Report contents
At a glance
- Business model: NXP earns revenue shipping analog, mixed-signal, and embedded processing chips, mainly into automotive and industrial sockets qualified over one to three years and rarely swapped mid-program. That design-in switching cost sustains mid-50s gross margins and a 27% operating margin, but the obstacle is that growth depends on content per vehicle outrunning annual price-downs from concentrated buyers.
- Financial quality: Free cash flow was positive every fiscal year from 2021 through 2025, totaling $11.7 billion against $12.0 billion of cumulative net income, with capex at only 4–9% of revenue. Balance-sheet capacity is more mixed: goodwill of $10.3 billion is roughly 90% of equity, and liabilities were 55.9% of assets in Q2 FY2026, after a Q1 FY2026 below-the-line item lifted net income above operating income.
Business model
I. Business model: design-in wins, then the cycle sets the year
NXP sells chips. Revenue arrives as purchase orders and is recognized when product ships — not as a subscription, not as licensing. That one fact shapes the rest. Results track customers' production schedules, prices are negotiated program by program rather than reset each quarter, and a soft quarter at an automaker reaches NXP with a lag.
The reference document is the FY2025 Form 10-K for the fiscal year ended December 31, 2025, filed in February 2026. One limitation belongs up front: the excerpt available for this section contains only the cover page and table of contents, so the segment, customer-concentration, and geographic splits in Item 1 and Item 7 cannot be quoted or verified here. The cover page does establish the basics — a Netherlands-incorporated company listed on Nasdaq under the ticker NXPI, 252,692,845 ordinary shares outstanding as of February 10, 2026, and an aggregate market value of shares held by non-affiliates of $54.7 billion as of June 27, 2025. That is a scale reference, not a current valuation.
Who pays, and what they are buying
The business type is a manufacturer, fab-lite: it produces some wafers internally and buys others, and sells analog, mixed-signal, and embedded processing chips into automotive, industrial and IoT, mobile, and communications infrastructure end markets. The stage is mature and cyclical, not expansion. Automotive is the anchor. A chip is designed into a vehicle platform and normally stays in production for that platform's life, so engineering spending today becomes revenue years later.
The customers are automotive Tier 1 suppliers and vehicle OEMs, industrial equipment builders, and distributors who resell to smaller buyers. Two consequences follow. The paying decision sits with platform owners and procurement organizations, not with end consumers, so demand is set by vehicle builds, factory automation budgets, and inventory policy in the channel. And distributors punch above their revenue share: sell-in to a distributor is not sell-through to an end user, and that gap is where cycle turns hide. The regional revenue mix — China, Europe, the Americas, and the rest of Asia — sits in the disclosures this excerpt does not include, and that split should be pulled before sizing regional exposure.
Content per vehicle grows while price per unit usually falls
The direct operating cause of growth is more semiconductor content in each vehicle: electrification, sensing, and the consolidation of computing into fewer, larger controllers. The deeper structural cause is the design-in model itself. Qualification takes years, changing suppliers mid-program is expensive, and the customer's own engineers build around NXP's part. That is a real switching cost.
It is not pricing power. Automotive buyers are concentrated, run annual cost-down negotiations, and dual-source anything critical. Content per vehicle can rise while the price of any given part falls. The tradeoff is measurable: if per-vehicle content grows faster than contractual price erosion, automotive revenue compounds on flat vehicle production; if it does not, growth is only as good as the global build rate. What to watch is whether automotive revenue outgrows vehicle production over the same stretch, and whether management describes price-downs as routine or worsening.
Industrial and IoT: shorter cycles and thinner visibility
Industrial and IoT demand is more disbursed, more distributor-mediated, and quicker to be cancelled. The operating cause of that volatility is short lead times and small order sizes; the structural cause is that the channel, not NXP, absorbs the first shock. When distributors carry excess inventory, orders fall before end demand does, so reported revenue undershoots consumption and the recovery looks late. The same mechanics flatter the upcycle, which is why one restocking quarter can pass for a demand recovery.
The structural tradeoff: owned fabs versus purchased wafers
Manufacturing the right wafers in-house protects differentiated process technology and lowers unit cost when the lines are full, but the fixed cost stays whether demand does or not. Buying wafers from foundries preserves flexibility and moves that risk to a supplier, in exchange for higher variable cost and less control over allocation during shortages. Neither side is free. Running lean leaves the company short in an upcycle; building ahead ties up capital that has to earn a return across the whole cycle.
What would overturn this view
The judgment here — a mature, cyclical manufacturer whose growth depends on content gains rather than customer count — would weaken if automotive revenue per vehicle stalled for several quarters while design-win counts still rose. It would strengthen if industrial orders broadened beyond the submarkets that led the last recovery. The evidence that matters is not the headline quarter. It is design-win conversion, distributor inventory relative to sell-through, and whether customers keep paying for NXP's differentiation after the annual price conversation.
