Six months ago we called Microsoft overvalued at $490 and got roasted for it. It's now $373. Here's where it stands.
In December, I [posted](https://www.reddit.com/r/ValueInvesting/comments/1p9ks0t/understanding_michael_burrys_nvidia_short_the/) a breakdown in this sub arguing Microsoft was overvalued at $490, in the context of its capex spend (a concern Michael Burry had raised). I got a fair amount of criticism for those valuations.
Here is exactly what I said at the time:
"Assuming the P/E ratio remains at 34, the stock price would drop to $412-427... However, if investors also lose confidence in AI infrastructure returns, the P/E multiple could compress further..."
"... the Burry-adjusted fair value becomes $311.49 per share, suggesting Microsoft is overvalued by 37% at the current price of $491.92."
Six months later, MSFT trades at $373 after touching a low of $349 on 25 June. Over the same period, Microsoft reported another step-up in AI capex, and we have also refined how our model treats capex. So I thought it was a good time to revisit its valuation.
**Microsoft's capex update**
The concern I raised in December was not a one off. Microsoft's capital spending as a share of revenue has climbed relentlessly, and the cash flow cost is now very clear. Look at the last two columns in the following table together. Operating cash flow has more than doubled since FY2020, to $136 billion. But free cash flow has stalled: FY2025 free cash flow ($71.6B) actually came in below FY2024 ($74.1B), even though operating cash flow grew 15%. The entire difference is capex, which has roughly doubled as a share of revenue, from about 11% in FY2020 to 23% in FY2025.
MSFT is not alone. The trend is industry wide. Across 2026, the four big hyperscalers (Microsoft, Google, Amazon and Meta) have collectively guided to roughly $725 billion of capital spending, up about 77% year over year.
|Fiscal year|Capex / revenue|Operating cash flow|Free cash flow|
|:-|:-|:-|:-|
|FY2020|10.8%|$60.7B|$45.2B|
|FY2021|12.3%|$76.7B|$56.1B|
|FY2022|12.0%|$89.0B|$65.1B|
|FY2023|13.3%|$87.6B|$59.5B|
|FY2024|18.1%|$118.5B|$74.1B|
|FY2025|22.9%|$136.2B|$71.6B|
**How we refined our model**
In December, our valuation model treated capex the conventional way, roughly holding current intensity across the forecast and into the terminal value. We went back and questioned that assumption over the past six months. Assuming capex remains elevated forever is unrealistic: a company cannot spend 20% of revenue on capex in perpetuity. So we changed two things:
1. Changed capex to glide toward maintenance over the forecast period (for MSFT, it drops from 20% to 19%, 18%... and 10% over 10 years)
2. The terminal uses maintenance (10%), the steady-state level that a perpetuity can actually sustain.
This is a more logical picture than holding a single crude number to the end of time, but it impacts the valuations.
**What the model says now**
Holding everything else constant and flexing only the long-run capex assumption, here is Microsoft's intrinsic value across the realistic range
|What you assume long-run capex does|Fair value|vs Price|
|:-|:-|:-|
|Glides down to maintenance (our default model now)|$487|\+31%|
|Glides from the FY2026 30% pace down to maintenance|$464|\+24%|
|Stays elevated permanently (20% of revenue)|$364|\-2%|
|Stays at the FY2026 pace (30%) permanently|$228|\-39%|
Two things stand out. The near-term barely matters (most of a DCF's value is in the terminal). And the terminal assumption is the whole ballgame. At $373, the market is sitting right around the "stays elevated" value, so it has already re-priced Microsoft for permanently high capex.
Note: We have added a "Hold Capex Elevated" switch to the valuation dashboard on our platform for users to toggle. So if you believe that capex should stay elevated, you can turn it on and watch fair value move from $487 to $364.
**So where do we stand now?**
On the refined model, it now screens 31% undervalued at $487, assuming that capex normalizes. If it stays elevated, fair value is $364, about where it trades. So there is no margin of safety at the bear assumption, but it is fairly valued rather than expensive, with upside to $487 if capex normalizes.
Can it go to $228? Sure, if AI capex runs at 30% of revenue forever. But the risk/reward has flipped. At $490, everyone was bullish, and the setup was terrible. Now, it is looking interesting.
I'm not telling you to buy it. I'm saying the asymmetry today is the opposite of what it was in December. The whole call now comes down to one assumption: does AI capex normalize, or stay elevated forever? Where do you land?
*Disclaimer: This article is for educational purposes only and is not investment advice. The author and Stockoscope may hold positions in the securities mentioned. Always do your own research.*