Linde (LIN) · Deep analysis

Report previewLIN.USReport date Last updated
← Back to reports
Report contents
01

Business model

Business model: density inside long-term contracts

Linde sells industrial gases, not the plants that consume them. Large customers in chemicals, refining, metals, electronics, healthcare and food buy oxygen, nitrogen, hydrogen, argon and specialty gases under multiyear on-site contracts with take-or-pay minima (minimum-volume commitments) and energy pass-throughs. Smaller customers buy merchant liquid or cylinders, while engineering clients buy plant design and equipment. Linde operates in the Americas, EMEA and Asia-Pacific. It is a mature industrial-gas producer with infrastructure-like contracts and a project backlog, not a commodity chemical manufacturer. Plants are usually built beside the customer, so local density and distribution economics decide the return.

The plant next door is the customer relationship

In on-site supply, Linde finances a dedicated plant, then sells the gas to the adjacent customer over a long contract. The customer pays for molecules and often a fixed capacity fee; Linde carries the capital and operating risk. That structure explains why the sale-of-gas business matters most. Linde’s investor summary puts the project backlog at $11 billion, split between $8.1 billion of sale-of-gas projects and $3 billion of sale-of-plant projects. The first creates recurring gas revenue; the second is a lumpier engineering and equipment sale.

Backlog componentUSD bnWhat it represents
Sale-of-gas projects8.1Recurring gas sales under long-term supply contracts
Sale-of-plant projects3.0One-time engineering and equipment revenue
Total project backlog11.0Multi-year pipeline, not current revenue

What the backlog does not do is remove execution risk. Linde spends capex years before the plant reaches design volumes. A delayed start-up or a slow customer ramp leaves capital idle and pushes the return out, even if the contract itself is firm. The model is best understood as a series of site-level infrastructure investments, not a single global gas network. Density helps because a new plant can share engineering, logistics and spare capacity with nearby operations, but the customer relationship remains local and often single-site.

Geography reinforces that. Linde’s FY2025 10-K, filed February 25, 2026, reports the business through Americas, EMEA, Asia-Pacific and Engineering. The contract template is global; the moat is regional. A plant in Texas, Bavaria or Singapore serves a local customer base, and its economics depend on regional utilization and the cost of moving gas, not on a single world price.

Merchant gas is where pricing gets tested

On-site contracts are designed to be dull: volume commitments, fixed fees and pass-through of energy costs. Merchant and packaged gas (liquid delivered by truck, or gas in cylinders) are not. They serve smaller manufacturers, laboratories, hospitals and food processors who buy as needed. Those sales carry higher unit margins and more pricing discretion, which also makes them the first place competition shows up. If a rival adds liquid capacity in a region, or industrial output slows, discounting appears in merchant realizations before it appears in group revenue. Fixed fees and cylinder rental add a small annuity layer, but they do not change the basic split: on-site revenue is contracted, merchant revenue is negotiated.

The company’s mix helps: it says about 62% of sales come from healthcare, food and beverage, and electronics, end markets it calls resilient. That label deserves care. Healthcare and food are steady; electronics has its own capital-spending cycle. The remaining sales are more tied to chemicals, refining, metals and general manufacturing. The 62% is a sales classification, not a margin guarantee: a hospital and a chip fab both buy gas, but one has steadier volumes and the other has sharper swings.

Where the model is fragile

The main fragility is not demand for gas. It is the price Linde pays to stay dense. On-site plants are capital-intensive and built for one customer’s site, so a weak contract, a customer shutdown or a delayed project can strand capital with few alternative uses. Bargaining power is strongest before construction, when the customer needs the gas and Linde can choose the site; it is weakest at renewal, when the plant is sunk.

Long-term contracts protect the revenue line but do not make it perpetual. Renewals can reset fixed fees, volume commitments or cost pass-through terms. Linde does not publish a contract-by-contract renewal schedule, so investors cannot see the rollover wall directly. Sale-of-plant projects add unevenness: they can lift reported growth in one period and disappear in the next, while contributing less recurring cash than sale-of-gas. The capital cycle is the deeper issue. Linde can add capacity only where it has a customer or a plausible merchant market. Building ahead of demand would raise fixed costs and pressure utilization, the opposite of the density strategy. Backlog quality therefore matters more than backlog size.

