The Fab Win Lands, the Margin Story Cracks: A Double Beat the Market Sold Down 5.9%
Key Takeaways
- A double beat the market refused to buy. Adjusted EPS of $4.50 (+10%) edged the ~$4.49 Street and sales of $9,289M (+9%) cleared a ~$9.0B bar, yet the stock fell 5.9% to $478.38 on 2.4x normal volume. The reason was one line: adjusted operating margin of 29.5%, 60 bps below prior year, on 9% revenue growth. Management led the call by naming it: "we are not satisfied with our margin performance for this quarter."
- US homecare is now a sized problem and a live divestiture candidate. Lincare drove the majority of the Americas margin decline; excluding it, management said Americas margin would have risen 20 bps ex-pass-through. Management put the drag roughly 30% above the ~$30M-a-quarter figure an analyst proposed, and disclosed it is evaluating "the strategic fit of this U.S. homecare business within Linde, both in part and as a whole."
- The backlog catalyst we were waiting for arrived two quarters early. A new long-term US electronics supply contract added $1B to lift the sale-of-gas backlog to a record $8.1 billion, hitting the "8 handle" management targeted for year-end. Total project backlog reached $11 billion, electronics grew 18%, and a Taiwan JV is investing a further ~$800M outside the reported backlog.
- The raise was exactly the size of the beat, and the back half was left untouched. The FY26 floor rose $0.10 to $17.70–$17.90, moving the midpoint $0.05 to $17.80, precisely the $0.05 by which Q2 beat its own guide midpoint. Q3 was guided to a $4.50 midpoint against a ~$4.53 Street. Meanwhile capex guidance jumped $500M to $5.5–$6.0B and Q2 free cash flow fell 13% to $833M.
- Rating: Maintaining Hold. Two of the three upgrade triggers we named in May have now fired (signed fab wins, and a multiple pullback to ~26.9x from ~28.6x), but the margin self-help pillar cracked in the same quarter. We cut fair value to roughly $500 and want one clean quarter of margin recovery, plus a decision on Lincare, before we pay up.
Results vs. Consensus
| Metric (Q2 2026) | Actual | Consensus / Guide | Beat/Miss | Magnitude |
|---|---|---|---|---|
| Sales | $9,289M | ~$9,000M | Beat | +3% |
| Adjusted diluted EPS | $4.50 | $4.49 | Beat | +$0.01 |
| Adjusted diluted EPS vs. company guide | $4.50 | $4.40 – $4.50 | Top of range | +$0.05 vs. midpoint |
| Adjusted operating profit | $2,744M | n/a | Up | +7% YoY |
| Adjusted operating margin | 29.5% | 30.1% prior year | Miss | -60 bps YoY |
| GAAP diluted EPS | $4.15 | n/a | Up | +11% YoY |
| Free cash flow | $833M | $954M prior year | Down | -13% YoY |
| FY2026 adjusted EPS guide (midpoint) | $17.80 | ~$17.89 Street | Below | -$0.09 |
| Q3 2026 adjusted EPS guide (midpoint) | $4.50 | ~$4.53 Street | Below | -$0.03 |
Year-over-Year (Q2 2026 vs. Q2 2025)
| Metric | Q2 2026 | Q2 2025 | Change |
|---|---|---|---|
| Sales | $9,289M | $8,495M | +9% |
| Cost of sales, excl. D&A | $4,861M | $4,306M | +13% |
| Cost of sales as % of sales | 52.3% | 50.7% | +160 bps |
| Selling, general and administrative | $891M | $870M | +2% |
| SG&A as % of sales | 9.6% | 10.2% | -60 bps |
| Operating profit (GAAP) | $2,554M | $2,354M | +8% |
| Adjusted operating profit | $2,744M | $2,556M | +7% |
| Adjusted operating margin | 29.5% | 30.1% | -60 bps |
| Adjusted EBITDA | $3,572M | $3,351M | +7% |
| Adjusted EBITDA margin | 38.5% | 39.4% | -90 bps |
| Net income (GAAP) | $1,928M | $1,766M | +9% |
| Adjusted net income | $2,089M | $1,937M | +8% |
| GAAP diluted EPS | $4.15 | $3.73 | +11% |
| Adjusted diluted EPS | $4.50 | $4.09 | +10% |
| Diluted shares outstanding (000s) | 464,523 | 473,573 | -2% |
| Adjusted effective tax rate | 23.9% | 24.3% | -40 bps |
| Operating cash flow | $2,271M | $2,211M | +3% |
| Capital expenditures | $1,438M | $1,257M | +14% |
| Free cash flow | $833M | $954M | -13% |
Sequential (Q2 2026 vs. Q1 2026)
| Metric | Q2 2026 | Q1 2026 | Change |
|---|---|---|---|
| Sales | $9,289M | $8,781M | +5.8% |
| Adjusted operating profit | $2,744M | $2,630M | +4.3% |
| Adjusted operating margin | 29.5% | 30.0% | -50 bps |
| Adjusted diluted EPS | $4.50 | $4.33 | +3.9% |
| Free cash flow | $833M | $898M | -7.2% |
| Sale-of-gas backlog | $8.1B | $7.1B | +$1.0B |
The sequential margin line is the one to sit with. Last quarter we wrote that the 50 bps sequential margin gain was the number that mattered, because it demonstrated the self-help engine the thesis depends on. This quarter that gain was handed straight back, and the year-over-year comparison went from a 10 bps decline to a 60 bps decline. On a 5.8% sequential revenue increase, adjusted operating profit rose only 4.3%. Linde's model is supposed to convert incremental volume at above-corporate margin. In Q2 it converted at below-corporate margin.
Assessment — Revenue
The top line is genuinely healthier than it has been at any point in our coverage, and the mix inside it is the best news in the quarter. Volume contributed 2 points, matching price for the first time in the cycle, and the CFO said "Almost half of the volume increase relates to project start ups in APAC, and Americas." with the remainder organic across the US, China, Korea, India and advanced materials. Electronics grew 18% year over year and manufacturing was the fastest-growing industrial market, with aerospace alone accounting for more than a third of manufacturing growth. That is a considerably broader base than the single-cylinder Americas story of Q1.
The caution is that the reported 9% flatters it. Currency added 2 points and will fade: the company's own Q3 guide assumes no year-over-year FX benefit at all, and the full year assumes only 1%. Cost pass-through added another point with, in the company's words, "minimal impact on operating profit," and the same pass-through is what dragged reported APAC and EMEA margins below their underlying trend. Underlying 4% growth on a 0%-base-volume guide is a beat against the plan. It is not the acceleration the reported number implies.
Assessment — Margins
This is where the quarter was lost. Adjusted operating margin of 29.5% sits 60 bps below prior year, or 30 bps excluding cost pass-through. Cost of sales rose 13% against 9% revenue growth and climbed 160 bps as a share of sales, to 52.3%. Management attributed the shortfall to three things in descending order: the US homecare business, a dilutive mix of higher hard-goods sales in the US package business, and lower-margin electronics equipment sales in APAC. It characterised the second and third as good problems, and we broadly agree, since both are demand signals that seed future gas volume. The first is not a good problem.
