Signs Of 2001 Bubble Crash Repeat Warns Economist | Peter Berezin
Summary
Gold: Structural bull market supported by central bank buying, unsustainable fiscal policy, and rising retail interest, though the guest warns it will eventually end badly.
AI: Earnings may be inflated by temporary excess demand for compute; a potential "metaverse moment" could flip sentiment and trigger a sharp unwind.
Hyperscalers: Massive capex needs and uncertain monetization raise downside risk; combined revenues of leaders may not justify $1T+ annual capex without a true AI takeoff.
Key Companies: Caution around NVDA, MSFT, META, and AMZN as AI/cloud demand, chip write-offs, and circular financing dynamics could pressure future earnings.
Semiconductors: GPU supply growth and rising competition may compress pricing and margins, challenging the sustainability of current valuations.
US Treasuries: Deficit dynamics create the risk of a bond market "revolt," though medium-term inflation expectations remain anchored for now.
Defensive Tilt: Favours Health Care and looks to Emerging Markets for value and fundamentals, while remaining tactically neutral but preparing to turn more defensive on rising layoffs.
Transcript
We're still kind of in the middle innings of a structural bull market in gold. It'll end badly. They always do. I don't think it's ended already. US tech was a great trade this year and past years. I'm skeptical that it'll be a great trade in 2026. If that happens, it'll be time to run for the hills, but it hasn't happened yet. >> The day ahead of the FOMC announcement, what are your expectations? And importantly, uh what is the guidance from the Fed that you're looking for? that's important for you. Peter Beren joins us once more. He's a director of research and the chief global strategist at BCA research. We'll be getting his views on the economy and uh equities markets. Welcome back to the show, Peter. Good to see you again. >> Good to be back on. >> One of the um we're speaking towards the end of October, so it's well into Q4. One of the um hallmarks of this year is marked by unprecedented prices in stocks, Bitcoin, gold, precious metals. And uh this comes in a backdrop of trade uncertainty and uh a lot of volatility with geopolitics. How would you characterize uh this year's performance in the stock market particularly when it comes to um this heightened sense of uh euphoria uh that some economists on my show have said is contradicting somewhat to the real economy. I'll let you comment on that. Yeah, I've been using the metaphor of walking on a tight rope. So, as long as you're still balanced, everything looks fine. Uh the economy is able to avoid a recession, it's able to avoid uh higher inflation, but it's kind of hard to stay there. Probably at some point you are going to fall either in direction of overheating or to rapid cooling and rising unemployment. That's sort of my worry that a lot of these risks are there. They may not be super visible now, but if they materialize given where valuations are, stocks can fall a lot. >> You had um you had a slightly more bearish outlook on the S&P. What why do you think the growth of the stock market exceeded your initial expectations? What were the uh wild cards this year? I >> I think there are a number of uh factors. one, Trump did uh change course on the trade war and so the sort of tit fortat retaliation that we could have seen uh starting in April never occurred. Uh fiscal policy has remained very very stimulative. There was an open question as to whether the bond market will kind of riot over the prospect of these large unfunded deficits. It never did. you know, bond yields are now slightly below 4% in the in the 10-year. And finally, perhaps most importantly, the whole AI trade that supported capital spending. And perhaps more more importantly, it supported consumption through the wealth effect. You know, we've had since August declining real disposable income in the US and yet consumer spending has remained pretty robust as people dig into those savings feeling that they have the wealth to do so. >> Uh just on the bond markets, you brought that up. We're now at $38 trillion of debt. It was announced last week. Uh the deficit continues to rise. At what point can we see some sort of revolt uh in the bond market in either in the form of much higher yields, much higher spreads, maybe a failed auction in the treasury market at some point or does none of this matter really, Peter? >> No, it matters, but it's very difficult to predict when such a revolt will occur. It's a bit like asking, you know, when will the avalanche be? You're looking at a mountain side with a lot of snow. You know that it's going to happen eventually if [laughter] snow continues to fall, but you don't know when. All you know is that there's fragility and there's risk. >> So, let's just pull up one of your posts on LinkedIn. Uh, I encourage people to check out Peter's LinkedIn, by the way. You post regular updates on your thoughts. The bursting of the internet bubble in 2000 was preceded by a manic boom bust episode in the most speculative tech stocks is a recent selloff in nuclear rare earth and quantum stocks sending a similar red flag. And here you have two charts uh from BCA. So you've got the first uh first chart is here uh from 1996 to 2004 you've got internet stocks um and then uh in their performance during that time and then now you have the current period uh you've got quantum computing rare earth and nuclear. My my first my first question is why are you putting in rare earths and nuclear uh in this analysis? Well, because it's part of a broader speculative uh theme. Investors have seen all these huge gains in large cap mega tech stocks. Uh they figure to themselves that okay, I'm not going to make that much more money in Nvidia. It's already a $4 trillion plus company. So, I'm going to go down the quality ladder. I'm going to go down the market cap ladder and find something small, find something speculative to buy. And that sort of attitude often does prevail towards the end of bull markets. We saw that sort of boom bust episode at the tail end of the dotcom bubble. We saw something again not too long ago with some of these rare earth nuclear uh quantum computing uh stocks. Is that a red flag? I mean it might be. It might be. I don't know for sure. What I am telling my clients is before you get really defensive, you know, wait for what I call a metaverse moment, an occasion where some big cap tech company comes out and says we're going to spend a bazillion dollars on GPUs and data centers and its stock price rather than going up goes down. If that happens, it'll be time to run for the hills. But it hasn't happened yet. the narrative that AI is in a bubble of AI stocks rather is in a bubble. Um that's been that's been talked about for months if not years. How do you make that kind of evaluation? How do you definitively say well a P of XYZ uh means it's a bubble versus just you know expensive or fairly valued relative to peers? Um you know at one point does something in the tech space become a bubble? >> Well I think you can have a bubble in valuations uh but you can also have a bubble in earnings. So think about banks in 2006 2007 you know they were