Meta Q1 2026 Earnings: It Beat Every Number and the Stock Fell 7% Anyway

On April 29, 2026, Meta Platforms reported first-quarter revenue of $56.3 billion, up 33% year-over-year — the fastest growth rate since 2021 — and net income of $26.8 billion, up 61%. Both beat Wall Street's estimates comfortably. The stock fell about 7% in after-hours trading anyway, because buried in the same release was a new full-year 2026 capital-expenditure guide of $125–145 billion, up $10 billion at the midpoint from the range Meta gave just three months earlier. Investors didn't get spooked by weak results. They got spooked by how much stronger results still weren't enough to slow the spending.

The Headline Numbers​

Two things are true simultaneously in this table. First, the underlying business is accelerating: 33% revenue growth on a base already north of $40 billion a quarter is the kind of number that shouldn't be possible five years into a company's life as a $100B+ revenue business, and it's the fastest Meta has grown since before the 2022 ad-market slump. Second, GAAP net income is inflated by an $8.03 billion one-time tax benefit tied to U.S. Treasury Notice 2026-7, which changed the tax treatment of previously capitalized R&D costs under the Corporate Alternative Minimum Tax (CAMT) — a direct partial reversal of the $15.93 billion charge Meta took in Q3 2025 when the "One Big Beautiful Bill Act" (OBBBA) first triggered a valuation allowance against its deferred tax assets. Strip the benefit out and diluted EPS was roughly $7.31, still ahead of the $6.79 consensus estimate, but a very different-looking headline than the reported $10.44.

Revenue Deep Dive: Advertising Carries, Reality Labs Steadies​

Meta reports through two segments, Family of Apps (FoA) and Reality Labs (RL), and the mix hasn't fundamentally shifted — but the dollar magnitude has:

Within Family of Apps, advertising revenue was $55,024 million, essentially the entire segment and essentially the entire company — Meta's "other revenue" line (WhatsApp Business Platform fees, subscription products) was just $885 million. This is still fundamentally a one-product company: sell attention across Facebook, Instagram, WhatsApp, and Threads, convert it to ad revenue, and reinvest the proceeds into AI infrastructure that makes the ad product better targeted. The 33% growth rate says that flywheel is spinning faster than it has in years, most credibly attributed to AI-driven improvements in ad ranking and Advantage+ automated campaign tools that let advertisers spend more per dollar of measurable return.

Reality Labs remains what it has been since Meta started breaking it out in 2021: a persistent, large-magnitude loss generator, this quarter $4.0 billion on just $402 million of revenue. Across five full fiscal years the segment has never come close to profitability, and Q1 2026 doesn't change that trajectory — but the loss also hasn't been growing as a share of the company the way it once threatened to, because FoA's operating income base has grown so much larger around it.

The Margin Story: Two Tax Swings Bookending a Stable Core​

Layering in the last five fiscal years shows a business whose operating margin has been remarkably resilient through a layoff cycle, an AI capex ramp, and now two enormous one-off tax events:

FY2022's margin collapse is the "Year of Efficiency" story arriving early — $4.6 billion of restructuring charges (the November 2022 layoffs, a data-center pivot away from an over-built plan, facilities consolidation) landed directly in cost of revenue, R&D, marketing, and G&A, with no separate line to strip out. FY2023 carried a smaller $3.5 billion restructuring tail as the efficiency drive continued with a second, ~10,000-person layoff round. By FY2024, restructuring charges were down to $389 million and operating margin hit a five-year high of 42.2% — the clearest evidence that the 2022–2023 cost discipline was real and sticky, not just a one-year panic.

The net-margin line is where the two tax events show up and where the "core business" story gets obscured if you only read the bottom line. FY2025's operating margin (41.4%) was essentially identical to FY2024's, but net margin fell more than 7 points because the OBBBA charge pushed the effective tax rate from 12% to 30% in a single year. Q1 2026 runs the trick in reverse: operating margin held at 40.6%, right in line with the FY2025 average, but the Treasury Notice 2026-7 benefit actually made the effective tax rate negative for the quarter (a $5.0 billion net benefit against $21.8 billion of pre-tax income), pushing net margin to 47.6% — the highest of any period in this series, and not a repeatable one. Anyone modeling Meta off trailing-twelve-month net income right now is modeling a number that two unrelated one-off tax events, netting to a combined swing of roughly $24 billion, have made temporarily unrepresentative of the underlying business.

