Coverage is the part of the bank that owns the client relationship. A coverage banker (also called senior banker, relationship manager or global relationship manager, GRM, when the client is multinational) sells no product of his or her own. The banker’s job is to know the client group better than any other bank does, to anticipate what it will need, and to bring in the product desks at the right moment: lending, debt capital markets, rates and foreign exchange, cash management, trade finance, advisory. Read this module before the product modules of blocks B and C, because every deal described there starts in a coverage conversation.
Where it sits. At Meridian Bank, coverage belongs to the Corporate and Institutional Bank. It is organised by client size (large corporates, mid-caps), by country and, for the largest clients, by sector: energy and utilities, industrials and building materials, consumer, technology, financial institutions. Sector teams matter for this course because sustainability questions are sector questions: the transition of a cement producer has little in common with that of a retailer.
Clients. A senior banker in large-corporate coverage typically looks after fifteen to thirty client groups. The contacts are the chief financial officer, the group treasurer, the head of funding, the head of mergers and acquisitions and, more and more often, the chief sustainability officer. The treasurer is the day-to-day counterpart; the CFO decides which banks belong to the core group.
How it earns money. Coverage has no revenue line of its own. Its performance is measured on the revenues that all product lines book with its clients, net of risk and capital: the relationship profit and loss account. The mechanics are simple and somewhat brutal. Large corporates expect their core banks to lend, mostly through a revolving credit facility that pays little (see B03). In exchange, lenders expect to be chosen for the business that pays: bond mandates, hedging, cash management, acquisition finance, advisory. The total amount of fees and margins that a client pays to all its banks in a year is called the wallet. For a large investment-grade group it runs to tens of millions of euros a year, more in a year with an acquisition. A bank’s portion of it is its share of wallet. The banker’s task is to make the share of wallet at least as large as the share of the lending, and the return on the capital committed to the client at least equal to the hurdle of 12% after tax.
The team and the adjacent functions. Around the senior banker sit one or two junior bankers or analysts who build the account plans, the pitch books and the profitability simulations. Next to them:
| Function | What it does for coverage | Document |
|---|---|---|
| KYC and onboarding | Identifies the client, its owners and its risks | KYC file, periodic review |
| Credit analysts | Rate the client, write the annual review | Credit application |
| ESG risk team | Screens the client against sector policies | ESG assessment memo |
| Sustainable finance advisory | Assesses the transition plan, structures labels | Transition scorecard |
| Product partners | Price and execute | Term sheets, pitches |
Credit risk (second line), compliance and the reputational risk committee give the independent opinions and decisions. The business manager of the coverage department produces the relationship profit and loss account every month from the management information system.
A realistic week. Monday: pipeline meeting with the product partners, client by client; the banker updates the customer relationship management tool (CRM) with the calls of the previous week. Tuesday: lunch with a treasurer who mentions that the board has just approved a carbon-capture investment; a call report goes to loan structuring, DCM and the carbon desk the same afternoon. Wednesday: credit committee for the annual review of another client; the banker presents, the credit analyst defends the rating, risk challenges. Thursday: internal rehearsal of a strategic dialogue meeting with a client’s CFO, with the sector head and the sustainable finance adviser. Friday: account-plan review with the head of coverage, where the banker has ten minutes per client to explain why the capital committed is, or will be, properly paid.
WHY IT MATTERS ON THE DESK — The coverage banker is the only person in the bank who sees the whole client: every exposure, every revenue, every risk, every opportunity. Sustainability has made that view more valuable. The client’s transition plan is at the same time a credit question (will carbon costs erode cash flows?), a reputational question (can we be seen financing this?), a regulatory question (what does this exposure do to our own metrics?) and the best source of new business in a decade (who finances the capex?). Only coverage is positioned to connect the four.
IN ONE SENTENCE — A bank does not judge a loan to a large corporate on its own return; it judges the whole relationship, of which the loan is the entry ticket.
INTUITION — A supermarket sells milk at almost no margin because customers who come for milk fill the rest of the trolley. The revolving credit facility is the milk of corporate banking. The coverage banker is the store manager who must check, client by client, that the trolley is in fact being filled.
The logic comes from the economics of lending to investment-grade companies. Margins are thin, capital consumption is not, and the stand-alone return of a facility is usually well under the hurdle: module B03 finds 3.96% on a EUR 750m back-up line, against a hurdle of 12%. The bank accepts this because the lenders form the group from which the client chooses its bookrunners, hedge providers and cash-management banks. Treasurers call it reciprocity and manage it explicitly, often with a spreadsheet that compares, for each bank, the capital it commits with the fees it has received.
Three notions structure the banker’s reasoning.
Wallet. The annual fees and margins paid by the client to all banks, split by product. It is estimated, never observed: the banker rebuilds it from the client’s financial statements (drawn debt and facilities, bond issuance, hedging volumes disclosed in the notes, number of countries for cash management), from league tables and from what the treasurer is willing to say.