The more serious issue is what this does to the full-year commitment. In May, management told the Street it expected full-year margin expansion "at the upper end or even above" its traditional 40–60 bps range, and that commitment was the mechanism behind the raised EPS floor. Half the year is now done. H1 adjusted operating margin of 29.7% is below the 30.1% Linde posted in H1 2025 and below the 29.8% it delivered for full-year 2025. Reaching even the low end of that expansion ambition would now require a second half that expands margin by well over 100 bps year over year. Management did not repeat the 40–60 bps commitment on this call. It said instead that it expects "many of these margin headwinds to be temporary," that "Actions are aggressively underway", and that it would "likely look to take some cost actions this quarter". That is a materially weaker forward statement than the one made in May, and the Street priced the difference.
Assessment — EPS
Adjusted EPS of $4.50 grew 10%, the same rate as Q1, and H1 adjusted EPS of $8.82 is up 10% against $8.04. On its face the earnings algorithm is working. The problem is composition and trajectory. Composition: 7 points of the 10 came from operating profit and the balance from tax, equity income and share count, at a moment when the buyback pace is slowing (H1 net share purchases of $1,656M against $2,207M a year ago, down 25%) to fund a rising capital budget. Trajectory: the full-year midpoint of $17.80 less the $8.82 already banked leaves roughly $8.98 for the second half against $8.42 in H2 2025, implying about 7% growth versus the 10% delivered in H1. With Q3 guided to a $4.50 midpoint, the implied Q4 sits near $4.48. The company is guiding to deceleration in the half where the margin recovery is supposed to arrive.
Segment Performance
| Segment | Sales | YoY | Underlying | Op. profit | Margin | Margin Δ YoY |
|---|---|---|---|---|---|---|
| Americas | $4,083M | +7% | +4% (2% price, 2% volume) | $1,272M | 31.2% | -50 bps |
| EMEA | $2,303M | +7% | +1% (2% price, -1% volume) | $823M | 35.7% | -40 bps |
| APAC | $1,870M | +13% | +8% (6% volume, 2% price) | $531M | 28.4% | -120 bps |
| Engineering | $625M | +13% | n/a | $100M | 16.0% | -30 bps |
| Other | $408M | +30% | +26% (volume/price) | $18M | 4.4% | from (4.1)% |
| Total | $9,289M | +9% | +4% | $2,744M | 29.5% | -60 bps |
Segment operating profit of $2,744M is the adjusted figure. Reported operating profit of $2,554M is arrived at after $190M of Linde AG merger purchase-accounting impacts (prior year: $202M).
Sales bridge by segment (Q2 2026 vs. Q2 2025)
| Factor | Group | Americas | EMEA | APAC |
|---|---|---|---|---|
| Volume (APAC: volume/equipment) | +2% | +2% | -1% | +6% |
| Price / Mix | +2% | +2% | +2% | +2% |
| Cost pass-through | +1% | 0% | +2% | +2% |
| Currency | +2% | +2% | +3% | +3% |
| Acquisitions / divestitures | +1% | +1% | +1% | 0% |
| Engineering | +1% | n/a | n/a | n/a |
| Total | +9% | +7% | +7% | +13% |
Americas
Sales of $4,083M grew 7% with 4% underlying, split evenly between price and volume, led by electronics and manufacturing. That is a solid quarter on the top line. The problem is beneath it: operating profit grew only 5% and margin fell 50 bps to 31.2%, breaking a run in which the Americas had been the margin bright spot (31.6% and up 60 bps in Q1). Management was explicit that the decline is not a gases problem. Excluding the US homecare business, it says Americas margin would have expanded on a pass-through-adjusted basis.
"So I think we--the slides itself, we have laid out the fact that The Americas business ex the US home care or Lincare business would be up 20-basis-points on margin ex pass through as we normally do. So that is a reflection of the gases business doing well." — Sanjiv Lamba, CEO
The second, smaller drag is mix. US hard-goods sales rose a double-digit percentage, which dilutes segment margin but is the leading indicator management watches most closely for a US manufacturing recovery. We treat that trade as clearly favourable: hard goods pull consumables and gas volume behind them, and the same dilution showed up in Q1 without derailing the margin.
Assessment: The Americas gases franchise is performing. A structurally challenged homecare distribution business bolted onto it is now large enough to swing the segment's reported margin by more than half a point, which is a portfolio problem rather than an operating one. That distinction is why the strategic review matters more than the size of the drag.
APAC
APAC was the volume star and the margin casualty. Sales rose 13% with 8% underlying, of which 6 points were volume and equipment, the second consecutive quarter at that rate after a 2025 that hovered around flat. Operating profit rose 8% but margin fell 120 bps to 28.4%, or 70 bps excluding cost pass-through. The composition is the same one flagged in Q1 and now confirmed as persistent rather than seasonal: a heavy slug of low-margin sale-of-equipment to electronics customers, which carries contracted merchant gas revenue behind it.
"There are 3 components to what is happening in the Asia volumes. Right? There is obviously base volume, which is positive. There are sale-of-equipment, significant sale of elements sitting within there for the electronics customers. That has had a somewhat disproportionate impact in this last quarter that we that we are talking about. And last but not least, there are some, you know, ramp ups." — Sanjiv Lamba, CEO
In Q1, management said it expected APAC margin to recover toward the roughly 29% level of the prior year. It did not. APAC margin improved 40 bps sequentially from 28.0% to 28.4% but sits 120 bps below the prior-year quarter's 29.6%. This is a missed forward commitment, even if the underlying cause is a trade we would take.
Assessment: Positive base volume in APAC for a second straight quarter is a real thesis input, and it is the clearest evidence yet that the volume recovery is broadening past one region. But the equipment mix that comes with it is proving durable, not transitory, and the promised return to 29% margin did not happen. Treat APAC as a volume win and a margin drag simultaneously, and stop modelling the equipment dilution as temporary.
EMEA
EMEA sales rose 7% to $2,303M on 3% currency, 2% price, 2% cost pass-through and 1% acquisitions, with volume down 1%, primarily manufacturing. Underlying growth of 1% is a genuine improvement over Q1's underlying decline of 2% and a 3% volume drop. Operating margin of 35.7% fell 40 bps on a reported basis, but rose 10 bps excluding cost pass-through, so the underlying margin actually expanded. EMEA remains, as it has all year, the company's highest-margin region on the weakest volumes.
The striking thing about EMEA this quarter is not the numbers. It is that the region was not mentioned once in prepared remarks and was not raised in a single analyst question. A quarter ago it was the subject of an extended exchange on structural decline, production relocation and management's open dissatisfaction. This quarter the entire call moved on.
Assessment: The volume trend is less bad and the underlying margin is up, both modest positives. But the total absence of EMEA from the discussion tells you the market has stopped underwriting a European recovery and started treating the region as a stable, high-margin annuity in slow decline. That is probably the right frame. It also means EMEA has stopped being a source of upside surprise.
Engineering
Engineering sales rose 13% to $625M on project timing plus 2% currency, reversing Q1's 8% decline, with operating profit up 11% to $100M at a 16.0% margin. Order intake of $871M was well ahead of Q1's $640M, and the third-party sale-of-equipment backlog stands at $3.0 billion. For the half, however, sales are up only 2% and operating profit is down 1%, which is the honest picture of a business whose quarterly revenue is a timing artefact.
Assessment: The order intake is the signal, and it is improving for a second consecutive quarter. The margin at 16.0% is below the 19.5% posted in Q1 and below H1's 17.6%, so the revenue recovery came at mix cost. Engineering is doing what it is supposed to do, which is convert third-party work into gas-network optionality; it is not a swing factor for the thesis.
