Alphabet’s $80 Billion Flex

0:01Travis Hoium:Why does one of the most profitable companies in the world need more money? Motley Fool Hidden Gems Investing starts now. Welcome to Motley Fool Hidden Gems Investing. I'm Travis Hoium joined today by Lou Whiteman and Tyler Crowe. And guys, one of the stranger news announcements this week was that Alphabet announced that they are raising another$80 billion. What got attention was that$10 billion of that is going to come from Berkshire Hathaway. But as NeoClouds and Oracle and even some of the other hyperscalers are taking on more and more debt to fund this AI build out, which is now moving past the operating cash that they're generating from their business, even Alphabet might pass that in 2026, they need more money.

0:46Travis Hoium:And Tyler, Alphabet said, hey, we want to sell equity, not debt. So this is just kind of an interesting move from them.

0:53Tyler Crowe:I think there's a little bit of like, we've been watching too many industry and business related television shows like Succession. We're all, we're thinking like strategy, like how can we really stick it to these other players? But let's be honest, they're just looking at the numbers and being like, we need more money. We're not trying to like necessarily beat somebody more so than the other. The first thing is the bottom line first. And I think this is really the case when it comes to this deal with Google raising$80 billion because just some rough numbers. I think their current capex is somewhere in like$170 billion range for 2026.

1:28Tyler Crowe:2027 is probably going to be more. And they're bringing in$175 billion in operating cash over the past 12 months. So if you're going to obviously do more, you're going to go past your operating cash flow. And I think this is getting ahead of that. They're obviously probably not going to go$80 billion over in a single year. This is probably like, yeah, let's get a little bit of a cash cushion, get ahead of it. We've got a decently priced stock right now, and this will be a good time to do it. We'll get Berkshire involved. It'll make it a little bit easier. And they've got things like that big Anthropik deal that they signed last year.

2:03Tyler Crowe:So there is demand for what they want to build.

2:06Travis Hoium:Yeah, when we look at some of these deals, I mean, Anthropik just raised$65 billion. Now they're selling several percentages of their company, 6%, 7 % of the company to raise that kind of money, Lou. this is 2 % of Alphabet. So it's not all that dilutive and it does keep that flexibility on the balance sheet. Like Tyler said, I mean, there is debt on the balance sheet, but they still have a net cash balance on the balance sheet. So, you know, is this the kind of thing that you like as those bills for the AI build out keep going up or would you rather have them use debt?

2:40Lou Whiteman:Yeah. Now, Tyler, first of all, two things can be true. Okay. This can both be that they need money and they are smiling and waving at SpaceX as they prepare for an IP.

2:50Travis Hoium:Yeah, this did come before the SpaceX IPO, before we heard, I think, hours afterwards that's anthropic.

2:57Lou Whiteman:Yeah, it's basically the same amount. I mean, part of it is, is just like every other company striking while the iron is hot, probably too, for the excitement around it. But look, I think that they have a compelling story to tell relative to those startups. So why not, you know, get your story out there? I think Tyler said it right. With very little dilution, this is the time where you do use equity when your stock is highly valued. And so to raise money this way, I think makes sense. Travis, you hit on, I think, maybe the most interesting thing about this. For even, I think, among the best of these companies, even the cash generation machine, last year's bull story is now over.

3:36Lou Whiteman:because last year at this time, all of the spending was justified as it's not 1999 all over again. They can fund this from operations. They can fund this from their cash. That is simply no longer the case. So that's over. The good news is, is that these are massive, well-capitalized companies that have a lot of options. So it's not like, I'm not trying to say the sky is falling. I do have questions, but I think this is the smart move is that this is really, really expensive. Strike while it's hot.

4:04Travis Hoium:All right, pop quiz. out of the hyperscalers that are spending hundreds of billions of dollars on this ai build out in the neoclouds what is the one other company that has a net cash balance do you guys know it is microsoft it's a mic only the only other one of these companies amazon net debt of about 67 billion dollars and then all obviously all the neoclouds you know nebius core weave iron They're all Oracle, a ton of debt. So debt is really fueling most of this. This is going to put Alphabet in a very different position. Tyler, what does this say about the return on investment that they're potentially seeing?