Questions worth following
- Will automotive content per vehicle grow fast enough to offset weaker vehicle production?
Automotive is NXP's largest end market, so the balance between vehicle volumes and electronics content drives most of the earnings swing. If content stalls while production is soft, revenue and internal fab loading fall together, magnifying the effect on gross margin.
- Can NXP defend its China socket share as local automotive and industrial suppliers scale?
China is NXP's largest country market, so share moving a few points changes revenue and the design pipeline for years. Because automotive sockets qualify over long cycles, a loss is durable rather than cyclical, which affects both the revenue base and the margin structure.
- Is NXP's industrial and IoT recovery broadening across submarkets or still resting on a few of them?
Industrial and IoT is the second earnings leg and the most channel-sensitive. A narrow recovery leaves NXP exposed to another inventory correction, while broad order growth across submarkets would show that end demand, not distributor restocking, is doing the work.
Financial quality
Financial quality: the cash has held up better than the chip cycle
NXP's five-year record is steady cash flow attached to shrinking revenue. Through FY2025 the company produced positive free cash flow every year, $11.7 billion in total against $12.0 billion of cumulative net income, while revenue slipped from $13.28 billion in FY2023 to $12.27 billion in FY2025. Net margin went from 21.1% to 16.5%, and ROE from 39.9% to 20.1%. This is still a highly profitable business. It is simply earning on a smaller base than the 2022 shortage peak.
| Metric (USD bn unless noted) | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | Q2 FY2026¹ | Assessment |
|---|---|---|---|---|---|---|---|
| Revenue | 11.06 | 13.21 | 13.28 | 12.61 | 12.27 | 3.50 | Contracting three years |
| Gross margin | 54.8% | 56.9% | 56.9% | 56.4% | 54.4% | 57.1% | Stable mid-50s |
| Operating margin | 23.3% | 28.8% | 27.6% | 27.1% | 27.0% | 30.6% | Strong |
| Net margin | 16.9% | 21.1% | 21.1% | 19.9% | 16.5% | 22.0% | Strong, cyclical |
| Net income | 1.87 | 2.79 | 2.80 | 2.51 | 2.02 | 0.77 | Down 28% from peak |
| ROE | 24.2% | 39.9% | 34.8% | 28.2% | 20.1% | Not disclosed | Strong, normalizing |
| Operating cash flow | 3.08 | 3.90 | 3.51 | 2.78 | 2.82 | 0.86 | Strong |
| Free cash flow | 2.15 | 2.67 | 2.51 | 2.06 | 2.28 | 0.94 | Exceptional consistency |
| OCF / net income | 1.65 | 1.40 | 1.25 | 1.11 | 1.40 | 1.12 | Exceptional |
| Capex (OCF − FCF) | 0.93 | 1.23 | 1.00 | 0.72 | 0.54 | Not disclosed | Very light (4–9% of revenue) |
| Liabilities / assets | 67.6% | 66.7% | 63.2% | 60.9% | 60.7% | 55.9% | Demanding, improving |
| Current ratio | 2.13 | 2.12 | 1.91 | 2.36 | Not disclosed | Not disclosed | Adequate |
| Goodwill | 9.96 | 9.94 | 9.96 | 9.93 | 10.30 | 10.27 | ~90% of equity: Weak |
| Overall period grade | Adequate | Strong | Strong | Strong | Adequate | Strong |
¹ Three months ended June 30, 2026; not comparable with the annual columns. *Supplied FCF exceeds OCF in this quarter, so I lean on OCF quarterly and the annual FCF series.
