What actually happened
Headline miss: Q3 net income collapsed to $2.7B and EPS to $1.05, thanks to a one-time, non-cash tax charge of ~$15.9B tied to the new U.S. tax law. Strip that out and EPS would’ve been $7.25—comfortably above estimates. Revenue hit $51.24B (+26% YoY).
Margins (the real story): Operating margin was 40% vs 43% last year—solid, but moving the wrong way as expenses rose 32%.
Spending drumbeat: Capex guidance for 2025 was raised to $70–72B, and management warned 2026 spend likely grows “notably larger” with higher depreciation and cloud costs.
New debt coming: Meta is seeking at least $25B via a multi-tranche bond sale—only its second ever (after 2022).
Is there a “margin problem”?
Short term, yes—mix and timing, not demand.
Demand looks fine: Ad impressions +14% and price per ad +10%—a healthy combo.
Accounting vs. cash: The tax hit crushed GAAP profit but is non-cash, and Meta expects lower federal cash taxes going forward. That helps future cash margins even if GAAP optics look noisy.
Cost curve rising: The headwind is infrastructure—servers, data centers, and AI talent. Management telegraphed higher capex and expenses again in 2026, meaning depreciation will nibble at operating margins even if revenue keeps growing.
Net: It’s not a demand or competitive-margin collapse; it’s an investment cycle that compresses reported margins while Meta builds AI capacity.
Where could the bottom be?
No one rings a bell, but we can frame it.
If you treat Q3 as “normal” (exclude the tax hit), annualized EPS ≈ $29. At $600, that’s roughly 21× earnings; at $660–670 where shares traded post-print, it’s ~23–23.5×. That’s not distressed, but it’s below many AI peers given +26% revenue growth. (Simple math; EPS and revenue from Meta’s release.)
Technical bottoms often form where fundamentals offer a valuation floor. A ~20× multiple on ~$29 implies ~$580 as a rough “fundamental support” level. Not a prediction—just a sanity marker for long-term buyers.
Would I buy the dip at $600?
My take (not advice): a measured “yes,” in tranches, if your thesis is Meta monetizing AI across ads, Reels, Llama, and devices—and you can stomach heavy capex.
Why consider it:
Core ads engine is sturdy (pricing +10%, impressions +14%).
The tax charge is one-off and non-cash; cash taxes should ease.
Balance sheet remains strong, and the $25B bonds term out funding during the heaviest build-out years.
Why go slow:
Management flagged even higher spend in 2026; margin/FCF optics may get worse before they get better.
Regulatory/legal overhangs (EU ad changes; U.S. youth cases) can hit revenue or add costs.
Reality Labs still burns cash, and bond supply adds a modest headwind to equity appetite.
Playbook in one chart
Aggressive growth holder: Start/add near $600; add more on weakness toward $580 if the long-term AI thesis is intact.
Quality at a reasonable price: Wait for either capex clarity (2026 outlook stabilizes) or the stock to trade closer to 20× “normalized” EPS.
Risk-off: Avoid until you see opex/capex decelerate or margin troughs confirmed in guidance.
Bottom line: This drop looks more like valuation + spending shock than a broken business. If you believe Meta’s AI bet converts to revenue, buying near $600 in steps makes sense. If margin optics keep you up at night, keep it on the watchlist and let 2026 spending plans settle first.
Comments