Ad strength intact but a looming >$100B capex plan raises the bar on Meta’s payoff

What happened

Meta’s advertising engine remains the core profit center: recent quarterly data show double‑digit revenue growth with rising ad impressions and price per ad, underpinning high margins and very strong cash generation. At the same time, a new report flags capital expenditures could surge to more than $100 billion next year — a step‑change from the ~$39 billion capex run‑rate in 2024 and consistent with DeepValue’s warnings about aggressive AI/data‑center spending (Louisiana facility, Scale AI) and Reality Labs losses. If management follows through, the company’s free‑cash‑flow profile and consolidated margins would be materially weaker in the near term, converting a relatively asset‑light, high‑margin ad business into a far more capital‑intensive enterprise with persistent cash burn. That outcome is not impossible — Meta frames these outlays as long‑duration, strategic investments — but the firm’s dual‑class governance and past willingness to prioritize platform bets mean shareholders may see extended heavy spending even without clear ROI. Given a current valuation that already embeds robust growth (c. 26x trailing P/E and a DCF base case modestly below price), a credible >$100B capex trajectory meaningfully reduces the margin of safety and argues for waiting for clearer evidence of returns before upgrading to a buy.

Implication

Monitor five concrete items before changing exposure: (1) confirmed capex guidance and quarterly capex run‑rate versus the >$100B figure; (2) operating cash flow and free cash flow trends to see the cash impact; (3) Reality Labs revenue and losses to judge whether the platform option is de‑risking; (4) measurable ROI from AI infrastructure (ad ARPU, improved targeting, new monetization paths) and specific milestones on the Louisiana data center/Scale AI; and (5) changes to buyback/dividend policy or debt levels. If capex proves transitory or demonstrably productive — lifting ARPU or creating new profitable revenue streams — the valuation case could revive; absent that proof, heavy spending will compress margins, reduce capital returns and justify a lower fair value. In short, evidence of ROI or a material price weakness is required before increasing allocation; absent that, maintain a cautious/underweight posture.

Thesis delta

Previously our stance was WAIT: Meta’s strong ad franchise and cash generation were balanced against heavy Reality Labs losses and elevated AI capex, leaving upside and downside roughly matched. The new article’s >$100 billion capex claim, if accurate, raises the probability that near‑term cash generation will be meaningfully impaired and that the company will sustain much higher capital intensity than assumed, shifting the risk/reward toward the downside. We are therefore more cautious: the core ad moat still protects downside to an extent, but valuation now looks less forgiving without transparent capex discipline and demonstrable ROI.