Stock Review: Microsoft
Microsoft reported earnings on April 30th for the third quarter of their fiscal year. Year-over-year revenue growth was 13%, with an operating margin of 45.6%. For a company generating $270 billion in trailing 12-month sales, this level of growth and profitability is exceedingly rare.
The Windows operating system and software tools like Word, Excel, and PowerPoint are still foundational to Microsoft’s business. But the company today is almost indistinguishable from the asset-light software company that defined the PC era of the 1980s and 1990s. Microsoft today is less a software company and more a global infrastructure platform powering the digital enterprise.
Here’s CEO Satya Nadella on the most recent earnings call in response to an analyst’s question about demand for cloud migrations:
“And if I step back, and Amy and I talk a lot about this, this time around, there’s nothing certain for sure in the future, except for one thing, which is our largest business is our infrastructure business. And the good news here is the next big platform shift [AI] builds on that.
It’s not a complete rebuild, having gone through some of these platform shift, where you have to come out on the other side with a full rebuild. If there is good news here, it’s that we have a good business in Azure that continues to grow, and the new platform depends on that. We want to stay disciplined and execute super well on that.”
While there are significant differences between an infrastructure company and a traditional software company, the common thread that runs through Microsoft’s history is that they have always served other businesses. That enterprise-first DNA still defines Microsoft’s strategy. Where Apple excels at running a consumer platform, Microsoft has become indispensable to enterprise infrastructure.
Microsoft’s evolution from a software company focused on local servers and PCs to a cloud infrastructure provider was far from guaranteed. But in hindsight, the shift feels natural—especially when you consider Microsoft’s long history of building and selling platforms, no matter how they are constructed, to enterprise customers. What began with Windows has grown into Azure, as the internet and SaaS (software-as-a-service) have come to define the modern enterprise.
Today, with the rise of generative AI, Microsoft finds itself at the center of a new infrastructure race—one that could deepen its strategic moat even further.
Let’s look at the assets on Microsoft’s balance sheet to better understand the transformation of their business over time. Here is a chart of property, plant, and equipment (PP&E) as a percentage of total assets going back 25 years:
This upward trend highlights just how capital-intensive Microsoft’s new reality has become.
This transformation has been underway for a decade under Nadella’s leadership, with the “Intelligent Cloud” business segment (Azure) quintupling sales over the last ten years to an annual run-rate of over $100 billion today.
The trends that gave rise to cloud computing in the 2010s are accelerating at an even faster clip in the 2020s with the rise of large language models (LLMs) and other AI applications. AI workloads require a tremendous amount of compute for both training (building a foundation model) and inference (ongoing use of the model). This translates into incredible demand for data center capacity that a company like Microsoft might use for their own purposes or rent out to an Azure customer.
The chart below shows that aggregate capital expenditures at companies like Microsoft, Alphabet, and Meta – all heavily involved in the build out of AI infrastructure – have exploded from about $5 billion (quarterly) in 2015 to over $45 billion during the first quarter of 2025. I excluded Amazon from the chart because their capital expenditures for the retail business distort the numbers for cloud computing, but they are also investing aggressively in data center capacity.
For Microsoft’s shareholders, the transition from a software company to an infrastructure company has, thus far, been rewarding. Over the 10 years ending 5/8/2025, the stock returned 26.7% annually, including dividends. Expansion of the PE multiple has contributed about 5.5% per year to that total return. Earnings grew 17% annually, with the remainder of the return coming from a growing dividend and stock buybacks.
In addition to being a major tailwind for growth, Microsoft’s Intelligent Cloud (IC) is incredibly profitable when you look at segment operating income. For the most recent quarter, IC posted $42 billion of operating income on $100 billion of sales, a 42% operating margin.
The question investors must grapple with today is: At what cost are those profits being generated? Microsoft’s shift to a capital-intensive model raises important questions about the durability of margins, the efficiency of capital reinvestment, and the underlying quality of reported earnings.