Share of wallet. Meridian’s revenues with the client divided by the wallet, overall and by product. The comparison that matters is with the lending share, Meridian’s portion of the client’s committed bank facilities. A bank with 10% of the lending and 5% of the wallet is subsidising its competitors.
Client tiering. Clients are classified by potential and by current return, usually into three or four tiers. The tier decides how much banker time, product-partner attention and balance sheet the client receives. An account plan is the yearly argument for keeping or changing the tier.
FORMULA
is the lending revenue (net margin over the funds transfer price, plus fees on credit products); the ancillary revenue (everything else: capital markets, hedging, cash management, advisory); and the cost-income ratios of the two families; the expected loss; the tax rate; the capital allocated per euro of risk-weighted assets; the risk weight of the credit exposure and its exposure at default; the euros of risk-weighted assets (operational and market risk, counterparty risk) carried by each euro of ancillary revenue.
Term by term. The numerator is an after-tax profit: revenues, minus the cost of the people and systems that produce them, minus the average credit loss. The denominator is the capital the bank must hold: almost all of it comes from the credit exposure, a little from the fee businesses. Ancillary revenue is therefore precious. One euro of it adds euro of profit for euro of capital, a marginal return of 92%, where the loan returns a few per cent.
Take a plain example, outside the case. Meridian lends EUR 100m for five years to a BBB client at a margin of 100 bp. The funds transfer price takes 30 bp of liquidity premium, so the net margin is 70 bp and the lending revenue EUR 0.70m. Costs are EUR 0.28m. With a one-year probability of default of 0.20% and a loss given default of 40%, the expected loss is EUR 0.08m. The foundation IRB risk weight is 39.02%, so risk-weighted assets are EUR 39.02m and allocated capital EUR 5.07m. The stand-alone RAROC is 5.03%. To reach 12%, the relationship must also generate EUR 1.36m of ancillary revenue a year: almost twice the lending revenue itself.
INTERACTIVE LAB — The entry ticket: how much ancillary revenue a loan must bring
Compute the stand-alone RAROC of a drawn loan and the ancillary revenue the relationship needs for the whole to reach the hurdle.
| Parameter you can move | Range | Default |
|---|---|---|
| Drawn loan | 20 to 500 EUR m | 100 EUR m |
| Margin over the reference rate | 40 to 400 bp | 100 bp |
| FTP liquidity premium | 0 to 80 bp | 30 bp |
| One-year PD | 0.03 / 0.07 / 0.2 / 1 / 4 | 0.2 % |
| Ancillary revenue actually earned | 0 to 5 EUR m | 0 EUR m |
| RAROC hurdle after tax | 8 to 16 % | 12 % |
| Result at default settings | Value |
|---|---|
| Lending revenue (net of FTP) | 0.700 EUR m |
| Expected loss | 0.080 EUR m |
| RWA of the relationship | 39.02 EUR m |
| Allocated capital | 5.07 EUR m |
| Relationship RAROC | 5.03 % |
| Ancillary revenue needed to reach the hurdle | 1.36 EUR m |
| Marginal RAROC of one euro of ancillary revenue | 92.3 % |
Try it. Set the PD to 1.00% (BB): at the same margin the loan returns almost nothing and the cross-sell needed more than triples. Pricing for the rating comes first.
PITFALL — Cross-sell is a promise, the loan is a commitment. The capital is consumed from the day of signing; the ancillary revenue arrives later, if it arrives. Account plans that reach the hurdle “on a fully loaded basis” with mandates the client has never mentioned are the standard failure of coverage. A good plan lists each expected mandate with a probability, a date and the name of the person at the client who decides.
The account plan. Once a year, for each client of the upper tiers, the banker writes a document of five to fifteen pages: the client’s strategy and financial profile, its banking group and Meridian’s position in it, the relationship profit and loss account over three years, the wallet and share of wallet, the exposures and limits, the opportunities by product with amounts and probabilities, the risks, and the request made to the bank (more balance sheet, less, a change of tier). It is reviewed by a client committee chaired by the head of coverage, with the product heads in the room.
The call report. After each client contact, a short note in the CRM: who was met, what was said, what follows. Call reports are the raw material of the pipeline and the evidence, later, of what the bank knew and when. That matters for conduct and for inside information: a banker who learns of an unannounced acquisition must wall-cross the right people and no others.
The credit application. Coverage is the first line of defence for credit risk. The banker sponsors each credit request, written with the credit analyst, and presents it to the credit committee. The annual review of the client’s rating and limits follows the publication of its accounts.
The KYC file. Know-your-customer due diligence is a legal obligation under anti-money-laundering rules: identity, beneficial owners, sanctions, politically exposed persons, source of funds, purpose of the relationship. The banker owns the file’s content; a dedicated team performs the checks. It is reviewed periodically, every one to five years depending on the risk class, and on trigger events.
The pitch. A tailored presentation proposing an idea: a refinancing, a hedge, an acquisition target. Senior bankers are judged by clients on the quality of the ideas they bring, which is why sector knowledge is the real currency of coverage.