The One Big Question: Does the AI Capex Bet Pencil Out?​

The market's verdict on Q1 2026 wasn't about the quarter that just closed — it was about the $125–145 billion Meta now says it will spend on capex in calendar 2026, up from the $115–135 billion range given in Q4 2025's guidance. That $10 billion midpoint raise, on top of full-year 2026 total expense guidance of $162–169 billion, is what sent the stock down 7–8.6% the next trading day, even against a quarter that beat revenue, EPS, and daily-active-people estimates across the board.

The balance sheet shows exactly how this is being financed. Meta carried zero debt as recently as year-end 2021. It issued its first-ever bonds in August 2022 ($10 billion), then $18.5 billion cumulative face value by 2023, $29 billion by 2024, and a $30 billion single issuance in November 2025 brought the total to $59 billion face value ($58.7 billion carrying value) by year-end 2025 — essentially unchanged through Q1 2026 at $58.7 billion. Property and equipment, net, has grown in lockstep: $57.8 billion at the end of FY2021 to $194.8 billion at the end of Q1 2026, a 3.4x increase in a little over four years, with the FY2025-to-Q1-2026 jump alone ($176.4B → $194.8B) representing more sequential-quarter PP&E growth than most companies' entire balance sheets.

The bull case is that this is exactly the kind of spending Meta has earned the right to make: operating margin has held at 40%+ through the entire ramp, meaning the AI infrastructure investment isn't cannibalizing the current ad business — it's funded by it, with room to spare, and every dollar of better ad-ranking infrastructure shows up in the 33% revenue growth this quarter. The bear case is that "we're spending more because it's working" is also exactly what a company under competitive pressure to keep pace with Alphabet, Microsoft, and Amazon's own AI infrastructure races would say regardless of the true marginal return on the incremental dollar — and S&P Global Visible Alpha's Melissa Otto flagged precisely this: real, unresolved questions about ROI on the spending, not about whether Meta can afford it.

What Wall Street Thinks​

Sell-side reaction split along almost exactly this line — bullish on the growth, cautious on the capex trajectory:

• UBS raised its price target to $908 from $872, maintaining Buy, citing generative AI as the key driver of further EPS revisions and valuation expansion through 2026.
• Citizens reiterated Market Outperform with a $900 target.
• BofA Securities maintained Buy but trimmed its target to $820 from $885, explicitly citing the capex raise as a near-term earnings-quality concern.

The broader picture: 42 of 50 covering analysts rate Meta Buy, with 8 Hold and zero Sell, and a consensus 12-month price target near $837 — solidly bullish in aggregate, but with the capex question now clearly the single biggest swing factor in individual analysts' models, visible in the wide spread between BofA's trimmed $820 and UBS's raised $908.

Tracking a Capex Supercycle in Plain Text​

Modeling Meta in Beancount, the open-source double-entry accounting system, makes the capex story mechanically undeniable — you can watch Assets:NonCurrent:PropertyPlantEquipment and Liabilities:NonCurrent:LongTermDebt grow in the same rows, period over period, with every dollar traceable back to a 10-K or 10-Q.

Here is Q1 2026's income statement as a single zero-sum Beancount transaction — note the convention: Income accounts carry negative (credit) balances, Expenses carry positive (debit) balances:

; Q1 2026 Income Statement — three months ended March 31, 2026
; 1 MUSD = USD 1,000,000 | All figures in millions USD
; Check: −56,311 + 33,439 + 1,120 + (−5,021) + 26,773 = 0 ✓
 
2026-03-31 * "Meta Platforms Inc" "Q1 2026 Income Statement"
  Income:Revenue                           -56311 MUSD
  Expenses:CostOfRevenue                    10218 MUSD
  Expenses:ResearchAndDevelopment           17699 MUSD
  Expenses:MarketingAndSales                 2908 MUSD
  Expenses:GeneralAndAdministrative          2614 MUSD
  Expenses:OtherNet                          1120 MUSD  ; interest and other income (expense), net (debit)
  Expenses:IncomeTax                        -5021 MUSD  ; tax benefit — Treasury Notice 2026-7
  Equity:Adjustments                        26773 MUSD  ; net income offset (RE set by balance assertion)