trading at pretty low P ratios sort of in the in the teens they didn't look that expensive based on P ratio but they were actually in a bubble because those earnings were massively overinflated by the uh housing uh bubble so I think the risk to investors is that the earnings that companies such as Nvidia and Microsoft uh Meta are reporting are exagger exaggerated by the fact that we have this unprecedented excess demand for compute. Like if you're Nvidia selling chips, you can make a lot of money. If you're Microsoft or AWS renting out cloud computing, you can charge very high fees. But the capex is ongoing. The supply is increasing and eventually we're going to see more competition and prices will come down as will earnings. So those P ratios may not be giving you a good signal of how expensive or cheap the stocks are. >> Uh and how comparable is this period right now to 1999 for example? >> It's in the same ball ballpark for sure. I mean if you look at the concentration of the stock market in tech it's comparable to what we saw in 1999. the P ratios are not as crazy but as I said that's partly because the earnings today may be greatly exaggerated by kind of cyclical factors if you look at price to sales ratios for example the S&P 500 is trading something like 50 times more 50 50% more expensive than where it was at the peak in 2000 and so from that perspective valuations are extremely stretched and unfortunately you know We all know the saying, stocks take the uh stairs up and the elevator down. Uh we could be getting close to one of these wy coyote moments where the stock market looks down, panics and follows a lot. >> This uh Reuters article uh summarized the situation as the following. A widely cited MIT study earlier this year found that of the more than 300 AI projects analyzed, only about 5% delivered measurable gains. Uh the four tech giants and other major cloud firms are together spending uh roughly $400 billion on AI infrastructure this year, but returns for businesses adopting the technology remain uncertain. Potentially, we'll see Peter returns come later. Uh as as an investor or strategist, you're looking at maybe data like this and other data points. Uh what do you what do you do with that? Do you do you say to yourself, well, you know, we're we're not potentially seeing returns right away. They're spending a lot on capex. In fact, the tech companies are spending more on capex than any other S&P sector. Uh and and potentially that's good uh for valuations now or do you look at this as kind of a an opportunity that is a bit overhyped here? >> So this is like the you know trillion dollar question. We don't know what sort of productivity gains AI will uh deliver. There was a study from Metr. arts, a think tank that studies AI capabilities just out a month ago or so. What they did in this study was that they uh looked at the extent to which veteran software developers benefited from the use of AI in their coding. What they found was that access to AI actually reduced their productivity by about 19 uh%. And so I think the jury is still out on how seismic AI will be. It could be the case that AI ends up being very useful in automating kind of very uh standard generic type of work but isn't very useful at sort of pushing out the envelope in terms of the knowledge that humanity has access to. In which case, you know, it'll have implications for our productivity growth. But will it justify the sort of investments we're seeing now? Probably not. >> Now, uh there's been this speculation that a lot of the valuations um have been boosted by tech companies investing in each other. Just today, for example, uh Microsoft is is investing um uh well, they're restructuring their investment. So now they're following the recapitalization. Microsoft holds an investment in Open AI group PBC valued at approximately $135 billion representing roughly 27% stake in the company. Uh Nvidia today uh takes a $1 billion stake in Nokia and now the uh company is up 26%. There's a lot of this what some people call on the internet um this this circular financing going on uh and uh it appears to be working for investors. Um how sustainable is this? And if is that really the reason for why valuations are high? >> Yeah. Don't don't call it a Ponzi scheme. It's a trapezoidal it's a trapezoidal AI opportunity. That's what it is. Uh yeah, I mean there's a lot of kind of uh monkey business going on here. You have to kind of ask yourself like why does Meta not want to put the debt for these data centers on its balance sheet? Why is it moving it to special purpose vehicles? So why is Microsoft uh keen to rent chips as opposed to buy them? You know, probably because there's some worry as to whether those chips will end up being all that valuable. Right now, you have a situation where Nvidia makes a sale, books a big profit, Microsoft or AWS or Meta, whoever buys the chip, but it's not a cost for them. It's capex, right? Gets written off over time. But if it doesn't prove to be all that useful, those chips, they'll get rid written off quite quickly and those companies will suffer losses. I think that's sort of the risk. >> All right. At this point, uh I'm going to ask you for maybe your revised S&P outlook if you have one and overall your tactical asset allocation um as of the end of October. >> So tactically, I'm neutral. Uh over a 12-month horizon, I would be more uh defensive, but tactically like over a 3-month horizon, I would be neutral. But I would be looking at a checklist of things to get more defensive. So what do I want to see? I mentioned that metaverse moment as a potential catalyst for the end of the EI trade. But on the economic side, I'd like to see, not I'd like to see, I would need to see uh an increase in layoffs before I turn more bearish on the economy. Right now, we have kind of this two-tier labor market. You got the core of the labor market, which is fine. If you have a job, you're not too worried because layoffs are low. But if you're on the periphery, if you're like a new entrant, a young student looking for work for the first time, if you're a minority, for example, things are much more difficult. If layoffs rise, then the core of the labor market will start to break apart and then you're looking at a recession, but we haven't seen it yet. So, we'll be looking for that to get much more defensive. Well, Amazon just announced today or this week that rather that they're cutting 14,000 jobs um at the corporate level. Uh and some of it may be due to automation. They want to make the company a lot leaner. They want to they want to call themselves the world's biggest tech startup uh or operate that way. I I don't know if this is the u sort of a start of of more job uncertainty in the tech sector or overall maybe this is company specific. Are you seeing signs on the periphery of more layoffs to come, Peter? Not a lot. Not a lot. I mean, you can look at initial unemployment claims and we don't have the federal government data, but the states still release data and you can sort of compile compile all of that data. It doesn't really show much of a pickup. So, you have to be careful with like headlines from Amazon. Amazon overhunded during the pandemic. And realistically, it's much