저자: valbitrage · 외국인
요약 보기
빌 애크먼이 2월부터 마이크로소프트 주식을 매수해 565만주 보유(평균 390~415달러). 현재 주가는 적정가치 범위 상단 근처로 평가. 클라우드·AI 성장(Azure 40% 성장, Copilot 유료좌석 250% 증가, AI 연매출 370억 달러, RPO 6270억 달러)을 근거로 AI 투자 성과 낙관. 주된 리스크는 CapEx 대비 자본수익률(ROIC) 하락과 클라우드 마진(60%대중반) 회복 실행 여부. PE 기준으로는 적정가치가 더 높다고 제시.
다듬은 글
Super investor Bill Ackman has been quietly buying Microsoft shares since February of this year. He now holds 5.65 million shares bought at an average cost of somewhere between 390 and 415 dollars per share. Now, as of today, the stock is still trading not far from those prices. And so, the key question is, does the current price offer a compelling entry point for the stock?
That's what we're going to try to answer in this video. Now, as always, this is not personal investment advice. This is just our opinion based on the research and analysis that we've done. All right, let's get into it. First, let's quickly go over what Microsoft does. Because whenever we evaluate a stock to understand where it's going, we need to understand what the business does and how it makes money.
Microsoft's business is primarily focused on software and cloud computing. And there are three segments to the business. The first segment is productivity and business processes. This covers products like Microsoft 365, Microsoft Teams, LinkedIn, and Dynamics 365. The second segment is the intelligent cloud segment.
That covers products like Azure, SQL, Windows servers, GitHub, and Nuance AI solutions. And then we also have the personal computing segment, which covers Windows licensing, Surface devices, PCs, tablets, Xbox gaming consoles, and Bing search services. Now, the productivity and business processes segment is still the main segment of the business, and it currently accounts for roughly 42% of total revenue.
The intelligent cloud segment is the fastest growing, and it's about to outpace the main segment. Microsoft expects that to happen next quarter. That part of the business accounts for another 42% of revenue. And finally, the personal computing segment is a much smaller portion of their business. That only makes up about 16% of total revenues.
And so, what's the business model? How does Microsoft make its money? Well, first, it sells those software products that we all know, like Microsoft 365, Teams, and Dynamics, and charges a recurring subscription fee for each license. Those licenses are typically structured as monthly or annual commitments per user, which is a pretty favorable and predictable pricing structure for Microsoft.
This part of the business, the productivity segment, is also the highest margin, which means it's the most profitable, and it still generates most of the company's total profits. The other way Microsoft makes money is by charging for those cloud computing services via Azure on a usage-based pay-as-you-go type of model.
That pricing model is less predictable and also less profitable than that of the productivity segment. But as we mentioned, it's the fastest growing and it's going to become the main driver of revenue soon. Microsoft now also makes money off of its AI products, which are embedded into all of its business segments.
And like Azure, those are going to be priced based on a usage pay-as-you-go type of business model. All right, so all of this is very important to keep in mind as we think about Microsoft's future. But before we get to that, let's quickly take a look at how they've been doing historically. Their recent historical performance has been solid.
Revenue CAGR at 15.3% and EPS CAGR at 18.7% since 2023. And the long-term trends look absolutely fantastic. We have this beautiful revenue arc over the last 10 years. Really can't do much better than this. Gross margins are very stable, operating leverage is also improving, and net margins are trending upwards over time.
EPS growth closely follows revenue, and free cash flow also had a nice upward trend as well. And we see here that net debt is going down as well. There's really not much off here. The historical performance has been great. Now, their most recent performance has also been fantastic. In the last quarter, they posted a double beat on analyst expectations.
Total revenue grew by 18% exceeding analyst expectations by about 2%, and EPS grew by 21% which also exceeded analyst expectations by about 5%. And so if their performance has been this good, then why has the stock sold off? And why is the market afraid to hold Microsoft at this time? Well, if we look carefully here, we can see that their ROIC, their return on invested capital, has been decreasing consistently over the past 3 years, although it still remains very high.
The reason for this is that Microsoft is taking a lot of its cash and deploying it towards AI investments across all of its segments. And it's investing a lot in AI. They're planning to spend over $40 billion per quarter on AI for a total of $190 billion throughout 2026. That's nearly triple the amount they spent in 2025.
