AI Market Dynamics: The podcast discusses the exponential growth in AI demand, driven by increased token consumption and decreasing costs due to advancements in large language models (LLMs) and Nvidia hardware.
Gold Market Insights: Gold has reached all-time highs, driven by geopolitical tensions, inflation risks, and increased central bank purchases, suggesting a long-term bullish trend despite potential short-term corrections.
Credit Market Concerns: The discussion highlights potential risks in the private credit markets, particularly focusing on the role of Business Development Companies (BDCs) as a potential escape valve during credit squeezes.
Investment Opportunities: OpenAI and gaming sectors are identified as significant beneficiaries of AI advancements, with OpenAI potentially challenging existing tech giants in various domains.
Fiscal Outlook: The U.S. fiscal deficit is expected to improve, with projections of 4.5% to 5.5% of GDP, driven by tariff revenues and reduced government spending, contributing to overall financial stability.
Energy and AI: The podcast notes the competitive advantage of China in energy costs, which could impact the global AI infrastructure landscape, emphasizing the need for the U.S. to enhance its energy production capacity.
Portfolio Strategy: There's a growing discussion around the traditional 60/40 portfolio model, with some strategists suggesting a shift away from bonds due to their lower returns compared to equities and other asset classes.
Transcript
So, what am I going to talk about today? Number uh basically three things and then some uh extra goodies at the end. Number one, what is really going on in the AI market? Uh number two, gold has performed like crazy. All-time highs $4,000 an ounce this year. Did I miss it? And then thirdly, what's going on in the credit markets? Uh we've had some headlines recently and have some ideas on what investors should be looking at as that uh as that plays out. [Music] Welcome to thoughtful money. I'm its founder and your host Adam Tagert. When market uncertainties is as high as it is now, I often emphasize that the most useful people to interview are asset managers because they don't have the luxury of merely having an opinion on the road ahead. They have to commit capital to their convictions and be judged upon the results. Today, we've got the great fortune of having the return appearance of one of the most respected capital allocators in the business, Yan Vanek. Janna is CEO of VANC, an asset management firm with over a hundred billion dollars in assets under management, invested across its wide family of ETFs and funds, spending equity, bond, commodity, digital, and regional asset classes. As we've done the past several quarters now, Jan and I will spend the next hour discussing his Q4 macro and market outlooks, as well as where he sees the biggest opportunities for investors right now. Yan, thanks so much for joining us today. It's good to see you again, Adam. >> It is too. Um, I got to tell you, Yan, it is just such a pleasure and a privilege to have you come on for these quarterly outlooks. Um, I very much look forward to them. The audience does as well. Um, I've been having people ping me a lot recently saying, "Okay, isn't it time for Jan's outlook?" So, here we go, folks. We're going to we're going to scratch that itch for you. Um, so Yan, I will stop talking really quickly because I know you've provided um a chart deck as you have done in the past for these. So, we'll get to that as quickly as possible. I guess right before we do, if you had to pick a a word or a phrase to sort of describe the macro situation right now, what comes to mind? >> Happy. >> Happy. >> Okay. Like like >> you only give me one word. >> Healthy kind of happy or like a like a manic kind of happy. >> I would say happy with eyes wide open. Um there's so many things that are changing so fast. So that's the eyes wide open part. >> Okay. All right. Well, I could ask you a lot of questions about the why of that, but I think your slides are going to answer that better than uh any question I could ask. So, why don't we why don't we jump in your material. >> All right. Yeah, thanks for thanks for doing this and uh you know, my uh my friends and clients look forward to us doing this together as well. So, thank you for the opportunity, Adam. Okay. Um just a little bit uh by way of background um in case uh people haven't heard me before just a little quick reminder of the perspective that VANC brings and I bring to looking at the markets um you know we look at the same big drivers that other people do government spending uh monetary policy fed policy and technology trends I think what differentiates us uh is a little bit what I'd call our historical perspective uh which is that change can be more dramatic um and rapid than people might expect and an odd range of things can happen uh because they've really happened before in history. So that's number one. And secondly, we try not to focus too much on uh daily, monthly or weekly trends, but focus more on multi-year trends and see how they're playing out uh today. And then I would say that if we think the puck is headed a particular direction over a multi-year period, that gives us higher conviction. Meaning sometimes you can have uh a stronger belief in something is going to happen in the next five or 10 years rather than it's going to happen tomorrow because tomorrow is a little bit uncertain. So um and also just by way of background there there are four big crazy things happening in the world, but I'm not going to focus on on all of these today. One is just that we are just getting through the worst ever uh peaceime fiscal deficit in US history meaning uh low unemployment uh but high government spending. I think that AI is the second biggest technology that's hit America. Uh the the first being railroads I'll refer to them very briefly. Uh third that India is headed the direction of being a major consumer economy in the next 10 years. uh to the scale maybe not quite of China but close. And lastly, that we've had uh a dramatic uh deregulatory act in the United States uh payments markets this year, but not going to talk to it, but as as big as what FDR did with the banking system in the 1930s. So um I will now say if you have no more time to listen to me uh very quickly I've I haven't done this before Adam but in a nutshell um you know what do we think is happening um number one quickly that the fiscal deficit is probably getting better than people think um estimating four and a half to five and a half% for next year that AI and we'll talk a bit about this is supporting the megga cap profitability, the fact that the S&P continues to rise on a market cap basis and that the bigger continued to take share market cap share. So we said this a quarter ago that we wouldn't be surprised if they had a good quarter and that continued and and indeed it happened. Um if there's a cloud to worry about, it's probably some kind of growth scare that would come out of the employment markets because we have so many things going on there including tariffs and um you know AI displacement. Uh we think it's early for the ddollarization or gold trend. We'll talk about that. Um in India, I don't have to talk about that. And then we'll talk about the AI plays. So what am I going to talk about today? number uh basically three things and then some uh extra goodies at the end. Number one, what is really going on in the AI market. Uh number two, gold has performed like crazy. All-time highs $4,000 an ounce this year. Did I miss it? And then thirdly, what's going on in the credit markets? Uh we've had some headlines recently and have some ideas on what investors should be looking at as that uh as that plays out. Does that sound like an okay agenda? >> It does. I was going to tell you on I can tell you with confidence that those are all questions the thoughtful money audience is asking in real time right now. So very timely. >> Great. Well, so for the AI stuff, um I really want to talk about what exactly is going on um on the on the technology front before we get to earnings projections. So we've talked about Nvidia, but I want to talk about what is driving this huge demand for compute and the deal frenzy that we've seen. And what I'm going to propose is that the craziness of technology is the following that I call that third bullet there crazy mainframe math which is that token demand is exploding and tokens are sort of like I'll call them words or uh units of of knowledge in in AI world. So it's basically the amount of demand for AI is going up exponentially at the same time that the costs are coming down exponentially and the costs being twofold. Number one the efficiency of the LLMs large language models and number two the efficiency of the NVIDIA hardware. So we'll drill into that. So part of the reason that I said eyes wide open at the beginning, Adam, is that there's this crazy frenzy for compute demand and the variables are moving at unbelievable speeds. So we'll look at some charts there. Uh we've just get a refresh on what's happened to the beneficiaries of the hyperscaler demand for energy as a second bullet. And then I want to highlight um a new company I think is the the new Mag 8 company. And then I want to talk about gaming as another AI winner that emerged this quarter. So all right. So here's the crazy mainframe math. This is tokens consumed are going up 38 times year. And that's not just because we have a a weird starting point a year ago. It's going up like 20% a month and there's >> Hey J, I'm I'm sorry to interrupt but just so people understand you mentioned it briefly but token is that is that like the inputs on which we train AI and those are going up or is it like the queries like questions people are asking of AI? >> Oh no, no. Yeah, good point. Not queries. It's to think of it as words. I'm going to say words but there's more I think more knowledge in there than words per se. But if most of what you're doing is processing language or words, these are the these are the units of of compute and they break it down into what they call tokens. >> All right. But again, is are those like the inputs? Are those like the the words or the knowledge they're putting into the LLMs? >> Not no not per se because you could have a query that says um when did thoughtful money start, right? That's a several word query. But then it might have to consume a thousand words to figure out the answer to that. And that second component is token demand. >> Got it. Okay. So it's the I'm going to use the word input again, but it's it's it's the inputs that the AI is is is grabbing to try to answer your question. >> Correct. Right. It's what's its processing. I gave probably a pretty bad example just there because it was a single piece of information. But if you're trying to establish causality, like what makes a great thoughtful money podcast, right? Then then that requires lots of compute. And even if you defined what great was, right? A lot of tokens would be consumed in that uh in that. And and one of the things that's come out, there's two things that have emerged in the last quarter or two. one is that um some of these LLMs are capable of reasoning, right? So, not just spitting out an answer to a question, but exploring the thought process and showing users the logic that it used to get to an answer. >> So, that's one. And then second of all, um it's not just you and I that are asking queries of of these AI chats. It's software itself. Right. So a whole new source of uh demand for for AI processing is coming out of things like robots or just software that's being developed to answer questions in a more sophisticated manner. >> Okay. And on that and I don't want to I don't want to take us too far field on this but I have heard recently that you know we've trained these large language models on human generated content and we have kind of gotten to the end of that where we put about all the content that's ever been generated by humans into these models and so the next wave of of kind of information that these models are going to be trained on is going to be information that is data that's being generated by AIS themselves. >> Yeah. Yes. I mean that's a that's another dimension and I don't want to go too deep there but that that's yes again software is now we used to write software and it would process data now software is kind of rewriting itself. >> Yeah >> it's learning from the queries that it's doing. >> So I'm gonna avoid asking the question that everybody wants to know which is how long until it gets so smart it kills all of us. But we'll save that till the end of the presentation. Yeah, that's it's like it gets super dark. I'm I'm just I'm frankly not worried about that sitting here today. That's why I'm happy. So, um we we can we can talk about that maybe uh in in a deep dive. So, you've got demand for tokens exploding at 38 times a year, >> right? Just exploding. And then >> the costs of processing. So, I'm going to break this down into hardware and software. So, this is the hardware slide. Oh, I'm realizing now it's probably too small to read, but these are the different uh hardware solutions that Nvidia has offered. So, you can see Kepler is the one on the on the left and and how much how many tokens how much energy it took to process a token. >> That's unbelievable the savings there. And it's going down like Jensen Hong who I I had the fortune of seeing last week. It's going down at 90% a year. So he plans on releasing new compute platforms every year and he's saying that their costs are going down 90%. So that's why I'm saying this is a crazy equation, right? Demand is, you know, exploding and costs are imploding. And what he says that took me a long time to understand is he says, "I don't care if people give away my competitors give away their chips because those people will still buy mine." And the reason he's saying that is we have a energy shortage in the US. And if his chips are 90 or 30 times more efficient than even the competitor's chips from last year, people want to consume way less energy, right? for the same amount of token processing power. >> So, we're kind of on this hamster wheel, but it's it's an amazing one. Uh that that's why I'm saying eyes wide open that we have to keep our eye as these things develop. And then the third component here is I'll call it software. So, these are the large language models. And um on the left hand scale is just the cost of processing tokens in dollars. And and you can just get from the shape of the two colored uh lines and I'll explain what they are in a second that they're coming down and amazing. And these are I think logarithmic charts. So for those of you into math, they're falling really fast for the English majors in the audience. So you've got two cost components falling fast. the hardware which we just saw and then this is the software. Now the two different lines are basically describing two types of LLMs. The more sophisticated ones are the ones in orange. So they score much higher on kind of if you will benchmarks, intelligence benchmarks of these large language models. Um and that's what MMLU stands for. So the ones that are that you and I would want to use are in orange, but the kind of lower quality ones costs are falling even faster. >> Even faster. But still the orange ones are still dropping from I don't know 50 to to 75. Just massive drops. >> And and this is why like I think hopefully it's pretty clear that if Deep Seek comes out with a model that drops it a lot, there's going to be Whoa. Maybe suddenly demand for compute goes down, right? But if the MA models start choking and don't uh improve on their efficiency, then all the efficiency has to come on the hardware side. >> Okay, >> so my mainly my main point is there's two this is the real technology behind what's going on and and it's pretty incredible. Now uh so who cares Yan? Okay. What I would say is what we are seeing and here's to Nvidia. Uh we'll just use this as the benchmark for all this. What we've seen over the summer, Adam, is a huge run to lock down compute. All the different deals that are happening, and I have a chart on that later, but Open AI wants high bandwidth memory. They want chips. Nvidia wants chips. Everyone wants more compute power because basically I think the industry realizes that demand is exploding so much that there will be a shortage of compute and what's odd is I think Wall Street is missing it or my colleagues think Wall Street is missing it. So you know I've gone luc uh cold and hot on Nvidia stock. um you know a year ago July I was a little bit cold on it uh because of the valuations now we think Nvidia is reporting that their demand is well clear visible into 2027 so two more years of of earnings and their their valuation their price to earnings ratio is I would say at the high end of a normal company but not of a company that's growing 40 to 50% a year which is what they're talking about. So, you know, if that if those growth rates are are achieved, then their forward pees should go up as well. So, the stock price should be getting a lift from higher earnings and a higher multiple. What we worry about is three to four years from now when the multiple comes down. But in the short term, we're facing a compute shortage. >> Okay. Um, I I totally get that and it's interesting. So, like you, Yan, um, and probably even more so over the past couple years, I've had trouble wrapping my brain around how a company could could be worth so much and worth so much so quickly. And I I I'm I'm kind of taking from your presentation here like I I I'll often talk about a trillion like when we talk about our trillions of debt and and when you actually try to tell somebody how much a trillion is it's a number so vast our brains really can't fully grasp it right um and to a certain extent I hear you saying that the the the the dynamics that are in play in this AI market are at at a such a scale and moving at such a speed that it is really hard for most people to really truly understand it. And and because of all this, you know, there's still a fair amount of demand left here that makes Nvidia's nosebleleed stock price maybe fairly valued here, maybe even a