Linde’s published growth model targets 8–12% annual EPS growth excluding macro effects, leaning on pricing, productivity, backlog execution and buybacks. That is a mature compounding model, not a volume-growth story. The judgment is that Linde owns a durable but capital-hungry position: local density, long contracts and pricing discipline make returns predictable, while growth depends on converting signed projects and defending merchant prices. The evidence that would overturn that view is not a single weak quarter. It would be a flat sale-of-gas backlog alongside rising capex and slipping project start-ups, which would show capacity being built ahead of committed demand.

Questions worth following

  • Will Linde's large on-site projects start up on schedule and reach contracted volumes?

    Start-up timing decides when capital already spent starts producing revenue. Delays push depreciation and interest ahead of earnings, and a project that misses its contracted volumes weakens the case that Linde's growth spending earns an acceptable return rather than just enlarging the asset base.

  • Does Linde's resilient end-market mix hold through a downturn?

    About 62% of sales come from healthcare, food and beverage, and electronics, which the company treats as resilient. If that mix keeps volumes steadier than the industrial cycle, revenue and plant utilization should be less cyclical; if electronics behaves like another capital-spending market, the defensive claim is weaker.

02

Financial quality

Financial quality: cash-backed earnings, thinner cover for the payout

Earnings quality: Strong. Linde turns reported profit into cash with unusual consistency. Net income of $6.90 billion in FY2025 arrived alongside $10.35 billion of operating cash flow, and every year since FY2021 has produced operating cash flow at 1.4 to 2.5 times net income. Free cash flow was positive in all five years shown. The strain sits on the uses side: free cash flow is 23% below its FY2021 level because capital spending has climbed, and the balance sheet now leans harder on debt against a smaller equity base.

Margins did the work, and volume has only just returned

Revenue went from $30.79 billion in FY2021 to $33.99 billion in FY2025, about 2.5% a year. Over the same span net margin rose from 12.4% to 20.3% and operating margin from 17.1% to 26.3%, lifting ROE from 8.4% to roughly 18% and ROA from 4.5% to 7.9%. Returns improved because the margin improved: asset turnover was 0.41 in both FY2021 and FY2025, so neither faster asset use nor leverage explains the step-up.

Gross margin needs a caveat. FY2025 comes in at 43.3% against 36.6% in FY2024, and the quarters inside 2025 jump from 37.5% in March to 49.3% in June. A break that sharp inside a single year points to a change in what sits in cost of sales, so the net and operating margin lines are the comparable ones.

Growth has restarted. Revenue rose 9.3% in the June 2026 quarter to $9.29 billion, and Linde reported $9.4 billion for the September quarter, up 9.2%, with volume up 2% and price/mix adding 3% (CNBC reported). Price and cost programs carried 2022 through 2025; holding a 20% net margin from here needs the volume leg.

Capital spending is absorbing the cash growth

Capex, measured as operating cash flow minus free cash flow, rose from $3.08 billion in FY2021 to $5.26 billion in FY2025, from 10.0% to 15.5% of revenue and from 32% to 51% of operating cash flow. Free cash flow fell from $6.64 billion to $5.09 billion even as profit grew. The June 2026 quarter repeats the pattern: operating cash flow of $2.27 billion against $2.21 billion a year earlier, but free cash flow of $0.83 billion against $0.95 billion, because capex rose to $1.44 billion from $1.26 billion.

This is the most consequential test of the next two years. On-site plants built against take-or-pay contracts turn capex into predictable, contracted cash, which is a different proposition from merchant capacity built on hope. But the cash arrives after start-up, and until it does, free cash flow growth has to come from volume and price rather than from capital discipline.

Leverage is modest, but the equity cushion is thin

Liabilities rose from $36.2 billion at the end of FY2021 to $47.7 billion in June 2026 while equity fell from $44.0 billion to $39.1 billion; liabilities are now 54% of assets. Buybacks plus borrowing produced that combination, and a smaller denominator mechanically lifts ROE. Return on assets doubling over the same period is the check that margins, not financial engineering, did most of the work.

Liquidity is adequate rather than comfortable. Cash of $4.90 billion against $7.33 billion of short-term debt is 0.67 times cover, the weakest in the five years shown, and the current ratio is 0.89. Interest expense of $0.21 billion in FY2025 is small against $12.95 billion of EBITDA, but the line looks low relative to roughly $25 billion of gross debt reported in mid-2025, so coverage built on it may overstate the cushion. Distributions have run close to free cash flow: financing outflows of $4.55 billion in FY2025 against $5.09 billion of free cash flow, similar in FY2023 and FY2024, and above free cash flow in FY2021.