Coverage carries three risks of its own. Concentration: the banker always wants more limit for a good client; single-name and sector limits exist to say no. Conflicts and tying: making a loan formally conditional on other mandates is restricted by competition and conduct rules in several jurisdictions, and by the client’s own governance; reciprocity is an expectation that is discussed, never a contractual condition. Information: the banker sits on the private side of the information barrier and must not pass confidential client information to the public side (sales, trading, research) except through formal wall-crossing procedures.
IN ONE SENTENCE — The ESG client dialogue is the same strategic dialogue as before, extended to one question that now drives the client’s capex, its risk and the bank’s own constraints: how will your business model survive and pay for the transition?
Three things have changed in practice.
First, the onboarding and review process has an ESG branch. Alongside the KYC questions, the client is screened against the bank’s sector policies and exclusion lists, and scored for environmental, social and governance risk. Bankers call it KYC-ESG, although the legal basis is different: prudential rules on risk management and the bank’s own commitments, where KYC rests on anti-money-laundering law.
Second, for clients in high-emitting sectors, the bank forms a view on the transition plan. That view feeds the credit rating, the appetite for long tenors, the eligibility for labelled products, and the bank’s own sector targets.
Third, the transition is a source of business. A hard-to-abate client must invest heavily, hedge new exposures (carbon allowances, power), restructure its portfolio (disposals, acquisitions) and explain itself to investors. Each of these is a mandate for some desk.
INTUITION — A doctor who treats a smoker does not refuse the patient, and does not pretend the habit is irrelevant either. The first consultation records it, the prescription depends on whether the patient has a believable plan to stop, and the insurer wants to know. The ESG dialogue with a cement producer has the same three parts: record, assess the plan, report.
A sector policy is an internal rule, usually published, that sets the conditions under which the bank serves a sensitive sector: thermal coal, oil and gas, mining, palm oil, defence, tobacco, and increasingly heavy industry. It contains exclusions (activities or companies the bank does not finance), conditions (criteria the client must meet, sometimes by a date) and a monitoring regime. Exclusion lists name the companies that fail; a hit blocks onboarding or any new transaction in the bank’s systems.
The screen produces one of three outcomes: compliant; compliant with conditions or under watch (enhanced due diligence, action plan agreed with the client); non-compliant (no new business, managed exit). Difficult cases, such as a serious controversy, a policy criterion open to interpretation or a transaction with high public visibility, are escalated. The usual chain is the ESG risk team, then compliance and the regional head, then a reputational risk committee at group level, which can approve, approve with conditions or refuse. Its decisions are recorded and bind the business.
WHO DOES WHAT — The banker fills in the ESG questionnaire with the client and owns the answers. The ESG risk analyst (first line, attached to the business) checks them against public information and data providers and writes the ESG assessment memo. Compliance and the risk function (second line) give an independent opinion when the case is escalated. The reputational risk committee decides. Internal audit (third line) reviews samples of files afterwards.
A transition plan is the part of a company’s strategy that sets out its emission targets, the actions and investments to reach them, and the governance that holds management to them. Several frameworks describe what it should contain, and in November 2026 they have largely converged.
| Framework | Status in November 2026 | Use for the banker |
|---|---|---|
| ESRS E1 (EU) | Revised standards adopted by the Commission on 3 July 2026 | What EU clients disclose |
| IFRS S2 and TPT materials | ISSB standard; TPT framework now with the IFRS Foundation | Disclosure outside the EU |
| SBTi Net-Zero Standard V2 | Published 11 June 2026 | Target validation |
| TPI, ACT, CA100+ benchmark | Public assessments of large emitters | Third-party cross-check |
| LMA Guide to Transition Loans | 16 October 2025, with exposure-draft principles | Entity-level test for loans |
The European Sustainability Reporting Standards require a company that has a transition plan for climate change mitigation to describe it (targets, decarbonisation levers, investments, locked-in emissions); a company without one must say so. The Omnibus I Directive (EU) 2026/470, published on 26 February 2026, kept this disclosure but narrowed its scope to companies with more than 1,000 employees and more than EUR 450m of turnover, and removed from the due diligence directive (CSDDD) the obligation to adopt and put into effect a climate transition plan. For a banker this means fewer clients with standardised data, and a plan that is disclosed where it exists but that no EU text obliges the client to carry out.
Version 2 of the SBTi Corporate Net-Zero Standard requires every company seeking validation to maintain a transition plan, with public disclosure for the larger ones, and replaces the single validation by an assessment at the end of each target cycle. The LMA, APLMA and LSTA Guide to Transition Loans asks the borrower of a transition-labelled loan to demonstrate a credible entity-level transition strategy, either a plan or a set of indicators tied to recognised pathways (public summaries; the text is discussed in A04).
From these sources, banks have built internal assessment grids. They differ in detail and agree on five questions:
PITFALL — A net-zero date of 2050 says nothing about credit risk over the life of a seven-year loan. The banker’s horizon is the next five to ten years: the 2030 target, the capex approved for it and the carbon price the client will actually pay in between. Assess the plan where it overlaps with the bank’s exposure.