little little cheap right now relatively. >> Yeah. Well, I wanna I also wanna maybe this is just something that everyone else knew but Yan, but Nvidia is is in a semiconductor ETF. It's a chip manufacturer, but it's really not anymore, right? It's gone from just a GPO card manufacturer to now it is the mainframe of today, right? They are supplying the whole compute platform. And what's different about that is you know they are the heart of the ecosystem. They are the ones cutting deals with memory suppliers or CPU with the different bits and components, but they through their CUDA software basically want it. They want everyone to live in an Nvidia world. And we and we do, which means that it's got more competitive modes than simply being it can't beclipsed just if someone comes out with a with a faster process. There is my point. They've got a whole eco competitors have a whole ecosystem. And may maybe that's just Jensen Wong's sales pitch that I'm repeating, but all the deals that we saw over the summer, I think support that thesis. >> Okay. Um All right. I'm I'm resisting about asking you about an AI bubble and all that stuff. We we'll save those questions for later to let you go through this. >> I'm trying to I'm trying to address it, right? And and we'll get Yeah. We'll have a couple more slides, but part of what I'm saying is demand is going up for something like we've hardly ever seen in our life. Yeah, it may not be a bubble. It may just be a super big demand surge that merits all this. >> I I think so. I mean, that that's what I would just say. And and it's easy. Let's keep your eyes on to part of part of the hope of these uh these quarterly sessions, Adam, is investors know what to focus on. What I would focus on if I had to focus on one thing is token demand, right? And if that's not going up exponentially, then I'm really worried. >> Really worried. Okay. So, two two quick questions out of this and I'll let you resume. Um I guess let's start on token demand. This kind of gets to the um the AI bubble question I I wanted to ask you, which is I totally get in this world why Nvidia is the darling, right? Because because of all this, we need more computer. You know, we we we got to invest in the platform to make all this happen, right? And largely that's happening on Nvidia technology. Um now at the end of the day, why does this matter? Right? We AI doesn't have a ton of value if we just make Nvidia a really rich company. It matters because we're doing all sorts of amazing new things that create a whole bunch of incrementally new value for us using AI. And to date, there hasn't been a ton of that coming out in terms of like incremental revenue or profit on the income statements of corporate America. And so the question I have for you is are we beginning to to are we beginning to see some of that or is this still being sort of all taken on faith that if we feed Nvidia enough, it'll support the AI valuations that all the stocks in the AI ecosystem have? >> Yeah. Um, so I would just sort of say yes. And that's sort of the the counter. What I would say is the at the super high level, the US is a consumer economy, not a manufacturing economy. And the consumer is now in a different regime of interfacing. And I'll explain this in in the next couple of slides >> with what it does, entertainment, purchasing, working, right? All of that. And the economics are well spoken for, but I'll get that to to that in in a second. So, uh it's not that this is a toy that is unhinged from revenue, but let me give you some nuance there because there is there is one weakness that I want to want to focus on. >> Okay. Um, so, so let let's just do a side trip to what we talked about three months ago, six months ago, which is there's going to be a lot of demand for electricity and nuclear and some of these other things are going to win. I mean, this is exceeded my wildest expectations. Nuclear stocks are up 90%. Olo, which has no, forget profits, has almost no revenue, is up seven times this year. >> They don't make any energy. They they haven't produced one electron yet. >> I I I am I am check me you know as my dad used to say take me out coach. I don't know what game this is. I I I I can't say sell because I believe in this as a multi-year trend but I I would more I don't know. I just don't know what to do. Take profits. Um don't don't take get out of the position. But this is this is really um I would say you know excessive um so uh and and just a a a consequence of that that I want to touch on briefly is just that and we've seen this bubble up already which is the costs of electricity if we have all this demand Adam the cost for electricities are going to go up and so you can see here that the cost of electricity in turquoise is above CPI high. Uh this is a very regressive uh fact, meaning that lower income people are much more affected by an increase in the cost of electricity. And um you know, we could spend a whole hour talking about it, but I just wanted to flag this uh because there were different states are dealing with this in different ways and I think there could be an entire reshuffleling of how electricity gets done, but at least I just wanted to highlight this. So, so real quick, Yan. Um, so I was thinking about this, asking you about this before you brought up those charts showing that the Nvidia uh technology is becoming more and more efficient in terms of the energy needed uh to to process tokens. >> Yeah. >> So, um I guess my question is is how how much do you expect this to spike given that there will be technology somewhat riding the rescue here? So we're we're going to be placing a lot more demands on AI queries going forward obviously as it continues to grow, but they will require less and less energy per query. So does that does that moderate the demand it's going to place on electrons or >> No, not not in the short term. This is part of the whole basically compute shortage and you know shortages and gas turbines and you know lack of nuclear. I mean we need all of the above. Uh so uh I I don't that's a good question. I don't have a a simple answer on cost projections. You know there's a difference between what consumers get charged which is this number and um you know what businesses get charged, >> right? But but as as there is increasing demand just for the overall electrons there are there people are seeing their bills creep up. We're seeing it all over. Right. I put out a tweet out I know a couple days ago asking if there was a betting market yet on Poly Market for when the first data center was going to get burned down by angry local residents. >> Oh my go Okay, that's a little uh it's a little hostile, but >> No, I know. Little little dark, but if this line keeps going up at some point there, you're going to get some real push back, right? >> No, no, no. I mean, sorry, I'm with you. Just not as dark. >> I know. which is actually the takeaway is less about AI here than about politics because there this is a whole different thing but there's a massive difference in American sentiment and polling on a range of topics postcoid >> so I don't want to get metaphysical but this I believe is one of the only one of the many ingredients to that like why do people feel so much pain in their pocketbook book relative to the quote unquote inflation of 2.3% and this is I think one of them. >> Okay. All right. Well, very important and we'll keep tracking this I'm sure in your future appearances. Real quick, I want to note that when you presented last quarter, I did own some Oaklo at that point in time. Um, but after your presentation, I did buy some of the um the Vanek uranium ETF Trust >> and that's up 76% since your last appearance here. So >> nice. Nice. You >> want to take a victory lap on that? Take a victory lap. >> Well, I don't know. I I I Yeah. I mean, that's why I love multi-year trends, you know, and hate them, right? Because you knew the puck was going there over several years, but you didn't know how much was going to get priced in in 2025. >> Yeah. >> Uh but that's a that's like a whole other thing. Okay. So, that that's kind of reviewing stuff that we've talked about before. Now, the two I'm going to give you two winners um of of AI. And again, um so I I know this is sort of stating the obvious, but I think we're kind of missing I was kind of missing how powerful Open AI is. And so I'm going to argue that um it's not only a MAG 8 company, it is at the doorstep or in fact on a daily basis competing with some of the other MAG7 companies and winning. So let's just start why I'm starting this. OpenAI I guess was launched I forget 2022 I think and it saw the fastest user adoption of anything. And so if you look at monthly active users on the right, which is the the main thing on this chart, they are really the leader in monthly engagement with AI chat. >> And I think it's amazing to think that even Google hasn't been able to get more uh more users on Gemini, right? So suddenly they are in interfacing directly with the US consumer. To your question of a couple of minutes ago, Mh. >> Um, and and we'll anyway, we'll we'll talk about that in a second, but they are uh I think they're just underappreciated. They're like our friendly little chat tool, but Sam Alman is literally competing against Gemini, I would say, and Amazon for sure. So, let's dig into that. Um, just another way of showing how chat is winning. This is year-on-year website visits relative to the Giants. like it's beating the giants. So, Google is on the far left, no growth. YouTube, no growth. Facebook a loss. Uh Instagram slight growth. Chat of course stands out. And you can see basically no one else is has a lot of um momentum. Maybe Reddit is positive. All right, let's keep going. Chat is powerful just the way Google is powerful if it's the first step for the US consumer. And so in terms of AI chats referring traffic to the sites on the left, which is where the American consumer goes, right? Amazon, Booking, Expedia, Etsy, Zillow, Carvana, look at chat market share. >> Yeah, that's really f that that's amazing. Um, it's funny because I I used to work at Yahoo and in the early early days when I didn't I had yet to work there. Yahoo was sort of synonymous for the internet. You know, there were other websites out there, but everybody just thought of Yahoo and you started your your whole journey there. It seems like chat GPT is that for AI right now. >> It's it's incredible. It's incredible. If you and I had been betting on Poly Market, would someone challenge Google for search traffic or AI chat traffic? I I think it would, you know, a long long long odd, >> right? So, Y, I got to ask then, how come Google's stock price isn't getting beaten up over this? This is threatening Google's, you know, cash cow monopoly. >> Well, I think they have a a whole range of businesses, number one, and number two, they let's just go back here. They're number two. They, you know, this may not be a winner take all. I actually I don't think it's going to be a winner take all, but we can talk let me let me try to explain that in a second. But that's that's not bad, right? Plus, they have so many other businesses. But let's talk about Okay, so this is just another way of showing the same thing. I'm sure you've probably never heard of these sites, right? Move, Zupla, and OTM, but they're basically real estate sites in the UK. And you can see that a whole bunch of their, well, I mean, maybe you can call that a small number. I don't know. 17% of their traffic is now coming through chat. So it's the same it's the same point which is um okay so now let me let me try and experiment here. I think that AI chats offer different types of services and I want to kind of differentiate them. One is like Instagram or Facebook where you kind of hang out with the people that you agree with and chat tries to basically please you and your experience is what makes chat will give you uh data and facts that make you happy that will bend you from the truth. >> Yeah. It it enriches your echo chamber. It it reinforces it, right? It reinforces your echo chamber. Exactly. And I I was listening to a professor uh talk about how like explaining how chat would um like it's like there's two e in beagle and the user kept saying no there's three and the chat would say oh I'm sorry you're right and it's just it's it's just and that's because they weren't appreciating yes there is a dimension of chat that's just trying to make you happy right that's number one number two there's a different type of chat I'll call that the nonpleasing where the you're looking for accuracy is the value, right? And and of of course you need to have accurate data, but if you're at work, you want or if you're answering a customer inquiry for commercial purposes, you don't want you want to sure you want to please the customer, but you need to give them an accurate answer, especially when you're in the health care fields or the finance fields, right? And also some of us just personally want accurate answers not you know not biased answers. So I call that a second entire use uh use case and then third there's there's shopping and then fourth there's gaming and and we'll talk about that >> but let's just dwell on shopping for example if open AI can integrate with Shopify they can basically compete against Amazon. So again here this is open AI is a mag 8 company that's competing with the other ones. I mean, what's Amazon's moat against shopping if someone can access all goods and services through Shopify? And that's why Shopify stock has gone berserk over the summer. Uh because it suddenly got effectively it can get around the Google not or not get around entirely, but it has another source of customers which is open AI. >> Interesting. So yeah, you can do the same thing you do on Amazon, which is search for a product, maybe have the product recommended to you based upon what it knows of you, too, and then you can actually transact. Um, there it's just AI plus Shopify versus having it happen all on Amazon. >> Exact. Literally, Shopify provides the payment mechanism and the logistics to deliver the product to you. It won't come in a Prime Amazon truck, right? But it'll still it'll still get to you. >> Yeah. Does does shop does does chat have an advantage on the agentic side of things over an Amazon? So, in other words, you can say, "Look, find me a you know, find me a a nutritious health bar and then buy it for me." And it just it goes off and does it all for you. >> Yeah, I'm not expert enough to know which which platform is the best for that right now, but they're all my main point is they're competing against each other. Mhm. >> And then if you go back to this commercial application here under research app chat, the winner is Open AI through their deal with Microsoft, right? Most companies work either in a Microsoft ecosystem or maybe a Google ecosystem. That's it. And they've won. They're work they're partners with Microsoft, right? So they're the leading solution. And the other thing that's happening now is that um as you and I use chat more and more, we're starting to they're starting to emerge switching costs between the different platforms, right? Because the more we query and use one chat, the more it knows who we are and the kind of answers that we're looking for. And I would argue either that's under the social or pleaser chat or the researcher chat. It's learning more about us, right? If you upload your calendar, your contacts, all that kind of stuff like in Microsoft Copilot, suddenly that life is really way more efficient for you as a user. So over time, these eyeballs are more that's why everyone's fighting for market share right now. >> Okay. And I'm going to ask a question that you might be asking answering later on, but why does not why does this not become a winner take all ecosystem, right? where the smartest AI that can come up with the best answers and do the most for me, then the network effect won't kick in and everybody wants to use that and then that makes that even better. And then why would I want to use the number two version? >> Yeah, I I don't know. There's a scale definitely a scale advantage that we've talked about before, right? And and let me let me actually let me jump ahead um and then I'll move backwards because this is such a stunning chart and I I stole this and steal it but uh I borrowed it from a firm called New Edge. So give them credit. Um >> Cameron Dawson's firm. >> Yeah, exactly. And as right here you see what happened is that retrenchment from the MAG7 in the 2022 time period in terms of their earnings. This is their earnings growth. Mhm. >> But as soon as AI kicked in, look at this. It's like a straight line to the sky. And so this is what I think is describing the stock market these days. So I'm not sure that all of them can't win against the rest of the economy. At least they're showing a lot of earnings growth partially because they're cutting their costs. So if you as you have that image in your mind, let me let me just go back and try to answer that. This is what Wall Street is all in a tizzy about, right? And part of what I try to do sitting here in Midtown Manhattan and dealing with our active investment approach, highlight, you know, kind of what I think they're missing. And what I said before is I think people are underestimating the compute shortage and and how earnings can go up over the next two years. But this sort of web of inter uh of of consumer buying and interin investment I think overconfuses it to me two there's two things and I've said them before but look at this chart. Number one, we're living in Nvidia's hardware world, right? Which is why Nvidia is cutting deals with memory suppliers, high high bandwidth memory suppliers, CPU suppliers, right? They're they're of course wanting to deal with all the the hypercomp hyperscaler compute platforms. So we're living in Nvidia's world and we're also open AI is building up a comp a compute infrastructure to compete against Google Gemini and every other Amazon and Meta against these others. So they are they are the the the risk you asked before like where's the revenue where the revenue question it can be very much narrowed down to where's open AAI's revenue because they're