Other
Other, which houses corporate costs, Linde Advanced Material Technologies and the global helium wholesale business, grew sales 30% to $408M and swung to an $18M operating profit from a $13M loss. Management attributed the improvement to higher volume and price in the materials business, helped by commercial-space build-out, plus lower corporate costs. This is also where the helium wholesale economics land.
Assessment: A $31M year-over-year operating-profit swing in a segment that is usually noise is worth flagging, because it is the one place the helium pricing cycle and the space ramp show up in the P&L today. It partially offsets the Lincare drag at the group level, and it is the quietest good news in the quarter.
Key Topics & Management Commentary
Overall Management Tone: Openly self-critical, and deliberately so. Management put the margin shortfall in the opening minute of prepared remarks rather than letting Q&A extract it, named the responsible business, and closed the financial review by conceding that "We know our owners expect more". Against the measured-but-warming posture of the prior quarter, this was a management team choosing to own a problem before the Street defined it, while holding the growth narrative intact through a record backlog. The confidence gap is specific rather than general: forward statements on demand, backlog and the earnings algorithm were as firm as ever, but every forward statement on margin was framed as an action underway rather than a commitment quantified.
1. The Margin Miss, Named in the Opening Minute
Linde has spent two years selling a self-help story in which price and productivity compound EPS through a flat-volume cycle. This quarter volume finally arrived and margin went the other way. Adjusted operating margin fell 60 bps year over year, 30 bps excluding cost pass-through, and the CEO chose to lead with it rather than bury it in the segment discussion.
"While these results demonstrate the strength of our core business and the future growth prospects, we are not satisfied with our margin performance for this quarter. Operating margins, excluding cost pass through, declined approximately 30-basis-point year over year. Primarily driven by the Americas segment." — Sanjiv Lamba, CEO
The CFO's decomposition ranked the causes and pre-emptively defended two of the three: US homecare first, then dilutive US hard-goods mix, then low-margin APAC electronics equipment.
"Operating margins of 29.5% decreased 60-basis-point from prior year, or 30-basis-point when excluding the impact of cost pass through. As Sanjiv mentioned, the U.S. homecare business negatively impacted the Americas. Excluding this, margins would have increased. But regardless, actions are underway to improve." — Matthew J. White, CFO
Assessment: Leading with the bad news is the correct disclosure choice and it does not change the number. The thesis pillar at risk is Bull 3, margin self-help, which was the mechanism behind the raised guidance floor in May. That pillar moves from confirmed to challenged today, and it will stay challenged until we see a quarter of year-over-year margin expansion.
2. Lincare Moves From Footnote to Strategic Review
The US homecare business, Lincare, has been a recurring aside on Linde calls for two years. This quarter it became the explanation for a 5.9% down day. Management framed the deterioration as structural rather than cyclical, driven by labour cost inflation and reimbursement policy, and confirmed that portfolio pruning has not been sufficient.
"The majority is driven by the U.S. homecare business. Even though we have been actively pruning this portfolio, it simply has not been enough to overcome the continued headwinds led by higher cost inflation, and policy changes. We have a series of actions underway. And I fully expect sequential improvement into the third quarter. At the same time, we continue to evaluate the strategic fit of this U.S. homecare business within Linde, both in part and as a whole." — Sanjiv Lamba, CEO
The phrase that matters is "both in part and as a whole." That is divestiture language, and it is the first time in our coverage that Linde has put a business of this scale explicitly on the table. Pressed on the history, the CEO was candid that this is a post-COVID normalisation the company has been slow to address.
"Look. The challenges at Lincare are not new. Right? The business has served us well through the COVID period and the immediate kind of couple of years after that. But over the last couple of years and in particular, you heard us reference it as well. It has faced persistent headwinds, right, from labor cost inflation and changes in reimbursement environment." — Sanjiv Lamba, CEO
Assessment: This is the most consequential disclosure of the quarter and it cuts both ways. Near term it is a confirmed, sized earnings drag with no completion date. Medium term, separating a low-multiple, policy-exposed distribution business from a 29%-margin industrial gases franchise would be accretive to both the margin profile and the multiple. We would rather own the outcome than the process, and the process has no clock on it.
3. A Record $8.1 Billion Sale-of-Gas Backlog, Two Quarters Early
In May management said it expected the sale-of-gas backlog to carry an "8 handle" by year-end on electronics wins it had not yet signed. It signed one, and the backlog got there in a single quarter.
"During the second quarter, we achieved record sales and EPS levels. with both growing at near double-digit percent. While increasing the backlog by $1 billion to a record $8.1 billion. After securing a new electronics win in the US." — Sanjiv Lamba, CEO
Total project backlog reached $11 billion. Management further committed that the backlog will still end the year with an 8 handle despite a heavy start-up schedule that consumes roughly $1.3 billion of it.
"For the remainder of the year, we are expecting to start up more than 20 projects that add up to approximately $1.3 billion in investments. Even after accounting for these startups, and based on the opportunities I see today, I expect our sale-of-gas backlog to finish the year with an 8 handle." — Sanjiv Lamba, CEO
Assessment: This is the single upgrade trigger we named in May that has now unambiguously fired, and it fired ahead of schedule with a harder commitment attached. Backlog replenishment was the concern we flagged at initiation, when the sale-of-gas figure was closer to $5.5 billion. It is resolved. Bull 2 strengthens.
4. Electronics Is Now the Engine, and Not All of It Is in the Backlog
Electronics grew 18% year over year and is the fastest-growing end market, driven by project start-ups plus AI-hardware-linked demand. The new US award supports advanced-node fab expansion in the Western US and adds to Linde's existing Arizona cluster, with construction already underway on reimbursable letters of intent while the supply contracts are papered.
"As I mentioned earlier, we added $1 billion of new electronic wins to the backlog to support the expansion of advanced node fabs in the Western US. Consistent with other backlog projects, we have already begun constructing the plants under reimbursable LOIs while the supply contracts are finalized." — Sanjiv Lamba, CEO
A second, separate disclosure is easy to miss and materially expands the growth picture, because it sits outside the reported backlog entirely.
"Not included in the backlog, are a couple of electronics wins by our Taiwan JV, which will invest approximately $800 million to build, own, and operate ASUs and hydrogen production units to supply to new semiconductor fab and advanced packaging facilities there." — Sanjiv Lamba, CEO
Assessment: The headline $8.1 billion understates committed growth by roughly $800 million of JV capital, and the pipeline behind it spans the US, Taiwan, Korea and China. Electronics has replaced clean energy as the reliable engine of the backlog, with better contract structure and faster construction cycles. This is the strongest part of the long-term case and it improved this quarter.
5. Capital Intensity Steps Up and Free Cash Flow Steps Down
Full-year capital expenditure guidance rose $500 million to $5.5–$6.0 billion, driven by the new backlog award plus rising base capex for commercial-space customers. In the quarter, capex of $1,438M was up 14% against operating cash flow up 3%, so free cash flow fell 13% to $833M. Year to date, net share purchases of $1,656M are 25% below the $2,207M spent in H1 2025.