On the bank. The EBA Guidelines on the management of ESG risks (EBA/GL/2025/01, published on 9 January 2025) have applied to large institutions since 11 January 2026 and apply to small and non-complex institutions from 11 January 2027 at the latest. They ask banks to identify and measure ESG risks at counterparty level, to use engagement with counterparties as a risk-management tool, and to prepare the prudential plan required by the banking directive (CRD6, Directive (EU) 2024/1619, Article 76(2)), with quantitative targets to address the risks arising from the transition. The EBA Guidelines on loan origination and monitoring (EBA/GL/2020/06) already required ESG factors in credit decisions. The ECB’s supervisory priorities for 2026 to 2028 keep the management of climate and nature-related risks among the areas where it expects persistent weaknesses to be remedied. None of these texts tells a bank to stop serving cement producers. They do oblige it to know what the transition does to each large exposure, and to be able to show the supervisor a file. That file is written in coverage.
On the client. For an EU cement producer the binding constraint is the carbon price. Cement is covered by the EU Emissions Trading System and by the Carbon Border Adjustment Mechanism (CBAM), whose definitive period began on 1 January 2026. Free allowances for CBAM sectors are being withdrawn: under the law in force they are multiplied by a factor of 97.5% in 2026, 95% in 2027, 90% in 2028, 77.5% in 2029 and 51.5% in 2030, then fall to zero in 2034. The product benchmark for grey cement clinker for 2026 to 2030 has been reported at 0.656 allowance per tonne (Implementing Regulation (EU) 2026/1412). A Commission proposal of July 2026 to review the ETS would slow the schedule after 2027 and end it in 2038; in November 2026 it is still under negotiation and is treated in this module as a sensitivity only (see C10 for the market).
Taxonomy. The EU Taxonomy treats cement manufacturing as a transitional activity (activity 3.7 of the Climate Delegated Act, Delegated Regulation (EU) 2021/2139). It contributes substantially to mitigation when specific emissions are below 0.722 tCO2e per tonne of grey clinker or 0.469 tCO2e per tonne of cement. The second figure is far below the sector’s current average of roughly 0.6 tonne, so in practice a plant qualifies only after deep clinker substitution or carbon capture.
The Net-Zero Banking Alliance, under which many banks had published sector targets including cement, ceased operations after a member vote in early October 2025; its target-setting guidance stays publicly available, rewritten as guidance for the banking industry in general. European banks have kept their sector targets, now anchored in their CSRD reports and in the prudential plans required by CRD6, with less reference to collective commitments. Client transition assessments are standard for the largest emitters in the portfolios of the main European banks, with methods that are proprietary and little disclosed, which is one of the weaknesses discussed in section 8. The labelled-product toolkit for hard-to-abate clients has widened with the transition-loan guide of October 2025 and ICMA’s Climate Transition Bond Guidelines of November 2025 (see A04, C02).
Coverage has no single deal. Its lifecycle is the annual account-planning cycle, from which deals are born. The flow below follows Meridian’s cycle for a client of the top tier in a high-emitting sector; the case study in section 5 walks through it with numbers.
INTERACTIVE LAB — The annual account-planning cycle for a hard-to-abate client
Follow the sequence from the relationship P&L to the approved plan and the first mandates.
| # | Who | Does what | Deliverable |
|---|---|---|---|
| 1 | Business manager, coverage | Extracts the three-year relationship P&L, exposures and capital from the management information system | Relationship P&L pack |
| 2 | Senior banker and analyst | Estimate the client’s wallet by product and Meridian’s share; compute the relationship RAROC | Wallet table, RAROC sheet |
| 3 | KYC analyst | Runs the periodic KYC review: ownership, sanctions, adverse media; sends the ESG questionnaire with the banker | Updated KYC file |
| 4 | ESG risk analyst (first line) | Screens the client against the sector policy and exclusion list; checks controversies | ESG assessment memo with policy outcome |
| 5 | Compliance and reputational risk committee | Review the escalated controversy; decide with conditions | Committee minutes, conditions |
| 6 | Sustainable finance adviser (sector) | Assesses the transition plan on targets, levers, capex, governance and track record | Transition scorecard and class |
| 7 | Credit analyst | Performs the annual review; integrates carbon-cost scenarios in the rating | Rating confirmation, credit review |
| 8 | Senior banker and product partners | Hold the pipeline workshop: loans, DCM, hedging, carbon, cash management | Opportunity list with amounts and probabilities |
| 9 | Senior banker | Writes the account plan; simulates RAROC with and without the new exposure | Draft account plan |
| 10 | Client committee (head of coverage, product heads, risk) | Challenges the plan; applies the ESG steering rules; sets ancillary targets | Approved plan and conditions |
| 11 | Senior banker and sector head | Hold the strategic dialogue meeting with the client’s CFO, treasurer and CSO | Call report, mandates to pursue |
| 12 | Credit committee | Approves the new limits with tenor and structure conditions | Credit decision |
| 13 | Deal teams, then senior banker | Execute the transactions; the banker tracks the plan quarterly in the CRM | Signed deals, quarterly tracking |
Steps 1 and 2: the numbers first. The cycle starts in the autumn, after the client’s half-year results. The business manager’s pack shows revenue by product and by booking entity, the average and year-end exposures, the risk-weighted assets and the expected loss. The banker’s analyst adds what the systems do not know: the wallet. The control at this stage is reconciliation: the revenues in the pack must match the general ledger, and the exposures must match the credit system. A plan built on a wrong relationship P&L is discovered at the committee, in public.