only printing 20 billion of revenue this year 40 billion next year and maybe a hundred billion. So they're everyone else has got revenue, Adam, to pay for their compute, right? Amazon's got AWS. Everyone has billions and billions of dollars of revenue. In fact, they haven't even been borrowing to build out their compute. They had to build it anyway. So they were spending money on computers anyway. They're just this is just the new Nvidia regime that they're switching over to. But Open AI is the one question I would say. So that's the I would have a a completely different takeaway from this chart. Not that it's interconnected, but that Open AAI is the one firm, you know, that doesn't necessarily have the revenue to pay for it. >> Okay. And I'm sorry, what what what scale did you say their current revenues are on? >> Uh, so I mean, as opposed to hundreds of billions of revenue, they have maybe 20 this year. >> Okay. >> And like 50 or 60 next year and then 100 billion the year afterwards. So they're borrowing a lot of money and raising a lot of money to build this compute platform. But the dream is huge, >> right? >> Right. Why are they getting that money? Because they're the winner so far. >> Because the potential prize is so large. Yeah. >> Well, they're winning. They have the consumer, right? The question is, can they monetize the consumer? But they do have the consumer. So, I think I'm just trying to reframe the the whole AI debate here. And partly because Open AI is not a public company, I think we're kind of missing a little bit what's going on. >> Yeah. And I was going to ask about that. Do we do we know about do they have plans to go public soon? I mean, I know that that when they initially announced thinking about it, Elon had a lot of huge issues with that, but >> I I I there's no no announcement at all on that front. >> Okay. All right. >> So, anyway, I'm sorry. I gota I got to go fast here. Um to finish the AI section because I know you're gonna we want to limit this. So the last winner, so the one winner that kind of emerged to me that's very clear that you want to watch is OpenAI and can they get their financing because that's the weak link in that whole web and then the second winner is gaming. And so very quickly the biggest we had the biggest take private in the history of finan capital markets that happened in the last quarter. I love like kind of highlighting some of these things that have happened Adam because things are happening so fast it's hard to kind of we don't even have time to think about the significance but I think what these investors are saying is gaming is a huge AI winner. So number one the video experience is going to be way better and number two the costs of programming these games and doing the QA quality assurance is dropping dramatically. So I think what they were saying by taking this company private is that the public markets are missing that gaming is a big AI winner. >> Okay. And that makes sense. We're we're entering into the AI is going to help us sort of enter into the era of Ready Player One where the the gaming experience just becomes so amazing and immersive, but the cost of providing that unbel you know much better experience is getting driven down at the same time by the technology. So, if you can own one of these companies, you're going to ride that wave. >> I think that's what this headline tells us. So, um we're finishing up the the AI story. So, who cares? Yan saying number one, focus on tokens. Number two, focus on whether Open AI uh can, you know, can raise the money to pay for the compute platform. uh and also watch it try to compete with Amazon and and you know frankly a lot of the other mag seven companies because that that's really a quite interesting thing to watch. So that those are the kind of in in a nutshell the two things I would say to look at as we watch this and also I mentioned that we think that the the stock market is underappreciating the compute shortage that's happening from this huge um you know demand for tokens. So what does that mean? we see the MAG7 earnings continue to explode. I like to say this is a chart of US stock market history. I love the railroads. Um and you can see how big the railroads became as a percent of the market. So all I'm saying is the growth in the MAG 87 um uh is not unprecedented in our in our financial market history and we don't have time to go down that rabbit hole but I just want to say it's not unprecedented. And then lastly, the >> sorry just back to that chart though. The key takeaway from that is you would expect >> the share of this chart that's made up of the dark and light blue. Sorry, the the >> uh the transports are even darker. So I don't want to confuse people, but the Yeah, the two blues near the bottom there are going to continue to get bigger as a percentage of the pie. >> Very well could be. Yeah. Um and yeah likely I would say and then the question is okay when does the earnings growth of the rest of the market catch up with these juggernauts and um you know some some people are looking at 2026 estimated earnings and they say okay the mag 7 here is expected to be 13% uh earnings growth uh small cap could be 22% %. >> Wow. >> So maybe there's other parts of the market, you know, Adam, people have been holding their breath for this for years or at least the last two years, but it's it's, you know, it's conceivable. I mean, value doesn't look like it's going to be um surpassing the mag seven in growth, but there anyway, this is I think the thing to watch is, right, does the market smooth out a little bit? And it's you saw a lot of that in Q3. So, um, anyway, so that's that's the market as a whole. I'm happy. >> Okay. And I will say that's consistent with recent, uh, experts in the channel like Mike Canterowitz from Piper Sandler and Anna Wong um, becoming a bit more optimistic about the economy next year and that the the profits are going to start broadening beyond just the the the big tech sector. >> Yeah. So, um, okay, I promised three topics and then a speed round. So, I've used up a lot of time here, uh, to talk about AI. I'm going to be pretty quick on gold. So, gold, um, the gold price here hitting all-time highs, well above 30 and 200 day moving averages on a technical basis. That means it's got no support. Um, but my perspective, and I don't have a target price. have to work on that for a future quarter. Adam is that we've got, you know, four things happening. Uh, and and I'll run through them very quickly. Uh, one is just the emerging markets are becoming wealthier, so they're just buying more gold, and that trend is not going to change. >> Number two, I would argue this is, and this is a long discussion, but military conflicts or the or the the threat or visibility of them are actually up. Now I know the Middle East has calmed down over the last couple of days, but I think the Ukraine Russia war is deepening and China by um stopping the export of rare earth and magnets has really told the US look your military needs my magnets. And I think that's caused more uncertainty uncertainty in the world infla what I call the inflation risk. I'll show you that, you know, maybe the Fed cuts rates too fast. Or put another way, Trump puts too much pressure on the Fred Fed and there's Fed independence issues. Um, again, I'm not going to dwell on all of these, but my main point is these things are not going away. And then the last one is budget problems with France and Japan causing government changes, but prime minister changes and turmoil. It's this I just don't see any of these things stopping. So just to to you know what I'm going to save some time when gold has gone into different regimes the prices has gone parabolic in an and and so yeah 4,000 feels high but it's gone up 20 times in past gold cycles. So, you know, we can again talk about that more some other time. All this >> just just so you know, Yan, I mean, take the time you need here. I'm I'm not backended. If you are, I understand, but I don't want you to cut things that are important to you. >> Okay. All right. Well, that's good to know. Sorry. Um, so so the Zoom won't go off in an hour. Um, yeah. I mean, I don't know if you have any questions. I think everyone's familiar with the gold bull market of the >> 70s. I I do have one question. So, you talked about the inflation risk there. Um, so you seem pretty sanguin on the economy for 2026. Um, if the Fed does continue cutting, do you have any worries about it restoing inflation because it's cutting into a growing economy? >> Well, uh, services inflation's still high and you know, sure goods inflation has come down, right? But, uh, this, you know, it's I don't know, Adam, but my point is the markets were are worried about this. >> And um I would say that it's interesting Wall Street split, right? Part of what I like to do is just to tell people who don't sit in the middle of portfolio managers all day long. Half half the room thinks the Fed should have been cutting that inflation's low and what's the problem? And the other is like no, they don't need to move. The economy feels like it's got a lot of momentum. The financial markets are frothy almost, right? We have IPOs. Um, you know, we have profitless companies like Olo going up seven times and the Fed doesn't need to cut. So, um, my my main point is uh the risk is there and um and and I think gold is pricing that in. >> Okay. >> So, all these things we'll be talking about I think in the future. uh the manifestation. So this is foreign central bank buying. So the the the context here is when central banks were selling in the dark blue line on the left. This is developed markets meaning Europe and not the United States but Europe let's call it and then as uh since the financial crisis in 209 the rest of the world said you know that US banking system isn't quite as stable as I thought it was. let me like start tucking some gold away. And frankly, I'm rich, so I can start tucking some gold away. And that's what started happening. And then in 2022, after the US froze Russia's uh assets overseas, they said, you know what, I'm even a little bit less confident um on what the United States is going to do with my money. I'm going to buy even more gold. Right? And so that and they can afford to. So, I just see this, like I said, I see this as a multi-deade trend. I see no reason for this to change. And so, that's why um, you know, gold, you know, gold's a beneficiary of this. >> So, and if you measure gold, now this is a little fake because gold is marked to market. Um and and so obviously the the bull market in gold has made that dark blue line go up, but it it's kind of noticeable, right, that now uh central banks own more gold than they own US treasuries. So this chart's been floating around over the last month or so. Um and but I but I think it's I think it's significant. I think >> notable >> gold gold has now reemerged. of a new regime as the world's leading currency or store of value. It's not the yen. It's not the Deutsch, sorry, the euro. I always say that I I I like I I do that informally to call the euro the Deutsch mark just to make fun of the Europeans. Um it's not the remman, it's not the rupee, right? It's it's um it's going to be gold um across those different regimes. So anyway, we can talk about more some other time. The other thing I like to point out, >> this was the next question I was going to ask you. So, great anticipation. >> No one in the US cares. I mean, I know that some of your listeners are passionate um and in following the gold market and they've benefited a lot and kudos to you and them. So, but you I you are a outlier. Most Americans are not do not care. And uh you can see the purple just shows the prices going up. But you can see in the gray line that as a percent of total investable assets measured by ETFs and mutual funds, gold funds are not growing. >> And so sorry, we we're at sorry that that's the the total assets of the gold funds. Okay. >> Correct. As a percent of total. >> Okay. All right. Um uh so is this Yan a tailwind to to be that you're anticipating here that eventually the the US investor will wake up and want to buy gold? I mean what's funny is is is uh you know I guess it took $4,000 an ounce for the mainstream financial media to start using the term basement trade. I mean, and to me that is the the financial media like finally waking up a little bit to this, you know, maybe cracking an eye open. But do you expect at some point the gold trade to get a nice tailwind as the US investor says, "Okay, look, I got to be in this thing." >> Only when the S&P goes down significantly. Otherwise, everyone is living in this world of I'm getting wealthier. All investors are getting wealthier. Unfortunately, half of America doesn't invest, but uh investors are just they're loving the S&P, right? What's not to like? I mean, if you're in the financial services industry, you literally don't have to do anything and your revenues are going up 10, 15% a year. It's crazy. >> So, no, I I I this could take years, Adam. I'm not holding my breath. Are you? >> I have no idea the timing of it. So, I value your your point of view there. And you're you're not the first to say that which is it's going to it's going to take the you know basically the status quo to stumble before people start thinking about that. >> Yeah. Okay. So not too late to buy gold um is the short version. So we covered AI, we talked about what to look for. We talked about gold. I don't think it's early because these trends are multi-year trends. >> And the third topic is just the private credit markets. >> Yep. And sir, real quick before you get there, just just to make it practical for folks. That said, you you showed the the chart of gold being pretty far above its moving averages. I am guessing you would not be shocked to see a pullback in gold in the near term, but that would be more or less sort of a a buy the dip moment. Correct. >> Yeah, I mean it could correct significantly. I mean, look, you're looking at this 200 day moving average at what 3,00 3,300 3,400. I can it could easily drop $700 an ounce easily. Um but I think the long-term trend is going higher. Yeah, great point. >> Yep. Just wanted to make sure folks know that it could go down before it goes up. >> Right. So, um let's just talk about the fixed income markets. And I basically again have a simple point is I have something for you to look at uh if pressures um develop in in the credit markets and I would argue so far they really haven't been u so um I'll get there in in two slides what to look at but number one is because so much lending since the financial crisis here in 2010 has happened in the private markets the rest of us have no idea what's going on there, right? So, that's a little bit why I wanted to talk about it. We know that private credit funds and managers like Apollos and Aries have taken in billions and billions of dollars. Um, and there have been recent events over the last month or so, Adam, two basic defaults, one with a an auto lender and the other with an auto supplier, um, that are making headlines. So I wanted to say okay what what is h what what what should we be thinking of as investors and so another another thing that shows this concern is that Aries which was involved with one of the credits is a you know private credit manager this is the uh this is the management company so it manages funds then invest in these private markets and you can say it's it's had quite a fall here recently which has gotten some headlines and um we manage the largest uh BDC uh ETF and BDC's are basically uh funds that invest in private credit. And so you can see here that um they're the price of these BDCs, Adam, trades above and below what they're supposedly worth. And what they're supposedly worth is their book value. So supposedly, you know, they're very smart accountants and everything looking at the financial statements of the what they're these these funds are lending to and saying, "Okay, that's the book value." And sometimes these these funds are very popular. You can see that people will pay over 10% above what they're worth because the yields are attractive. And sometimes people freak out like during the COVID crisis or other kinds of bare markets and bonds like uh 2022, they'll trade at a 20% discount. So I basically have this theory that you can't redeem from these private credit funds, Adam. It's right. That's the kind of structure that they have. >> Yep. >> So there's Jan's theory is BDC's could be the escape valve. So if you own both private credit and BDC's and you suddenly became concerned about defaults, you would sell you can sell your BDC's in the market every single day. >> Correct? >> And they have fallen a lot in the last couple of weeks like 10%. I know maybe to you and me that doesn't look like a lot. That feels very painful to BDC investors. Mhm. >> And my basic theory is that this could easily go to a 20% discount um and because this or even maybe more if there were some kind of sudden concern about credit um in the in the markets. And so um anyway, that's my that's my thing to look for is look at these BDC prices. I'm I'm not try I'm obviously making my our shareholders unhappy that own our BISDY fund. I guess or from a perverse perspective, giving them a target price for when it might be attractive to buy, which would be at a 20 or 25% uh discount to uh to book value. So, we're looking to put this chart, I mean, I'll share these charts through you, um, Adam, as always, but we're looking to put this information on our website because I think it's something to to look at because there aren't many escape vows. you know, people can't redeem from their private credit funds except for on a quarterly limited basis. >> So, um I totally understand that. So, this this is interesting. So um one just in general you're sort of raising concerns on a topic that's getting increasingly discussed of late which is you know private equity sorry private credit um they both have but private credit um has sort of really exploded over the past 5 10 years and it's not regulated. It's much more opaque than the public lending space. And so there are a lot of questions of are these guys making good loans. You know, we we kind of hope