"Year to date, we have deployed $6 billion of capital. Split evenly between business investments, and shareholder returns. $1.9 billion of secured growth represents capital deployed for acquisitions, and the project backlog. When considering the record $8.1 billion sale-of-gas backlog, continued roll up acquisition targets and project pipeline opportunities we expect this number to remain a significant use of capital for the foreseeable future." — Matthew J. White, CFO
Return on capital of 23.5% compares with 24.2% for full-year 2025, extending the decline management flagged in February when it said the metric would sit in the low-to-mid 20s for several years as growth turns more capital-intensive.
Assessment: This is the honest cost of the backlog win, and it is the right trade at a post-tax double-digit unlevered IRR. But it changes the shareholder-return arithmetic for the next two years: capex up, buyback down, ROC drifting lower, free cash flow shrinking in absolute terms. Investors who own Linde for the compounding buyback rather than the growth pipeline are being asked to accept a different mix than they had in 2024 and 2025.
6. Helium: The Free Option Gets Repriced
Last quarter's most important disclosure was that helium had flipped from a multi-year headwind to an acute-shortage tailwind, entirely excluded from guidance and therefore pure upside. Three months later, the mechanism is working but the payoff is smaller and slower than the framing implied.
"So I think first on helium, Yeah. So we left the guidance intact. So by default, that kind of means no material change and helium would also be part of that... You know, when you think about the helium business right now, what we are seeing, we are seeing strong price improvement. But we are also seeing higher costs for dislocation, as Sanjiv mentioned. So the contribution on dollar basis, it is positive, but it is not as large as we would like it, but it is positive. But on a margin basis, that grossing up effect right now is a little bit dilutive." — Matthew J. White, CFO
Management also pushed normalisation out. The CEO said the Strait of Hormuz disruption will "have a lasting impact for the rest of the year," that restarting Qatari production and realigning tanks and shipping will proceed "at a slower pace than most of us would like," and that a normalised market is an early-2027 event. Asked directly whether helium is a 2027 tailwind, the answer was hedged.
"Helium will be normalized next year. I think we will have to wait and see what that means. The complexity of volume and price mix, I think, will play a role what helium does next year." — Sanjiv Lamba, CEO
Assessment: Downgrade the helium option. It is contributing positive dollars, which is better than the drag of a year ago, but it is dilutive to reported margin because the cost of dislocated supply grosses up alongside the price, and management explicitly declined to put any of it into guidance. The Q1 read that a mid-year helium-driven guide bump was a live upgrade trigger did not survive contact with this quarter. Helium is now a modest positive, not an asymmetric one.
7. The Raise That Was Exactly the Size of the Beat
Management lifted the full-year floor $0.10 to $17.70–$17.90 and left the top untouched, moving the midpoint from $17.75 to $17.80. Q2 beat the midpoint of its own guide by $0.05. The full-year midpoint moved $0.05. The second half was explicitly left alone.
"The updated full year range is $17.70 to $17.90. Or 8% to 9 percent growth excluding a 1 percent FX tailwind assumption. This range raises the prior bottom end by $0.10 but leaves the top unchanged. While base volumes showed some recovery in the second quarter, we would like a few more quarters under our belt before incorporating this trend. Into future guides. Therefore, we are leaving the back half guidance assumption the same as before." — Matthew J. White, CFO
The Q3 range of $4.45–$4.55 assumes no year-over-year currency benefit and a 1% sequential FX headwind, which is why the midpoint sits flat against the $4.50 just delivered even though management describes a $0.05 sequential improvement excluding currency. It closed the guidance discussion with an unusually direct acknowledgement of investor frustration.
"Of course, this is merely a guide. How we perform is what matters most. We know our owners expect more, and the organization is committed to delivering on those expectations." — Matthew J. White, CFO
Assessment: A raise that exactly offsets the quarter's beat is arithmetically a maintained guide. Management's stated reason, wanting more quarters of confirmed base volume before extrapolating, is the same conservatism that has produced four consecutive beats, so we would normally read it as sandbagging. The difference this time is that the margin shortfall gives the unchanged back half a second and less benign explanation, and the market chose that one.
8. Cost Actions Are Coming, Unsized, in Q3
Asked what specifically drives the margin recovery, the CFO pointed first to easier second-half comparisons and then previewed a discrete cost programme, without a number attached.
"But we will likely look to take some cost actions this quarter, depending on the size. that is something we wanna get ahead of. I mean, it is clear you are seeing more inflation around the world, and that is something that we have to manage through our productivity and our actions. And in some regions, you are seeing growth, which supports it. In other regions, you are seeing inflation without the growth." — Matthew J. White, CFO
He also leaned on the shape of the prior year, noting Linde ran strong front-half and weaker back-half margins in 2025, which mechanically makes the H2 2026 comparison easier.
"So if you may recall, 2025 we had strong front half margins, weaker back half margins. So when you think about the whole year in the context, I am fully expecting us to see better year over year just given how last year played out." — Matthew J. White, CFO
Assessment: Linde has done this before and done it well; the 2025 restructuring delivered. But an unquantified charge landing in Q3 with details deferred to the October call means the margin question cannot be settled for another three months, and it introduces a GAAP-versus-adjusted wedge in a quarter where the Street will be scrutinising exactly that. Note also that the easier-comparison argument is a statement about the base, not about operating improvement.
9. The US Manufacturing Recovery Broadens
The demand commentary was the most constructive of the past four calls. Manufacturing was the fastest-growing industrial end market with volume growth in both the Americas and APAC, and the composition matters: aerospace, not general industrial, is doing the heavy lifting, alongside construction tied to data centres.
"Manufacturing remains the fastest growing market. We experienced volume growth across APAC and the Americas, although the US is still the primary driver with both aerospace and construction activity, related to data centers. In fact, aerospace accounted for more than a third of the manufacturing growth during the quarter." — Sanjiv Lamba, CEO
Metals and mining and chemicals and energy each grew low single digits, solid in the US and Brazil and flat elsewhere. Management summarised the picture in terms it has avoided for two years.
"In summary, we have lapped the more difficult comps, and are starting to see green shoots of growth across certain geographies and end markets." — Sanjiv Lamba, CEO
Assessment: The volume pillar keeps improving quietly while the margin pillar deteriorates loudly. Aerospace and data-centre-linked construction are better demand drivers than generic PMI recovery because they carry multi-year visibility. The bear point on absent base volume, which we downgraded from absent to emerging in May, is now closer to resolved than open.
10. Commercial Space, and the Make-Versus-Buy Question
Space is scaling and is now consuming base capex, but the quarter introduced a wrinkle. Some launch customers, with capital access and large propellant requirements, would prefer to own their air-separation units rather than buy the molecules.
"I would say with commercial space, that given the quantities of propellant they require, you are seeing a similar dynamic at least with certain players that have comfort and the access of capital to have a desire to vertically integrate. Now this right now is primarily only with certain players for atmospherics. We are not seeing it on the hydrogen side, which is a very, very different dynamic." — Matthew J. White, CFO
Management's answer is that it will sell either way, and that sale-of-plant deals typically come with operate-and-maintain contracts. On sizing, the CEO held the prior framing rather than raising it.
