What Earnings Are Telling Us About the Next Market Move

All right, it is 11 o'clock. I think we'll go ahead and get this thing kicked off. Welcome in, everybody. Thanks for joining us. Start it off with a quote. It covers a lot. The problem with the internet is that it's full of bad data, conflicting information, and opinions. It's a topic we've been touching on a lot lately. Covers a lot. Yeah, I think that's a good one. A whole bunch of different topics, delves into all sorts of things AI-related that we'll get into throughout today's live podcast. Welcome in. This is our monthly live podcast. This one's titled What Earnings Are Telling Us About the Next Market Move. We'll get into a whole bunch of stuff today. Here, Kyle Guscik II with David Trainer. I'm glad to have you guys here today. Started off with the disclosures and disclaimers. We are publishing research. We are not providing advice. And you can read all of these in the replay that we'll provide later once this is a finished podcast. Kick it off with our agenda. Next one, podcast will be on September 11th. If you are signed up for this one, you're already signed up for the next one. So congratulations. Hope to see you there. Since we last talked, we put out a bunch of different research, some new long ideas, danger zones, some interesting looks at stocks that passed and failed the core earnings leaders test, ones that were added or removed From the Core Earnings Leaders Index, which is based on our core earnings data. Hosted quite a few earnings watch parties. We're right in the middle of earnings season. David goes in and looks at the earnings of the week, what it means, what it looks like in our system. And we'll do that a little bit later today. Published some sector style fund research, model portfolio updates, and a whole bunch more. But today we'll get into some upcoming training and announcements and a little overview of where does AI take the market next. And then we'll get into an earnings watch party. To finish it off, we got some big AI names reporting this week that we'll dive into. But wanted to kick off with the first announcement. We're proud that we've now had 62 consecutive months of number 1 stock pick ranks from SumZero. You can see our categories that we ranked in this month. Awesome to see the research that we put out continue to outperform. And this is SumZero, 16,000+ professionals competing for these rankings. So we're very proud to Be recognized for this and for 5+ years. That's no small feat. And if you want to learn more about how we do it, we have some upcoming trainings. The Ideas Dinner, which is for pro and institutional members. The next one will be August 18th. I'm actually going to share a link to our last one. We give a little sneak peek free training or free replay for everyone in our online community. So you guys can check out and see what those are. I'll share that link in the chat here in a second. And then. On August 20th, we've got a training, The New AI Gold Rush. We're going to show you how we're finding the stocks that are going to lead the next leg of the market beyond the AI hype and AI narrative and show you what we're seeing in the data and how we're finding some opportunities there. But in the chat, I said I'll share it, so I'm gonna put that here. That is the link to the free replay of the Ideas Dinner. So if you're not a member of our community, you can sign up to watch that. give some great insight into long ideas, competitive analysis. David did a breakdown of the upcoming potential PayPal acquisition. So I think it was a great time, awesome insights into a lot of different ideas.

Those dinners are probably the most fun thing we do at New Constructs. If you are a pro member and you've not been to one, you're missing out on probably the most valuable feature of your pro membership. And if you haven't seen this one, we decided to share with everybody. Take a look. It's, I think you'll see one of the coolest things there is in all of investing, which is a bunch of very focused and interested and smart people getting together with great tools to talk about stocks. Yep.

Agreed. Share that out. Next up, David, I'm going to turn it over to you. We've got the first topic of the day. Where does AI take the market from here? So if you want to jump on screen share, we can go from there.

