The kiln that cannot be moved: Vicat, Montalieu-Vercieu and the arithmetic Isère cannot yet see

I. The retrieval limitation, stated before the analysis

A publication that reports on evidence has to report on the condition of its own access. During the research window for this dossier, repeated automated requests to vicat.fr and vicat.com — corporate pages, investor pages and press-release PDFs alike — returned HTTP 202 responses with empty bodies. For sourcing purposes that is indistinguishable from a refusal: no page content could be read end to end. Two of the most substantial press accounts of the Isère carbon-capture project, in Le Monde and Les Echos, are behind subscriber paywalls, and only their headlines and leads were visible.

So each figure in this dossier carries its access condition in the source record: opened in full, or read from a search-engine snippet of the company's own page, or paywalled and therefore used only for the fact of publication. Where a number exists only inside a document we could not open, it is refused rather than approximated. That is slower and thinner than a brochure. It is also the only version of this dossier that can be audited.

II. The identity: an Isère headquarters running a twelve-country group

Vicat is a French société anonyme with a board of directors, SIREN 057 505 539, registered office at 4 rue Aristide Bergès, Les Trois Vallons, 38080 L'Isle-d'Abeau, NAF code 2351Z (cement manufacture). Its shares trade on Euronext Paris under the mnemonic VCT, ISIN FR0000031775. The company's own legal-information page states a share capital of €179,600,000; we record that as a single-source figure read from a snippet, not from a filing we opened.

Two facts about the group are worth holding together. Its own material describes it as "un groupe industriel international français, créé il y a plus de 170 ans dans la lignée de Louis Vicat, toujours contrôlé par la famille fondatrice" — family control, stated by the company. And its recruitment listings describe a group "présent dans 12 pays", employing "près de 10 000 collaboratrices et collaborateurs", with "65% à l'international" (Vicat recruitment listings, read 4 September 2026).

What we do not print is the ownership table. The only percentage breakdown we located — a free float around 34%, two holding companies around 32% and 27%, the founding family under 5% — comes from a market-data aggregator, not from Vicat's own registration document, which we could not open. A shareholding structure is exactly the kind of claim that should rest on a regulated filing, so it is refused here. The qualitative statement stands, sourced to the company; the arithmetic does not.

III. The financial frame: two audited years, and one number we still refuse

For the year ended 31 December 2024, the group's results release of 18 February 2025 reports consolidated sales of €3,884m against €3,937m in 2023 — a reported decline of 1.3% alongside organic growth of 2.3%, the gap being currency and scope. EBITDA was €783m, a 20.2% margin. EBITDA in the United States rose 25.8% to €190m. Free cash flow rose 26.7% to €373m, and leverage stood at 1.6x. The chief executive called it a record year.

The following year, published 16 February 2026, is the one that matters for a reader deciding whether this decarbonisation programme is being funded from strength. Sales came in at €3,854m: down 0.8% as reported, but up 3.3% at constant scope and exchange rates, with the fourth quarter accelerating to +8.1%. EBITDA was €771m, down 1.6% as reported and up 3.7% like-for-like, at a 20% margin, with current EBIT of €445m (11.5% margin). Net debt fell by a further €85m. The release attributes the pattern to stabilisation in French cement, recovery in Switzerland and strong growth in the Mediterranean — against unfavourable currency effects.

Two things follow. First, the reported-versus-like-for-like split is the whole story of this company's last two years, and either half alone is misleading: a reader shown only "sales down 0.8%" sees a shrinking cement group, and a reader shown only "+3.3%" sees one that never met a currency headwind. Both are printed here, in that order, in every year. Second, capital expenditure is refused: no capex figure appeared in any text we could retrieve, and for a group financing kilns, a calcined-clay line and a capture unit at once, capex is far too load-bearing to reconstruct from margins. A 2025 net income of €275m appears on a financial-data aggregator, not in a company text we could open; it is recorded with that provenance and is used to support no argument here.

One conflation is worth naming because a casual reader will make it. The corporate summary pages carry roughly €3.8bn and close to 10,000 employees as a current-state description. That rounding sits between two different audited years — €3,884m for 2024 and €3,854m for 2025 — and quoting it as either is how a 0.8% decline disappears. The two audited figures are printed side by side above, and the rounded corporate figure is not used as a substitute for either.

IV. Montalieu-Vercieu: what a kiln in Isère is actually being asked to do

The Montalieu-Vercieu plant sits on the Rhône in northern Isère. The only production-capacity figure we could source is 2,000,000 tonnes, and it comes from the plant's own 2019 regulatory annual activity report under the French ICPE regime. That is a 2019 number; it is printed with its date and must not be read as current capacity, because no more recent published figure was found on any page we could open. A prefectural formal-notice order of 26 June 2020 from the Isère departmental services adds a detail no corporate page states: the plant known as "usine de Montalieu" is administratively located on the commune of Bouvesse-Quirieu, under ICPE registration 061.02824. Regulatory paperwork is where an industrial site stops being a brand and becomes an address.