On the surface, Microsoft’s profitability and ROIC look healthy even as they’ve dramatically ramped up investments in physical equipment and data centers. On an absolute basis, a simple measure of ROIC (net operating profit after tax divided by average invested capital) over the last 12 months is a respectable 21.9%. However, a few things are worth monitoring:
- The trend in ROIC may matter more than the absolute level. If returns on average invested capital drift lower, it may mean that the company is destroying value with the incremental capital that’s being reinvested back into the company. If this trend were to persist for some time, investors may not be as willing to pay ~30x earnings to own shares in Microsoft.
- The quality of Microsoft’s reported earnings may deteriorate as they aggressively invest in cloud data centers and AI. The company must make accounting decisions about how to depreciate the useful life of capital investments. Operating profits could be overstated if the depreciation schedule does not accurately reflect the ongoing investments that will be required to maintain data center investments. For example, in July 2022, the company changed the useful life of server and networking equipment from four years to six years.
These are classic transition risks—ROIC compression, free cash flow lag, and accounting distortions—all worth watching, even for a business that looks dominant on the surface.
With those concerns in mind, it still makes sense for Microsoft to be aggressively investing in cloud and AI infrastructure. Moreover, Microsoft’s moat doesn’t rest on data centers alone. It’s the interplay between software, services, and physical infrastructure that sets them apart.
For more context, I’ll highlight another quote from the most recent earnings call:
“Amy Hood (CFO): And over time, you see software efficiencies and hardware efficiencies build on themselves. And you saw that process for us, for, goodness, now, quite a long time. And what Satya’s talking about is how quickly that’s happening on the AI side of the business, and you add to that model diversity. Think about the same levers, plus model efficiency, those compound. Now, the one thing that’s a little different this time is just the pace.
And so, when you’re seeing that happen, pace in terms of efficiency time, but also pace in terms of the buildout, so it can mask some of the progress. But we are working hard across all of the teams, hardware, software, even the build teams, to get things in place as quickly as possible, dock to live times. All of that is improving, and all of that actually is benefiting us. And I’ll go ahead and say our margins on the AI side of the business are better than they were at this point, by far than when we went through the same transition in the server to cloud transition.
Satya Nadella: Yeah, I mean, I think at a macro level, I think the way to think about this is, you can ask the question, what’s the difference between a hosting business and a hyperscale business? It’s software. That’s, I think, the gist of it. Yes, for sure, it’s a capital intensive business, but capital efficiency comes from that system wide software optimization. And that’s what makes the hyperscale business attractive, and that’s what we want to just keep executing super well on.”
What Microsoft is describing here is not just scale but improving capital efficiency across the AI stack. Hardware and software teams are working in tandem to reduce build times, improve model performance, and drive better utilization of data center infrastructure. That’s critical in order for Microsoft’s reported profits to translate into durable, high-quality free cash flow.
As I wrote last summer in Hardware is Eating the World, investors take for granted the idea that asset-light software businesses are an ideal investment. The fixed costs in a software business are typically quite low, so there’s a tremendous amount of operating leverage and profit in a software business if it successfully scales.
What if that’s no longer true in a world dominated by AI? Many companies are starting to report the amount of code that their engineers are generating with AI itself. If the marginal cost of code approaches zero—tied only to energy and compute—then much of what once made software special may no longer hold. The data underneath a software tool will likely be valuable in an AI world, but the code wrapped around it may not be as economically valuable as it was in the early 2000s.
If that’s true, the winners in an AI era may not be pure-play software companies—but the owners of compute, data, and distribution.
There’s a knee jerk reaction in the investment community to see the rising capital intensity underneath a company like Microsoft as a risk. To be sure, it is a risk factor! Yet the infrastructure business and the friction of building in the real world may create an even wider moat around their business. The real risk may not be Microsoft’s capital intensity—but everyone else’s inability to match it.
Advisory Services offered through Peak Asset Management, LLC, an SEC registered investment advisor. The opinions expressed and material provided are for general information, and they should not be considered a solicitation for the purchase or sale of any security. The information in this material is not intended as tax or legal advice. Please consult legal or tax professionals for specific information regarding your individual situation. This content is developed from sources believed to be providing accurate information and may have been developed and produced by a third party to provide information on a topic that may be of interest. This third party is not affiliated with Peak Asset Management. It is not our intention to state or imply in any manner that past results are an indication of future performance. Copyright © 2025 Peak Asset Management