All right. Thankfully, do we think AI is going? Because I think there's so many different perspectives on this out there these days that there's a void in leadership, honest leadership about what's going to happen. You got people saying that the market's going to go on the greatest run over the next few years that we've ever seen, Tom Lee. And you've got guys like Michael Burry saying they're going to short this again. It's the biggest opportunity since The Big Short. So, I think the truth probably lies somewhere in between. Let's talk about where we think AI takes the market from here. First of all, I think I wanna point out that we are in what I'd call the parrot phase. AI is really good at repeating back to us what we already know or what's on the internet. What it's not good at is figuring out what's right in terms of understanding how to answer hard questions. Parroting back things is easy, but If you're needing to actually discern something nuanced, AI has a really hard time doing that because it's looking at the internet and it's hard to know what's true. You all have seen our webinar in terms of why does AI lie so much? Well, you know that AI is trained to give you an answer, true or false. It's trained to look confident, true or false. And it can do that because The internet's got an answer for everything. Whether or not that answer is true is really a tough way to go. And what people don't understand is that the concept around AI is the same as it has been for every model in the history of the world— garbage in, garbage out. You can't— the model is never any better than the underlying data going into it. Solving expert-level problems is not possible without accurate data. Just like training a doctor to do good work, is not possible without teaching them how to do things the right way. The human brain, no matter how strong a brain might be, if it is filled with the bad information, it's going to be a bad brain. We all know people really smart, not always all that productive. So no matter how big the brain, no matter how smart the model, it doesn't really matter if it's not filled with good Data. And this is something that we've been hearing other people say for a while. Larry Ellison all over the internet saying flatly, the AI models are all converging, commoditizing to the same level of value because they're all based on the same data. And then what do we see? The Chinese models started coming in and pricing got blown up. No surprise. And we're seeing prices from AI and cloud come way down. And the Chinese models were super cheap. They're coming up a little bit because they can. they know they can make some money. If someone's charging 25 times as much as you, maybe they only need to be charging 10 or 5 times as much, but you're still a lot cheaper. The next few slides, I want to talk about the AI stack because the AI stack, I think, is complicated. There's a lot that goes into AI. It's not just a model. You've got, at the core, you got data collection, you got model development, training and optimization, deployment integration. Monitoring and maintenance. That's one way to think about the stack. Another way to think about the stack is data sources, transformers, vector databases, embeddings, LLMs, and AI agents. Another way to think about the stack, infrastructure, the data center and the data. Then you've got the AI, then you've got inference, and you got governance. That's important too. There's all these layers. But consistent across all these charts, data. The least talked about item in all these, data. We hear things about infrastructure and data centers and billions being spent. We hear things about governance and cybersecurity and inference chips. For sure, we hear a lot about AI and machine learning models. How come they don't talk about data so much? They don't want you to know how much of a problem it is. Without data, none of this works. But they don't want you to know that because they don't want you to ask the question, where do you get your data? How good is your data? Do you think that they would probably be more proactive talking about data if they were proud of their data? Probably. But they don't want you to think about that. Instead, instead, they wanna distract you with news about how the AI is so powerful that it broke out of its cage and attacked Hugging Face. Remember that a week or two ago? The OpenAI latest version broke out and attacked another company and stole its data. It's so powerful, we can't even keep it under control. Look over here. Don't look over here. When I asked my friend who ran Google Cloud Professional Services about data, this was a little over a year ago. I said, hey, I know you're spending $15 billion a year on, on Data centers. What kind of data are you putting in there, Brad? And Brad was like, oh yeah, we don't really like to talk about that. I'm like, yeah, I know. It's all internet. I get it. What else are you going to do? He's like, yeah, I don't know. I said, what if I told you there was a dataset out there that was proven to be accurate? He said, that's exactly the special sauce we're looking for to build our AI on. They just don't exist. I said, how about this? How about this Harvard Business School paper, these Bloomberg indices? That's proven. He's like, we're all over it. And they immediately put $2 million into building an agent based on our data. They're dying for data. I'm getting calls from data brokers. Hey, Amazon's looking for better financial data. Yeah, of course they are. Because as long as they're offering the same old stuff, the same old ingredients, they're all moving around the same pile of dirt. There's no way to make that any more valuable. There's no way to differentiate themselves. And as long as they're undifferentiated from a super cheap Chinese model, Then they got to be super cheap. And if they're super cheap, they can't make money. So the commodity nature of the data is why DeepSeek, KIMI, and all these other Chinese models, open source models, can catch up so fast. By the way, it's not just going to be those guys. There's going to be others. And that creates an existential valuation crisis for these stocks because if there's not margin anywhere in the stack, Cybersecurity, competitive. Inference, those are chips, that's a competitive business. Infrastructure, competitive business. There's nowhere to make huge amounts of profits, in which case the valuations of all these stocks are set to plummet. Now maybe a few will survive, but not in their current form. All these stocks are priced as if they're all going to win. And Economics 101 says people will not pay a premium for something that is— that does not produce premium value, period. And if there's no premium value in any element of the stack, where do these premium valuations go? They go down. It is tough to make a straight-faced argument for investing in businesses like OpenAI and Anthropic that charge significantly more for a product produced and sold for much less by multiple competitors. When you go to the grocery store, would you pay 23 times, 5 times, 2 times as much for something that's the same? Would you do that for anything?