Steps 3 to 5: can we, and under what conditions. The KYC review and the ESG screen run in parallel because they use the same client contacts and many of the same sources. The ESG questionnaire is sector-specific: for cement it asks for emissions by scope, clinker ratio, fuel mix, targets, exposure to carbon pricing, quarry permits and biodiversity management, health and safety records, and pending litigation. The ESG risk analyst compares the answers with the client’s sustainability statement, with data providers and with a controversy database. The memo ends with the policy outcome. Any escalation must be closed before new business is proposed: the committee’s minutes are attached to the account plan.
Step 6: how good is the plan. The sustainable finance adviser is a sector specialist, often an engineer by training, placed in the first line next to coverage. The scorecard is filled in from public documents first, then completed in a working session with the client’s sustainability and treasury teams. The second-line ESG risk function validates the method and reviews a sample of scorecards.
Step 7: the credit view. The credit analyst owns the rating. The transition assessment enters through the financial projections (carbon cost, capex, leverage) and, at Meridian, through a qualitative overlay that can move the internal rating by one notch. The analyst and the adviser must use the same carbon-price and pass-through assumptions; a classic control finding is two teams of the same bank holding two different views of the same client.
Steps 8 to 10: from assessment to plan. The pipeline workshop is where the assessment becomes business. Each product partner states what it could do, for how much and with what probability; the banker arbitrates. The draft plan then goes to the client committee. This is the real decision point of coverage: the committee can change the tier, cap the balance sheet, and set the ancillary revenue the banker must deliver. ESG steering rules are applied here, before any credit request is written.
Steps 11 to 13: back to the client. The strategic dialogue meeting is prepared like a pitch: a short document that shows the client its own carbon-cost trajectory as the bank sees it, the financing options for the capex, and what the bank is ready to commit. The credit request follows the mandate. After signing, the plan is a tracking tool: every quarter the banker records which opportunities were won, lost or delayed, and the relationship RAROC is recomputed.
Question. What does Castellan pay its banks, and how much of it reaches Meridian?
Data. Castellan Cement is a fictional listed European cement group. All figures are case assumptions for the financial year 2025 unless stated otherwise.
| Item | Value |
|---|---|
| Revenue; EBITDA | EUR 4,200m; EUR 900m |
| Net debt; leverage | EUR 2,160m; 2.4x |
| Enterprise value including cash (EVIC) | EUR 6,000m |
| Cementitious production, all in the EU | 20 Mt |
| Clinker ratio; clinker output | 74%; 14.8 Mt |
| Scope 1 intensity (2020: 655) | 590 kg CO2 per tonne |
| Scope 1; scope 2 emissions | 11.8 Mt; 0.9 Mt |
| Internal rating at Meridian | BBB (PD 0.20%) |
It is November 2026. Ilse Vandermeer, senior banker in Meridian’s industrials and building materials team, prepares the account plan for the client committee of 3 December. Castellan’s banking group has twelve banks. Its committed bank facilities total EUR 2,500m: a syndicated revolving credit facility (RCF) of EUR 1,500m maturing in 2028, term loans of EUR 600m and guarantee lines of EUR 400m. Meridian holds EUR 250m of them:
| Meridian exposure | Amount | Price paid by Castellan |
|---|---|---|
| RCF hold, 10% drawn on average | EUR 150m | Margin 80 bp; commitment fee 28 bp |
| RCF upfront fee, spread over 5 years | 4 bp a year on the hold | |
| Term loan, drawn | EUR 60m | Margin 110 bp |
| Performance guarantees outstanding | EUR 40m | Fee 60 bp |
On top of the credit products, Meridian earned EUR 1.15m of ancillary revenue with Castellan in the last twelve months: cash management in one country (EUR 0.50m), foreign exchange and interest-rate hedging (EUR 0.45m) and a passive bookrunner role on a bond (EUR 0.20m). Ilse’s team estimates Castellan’s wallet at EUR 36m a year: EUR 12m for credit products and EUR 24m for everything else (capital markets, hedging including carbon allowances, cash management and trade, advisory).
Calculation. What Castellan pays Meridian on credit products: , that is 0.120 + 0.378 + 0.060 + 0.660 + 0.240 = EUR 1.458m. With the ancillary revenue the total is EUR 2.608m.