"we are on track for that billion-dollar opportunity that we laid out over the next few years. I think 2030 was the timeline billion plus is what our expectation around the space markets was. Once it reaches a certain size, you will see us split that out in our end markets to have more visibility around it." — Sanjiv Lamba, CEO
Assessment: A quiet negative revision. Last quarter the space opportunity was described as scaling faster than the $1 billion-by-2030 frame; this quarter it is back to on track for that frame, with a new structural caveat that the best customers may self-supply atmospherics. Sale-of-plant economics are lower-return and non-recurring compared with sale-of-gas. Space remains real optionality, but it is a smaller and lower-quality option than it looked three months ago.
11. The 2027 Algorithm, Restated Without Macro Help
Asked whether next year's double-digit EPS growth requires macro cooperation, the CEO restated the standing algorithm and declined to lean on the cycle.
"As you know, our EPS algorithm lays out the fact that between management actions and capital allocation combined, we should be delivering 8% to 12%. We are not looking for macro as long as macro is not taking away from that. You should expect us to look at that 8% to 12% range And I think we will be, you know, consistent on that as we as we look at next year as well." — Sanjiv Lamba, CEO
Assessment: The algorithm is intact and the guide is at the bottom of it. FY2026 at the $17.80 midpoint is roughly 8% growth, the low end of an 8–12% range, in a year with a record backlog, positive volume in two of three regions and a rising helium price. That gap is the whole investment question. It is not a demand problem; it is a margin and capital-intensity problem, and both are inside management's control.
Guidance & Outlook
| Metric | Period | Prior guide | New guide | Change |
|---|---|---|---|---|
| Adjusted EPS | Q3 2026 | n/a | $4.45 – $4.55 (+6% to +8%) | New |
| Adjusted EPS | FY 2026 | $17.60 – $17.90 | $17.70 – $17.90 (+8% to +9%) | Floor raised $0.10; top unchanged |
| Capital expenditures | FY 2026 | $5.0B – $5.5B | $5.5B – $6.0B | Raised $0.5B |
| FX assumption | Q3 2026 | n/a | No YoY impact; 1% sequential headwind | New |
| FX assumption | FY 2026 | 1% tailwind | 1% tailwind | Maintained |
| Base volume assumption | H2 2026 | 0% at midpoint | 0% at midpoint | Maintained |
The full-year raise is arithmetically neutral. Q2 adjusted EPS of $4.50 beat the midpoint of the company's own $4.40–$4.50 guide by $0.05, and the full-year midpoint moved from $17.75 to $17.80, also $0.05. Management said so directly: the back-half assumption is unchanged, because it wants "a few more quarters under our belt" of confirmed base-volume recovery before it will put that recovery into a guide. The capex guide, by contrast, moved by half a billion dollars.
Implied second-half ramp: H1 adjusted EPS of $8.82 against the $17.80 full-year midpoint leaves roughly $8.98 for the second half, versus $8.42 in H2 2025. That is about 7% growth, against 10% delivered in H1. With Q3 guided to a $4.50 midpoint, the implied Q4 sits near $4.48 on the same midpoint arithmetic, or a range of roughly $4.43 to $4.53 across the guide. Linde is guiding to slower earnings growth in the half where the margin recovery, the cost actions and the easier prior-year comparisons are all supposed to land.
Street at: Consensus for the full year sat near $17.89 and for Q3 near $4.53. Both sit above the respective guide midpoints of $17.80 and $4.50, which means the Street had effectively parked itself at the top of Linde's range. A guide that raises the floor without moving the ceiling does nothing for a model anchored to the ceiling. That, more than the margin line in isolation, explains why a beat produced a 5.9% decline.
Guidance style: Conservative, and this quarter conservatism was read as caution rather than sandbagging. The pattern is consistent with Linde's history of guiding to the achievable and beating modestly, four quarters running. The tell that something changed is not in the range itself but in the language around it. In May the company committed to full-year margin expansion at the upper end of or above 40–60 bps. In July it committed to actions, sequential improvement and an easier comparison, without restating the number.
Analyst Q&A Highlights
Sizing the homecare drag, and whether the business is even profitable
The dominant line of questioning on the call was the US homecare business, raised in the first question and returned to repeatedly. One exchange forced management to correct an analyst's arithmetic upward, which is the only quantification of the drag anywhere in the disclosure.
Q: "If I did the math correctly, the home care penalty was 30 million in the second quarter. So order of magnitude, is it a $100 million penalty for this year? And is Lincare all of your 23% of healthcare revenues for The Americas?"
— Jeffrey Zekauskas, JPMorgan
A: "So I think the numbers a little higher than what you have. So you are close, but I would say it is probably, you know, higher, though. But you could probably say 30% higher than that number, give or take. Okay. So that is the headwind we have. that is what we are facing."
— Matthew J. White, CFO
Assessment: Management volunteered that the drag is roughly 30% larger than the ~$30 million-per-quarter, ~$100 million-per-year framing put to it, which implies something closer to $130 million annualised. That figure appears nowhere in the release. Volunteering an upward correction is a mark of candour and a reminder that the disclosure around this business has been thin for two years.
Whether the helium assumption moved, and why Americas price stopped accelerating
A two-part question sought to establish whether the helium shortage that dominated the prior quarter's narrative had been formalised into numbers, and separately why Americas pricing stepped down from the prior quarter's 4%.
Q: "First on helium, just to clarify. Did you all change anything in your guidance assumptions relative to what you had assumed back at the start of the year? And then secondly, in Americas, the kind of year-on-year price step down, I think it was flat sequentially. Was that the hard goods mix issue, or is underlying sequential price leveling off?"
— Vincent Andrews, Morgan Stanley
A: "So I think first on helium, Yeah. So we left the guidance intact. So by default, that kind of means no material change and helium would also be part of that. So to your first point, we did not change it... You know, on The Americas, just to make sure I understand, I mean, price is up 2 percent year over year. Sequentially, we are flat... I would say pricing in Americas on the year over year is tracking where we would expect and what we wanna see."
— Matthew J. White, CFO
Assessment: Two answers, both mildly negative for the thesis. Helium was explicitly not upgraded into guidance, which retires the mid-year-helium-bump upgrade path we identified last quarter. And Americas price at 2% is half the 4% of Q1, with management steering the discussion to the year-over-year frame rather than the sequential one. Pricing power is intact but no longer accelerating, which matters more in a quarter when cost inflation is running ahead of it.
Where the extra half-billion of capital expenditure goes, and the space make-versus-buy risk
The capex increase drew the most analytically interesting exchange of the call, because the follow-up surfaced a structural question about whether Linde's largest new end market wants to buy gas at all.
Q: "I wanted to ask on the CapEx for this year... did you indicate that a lot of that increase was linked to commercial space? And I guess if you can give maybe any other breakdown of that $500 million increase, that would be helpful. And I am just curious with that, if you are building more for that market through your merchant pipeline, What does that mean for space customers' approach in your view? To make versus buy in terms of oxygen, nitrogen and the gases for that market?"
— Joshua Spector, UBS
A: "So starting on the CapEx, yes, you are correct. the CAPEX number on the estimate, was bumped up. Clearly, with the backlog wins, that will drive that... And, yes, there are going to be more commercial space activities in the base CAPEX that also are contributing to that as well... So, absolutely, I expect, you will see a blend of sale of gas and some sale of plant. Generally, those sale of plants can come with what is called an operate and maintain So you tend to run it all as a system."