No.

So where's AI gonna create the most value? For Wall Street, $100 million payday on SpaceX. They're looking to do the same thing, run that same playbook for Anthropic. for OpenAI, for any other kind of new stock like CoreWeave, anything related. All this trading in the existing stocks, let's do it. They are all over it. Let's create ETFs based on AI. Let's do it. That's where the money's going to go, folks. That's where it's already going. And these insiders, they are cashing out. SpaceX insiders, stock's at, I don't know, $130 or something now. $135. Private equity people were in there at 25 cents a share, $5 a share, $10 a share. Later rounds, probably 70, 80 a share. I don't know for sure. And they sold it for $135, $140. Yeah, they're cashing out big time on your momentum, on your belief that AI is going to take over the world. Folks, transforming technologies The ones that are most transformative, the most impactful, the value accrues to the consumer, to society, not to some company that's going to be a gatekeeper on something that's so valuable. Electricity is a great example. The pundits refer to AI being as impactful as electricity all the time. That's the opposite argument they should be making. The electric company, that's a utility. We can't in our right mind restrict access to utility or make it a profit-seeking enterprise so that somebody can squeeze people out of having access to power. Aren't they saying AI is gonna have— everyone's gonna have access to AI. Everyone needs it. It's gonna change the world. That's why I show you that stack. All these things, no profit, commodity profit. And if it becomes a monopoly, it's too profitable, then it becomes a utility. They can't let that happen. They can't do that and put AI in the hands of everybody. So all this stuff really doesn't make a lot of sense. And we're not the only ones saying this. Steve Eisman said it really well, CNBC. Just a couple days ago. This whole thing has an Achilles heel. That Achilles heel is that if people aren't willing to spend huge money for access to OpenAI and Claude and Grok, any private model, any closed model, where do the profits come from? They disappear. There aren't any. And with all these stocks priced as if they're all going to be winners, not just be profitable, but be monopolistic winners. SpaceX stock price implies that its profits will be twice as large as Apple's profits today. Apple's the most profitable company of all time. SpaceX is already priced to be twice as profitable. Is that right, Kyle, or is the multiple even higher? Might be even higher.

I know it's the most profitable company. I'm not sure twice or 3 times, but yeah, it definitely has to be the most profitable company in the market.

Yeah. That's what's priced in. And is that going to happen? What happens to businesses that are super profitable? They attract competition. You've got to be super profitable and do things that no one else can do in order to justify these premium valuations. And they can't all do that. All right.

I think the upcoming Earnings Watch will see some of those examples of the astronomical valuations with some of these AI names. Yes.

Let's jump right into that. All right. Let's go. To the stocks on the list this week. Cisco. This is, this was actually a long idea for a while. We loved the stock when it was cheaper, but what we're seeing across the board here, folks, we don't see any good ratings and we're seeing declining economics. But yeah, Cisco is a great company. Plumbing. We love it. We love this idea of the internet needing plumbing. And look, as, as people have been saying forever about basically the internet, which I think is for sure true. The more tools that grow, the more the bandwidth we need, the more plumbing we need. All these things are true, but it's all priced in. It's all priced in. How did they do this week? Yeah, I think Cisco had a tough one. For whatever reason, people don't like this stock as much as they like the other stocks. It's not as sexy. Been around for a while. Yeah, I don't know what to say. Good company, bad stock. I say this a lot. I think this business is doing well, but even these guys are probably stretching a little bit. Free cash flow to debt. Not as great. But yeah, Cisco, good company, bad stock, neutral rating. Don't know what else to say. Not that— not just nothing terrible, nothing good, neutral. Stay away.