more convenient for a company to say that they're laying people off because of AI than because of performance issues. performance of shoes, you know, you get sued for wrongful dismissal. So, I would take some of these uh headlines with a grain of salt. >> Uh we don't have government data on on uh more official government data on employment right now. So, I'm just curious, what are you as an economist doing to gauge uh the performance of the labor market? >> Well, it's it's hard. It's much it's much harder to do so. I mean, we do have some alternatives from the ADP. They just released some data uh this morning showing fairly sluggish job growth uh through to the uh through to October 11th if if I recall. Um so the numbers on the job market don't look particularly strong but they don't look recessionary either. I don't buy this argument that the weakness is entirely due to supply. If it was just sort of slower labor force growth that was driving weak employment growth, you would expect to see those sectors where labor shortages are most acute, like healthcare, seeing the least job gains. In fact, they're seeing the largest job gains. So, this is partly a demand story, but it hasn't gotten to the point where layoffs have increased. Until that happens, u you're not going to see a recession. when we do get uh more of a slowdown or perhaps even a recession if it eventually comes, how low will the Fed funds rate get? Can we see a return to uh zero? It's possible. I mean, I don't think that would be my base case, but uh I do think that the next recession could resemble the one in 2001 where the recession was more the uh consequence of a stock market crash than the cause of a stock market crash. Usually the economy goes down and stocks fall, but sometimes the opposite can happen if the AI trade starts to fizzle. As I said, that's been really one of the few things that has been supporting the economy. The economy will weaken and that will sort of feed on itself. In that sort of a context, the Fed will cut rates pretty aggressively. That's what they did in 2001. To zero, probably not, but to 2% possibly. Of course, what's different now is that in 2001 when the Fed cut rates that created a new bubble which sort of rescued the economy temporarily, the housing bubble. uh we might not get a second bubble this time around in which case the downturn could be somewhat deeper. >> Scott Bent named his picks for the next Fed chair. Um most of them I believe are uh very doubbish. Correct me if I'm wrong. Um is the market just essentially pricing in that the Fed will cut aggressively despite uh or regardless of where inflation is headed from here? Peter, >> well dovish or obsequious? I mean, it's it's a fine line, I suppose. Um, you know, I I think right now it's okay to be doubbish because the economy isn't in great health. Certainly, the labor market looks very very uh vulnerable. I think the problem would be is if the Fed were to cut rates against the backdrop where the economy is starting to heat up, inflation is rising. That could cause real problems for the uh bond market. I don't think it's going to happen though. Like, you know, I I worked in Washington for a very long time. Like, the Fed is like the Borg. Once you're there, you sort of get uh absorbed by the whole ideology, the kind of evidence-based approach to monetary policym. You can't really be a renegade in that sort of institution. Uh but over time, I suppose if you keep on replacing uh people with political lackey, you can change the the dynamics. But it'll take several years for that to happen. It's not going to happen like over the next 6 months. >> What what's your and your uh firm's outlook on inflation? And the second part of the question is at what upper band of inflation will the Fed actually start to pause their pivot? >> Well, I think they would get worried if inflation expectations were to increase meaningfully right now. If you look at say the one-year CPI swap, like it's trading at a bit over 3%. Which is not where the Fed wants it. But if you look at the one-year forward CPI swaps or the expectation of where inflation will be, not over the next 12 months, but over the subsequent 12 months, it gets closer to 2%. So longerterm inflation expectations look pretty well anchored. Uh, as long as that remains the case, the Fed could cut rates, but if they were to rise, if they were to rise, then you're going to have real problems, not just for the Fed, but for the bond market and for the broader broader broader economy. Do you believe that the effects of tariffs are backloaded in the sense that in the first half of the year maybe economic growth didn't contract as much as maybe anticipated and perhaps inflation data, official inflation data didn't reflect it yet, but it's going to come much later this year or perhaps into spring next year? Well, the data is very clear that companies are now starting to pass on tariffs to customers and that in turn will erode real income. So, that's a negative for the economy. On the flip side, you can argue maybe not too plausibly, but at least you could argue that trade uncertainty has fallen significantly. Uh, and that should give companies sort of the green light to invest more. I'm a bit skeptical of that argument because for a firm to, you know, build a factory that's only going to be profitable if those tariffs remain in place, you got to be pretty certain the tariffs will remain in place. And given the possibility that the Supreme Court could rule against the tariffs, given the possibility that Trump could just strike a deal that reduces tariffs, you're not going to see too many people wanting to make multi-year investment commitments. >> Let's talk about precious metals. There's a lot of speculation as to what this rise in gold is signing. You even posted yourself on your views on the recent search. This is what you had to say 3 days ago on LinkedIn. Despite the surge in gold prices this year, the amount of gold held in US ETFs is still below where it was in 2022, the tactical where my tactical quant model, which brings together dozens of leading indicators for gold prices, remains bullish. Here you have a price of gold versus ETF holdings. Uh, as you can see, it's more or less parabolic. I'm talking about the price. Um, and this is your quant model. Maybe explain how this model works and ultimately why you're still bullish at $4,000. I mean, gold has fallen somewhat, like 12, 11, 12% since its highs. Uh but it's still $4,000 right now. >> Yeah. I mean, the model's been bullish for a while and, you know, it's done really well uh for for our readers. Um it's still bullish because kind of the medium-term trend remains to the upside even if the short-term trend has deteriorated. Now we have kind of this daily model as well for gold which did turn a little bit bearish but kind of the threemonth model is bullish because it senses that the underlying trend is still to the upside that you have these fundamental tailwinds for gold. The fact that central banks are uh buying gold in large quantities. the fact that fiscal policy remains unsustainable across much of the uh developed world and the fact that retail interest is only now perking up. You know what that chart that you just showed