they are, but we don't really know, right? We're not going to know until the bodies start floating to the surface. And then we just had and first brands go out and and now people are beginning to get a little scared like, oh, whoa, maybe the maybe these, you know, these credit books are a lot worse than we were we were pricing in. Right. >> Yeah. So, you know, you're saying, hey, the the the escape valve, if you're an investor in that space and you're nervous is if you own some of these BDCs, yeah, you can sell them real quick. They're liquid, right? >> So, it it pushes the discount down here. So, part of what you're saying is a warning of like, hey, if you own some of these BDC's, be careful. They might get caught in this this selling pressure. Maybe you're telling people if they're real aggressive and want to try to short something, this is a short candidate to go look at. But um a lot of the viewers of this channel, Yan, they first heard about BDC's from a fellow named um Stephen Bavaria who's been on the channel a couple of times who builds these income investing portfolios. And BDC's are part of it, right? Um and one of the reasons why Stephen likes to do um income investing is because it's it's all debt. And at the end of the day, debt is a contract, right? And he's like, "The nice part about it is you only have the existential risk of the company to worry about." Like the company is going to pay its debt as long as it's still alive. You just have to hope the company's going to stay alive, right? Whereas if you're buying a stock, you have the existential risk, but then you have a whole bunch of other risks on top of that, right? And so what I'm where I'm getting here is there might be an opportunity here, as you're saying, Yan, where the market kind of freaks out. these things go on sale, but they're go they then get to a place where they're priced really attractively versus the true risk inside the company itself. And so you might be able to pick up some great assets at a good value here if you b your time and find a good entry point. >> Yeah. I mean, as I said, like I I'm I like to be like a the human highlighter pen and say, you know, this is something that's coming to our attention, number one. Um, that you might not see in the headlines every day. Um, and what you make of it is a little bit up to you, right? There's a lot of pro pros and cons, but it's again, I would say it's something to keep your eye on. >> Okay? And the shorthand if you don't until we get this up on our website look just look at the area stock price there's definitely a two-way market on that Adam >> absolutely yep um and >> and and and part I would just say two things I mean there's other reasons BDC's may go down because if interest rates go down then demand for fixed income may just go down in general but I think the question in this whole private market world is is there are there looser underwriting standards than you have when you have a public security, >> right? >> And these two situations have what you have seen is that some concentrated lending to to these bankrupt counterparties, right? Um with with several firms having over hundred million dollars of exposure, that's that's a lot of money. So anyway, just uh something something to watch. >> Well, well, so sorry. I just I I've asked this question a couple times recently, so I'm glad you just put on the table. Do you have a a point of view on that yet of like is there a potential I word reckoning might be too extreme, but like you know, how big is the risk that we realize, oh my god, there was a lot more poor lending done in this private industry than we realized and that that could create a contagion at some point. Yeah, I I think the credit risks and I'm just sort of repeating from what I understand from their public representation. So I don't have like an insider's view on this at all is relatively well managed and diversified across the industry. So it's very hard for me to kind of see some kind of systemic risk at this point. But I I do think there's a governance risk and I'm just saying that again it's not the risk per se, it's Wall Street's perception of the risk. Yeah. >> And so that's what I'm kind of saying here is that um look in my heart, I should say I'm an ETF issuer and I love liquidity, right? But sometimes in the fixed income markets that the demand for liquidity will drive ETFs and and you know BDC's here to a discount. Um and that's really been a a buying opportunity but but it's some I don't want investors to be surprised by it. >> Okay, great point. >> Okay, so um I'm did I'm not even going to talk about that chart. All right. So, I think we're done for the main portion. Look at exploding AI token demand, right, as something to watch. Uh, and then whether OpenAI gets the financing to build out its compute. But it's it's a really interesting space. uh gold is part of a longer term trend I think and um if if there's a squeeze on private credit um liquid vehicles like BDC's may be the escape valve in a negative way but it might also be a buying opportunity. So those are the kind of three points uh that I thought were unique. Now um I want to do like a speed round um because I conscious of not being able to talk for four hours. Uh so maybe just to take um another five or 10 minutes and and and just flag some things that I thought were interesting, Adam, for as you and I catch up. So the the the benchmark as the headline of putting together an invest client's portfolio is kind of 6040 for moderate risk. And I'm just personally hearing a lot more people question the 40% Adam. >> Um and this is just one chart I use. This is in JP Morgan does guide to the markets which I know a lot of people are familiar with and the chart I forget the chart on the left but the chart on the right just shows that bonds don't always move opposite to stocks they can move in the same direction and >> we learned that in 2022 the hard >> right exactly um and and the temper tant taper tant temper tantrum the tariff temper tantrum of April, you know, the dollar didn't strengthen, it weakened and bonds rallied a little bit, but then they didn't rally that much. So, I would say, you know, people are questioning the 40%. I don't want to get into it too much. I just I just think that's kind of interesting because so much of us benchmark against that. The one slide I put together for people to think about is the total return from a bond. And we've been in a 40-year bull market for bonds. So the total return, which is the third column from the right, is uh you can see annualized people made a lot of money in the 80s and 90s, a little bit less so recently. Okay. More as interesting, you basically don't make any money from the appreciation of bonds over time. You really make it all in the yield. And so um anyways, I I think that's just something for people to think about. So bonds are yielding four or five%. That's what you're going to get. All right? Don't expect to get more than that. And and I think people are saying, "Wow, I'm getting like double that from stocks. Why do I own bonds again?" >> Um and gold has beaten bonds over almost every time period. Well, definitely over every time period over recent history. Now, >> why do I own bonds? I I'm I'm not I'm not trying to revisit 6040 single-handedly. I'm just hearing a lot of strategists and other people sort of saying let's have a discussion about 6040. Okay. Just let you know I'm hearing that too from a number of the capital managers that I I talked to on this channel. You know, some are developing a 603010 model. Some are doing some variance of that. But the the thing that they're taking away from is bonds. >> Exactly. I think exactly right. The starting point is stocks are fine in people's portfolios, right? It's the bond portion that people are taking money away from and that's often the biggest question um kind of in our industry. Um I am I'm going to skip a couple of slides here. Um uh and leave India for some other time. But I want to talk about Alibaba. Um and the question is um so a year ago uh talking to someone named Ab Adam Tagert I talked about my personal personal buying of Alibaba. Um, and one of the reasons was I thought the AI component was underappreciated and Alibaba is still cheap on that basis. And I just wanted to share with you something that I don't think is hardly known at all. Um, which is and this is actually original Vanc research which is I can't always say that. This is US electricity prices per kilowatt hour compared to China per kilowatt hour. Now, the labeling isn't really great here, but the blue line is China. Bottom line is the headline, US is twice the electricity costs of China. >> Wow. And their electrical um generation capacity is some multiple of ours too, >> exploding. And so I hate to like bum you out, but what if China became a cheaper place to do hyperscaler activity and that's an area that we could be competing with in the world. So, it's a little bit more Adam a pin to put in here to be thought provocative because I'm probably, you know, I don't want to spend a ton of time talking about it and I certainly am not doing some grand predictions here, but it's it's um ultimately economies are competing on the basis of the cost of electricity. Europe, the cost of electricity, right, have gone off the charts. They're now trying to deal with it. The Middle East is very cost competitive. Um, so anyway, I just think that it's it's sort of in in the background here. This was something helpful to to point out for people. >> Okay. Yeah. And and this this does add tailwinds to the AI energy trends that we were talking about earlier, right? Um trade meaning, you know, in the US, let's just look at the US lease. >> That is like a top sovereign priority, right? is to to build out our energy production grid so that a we can match China in production capacity but also not have this delta in the cost. >> Yeah. And then one last slide. You know, I I I can't stay away from the budget deficit, but I'm just going to get I'm just going to focus on one statistic. So, our budget federal budget uh fiscal years end at the end of September. And you've seen this slide for me before, at least the the column in the middle, where uh in fiscal uh 20 24, sorry, that was last year, right? Um our budget deficits was 6.4% of GDP. Simple math. We just printed fiscal 2025 and it dropped to 5.9% of GDP. So in prior quarters we've been talking about what will it be like for next year because this is a hybrid Biden Trump year. A lot of the spending that happened in Septe all of the spending in September, October, November, December and January was Biden spending. So a third of the year, this is a third Biden and let's call the twothirds Trump. So um this is just Jan has been optimistic that they're using a lot of tariff revenue and getting this deficit down as a percent of GDP. And what's happening is you saw revenue going up slightly about revenue and outlays went up about the same 300 billion but that's shrinking as a percent of GDP right so if the GDP is growing that 1.8 8 becomes less and less as a percent. >> Right. Right. >> So, I'm I'm still shooting for 4 and a half to 5 and a half% of GDP for this fiscal year that we just started a couple of days ago um on October 1st. What does that mean? It means less government stimulus for the economy which has been offset by AI compute spending. Um and it but it means better overall uh financial stability hopefully um as we crawl out from what I had been calling the reckoning of this last year. >> Yeah. And does that mean you think we'll largely get out of this? Okay. >> No, I'm just I'm just like we should celebrate this like we we snuck in, right? It was six and a half% or 6.4% 4% last year. And if you read a lot of headlines in in 2025 from the CBO and all that kind of stuff, they were like, "Trump's not shrinking the budget deficit. It's going up even further." Wrong. Right. And And is it largely wrong because the tariff revenues are offsetting things like the I mean, it just got passed, but things like the increased spending in the OBB. >> Yep. Yeah, I I can we can go through the the the details, but basically yes, and that's yes, tariffs are are are the most important part of this. Um, but you know, he listen, it's not talked about a lot and it it just flew by in the press like it barely got a headline. So, I just think that if you think about the big macro factors, this this the risk of crisis coming from DC is marginally less. and and I'm not celebrating or anything yet, but um it's headed the right direction. If we have a 4% growing uh economy next year and you know revenue and outlays stay in sync with each other, this number will come down again. >> Okay. Um All right. So, sorry, you were expecting what? Four and a half for fiscal year 2025. Uh, no. For next year, I never guessed it this year because I said it's a hybrid year and I I just had no idea. So, my I've always been talking about fiscal 2026 next >> fiscal. Okay. So for fiscal and I guess the administration would say hey yan if you like that you'll like the years even further because you know the passing of the one big beautiful bill we were frontloading a lot of the spending on it and that spending should diminish in future years um as a percent of GDP and we should start getting tailwinds from all the other policy stuff that we're doing and the economy should grow even more and that should narrow the deficit even further. Will that happen? Who knows? But that I think that's what you Yeah. >> I just think it's worth focusing on the scorecard, right? And and what we what we miss, again, this is why I love doing this outlooks with you, is sometimes the scorecard like flashes by so fast and we lose it and suddenly we're focusing on Taylor Swift's wedding or something else. >> No, I appreciate you doing that very much. If I can, let me just ask you one question on something you flew over and we can we can do the deeper dive next time, but you skipped over India. I know you've been pretty bullish about India um for the a number of the different quarterly outlooks we've done together. >> Um I don't follow it as nearly as close as you do, but I think Indian stocks have kind of taken it uh some of the winds come out of their sales over the past quarter. >> Um so just in a sentence or two, >> what are your thoughts here? Hey, temporary the party's going to continue again or is there something more material going on here? >> They've done the structural reforms that they need to and now the economy just needs to grow into them was my base case. I've got even more good news because they're passing a bankruptcy law reform which basically means that companies will go bankrupt in a more orderly and quicker manner which means more efficient capital usage and better risk um better risk management if you will in the economy which is fantastic. So yet another structural reform seems to be in the works. It's not done done, but it's almost done. I would say in terms of buying, um, you know, we don't quite, yes, markets are down. We don't quite have the full buy valuations that I would love. I would say we're close to the price to book being pretty attractive. PE not quite as much. So, um, if you don't own any, I'd certainly buy half a position. Um, but we don't have so much as to, you know, overweight. Okay, got it. Like I said, next time to the extent that that you know things are still somewhat similar um >> Oh, give a chance to really go deep into it. >> And because I love making predictions, um and I love having these recorded um I predict that the US will cut a tariff deal with India in the next month at 15% tariffs. >> Okay. All right. Um well, I I hope you're right first off, first and foremost. Um, I will say too, Yan, just I think we noted several of them in this this past hour, but I I think more often than not, your predictions have actually come pretty close to the mark. At least the ones that you've made here with me. You got a good track record, so hopefully you don't spoil it. >> Some good some good wins. Yep. >> All right. Well, look, um, Yan, fantastic as always. Um most important question um for folks that would like to um follow and track Vanx work and learn more about their family of funds where should they go? >> Yeah, our website has all our fund information uh van.com and uh I'm on I'm on X uh yanvan number three and I'm also on LinkedIn. I will share great great pods like some of some of Adam's pods. Um, so thanks. Yeah, thanks. And and uh appreciate again the platform. As you know, I did a shout out to you on on Josh Brown. >> You did. You are such a good friend of this this platform, >> which a lot of people have commented to me on. So, um, it's good. >> That's great, Yan. Well, look, thank you. And again, we'll continue doing this as as long as you've got the interest and the energy. It really is wonderful. Uh, and Yan, when I edit this, I'll put up the links to van.com as well as your social media handles there so folks know exactly where to go. But yeah, my friend, thanks so much. I know these um require a fair amount of time to put together. You know, I think you said they're they're helpful in helping you sort of structure your thinking for the next quarter, which is great, but I know that it's an expenditure of energy on your part. Really appreciate it. >> And and we'll get you the chart pack as well because I know that always comes up. So, >> yes, thank you. And yes, so folks, the chart pack will be available um in the Substack uh write up of this. It'll be to available for free to anybody. So, if you don't already sign up for our Substack, just go to thoughtfulmoney.com/newsletter. You can sign up for free >> and it's we'll get you latest tomorrow morning. You know, compliance always needs to polish every uh every vow. >> All right. Well, it should be ready by the time this video launches. But anyways, folks, please do me a favor. Thank Yan for just doing his usual bangup job here by hitting the like button, then clicking on the subscribe button below. Yan, my friend, again, I'm just super appreciative of this and look forward to doing this with you in a quarter. >> Great. Thanks, Adam. Good luck. >> All right, everybody else. Thanks so much for watching.
