RTX Q3 2026: How Much of the Earnings Surge Repeats?

The 60-second read

Five things to know

1. Growth extends across the portfolio. Sales reached $24.965B, with higher commercial aftermarket and defense activity. Backlog reached $340B.
2. The earnings surge includes a recovery benefit. Adjusted EPS of $2.26 includes $0.25 from IEEPA recovery. Removing only that benefit leaves $2.01, still above last year’s $1.70.
3. Cash timing cuts both ways. Quarterly FCF declined to $2.119B, while nine-month FCF rose to $6.306B. Contract-liability movements explain a large part of the quarterly reversal.
4. Pratt has both momentum and customer risk. Adjusted operating profit rose to $968M, but GAAP profit fell to $608M after a $363M customer-bankruptcy charge.
5. Valuation remains the constraint. Our three routes support $165 base value and $173.75 probability-weighted value versus $192.11 before the print. A stronger business need not be a cheap stock.

Three scenarios, one probability-weighted number

Analyst twelve-month valuation scenarios. Probabilities are judgments, not statistical forecasts.

Signposts: what would change our mind

Thresholds are our monitoring rules, not management commitments. The $132 entry screen is a 20% margin of safety to $165; a falling price alone does not establish a BUY.

The tape

The price anchor is $192.11, the October 19 regular-session close. The retrieved quote and daily-price history agree on that level. This packet does not treat that price as the market’s response to the October 20 release. The current session has not closed, and no verified post-print reaction is used in the valuation. The company scheduled its earnings call for 8:30 a.m. Eastern, or 7:30 a.m. Chicago time. The release is the source of the analysis; no call quotation or answer is inferred from a scheduled event.

The range in past reactions is context rather than a trading system. A positive earnings surprise does not uniquely determine a stock’s move, and an old reaction cannot substitute for today’s expectations. Our valuation uses a consistent diluted-share denominator. That makes its equity and enterprise proxies slightly larger than the provider’s market capitalization, which uses a different share basis. We show both rather than silently presenting them as the same measure.

The print

Revenue rose to $24.965B and adjusted EPS reached $2.26. FMP’s current earnings row agrees with those headline actuals and carries estimates of $24.10919B and $1.78. Computed against that snapshot, revenue beat by 3.5% and EPS by 27.0%. The EPS basis is provider-defined and should not be confused with the company’s GAAP $1.89. In particular, the $0.25 IEEPA benefit remains within adjusted EPS; the adjusted label does not mean every favorable unusual item has been removed.

Removing only the disclosed recovery from this quarter produces $2.01 per share, still 18.2% above last year’s adjusted $1.70. That is a useful recurrence test, not a replacement accounting measure and not a claim that all the remaining earnings are recurring. It also avoids the opposite error: calling the entire earnings improvement artificial. The company attributes growth to volume, mix and productivity across the businesses, with higher taxes and lower pension income partly offsetting those gains.

Management raised all four outlook measures. Forward adjusted EPS and free cash flow are not fully reconciled to GAAP because the company says some future excluded items cannot be reliably estimated without unreasonable effort. The packet retains those labels. A higher annual forecast is constructive evidence, but its midpoint should not be multiplied by a premium multiple without considering what generated the increase and which benefits will survive into the following year.

Segments and backlog

Collins sales increased to $8.382B. Commercial original equipment grew 23%, aftermarket 9% and defense 7% on the company’s organic comparison. Adjusted operating profit reached $1.788B and a 21.3% margin, including approximately $250M of IEEPA recovery. The prior-year GAAP comparison includes the actuation and flight-control business sale gain, so the reported and adjusted growth rates answer different questions. The volume improvement is valuable; the entire margin step-up is not a safe permanent run rate.

Pratt’s commercial aftermarket grew 21%, while commercial original equipment fell 5% and military rose 2%. Adjusted profit increased to $968M, including approximately $160M of IEEPA recovery. GAAP segment profit instead declined to $608M after a $363M customer-bankruptcy charge, primarily related to specific customer contract assets. A repair-driven aftermarket can generate revenue while customer credit risk still consumes economic value. We therefore monitor the profitability and collectability of that activity together.

Raytheon sales rose 14% to $8.031B and adjusted profit reached $1.134B, a 14.1% margin. The release identifies higher defense volume and improved productivity, including approximately $85M of benefits from an international-contract modification and an offset-obligation adjustment. Those benefits are different from IEEPA recovery and are not added to its disclosed EPS amount. Contract-level estimate improvements can be legitimate earnings while still requiring careful recurrence assumptions.