illustrates is that the amount of gold in US listed ETFs is still below where it was in 2022. So if you look at the amount of gold in like GLD, it hasn't even returned to where it was back then. Typically, the end of a bull market occurs when you have this massive wave of retail participation. That's happening now, but we haven't reached a crescendo. The um I'm going to pull up this chart here. This is the tip ETF tracks the Treasury inflation protection security. Um kind of a proxy for what you were saying earlier, the one-year forward CPI inflation rate, but I'm going to overlay this with gold. Historically, they've held a close correlation um or relatively close. I'm going to zoom in to the last couple years now. Uh and you can see that um the past couple years they've kind of been moving together. Um and if you keep scrolling to this here, that's when that's when things start to break apart. If you if you just take a if you just zoom out a little bit here, um they have moved up in the same direction, but gold has way overshot uh the growth. There we go. That's a little bit clear. Gold has way overshot the growth of uh of tips. And um to your earlier point, inflation expectations are not rising dramatically. They've gone up a little bit. Uh but it doesn't seem like gold is following that this year. So what's what's what's going on here? >> Well, the way bull markets usually work is that you begin with a very sensible narrative and then investors extrapolate. They get too far ahead of themselves. the prices move so much that that narrative is more than fully discounted. So I think with gold it'll be that same story. It'll all end in tears. The question is one of timing. Is that going to happen this year? Has it happened already? Or is it going to happen further down the road? My guess is that we're still kind of in the middle innings of a structural bull market in gold. It'll end badly. They always do. Um but I don't think it's ended already. >> The uh the amount of gold held in foreign reserves is now exceeded uh according to some sources exceeded uh the amount of treasuries held by said foreign reserves. Uh for the first time in I believe decades. Uh now Peter what what is your view on the US dollar uh especially given this particular trend that foreign central banks and uh foreign reserves are now accumulating more gold and dumping dollars presumably. What what is the signal for the for the future of the dollar? >> Well I think I mean for the most part most central banks hold a lot more dollars. They might be not treasury bonds might be treasury bills or bank accounts or things like that and they hold gold. So, so this is not something that is close to uh ex is exhaustion. What I think is true is that a lot of central banks are looking for something to hold other than dollars. They saw the experience of Russia here. I'm talking about kind of em central banks. They don't want to experience that themselves. And so, gold just becomes kind of a natural alternative asset for them to hold. They're not going to hold crypto because crypto is a threat to fiat currencies which they themselves administer. Holding gold is a lot more innocuous and I think that's sort of where they're heading and that trend isn't going to end anytime soon. >> Uh Peter, let's finish off then on your allocation to other assets and uh sectors other sectors that you do like. Um I'll let you comment on that, but I have some specific sectors I'd like you to address. So, people have been talking about nuclear power as the future, and Trump has made nuclear um power a little more accessible to to industry. He's um he's deregulated some of the mining restrictions around uranium. Uh what's your take on nuclear? I >> mean, it's it's it's it's it's a good technology in the current age, but it's also one that is not undiscovered. People in the investment community know the story and they know the story well and some of the stocks that are levered to this uh theme are very expensive. Some of them don't even have any revenue yet. Uh I think that trade has uh gone past this expiration date. >> What what's the investment thesis behind hyperscalers? Are we are we investing in real estate um in urban centers? Are we investing in cloud services uh big tech names? What's the play here? Well, you're investing in the AI dream. You're investing in the hope that uh these companies will be able to monetize the investments that they've undertaken. But the bar for that is just really really high. Uh probably by the end of this decade, at least based on current analyst consensus estimates, you're going to have close to like a mill like close to a trillion dollars in annual capex. Uh now to profitably monetize that you're going to need something like $2 trillion in annual uh revenue uh that's going to be really hard to generate. I mean the combined sales if you look at sort of trailing 12 month sales of Microsoft, Google, AWS, Oracle, Meta, it only adds up to about a trillion. Um so where are we going to get two trillion in annual sales from? Unless AI really takes off, uh, it's just not going to happen. In which case, those stocks are going to go down. That's the risk for investors. That's the risk for the economy. >> Any other sectors that you do like or assets? Well, I think right now, as I said, you know, I'm modestly defensive, not wildly so, but within that kind of defensive uh uh theme, I would be looking to, you know, buy more consumer staples, buy more healthcare, like those sectors haven't done particularly well this year. They're relatively cheap. I would also be looking to look abroad. You know, emerging markets are coming up from behind. Uh they're cheaper. They've got pretty good valuations in many cases. They got pretty good economic fundamentals. Uh so US tech was a great trade this year and past years. I'm skeptical that it'll be a great trade in 2026. >> Uh finally, Peter, so we're speaking ahead, right? The day ahead of the FOMC announcement. What are your expectations and importantly uh what is the guidance from the Fed that you're looking for that's important for you? >> [gasps] >> I mean, I think the Fed will sort of continue to sound as though they're in an easing mode, but they're in no hurry to ease aggressively. They're going to talk about how they're focused on the labor market, how the recent data hasn't been abysmal, but hasn't really been very supportive either. They're going to talk about inflation, uh, what's happening to actual inflation, inflation expectations. That I think is a concern for them. Uh and so they're in kind of this phase where they're cutting but not particularly aggressively. >> Very good. Thank you very much, Peter. Uh we'll follow up with you next time. Appreciate your time today. Where can we uh learn more about you and uh read your work. >> You can go on LinkedIn as you mentioned, but you know, I also have a following on Twitter X, whatever you want to call it these days. So you can go there and of course visit the VCA website and sign up for for a trial if you'd like. We'll put all those links down below. Follow BCA, follow Peter, and uh we'll speak next time. Appreciate it, Peter. Thank you. >> Thank you. >> Thank you for watching. Don't forget to like, subscribe.