Jan van Eck: Q4 Macro & Market Outlook
Summary
Transcript
So, what am I going to talk about today? Number uh basically three things and then some uh extra goodies at the end. Number one, what is really going on in the AI market? Uh number two, gold has performed like crazy. All-time highs $4,000 an ounce this year. Did I miss it? And then thirdly, what's going on in the credit markets? Uh we've had some headlines recently and have some ideas on what investors should be looking at as that uh as that plays out. [Music] Welcome to thoughtful money. I'm its founder and your host Adam Tagert. When market uncertainties is as high as it is now, I often emphasize that the most useful people to interview are asset managers because they don't have the luxury of merely having an opinion on the road ahead. They have to commit capital to their convictions and be judged upon the results. Today, we've got the great fortune of having the return appearance of one of the most respected capital allocators in the business, Yan Vanek. Janna is CEO of VANC, an asset management firm with over a hundred billion dollars in assets under management, invested across its wide family of ETFs and funds, spending equity, bond, commodity, digital, and regional asset classes. As we've done the past several quarters now, Jan and I will spend the next hour discussing his Q4 macro and market outlooks, as well as where he sees the biggest opportunities for investors right now. Yan, thanks so much for joining us today. It's good to see you again, Adam. >> It is too. Um, I got to tell you, Yan, it is just such a pleasure and a privilege to have you come on for these quarterly outlooks. Um, I very much look forward to them. The audience does as well. Um, I've been having people ping me a lot recently saying, "Okay, isn't it time for Jan's outlook?" So, here we go, folks. We're going to we're going to scratch that itch for you. Um, so Yan, I will stop talking really quickly because I know you've provided um a chart deck as you have done in the past for these. So, we'll get to that as quickly as possible. I guess right before we do, if you had to pick a a word or a phrase to sort of describe the macro situation right now, what comes to mind? >> Happy. >> Happy. >> Okay. Like like >> you only give me one word. >> Healthy kind of happy or like a like a manic kind of happy. >> I would say happy with eyes wide open. Um there's so many things that are changing so fast. So that's the eyes wide open part. >> Okay. All right. Well, I could ask you a lot of questions about the why of that, but I think your slides are going to answer that better than uh any question I could ask. So, why don't we why don't we jump in your material. >> All right. Yeah, thanks for thanks for doing this and uh you know, my uh my friends and clients look forward to us doing this together as well. So, thank you for the opportunity, Adam. Okay. Um just a little bit uh by way of background um in case uh people haven't heard me before just a little quick reminder of the perspective that VANC brings and I bring to looking at the markets um you know we look at the same big drivers that other people do government spending uh monetary policy fed policy and technology trends I think what differentiates us uh is a little bit what I'd call our historical perspective uh which is that change can be more dramatic um and rapid than people might expect and an odd range of things can happen uh because they've really happened before in history. So that's number one. And secondly, we try not to focus too much on uh daily, monthly or weekly trends, but focus more on multi-year trends and see how they're playing out uh today. And then I would say that if we think the puck is headed a particular direction over a multi-year period, that gives us higher conviction. Meaning sometimes you can have uh a stronger belief in something is going to happen in the next five or 10 years rather than it's going to happen tomorrow because tomorrow is a little bit uncertain. So um and also just by way of background there there are four big crazy things happening in the world, but I'm not going to focus on on all of these today. One is just that we are just getting through the worst ever uh peaceime fiscal deficit in US history meaning uh low unemployment uh but high government spending. I think that AI is the second biggest technology that's hit America. Uh the the first being railroads I'll refer to them very briefly. Uh third that India is headed the direction of being a major consumer economy in the next 10 years. uh to the scale maybe not quite of China but close. And lastly, that we've had uh a dramatic uh deregulatory act in the United States uh payments markets this year, but not going to talk to it, but as as big as what FDR did with the banking system in the 1930s. So um I will now say if you have no more time to listen to me uh very quickly I've I haven't done this before Adam but in a nutshell um you know what do we think is happening um number one quickly that the fiscal deficit is probably getting better than people think um estimating four and a half to five and a half% for next year that AI and we'll talk a bit about this is supporting the megga cap profitability, the fact that the S&P continues to rise on a market cap basis and that the bigger continued to take share market cap share. So we said this a quarter ago that we wouldn't be surprised if they had a good quarter and that continued and and indeed it happened. Um if there's a cloud to worry about, it's probably some kind of growth scare that would come out of the employment markets because we have so many things going on there including tariffs and um you know AI displacement. Uh we think it's early for the ddollarization or gold trend. We'll talk about that. Um in India, I don't have to talk about that. And then we'll talk about the AI plays. So what am I going to talk about today? number uh basically three things and then some uh extra goodies at the end. Number one, what is really going on in the AI market. Uh number two, gold has performed like crazy. All-time highs $4,000 an ounce this year. Did I miss it? And then thirdly, what's going on in the credit markets? Uh we've had some headlines recently and have some ideas on what investors should be looking at as that uh as that plays out. Does that sound like an okay agenda? >> It does. I was going to tell you on I can tell you with confidence that those are all questions the thoughtful money audience is asking in real time right now. So very timely. >> Great. Well, so for the AI stuff, um I really want to talk about what exactly is going on um on the on the technology front before we get to earnings projections. So we've talked about Nvidia, but I want to talk about what is driving this huge demand for compute and the deal frenzy that we've seen. And what I'm going to propose is that the craziness of technology is the following that I call that third bullet there crazy mainframe math which is that token demand is exploding and tokens are sort of like I'll call them words or uh units of of knowledge in in AI world. So it's basically the amount of demand for AI is going up exponentially at the same time that the costs are coming down exponentially and the costs being twofold. Number one the efficiency of the LLMs large language models and number two the efficiency of the NVIDIA hardware. So we'll drill into that. So part of the reason that I said eyes wide open at the beginning, Adam, is that there's this crazy frenzy for compute demand and the variables are moving at unbelievable speeds. So we'll look at some charts there. Uh we've just get a refresh on what's happened to the beneficiaries of the hyperscaler demand for energy as a second bullet. And then I want to highlight um a new company I think is the the new Mag 8 company. And then I want to talk about gaming as another AI winner that emerged this quarter. So all right. So here's the crazy mainframe math. This is tokens consumed are going up 38 times year. And that's not just because we have a a weird starting point a year ago. It's going up like 20% a month and there's >> Hey J, I'm I'm sorry to interrupt but just so people understand you mentioned it briefly but token is that is that like the inputs on which we train AI and those are going up or is it like the queries like questions people are asking of AI? >> Oh no, no. Yeah, good point. Not queries. It's to think of it as words. I'm going to say words but there's more I think more knowledge in there than words per se. But if most of what you're doing is processing language or words, these are the these are the units of of compute and they break it down into what they call tokens. >> All right. But again, is are those like the inputs? Are those like the the words or the knowledge they're putting into the LLMs? >> Not no not per se because you could have a query that says um when did thoughtful money start, right? That's a several word query. But then it might have to consume a thousand words to figure out the answer to that. And that second component is token demand. >> Got it. Okay. So it's the I'm going to use the word input again, but it's it's it's the inputs that the AI is is is grabbing to try to answer your question. >> Correct. Right. It's what's its processing. I gave probably a pretty bad example just there because it was a single piece of information. But if you're trying to establish causality, like what makes a great thoughtful money podcast, right? Then then that requires lots of compute. And even if you defined what great was, right? A lot of tokens would be consumed in that uh in that. And and one of the things that's come out, there's two things that have emerged in the last quarter or two. one is that um some of these LLMs are capable of reasoning, right? So, not just spitting out an answer to a question, but exploring the thought process and showing users the logic that it used to get to an answer. >> So, that's one. And then second of all, um it's not just you and I that are asking queries of of these AI chats. It's software itself. Right. So a whole new source of uh demand for for AI processing is coming out of things like robots or just software that's being developed to answer questions in a more sophisticated manner. >> Okay. And on that and I don't want to I don't want to take us too far field on this but I have heard recently that you know we've trained these large language models on human generated content and we have kind of gotten to the end of that where we put about all the content that's ever been generated by humans into these models and so the next wave of of kind of information that these models are going to be trained on is going to be information that is data that's being generated by AIS themselves. >> Yeah. Yes. I mean that's a that's another dimension and I don't want to go too deep there but that that's yes again software is now we used to write software and it would process data now software is kind of rewriting itself. >> Yeah >> it's learning from the queries that it's doing. >> So I'm gonna avoid asking the question that everybody wants to know which is how long until it gets so smart it kills all of us. But we'll save that till the end of the presentation. Yeah, that's it's like it gets super dark. I'm I'm just I'm frankly not worried about that sitting here today. That's why I'm happy. So, um we we can we can talk about that maybe uh in in a deep dive. So, you've got demand for tokens exploding at 38 times a year, >> right? Just exploding. And then >> the costs of processing. So, I'm going to break this down into hardware and software. So, this is the hardware slide. Oh, I'm realizing now it's probably too small to read, but these are the different uh hardware solutions that Nvidia has offered. So, you can see Kepler is the one on the on the left and and how much how many tokens how much energy it took to process a token. >> That's unbelievable the savings there. And it's going down like Jensen Hong who I I had the fortune of seeing last week. It's going down at 90% a year. So he plans on releasing new compute platforms every year and he's saying that their costs are going down 90%. So that's why I'm saying this is a crazy equation, right? Demand is, you know, exploding and costs are imploding. And what he says that took me a long time to understand is he says, "I don't care if people give away my competitors give away their chips because those people will still buy mine." And the reason he's saying that is we have a energy shortage in the US. And if his chips are 90 or 30 times more efficient than even the competitor's chips from last year, people want to consume way less energy, right? for the same amount of token processing power. >> So, we're kind of on this hamster wheel, but it's it's an amazing one. Uh that that's why I'm saying eyes wide open that we have to keep our eye as these things develop. And then the third component here is I'll call it software. So, these are the large language models. And um on the left hand scale is just the cost of processing tokens in dollars. And and you can just get from the shape of the two colored uh lines and I'll explain what they are in a second that they're coming down and amazing. And these are I think logarithmic charts. So for those of you into math, they're falling really fast for the English majors in the audience. So you've got two cost components falling fast. the hardware which we just saw and then this is the software. Now the two different lines are basically describing two types of LLMs. The more sophisticated ones are the ones in orange. So they score much higher on kind of if you will benchmarks, intelligence benchmarks of these large language models. Um and that's what MMLU stands for. So the ones that are that you and I would want to use are in orange, but the kind of lower quality ones costs are falling even faster. >> Even faster. But still the orange ones are still dropping from I don't know 50 to to 75. Just massive drops. >> And and this is why like I think hopefully it's pretty clear that if Deep Seek comes out with a model that drops it a lot, there's going to be Whoa. Maybe suddenly demand for compute goes down, right? But if the MA models start choking and don't uh improve on their efficiency, then all the efficiency has to come on the hardware side. >> Okay, >> so my mainly my main point is there's two this is the real technology behind what's going on and and it's pretty incredible. Now uh so who cares Yan? Okay. What I would say is what we are seeing and here's to Nvidia. Uh we'll just use this as the benchmark for all this. What we've seen over the summer, Adam, is a huge run to lock down compute. All the different deals that are happening, and I have a chart on that later, but Open AI wants high bandwidth memory. They want chips. Nvidia wants chips. Everyone wants more compute power because basically I think the industry realizes that demand is exploding so much that there will be a shortage of compute and what's odd is I think Wall Street is missing it or my colleagues think Wall Street is missing it. So you know I've gone luc uh cold and hot on Nvidia stock. um you know a year ago July I was a little bit cold on it uh because of the valuations now we think Nvidia is reporting that their demand is well clear visible into 2027 so two more years of of earnings and their their valuation their price to earnings ratio is I would say at the high end of a normal company but not of a company that's growing 40 to 50% a year which is what they're talking about. So, you know, if that if those growth rates are are achieved, then their forward pees should go up as well. So, the stock price should be getting a lift from higher earnings and a higher multiple. What we worry about is three to four years from now when the multiple comes down. But in the short term, we're facing a compute shortage. >> Okay. Um, I I totally get that and it's interesting. So, like you, Yan, um, and probably even more so over the past couple years, I've had trouble wrapping my brain around how a company could could be worth so much and worth so much so quickly. And I I I'm I'm kind of taking from your presentation here like I I I'll often talk about a trillion like when we talk about our trillions of debt and and when you actually try to tell somebody how much a trillion is it's a number so vast our brains really can't fully grasp it right um and to a certain extent I hear you saying that the the the the dynamics that are in play in this AI market are at at a such a scale and moving at such a speed that it is really hard for most people to really truly understand it. And and because of all this, you know, there's still a fair amount of demand left here that makes Nvidia's nosebleleed stock price maybe fairly valued here, maybe even a little little cheap right now relatively. >> Yeah. Well, I wanna I also wanna maybe this is just something that everyone else knew but Yan, but Nvidia is is in a semiconductor ETF. It's a chip manufacturer, but it's really not anymore, right? It's gone from just a GPO card manufacturer to now it is the mainframe of today, right? They are supplying the whole compute platform. And what's different about that is you know they are the heart of the ecosystem. They are the ones cutting deals with memory suppliers or CPU with the different bits and components, but they through their CUDA software basically want it. They want everyone to live in an Nvidia world. And we and we do, which means that it's got more competitive modes than simply being it can't beclipsed just if someone comes out with a with a faster process. There is my point. They've got a whole eco competitors have a whole ecosystem. And may maybe that's just Jensen Wong's sales pitch that I'm repeating, but all the deals that we saw over the summer, I think support that thesis. >> Okay. Um All right. I'm I'm resisting