I think Cisco also provides somewhat of an example of the dot-com boom where they soared because of the potential profits, and then when the dot-com crashed, it took Cisco 20 years to get back to That previous high.

Yeah, they were— they've been through— they've seen this party before right here. They were our poster child, to your point, Kyle. They were a poster child of a good company, good company, gonna survive, but way overpriced. They weren't gonna have a monopoly on all the internet build-out and plumbing, right? None of these companies are. Just the idea that these companies are gonna have this kind of profit is honestly— it's counterfactual. To what happens anytime people are like, oh, that's going to be a super profitable business. Why don't we let that guy have it? I don't want it. I want to go, I want to stick with restaurants and less profitable businesses. Yeah, let's let somebody else do that. No, they're all going to do it. They're going to do it. They're going to— a lot of them are going to be terrible at it and they're going to charge the same price anyway because they got capital to burn. They just think they're going to get in. It's called irrational competition. Happens all the time. Netflix was irrational competition for a decade before they made any money. That can happen. There's a lot of capital out there chasing AI, which means there's gonna be a lot of companies offering the same thing at terrible prices because they just want to get a piece of that market share. That's what Netflix did. They basically were subsidized and burned money to get market share until the— until they could raise prices enough and it had enough product to be profitable. And that's a strategy that works for maybe 1 or 2 companies ever, but all it does is just ruin profits for everybody else in the interim. All right, let's talk about Applied Materials. Super expensive stock here and not that good. These— all these guys have to really ramp up some spending. Their returns on capital— see, economic earnings or accounting earnings taking off, but returns on capital coming down because they're all spending. And a lot of that's on the balance sheet. You don't see it. We're going to update our winners and losers report here very soon. You're going to see off-balance sheet debt skyrocketing. We're talking about over a trillion now. And when you look at the real numbers and you see the cash flow, and then you compare that to valuation, again, all priced as if they're going to be winners, and most of them are going to be losers. And that's why we're saying beware of this final cash out. And there are good stocks out there, and New Constructs has made a living finding those stocks consistently over time. There are always good stocks, always some good stocks. You just got to have the discipline to go buy the good stock, even if it's not popular. And you got to have the analytical capability at scale so we can scan the whole market. We can find those good stocks and help you avoid being the bag holder in the final cash out. Brookfield Corp. Big earnings report this week as well. How did they do? Second quarter earnings report, solid growth, earnings distributions, net income volatility. It's a Canadian company. We tried to find some exciting names this week. We failed. Unattractive. Just doesn't look so great. A lot of debt here. So the economic book value is negative. I think we go in here and take a look here. We'll see. Not a lot of excess cash. Huge tax-deferred liability. Huge minority interest.

Wow.

And then debt, not too crazy. That's $118 billion in minority interest. So that's a senior claim on anybody else out there. We're looking at all the liabilities, folks. Let's talk about a company we used to love, Simon Property Group. I remember loving these guys out of COVID Kyle. This is one of our COVID picks. And I was like, you know what, man? We are going to lean in right now because a lot of these stocks, the world is not going to end. And we crushed it on a bunch of stock picks. That's a big part of, I think, part of our sum-zero ranking is we leaned in on a bunch of them. One of them was Simon Property Group because they were like, they owned a lot of malls and everyone's like, oh, all the malls are going to die. Not the really good malls. Those survive. I'm lucky. I've had one in my neighborhood since I've been growing up here in Nashville, the Green Hills Mall. I had no idea how nice that mall was until I went to other places and I'm like, whoa. Okay. Yeah. This is— Some malls absolutely need to go. Some of them are really nice. People like them. And Simon Property Group, we loved it. And we rode this stock from really low levels to I don't know how high. We— this was a great stock for our focus list and model portfolios. But guess what? It got expensive, folks. Got really expensive. And it's a competitive industry. This is not AI. This is commercial real estate. But I think even if you're into the data center stuff of commercial real estate, that's competitive too. People aren't just handing profits out. Everyone's out there fighting for a piece of it. And as long as there's competition, profits will be de minimis eventually. And the higher they are in the short term, the more competition you attract. All right. Coherent Corp. 32.3 price to economic book value. What happened with these guys? So down 14% this week, up 50%, up 70% year to date, down 14% this week. What happened? They had a strong quarter, upbeat guidance. They beat and they had a guidance raise, capacity expansion, optical AI infrastructure demand. Demand center scaling is a massive secular catalyst. Stretch valuation and post-earnings profit taking. Yeah. So folks, when you see something like this, 32.3, that's, I think Nvidia is 5 or 6. I think SpaceX isn't even that bad. Is it? Actually, SpaceX is maybe negative.