4:48Travis Hoium:And this is one of the things I think we're kind of trying to draw conclusions that are hard to draw today because you look at the GCP numbers within Google Cloud. Those are phenomenal. but the$180 billion that you mentioned that they're spending this year to do this AI build-out, very little of that is actually contributing revenue today. It's gonna be next year, the year after, if token prices start to come down, which they've actually gone up recently, the economics could change. So what does this at least indicate about what they're seeing about the ROI for the business?

5:23Tyler Crowe:I feel like that's a great question. I don't have a great answer to. I'm kind of throwing out into the wind here But that's the case with all of AI, right? Where we can see some tangible benefits. I mean, here at Motley Fool, using Claude and GPC and several other products, much more prominently in our kind of everyday workflows. So we see the use cases. However, the economics of what those businesses are doing is a little bit questionable. for the hyperscalers you know one thing that you can at least rely on to a certain degree is companies like anthropic are pledging large amounts of money to them for renting this stuff and just to use the spacex example that you were mentioning earlier as part of the s1 for spacex that deal actually said that anthropic is going to pay them i think it's 1.25 billion dollars a month for compute power.

6:17Tyler Crowe:And so it shows like there is a monetizable way of doing this. And it's not just some, hey, we don't really know what it's going to work with. There are actual like revenue numbers you can put behind the compute. And so similar to that, the Anthropik deal that they signed was something like$200 billion worth of compute power over the next five years. So there is a return.

6:40Travis Hoium:But I want to push you on that. You and I have both been following energy for a very long time. If you go back to one of the massive growth stories of the 2010s, it was the solar industry, right? Solar installations exploded. And there were companies that were supply constrained, manufacturers that were supply constrained of silicon. So they needed to get silicon. How do you do that? You sign a long-term contract with these suppliers to get silicon. The problem was for these companies, they signed a contract for four or five years to get this silicon at a certain price. And then the price collapsed.

7:14Travis Hoium:And so suddenly you have a contract for, I don't know, let's say$100 a pound. I'm making up these numbers here. But the spot price is$5 a pound. Suddenly you're losing your shirt. So this is what I really struggle with is what are the long-term economics? Because yes, that Anthropik deal is great for Google because they have a contract and we could go with CoreWeave or iron or any of these companies but eventually the economics doesn't work for somebody and we don't know exactly who that's going to be so it does that ring a bell to you tyler we haven't gone through some of this with that and you could go to the the drilling companies you know in energy

7:52Tyler Crowe:this kind of happens over and over again i almost feel like we might be stepping on uh some stuff we're going to talk about later on this but it is kind of this idea of like what we're doing at the current pace doesn't necessarily like work there's some incongruencies here and to that point like yes we could see compute power costs go way down for reasons xyz but there's also like the possibility of algorithm efficiencies right now all these companies are incented to put out the best product in turn and kind of compute power be damned eventually it's going to come down to like a cost per token you know efficiency from the actual computing that you're doing with these algorithms that will incent these companies to, you know, be within the range or try to bring down their costs commensurate with what they're going to see with compute power.

8:41Lou Whiteman:To your point, though, I would love to see the contracts because it could be the other way to like how how how rock solid is this future revenue? What are the escape clauses? Yeah, but it does feel like there's a tension there. Here's what I'll say, though, and this is kind of the silver lining. I'm confused by ROI. I don't see it. But having Berkshire along for the ride is a pretty good signaling tool, whether it's right or not. You know, maybe Greg Abel, his legacy will be he's just a, you know, LOL day trader, nothing matters. And he's just jumping in on a momentum deal. But I doubt it. And Berkshire, with its kind of good housekeeping seal of approval, seeing a path to value here, seeing something there, you know, as things go, and especially as everyone's competing, as I said at the top, that's a pretty good endorsement.

9:28Lou Whiteman:So, I mean, maybe if Berkshire sees the reason to believe, we should too.

9:31Travis Hoium:When we come back, we're going to get to the supplier whack-a-mole. Who are these companies paying for all of this compute? Why are Dell and HP two of the hottest names on the market? You're listening to Motley Fool, Hidden Gems Investing.