— Matthew J. White, CFO
Assessment: The capex increase is confirmed as backlog- and space-driven rather than cost-driven, which is the right kind of increase. The make-versus-buy answer is more equivocal than management's tone suggests. Sale-of-plant revenue is one-time and lower-return than a fifteen-year sale-of-gas contract, and management conceded that certain well-capitalised space customers prefer to integrate atmospherics. The space option should be marked down accordingly.
What actually reverses the negative operating leverage
A direct question on the mechanics of margin recovery produced the call's only forward commitment on costs, and it arrived without a number.
Q: "But you could just elaborate a little bit more on some of the actions you are taking to drive a little bit of the margin recovery. I know that you did have some of that within The Americas, some compression. Then and if you could look into maybe the back half or next year, do you expect that negative operating leverage to be resolved? And what would drive that? Is it increased management actions and pricing or productivity?"
— Arun Viswanathan, RBC Capital Markets
A: "But we will likely look to take some cost actions this quarter, depending on the size. that is something we wanna get ahead of... And in some regions, you are seeing growth, which supports it. In other regions, you are seeing inflation without the growth. And that is an area we are gonna focus on specifically for this quarter. Above and beyond our normal productivity initiatives we normally take as part of our everyday DNA. So more to come on that. it is something we will probably give a little more color on and what we have done, in the October call."
— Matthew J. White, CFO
Assessment: The phrase "inflation without the growth" is the most precise diagnosis offered all call, and it is not a mix problem or a timing problem. Deferring both the size of the action and its regional target to the October call leaves the central question of the quarter open for a full reporting cycle. Management is asking to be trusted on margin for one more quarter, immediately after a quarter in which the May margin commitment was quietly not repeated.
Whether the second-half demand outlook supports the volume the guide excludes
A question on the composition of the North American recovery drew the most complete end-market walk of the call, notable for how constructive it was against a guide that assumes none of it.
Q: "Just talking about some of the manufacturing growth assumptions, particularly in North America, it does not seem like you have some of the base volume assumption trend sort of baked into the outlook. But is the bulk of that inflection that you have seen coming from commercial space? I was hoping maybe you could dig into the health of some of the other end markets and what you are sort of in anticipating for the second half?"
— Patrick Cunningham, Citi
A: "Electronics, you know, as you saw year on year had 18 percent growth in second quarter. We expect electronics momentum to carry on for the rest of the year as well... On the industrial markets, and I would say to you, manufacturing, which you kind of specifically mentioned, looks robust. Signals from The US market in particular where the recovery is most prominent looks good... the gases side has been growing mid- to high-single-digit with the hard goods themselves growing double-digit."
— Sanjiv Lamba, CEO
Assessment: The qualitative outlook is meaningfully better than the quantitative guide, which is the classic Linde setup and the reason four straight beats have happened. It also means the unchanged back half is a choice, not a forecast. If the margin actions land, the volume is there to beat on.
Whether APAC's 6% volume is real demand or equipment sales
Two consecutive quarters of 6% APAC volume, against roughly flat throughout 2025, prompted questioning on how much of that is durable base demand rather than one-time equipment revenue.
Q: "Sanjiv, if I look at your volume trend in Asia, it was up 6 percent for a second consecutive quarter. You know, versus, call it, either side of flat, throughout 2025. Can you unpack that a little bit for us? My sense is you have had project startups there and maybe some sale of equipment. Just trying to get a better sense of whether the baseline demand is improving in APAC."
— Kevin McCarthy, Vertical Research Partners
A: "There are 3 components to what is happening in the Asia volumes. Right? There is obviously base volume, which is positive. There are sale-of-equipment, significant sale of elements sitting within there for the electronics customers. That has had a somewhat disproportionate impact in this last quarter that we that we are talking about. And last but not least, there are some, you know, ramp ups."
— Sanjiv Lamba, CEO
Assessment: Management confirmed base volume is positive in APAC but declined to split the 6% into its three components, and described the equipment contribution as disproportionate. So the honest read is that underlying APAC demand is improving and is materially less than 6%. That is still the second cylinder the thesis needed, and it also explains the 120 bps margin decline in the same segment.
Whether 2027 double-digit EPS growth needs macro help
A forward-looking question tested whether the accumulating tailwinds are sufficient to return the company to the top half of its earnings algorithm without a cyclical assist.
Q: "I know it is early, but if you look at next year, 2027, given project start ups, helium maybe being a tailwind next year, helium, growth in space, pricing, productivity. Do you need much of any macro improvement to get to double digit 10 percent E.P.S. growth next year?"
— David Begleiter, Deutsche Bank
A: "As you know, our EPS algorithm lays out the fact that between management actions and capital allocation combined, we should be delivering 8% to 12%. We are not looking for macro as long as macro is not taking away from that. You should expect us to look at that 8% to 12% range And I think we will be, you know, consistent on that as we as we look at next year as well."
— Sanjiv Lamba, CEO
Assessment: A restatement of the standing algorithm rather than a commitment to its upper half, and a refusal to confirm double digits for 2027. Given that 2026 is guided to roughly the bottom of the range in a year with record backlog additions and improving volume, the burden of proof on the algorithm is rising rather than falling.
The Middle East, helium logistics and the knock-on to Asian industrial demand
A question on the Strait of Hormuz produced both the timeline for helium normalisation and an under-appreciated second-order effect on Asian industrial activity.
Q: "Question just around the impact that you have seen on your business and on your customers from what is happening with the Strait Of Hormuz and kind of the Greater Persian Gulf area. Obviously, particularly with helium, but then just with the general business. And then you know, if that issue reside or resolves itself this year, what do you think the impact will be a year out as that starts to normalize?"
— Duffy Fischer, Goldman Sachs
A: "I think we will have a lasting impact for the rest of the year. I do not think you will see normalization. This year, it will, you know, once those issues are resolved, of course, still remains a little bit of a question mark today, Once the issues are resolved, we will see normalization progress at a slower pace than most of us would like. And it will kinda probably take us into the early part of next year... where we have seen an impact of the Middle East crisis is the fact that in Asia, countries highly dependent on hydrocarbons coming out of the Middle East have had to scale back industrial activity."
— Sanjiv Lamba, CEO
Assessment: The most useful new information in the Q&A. Helium normalisation is now an early-2027 event, so the disruption premium and its cost drag both persist through 2026. Separately, the disclosure that India, parts of ASEAN, Australia and to a lesser degree China have scaled back industrial activity on Middle East hydrocarbon dependence is a demand headwind that did not appear in the release and partly offsets the constructive APAC volume story.
What They're NOT Saying
- The 40–60 bps full-year margin commitment was never repeated. In May management committed to full-year margin expansion at "the upper end or even above" its traditional 40–60 bps range, and that commitment underpinned the raised EPS floor. The phrase does not appear anywhere in this quarter's prepared remarks or Q&A. Nobody asked, and management did not volunteer. Withdrawing a quantified commitment by silence is the quietest way to withdraw it.
- Woodside and OCI went unmentioned for a full quarter. Last quarter the ATR and sequestration phases of the large US Gulf Coast project slipped to Q1 2027 and the nitrogen phase was on track for a mid-2026 start. Mid-2026 has arrived. There was no start-up confirmation, no revised timeline and no question. For a project management flagged in February as a multi-billion-dollar backlog contributor, a full quarter of silence is conspicuous.
- EMEA received zero airtime. Neither "EMEA" nor Europe as a region appears in the prepared remarks, and no analyst raised it. A quarter ago it was a major topic. EMEA volumes are still negative and the segment is 25% of sales. The silence suggests both sides of the call have written off a European recovery rather than that the problem was solved.