The $340B backlog includes $176B commercial and $164B defense, after more than $81B of new awards. Tomahawk and AMRAAM awards include multiyear undefinitized production contracts. Backlog demonstrates customer commitment, but profitability also depends on final terms, production capacity and delivery. The annual report defines backlog as remaining performance obligations on firm orders and excludes unexercised options. Its prior year-end conversion estimate is not automatically a forecast for the much larger current balance.

The profit bridge

Consolidated operating profit was $3.337B. It differs from the sum of the three segment profits because eliminations, corporate expenses, pension accounting and acquisition accounting sit outside those headline business results. The adjusted consolidated figure is $3.987B. The $650M net difference includes a $448M acquisition-accounting expense, offsetting restructuring credits, and a net collection of significant items. Treating the whole gap as cash savings would confuse accounting adjustments with spendable cash.

Non-service pension income adds $345M below operating profit, while net interest removes $381M. Taxes then remove $593M and minority interests $125M, leaving $2.583B for common owners. This distinction also explains a provider trap: historical FMP net-income rows generally use common-owner income, while its newest row uses consolidated income before minorities. We preserve primary definitions throughout the nine-quarter series instead of allowing that change of basis to manufacture growth.

The Pratt customer charge and investment gain partly offset each other within the broader adjustments. The former remains an economic risk even if management considers its nature unusual; the latter should not be extrapolated simply because it increased GAAP income. A good earnings-quality assessment asks whether the contract assets are collectible and whether the business can replace temporary benefits with production and service economics. It does not equate every exclusion with misconduct.

Earnings quality

Quarterly operating cash flow exceeds consolidated net income, but that single ratio does not settle recurrence. Receivables and contract assets together are substantial because aerospace and defense work recognizes revenue across long production and service cycles. Contract assets are not identical to overdue trade invoices. Adding them to receivables creates a useful economic exposure measure, provided the label is explicit and comparisons retain the same method.

Inventory reached $14.953B versus $13.364B at year-end. Higher output can justify more inventory, while delays or customer distress can make it harder to monetize. The cash-flow statement, rather than the balance-sheet difference alone, identifies the period’s inventory cash use because currency, portfolio and noncash movements can also change balances. The customer-bankruptcy charge makes this distinction particularly relevant at Pratt.

Stock compensation is modest relative to sales, but it is still a cost of ownership. We do not remove it again from the company’s reported cash-flow calculation and then silently change the share count. The valuation uses current diluted weighted-average shares, and the cash-flow history separately shows share-based retirement-plan matching contributions. Equity dilution, employee compensation and cash distributions are related economic questions with different accounting lines.

We do not attach a forensic fraud score to incomplete or mixed-basis inputs. The nine recurring checks are more useful here than a mechanical model whose definitions do not match long-cycle contract accounting. An improvement in headline margin should ultimately appear in common-owner earnings, collections and a stable financing requirement. That is the standard against which the next results should be judged.

Cash and the balance sheet

Quarterly free cash flow fell 47.4% to $2.119B, yet nine-month free cash flow increased 32.9% to $6.306B. Both statements are true. Last year’s third quarter benefited from a $986M contract-liability cash inflow; this quarter recorded a $980M outflow. The $1.966B swing is larger than the $1.771B decline in operating cash flow. Other lines partly offset it, so it would be wrong to describe the swing as the only cause or to ignore the stronger year-to-date position.

To reach the $9.0–9.2B annual free-cash-flow guide, the fourth quarter needs $2.694–2.894B. That is a concrete test rather than an assertion that a weak quarter makes the guide impossible. The current year-to-date total already includes approximately $300M of IEEPA cash recovery. Our valuation stress removes that disclosed recovery from the annual midpoint to avoid capitalizing it forever, while acknowledging that future operating improvement could replace it.

FMP’s current total-debt field omits current maturities while incorporating non-current operating leases. We therefore use the primary borrowing total. Net debt is meaningful, but subtracting it a second time from an equity cash-flow valuation would understate value: the selected cash-flow measure is already after interest. The enterprise-value proxy is shown separately for context.