What the site is now known for is VAIA — Vicat Advanced Industrial Alliance. The company's own release states that the project follows the signature of an ecological transition contract with the French government under the national strategy for decarbonising the fifty most CO₂-emitting industrial sites, and that it "permet de capturer et séquestrer 1,2 million de tonnes de CO₂ par an". That framing is the finding, not the tonnage: the state did not subsidise a cement plant, it contracted a site on a list. French construction-industry press reported the same order of magnitude — capture and storage of 1.2 Mt CO₂/year at Montalieu-Vercieu — in April 2025. National press reported in November 2025 that the captured CO₂ would be transported for storage under the Adriatic; that account is paywalled, so it is recorded as published and not quoted beyond its lead.

One dating detail, kept visible: the launch is dated 3 March 2025 on the company's news page and 4 March 2025 in the dateline of the press-release PDF it links to. It is a one-day discrepancy of no consequence to the argument, and it is printed because a dossier that silently picks one date cannot be checked against the record.

The transport and storage leg is not a cement problem, which is why the public consultation names the counterparties: the Rhône décarbonation project is carried by Vicat with SPSE, Elengy and RTE — a pipeline operator, an LNG terminal operator and the electricity transmission system operator — and runs its own consultation site, with published citizen contributions. That combination is the actual finding of this section. A cement plant cannot decarbonise as a plant. It can only decarbonise as a node on a pipeline, a grid connection and a storage contract, each with its own regulator, its own consultation and its own timetable.

The engineering economics, from the one account we could read in full

The company's own domains would not open for us. A news-agency dispatch of 14 November 2025, filed after a site visit and carried openly by two named outlets, did — and it contains the magnitudes the corporate record withholds. According to that account: the Montalieu-Vercieu plant is the largest cement works in France and sits among the country's fifty most CO₂-emitting industrial sites; its rotary kiln reaches 1,450°C and burns coal, biomass and a rising share of waste — shredded tyres, old furniture, animal meal — around the clock; the captured CO₂ would leave through roughly 300 km of repurposed gas pipeline turned "carboduc", be liquefied at Fos-sur-Mer using the Fos-Tonkin terminal, shipped, and stored in geological reservoirs operated by ENI off the Italian coast; the network is sized at 4 million tonnes of CO₂ a year, of which 1.2 million for Vicat; the whole Rhône décarbonation programme is costed at €1bn to €1.5bn; Vicat's own intended share is €600m to €900m; studies began around 2023; and the company expects an investment decision at the end of 2027 for commissioning three years later. Its programme director is named in that account and concedes the timetable is ambitious.

Three consequences, and they are the reason this section exists. First, the sequence is the risk: nothing here is under construction. A final investment decision at end-2027 means the capture unit is a study with a grant behind it, and a reader deciding a career on it should date the commitment, not the ambition. Second, the reuse is the strategy. An old gas pipeline and a mothballed LNG terminal are what make a €1–1.5bn number thinkable at all; this is not a greenfield carbon industry but a repurposing of Rhône-valley infrastructure, which is also why a grid operator sits among the co-developers. Third, the commercial motive is stated, not inferred: the programme director's argument is that the first movers into low-carbon cement gain an advantage as construction rules tighten — one of the consultation's guarantors puts the regulatory deadline at 2031 for a massive reduction in construction's carbon footprint. That is a market being created by regulation, and it dates when the skills are wanted.

These figures are announced intentions reported by a news agency, and this dossier keeps them in that class. They are not an audited capital-expenditure line: no capex figure appears in any Vicat financial text we could retrieve, a €600–900m intention is a range and not a commitment, and the ratio a reader will want — public grant against private spend — cannot honestly be computed from a subsidy figure read in a snippet and an intention range read in a dispatch. Both numbers are printed with their provenance so that anyone can see exactly how firm each one is.

The funding record: one amount printed, one refused, one date order left crooked

Public support for VAIA appears in the company's own releases under two distinct schemes: the European Innovation Fund and the French Grands Projets Industriels de Décarbonation (GPID). The order of the dates does not line up, and we do not straighten it:

• a release dated 12 February 2026 announces the GPID award and refers back to "la sélection du projet par le programme européen Innovation Fund en novembre 2025";
• a release dated 30 March 2026 announces that the group is "officiellement lauréat" of the Innovation Fund subsidy, after CINEA formalised the list of successful Innovation Fund 2024 projects.

So the formal confirmation of the European award is dated after the national release that already treats it as settled — a selection in November 2025, an official award list in 2026. It is printed, not resolved, because the two releases are by the same publisher: the company.

On amounts, the record is asymmetric and this dossier reflects that asymmetry rather than smoothing it. The Innovation Fund press release states a subsidy of €149,998,000. We print that figure with its provenance stated plainly: it was read from the indexed text of Vicat's own press-release PDF, a document our fetcher could not open directly (see Section I). No amount is printed for the GPID award — none appeared in any text we could retrieve, and a national decarbonisation grant with no stated figure is not a number waiting to be estimated. The consequence is worth stating: the best-documented public money behind Isère's largest carbon project is European, not French, in the sense that only the European figure is retrievable. Separately, we found no page labelling the VAIA subsidy "France 2030". The named schemes in our evidence are the Innovation Fund, the GPID and the ecological transition contract, plus the ADEME-operated Plan de Relance for a different site. The France 2030 label is therefore refused here.

