Meta is undervalued with high upside; expected pivot to renting out AI compute will remove valuation discount, supported by technical floor at $540.

Do you buy Meta stock or stay away? There's something unique going on with this stock. It is obviously one of our massive hyperscalers. It is the only hyperscaler that does not rent compute, and Wall Street has a few choice things to say about it.

You've got Meta stock on the one-year basis compared to Amazon and Apple, destroyed versus these other companies. Meta stock down 23%. Amazon's up 15, Apple is up 37% on a one-year timeframe.

If I go back uh to the beginning of the year, Meta also underperforms with a 12% decline, 15% gain for Amazon, and 17% gain for Apple. So, what is wrong with this company, and why is it getting crushed?

Well, a lot of it, at least according to Wall Street, has to do with the fact that yes, indeed, they do not rent out their compute.

But, if they change their mind on that, would it be worth buying? Is there a chance they're going to change their mind? What else is going on? Let's look at a few pieces on this.

First, via Barron's, we have an argument from New Street Research that if Meta rented out some of its extra data center compute, they would end up having a positive return on investment on their uh AI capex spend.

Right now, without them renting out their compute, they sit at return on invested capital estimates of negative 2.2%. Not good. This is seen as a big overhang on the stock.

Uh and so obviously, when Zuck in the uh in a conference call last month said that, "Hey, we're getting a lot of offers for compute at a significant premium over what we paid," which is probably the likes of like an Anthropic or Open AI who really don't care right now how much they're spending, They're just kind of trying to attract users.

That's obviously not sustainable, but if we're in a shortage or scarcity mindset and OpenAI and Anthropic want to be drunk sellers sailors, so to speak, and blow a bunch of money, let them.

Why not stand there and monetize? It's like selling lemonade to people who are dying of thirst in the desert. You could charge whatever you want.

We know Elon wants to be the desert lemonade seller cuz he literally tells us, "Hey, man, if I could build compute for and and end up monetizing it at 30 to 50 billion dollars per gigawatt, I'll do that all day long because I could basically build my data center in 1 year and have a 1-year payback period."

Which, if you think about it, Elon is pitching SpaceX as doing that between now and the future in the build-outs.

Obviously, they have some already established in terms of compute, but they want to almost 10x the compute that they have. I think they want to go from about 1.6 gigawatts to 10 gigawatts.

Meta already has about 6 gigawatts and they're trying to get to 10. So, technically, Meta could flip the switch and be about four times the size of SpaceX's current compute now and monetize on that scarcity now.

So, Meta is not advertising that they can monetize on that scarcity even though they have it. Elon doesn't really have it and is pitching monetizing on that scarcity. This is why people think Meta might be trading for a discount and that that could change really rapidly if they actually came out and flipped the script.

In fact, they say that Zuckerberg will end up acquiescing. They say, "We think there will be cloud revenue this year and it will be a catalyst for the shares." Now, another thing that is going to be a catalyst, which I think is really cool, is obviously the expiration of the coupon code today.

Remember, you can join us over at meetkevin.com. All nine courses, every trade alert, every private live stream, every Alpha report, every day the market is open. Tax professionals say it could be a tax write-off.

Use coupon code J Hole expiring tonight at 11:59 p.m. Take a look at that. Make sure you get in and then remember you can also bundle up with Homes AI if you have not yet. You should do that because we are getting rid of the lifetime access for Homes AI.

The product is going to get better every single month. We've already gotten a lot of additional feedback from the community that we already have in progress to make the app even better.

And so we couldn't be more excited to have you as part of the startup. Consider bundling up and joining us. If you're already course member, you can also bundle by emailing us at staff@mevin.com for a special coupon code.

Uh but actually looking at Meta catalyst, it's not just New Street that suggests this. It's also Deutsche Bank. So, let's jump into the um well, it's not just Deutsche Bank either. It's also Goldman Sachs.

We're actually going to start with Goldman Sachs. Goldman Sachs argues the following, that there is a quote compelling risk reward with a focused on increase AI clarity in the coming months if Meta can start monetizing their compute.

First, there's a whole section about how great it is that they settled this lawsuit because there was talk about how the lawsuit could potentially end up costing 1.4 trillion dollars in fees.

At least that was the kind of crap that was advertised in PR, which probably helped kind of wrestle them into a settlement in the low, you know, five-ish billion-dollar range. Still a lot of money, but it's certainly not 1.4 trillion dollars.

They say that removes a big overhang over the stock and some of these limits like a time limit for children or notifications off during night hours or parental supervision options, age limits and verifications, that these things aren't really that big of a deal and they actually remove more of a negative catalyst from the stock than you end up losing by having those limitations.

On the flip side though, Goldman Sachs thinks that refocusing on operating fundamentals and AI execution is where you can really make the money. Deutsche Bank agrees with this, by the way.