Signs Of 2001 Bubble Crash Repeat Warns Economist | Peter Berezin
Summary
Transcript
We're still kind of in the middle innings of a structural bull market in gold. It'll end badly. They always do. I don't think it's ended already. US tech was a great trade this year and past years. I'm skeptical that it'll be a great trade in 2026. If that happens, it'll be time to run for the hills, but it hasn't happened yet. >> The day ahead of the FOMC announcement, what are your expectations? And importantly, uh what is the guidance from the Fed that you're looking for? that's important for you. Peter Beren joins us once more. He's a director of research and the chief global strategist at BCA research. We'll be getting his views on the economy and uh equities markets. Welcome back to the show, Peter. Good to see you again. >> Good to be back on. >> One of the um we're speaking towards the end of October, so it's well into Q4. One of the um hallmarks of this year is marked by unprecedented prices in stocks, Bitcoin, gold, precious metals. And uh this comes in a backdrop of trade uncertainty and uh a lot of volatility with geopolitics. How would you characterize uh this year's performance in the stock market particularly when it comes to um this heightened sense of uh euphoria uh that some economists on my show have said is contradicting somewhat to the real economy. I'll let you comment on that. Yeah, I've been using the metaphor of walking on a tight rope. So, as long as you're still balanced, everything looks fine. Uh the economy is able to avoid a recession, it's able to avoid uh higher inflation, but it's kind of hard to stay there. Probably at some point you are going to fall either in direction of overheating or to rapid cooling and rising unemployment. That's sort of my worry that a lot of these risks are there. They may not be super visible now, but if they materialize given where valuations are, stocks can fall a lot. >> You had um you had a slightly more bearish outlook on the S&P. What why do you think the growth of the stock market exceeded your initial expectations? What were the uh wild cards this year? I >> I think there are a number of uh factors. one, Trump did uh change course on the trade war and so the sort of tit fortat retaliation that we could have seen uh starting in April never occurred. Uh fiscal policy has remained very very stimulative. There was an open question as to whether the bond market will kind of riot over the prospect of these large unfunded deficits. It never did. you know, bond yields are now slightly below 4% in the in the 10-year. And finally, perhaps most importantly, the whole AI trade that supported capital spending. And perhaps more more importantly, it supported consumption through the wealth effect. You know, we've had since August declining real disposable income in the US and yet consumer spending has remained pretty robust as people dig into those savings feeling that they have the wealth to do so. >> Uh just on the bond markets, you brought that up. We're now at $38 trillion of debt. It was announced last week. Uh the deficit continues to rise. At what point can we see some sort of revolt uh in the bond market in either in the form of much higher yields, much higher spreads, maybe a failed auction in the treasury market at some point or does none of this matter really, Peter? >> No, it matters, but it's very difficult to predict when such a revolt will occur. It's a bit like asking, you know, when will the avalanche be? You're looking at a mountain side with a lot of snow. You know that it's going to happen eventually if [laughter] snow continues to fall, but you don't know when. All you know is that there's fragility and there's risk. >> So, let's just pull up one of your posts on LinkedIn. Uh, I encourage people to check out Peter's LinkedIn, by the way. You post regular updates on your thoughts. The bursting of the internet bubble in 2000 was preceded by a manic boom bust episode in the most speculative tech stocks is a recent selloff in nuclear rare earth and quantum stocks sending a similar red flag. And here you have two charts uh from BCA. So you've got the first uh first chart is here uh from 1996 to 2004 you've got internet stocks um and then uh in their performance during that time and then now you have the current period uh you've got quantum computing rare earth and nuclear. My my first my first question is why are you putting in rare earths and nuclear uh in this analysis? Well, because it's part of a broader speculative uh theme. Investors have seen all these huge gains in large cap mega tech stocks. Uh they figure to themselves that okay, I'm not going to make that much more money in Nvidia. It's already a $4 trillion plus company. So, I'm going to go down the quality ladder. I'm going to go down the market cap ladder and find something small, find something speculative to buy. And that sort of attitude often does prevail towards the end of bull markets. We saw that sort of boom bust episode at the tail end of the dotcom bubble. We saw something again not too long ago with some of these rare earth nuclear uh quantum computing uh stocks. Is that a red flag? I mean it might be. It might be. I don't know for sure. What I am telling my clients is before you get really defensive, you know, wait for what I call a metaverse moment, an occasion where some big cap tech company comes out and says we're going to spend a bazillion dollars on GPUs and data centers and its stock price rather than going up goes down. If that happens, it'll be time to run for the hills. But it hasn't happened yet. the narrative that AI is in a bubble of AI stocks rather is in a bubble. Um that's been that's been talked about for months if not years. How do you make that kind of evaluation? How do you definitively say well a P of XYZ uh means it's a bubble versus just you know expensive or fairly valued relative to peers? Um you know at one point does something in the tech space become a bubble? >> Well I think you can have a bubble in valuations uh but you can also have a bubble in earnings. So think about banks in 2006 2007 you know they were trading at pretty low P ratios sort of in the in the teens they didn't look that expensive based on P ratio but they were actually in a bubble because those earnings were massively overinflated by the uh housing uh bubble so I think the risk to investors is that the earnings that companies such as Nvidia and Microsoft uh Meta are reporting are exagger exaggerated by the fact that we have this unprecedented excess demand for compute. Like if you're Nvidia selling chips, you can make a lot of money. If you're Microsoft or AWS renting out cloud computing, you can charge very high fees. But the capex is ongoing. The supply is increasing and eventually we're going to see more competition and prices will come down as will earnings. So those P ratios may not be giving you a good signal of how expensive or cheap the stocks are. >> Uh and