And so the major fear the market is experiencing right now is that these investments may not translate into real returns that will generate additional value for shareholders. So when we think about the future of Microsoft stock, this is where we need to focus our attention. Stocks only grow when the business generates excess returns on capital.
If it doesn't generate those excess returns, then the stock will not grow or at least it shouldn't really have a compelling reason to grow. And so when we think about Microsoft, we need to consider what needs to happen over the next 3 to 5 years in order for those investments to be worthwhile. And there are really mainly two things.
First, the areas Microsoft is investing in, namely AI productization and Azure, need to compound at at least an annual growth rate of 35%. If they don't, then the increase in revenue and profit won't be enough to justify the spend. The second thing that needs to happen is that in the long run, Microsoft will have to reduce the amount it keeps spending in order to provide the same cloud and AI services to customers.
Right now, the gross margins in the cloud segment have dipped into the mid-60s. But in the long run, those margins aren't acceptable. Microsoft will have to achieve economies of scale on its AI data centers and lower the cost of running them so that cloud growth margins can expand back upwards into the low 70s.
And so, the risk here is mainly execution and the open question is whether these targets, both for cloud and AI growth, also for long-term cost efficiency, are these targets actually attainable? Well, Microsoft's most recent quarterly performance is showing us that those AI investments may be worth it.
You see, the vast majority of their growth in the last quarter was driven by growth in the cloud segment and also in the adoption of AI-driven products like Microsoft 365 Copilot. Azure revenue growth accelerated to 40% year-over-year and Copilot growth saw a 250% increase in paid seats year-over-year.
Beyond that, their AI business also surpassed an annual revenue run rate of $37 billion. That's a 123% year-over-year increase. Barring any major constraints to the business, that type of run rate proves the sheer velocity of AI adoption and demand that Microsoft is currently experiencing. And even more importantly, at least from a long-term perspective, Microsoft's RPO backlog reached $627 billion.
In other words, they have signed deals that guarantee them $627 billion in future revenue. And a lot of that backlog is captured by Azure. Again, this is proof that the AI boom is creating demand that is a true accelerant for the company. Now, why are these figures important? Well, first off, they show that Microsoft is currently meeting the demanding growth rates that its CapEx requires.
And what's important to note here is that we're still quite early on in AI adoption. I'm personally of the camp that spending on AI will continue to increase for two reasons. First, the technology is still developing and there are far more advances to be made than what is currently available. And second, people are still trying to figure out how to use AI and what use cases provide the most compelling force multipliers for their operations.
I think that as the technology improves and as the rate of learning among users also improves, those use cases will not only become more evident, but they'll also expand in number. I personally have a hard time seeing how growth can slow down when we're this early on in the technological shift and also especially over the long run.
When you look at any other technological shift historically, what typically happens is that people do overspend and then retreat. But after they retreat, spending starts to increase again, but at a more gradual and steady rate. That's because after the initial frenzy, the new tech becomes embedded in workflows and operations as an essential utility.
I think the same thing is going to happen with AI. Now, the other thing that will help Microsoft profit from AI adoption is their shifting pricing strategy. A lot of their AI business, especially the AI components of the productivity segment, currently run on a per seat subscription revenue model. But they're going to be changing that and they're going to start charging customers on a usage basis.
This pricing will be much more favorable to them from a monetization perspective. Of course, there is a risk that customers may curb spending if costs become too high. But unlike other companies whose revenue models are being completely disrupted by pricing changes, the effects on Microsoft's business model will be far more tapered.
And the reason is this, the intelligent cloud segment, which they're banking on being their main segment soon, is already 100% consumption based. So, that part of the business isn't getting affected at all by these pricing changes. The part of the business that does get affected is a productivity segment.
But again, the revenue model isn't shifting for that entire segment, just for the AI driven components. Microsoft will continue charging base per seat prices for its traditional software products and they will add on usage based pricing for the AI layers. And so, the pricing shift isn't a complete revamp to Microsoft's business model as it is for many other software firms.
In our view, as long as their product capabilities improve, given that they're already highly embedded in enterprise workflows, we think that these pricing changes will better support the top line. Now, as we mentioned, the second thing Microsoft needs to do in order to ensure that its CapEx doesn't become a drag on future business performance is to ensure that it scales spending sustainably.
In the future, its gross margins need to improve and rebound back to those 70% levels. Now, while it's not immediately evident that this will happen, what is encouraging is that the company is already being careful with its spending, and they're trying to make sure that every dollar they're spending goes farther.