- The size of the Q3 cost action is deferred to October. Management confirmed cost actions are coming "this quarter," would not size them, and pushed detail to the October call. That means a GAAP charge of unknown magnitude lands in the middle of the very quarter the margin recovery is supposed to be visible in.
- No timeline, no perimeter, and no financials for the Lincare review. Management said it is evaluating strategic fit "both in part and as a whole" but disclosed no revenue, no operating loss, no book value, no perimeter and no decision date. Investors are being asked to price an outcome with none of the inputs.
- Free cash flow, the buyback and return on capital were absent from the call. Free cash flow fell 13%, the buyback pace is down 25% year to date, and ROC has drifted to 23.5% from 24.2%. None of the three was discussed in prepared remarks or raised in Q&A, in a quarter when capex guidance rose half a billion dollars.
- Helium remains unquantified in either direction. Management characterised the contribution as positive on dollars, dilutive on margin, and not in the guide. It has now declined to size helium for two consecutive quarters, which preserves beat potential and leaves the single most-discussed swing factor unmodellable.
Market Reaction
- Pre-print setup: LIN closed at $508.64 on July 30, entering the July 31 before-the-open print up 19.3% year to date against 8.7% for the S&P 500, up 10.5% over the trailing twelve months, and near the upper end of a $389.38 to $546.64 52-week closing range. The trailing 30 days had been softer at -4.7%, so the stock came in off its highs but still with a substantial year-to-date cushion to give back.
- Reaction session (July 31, print before the open): Shares gapped down 6.6% to open at $474.98, traded a $466.88 to $483.95 range, and closed at $478.38, down 5.9% or $30.26. Volume of 5.2 million shares was 2.4 times the 2.2 million 30-day average. The S&P 500 rose 0.7% the same session, so the entire move was stock-specific.
- Scale: The decline erased more than a third of the year-to-date gain in a single session and left the stock 12.5% below its $546.64 52-week closing high.
The sell-off was not about the print. Revenue and EPS both beat, the backlog set a record, and the full-year floor went up. It was about three things the market added together and did not like: a margin line moving the wrong way in the quarter volume finally arrived, a guide that raised the floor while the Street was modelled to the ceiling, and a half-billion-dollar increase in the capital budget landing in the same release as a 13% decline in free cash flow.
The second of those deserves emphasis, because it explains the magnitude. Consensus sat near $17.89 for the year, effectively at the $17.90 top of Linde's own range. A raise that lifts the floor $0.10 and leaves the ceiling untouched gives a model anchored to $17.90 precisely nothing, while confirming that management will not underwrite the volume recovery it just described qualitatively. Combined with a Q3 midpoint of $4.50 against a ~$4.53 Street, the guide was a downward revision for anyone positioned at the top of the range, which after a 19% year-to-date run was most of the marginal buyer base.
The third factor is the one likely to persist. Linde has been owned for years as a low-volatility compounder with a shrinking share count and rising free cash flow. This quarter it presented as a growth-capex story: capex guidance up $500 million, free cash flow down 13%, buybacks down 25% year to date, and return on capital drifting toward the low 20s. That is a defensible use of capital at double-digit unlevered IRRs, and it is a different security than the one many holders thought they owned. The gap-down at the open, before any of the call's nuance was available, says the release alone was enough.
We read the 5.9% decline as broadly rational rather than an overreaction. A 60 bps margin miss on a franchise whose entire premium rests on margin compounding is exactly the kind of news that should compress a 28.6x multiple. What the sell-off does not do is fully discount the offsetting positives in the same release: a record backlog with an early-delivered commitment, positive volume in two of three regions, and a portfolio action that could remove the source of the problem.
Street Perspective
Debate: Is the margin decline a mix artefact or the start of a trend?
Bull view: The bull case on the Street holds that two of the three drivers are self-evidently good news. Higher US hard-goods sales and low-margin electronics equipment in APAC are demand signals that seed contracted gas volume behind them, and both dilute margin while expanding the earnings base. Strip those and the third driver, Lincare, is a non-core distribution business management has now put explicitly on the table. Ex-homecare, Americas margin expanded.
Bear view: The bear camp points out that the same temporary-mix explanation was given for APAC in Q1 with a commitment to recover to 29%, and APAC is still 120 bps below prior year. Cost of sales rose 160 bps as a share of sales, which is broader than any single business. Management's own diagnosis, inflation without growth in certain regions, describes a cost problem, not a mix problem, and it is now taking a discrete cost action to address it.
Our take: The bears have the better of it for the next two quarters and the bulls for the two after that. The mix explanation is genuine but has now failed one forward test (APAC), and the withdrawal by silence of the 40–60 bps commitment is the more informative signal. That said, none of the three drivers impairs the pricing franchise, and a Lincare separation would resolve the largest of them structurally. We expect margin to trough this year rather than deteriorate further.
Debate: Does the record backlog justify the capital intensity?
Bull view: A $1 billion single-quarter backlog addition into advanced-node fabs, on top of a Taiwan JV building a further $800 million outside the reported figure, is the highest-quality growth Linde has secured in years. These are fifteen-year contracted, take-or-pay-structured, inflation-indexed assets at post-tax double-digit unlevered IRRs. Spending $5.5–$6.0 billion to lock in that revenue is straightforwardly value-creating, and the free cash flow dip is a timing artefact of building ahead of start-up.
Bear view: The skeptical view is that the market pays Linde a premium multiple for cash returns and a shrinking share count, not for a growth capex cycle. Capex has now been raised twice from the original plan, buybacks are down a quarter year on year, return on capital has fallen for a second straight period, and the backlog takes two to three years to build and further time to ramp. Investors are being asked to fund 2029 earnings out of 2026 cash flow at a 27x multiple.
Our take: We side with the bulls on the economics and the bears on the timing of the payoff. Management's own framing is that projects hit their return threshold only after a multi-year ramp, so the earnings contribution from this backlog is a 2028-and-beyond event. The correct response is not to sell the capital allocation, it is to stop paying a cash-return multiple for a growth-capex phase. That is a valuation adjustment, and it happened on July 31.
Debate: Is Lincare an overhang or a catalyst?
Bull view: One camp argues the strategic review is the best news in the release. Separating a reimbursement-exposed, labour-intensive homecare distribution business from a 29%-margin industrial gases franchise would lift group margin, remove a persistent explanation-requiring drag, simplify the story, and free capital. The language used, evaluating fit "in part and as a whole," is the strongest signal Linde has given that it will act.
Bear view: The counter is that a business under visible margin pressure with a policy-driven headwind is a poor thing to be selling, that Linde has referenced Lincare's difficulties for two years without acting, and that a review with no perimeter, no timeline and no disclosed financials could easily run for a year. In the meantime it is a confirmed drag of roughly $130 million annualised on the company's own upward correction.
Our take: A catalyst with no clock is an overhang. We give management credit for the disclosure and none yet for the outcome. The asymmetry favours patience: if a sale happens the multiple and the margin both improve, and if it does not, the drag is already in the numbers. What we will not do is pay in advance for an outcome with no announced perimeter.
Debate: Has the multiple corrected enough?
