at the incentive structure of semiconductor companies right now, they're obviously moving production lines towards the highest margin products uh all going into the data centers. Um and it basically means that I think the biggest equity trade for the second half of the year uh could be found among companies with a very strong supply chain and the pricing power allowing them to pass on the increase in prices and semis uh to the end consumer. Uh good example of that could be HP uh doing very well as we as we record here also on the back of Lenovo's results. Uh but some of these companies with decent supply chains uh and consumers that can you know swallow a price increase uh in in in many cases enterprises uh I think they're they're up for a fantastic second half of the year. And uh it actually strikes me, you know, amidst this AI and Iran focus that look at everything related to the US consumer max and equity markets. It, you know, it's basically been bombed out over the past couple of months relative to everything else. And I think that's an incredible value trade here. So if you don't, you know, enjoy the bizarre volatility in these semiconductor trades, try and find some names with a with with a very decent end consumer footprint and pricing power. I think they'll do well.
>> What you just described sounds inflationary to me. One of the things that has kept inflation lower than a lot of people say is is uh the hedonic adjustments and the the falling prices of electronics. So one of the sort of just steady consistent deflationary forces technological improvement the lowering costs of the electronic devices that we all use might be going away.
>> Yeah. Um, that is actually an incredibly fair point. Um, so, so why is an iPhone seen as disinflationary in the CPI basket when the nominal price of an iPhone has gone up? It's basically because you get more memory in your iPhone on a running basis, it's because you get a better camera, uh, etc. Uh, would I bet that you get more memory for the same price, uh, in an iPhone in in a year or two from now? No. Uh so you're absolutely right that this this is probably an overlooked angle on inflation um and one that holds the potential to to add to inflation in in pockets of the uh goods category. Uh without a doubt. Um on the flip side, I still struggle to see, you know, the border surfaces inflation in in in this environment. Um especially since, you know, it actually holds true when you look at it um also outside of the US. We're probably seeing the the early innings of a jobs comeback here. Um but nothing major. And uh if anything, I guess this AI trade tells you that inflation will show up in goods and not not in services. Uh given the composition of the inflation basket, I'm not overly worried about seeing goods inflation. Um, you know, it's basically an 80/20 um split or something like that, right? So, uh, you're right that it this environment speaks in favor of goods inflation, but not necessarily services inflation.
>> Obviously, the Fed has a dual mandate at periods in time. It was actually the labor market that was giving the Fed its easing bias. Um, and then that sort of went away. There was concern about weakness in the labor market. It looked to be a little bit of a flash in the pan. Then it started to be, okay, we're making progress on inflation. that progress has stalled. Layoffs are coming. We're we're hearing about big layoffs. Meta just had a huge one. Do you think that the labor market could be where uh Kevin Worsh gets his cover to cut?
>> I think he he'll use the labor market, but more in an indirect way via the, you know, productivity discussion as his excuse to at least remain patient here. Everything that I've heard from Kevin Wars, especially since he got got the nod from from Trump, is that he wants to be the Greenspan of the 2020s, right? I mean, he he's referring to himself as, you know, a tech bro and, you know, a Bitcoin evangelist, uh, a productivity guy, and I'm not necessarily sure he's been that throughout his career. He hasn't. Um but that's at least the picture he's trying to paint uh of himself here in the early innings of his chairmanship. Uh so sure I mean there there's certainly an angle uh to it that is related to the labor market, but more from a productivity standpoint than I at least I'm yet to hear him being really scared of of layoffs. Um and the interesting thing is that exactly as you've referred to I mean when you look at the job market post AI the layoffs are happening in technology not outside of technology which is quite interesting right um uh that that's probably it's been a surprise to me but um in many ways I would consider this industrial robotics and uh, you know, the real-life implementation of AI to be the next big thing. Um, but I'm not sure it's going to happen this cycle. And I mean, the robotics trade has been terrible uh this year. Um, I can only think of one true robotics name with with an actual earnings momentum, which is Symbotic. Uh, you know, they've had a tremendous partnership with Walmart through this, and it's actually an industrial robot in use. Um, quite a few of these, you know, AI-linked robotics plays, they're yet to make any um major progress. I mean uh even Tesla's is kind of suffering from from that lack of momentum. Uh so for it to turn into a big story for the labor market in logistics, in uh warehousing, uh in transportation, etc. You need to see those self-driving cars to a much larger extent than what you already you need to see the robotics uh actually making progress in real life. And we're not there there yet.