V. Argilor: the technology that is not in Isère, and why it belongs in this dossier

The group's most advanced clinker-substitution asset is at Xeuilley, in Meurthe-et-Moselle — not in Isère. Vicat announced in June 2021 that it was a laureate of the first Plan de Relance programme financed by the government and operated by ADEME, accelerating its decarbonisation plan through Argilor, a calcined-clay line. The trade press reported the same month that the project would use a flash calciner supplied by FLSmidth and would "replace clinker with an environmentally-friendly clay and cut up to 16 per cent of CO₂ emissions, compared to existing cement products". Commissioning was reported in May 2023 as scheduled for the first quarter of 2024, and an international decarbonisation coalition describes the result as the first European industrial-scale flash calciner to enter operation. The ADEME grant amount is refused; it appeared in no text we could open.

Argilor matters to an Isère reader for a reason that has nothing to do with geography. Clinker substitution and carbon capture are not alternatives — they are sequential levers on the same balance sheet, and a group runs them in different departments at different times. In May 2026 the company announced both together at site level: electrification at Xeuilley and e-fuel production at Montalieu. The engineer who commissions a calciner in Lorraine is the engineer a capture project in Isère will want three years later. Career mobility inside this group runs along the decarbonisation programme, not along the map.

What is refused here is the headline every cement decarbonisation story wants: a number. No CO₂-intensity target (kg CO₂ per tonne of cement) and no clinker-factor target is printed. The company's decarbonisation page describes four levers qualitatively — modernising installations, alternative fuels, reducing the clinker rate, capturing residual emissions — and the retrievable text contains no numeric target. The "up to 16%" figure above is a supplier-reported product comparison for one line, not a group intensity target, and it is not promoted into one.

VI. What this actually means for the people who get hired

Vicat's own recruitment portal, job.vicat.net, showed 95 open positions when we indexed it, alongside an open-application channel. The postings that are useful as evidence are the ones that name a site and a discipline. In Isère: an apprenticeship Ingénieur Structures & Matériaux at Vicat Sysnergie in L'Isle-d'Abeau, a Chargé(e) Qualité at Sigma Béton at the same location, and an Ingénieur Digitalisation posting spanning Grenoble and Créchy. In the wider group: an Ingénieur Digitalisation at Xeuilley — the Argilor site — and an Ingénieur Conception & Programmation robotique at Vicat Sysnergie in Chambéry.

Read as a set rather than as adverts, those five postings describe a hiring pattern that contradicts how cement is usually imagined as a career. Four of the five are engineering roles, two are digital, one is robotics, and all five are work-study contracts — meaning the group is building this capability through apprenticeship rather than buying it finished. An internal engineering entity (Vicat Sysnergie) and a materials-testing entity (Sigma Béton) appear as the employers of record in Isère, which tells a candidate something a group careers page never says: the interesting technical work is organised in subsidiaries with their own names.

The qualifications that travel here are process engineering, thermal and combustion knowledge, materials characterisation, industrial data and automation. The decarbonisation programme does not create "green jobs" as a separate category; it converts existing process roles into instrumented ones — a capture unit is a chemical plant bolted onto a kiln, and it has to be operated by people who can read both. We print no salary figure, no site headcount and no hiring volume: no Vicat-published number exists for any of them in our evidence, no job board or salary aggregator was consulted, and there is no structured job-family taxonomy or named apprenticeship programme page we could open. A candidate deserves the shape of the demand, which is documented. The size of it is not.

VII. The finding

The materials layer is the part of the sovereignty stack that cannot be relocated and cannot be hidden. A fab can be announced; a kiln has a permit file, a 2019 capacity declaration and a measurable emissions stack. That makes Vicat unusually easy to hold to account and unusually hard to write about: the documents that would settle the numbers — the registration document, the full results release, the grant decisions — are precisely the ones our fetcher could not open.

So the honest verdict is two-sided. The direction is documented to a standard almost no other Isère anchor matches: a named project, a stated tonnage, three named infrastructure counterparties, a state ecological-transition contract on a fifty-site list, two public schemes, a live consultation with published citizen contributions, one European subsidy of €149,998,000, a reported programme cost of €1bn to €1.5bn with an intended Vicat share of €600m to €900m, a stated investment decision at the end of 2027, and an operating asset in another department already running the substitution lever. The firmness of the rest is not: no GPID amount, no audited capex line behind the announced range, no CO₂-intensity or clinker-factor target, no current plant capacity, no site workforce. That asymmetry is itself the story — the European instrument publishes its number, the national one does not, and the private half of the ledger exists so far as an agency-reported intention rather than a board decision. Anyone who tells you the cement transition in Isère is funded to a known total is quoting a document we could not open. We will re-read this anchor when those documents are retrievable, and the corrections will be published with dates.