about asking you about an AI bubble and all that stuff. We we'll save those questions for later to let you go through this. >> I'm trying to I'm trying to address it, right? And and we'll get Yeah. We'll have a couple more slides, but part of what I'm saying is demand is going up for something like we've hardly ever seen in our life. Yeah, it may not be a bubble. It may just be a super big demand surge that merits all this. >> I I think so. I mean, that that's what I would just say. And and it's easy. Let's keep your eyes on to part of part of the hope of these uh these quarterly sessions, Adam, is investors know what to focus on. What I would focus on if I had to focus on one thing is token demand, right? And if that's not going up exponentially, then I'm really worried. >> Really worried. Okay. So, two two quick questions out of this and I'll let you resume. Um I guess let's start on token demand. This kind of gets to the um the AI bubble question I I wanted to ask you, which is I totally get in this world why Nvidia is the darling, right? Because because of all this, we need more computer. You know, we we we got to invest in the platform to make all this happen, right? And largely that's happening on Nvidia technology. Um now at the end of the day, why does this matter? Right? We AI doesn't have a ton of value if we just make Nvidia a really rich company. It matters because we're doing all sorts of amazing new things that create a whole bunch of incrementally new value for us using AI. And to date, there hasn't been a ton of that coming out in terms of like incremental revenue or profit on the income statements of corporate America. And so the question I have for you is are we beginning to to are we beginning to see some of that or is this still being sort of all taken on faith that if we feed Nvidia enough, it'll support the AI valuations that all the stocks in the AI ecosystem have? >> Yeah. Um, so I would just sort of say yes. And that's sort of the the counter. What I would say is the at the super high level, the US is a consumer economy, not a manufacturing economy. And the consumer is now in a different regime of interfacing. And I'll explain this in in the next couple of slides >> with what it does, entertainment, purchasing, working, right? All of that. And the economics are well spoken for, but I'll get that to to that in in a second. So, uh it's not that this is a toy that is unhinged from revenue, but let me give you some nuance there because there is there is one weakness that I want to want to focus on. >> Okay. Um, so, so let let's just do a side trip to what we talked about three months ago, six months ago, which is there's going to be a lot of demand for electricity and nuclear and some of these other things are going to win. I mean, this is exceeded my wildest expectations. Nuclear stocks are up 90%. Olo, which has no, forget profits, has almost no revenue, is up seven times this year. >> They don't make any energy. They they haven't produced one electron yet. >> I I I am I am check me you know as my dad used to say take me out coach. I don't know what game this is. I I I I can't say sell because I believe in this as a multi-year trend but I I would more I don't know. I just don't know what to do. Take profits. Um don't don't take get out of the position. But this is this is really um I would say you know excessive um so uh and and just a a a consequence of that that I want to touch on briefly is just that and we've seen this bubble up already which is the costs of electricity if we have all this demand Adam the cost for electricities are going to go up and so you can see here that the cost of electricity in turquoise is above CPI high. Uh this is a very regressive uh fact, meaning that lower income people are much more affected by an increase in the cost of electricity. And um you know, we could spend a whole hour talking about it, but I just wanted to flag this uh because there were different states are dealing with this in different ways and I think there could be an entire reshuffleling of how electricity gets done, but at least I just wanted to highlight this. So, so real quick, Yan. Um, so I was thinking about this, asking you about this before you brought up those charts showing that the Nvidia uh technology is becoming more and more efficient in terms of the energy needed uh to to process tokens. >> Yeah. >> So, um I guess my question is is how how much do you expect this to spike given that there will be technology somewhat riding the rescue here? So we're we're going to be placing a lot more demands on AI queries going forward obviously as it continues to grow, but they will require less and less energy per query. So does that does that moderate the demand it's going to place on electrons or >> No, not not in the short term. This is part of the whole basically compute shortage and you know shortages and gas turbines and you know lack of nuclear. I mean we need all of the above. Uh so uh I I don't that's a good question. I don't have a a simple answer on cost projections. You know there's a difference between what consumers get charged which is this number and um you know what businesses get charged, >> right? But but as as there is increasing demand just for the overall electrons there are there people are seeing their bills creep up. We're seeing it all over. Right. I put out a tweet out I know a couple days ago asking if there was a betting market yet on Poly Market for when the first data center was going to get burned down by angry local residents. >> Oh my go Okay, that's a little uh it's a little hostile, but >> No, I know. Little little dark, but if this line keeps going up at some point there, you're going to get some real push back, right? >> No, no, no. I mean, sorry, I'm with you. Just not as dark. >> I know. which is actually the takeaway is less about AI here than about politics because there this is a whole different thing but there's a massive difference in American sentiment and polling on a range of topics postcoid >> so I don't want to get metaphysical but this I believe is one of the only one of the many ingredients to that like why do people feel so much pain in their pocketbook book relative to the quote unquote inflation of 2.3% and this is I think one of them. >> Okay. All right. Well, very important and we'll keep tracking this I'm sure in your future appearances. Real quick, I want to note that when you presented last quarter, I did own some Oaklo at that point in time. Um, but after your presentation, I did buy some of the um the Vanek uranium ETF Trust >> and that's up 76% since your last appearance here. So >> nice. Nice. You >> want to take a victory lap on that? Take a victory lap. >> Well, I don't know. I I I Yeah. I mean, that's why I love multi-year trends, you know, and hate them, right? Because you knew the puck was going there over several years, but you didn't know how much was going to get priced in in 2025. >> Yeah. >> Uh but that's a that's like a whole other thing. Okay. So, that that's kind of reviewing stuff that we've talked about before. Now, the two I'm going to give you two winners um of of AI. And again, um so I I know this is sort of stating the obvious, but I think we're kind of missing I was kind of missing how powerful Open AI is. And so I'm going to argue that um it's not only a MAG 8 company, it is at the doorstep or in fact on a daily basis competing with some of the other MAG7 companies and winning. So let's just start why I'm starting this. OpenAI I guess was launched I forget 2022 I think and it saw the fastest user adoption of anything. And so if you look at monthly active users on the right, which is the the main thing on this chart, they are really the leader in monthly engagement with AI chat. >> And I think it's amazing to think that even Google hasn't been able to get more uh more users on Gemini, right? So suddenly they are in interfacing directly with the US consumer. To your question of a couple of minutes ago, Mh. >> Um, and and we'll anyway, we'll we'll talk about that in a second, but they are uh I think they're just underappreciated. They're like our friendly little chat tool, but Sam Alman is literally competing against Gemini, I would say, and Amazon for sure. So, let's dig into that. Um, just another way of showing how chat is winning. This is year-on-year website visits relative to the Giants. like it's beating the giants. So, Google is on the far left, no growth. YouTube, no growth. Facebook a loss. Uh Instagram slight growth. Chat of course stands out. And you can see basically no one else is has a lot of um momentum. Maybe Reddit is positive. All right, let's keep going. Chat is powerful just the way Google is powerful if it's the first step for the US consumer. And so in terms of AI chats referring traffic to the sites on the left, which is where the American consumer goes, right? Amazon, Booking, Expedia, Etsy, Zillow, Carvana, look at chat market share. >> Yeah, that's really f that that's amazing. Um, it's funny because I I used to work at Yahoo and in the early early days when I didn't I had yet to work there. Yahoo was sort of synonymous for the internet. You know, there were other websites out there, but everybody just thought of Yahoo and you started your your whole journey there. It seems like chat GPT is that for AI right now. >> It's it's incredible. It's incredible. If you and I had been betting on Poly Market, would someone challenge Google for search traffic or AI chat traffic? I I think it would, you know, a long long long odd, >> right? So, Y, I got to ask then, how come Google's stock price isn't getting beaten up over this? This is threatening Google's, you know, cash cow monopoly. >> Well, I think they have a a whole range of businesses, number one, and number two, they let's just go back here. They're number two. They, you know, this may not be a winner take all. I actually I don't think it's going to be a winner take all, but we can talk let me let me try to explain that in a second. But that's that's not bad, right? Plus, they have so many other businesses. But let's talk about Okay, so this is just another way of showing the same thing. I'm sure you've probably never heard of these sites, right? Move, Zupla, and OTM, but they're basically real estate sites in the UK. And you can see that a whole bunch of their, well, I mean, maybe you can call that a small number. I don't know. 17% of their traffic is now coming through chat. So it's the same it's the same point which is um okay so now let me let me try and experiment here. I think that AI chats offer different types of services and I want to kind of differentiate them. One is like Instagram or Facebook where you kind of hang out with the people that you agree with and chat tries to basically please you and your experience is what makes chat will give you uh data and facts that make you happy that will bend you from the truth. >> Yeah. It it enriches your echo chamber. It it reinforces it, right? It reinforces your echo chamber. Exactly. And I I was listening to a professor uh talk about how like explaining how chat would um like it's like there's two e in beagle and the user kept saying no there's three and the chat would say oh I'm sorry you're right and it's just it's it's just and that's because they weren't appreciating yes there is a dimension of chat that's just trying to make you happy right that's number one number two there's a different type of chat I'll call that the nonpleasing where the you're looking for accuracy is the value, right? And and of of course you need to have accurate data, but if you're at work, you want or if you're answering a customer inquiry for commercial purposes, you don't want you want to sure you want to please the customer, but you need to give them an accurate answer, especially when you're in the health care fields or the finance fields, right? And also some of us just personally want accurate answers not you know not biased answers. So I call that a second entire use uh use case and then third there's there's shopping and then fourth there's gaming and and we'll talk about that >> but let's just dwell on shopping for example if open AI can integrate with Shopify they can basically compete against Amazon. So again here this is open AI is a mag 8 company that's competing with the other ones. I mean, what's Amazon's moat against shopping if someone can access all goods and services through Shopify? And that's why Shopify stock has gone berserk over the summer. Uh because it suddenly got effectively it can get around the Google not or not get around entirely, but it has another source of customers which is open AI. >> Interesting. So yeah, you can do the same thing you do on Amazon, which is search for a product, maybe have the product recommended to you based upon what it knows of you, too, and then you can actually transact. Um, there it's just AI plus Shopify versus having it happen all on Amazon. >> Exact. Literally, Shopify provides the payment mechanism and the logistics to deliver the product to you. It won't come in a Prime Amazon truck, right? But it'll still it'll still get to you. >> Yeah. Does does shop does does chat have an advantage on the agentic side of things over an Amazon? So, in other words, you can say, "Look, find me a you know, find me a a nutritious health bar and then buy it for me." And it just it goes off and does it all for you. >> Yeah, I'm not expert enough to know which which platform is the best for that right now, but they're all my main point is they're competing against each other. Mhm. >> And then if you go back to this commercial application here under research app chat, the winner is Open AI through their deal with Microsoft, right? Most companies work either in a Microsoft ecosystem or maybe a Google ecosystem. That's it. And they've won. They're work they're partners with Microsoft, right? So they're the leading solution. And the other thing that's happening now is that um as you and I use chat more and more, we're starting to they're starting to emerge switching costs between the different platforms, right? Because the more we query and use one chat, the more it knows who we are and the kind of answers that we're looking for. And I would argue either that's under the social or pleaser chat or the researcher chat. It's learning more about us, right? If you upload your calendar, your contacts, all that kind of stuff like in Microsoft Copilot, suddenly that life is really way more efficient for you as a user. So over time, these eyeballs are more that's why everyone's fighting for market share right now. >> Okay. And I'm going to ask a question that you might be asking answering later on, but why does not why does this not become a winner take all ecosystem, right? where the smartest AI that can come up with the best answers and do the most for me, then the network effect won't kick in and everybody wants to use that and then that makes that even better. And then why would I want to use the number two version? >> Yeah, I I don't know. There's a scale definitely a scale advantage that we've talked about before, right? And and let me let me actually let me jump ahead um and then I'll move backwards because this is such a stunning chart and I I stole this and steal it but uh I borrowed it from a firm called New Edge. So give them credit. Um >> Cameron Dawson's firm. >> Yeah, exactly. And as right here you see what happened is that retrenchment from the MAG7 in the 2022 time period in terms of their earnings. This is their earnings growth. Mhm. >> But as soon as AI kicked in, look at this. It's like a straight line to the sky. And so this is what I think is describing the stock market these days. So I'm not sure that all of them can't win against the rest of the economy. At least they're showing a lot of earnings growth partially because they're cutting their costs. So if you as you have that image in your mind, let me let me just go back and try to answer that. This is what Wall Street is all in a tizzy about, right? And part of what I try to do sitting here in Midtown Manhattan and dealing with our active investment approach, highlight, you know, kind of what I think they're missing. And what I said before is I think people are underestimating the compute shortage and and how earnings can go up over the next two years. But this sort of web of inter uh of of consumer buying and interin investment I think overconfuses it to me two there's two things and I've said them before but look at this chart. Number one, we're living in Nvidia's hardware world, right? Which is why Nvidia is cutting deals with memory suppliers, high high bandwidth memory suppliers, CPU suppliers, right? They're they're of course wanting to deal with all the the hypercomp hyperscaler compute platforms. So we're living in Nvidia's world and we're also open AI is building up a comp a compute infrastructure to compete against Google Gemini and every other Amazon and Meta against these others. So they are they are the the the risk you asked before like where's the revenue where the revenue question it can be very much narrowed down to where's open AAI's revenue because they're only printing 20 billion of revenue this year 40 billion next year and maybe a hundred billion. So they're everyone else has got revenue, Adam, to pay for their compute, right? Amazon's got AWS. Everyone has billions and billions of dollars of revenue. In fact, they haven't even been borrowing to build out their compute. They had to build it anyway. So they were spending money on computers anyway. They're just this is just the new Nvidia regime that they're switching over to. But Open AI is the one question I would say. So that's the I would have a a completely different takeaway from this chart. Not that it's interconnected, but that Open AAI is the one firm, you know, that doesn't necessarily have the revenue to pay for