how comparable is this period right now to 1999 for example? >> It's in the same ball ballpark for sure. I mean if you look at the concentration of the stock market in tech it's comparable to what we saw in 1999. the P ratios are not as crazy but as I said that's partly because the earnings today may be greatly exaggerated by kind of cyclical factors if you look at price to sales ratios for example the S&P 500 is trading something like 50 times more 50 50% more expensive than where it was at the peak in 2000 and so from that perspective valuations are extremely stretched and unfortunately you know We all know the saying, stocks take the uh stairs up and the elevator down. Uh we could be getting close to one of these wy coyote moments where the stock market looks down, panics and follows a lot. >> This uh Reuters article uh summarized the situation as the following. A widely cited MIT study earlier this year found that of the more than 300 AI projects analyzed, only about 5% delivered measurable gains. Uh the four tech giants and other major cloud firms are together spending uh roughly $400 billion on AI infrastructure this year, but returns for businesses adopting the technology remain uncertain. Potentially, we'll see Peter returns come later. Uh as as an investor or strategist, you're looking at maybe data like this and other data points. Uh what do you what do you do with that? Do you do you say to yourself, well, you know, we're we're not potentially seeing returns right away. They're spending a lot on capex. In fact, the tech companies are spending more on capex than any other S&P sector. Uh and and potentially that's good uh for valuations now or do you look at this as kind of a an opportunity that is a bit overhyped here? >> So this is like the you know trillion dollar question. We don't know what sort of productivity gains AI will uh deliver. There was a study from Metr. arts, a think tank that studies AI capabilities just out a month ago or so. What they did in this study was that they uh looked at the extent to which veteran software developers benefited from the use of AI in their coding. What they found was that access to AI actually reduced their productivity by about 19 uh%. And so I think the jury is still out on how seismic AI will be. It could be the case that AI ends up being very useful in automating kind of very uh standard generic type of work but isn't very useful at sort of pushing out the envelope in terms of the knowledge that humanity has access to. In which case, you know, it'll have implications for our productivity growth. But will it justify the sort of investments we're seeing now? Probably not. >> Now, uh there's been this speculation that a lot of the valuations um have been boosted by tech companies investing in each other. Just today, for example, uh Microsoft is is investing um uh well, they're restructuring their investment. So now they're following the recapitalization. Microsoft holds an investment in Open AI group PBC valued at approximately $135 billion representing roughly 27% stake in the company. Uh Nvidia today uh takes a $1 billion stake in Nokia and now the uh company is up 26%. There's a lot of this what some people call on the internet um this this circular financing going on uh and uh it appears to be working for investors. Um how sustainable is this? And if is that really the reason for why valuations are high? >> Yeah. Don't don't call it a Ponzi scheme. It's a trapezoidal it's a trapezoidal AI opportunity. That's what it is. Uh yeah, I mean there's a lot of kind of uh monkey business going on here. You have to kind of ask yourself like why does Meta not want to put the debt for these data centers on its balance sheet? Why is it moving it to special purpose vehicles? So why is Microsoft uh keen to rent chips as opposed to buy them? You know, probably because there's some worry as to whether those chips will end up being all that valuable. Right now, you have a situation where Nvidia makes a sale, books a big profit, Microsoft or AWS or Meta, whoever buys the chip, but it's not a cost for them. It's capex, right? Gets written off over time. But if it doesn't prove to be all that useful, those chips, they'll get rid written off quite quickly and those companies will suffer losses. I think that's sort of the risk. >> All right. At this point, uh I'm going to ask you for maybe your revised S&P outlook if you have one and overall your tactical asset allocation um as of the end of October. >> So tactically, I'm neutral. Uh over a 12-month horizon, I would be more uh defensive, but tactically like over a 3-month horizon, I would be neutral. But I would be looking at a checklist of things to get more defensive. So what do I want to see? I mentioned that metaverse moment as a potential catalyst for the end of the EI trade. But on the economic side, I'd like to see, not I'd like to see, I would need to see uh an increase in layoffs before I turn more bearish on the economy. Right now, we have kind of this two-tier labor market. You got the core of the labor market, which is fine. If you have a job, you're not too worried because layoffs are low. But if you're on the periphery, if you're like a new entrant, a young student looking for work for the first time, if you're a minority, for example, things are much more difficult. If layoffs rise, then the core of the labor market will start to break apart and then you're looking at a recession, but we haven't seen it yet. So, we'll be looking for that to get much more defensive. Well, Amazon just announced today or this week that rather that they're cutting 14,000 jobs um at the corporate level. Uh and some of it may be due to automation. They want to make the company a lot leaner. They want to they want to call themselves the world's biggest tech startup uh or operate that way. I I don't know if this is the u sort of a start of of more job uncertainty in the tech sector or overall maybe this is company specific. Are you seeing signs on the periphery of more layoffs to come, Peter? Not a lot. Not a lot. I mean, you can look at initial unemployment claims and we don't have the federal government data, but the states still release data and you can sort of compile compile all of that data. It doesn't really show much of a pickup. So, you have to be careful with like headlines from Amazon. Amazon overhunded during the pandemic. And realistically, it's much more convenient for a company to say that they're laying people off because of AI than because of performance issues. performance of shoes, you know, you get sued for wrongful dismissal. So, I would take some of these uh headlines with a grain of salt. >> Uh we don't have government data on on uh more official government data on employment right now. So, I'm just curious, what are you as an economist doing to gauge uh the performance of the labor market? >> Well, it's it's hard. It's much it's much harder to do so. I mean, we do have some alternatives from