Result. Meridian’s share of wallet is 7.2%. It holds 10.0% of the committed facilities and collects 12.2% of the credit wallet, and only 4.8% of the non-credit wallet.
INTERACTIVE LAB — Castellan’s wallet and Meridian’s share of it
Compare Meridian’s share of the client’s lending with its share of the fees and margins the client pays to all banks.
| Parameter you can move | Range | Default |
|---|---|---|
| Meridian RCF hold | 50 to 300 EUR m | 150 EUR m |
| Average RCF utilisation | 0 to 100 % | 10 % |
| Term loan drawn | 0 to 200 EUR m | 60 EUR m |
| Guarantees outstanding | 0 to 100 EUR m | 40 EUR m |
| Meridian ancillary revenue | 0 to 6 EUR m | 1.15 EUR m |
| Client credit wallet | 5 to 30 EUR m | 12 EUR m |
| Client non-credit wallet | 5 to 60 EUR m | 24 EUR m |
| Client committed bank facilities | 1000 to 5000 EUR m | 2500 EUR m |
| Result at default settings | Value |
|---|---|
| Paid to Meridian on credit products | 1.458 EUR m |
| Total paid to Meridian | 2.608 EUR m |
| Share of wallet | 7.2 % |
| Share of committed facilities | 10.0 % |
| Share of credit wallet | 12.2 % |
| Share of non-credit wallet | 4.8 % |
Try it. Raise the ancillary revenue until the non-credit share equals the 10% lending share: that is what reciprocity would mean in euros.
Reading. Meridian lends like a core bank and is paid like a second-tier one. To be treated as it lends, it should earn 10% of the non-credit wallet, EUR 2.4m, against EUR 1.15m today. The wallet estimate is the weakest number of the plan: an error of a quarter on EUR 24m moves the share by more than a point. Ilse states it as a range in the document and uses the central value in the arithmetic.
Question. Is the capital committed to Castellan paid at the hurdle?
Data. The desk pays Treasury a liquidity premium of 30 bp on drawn five-year amounts and a contingent liquidity charge of 4 bp on undrawn amounts (10% outflow rate times the 40 bp carry of the liquidity buffer, see section 6). Guarantee fees are treated as lending revenue. Credit conversion factors under CRR3: 40% on the undrawn RCF, 50% on performance guarantees. Other parameters are those of Meridian: F-IRB, LGD 40%, capital at 13% of RWA, tax 25%.
Calculation.
| Line | Computation | EUR m |
|---|---|---|
| RCF drawn, net of FTP | 15 x (80 - 30) bp | 0.075 |
| RCF undrawn, net of liquidity charge | 135 x (28 - 4) bp | 0.324 |
| RCF upfront fee | 150 x 4 bp | 0.060 |
| Term loan, net of FTP | 60 x (110 - 30) bp | 0.480 |
| Guarantee fees | 40 x 60 bp | 0.240 |
| Lending revenue | 1.179 | |
| Ancillary revenue | 1.150 | |
| Operating costs | 40% x 1.179 + 60% x 1.150 | -1.162 |
| Expected loss | 0.20% x 40% x 149 | -0.119 |
| Pre-tax result | 1.048 |
Exposure at default is . At a risk weight of 39.02% the credit RWA are EUR 58.14m; the ancillary revenue adds . Total RWA are EUR 61.01m and allocated capital EUR 7.93m.
Result. The net result is EUR 0.786m and the relationship RAROC 9.91%. To reach 12% with the same exposures, ancillary revenue should be EUR 1.78m, EUR 0.63m more than today.
INTERACTIVE LAB — Relationship RAROC of Castellan before the plan
Rebuild the relationship P&L, the capital and the RAROC from the exposures, and find the ancillary revenue that would bring it to the hurdle.
| Parameter you can move | Range | Default |
|---|---|---|
| RCF hold | 50 to 300 EUR m | 150 EUR m |
| RCF utilisation | 0 to 100 % | 10 % |
| Term loan drawn | 0 to 200 EUR m | 60 EUR m |
| Guarantees outstanding | 0 to 100 EUR m | 40 EUR m |
| Ancillary revenue | 0 to 6 EUR m | 1.15 EUR m |
| One-year PD | 0.07 / 0.2 / 1 / 4 | 0.2 % |
| Result at default settings | Value |
|---|---|
| Lending revenue net of FTP | 1.179 EUR m |
| Operating costs | 1.162 EUR m |
| Exposure at default | 149.0 EUR m |
| Expected loss | 0.119 EUR m |
| Pre-tax result | 1.048 EUR m |
| Net result | 0.786 EUR m |
| RWA | 61.01 EUR m |
| Allocated capital | 7.93 EUR m |
| Relationship RAROC | 9.91 % |
| Ancillary revenue needed for 12% | 1.78 EUR m |
| Gap to current ancillary revenue | 0.63 EUR m |
Try it. Change the PD to 1.00%: the same relationship returns 2.63%. Keep this number in mind for stage 5.9.