it. >> Okay. And I'm sorry, what what what scale did you say their current revenues are on? >> Uh, so I mean, as opposed to hundreds of billions of revenue, they have maybe 20 this year. >> Okay. >> And like 50 or 60 next year and then 100 billion the year afterwards. So they're borrowing a lot of money and raising a lot of money to build this compute platform. But the dream is huge, >> right? >> Right. Why are they getting that money? Because they're the winner so far. >> Because the potential prize is so large. Yeah. >> Well, they're winning. They have the consumer, right? The question is, can they monetize the consumer? But they do have the consumer. So, I think I'm just trying to reframe the the whole AI debate here. And partly because Open AI is not a public company, I think we're kind of missing a little bit what's going on. >> Yeah. And I was going to ask about that. Do we do we know about do they have plans to go public soon? I mean, I know that that when they initially announced thinking about it, Elon had a lot of huge issues with that, but >> I I I there's no no announcement at all on that front. >> Okay. All right. >> So, anyway, I'm sorry. I gota I got to go fast here. Um to finish the AI section because I know you're gonna we want to limit this. So the last winner, so the one winner that kind of emerged to me that's very clear that you want to watch is OpenAI and can they get their financing because that's the weak link in that whole web and then the second winner is gaming. And so very quickly the biggest we had the biggest take private in the history of finan capital markets that happened in the last quarter. I love like kind of highlighting some of these things that have happened Adam because things are happening so fast it's hard to kind of we don't even have time to think about the significance but I think what these investors are saying is gaming is a huge AI winner. So number one the video experience is going to be way better and number two the costs of programming these games and doing the QA quality assurance is dropping dramatically. So I think what they were saying by taking this company private is that the public markets are missing that gaming is a big AI winner. >> Okay. And that makes sense. We're we're entering into the AI is going to help us sort of enter into the era of Ready Player One where the the gaming experience just becomes so amazing and immersive, but the cost of providing that unbel you know much better experience is getting driven down at the same time by the technology. So, if you can own one of these companies, you're going to ride that wave. >> I think that's what this headline tells us. So, um we're finishing up the the AI story. So, who cares? Yan saying number one, focus on tokens. Number two, focus on whether Open AI uh can, you know, can raise the money to pay for the compute platform. uh and also watch it try to compete with Amazon and and you know frankly a lot of the other mag seven companies because that that's really a quite interesting thing to watch. So that those are the kind of in in a nutshell the two things I would say to look at as we watch this and also I mentioned that we think that the the stock market is underappreciating the compute shortage that's happening from this huge um you know demand for tokens. So what does that mean? we see the MAG7 earnings continue to explode. I like to say this is a chart of US stock market history. I love the railroads. Um and you can see how big the railroads became as a percent of the market. So all I'm saying is the growth in the MAG 87 um uh is not unprecedented in our in our financial market history and we don't have time to go down that rabbit hole but I just want to say it's not unprecedented. And then lastly, the >> sorry just back to that chart though. The key takeaway from that is you would expect >> the share of this chart that's made up of the dark and light blue. Sorry, the the >> uh the transports are even darker. So I don't want to confuse people, but the Yeah, the two blues near the bottom there are going to continue to get bigger as a percentage of the pie. >> Very well could be. Yeah. Um and yeah likely I would say and then the question is okay when does the earnings growth of the rest of the market catch up with these juggernauts and um you know some some people are looking at 2026 estimated earnings and they say okay the mag 7 here is expected to be 13% uh earnings growth uh small cap could be 22% %. >> Wow. >> So maybe there's other parts of the market, you know, Adam, people have been holding their breath for this for years or at least the last two years, but it's it's, you know, it's conceivable. I mean, value doesn't look like it's going to be um surpassing the mag seven in growth, but there anyway, this is I think the thing to watch is, right, does the market smooth out a little bit? And it's you saw a lot of that in Q3. So, um, anyway, so that's that's the market as a whole. I'm happy. >> Okay. And I will say that's consistent with recent, uh, experts in the channel like Mike Canterowitz from Piper Sandler and Anna Wong um, becoming a bit more optimistic about the economy next year and that the the profits are going to start broadening beyond just the the the big tech sector. >> Yeah. So, um, okay, I promised three topics and then a speed round. So, I've used up a lot of time here, uh, to talk about AI. I'm going to be pretty quick on gold. So, gold, um, the gold price here hitting all-time highs, well above 30 and 200 day moving averages on a technical basis. That means it's got no support. Um, but my perspective, and I don't have a target price. have to work on that for a future quarter. Adam is that we've got, you know, four things happening. Uh, and and I'll run through them very quickly. Uh, one is just the emerging markets are becoming wealthier, so they're just buying more gold, and that trend is not going to change. >> Number two, I would argue this is, and this is a long discussion, but military conflicts or the or the the threat or visibility of them are actually up. Now I know the Middle East has calmed down over the last couple of days, but I think the Ukraine Russia war is deepening and China by um stopping the export of rare earth and magnets has really told the US look your military needs my magnets. And I think that's caused more uncertainty uncertainty in the world infla what I call the inflation risk. I'll show you that, you know, maybe the Fed cuts rates too fast. Or put another way, Trump puts too much pressure on the Fred Fed and there's Fed independence issues. Um, again, I'm not going to dwell on all of these, but my main point is these things are not going away. And then the last one is budget problems with France and Japan causing government changes, but prime minister changes and turmoil. It's this I just don't see any of these things stopping. So just to to you know what I'm going to save some time when gold has gone into different regimes the prices has gone parabolic in an and and so yeah 4,000 feels high but it's gone up 20 times in past gold cycles. So, you know, we can again talk about that more some other time. All this >> just just so you know, Yan, I mean, take the time you need here. I'm I'm not backended. If you are, I understand, but I don't want you to cut things that are important to you. >> Okay. All right. Well, that's good to know. Sorry. Um, so so the Zoom won't go off in an hour. Um, yeah. I mean, I don't know if you have any questions. I think everyone's familiar with the gold bull market of the >> 70s. I I do have one question. So, you talked about the inflation risk there. Um, so you seem pretty sanguin on the economy for 2026. Um, if the Fed does continue cutting, do you have any worries about it restoing inflation because it's cutting into a growing economy? >> Well, uh, services inflation's still high and you know, sure goods inflation has come down, right? But, uh, this, you know, it's I don't know, Adam, but my point is the markets were are worried about this. >> And um I would say that it's interesting Wall Street split, right? Part of what I like to do is just to tell people who don't sit in the middle of portfolio managers all day long. Half half the room thinks the Fed should have been cutting that inflation's low and what's the problem? And the other is like no, they don't need to move. The economy feels like it's got a lot of momentum. The financial markets are frothy almost, right? We have IPOs. Um, you know, we have profitless companies like Olo going up seven times and the Fed doesn't need to cut. So, um, my my main point is uh the risk is there and um and and I think gold is pricing that in. >> Okay. >> So, all these things we'll be talking about I think in the future. uh the manifestation. So this is foreign central bank buying. So the the the context here is when central banks were selling in the dark blue line on the left. This is developed markets meaning Europe and not the United States but Europe let's call it and then as uh since the financial crisis in 209 the rest of the world said you know that US banking system isn't quite as stable as I thought it was. let me like start tucking some gold away. And frankly, I'm rich, so I can start tucking some gold away. And that's what started happening. And then in 2022, after the US froze Russia's uh assets overseas, they said, you know what, I'm even a little bit less confident um on what the United States is going to do with my money. I'm going to buy even more gold. Right? And so that and they can afford to. So, I just see this, like I said, I see this as a multi-deade trend. I see no reason for this to change. And so, that's why um, you know, gold, you know, gold's a beneficiary of this. >> So, and if you measure gold, now this is a little fake because gold is marked to market. Um and and so obviously the the bull market in gold has made that dark blue line go up, but it it's kind of noticeable, right, that now uh central banks own more gold than they own US treasuries. So this chart's been floating around over the last month or so. Um and but I but I think it's I think it's significant. I think >> notable >> gold gold has now reemerged. of a new regime as the world's leading currency or store of value. It's not the yen. It's not the Deutsch, sorry, the euro. I always say that I I I like I I do that informally to call the euro the Deutsch mark just to make fun of the Europeans. Um it's not the remman, it's not the rupee, right? It's it's um it's going to be gold um across those different regimes. So anyway, we can talk about more some other time. The other thing I like to point out, >> this was the next question I was going to ask you. So, great anticipation. >> No one in the US cares. I mean, I know that some of your listeners are passionate um and in following the gold market and they've benefited a lot and kudos to you and them. So, but you I you are a outlier. Most Americans are not do not care. And uh you can see the purple just shows the prices going up. But you can see in the gray line that as a percent of total investable assets measured by ETFs and mutual funds, gold funds are not growing. >> And so sorry, we we're at sorry that that's the the total assets of the gold funds. Okay. >> Correct. As a percent of total. >> Okay. All right. Um uh so is this Yan a tailwind to to be that you're anticipating here that eventually the the US investor will wake up and want to buy gold? I mean what's funny is is is uh you know I guess it took $4,000 an ounce for the mainstream financial media to start using the term basement trade. I mean, and to me that is the the financial media like finally waking up a little bit to this, you know, maybe cracking an eye open. But do you expect at some point the gold trade to get a nice tailwind as the US investor says, "Okay, look, I got to be in this thing." >> Only when the S&P goes down significantly. Otherwise, everyone is living in this world of I'm getting wealthier. All investors are getting wealthier. Unfortunately, half of America doesn't invest, but uh investors are just they're loving the S&P, right? What's not to like? I mean, if you're in the financial services industry, you literally don't have to do anything and your revenues are going up 10, 15% a year. It's crazy. >> So, no, I I I this could take years, Adam. I'm not holding my breath. Are you? >> I have no idea the timing of it. So, I value your your point of view there. And you're you're not the first to say that which is it's going to it's going to take the you know basically the status quo to stumble before people start thinking about that. >> Yeah. Okay. So not too late to buy gold um is the short version. So we covered AI, we talked about what to look for. We talked about gold. I don't think it's early because these trends are multi-year trends. >> And the third topic is just the private credit markets. >> Yep. And sir, real quick before you get there, just just to make it practical for folks. That said, you you showed the the chart of gold being pretty far above its moving averages. I am guessing you would not be shocked to see a pullback in gold in the near term, but that would be more or less sort of a a buy the dip moment. Correct. >> Yeah, I mean it could correct significantly. I mean, look, you're looking at this 200 day moving average at what 3,00 3,300 3,400. I can it could easily drop $700 an ounce easily. Um but I think the long-term trend is going higher. Yeah, great point. >> Yep. Just wanted to make sure folks know that it could go down before it goes up. >> Right. So, um let's just talk about the fixed income markets. And I basically again have a simple point is I have something for you to look at uh if pressures um develop in in the credit markets and I would argue so far they really haven't been u so um I'll get there in in two slides what to look at but number one is because so much lending since the financial crisis here in 2010 has happened in the private markets the rest of us have no idea what's going on there, right? So, that's a little bit why I wanted to talk about it. We know that private credit funds and managers like Apollos and Aries have taken in billions and billions of dollars. Um, and there have been recent events over the last month or so, Adam, two basic defaults, one with a an auto lender and the other with an auto supplier, um, that are making headlines. So I wanted to say okay what what is h what what what should we be thinking of as investors and so another another thing that shows this concern is that Aries which was involved with one of the credits is a you know private credit manager this is the uh this is the management company so it manages funds then invest in these private markets and you can say it's it's had quite a fall here recently which has gotten some headlines and um we manage the largest uh BDC uh ETF and BDC's are basically uh funds that invest in private credit. And so you can see here that um they're the price of these BDCs, Adam, trades above and below what they're supposedly worth. And what they're supposedly worth is their book value. So supposedly, you know, they're very smart accountants and everything looking at the financial statements of the what they're these these funds are lending to and saying, "Okay, that's the book value." And sometimes these these funds are very popular. You can see that people will pay over 10% above what they're worth because the yields are attractive. And sometimes people freak out like during the COVID crisis or other kinds of bare markets and bonds like uh 2022, they'll trade at a 20% discount. So I basically have this theory that you can't redeem from these private credit funds, Adam. It's right. That's the kind of structure that they have. >> Yep. >> So there's Jan's theory is BDC's could be the escape valve. So if you own both private credit and BDC's and you suddenly became concerned about defaults, you would sell you can sell your BDC's in the market every single day. >> Correct? >> And they have fallen a lot in the last couple of weeks like 10%. I know maybe to you and me that doesn't look like a lot. That feels very painful to BDC investors. Mhm. >> And my basic theory is that this could easily go to a 20% discount um and because this or even maybe more if there were some kind of sudden concern about credit um in the in the markets. And so um anyway, that's my that's my thing to look for is look at these BDC prices. I'm I'm not try I'm obviously making my our shareholders unhappy that own our BISDY fund. I guess or from a perverse perspective, giving them a target price for when it might be attractive to buy, which would be at a 20 or 25% uh discount to uh to book value. So, we're looking to put this chart, I mean, I'll share these charts through you, um, Adam, as always, but we're looking to put this information on our website because I think it's something to to look at because there aren't many escape vows. you know, people can't redeem from their private credit funds except for on a quarterly limited basis. >> So, um I totally understand that. So, this this is interesting. So um one just in general you're sort of raising concerns on a topic that's getting increasingly discussed of late which is you know private equity sorry private credit um they both have but private credit um has sort of really exploded over the past 5 10 years and it's not regulated. It's much more opaque than the public lending space. And so there are a lot of questions of are these guys making good loans. You know, we we kind of hope they are, but we don't really know, right? We're not going to know until the bodies start floating to the surface. And then we just had and first brands go out and and now people are beginning to get a little scared like, oh, whoa, maybe the maybe these, you know, these credit books are a lot worse than we were we were pricing in. Right. >> Yeah. So, you