Bull view: At $478.38 the stock trades near 26.9 times the raised full-year midpoint, down from roughly 28.6 times at the Q1 reaction close, for a franchise with oligopoly pricing, a record backlog, 33 consecutive years of dividend growth and a volume recovery in two of three regions. Quality compounders rarely offer this kind of entry, and the de-rating already discounts a margin problem management says is temporary.
Bear view: The bear response is that 27 times for roughly 8% earnings growth is still a PEG above three, that the growth is guided to decelerate to about 7% in the second half, and that the buyback support is being redirected to capex. On a cash-return basis the stock has become more expensive, not less, and a 1.3% dividend yield does not bridge the gap.
Our take: Closer to fair than at any point in our coverage, and not yet compelling. The de-rating is real and it is the single largest change in the risk/reward this quarter. But it corrected a multiple that was priced for margin expansion into a quarter that delivered margin contraction, which means it removed an excess rather than created a discount. We want either one clean quarter of year-over-year margin expansion or a Lincare resolution before paying up.
Model Update Needed
| Item | Prior assumption | Suggested | Reason |
|---|---|---|---|
| FY26 adjusted EPS | $17.75 (old midpoint) | $17.80 (new midpoint) | Floor raised $0.10; anchor to the new $17.70–$17.90 range rather than the Street's $17.89. |
| FY26 adjusted operating margin | +50–70 bps (upper end of 40–60) | Flat to modestly down vs. FY2025's 29.8% | H1 came in at 29.7% against 30.1%. The upper-end commitment was not restated on this call. |
| Helium contribution | Neutral-to-positive; asymmetric upside | Small positive on dollars, dilutive on margin | Guidance explicitly unchanged; dislocation costs gross up alongside price; normalisation pushed to early 2027. |
| US homecare (Lincare) | Not separately modelled | ~$130M annualised drag (implied); carve out for scenario analysis | Management corrected an analyst's ~$100M framing upward by about 30%. Strategic review now live. |
| FY26 capital expenditure | $5.0B – $5.5B | $5.5B – $6.0B | New backlog award plus rising base capex for commercial space. |
| FY26 free cash flow | Growing with earnings | Down year over year | Q2 FCF fell 13% on 14% capex growth against 3% operating cash flow growth. |
| Buyback pace | ~$800M–$1.4B per quarter | Step down toward the low end | H1 net share purchases of $1,656M are 25% below H1 2025 as capital shifts to the project backlog. |
| Sale-of-gas backlog | $7.1B rising toward an "8 handle" | $8.1B, held with an 8 handle through year-end | Delivered two quarters early; ~$1.3B of start-ups in H2 to be replaced from the pipeline. |
| APAC segment margin | Recovering toward 29% | 28–28.5% while electronics equipment mix persists | The promised recovery did not happen; equipment dilution is proving structural, not seasonal. |
| Commercial space | Scaling faster than $1B by 2030 | $1B+ by 2030, with a sale-of-plant mix | Framing reverted to the original timeline; certain customers prefer to self-supply atmospherics. |
| Return on capital | ~24% | Low-to-mid 20s | 23.5% in the quarter against 24.2% for FY2025, consistent with management's own guidance. |
Valuation impact: At the $478.38 reaction close, LIN trades near 26.9 times the $17.80 full-year midpoint, against roughly 28.6 times the then-midpoint at the Q1 reaction close, with a dividend yield near 1.3% on the $1.60 quarterly rate. We cut our 12-month fair value to approximately $500, or about 28 times the full-year midpoint, from $525. The multiple haircut reflects reduced visibility on margin expansion and a capital-allocation mix that is shifting away from cash returns; the smaller cut than the share price decline reflects the record backlog and the early-delivered fab win. That is roughly 4% of price upside plus the yield, which is a market-perform total return.
Thesis Scorecard Post-Earnings
| Thesis Point | Status | Notes |
|---|---|---|
| Bull #1 — Pricing power compounds EPS through the cycle | Confirmed, decelerating | +2% price group-wide and in every region, tracking local inflation. But Americas price halved from 4% to 2%, and for the first time price plus productivity failed to offset cost inflation. Status tag holds at ON TRACK. |
| Bull #2 — Secular backlog (electronics, clean energy, space) drives durable growth | Confirmed, strengthened | Sale-of-gas backlog +$1B to a record $8.1B, hitting the "8 handle" two quarters early; total project backlog $11B; electronics +18%; a further ~$800M in the Taiwan JV outside the backlog. Offsetting: space reverted to the original $1B-by-2030 framing with a sale-of-plant caveat. Status tag holds at ON TRACK. |
| Bull #3 — Best-in-class capital allocation and margin self-help | Challenged | Adjusted operating margin -60 bps YoY and -50 bps sequentially; H1 at 29.7% against 30.1%; the May commitment to 40–60 bps of full-year expansion was not restated. Capital allocation intact but shifting: capex +$500M, buyback pace -25% YTD, ROC 23.5% from 24.2%. Status tag moves ON TRACK → AT RISK. |
| Bear #1 — Full valuation caps upside | Easing | The multiple corrected from roughly 28.6x to 26.9x forward in one session. Still above three times the growth rate, but the excess has been removed. Status tag moves MATERIALIZING → EMERGING. |
| Bear #2 — Base-volume growth remains absent | Easing further | Group volume +2%, matching price for the first time this cycle. Americas +2%, APAC +6% for a second straight quarter, EMEA improved to -1% from -3%. Two cylinders are now firing. Status tag moves EMERGING → CONTAINED. |
| Bear #3 — Guide below aspiration and below Street | Confirmed | FY midpoint of $17.80 sits below the ~$17.89 Street; Q3 midpoint of $4.50 below the ~$4.53 Street; the raise exactly offset the beat and the back half was left unchanged; implied H2 growth of about 7% against 10% in H1. Status tag moves EMERGING → MATERIALIZING. |
| Watch item — Helium shortage optionality | Downgraded | Guidance explicitly unchanged, dollar contribution "not as large as we would like," margin-dilutive on the dislocation-cost gross-up, normalisation pushed into early 2027. No longer an asymmetric option. Status tag moves EMERGING (positive) → CONTAINED. |
| New watch item — US homecare (Lincare) portfolio risk | New, emerging | ~$130M annualised drag implied by management's own upward correction; majority of the Americas margin decline; strategic review live with no perimeter, timeline or disclosed financials. Set at EMERGING. |
Overall: The thesis is more balanced and less coherent than it was in May. The growth half improved on every measure that matters: the backlog trigger we named as an upgrade condition fired early and large, volume turned positive in a second region, and the demand commentary was the most constructive of our coverage window. The quality half went the other way: the margin self-help pillar that justified the premium multiple was challenged for the first time, the helium option we treated as free upside was repriced to a small positive with a margin cost, and the guide raise was arithmetically neutral. On net, this is a quarter in which the business got bigger and the model got worse.
Action: Maintain Hold, fair value reduced to approximately $500 from $525. Two of the three upgrade triggers we set out in May have now fired, the signed fab win and the multiple pullback, and the third, a broadening volume recovery, is most of the way there. We are not upgrading on them because the same quarter broke the pillar those triggers were supposed to be layered on top of. The upgrade conditions from here are narrower and more concrete than they were: one quarter of year-over-year adjusted operating margin expansion, or a definitive resolution on the US homecare portfolio. Either would restore the compounding logic; the backlog and the price have already done their part.