So, Deutsche Bank argues that Facebook, Meta, could generate 15 to 36 billion dollars of extra revenue if they decided to monetize their compute right now.

Now, keep in mind that's actually not that much relative to their total revenue. They're expected to make a quarter of a trillion dollars at the end of 2026. Like, think about that for a moment.

A quarter of a trillion dollars of revenue, 254 billion dollars of revenue by the end of 2026.

So, generating an extra 15 to 36 billion dollars by monetizing some compute, I mean, that would be what? Maybe a gigawatt or two? Hopefully, less, you know, for the purposes of the whole AI industry.

Uh otherwise, if you do two gigawatts at those prices, it starts getting a little cheap per gigawatt. But, anyway, that adds about 6 to 14% on revenue. That's not really the biggest driver.

Instead, Goldman Sachs thinks that plus monetizing some of the other tools they're coming out with could really be the revenue drivers. Because Meta and Zuck basically come out and say, "Look, we're providing businesses a business in a box.

We're providing ways for other people to monetize, which enables them to pay us more money."

And I think that's the thesis here with Zuck, at least per Goldman Sachs. Goldman Sachs suggests that a business in a box, and and I put some of these things in here, but could potentially be, "Hey, I want to run a business.

I want to be an entrepreneur. I want to run ads." Boom, base case for Meta. "I want content created." No problem, we'll design it for you. Or, no problem, we'll automatically link you with creators who can make shorts or video content or whatever, and they can have monetized brand partnerships all through AI with real people getting basically hooked up and connected, which in the creator economy, it's a lot of work to coordinate on, oh, I I need to do this ad read or you know, they need adjustments to the content or this is what they want you to say, this, that, or whatever.

Goldman is like, dude, they can make some freaking money on monetizing AI here by facilitating agentic AI behavior, and also entering into more neo cloud type agreements. And they believe that the current compute shortage will persist into 2028, which is obviously where we have the risk of, you know, concerns around oversupply coming where, you know, now more SpaceX stuff comes online, Google stuff comes online, Microsoft, Amazon, Meta, the Neo clouds, the Iren, the Enbis, the Coreweave, whatever. A lot of money going into this, right?

Uh and Zuck has kind of been hinting at this. He said in a meeting with uh in a conference call with um uh employees that sometimes they get offered ridiculous premiums to open up API services and that, you know, maybe they should really take advantage of of doing that.

So, they think that if management continues to demonstrate monetization while providing greater visibility into long-term AI economics, the stock's primary valuation discount could narrow materially.

So, you know, a lot of folks keep asking me, they're like, "Kevin, what's a good stock I can buy that has some serious upside potential?" Obviously, no guarantees. I don't know your personal financial situation, so I can't give you personalized advice, but when you look at Meta, I personally think there's some compelling components to it.

I believe that the stock has a fair value of around $1,300. My stock AI product thinks Meta has a fair value of $1,071. So, somewhere in that range suggests that the stock has like a two or two and a half X built into it and maybe not that far off.

Now, when we look at the actual fundies, the real fundamentals for the company, we can see where there is some concern. Spending did go up a little bit in this quarter compared to prior quarters.

We went a little Girls Gone Wild here. We got cost of goods sold up 33%, which shrunk their PP, their pricing power. R&D and SG&A SG&A nearly doubled. How? Why? Why did Maybe not the S part, but the G&A.

Why did G&A double and R&D double? The sales portion actually only went up about 15%. So, less than revenue, which is good. But, they're plowing more money into R&D, and they're plowing more money into uh their administrative expenses, and their cost of goods sold went up more than the revenue growth that they have.

All of those things kind of weakened some of their appearance of pricing power, and they set up an opportunity for Mark Zuckerberg to turn that crap around.

We're not seeing repurchases. We're not seeing dividends. We got $25 billion in debt, which is less than Salesforce has. Uh Salesforce literally borrowed an extra $25 billion to go buy back about $25 billion of their stock.

Uh their stock comp rose about 52%. Their net income's up 22%. They've got free cash flow about $15 billion in a 6-month period. That works out to about 30 in uh an annual period, but that's down from where it has been previously.

And unfortunately, markets think it's going to keep going down. If they do it stay at 30 for the year divided by their current market cap of 1470, it's about a 2% free cash flow yield.

So, it's nowhere near as juicy as like a Salesforce at a 7% free cash flow yield. But, it's still better than the cybersecurity plays. It's still better than Palantir. So, a lot of people think there's a real opportunity in Meta stock.

If you look at their balance sheet, it's not as indebted as some other companies. Some companies like Google and Microsoft have really gone deep deep into debt. They've actually got cash, marketables, and receivables of $112 billion.

