Netflix's 2026 third-quarter 10-Q (accession 0001065280-26-000291, filed 2026-10-21) shows a clear pattern: revenue and operating income rose every quarter, while diluted EPS fell from $1.23 in Q1 to $0.80 in Q2 before recovering to $0.87 in Q3. The reported line items explain most of that gap, and the explanation is mostly one line in Q1 that the data does not describe.
The operating business: steady, and slightly accelerating
Reported revenue went from $12.25B in Q1 to $12.56B in Q2 and $12.87B in Q3. Operating income went from $3.96B to $4.19B to $4.46B. Operating income grew a little faster than revenue in both sequential steps, so the operating margin widened from about 32.3% to 33.4% to 34.6%. Those margins are my arithmetic on the reported revenue and operating income lines. Cost of revenue rose from $5.89B to $6.04B to $6.08B, and G&A fell from $603M to $499M to $466M over the same period, which is the main source of the operating leverage.
The platform's own year-over-year metrics point the same way: revenue growth of about 15.9% and a 3-year revenue CAGR of about 12.6%. I did not pull prior-year quarters, so I cannot reconcile those annual figures to the quarterly series here. The fiscal 2025 MD&A summary reports revenue of $45.2B, operating income of $13.3B and an operating margin of 29.5%, so the 2026 quarterly margins are roughly five points above the prior-year full-year level.
| Line item (USD) | Q1 2026 | Q2 2026 | Q3 2026 |
|---|---|---|---|
| Revenue | $12.25B | $12.56B | $12.87B |
| Operating income | $3.96B | $4.19B | $4.46B |
| Operating margin (derived) | 32.3% | 33.4% | 34.6% |
| Nonoperating income (as tagged) | $2.85B | $0.05B | $0.04B |
| Interest expense | $0.26B | $0.18B | $0.18B |
| Pre-tax income | $6.55B | $4.07B | $4.32B |
| Income tax expense | $1.26B | $0.67B | $0.65B |
| Effective tax rate (derived) | 19.3% | 16.4% | 15.1% |
| Net income | $5.28B | $3.40B | $3.66B |
| Diluted weighted shares | 4.30B | 4.26B | 4.20B |
| Diluted EPS (reported) | $1.23 | $0.80 | $0.87 |
Table notes: NFLX fiscal 2026 quarterly line items from SEC filings. Margins and effective tax rates are arithmetic on those line items. Pre-tax income ties to operating income plus nonoperating income less interest expense in each quarter, and net income ties to pre-tax income less tax.
Why EPS did not follow operating income
The reconciliation is clean. Pre-tax income equals operating income, plus the tagged nonoperating income line, minus interest expense, within rounding in all three quarters. Net income then equals pre-tax income less tax. Diluted EPS equals net income divided by diluted shares.
The gap comes almost entirely from Q1. Nonoperating income was $2.85B in Q1 and about $0.05B in each of Q2 and Q3. Take that line out and Q1 pre-tax income would be roughly $3.70B, below Q2's reported $4.07B, which is the direction operating income implies. The Q1 EPS of $1.23 therefore rests on a single nonoperating item that is about 70% of Q1 pre-tax income on a reported basis.
The reported data does not say what that item is. The tag is a generic nonoperating income line, and the fiscal 2025 MD&A in the profile describes a different situation: interest and other income fell 35% that year, reflecting foreign exchange losses. That context does not explain a $2.85B gain in Q1 2026. A reader should read the other income note in the 10-Q before drawing any conclusion about the quality of Q1 earnings. The 10-Q body was not available to me, only the filing index and the extracted line items.
Tax is a secondary factor but it points in the same direction. Tax expense fell from $1.26B to $0.67B to $0.65B, and the effective rate fell from about 19.3% to 16.4% to 15.1%. The Q1 rate is elevated relative to Q2 and Q3, which is consistent with the nonoperating gain being taxed at a higher marginal rate, though the filing data does not confirm that. The lower rate in Q2 and Q3 helps Q3 EPS slightly, but it does not change the main story. Even at the Q2 and Q3 tax rates, EPS is driven by the much smaller nonoperating line in those quarters.
The share count: real but small
Diluted weighted shares fell from 4.30B in Q1 to 4.26B in Q2 and 4.20B in Q3, a decline of about 2.4% over two quarters. The balance sheet shows the buyback activity behind it: treasury stock at book value rose from $23.7B at Q1 to $28.4B at Q2 and $33.2B at Q3. The repurchase outflows tagged in the cash flow statement are $4.71B for Q2 and $4.80B for Q3. Those amounts are large against quarterly net income of $3.4B and $3.7B, so the company is returning more cash than it earned in each of those quarters. The cash flow line items may not share a consistent period basis across the three filings, so I would not infer a precise quarterly run-rate from them.
The per-share effect is modest. Applying Q3 net income of $3.66B to the Q1 diluted share count gives about $0.85, against the reported $0.87. The roughly two cents from the lower share count is real but small next to the $0.43 drop in EPS from Q1 to Q2, which is almost entirely the nonoperating swing. Debt moved in the other direction: long-term debt was $13.4B at Q1, $11.8B at Q2 and $12.7B at Q3, and short-term borrowings rose from $1.0B to about $2.5B across the same period. Buybacks in these quarters were funded partly by that borrowing and partly by operating cash flow.
Valuation and open items
At the platform-computed values shown above, the market is pricing Netflix at about 19 times trailing earnings and about 17.6 times EV/EBIT, with a free cash flow yield of about 4.3%. Those multiples are on earnings that include the Q1 nonoperating item, so the trailing P/E is flattered by that line relative to the operating run-rate. A reader who wants to value the operating business should strip out the nonoperating income and look at the Q2–Q3 run-rate. I am not offering a view on what the shares are worth.
Three items are open. First, the nature of the $2.85B Q1 nonoperating line needs to be read from the 10-Q. Second, the fiscal 2025 MD&A refers to a pending acquisition of Warner Bros. Discovery, with an equity value of about $72.0B and bridge and transaction facilities in place. The reported data does not show its 2026 status, and it could change the debt and share count picture materially. Third, the company stopped reporting subscriber counts in 2025, so the revenue trend cannot be split into membership and pricing contributions from this data set. No segment data was returned for fiscal 2026.
The quarterly picture is simple once the nonoperating line is set aside: revenue growing about 2.5% a quarter, operating income growing about 6% a quarter, margins widening, and a modest tailwind from buybacks. The reported EPS decline from Q1 to Q2 is not an operating story. It is a question about one line in the Q1 income statement, and that is the question a careful reader should answer before drawing conclusions about earnings power.