Company free cash flow is operating cash flow less capital expenditure. It does not deduct every investing outflow; the quarter also includes $73M of additions to intangible assets. We retain the company definition for comparability, disclose that additional investment and demand a margin of safety. Cash dividends were $984M in the quarter and $2.882B over nine months, while reported common-stock repurchases were zero over the nine-month period. This is not a buyback-driven EPS story.

Three valuation routes

The valuation is intentionally explicit about recurrence. Our earnings baseline is the $7.70 full-year adjusted EPS midpoint less the disclosed $0.25 third-quarter IEEPA benefit, or $7.45. Our cash baseline is the $9.1B FCF midpoint less approximately $0.3B of related cash recovery, or $8.8B. These are analyst stress assumptions, not revised management guidance. They do not remove the customer-bankruptcy charge again because adjusted EPS already excludes that charge.

The three routes average $165.16; rounding to the nearest five dollars gives our $165 base value. That is 14.1% below the prior close. The models are related, so averaging them does not create three independent proofs. The difference between the earnings route and discounted-cash route highlights how much of the debate rests on conversion and the price of distant growth. The terminal value contributes most of the DCF, making the discount and long-run growth assumptions consequential.

At $192.11, the same cash model requires approximately 14.1% annual cash-flow growth for five years, holding the 8.5% discount rate and 3% terminal growth constant. That is possible, particularly if program economics normalize, but it exceeds our 8% base assumption. A reverse DCF describes what the price needs under chosen parameters; it is not the market’s literal forecast and it is not an independently observed growth rate.

High uncertainty calls for a 20% margin of safety, giving a $132 entry screen. The $225 bull scenario assumes $9 of forward normalized EPS at 25×; the $140 bear assumes $7 at 20×. Neither is company guidance. Our SELL rating is a relative valuation judgment for existing exposure, not a short-sale instruction or a claim of imminent distress. Stronger repeatable cash evidence could lift the cash routes; a cheaper price could improve prospective returns without changing the business.

The Street comparison

The retrieved target consensus is $231.36, with a $208–250 range. The latest target-news records predate this release: RBC’s October 15 target is $235, Bernstein’s October 15 target $214 and Deutsche Bank’s October 13 target $208. They are context for expectations, not confirmed responses to today’s earnings. An aggregated target is a set of opinions with different models, horizons and publication dates; it is not a company promise.

The annual 2026 snapshot has revenue of $96.282B and EPS of $7.2635, below the new company midpoints of $97.25B and $7.70. That gap is consistent with a forecast set that has not fully absorbed the new release, but we cannot infer each analyst’s revision timing. The 2027 provider EPS mean is $7.93029. Our bull case requires a stronger earnings trajectory than that snapshot, and our base case discounts the recurrence risk rather than copying a target consensus.

The disagreement is about price and repeatability, not whether commercial aviation and defense demand are growing. The provider’s positive ratings and our cautious valuation can coexist because the assumed growth, conversion and discount rates differ. We make those assumptions visible so readers can substitute a more optimistic or pessimistic view. Fresh analyst revisions should be compared by firm and date when actually available; this packet does not manufacture a post-call consensus.

Management and the prior call

The prior Q2 episode carried HOLD, a $190 base value and a $195 price anchor. Its trigger emphasized GTF aftermarket improvement and Pratt margins approaching the teens, alongside tariff and powder-metal risk. The new release provides genuine progress: Pratt aftermarket rose 21% and adjusted margin reached 10.5%. However, that margin includes approximately $160M of recovery, and GAAP profit absorbed a new customer exposure charge. We therefore grade the operating trigger as partial progress rather than a completed clean-margin inflection.

The lower base value is driven by a more explicit cash-recurrence framework and three displayed routes, not by pretending the results deteriorated across the board. We preserve the prior call instead of rewriting its history. The older local episode does not contain a complete packet signpost set for retrospective grading; the available metadata trigger is the comparison used here. Future packets can grade the numeric tests established above on a consistent basis.

Capital allocation also deserves context. The 2025 annual report describes an $11B repurchase authorization initiated in October 2023, with $615M remaining at year-end 2025. That is a dated authorization balance, not an assumption of current buying power or a forecast that repurchases will resume. The same annual report identifies IAE and IAE LLC as consolidated variable-interest entities with Pratt as primary beneficiary. Their ownership economics and minority claims matter even though RTX itself is ordinary US common equity.