the ADP. They just released some data uh this morning showing fairly sluggish job growth uh through to the uh through to October 11th if if I recall. Um so the numbers on the job market don't look particularly strong but they don't look recessionary either. I don't buy this argument that the weakness is entirely due to supply. If it was just sort of slower labor force growth that was driving weak employment growth, you would expect to see those sectors where labor shortages are most acute, like healthcare, seeing the least job gains. In fact, they're seeing the largest job gains. So, this is partly a demand story, but it hasn't gotten to the point where layoffs have increased. Until that happens, u you're not going to see a recession. when we do get uh more of a slowdown or perhaps even a recession if it eventually comes, how low will the Fed funds rate get? Can we see a return to uh zero? It's possible. I mean, I don't think that would be my base case, but uh I do think that the next recession could resemble the one in 2001 where the recession was more the uh consequence of a stock market crash than the cause of a stock market crash. Usually the economy goes down and stocks fall, but sometimes the opposite can happen if the AI trade starts to fizzle. As I said, that's been really one of the few things that has been supporting the economy. The economy will weaken and that will sort of feed on itself. In that sort of a context, the Fed will cut rates pretty aggressively. That's what they did in 2001. To zero, probably not, but to 2% possibly. Of course, what's different now is that in 2001 when the Fed cut rates that created a new bubble which sort of rescued the economy temporarily, the housing bubble. uh we might not get a second bubble this time around in which case the downturn could be somewhat deeper. >> Scott Bent named his picks for the next Fed chair. Um most of them I believe are uh very doubbish. Correct me if I'm wrong. Um is the market just essentially pricing in that the Fed will cut aggressively despite uh or regardless of where inflation is headed from here? Peter, >> well dovish or obsequious? I mean, it's it's a fine line, I suppose. Um, you know, I I think right now it's okay to be doubbish because the economy isn't in great health. Certainly, the labor market looks very very uh vulnerable. I think the problem would be is if the Fed were to cut rates against the backdrop where the economy is starting to heat up, inflation is rising. That could cause real problems for the uh bond market. I don't think it's going to happen though. Like, you know, I I worked in Washington for a very long time. Like, the Fed is like the Borg. Once you're there, you sort of get uh absorbed by the whole ideology, the kind of evidence-based approach to monetary policym. You can't really be a renegade in that sort of institution. Uh but over time, I suppose if you keep on replacing uh people with political lackey, you can change the the dynamics. But it'll take several years for that to happen. It's not going to happen like over the next 6 months. >> What what's your and your uh firm's outlook on inflation? And the second part of the question is at what upper band of inflation will the Fed actually start to pause their pivot? >> Well, I think they would get worried if inflation expectations were to increase meaningfully right now. If you look at say the one-year CPI swap, like it's trading at a bit over 3%. Which is not where the Fed wants it. But if you look at the one-year forward CPI swaps or the expectation of where inflation will be, not over the next 12 months, but over the subsequent 12 months, it gets closer to 2%. So longerterm inflation expectations look pretty well anchored. Uh, as long as that remains the case, the Fed could cut rates, but if they were to rise, if they were to rise, then you're going to have real problems, not just for the Fed, but for the bond market and for the broader broader broader economy. Do you believe that the effects of tariffs are backloaded in the sense that in the first half of the year maybe economic growth didn't contract as much as maybe anticipated and perhaps inflation data, official inflation data didn't reflect it yet, but it's going to come much later this year or perhaps into spring next year? Well, the data is very clear that companies are now starting to pass on tariffs to customers and that in turn will erode real income. So, that's a negative for the economy. On the flip side, you can argue maybe not too plausibly, but at least you could argue that trade uncertainty has fallen significantly. Uh, and that should give companies sort of the green light to invest more. I'm a bit skeptical of that argument because for a firm to, you know, build a factory that's only going to be profitable if those tariffs remain in place, you got to be pretty certain the tariffs will remain in place. And given the possibility that the Supreme Court could rule against the tariffs, given the possibility that Trump could just strike a deal that reduces tariffs, you're not going to see too many people wanting to make multi-year investment commitments. >> Let's talk about precious metals. There's a lot of speculation as to what this rise in gold is signing. You even posted yourself on your views on the recent search. This is what you had to say 3 days ago on LinkedIn. Despite the surge in gold prices this year, the amount of gold held in US ETFs is still below where it was in 2022, the tactical where my tactical quant model, which brings together dozens of leading indicators for gold prices, remains bullish. Here you have a price of gold versus ETF holdings. Uh, as you can see, it's more or less parabolic. I'm talking about the price. Um, and this is your quant model. Maybe explain how this model works and ultimately why you're still bullish at $4,000. I mean, gold has fallen somewhat, like 12, 11, 12% since its highs. Uh but it's still $4,000 right now. >> Yeah. I mean, the model's been bullish for a while and, you know, it's done really well uh for for our readers. Um it's still bullish because kind of the medium-term trend remains to the upside even if the short-term trend has deteriorated. Now we have kind of this daily model as well for gold which did turn a little bit bearish but kind of the threemonth model is bullish because it senses that the underlying trend is still to the upside that you have these fundamental tailwinds for gold. The fact that central banks are uh buying gold in large quantities. the fact that fiscal policy remains unsustainable across much of the uh developed world and the fact that retail interest is only now perking up. You know what that chart that you just showed illustrates is that the amount of gold in US listed ETFs is still below where it was in 2022. So if you look at the amount of gold in like GLD, it hasn't even returned to where it was back then. Typically, the end of a bull market occurs when you have this massive wave of retail