Earlier this week, I posted a breakdown here revisiting [Microsoft's](https://www.reddit.com/r/ValueInvesting/comments/1uio0by/six_months_ago_we_called_microsoft_overvalued_at/) valuation in light of its AI capex. Here's the same analysis on Meta. I won't repeat the methodology, it's all in the Microsoft post. Just the Meta setup and results.
**Capex**
Meta's capital spending is the heaviest of any hyperscaler. In FY2025 it spent $70B on capex, about 35% of revenue, up 87% in a single year. As a result, operating cash flow grew 27% last year, but free cash flow actually fell 15% (from $54B to $46B).
Interestingly, for Meta this has happened once before. In 2022 a capex surge plus a weak ad market took Meta's free cash flow to $19B and the stock under $90, before it ran back to $790. So capex panic is not new for Meta.
|Fiscal year|Capex|Capex / revenue|Operating cash flow|Free cash flow|
|:-|:-|:-|:-|:-|
|FY2020|$15.1B|17.6%|$38.7B|$23.6B|
|FY2021|$18.6B|15.7%|$57.7B|$39.1B|
|FY2022|$31.4B|27.0%|$50.5B|**$19.0B**|
|FY2023|$27.3B|20.2%|$71.1B|$43.8B|
|FY2024|$37.3B|22.6%|$91.3B|$54.1B|
|FY2025|$69.7B|34.7%|$115.8B|**$46.1B**|
**Valuation**
Similar to MSFT, instead of holding today's elevated capex flat forever, the model glides it down toward maintenance (roughly depreciation, 11% of revenue for Meta) over the forecast, with the terminal at maintenance. This is what you get holding everything else constant and flexing only the capex assumption (price $583):
|What you assume long-run capex does|Fair value|vs Price|
|:-|:-|:-|
|Glides down to maintenance (default)|$659|\+13%|
|Glides from the FY2026 35% pace down to maintenance|$622|\+7%|
|Stays elevated permanently (28% of revenue)|$331|\-43%|
|Stays at the FY2025 pace (35%) permanently|$160|\-73%|
So Meta appears fairly valued on the default model (13% undervalued at $659). But that assumes the build-out is temporary. If you think 28-35% of revenue on capex is permanent, fair value changes dramatically.
Disclaimer: This 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.