It is negative at 41.

Yeah, SpaceX. Yeah, negative 4. All right, that's even worse. Okay, so, but 32.3 is— that's amongst the, amongst the worst in the, probably in the S&P 500. That's crazy expensive, folks. Here's how it looks on a chart. Economic book value, it was, yeah, up $27 a share. It's gone negative. Now it's positive again, slightly. Divergence. That's big, folks. It's just, this is just risk, folks. Just taking a lot of risk. And what I'm here to tell you is you don't have to take that risk. There are lots of great stocks out there right now that you can buy. A lot of them have nothing to do with AI. And I think anything that's related to AI is pretty much expensive. A couple years ago, Kyle, we picked out a couple. We went out of our way and we said, we're going to find some AI stocks. We found Arrow Technology, Kyle, Vishay, PhotoLab. And those stocks did great, but they're too expensive now. We found the last few, like when the AI thing was like picking up, we're like, all right, we got to talk about AI. Our marketing people are like, you got to talk about AI. I'm like, it's all expensive. And we're like, all right, you know, Kyle, let's go dig. We got the whole market in our, in our available to us. We dug and we found, we can only find 2 or 3. And they're like, man, these are small. Is it really even AI? I'm like, okay. Yeah, no, these are AI related downstream. If AI does what it says it's going to do, Then these businesses will do really well too. And they were barely attractive. I think some of them may have been neutral and we're like, no, I think we're going to go with this and boom. But yeah, now the AI money pouring into so much that now the bowl is overflowed and it's flowing into everything else. Yeah, there's nothing. The AI money has found everything related to AI. You're going to have to look a little bit harder, but that's what we do. Supermicrocomputer. This is a great one because we were really out in front of this crashing. A while ago, I think it was about a year ago, or was it 2 years ago, Kyle, when we identified the poor internal controls? I think it maybe is 2024. I don't know.

Yeah, I think it was more than a year ago.

Yeah, 2 years ago.

Actually looks like August 2023 is when Garrett first flagged it.

Okay. So yeah, we found the internal controls as an issue back here. Stock went in, boom. Yeah, we flagged it big time. And it's still a flag. Guys, internal controls is an issue. You're a major publicly traded company. You can't get your accounting right? You can build fancy semiconductor chips. You can build AI models. You can't get your accounting right? Is that by accident? Come on. Right? That's another red flag on SpaceX. They can't even get their accounting right either. They're like emerging companies, so they don't have to admit that their internal controls are weak. They just say, we have no opinion. Come on. You can build rocket ships and AI, but you can't get your accounting right? And then you expect American investors to say, we're going to pour all this, give you all this money? Tell me how that's a good trade. It's not. Supermicro Computer. Yeah. So this one's been all over the place. This feels like almost an also-ran, as my mother-in-law used to say, been rode hard and put up wet. This thing's— this stock looks tired. It's no longer a go-go stock for AI. All right, Doug, what did Doug say here? Look in the Q&A here. If these companies crash, the term AI may not be popular. It may even be seen as a negative in investors' minds. AI equals artificial profits. Yeah, absent profits at this point. Good point. Great point. Okay, folks, we are at our time limit. Thank you very much for being with us today. Appreciate the comments and the questions, and glad you could be with us. Pay attention to this final cash out because there's more on that to come. Thank you and have a great weekend.