The balance sheet lines that tell the real story sit right next to each other: Assets:NonCurrent:PropertyPlantEquipment at $194,776M against Liabilities:NonCurrent:LongTermDebt at $58,748M as of March 31, 2026 — a physical-infrastructure asset more than three times the size of the debt raised to help fund it, with the remainder coming from a business throwing off $83 billion of annual operating income. There's also a fast-growing balance sheet line worth watching that has nothing to do with capex directly: Assets:NonCurrent:NonMarketableEquityInvestments grew from roughly $6 billion, flat for three straight years (FY2022–FY2024), to $27.5 billion at the end of FY2025 and $28.4 billion by Q1 2026 — a 4.5x jump in a single year that the 10-K discloses only in aggregate, not by holding.

The complete FY2021–Q1 2026 ledger — every balance sheet and income statement, sourced line-by-line from SEC filings — is open and auditable:

The Multi-Year Arc: From Efficiency Drive to Capex Supercycle​

Four years bracket two distinct eras of Meta's capital allocation. FY2021–FY2023 is the "Year of Efficiency" era: revenue barely grew in 2022, layoffs and cost discipline dominated the story, and PP&E growth ($57.8B → $96.6B) was funded mostly out of a business that had essentially zero debt. FY2024–FY2025 is the capex-supercycle era: revenue nearly doubled to $201 billion, PP&E more than doubled again to $176 billion, and debt went from a rounding error to $58.7 billion — the fastest balance-sheet transformation in the company's history, and one that accelerated further, not slower, in the first quarter of 2026.

The Verdict: Bull vs. Bear​

The Bull Case:

• Q1 2026 revenue grew 33% YoY, the fastest rate since 2021, on a base already exceeding $200 billion annually — evidence AI-driven ad-ranking improvements are compounding, not plateauing
• Operating margin held at 40%+ through the entire capex ramp (39.6% → 41.4% → 40.6% from FY2021 to Q1 2026), meaning infrastructure spending is being funded by an expanding, not shrinking, profit base
• The "Year of Efficiency" cost discipline (FY2022–FY2023 restructuring) proved durable — FY2024's 42.2% operating margin, a five-year high, came after the layoffs, not despite them
• 42 of 50 covering analysts rate the stock Buy with zero Sell ratings, and UBS's raised $908 target reflects genuine conviction in the AI monetization thesis
• Long-term debt of $58.7 billion against $194.8 billion of PP&E and $83 billion of annual operating income leaves ample balance-sheet capacity for the spending Meta has guided to

The Bear Case:

• The stock fell 7–8.6% on the Q1 2026 print despite beating every headline estimate — the market is explicitly pricing capex risk higher than growth reward right now
• FY2026 capex guidance of $125–145 billion, raised $10 billion at the midpoint in a single quarter, shows spending discipline may be eroding just as it did before the 2022 "Year of Efficiency" correction
• Reality Labs has lost money in every full year since Meta began breaking it out, with no profitability path yet disclosed — a permanent $15–20 billion annual drag with an uncertain payoff horizon
• Two unrelated one-off tax events (a $15.93B OBBBA charge in Q3 2025, an $8.03B benefit in Q1 2026) have made trailing net income and EPS comparisons unreliable for at least another two quarters
• Non-marketable equity investments jumping 4.5x to $27.5 billion in a single year, with no per-holding disclosure, adds an opaque, fast-growing balance-sheet line that's hard for outside analysts to underwrite

Our Take: The market's reaction to Q1 2026 was rational, not panicked — a $10 billion guidance raise on top of an already-enormous capex base is a legitimate reason to reprice, even against a genuinely excellent quarter. But the ledger tells a more reassuring story than the one-day stock move does: operating margin has been essentially flat at 40%+ for three straight periods spanning the entire capex acceleration, debt remains a fraction of the PP&E it's helping fund, and the "efficiency" muscle Meta built in 2022–2023 hasn't atrophied — it's what's paying for 2026's spending without margin compression. The real risk isn't that Meta can't afford $135 billion of capex. It's that two years from now, the ledger needs to show that PP&E line converting into revenue growth at something close to the 33% this quarter delivered, rather than into a Reality-Labs-style perpetual cost center with no visible return. Q1 2026 is evidence for the bull case. FY2026's full-year numbers will be the actual test.