In the last quarter, Microsoft did very well to squeeze as much computer as they could out of all the available square footage they have in their data centers. They also reduced costs significantly by pushing their own custom chips into production. Management was happy with these operational improvements and efficiency gains.
And this demonstrates that the spending is not only happening in a very controlled and calculated manner, but that a lot of the investments are being directed towards vertical integration efforts across their tech stack that will drive costs down in the future. So, in short, yes, the CapEx spending is high, and Microsoft will need to execute successfully on some lofty targets.
But the data we're looking at suggests that everything is moving forward in the right direction, and that the targets they have in front of them should be attainable. Moving on, let's take a look at their key metrics. And first, let's look at their returns on capital. Microsoft has a great history of producing awesome returns on capital.
That right here should provide investors with a bit of confidence regarding their AI spending. Microsoft over the last 10 years and over its entire lifetime has consistently made bets in the right direction, and we think it's very reasonable to assume that the CapEx spending now is going to follow a similar trajectory.
In addition to that, their revenue and EPS CAGRs are not bad as well. Again, this is a mature giant company. And so, within the software space, where growth early on tends to be very, very high, this performance is acceptable given where they are in their life cycle. Their balance sheet also remains fortress-like.
They have strong liquidity, strong solvency, and the operational ability to handle long-term debt very well. So, overall, their financial position still looks incredible. So, in terms of valuation, right now, we think that Microsoft is trading just above its fair value range, at around a 10% premium to the midpoint of that range, which is around $390.
You can see here that the upper end of that fair value range is $415. So, at the current price, we're really close to that fair value. Now, the multiples that are driving this fair value calculation are a PE of 29, an EV to revenue of around 7.5, and an EV to free cash flow of around 33. We weight these with an 18 41 41 split.
These multiples are all data-driven. We took the 7-year industry medians for each of these and adjusted them upwards for Microsoft's immediate peers' multiples, and also for their own historical multiples, as well. There's very high agreement between the EV to revenue and EV to free cash flow fair values, but the PE ratio gives us a fair value that's about 45% higher than the final assessment.
And so, if that's the metric you want to rely on more, then Microsoft arguably has a higher fair value than what our models are indicating here. All right. So, let's do a quick recap of the stock and provide our final judgment. So, first off, Microsoft is a powerhouse of a company with a highly diversified product line that is highly embedded in both retail and enterprise settings.
This company and its products are not going anywhere. Many software companies will be disrupted negatively by AI, but Microsoft is not one of them. Windows, Microsoft 365, Dynamics 365, Teams, all of these products are sticking around. They're not going to be replaced. Other than Amazon's AWS, we're not too concerned about competitive pressures within the industry as well.
We don't see that as a significant threat to Microsoft's business. On the contrary, we think that AI is a force multiplier for Microsoft. And so the risks they're facing and the challenges they're facing are mainly execution They need to translate those massive capital expenditures into actual returns for shareholders.
Given how much they're spending, the targets they need to meet are high. But the data we have shows that Microsoft is demonstrating the ability to meet those targets well. We also think that their ability to meet those targets in the future is not only possible, but it's also predictable. Although their pricing model is changing, it's not affecting the vast majority of the business as is happening with many other software firms.
It's fast-growing arm Azure already runs on the usage-based pricing model. And its traditional software products will continue to follow the seat-based pricing, which in turn provides a predictable hedge around some of the other pricing changes. Now, what will drive the business forward in the future is AI usage intensity.
As we mentioned previously, we only see that increasing in the future given how early on we are in this technological shift. Yes, there may be some overspending from customers and then a pullback. But in the long run, AI is going to become a utility technology that companies and organizations will have to use in order to keep up with the demands of the new economy.
Now, on price, we preferred Microsoft when it was trading closer to $350. But even now, we think that the stock is trading within a reasonable range of its fair value. So, if you're looking to buy the stock for the long run, we don't think the price is unreasonable as it is with many other companies.
All right, guys. So, that does it. That's what we think of Microsoft stock. If you like this video and you found it useful, hit the the button below. We'll keep the analysis coming your way. If you're looking for a platform that can help you do all this type of analysis by yourself, check out val-u-tra.com.
That's what we use to do this. It's free to sign up, so definitely go check it out. All right, we'll see you at the next one.
Comments 0