know, you're saying, hey, the the the escape valve, if you're an investor in that space and you're nervous is if you own some of these BDCs, yeah, you can sell them real quick. They're liquid, right? >> So, it it pushes the discount down here. So, part of what you're saying is a warning of like, hey, if you own some of these BDC's, be careful. They might get caught in this this selling pressure. Maybe you're telling people if they're real aggressive and want to try to short something, this is a short candidate to go look at. But um a lot of the viewers of this channel, Yan, they first heard about BDC's from a fellow named um Stephen Bavaria who's been on the channel a couple of times who builds these income investing portfolios. And BDC's are part of it, right? Um and one of the reasons why Stephen likes to do um income investing is because it's it's all debt. And at the end of the day, debt is a contract, right? And he's like, "The nice part about it is you only have the existential risk of the company to worry about." Like the company is going to pay its debt as long as it's still alive. You just have to hope the company's going to stay alive, right? Whereas if you're buying a stock, you have the existential risk, but then you have a whole bunch of other risks on top of that, right? And so what I'm where I'm getting here is there might be an opportunity here, as you're saying, Yan, where the market kind of freaks out. these things go on sale, but they're go they then get to a place where they're priced really attractively versus the true risk inside the company itself. And so you might be able to pick up some great assets at a good value here if you b your time and find a good entry point. >> Yeah. I mean, as I said, like I I'm I like to be like a the human highlighter pen and say, you know, this is something that's coming to our attention, number one. Um, that you might not see in the headlines every day. Um, and what you make of it is a little bit up to you, right? There's a lot of pro pros and cons, but it's again, I would say it's something to keep your eye on. >> Okay? And the shorthand if you don't until we get this up on our website look just look at the area stock price there's definitely a two-way market on that Adam >> absolutely yep um and >> and and and part I would just say two things I mean there's other reasons BDC's may go down because if interest rates go down then demand for fixed income may just go down in general but I think the question in this whole private market world is is there are there looser underwriting standards than you have when you have a public security, >> right? >> And these two situations have what you have seen is that some concentrated lending to to these bankrupt counterparties, right? Um with with several firms having over hundred million dollars of exposure, that's that's a lot of money. So anyway, just uh something something to watch. >> Well, well, so sorry. I just I I've asked this question a couple times recently, so I'm glad you just put on the table. Do you have a a point of view on that yet of like is there a potential I word reckoning might be too extreme, but like you know, how big is the risk that we realize, oh my god, there was a lot more poor lending done in this private industry than we realized and that that could create a contagion at some point. Yeah, I I think the credit risks and I'm just sort of repeating from what I understand from their public representation. So I don't have like an insider's view on this at all is relatively well managed and diversified across the industry. So it's very hard for me to kind of see some kind of systemic risk at this point. But I I do think there's a governance risk and I'm just saying that again it's not the risk per se, it's Wall Street's perception of the risk. Yeah. >> And so that's what I'm kind of saying here is that um look in my heart, I should say I'm an ETF issuer and I love liquidity, right? But sometimes in the fixed income markets that the demand for liquidity will drive ETFs and and you know BDC's here to a discount. Um and that's really been a a buying opportunity but but it's some I don't want investors to be surprised by it. >> Okay, great point. >> Okay, so um I'm did I'm not even going to talk about that chart. All right. So, I think we're done for the main portion. Look at exploding AI token demand, right, as something to watch. Uh, and then whether OpenAI gets the financing to build out its compute. But it's it's a really interesting space. uh gold is part of a longer term trend I think and um if if there's a squeeze on private credit um liquid vehicles like BDC's may be the escape valve in a negative way but it might also be a buying opportunity. So those are the kind of three points uh that I thought were unique. Now um I want to do like a speed round um because I conscious of not being able to talk for four hours. Uh so maybe just to take um another five or 10 minutes and and and just flag some things that I thought were interesting, Adam, for as you and I catch up. So the the the benchmark as the headline of putting together an invest client's portfolio is kind of 6040 for moderate risk. And I'm just personally hearing a lot more people question the 40% Adam. >> Um and this is just one chart I use. This is in JP Morgan does guide to the markets which I know a lot of people are familiar with and the chart I forget the chart on the left but the chart on the right just shows that bonds don't always move opposite to stocks they can move in the same direction and >> we learned that in 2022 the hard >> right exactly um and and the temper tant taper tant temper tantrum the tariff temper tantrum of April, you know, the dollar didn't strengthen, it weakened and bonds rallied a little bit, but then they didn't rally that much. So, I would say, you know, people are questioning the 40%. I don't want to get into it too much. I just I just think that's kind of interesting because so much of us benchmark against that. The one slide I put together for people to think about is the total return from a bond. And we've been in a 40-year bull market for bonds. So the total return, which is the third column from the right, is uh you can see annualized people made a lot of money in the 80s and 90s, a little bit less so recently. Okay. More as interesting, you basically don't make any money from the appreciation of bonds over time. You really make it all in the yield. And so um anyways, I I think that's just something for people to think about. So bonds are yielding four or five%. That's what you're going to get. All right? Don't expect to get more than that. And and I think people are saying, "Wow, I'm getting like double that from stocks. Why do I own bonds again?" >> Um and gold has beaten bonds over almost every time period. Well, definitely over every time period over recent history. Now, >> why do I own bonds? I I'm I'm not I'm not trying to revisit 6040 single-handedly. I'm just hearing a lot of strategists and other people sort of saying let's have a discussion about 6040. Okay. Just let you know I'm hearing that too from a number of the capital managers that I I talked to on this channel. You know, some are developing a 603010 model. Some are doing some variance of that. But the the thing that they're taking away from is bonds. >> Exactly. I think exactly right. The starting point is stocks are fine in people's portfolios, right? It's the bond portion that people are taking money away from and that's often the biggest question um kind of in our industry. Um I am I'm going to skip a couple of slides here. Um uh and leave India for some other time. But I want to talk about Alibaba. Um and the question is um so a year ago uh talking to someone named Ab Adam Tagert I talked about my personal personal buying of Alibaba. Um, and one of the reasons was I thought the AI component was underappreciated and Alibaba is still cheap on that basis. And I just wanted to share with you something that I don't think is hardly known at all. Um, which is and this is actually original Vanc research which is I can't always say that. This is US electricity prices per kilowatt hour compared to China per kilowatt hour. Now, the labeling isn't really great here, but the blue line is China. Bottom line is the headline, US is twice the electricity costs of China. >> Wow. And their electrical um generation capacity is some multiple of ours too, >> exploding. And so I hate to like bum you out, but what if China became a cheaper place to do hyperscaler activity and that's an area that we could be competing with in the world. So, it's a little bit more Adam a pin to put in here to be thought provocative because I'm probably, you know, I don't want to spend a ton of time talking about it and I certainly am not doing some grand predictions here, but it's it's um ultimately economies are competing on the basis of the cost of electricity. Europe, the cost of electricity, right, have gone off the charts. They're now trying to deal with it. The Middle East is very cost competitive. Um, so anyway, I just think that it's it's sort of in in the background here. This was something helpful to to point out for people. >> Okay. Yeah. And and this this does add tailwinds to the AI energy trends that we were talking about earlier, right? Um trade meaning, you know, in the US, let's just look at the US lease. >> That is like a top sovereign priority, right? is to to build out our energy production grid so that a we can match China in production capacity but also not have this delta in the cost. >> Yeah. And then one last slide. You know, I I I can't stay away from the budget deficit, but I'm just going to get I'm just going to focus on one statistic. So, our budget federal budget uh fiscal years end at the end of September. And you've seen this slide for me before, at least the the column in the middle, where uh in fiscal uh 20 24, sorry, that was last year, right? Um our budget deficits was 6.4% of GDP. Simple math. We just printed fiscal 2025 and it dropped to 5.9% of GDP. So in prior quarters we've been talking about what will it be like for next year because this is a hybrid Biden Trump year. A lot of the spending that happened in Septe all of the spending in September, October, November, December and January was Biden spending. So a third of the year, this is a third Biden and let's call the twothirds Trump. So um this is just Jan has been optimistic that they're using a lot of tariff revenue and getting this deficit down as a percent of GDP. And what's happening is you saw revenue going up slightly about revenue and outlays went up about the same 300 billion but that's shrinking as a percent of GDP right so if the GDP is growing that 1.8 8 becomes less and less as a percent. >> Right. Right. >> So, I'm I'm still shooting for 4 and a half to 5 and a half% of GDP for this fiscal year that we just started a couple of days ago um on October 1st. What does that mean? It means less government stimulus for the economy which has been offset by AI compute spending. Um and it but it means better overall uh financial stability hopefully um as we crawl out from what I had been calling the reckoning of this last year. >> Yeah. And does that mean you think we'll largely get out of this? Okay. >> No, I'm just I'm just like we should celebrate this like we we snuck in, right? It was six and a half% or 6.4% 4% last year. And if you read a lot of headlines in in 2025 from the CBO and all that kind of stuff, they were like, "Trump's not shrinking the budget deficit. It's going up even further." Wrong. Right. And And is it largely wrong because the tariff revenues are offsetting things like the I mean, it just got passed, but things like the increased spending in the OBB. >> Yep. Yeah, I I can we can go through the the the details, but basically yes, and that's yes, tariffs are are are the most important part of this. Um, but you know, he listen, it's not talked about a lot and it it just flew by in the press like it barely got a headline. So, I just think that if you think about the big macro factors, this this the risk of crisis coming from DC is marginally less. and and I'm not celebrating or anything yet, but um it's headed the right direction. If we have a 4% growing uh economy next year and you know revenue and outlays stay in sync with each other, this number will come down again. >> Okay. Um All right. So, sorry, you were expecting what? Four and a half for fiscal year 2025. Uh, no. For next year, I never guessed it this year because I said it's a hybrid year and I I just had no idea. So, my I've always been talking about fiscal 2026 next >> fiscal. Okay. So for fiscal and I guess the administration would say hey yan if you like that you'll like the years even further because you know the passing of the one big beautiful bill we were frontloading a lot of the spending on it and that spending should diminish in future years um as a percent of GDP and we should start getting tailwinds from all the other policy stuff that we're doing and the economy should grow even more and that should narrow the deficit even further. Will that happen? Who knows? But that I think that's what you Yeah. >> I just think it's worth focusing on the scorecard, right? And and what we what we miss, again, this is why I love doing this outlooks with you, is sometimes the scorecard like flashes by so fast and we lose it and suddenly we're focusing on Taylor Swift's wedding or something else. >> No, I appreciate you doing that very much. If I can, let me just ask you one question on something you flew over and we can we can do the deeper dive next time, but you skipped over India. I know you've been pretty bullish about India um for the a number of the different quarterly outlooks we've done together. >> Um I don't follow it as nearly as close as you do, but I think Indian stocks have kind of taken it uh some of the winds come out of their sales over the past quarter. >> Um so just in a sentence or two, >> what are your thoughts here? Hey, temporary the party's going to continue again or is there something more material going on here? >> They've done the structural reforms that they need to and now the economy just needs to grow into them was my base case. I've got even more good news because they're passing a bankruptcy law reform which basically means that companies will go bankrupt in a more orderly and quicker manner which means more efficient capital usage and better risk um better risk management if you will in the economy which is fantastic. So yet another structural reform seems to be in the works. It's not done done, but it's almost done. I would say in terms of buying, um, you know, we don't quite, yes, markets are down. We don't quite have the full buy valuations that I would love. I would say we're close to the price to book being pretty attractive. PE not quite as much. So, um, if you don't own any, I'd certainly buy half a position. Um, but we don't have so much as to, you know, overweight. Okay, got it. Like I said, next time to the extent that that you know things are still somewhat similar um >> Oh, give a chance to really go deep into it. >> And because I love making predictions, um and I love having these recorded um I predict that the US will cut a tariff deal with India in the next month at 15% tariffs. >> Okay. All right. Um well, I I hope you're right first off, first and foremost. Um, I will say too, Yan, just I think we noted several of them in this this past hour, but I I think more often than not, your predictions have actually come pretty close to the mark. At least the ones that you've made here with me. You got a good track record, so hopefully you don't spoil it. >> Some good some good wins. Yep. >> All right. Well, look, um, Yan, fantastic as always. Um most important question um for folks that would like to um follow and track Vanx work and learn more about their family of funds where should they go? >> Yeah, our website has all our fund information uh van.com and uh I'm on I'm on X uh yanvan number three and I'm also on LinkedIn. I will share great great pods like some of some of Adam's pods. Um, so thanks. Yeah, thanks. And and uh appreciate again the platform. As you know, I did a shout out to you on on Josh Brown. >> You did. You are such a good friend of this this platform, >> which a lot of people have commented to me on. So, um, it's good. >> That's great, Yan. Well, look, thank you. And again, we'll continue doing this as as long as you've got the interest and the energy. It really is wonderful. Uh, and Yan, when I edit this, I'll put up the links to van.com as well as your social media handles there so folks know exactly where to go. But yeah, my friend, thanks so much. I know these um require a fair amount of time to put together. You know, I think you said they're they're helpful in helping you sort of structure your thinking for the next quarter, which is great, but I know that it's an expenditure of energy on your part. Really appreciate it. >> And and we'll get you the chart pack as well because I know that always comes up. So, >> yes, thank you. And yes, so folks, the chart pack will be available um in the Substack uh write up of this. It'll be to available for free to anybody. So, if you don't already sign up for our Substack, just go to thoughtfulmoney.com/newsletter. You can sign up for free >> and it's we'll get you latest tomorrow morning. You know, compliance always needs to polish every uh every vow. >> All right. Well, it should be ready by the time this video launches. But anyways, folks, please do me a favor. Thank Yan for just doing his usual bangup job here by hitting the like button, then clicking on the subscribe button below. Yan, my friend, again, I'm just super appreciative of this and look forward to doing this with you in a quarter. >> Great. Thanks, Adam. Good luck. >> All right, everybody else. Thanks so much for watching.