Quarter by quarter: a 2025 trough that turned in 2026
| Metric (USD bn) | Q1 FY25 | Q2 FY25 | Q3 FY25 | Q4 FY25 | Q1 FY26 | Q2 FY26 | Assessment | Trend |
|---|---|---|---|---|---|---|---|
| Revenue | 2.83 | 2.93 | 3.17 | 3.33 | 3.18 | 3.50 | Adequate | Trough to recovery |
| Gross margin | 55.0% | 53.4% | 56.5% | 53.8% | 56.3% | 57.1% | Adequate | Rising |
| Net income | 0.49 | 0.45 | 0.63 | 0.45 | 1.12 | 0.77 | Distorted in Q1 FY26 |
| Net margin | 17.3% | 15.2% | 19.9% | 13.5% | 35.2% | 22.0% | Strong, not smooth |
| Total assets | 25.18 | 25.25 | 26.35 | 26.56 | 27.11 | 26.66 | — |
| Total liabilities | 15.50 | 15.31 | 15.93 | 16.11 | 15.84 | 14.90 | Improving |
| Current liabilities | 3.85 | 4.43 | 3.58 | 3.88 | 3.64 | 3.77 | — |
| Fixed assets | Not disclosed | Not disclosed | Not disclosed | Not disclosed | Not disclosed | Not disclosed | — |
| Operating cash flow | 0.57 | 0.78 | 0.58 | 0.89 | 0.79 | 0.86 | Adequate | Steady |
| Free cash flow | 0.40 | 0.94 | 0.51 | 0.79 | 0.71 | 0.94 | Adequate | Volatile quarterly |
One quarter in this table is not what it looks like. Q1 FY2026 net income of $1.12 billion exceeded operating income of $0.88 billion; EBIT of $1.51 billion implies roughly $0.63 billion arrived below the operating line, and investing cash flow was a $0.51 billion inflow. That quarter's 35% net margin is not a run rate. The company's October 27, 2026 release reports Q3 FY2026 revenue of $3.79 billion, up 19% year over year, GAAP gross margin of 57.8%, and $1.21 billion of non-GAAP free cash flow, alongside a new $5 billion repurchase authorization and gross financial leverage of 2.0x trailing EBITDA (1.3x net).
Where the returns actually come from
ROE has fallen for arithmetic reasons more than competitive ones. Net income dropped about 28% from FY2022 while equity grew from $7.45 billion to $11.4 billion by Q2 FY2026, so the same profitability spread over a larger base produces a lower ratio. Leverage props that ratio up: liabilities are still 56–68% of assets, and goodwill of $10.3 billion equals roughly 90% of equity, so ROE is measured on a base with almost no tangible content. Return on tangible capital is the number NXP doesn't disclose.
Cash conversion is genuinely good, but partly mechanical. OCF/net income of 1.11–1.65 across five years is exceptional, and capex at 4–9% of revenue is why free cash flow stays near net income. The catch is that FY2025 investing outflow was $2.36 billion against only $0.54 billion of capex, meaning roughly $1.8 billion went to acquisitions, investments, or joint-venture contributions. The Singapore VSMC fab is capacity that sits outside NXP's consolidated capex line; if those commitments migrate onto the balance sheet, free cash flow and leverage both move.
Working capital has been flattering the cash line too. The cash conversion cycle improved from 134 days to 116 days as DSO fell to 30 and DIO to 148. When that tailwind stops, OCF/net income drifts toward 1.0 without any change in the business.
Red flags worth checking
- The Q1 FY2026 below-the-line gain. Roughly $0.63 billion lifted net income above operating income in a quarter when investing cash flow turned positive. Whether that came from a divestiture, the VSMC joint venture, or asset sales determines whether it is repeatable.
- Goodwill of $10.3 billion against $11.4 billion of equity. Automotive is about half of revenue. A second auto or industrial downturn would test the annual impairment headroom, and a write-down would cut book value and lift debt ratios at once.
- Channel inventory of 11 weeks versus 9 a year earlier in the latest release, against 19% revenue growth. That gap could mean restocking rather than end demand, which tends to reverse into order cuts.
- Capital returns against 2.0x gross leverage. A $5 billion authorization plus $1.3 billion remaining and roughly $256 million of quarterly dividends are affordable while free cash flow runs above $2 billion a year. They stop being self-funding if operating cash flow slips and borrowing takes over.
Questions worth following
- Can NXP keep funding its dividends and buybacks from free cash flow rather than new debt?
Capital returns now run against a softer earnings base and roughly $10 billion of goodwill. If free cash flow stops covering the dividend and repurchases, funding shifts to the balance sheet, where gross leverage was already 2.0x trailing EBITDA in the latest reported quarter.
- Will NXP's reported profit keep coming from chip sales rather than one-off, below-the-line gains?
Q1 FY2026 net income of $1.12 billion exceeded operating income of $0.88 billion, so roughly $0.63 billion arrived below the operating line. If such items recur, headline earnings overstate the earning power of the semiconductor business and mislead anyone extrapolating that quarter.
- Could another automotive or industrial downturn force NXP to impair its $10.3 billion of goodwill?
Goodwill equals about 90% of shareholders' equity and 39% of total assets, so a write-down would cut book value and lift debt ratios at the same time. Automotive and industrial demand has already cycled once, with revenue falling for three straight years through FY2025.
Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.
Continue reading the full report
- Key risksWhat could break the investment case
- Valuation and margin of safetyWhat the current price assumes, and how much room for error it leaves
- SummaryOur overall judgement and what to watch next
Free to sign up · 20 research points · unlocking this report uses 10
20 points cover 2 full reports. No card needed.
Research reflects its generation date. Preview the first two modules.
More on NXP Semiconductors NV
NXP Semiconductors NV
Business, moat and key questions
Open company page