How much of of the white-collar workforce though do you think might be affected by this? I think it does make sense actually that tech firms would be the first, right? They're the most tech-forward. They've been implementing this earlier than anybody else. Um, and you know, Anthropic when they launched their Claude for finance, they said outside of tech, finance is the number one adapter and and um and implement AI. And I think that makes sense because they're the people who have to cover this uh for for their investors and for their clients. And so, you know, if you think about it like that, we're getting the layoffs first in tech. Are they coming next in finance? And then sort of what are the follow-on industries that they have forward-looking CEOs who are are implementing this, but they're not quite as far ahead as the tech CEOs. And then, you know, Jamie Diamond was saying he's asking Claude to to put together information on on interest rate swaps for him. You know, that that's the stage we're at in finance.
>> The reason why I was staring at my keyboard here and my mouse is that, you know, I made a big study of this going into this year um hasn't really paid off for me in terms of my investments yet, but um I I agree with the thesis that you laid out on uh on finance being an early adopter of this. um we can actually see it uh in in the net hiring data of graduates uh among some of the big banks. Um we also see it in Europe. So that trend albeit very early is there um especially in in in entry-level jobs if you look at it outside of finance and and tech. Um I was speaking about logistics and warehousing. I was speaking about uh transportation um and you know overall manual work right? If you look at the workforce in logistics and warehousing, we have probably probably a couple of million uh in that workforce uh just in the US and um if you look at it slightly broader in in transportation and and warehousing and I mean broader logistics, we're probably talking six, 7 million people. So if you can replace I don't know 10% of that, it it's it's already meaningful, right? Um, and we're talking about 600, 700,000 people. Uh, in in terms of robo taxis or self-driving cars, um, as far as I can, uh, gauge, we we see roughly 500,000 jobs related to, you know, taxi, shuttle, and chauffeur jobs in the US. Uh, so could, you know, quite a big thing. Uh, if if if robo taxis became a thing. Um I'm I'm yet to you know see a lot of momentum out outside of those whimo tests right so um it's also been you know increasingly clear to me that the robo taxi trade uh has become a negative surprise for the Tesla shareholders. Uh I'm a small Tesla shareholder myself but it it's not a big position of mine and um I've actually moved my attention uh towards some of the lead names. Uh so you know companies working with that radar technology that Whimo is using as well or at least a similar technology um because it seems like there's more momentum behind that and actually getting some cars on the street uh which is basically what you want to see uh as an investor in AI trends becoming uh a thing in real life. uh so you know overall I think that you know the process is as follows you'll see it in technology, finance is an early adopter, legal is probably another early adopter, you'll see that very shortly, and then um we'll come to the real world people actually driving around uh in trucks and cars and um, you know, doing plumbing and whatever um, you know, they're going to be hit the last here. Uh, and from a societal perspective, that's a pretty interesting line of events is that I mean, we're we're we're going to challenge the the desktop workers before we're going to challenge those um, yeah, working in the field.
>> It's funny you said legal. I I played tennis this morning with a with a lawyer who was talking to me about Harvey AI, which is the AI tool specifically for legal, and he was like, "Yeah, there are things that we can do in 10 minutes that you used to send a couple of associates and and paralegals, and it would take them a week to do."
>> Um, and you know, obviously, uh, I think we all heard Ken Griffin's comments about how depressed that made him. I I have been thinking about the the way that it it changes human behavior somewhat especially for younger people. You know, we obviously there's the there's the actual squeeze that inflation causes, but a lot of uh monetary policy makers have talked about that inflation expectations, the way it gets embedded into the psyche of society is is equally important in the way that that we start to behave and act and make purchasing decisions and plans. And I think that AI and concerns around it is going to have similar effects that the way the frameworks that we're that we're using to to determine the way people are going to act might break down.
>> Do you are you concerned at all and Kevin Worsh has talked about this in relation to monetary policy that like the frameworks need to adapt? Like in what ways do you think these changes require investors to to rethink their their frameworks and their priors for for what data means for the the way human beings make decisions.
>> So to begin with, allow me to share a bit of anecdotal evidence from my own little research shop that I u formed in in 2021 as far as I remember and sold last year. Um, you know, at peak, we were 15 FTEs. uh when we started implementing AI in our workflow, uh it quite quickly became clear to me that, you know, first of all, our research offering had to change um because we frankly spend a lot of time during the training day um manually creating statistics for for example when you had this um sudden closure of the strait of a moose before AI, it actually took a little while to calculate okay, uh China receives this amount of barrels, uh India's um is also a big um recipient of Iranian oil and so on and so forth. Those calculations actually took a couple of hours. We put that on chats, we discussed with clients what that meant. Uh now everyone's just asking Claude or ever, uh, so we're on the same page there. So instead of us spending time on that, we we only spend time on on discussing the bottom line, right? We we didn't to spend time on discussing the process. Uh, and it has actually annoyed me a lot as a researcher that no one is willing to discuss the process of getting from A to B in an analysis anymore because they just ask an LLM and then okay, here's here's the bottom line of something uh and take that for granted. Um, the best example I can give you is is uh after the, you know, the little excursion to Venezuela uh that the US Navy had in the early innings of the year. Uh all of a sudden I saw at least a handful of pretty respected pundits uh writing research papers around the massive cobalt um storages in Venezuela. Um, and they also refer to the magnificent um oil reserves down there. But first and foremost, where where did this this claim of cobalt reserve come from? Um, when I challenged ChatGPT and Gemini and Claude at the time, um, it came up with a reference to an old Reddit thread around, uh, old claims from Hukushave that, uh, Venezuel.