9:48Tyler Crowe:Hey, Fidelity. What's it cost to invest with the Fidelity app? Start with as little as$1 with no account fees or trade commissions on U.S. stocks and ETFs. Hmm, that's music to my ears. I can only talk. Investing involves risk, including risk of loss. Zero account fees apply to retail brokerage accounts only. Zero dollar commission does not apply to customers designated by Fidelity as a professional equity trader. A limited number of ETFs are subject to a service fee of$100. See details at Fidelity.com slash commissions. Fidelity Brokerage Services, LLC, member NYSE SIPC.

10:18Travis Hoium:Welcome back to Motley Fool and Jim's Investing. what's been crazy this year is how much we have gone from supply constrained environment to supply constrained environment all these companies have explosive growth and the next growth story apparently tyler is these old companies that have been around forever dell and hp hp enterprise they are on fire right now but when you look in the growth and valuation of these companies What excites you about this segment of that supplier space? Because it seems like, you know, memory was hot. Now it's just going up and down the gamut.

10:55Tyler Crowe:Yeah, to me, this is a little like this isn't just the tide that raises all boats. This is the tsunami that is bringing everyone 30 feet up higher than they thought they were going to go. It's not a tsunami.

11:08Travis Hoium:I like that.

11:09Tyler Crowe:Yeah, just think about, again, we're talking about Alphabet spending like$180 to$200 billion this year. If we're going to take the over or under of like how many revisions up there we're going to see on their CapEx, it's going up. 2027, it's going up. They're raising more money to go higher. All of them are going higher. Meta's doing weird off-balance sheet stuff so they can go higher. it's not it almost is to the point where it's not like we need to get the best equipment to make this happen it's like grab every spare part you can and slap it together and get some sort of compute power available i'm sure that a lot of these like neoclouds and some of these you know fly by night operations are just grabbing whatever they can to make some sort of profit or make some sort of return supplying this compute, which does appear to be constrained.

12:03Tyler Crowe:You know, Anthropic has said for a while that they are supply constrained when it comes to compute. Chep-GPT, not quite sure there because they spent loads of money before that for the compute. So to me, this is not necessarily a case of like, oh, yeah, these companies are all doing awesome. I think it points to the desperation that all these companies are going towards to make sure that they have enough equipment in the ground when these contracts are supposed to come due. I don't know if I'm necessarily excited by that because this is, you know, you think of it almost like a post hurricane. I know I just use tsunami, but you know, that kind of time where like supplies are so constrained that prices go up infinitely and whatever you can get your hands on is effective.

12:46Tyler Crowe:And this, that's kind of what this feels like. It's, you know, you don't see some discernment of like, oh, NVIDIA has got the better stuff. We're just going to use NVIDIA. It's like, no, we'll take some low end processor just as long as it will actually compute something.

12:59Travis Hoium:Yeah, Lou, I think this to go to our ROI point for Alphabet, this is what makes this also confusing, because if you look at their revision last quarter, they revised their CapEx number a little bit higher. But they basically said, hey, we're not going to get any more compute. We're just getting less compute for more money. That's flowing down to some of these suppliers. But you would think eventually, eventually he's doing a lot of work there. But eventually there is some sort of supply demand balance and the commodity companies become commodity companies again, which can be a really good business.

13:28Travis Hoium:But it doesn't necessarily mean they're going to be rocket stocks like they are today.

13:32Lou Whiteman:Yeah. And it tends to be like the longer we go through these cycles, the more our attention span goes quickly. So we'll see how sustainable these are. You know, no one's talking about Micron anymore because of HP Enterprises, it feels like. But I think what we're all dancing around, and I think, you know, we're not, because it's hard. I don't think any of us know the answer here. But there is just this weird tension where all of these stocks are based on projections that hyperscaler spending will continue basically through the end of the decade. With Micron, it was through 2029, the order book is filled.