What needs checking
  • Goodwill of $27.93 billion equals 71% of equity. With Lincare under strategic review, a $425 million restructuring charge taken in the September 2026 quarter and the business moved into the Other segment, impairment testing is the item to watch. A write-down would not touch cash, but it would cut the book value backing the dividend.
  • Receivables of $5.49 billion in June 2025 were up 8.5% from $5.06 billion at the end of FY2024 on roughly 3% revenue growth, with turnover drifting from 6.8 times in FY2022 to 6.5 times. That is drift, not deterioration, and earnings are still converting to cash; one more year of observation will settle whether it is mix or looser terms.

Note that reported operating income in the March 2026 quarter ($3.26 billion) exceeds that period's EBIT ($2.49 billion), a reminder that the two statements classify items differently. If volume growth holds and capex converts, free cash flow can resume growing from a lower base; if volume stalls while capex keeps climbing, the buyback leans increasingly on the debt market.

USD bn unless notedFY2021FY2022FY2023FY2024FY2025Q2 2026Assessment
Revenue30.7933.3632.8533.0133.999.29Adequate
Operating margin %17.119.424.726.126.327.3Strong
Net margin %12.412.418.919.920.320.8Strong
ROE %8.49.915.516.918.04.6Strong
ROA %4.55.17.78.27.92.1Adequate
Liabilities / assets %44.348.049.150.754.254.0Adequate
Cash / short-term debt (x)0.900.920.750.740.800.67Adequate
Operating cash flow9.728.869.309.4210.352.27Strong
Capex (OCF − FCF)3.083.173.784.495.261.44Adequate
Free cash flow6.645.695.524.935.090.83Strong
OCF / net income (x)2.542.131.501.431.501.18Exceptional
Capex / revenue %10.09.511.513.615.515.5Adequate
FCF / financing outflow (x)0.871.841.021.131.12n.m.Adequate

*Single quarter; ROE and ROA not annualized. Capex is derived as operating cash flow less free cash flow. FY2026 Q2 lacked current assets in the reporting period, so ratios using them are omitted.

USD bnQ1 2025Q2 2025Q3 2025Q4 2025Q1 2026Q2 2026AssessmentTrend
Revenue8.118.508.628.768.789.29AdequateRising
Net income1.671.771.931.571.861.93StrongSteady
Operating income2.222.352.342.023.262.54StrongSteady
Total assets82.7086.0885.9986.8286.3188.35AdequateRising
Total liabilities43.2446.0945.9147.0846.2247.73WeakRising
Operating cash flow2.162.212.953.032.242.27StrongSteady
Free cash flow0.890.951.671.570.900.83AdequateSoftening

Net PP&E is not disclosed in these quarterly periods; the last reported figure is $23.55 billion at the end of FY2022. Q3 and Q4 cash flow is seasonally stronger than the first half.

Questions worth following

  • Will Linde write down goodwill as it restructures Lincare?

    Goodwill of $27.93 billion is 71% of a shrinking equity base, and the home-health business that sits inside it is being restructured and reviewed for strategic alternatives. An impairment would not affect cash but would cut book value and the equity supporting the dividend.

  • Can Linde lift free cash flow sharply from FY2025's 5.09 billion base without a step-up in capital spending?

    The model needs 13% growth in year one and mid-single-digit growth afterward, but free cash flow has been flat to lower since FY2021. If capex keeps absorbing operating cash, the implied growth path and the premium to modeled value both weaken.

  • Can Linde sustain an 18% ROE as its equity base shrinks?

    ROE nearly doubled from 8.4% in FY2021 to about 18% in FY2025 while equity fell $5 billion through buybacks and liabilities rose $11.6 billion. If margins drive the return it is durable; if a thinner equity base does, the same return carries more refinancing risk.

Generated by DeepVal AI. Data and models may contain errors or be outdated. For research only; not personalized investment advice.

Go beyond the preview

Continue reading the full report

  1. Key risksWhat could break the investment case
  2. Valuation and margin of safetyWhat the current price assumes, and how much room for error it leaves
  3. 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 Linde plc Ordinary Shares