participation. That's happening now, but we haven't reached a crescendo. The um I'm going to pull up this chart here. This is the tip ETF tracks the Treasury inflation protection security. Um kind of a proxy for what you were saying earlier, the one-year forward CPI inflation rate, but I'm going to overlay this with gold. Historically, they've held a close correlation um or relatively close. I'm going to zoom in to the last couple years now. Uh and you can see that um the past couple years they've kind of been moving together. Um and if you keep scrolling to this here, that's when that's when things start to break apart. If you if you just take a if you just zoom out a little bit here, um they have moved up in the same direction, but gold has way overshot uh the growth. There we go. That's a little bit clear. Gold has way overshot the growth of uh of tips. And um to your earlier point, inflation expectations are not rising dramatically. They've gone up a little bit. Uh but it doesn't seem like gold is following that this year. So what's what's what's going on here? >> Well, the way bull markets usually work is that you begin with a very sensible narrative and then investors extrapolate. They get too far ahead of themselves. the prices move so much that that narrative is more than fully discounted. So I think with gold it'll be that same story. It'll all end in tears. The question is one of timing. Is that going to happen this year? Has it happened already? Or is it going to happen further down the road? My guess is that we're still kind of in the middle innings of a structural bull market in gold. It'll end badly. They always do. Um but I don't think it's ended already. >> The uh the amount of gold held in foreign reserves is now exceeded uh according to some sources exceeded uh the amount of treasuries held by said foreign reserves. Uh for the first time in I believe decades. Uh now Peter what what is your view on the US dollar uh especially given this particular trend that foreign central banks and uh foreign reserves are now accumulating more gold and dumping dollars presumably. What what is the signal for the for the future of the dollar? >> Well I think I mean for the most part most central banks hold a lot more dollars. They might be not treasury bonds might be treasury bills or bank accounts or things like that and they hold gold. So, so this is not something that is close to uh ex is exhaustion. What I think is true is that a lot of central banks are looking for something to hold other than dollars. They saw the experience of Russia here. I'm talking about kind of em central banks. They don't want to experience that themselves. And so, gold just becomes kind of a natural alternative asset for them to hold. They're not going to hold crypto because crypto is a threat to fiat currencies which they themselves administer. Holding gold is a lot more innocuous and I think that's sort of where they're heading and that trend isn't going to end anytime soon. >> Uh Peter, let's finish off then on your allocation to other assets and uh sectors other sectors that you do like. Um I'll let you comment on that, but I have some specific sectors I'd like you to address. So, people have been talking about nuclear power as the future, and Trump has made nuclear um power a little more accessible to to industry. He's um he's deregulated some of the mining restrictions around uranium. Uh what's your take on nuclear? I >> mean, it's it's it's it's it's a good technology in the current age, but it's also one that is not undiscovered. People in the investment community know the story and they know the story well and some of the stocks that are levered to this uh theme are very expensive. Some of them don't even have any revenue yet. Uh I think that trade has uh gone past this expiration date. >> What what's the investment thesis behind hyperscalers? Are we are we investing in real estate um in urban centers? Are we investing in cloud services uh big tech names? What's the play here? Well, you're investing in the AI dream. You're investing in the hope that uh these companies will be able to monetize the investments that they've undertaken. But the bar for that is just really really high. Uh probably by the end of this decade, at least based on current analyst consensus estimates, you're going to have close to like a mill like close to a trillion dollars in annual capex. Uh now to profitably monetize that you're going to need something like $2 trillion in annual uh revenue uh that's going to be really hard to generate. I mean the combined sales if you look at sort of trailing 12 month sales of Microsoft, Google, AWS, Oracle, Meta, it only adds up to about a trillion. Um so where are we going to get two trillion in annual sales from? Unless AI really takes off, uh, it's just not going to happen. In which case, those stocks are going to go down. That's the risk for investors. That's the risk for the economy. >> Any other sectors that you do like or assets? Well, I think right now, as I said, you know, I'm modestly defensive, not wildly so, but within that kind of defensive uh uh theme, I would be looking to, you know, buy more consumer staples, buy more healthcare, like those sectors haven't done particularly well this year. They're relatively cheap. I would also be looking to look abroad. You know, emerging markets are coming up from behind. Uh they're cheaper. They've got pretty good valuations in many cases. They got pretty good economic fundamentals. Uh so US tech was a great trade this year and past years. I'm skeptical that it'll be a great trade in 2026. >> Uh finally, Peter, so we're speaking ahead, right? The day ahead of the FOMC announcement. What are your expectations and importantly uh what is the guidance from the Fed that you're looking for that's important for you? >> [gasps] >> I mean, I think the Fed will sort of continue to sound as though they're in an easing mode, but they're in no hurry to ease aggressively. They're going to talk about how they're focused on the labor market, how the recent data hasn't been abysmal, but hasn't really been very supportive either. They're going to talk about inflation, uh, what's happening to actual inflation, inflation expectations. That I think is a concern for them. Uh and so they're in kind of this phase where they're cutting but not particularly aggressively. >> Very good. Thank you very much, Peter. Uh we'll follow up with you next time. Appreciate your time today. Where can we uh learn more about you and uh read your work. >> You can go on LinkedIn as you mentioned, but you know, I also have a following on Twitter X, whatever you want to call it these days. So you can go there and of course visit the VCA website and sign up for for a trial if you'd like. We'll put all those links down below. Follow BCA, follow Peter, and uh we'll speak next time. Appreciate it, Peter. Thank you. >> Thank you. >> Thank you for watching. Don't forget to like, subscribe.