14:07Lou Whiteman:And at the same time, corporate America is starting to talk about measuring ROI on AI instead of just throwing as much of it at the wall to see what sticks and put it in your conference call. If both of those things are true, then how are the hyperscalers holding up so well? You know, like how can everybody win here? Things are going to be massively expensive. Pricing power may not or there's there's this pushback on pricing power improving. And yet the ones both spending the money and providing the service are winners here. It feels like not all of those things can be true. And I don't think any of us know which part breaks or how or, you know, how this how the maths resolve here.

14:53Travis Hoium:Yeah, I saw an interview with an IBM executive this week, and he was making the argument that this is not a bubble because there's so much demand. And then went on to explain that, you know, if there's 10 or 15 suppliers, well, not all those suppliers are going to survive. It's just going to be two or three that are going to survive. And I was thinking that's exactly what a bubble is.

15:13Lou Whiteman:So, well, here's the other part, though. It is not over by a long time. Yeah. And I don't want to I don't want to name the company because I don't want to start this. But I just got a research note yesterday from an auto about an auto parts company where the analyst is basically projecting revenue to, I think, like a 20 percent CAGR. as soon as the market realizes that you can retrofit their parts into gas turbines for data centers. You know, I'm not I'm not predicting that's what's going to happen. But that is that doesn't imply to me that we are getting out of silly season anytime soon. Yeah, shows the level to that one as well.

15:49Tyler Crowe:Just in the wild stories of all of a sudden becoming AI compute supplies. There's a jet engine servicing company that owns some turbines that they also lease out to planes and like, hey, we can start leasing these out to AI data centers and we can make way more money. So it's just like it's all coming out of the woodwork now in the weirdest of places.

16:09Travis Hoium:Well, it's a good time to be a supplier of anything in AI today. When we come back, we're going to change directions here a little bit and talk about the future of Bitcoin. You're listening to Motley Fool, Hidden Gems, Investing.

16:24Tyler Crowe:Hey, Fidelity. What's it cost to invest with the Fidelity app? Start with as little as$1 with no account fees or trade commissions on U.S. stocks and ETFs. Hmm, that's music to my ears. I can only talk. Investing involves risk, including risk of loss. Zero account fees apply to retail brokerage accounts only. Zero dollar commission does not apply to customers designated by Fidelity as a professional equity trader. A limited number of ETFs are subject to a service fee of$100. See details at Fidelity.com slash commissions. Fidelity Brokerage Services, LLC, member NYSE SIPC.

16:55Travis Hoium:Welcome back to Motley Fool, Hidden Gems. investing every four years. Bitcoin drops like a rock. Last time it did was 2022. So right on cue, it's falling. But this time, it's micro strategy that is causing some of the consternation. Michael Saylor's firm, who has really made his name, being the loudest Bitcoin bull in the room, they have deployed a strategy called Bitcoin yield. That involves selling debt and equity to buy more Bitcoin as long as your net asset value of your equity is higher than the value of the Bitcoin that each share holds. That works. But Lou, this is starting to kind of come undone.

17:35Travis Hoium:That premium has started to come down. They have to pay down debt and dividends and things like that. And now they're actually selling some Bitcoin. So is this, you know, don't call it a Ponzi scheme kind of over for them?

17:50Lou Whiteman:So first of all, that four years thing, come on, Carl Jung called, okay? Like we do not have enough. That is a coincidence until we see seven or eight more patterns, but yeah, it's happening, right? Look, Sailor's plan reminds me of a lot of option strategies. You know, it works as long as the underlying asset you're looking at only goes in one direction. And the second it doesn't, you figure out it's, why isn't everyone doing this to, oh, this is why not everyone is doing this. This thing to me here is, you know, Saylor, if only he hadn't gone full on evangelist, because if they would have been willing to manage this at the first sign of cracks, they could have kind of unwound this orderly.

18:33Lou Whiteman:But no, he was he is the evangelist for Bitcoin. So therefore, he will never sell. He is going to be the, you know, the the the hands that never sell. he kicked the can and now he's paying the toll for that. You know, I don't know what this means for Bitcoin. I think it's important to separate Michael Saylor from Bitcoin. But I will say that if I if I was interested in Bitcoin, I would definitely be waiting for this to play out. We don't we don't want to catch this falling knife simply because we don't know how many of how many things are going to break if and when this strategy unwinds.