Minus their bills of 56, that gives me free cash uh of 56. And they have long-term debts of about 83. I could pay most of that off with the extra 56 in free cash I have and 1 year of cash flow.

Because that'd be an extra 27 billion of long-term debts and obviously, the long-term leases that they have of another 26. Yeah, some people are a little worried about that regarding off-balance sheet risk, and that's exactly why the monetization of artificial intelligence could be really useful for them if they pivot in the strategy.

Of course, that's all hopium. You know, just because Deutsche Bank thinks they should do it, and New Street Research thinks they should do it, and Goldman thinks they should do it, doesn't mean that they're going to do it.

And if they don't, and they decide, "You know what? We're going to keep using this to build out more subscription-based services on tokens, which now they're charging for like token usage on WhatsApp."

I went on vacation to Europe, and and this is one of the reasons why when I go on vacation, I really like studying companies. What I notice is I go rent a car with like Six Rental Car.

I'll get a message on WhatsApp because everybody in Europe uses WhatsApp, apparently. Uh and and they're like, "Hey, your rental car is ready. You're this, that, or whatever. Message us here if you have questions."

And 90% of it is just their chatbot replying, but you could ask for a human, too. I'm like, "All right. All right, this is this is actually really functional. I don't have to call a stupid number or wait on hold.

I get a pretty rapid experience with a chatbot, and you know, not that long of a wait if I wait for a human." And and it's all through WhatsApp, which Facebook can monetize.

Barclays actually thinks that you could see $65 billion come to Facebook through API revenue. That'd be pretty good. That'd be like 20%, uh you know, just through APIs of their total revenue. Not bad.

And keeping in mind that they're the only hyperscaler without a monetization plan on their CapEx right now, does set them up for a rapid turnaround post lawsuits getting settled, and coming up with some kind of CapEx plan for monetization.

You know, Meta did say that their guidance remains unchanged after these lawsuits. Uh the you know, judge obviously still has to sign off on the settlement. Uh to really complete the settlement, you need TikTok and YouTube to really join.

I as a parent, a lot of people have asked me about these default protections and these parental controls. I'm a big fan of trusting in children, but if I can also by default set up reasonable restrictions, I'm okay with that.

I don't think those things are bad because you can disable these protections if you want, but it's kind of nice. Like, for example, I use, you know, my kids Microsoft, their PC computers have the What is it called?

The Microsoft family thing or whatever it's called. Is it family security or is that the Apple one? I can never get it straight. I think it's family security. I think that's the Windows one.

And so, I can get time that the computer has been used. I can set time limits. I can approve apps that they're installing. You know, like, I think that's cool as a parent for an 8-year-old and a 10-year-old and then, you know, five other 2-year-olds.

I think that's a great tool. I also like that I can set up a separate VLAN for them through my Ubiquity router and block things that they don't even know are blocked. It's just, oh, I guess that website's not working.

How weird. I don't think that's bad as a parent as much as I like to teach independence. It's like a little safety net. So, um, I'm okay with that. People have asked me about that and I think that's These are These are very reasonable restrictions.

Restrictions on how long you can control, modifiable by parents, being off between 6:00 a.m. uh, and midnight for users under 18. So, that's midnight to 6:00 a.m. which can be lifted by the parents.

These resolve all state claims, but unfortunately, there are still lawsuits from individuals and private parties, so it doesn't resolve everything.

But, when we look at the valuation of the company, I mean, there are various different ways you can value it depending on what kind of PEG ratio you want to use. If you take a, you know, 2.4 PEG, you're at about 12.70. If you take a 2.1 peg, you can get to $1,100.

But, we've got 15% growth forecast for the company with you know, a substantial amount of EPS for the end of next year. This is all pretty impressive. So, the company seems cheap.

On a technical base, obviously full disclosure exposed to it. On a technical basis, it also feels like you should enjoy the fact that there is what I call a legendary support line at $540.

So, 577 / 540, I am 6.8% away from what I call a legendary support line. Those are the strongest technical support lines that I ever have. And we've got here in the last 6 months, a bounce in March, a bounce in June, a bounce in July, and a bounce in August at 540.

While the lawsuits are gone going on. Now, the lawsuits are gone and we're only six Well, the biggest ones from the states. We're only 6.8% above that support line. This is setting up for a massive breakout.

And we've got this downtrending Well, well, this sort of converging trend. I wouldn't necessarily call it I mean, this is a downtrend, but this is a convergence right here. So, maybe it's possible it goes lower.

I personally think our floor is that 540 level. Poop hits the fan, we go into a recession, all bets are off, obviously. But, this is a nice technical level that could really set up for a bounce.

You've got fundamentals on your side. You've got optionality on your side. You've got the legendary support line on your side of 540. And in my opinion, that's a great opportunity to break up through this convergence.