Reading. At 9.91% the relationship is close to the hurdle and below it. Castellan would not be in danger of losing its tier on these numbers. The plan’s financial objective is clear and modest: EUR 0.63m of additional fee business. The difficulty is that Castellan’s next request will be for more lending.
Question. Can Meridian do new business with Castellan, and on whose authority?
Data. Meridian’s cement sector policy (fictional, modelled on the structure of published bank policies) has four criteria:
| Criterion | Requirement | Castellan |
|---|---|---|
| Disclosure | Scope 1 and 2 emissions published and assured | Met: limited assurance |
| Target | Public 2030 scope 1 intensity target | Met: 480 kg per tonne |
| Plan | Transition scorecard class above “Insufficient” | Pending (stage 5.7) |
| Controversies | No unresolved severe controversy | One open case |
The open case: a regional environmental authority has fined a Castellan plant for repeated dust-emission exceedances, and a local association has filed a complaint about the extension of the adjoining quarry near a protected wetland. The controversy database used by the ESG risk team rates the case 3 on a scale of 5 (“significant”). Meridian’s policy escalates any case rated 3 or above when new financing is envisaged.
Process. The ESG risk analyst asks Ilse to obtain Castellan’s position in writing. Castellan provides the remediation plan for the filters (EUR 14m, completion mid-2027), the status of the appeal and the biodiversity assessment of the quarry extension. Compliance gives a second-line opinion: the case is real, local, and being remedied; the wetland question is open. The reputational risk committee meets on 19 November 2026.
Result. The committee approves the continuation of the relationship and new financing with three conditions: no financing dedicated to the quarry extension until the permit is confirmed; a six-monthly update on the remediation from the banker; re-escalation if the database rating rises to 4.
Reading. Nothing here is about carbon, which is typical: reputational escalations in hard-to-abate sectors come more often from local pollution, biodiversity, safety or human-rights cases than from emissions, which are known and priced. The committee’s conditions enter the account plan as constraints on the use of proceeds of any new loan. The policy’s third criterion, the transition plan, still has to be assessed, and before scoring the plan Ilse wants to know what is at stake financially.
Question. How much will carbon cost Castellan in 2030, with and without its plan, and what does that do to the credit profile?
IN ONE SENTENCE — A cement producer buys on the market the allowances it emits beyond its free allocation, and the free allocation is shrinking to zero by 2034.
FORMULA
is the gross carbon cost of year ; the allowance price; the scope 1 intensity in tonnes of CO2 per tonne of cementitious product; the production (20 Mt); the clinker benchmark (0.656 allowance per tonne of clinker); the clinker output on which the allocation is based (14.8 Mt); the CBAM factor of the year (97.5% in 2026, 51.5% in 2030, 0% in 2034).
Data. Allowance price EUR 85 per tonne, close to the auction prices of October 2026, held flat. Without the plan, intensity stays at 590 kg. With the plan it falls in a straight line to 480 kg in 2030 and is held there. The allocation is kept on the historical clinker output on both paths, because allocations are only adjusted when activity moves by more than 15% and the plan cuts clinker output by 8% (a simplification). The plan costs EUR 900m of capex over 2026 to 2030, of which EUR 180m is covered by public grants, and EUR 45m a year of additional operating cost from 2030 for the transport and storage of captured CO2. Selling prices are assumed to rise by 75% of the carbon-cost increase of an unabated producer, because the least efficient plants set the market price and CBAM applies the same cost to imports. The EBITDA of EUR 900m is stated at the 2026 carbon cost of the unchanged plant.
Calculation. Free allocation at full benchmark: Mt.
Result. By 2030 the plan saves 2.2 Mt of allowances a year, EUR 187.0m at EUR 85. Without the plan, the cost increase of EUR 379.6m is passed on at 75%, and EBITDA falls to EUR 805.1m; leverage on unchanged debt rises to 2.68x. With the plan, Castellan receives the same price increase, bears a smaller cost increase and pays the storage cost: EBITDA is EUR 947.1m, but net debt has risen by the EUR 720m of capex net of grants to EUR 2,880m, and leverage is 3.04x. The net capex is paid back in 5.1 years by the annual gain of EUR 142.0m. By 2034 the producer that did nothing has an EBITDA of EUR 698.9m and a leverage of 3.09x.
INTERACTIVE LAB — Carbon cost of Castellan under the free-allocation phase-out
Project the gross carbon cost with and without the transition plan and read the effect on EBITDA and leverage in 2030.