1. How much of the IEEPA recovery is already reflected in the raised annual cash and EPS outlook, and what should recur?
2. What explains the specific customer exposure and how much additional contract-asset risk remains?
3. How does the new multiyear defense backlog convert into funded deliveries and margins?
4. Which Pratt margin gains remain after recovery benefits and mix effects?
5. What cash investment is needed to support the larger backlog without stretching suppliers or the balance sheet?

These are analytical questions, not invented call answers. Management’s execution language is supported by higher sales and the raised outlook, but the remaining tests are numerical. The annual audit is by PricewaterhouseCoopers, which reports serving as auditor since 1947. RTX had approximately 180,000 employees at year-end 2025, compared with 186,000 in 2024 and 185,000 in 2023. Those scale and governance facts provide context; they do not eliminate program or credit risk.

Ranked risks and counterarguments

The strongest bull argument is operational: an installed engine base supports service demand, Collins participates in higher aircraft output, and Raytheon has a larger multiyear defense opportunity. If those forces lift repeatable cash flow faster than our 8% assumption, the current premium can be earned. The raised outlook and improved nine-month cash flow are evidence in favor of that case. A cautious valuation should not erase those positives.

The strongest bear argument is that investors may pay a premium for a peak rate of earnings improvement that contains recovery and contract-accounting benefits. Customer distress can also reveal risk in assets whose revenue was recognized earlier. A large backlog does not guarantee attractive final margins or prompt collections. If the company needs more inventory and capital to deliver it, revenue growth can outrun cash growth even without an accounting problem.

Government business creates both stability and concentration. The annual report places direct US-government sales at 38% of 2025 revenue, excluding foreign military sales through the government. Appropriations, export permissions, procurement priorities and contract execution can affect timing and profitability. We do not turn geopolitical headlines into guaranteed awards or extrapolate a program announcement into cash before an actual contract and delivery support it.

Our judgment lands between an operational collapse thesis and an unquestioned premium-growth thesis. The business has improved; the price still needs a demanding cash outcome. Conviction is 3/5 because outcomes are sensitive to both execution and valuation assumptions. A failure to meet the annual cash guide or another material customer charge would weaken the base case. Clean margin gains and demonstrably stronger recurring cash would challenge the SELL call.

Catalysts and decision checkpoints

A catalyst is useful only if it can change a decision. We will compare the annual cash outcome with the explicit remaining requirement rather than celebrate or condemn a quarter in isolation. We will also distinguish the timing of customer advances from the profitability of the underlying contracts. A collection rebound that merely restores a prior timing imbalance is valuable liquidity, but it should not automatically receive the multiple of permanent incremental earnings.

The next review should retain the same model definitions: common earnings for equity ownership, consolidated net income for the operating-cash conversion check, borrowing debt separated from leases, and diluted shares for per-share valuation. Changing those definitions in response to a favorable headline would make the track record less informative. The signpost thresholds above are analyst choices and should be updated only with an explained economic reason, not retroactively moved to turn a miss into a pass.

Data, sources and method

Every quarter is taken from its actual earnings exhibit, including the separately reported fourth quarter. EPS and average shares are not derived by subtracting year-to-date averages. Complete income, balance-sheet, cash-flow, segment and adjustment tables are retained in the underlying dataset. The three-year cash-flow series uses the audited 2025 annual report’s comparative statements. Dollar amounts are in millions unless marked billions or per share; operating margins and days are computed using the definitions beside their charts.

Primary statements take precedence where provider parsing differs. Historical provider net income matches common-owner income, while the newest quarter uses income before minority interests. Provider debt and minority-equity fields also omit or mix primary lines. These discrepancies remain recorded in the research evidence. They are reasons to reconcile definitions, not evidence that the company’s statements fail to balance.

DCF means discounted cash flow. FCF here means the company’s operating cash flow less capital expenditures. Contract assets represent recognized revenue whose collection depends on contractual milestones or other conditions; contract liabilities generally reflect customer funding ahead of recognized revenue. IEEPA refers to the International Emergency Economic Powers Act. An adjusted figure is a company-defined non-GAAP measure and may still contain benefits whose recurrence deserves separate analysis.

Scenario probabilities, discount rates, multiples and growth rates are Charged Alpha judgments. The weighted scenario value differs from the base value because the bull and bear distances are asymmetric. No valuation is a guaranteed outcome. Sources were obtained for the October 20 release; the market anchor is explicitly the prior completed session. Later call comments, prices or guidance can change the conclusion and should be incorporated only after verification.