>> Yeah. Um when you look at uh, you know, the the minerals report from the minerals resource um agency in the US uh and their, you know, table of known resources of metals in Venezuela. There is not even anything mentioned about cobalt. Um, and this was a very good example of respected pundits jumping to the conclusion without discussing the process of getting there on something as um relevant as the natural resources of a country that the US suddenly seized more or less, right? Uh, the same was true for the whole discussion on oil reserves in Venezuela. Initially, I think JP Morgan wrote that it is the biggest oil reserve in the world. I know that's been claimed by many for years. Uh, but it was sort of amplified by everyone asking ChatGPT whether that was true and it said yes. Um, then I took one step back and said, okay, where does this claim come from? Um, it basically comes from OPEC. Then you ask OPEC, where where do you know the Venezuelan reserves from? They refer to the national oil uh company of Venezuela. And when you ask them where they have this claim from, they refer to the old administration.
>> Yeah. So uh, you know, I don't know whether they have the biggest oil reserves in the world, but I'm highly doubtful. uh and uh I think AI amplifies this or, you know, worsens this kind of behavior from humans where you don't really, you don't question the process. You only discuss the the bottom line. Uh, so what do I what do I do now that I know that the US is in charge or sees the biggest oil reserve in the world? Well, in my opinion, what you do is that you challenge that the process of getting to that conclusion before you trade that conclusion. Uh, and that's been annoying me a lot this year that we've seen human behavior shifting towards discussing process to only discussing bottom lines.
>> The supposed facts.
>> Yes. The supposed facts. Yeah.
>> Well, but and then that that becomes a question about what is the the proper strategy. You know, maybe a classical value style investor would say, "Okay, well, I know what the truth is. I know what it's worth." and that will eventually win out the the weighing machine of the market. Um, but in the short term you have the voting machine of the market and if everyone is getting the same data and they're accepting it at face value. Um, you know, as an investor, do you do you think that going through doing the analysis, finding the real number, not trusting the AI number is going to be edge or or a hindrance to outperformance?
It's edge if you combine it with a momentum filter because essentially everything that you described there is going to amplify herd behavior and the momentum factor in equity markets. Uh, I think it's pretty safe to say that m the momentum factor will be even stronger due to AI. Um, so if you're running a momentum strategy, you'll do well. Uh, but if you're running a very, very patient value strategy, you can also do well, but it requires an investor base that is willing to take incredible volatility to get there. That's probably my bottom line.
>> All right. Well, I want to ask you a few questions about the bottom line. Uh, maybe get some good sound bites from you here, Andrea. So, where do you think the the 10-year is headed for the rest of the year? Are we going to see 5% on the 10-year?
>> Uh, yes. Plus 5% is very likely.
>> All right. You said Q4 maybe early next year, the end of the semiconductor trade. Is that still your call?
>> I I think the semiconductor cycle will roll over towards the end of the year, early next year, but I think we have a euphoric part ahead of us before we get to that part. So I count on my semiconductor stock toppling before we get there.
>> Okay. What are the other aspects of the AI trade that you think the market is underappreciating the most?
The market is not on top of the scarcity that the current AI trade creates in in consumer products such as desktop computers, um iPhones, etc. So the whole wafer cat can can cannibalization that leads all the production lines to create or manufacture chips for the uh for the AI supply chain will lead to scarcities within the consumer supply chain and that's a super good trade. Find companies with a strong supply chain within that space.
>> What is the macro risk you think that is the most underappreciated right now?
The most underappreciated macro risk right now is if the strait of Amuse is still closed uh when we get to August because then this crisis will start to bite. Uh and, you know, the current setup where China is on a buyer strike and the US is exporting more than it's capable of long-term um is certainly not um one that is without an end date and that end date is probably already in your leanings of the second half of the year.
>> All right, Andreas. Well, we will leave it right there. Thank you so much for joining us. Where can people find you these days?
>> Uh at Realvision. So I run an equity portfolio in there outside of editing the macro stuff. Uh then I also run an outcasting company called Nowcast IQ. Uh if you're interested in nowcasting data from all over the globe, that's your go-to.
>> All right. Well, hope to do it again soon. Thank you so much.
>> Thanks for having me, Max.
>> Thanks for tuning in. Looking into HFGM by Unlimited? Head to unlimitedfs.com/hfgm to learn more. That's unlimitedfs.com/hfgm. Until next time.