| Parameter you can move | Range | Default |
|---|---|---|
| Allowance price | 40 to 160 EUR/t | 85 EUR/t |
| Pass-through of the unabated cost increase to prices | 0 to 100 % | 75 % |
| 2030 intensity with the plan | 400 to 590 kg/t | 480 kg/t |
| Clinker benchmark | 0.6 to 0.7 t/t | 0.656 t/t |
| Plan capex net of grants | 300 to 1500 EUR m | 720 EUR m |
| CO2 transport and storage cost from 2030 | 0 to 120 EUR m | 45 EUR m |
| Result at default settings | Value |
|---|---|
| Free allocation before CBAM factor | 9.709 Mt |
| 2026 cost, no plan | 198.4 EUR m |
| 2030 cost, no plan | 578.0 EUR m |
| 2030 cost, with plan | 391.0 EUR m |
| 2030 allowance saving from the plan | 187.0 EUR m |
| 2034 cost, no plan | 1,003.0 EUR m |
| 2030 cost without plan as a share of today’s EBITDA | 64.2 % |
| 2030 EBITDA, no plan | 805.1 EUR m |
| 2030 EBITDA, with plan | 947.1 EUR m |
| 2030 leverage, no plan | 2.68 x |
| 2030 leverage, with plan | 3.04 x |
| Annual EBITDA gain of the plan from 2030 | 142.0 EUR m |
| Payback of net capex | 5.1 years |
| 2034 EBITDA, no plan | 698.8 EUR m |
| 2034 leverage, no plan | 3.09 x |
Try it. Lower the allowance price to EUR 60: the payback lengthens from 5.1 to 8.3 years. The plan’s economics are a bet on the carbon price that the bank shares once it lends for seven years.
PITFALL — “Transition is good for credit” is wrong over the life of the loan Castellan will ask for. In 2030 the company that invests is more leveraged (3.04x) than the one that does not (2.68x). The gain comes afterwards, and only if the allowance price holds and prices pass through. A lender to a transition is taking a view on carbon policy; it should say so in the credit application.
Reading. Three messages go into the account plan. Castellan’s exposure is large: the 2030 cost without the plan is 64.2% of today’s EBITDA before pass-through. The plan is economically rational at current prices, which is the strongest evidence of credibility there is. And the ETS review under negotiation would soften the schedule after 2027: good for Castellan’s cash flows, bad for the payback of its capture investment.
Question. Is the 2030 target ambitious, and where do the tonnes come from?
Data. Castellan’s target is 480 kg in 2030, validated by SBTi under its cement guidance. Meridian uses as its 2030 reference point 469 kg, the Taxonomy threshold for cement, a design choice of its sector policy. Intensity was 655 kg in 2020 and 590 kg in 2025.
Calculation, targets. The target covers of the distance to the reference. The historical pace is a year; the required pace is a year, 1.95 times faster.
INTERACTIVE LAB — Target ambition and required pace
Compare the client’s 2030 target with a reference point and its past pace of reduction with the pace it now needs.
| Parameter you can move | Range | Default |
|---|---|---|
| Intensity 2020 | 550 to 800 kg/t | 655 kg/t |
| Intensity 2025 | 500 to 750 kg/t | 590 kg/t |
| 2030 target | 380 to 600 kg/t | 480 kg/t |
| Reference point 2030 | 380 to 560 kg/t | 469 kg/t |
| Result at default settings | Value |
|---|---|
| Share of the distance to the reference covered by the target | 90.9 % |
| Historical annual change 2020-2025 | -2.07 % |
| Required annual change 2025-2030 | -4.04 % |
| Required pace as a multiple of historical pace | 1.95 x |
| 2030 intensity at the historical pace | 531.5 kg/t |
Try it. Read the dashed line in 2030: at the historical pace Castellan would reach 531.5 kg, not 480. The gap is what the new levers must deliver.
Data, levers. Cement emissions have two sources. About two thirds are process emissions: heating limestone releases CO2 whatever the fuel, about 525 kg per tonne of clinker. The rest comes from the kiln fuel. Castellan’s clinker therefore emits kg per tonne, of which 272 kg from fuel. The plan has three levers:
Calculation, levers. Lever 1: kg, a cut of 47.8 kg. Lever 2: fuel emissions fall to 217.8 kg and clinker intensity to 742.8 kg; cement intensity is kg, a further cut of 37.0 kg. Lever 3: kg. The plan lands at 480.1 kg.
INTERACTIVE LAB — Where Castellan’s tonnes come from: the lever waterfall
Decompose the reduction from 590 to 480 kg per tonne into clinker substitution, fuel switch and carbon capture.
| Parameter you can move | Range | Default |
|---|---|---|
| Clinker ratio in 2030 | 55 to 74 % | 68 % |
| Cut in fuel emissions per tonne of clinker | 0 to 60 % | 20 % |
| CO2 captured and stored | 0 to 3 Mt | 0.5 Mt |
| Process emissions per tonne of clinker | 500 to 550 kg | 525 kg |
| Result at default settings | Value |
|---|---|
| Clinker intensity today | 797.3 kg/t clinker |
| Lever 1: clinker ratio | -47.8 kg/t |
| Lever 2: fuel | -37.0 kg/t |
| Lever 3: capture | -25.0 kg/t |
| 2030 intensity | 480.1 kg/t |
| Share of the reduction that depends on capture | 22.8 % |
| Intensity floor without capture (process emissions only) | 357.0 kg/t |
Try it. Set capture to zero and push the two other levers to their limits: the intensity cannot fall below the process emissions of the clinker that remains. Deep cuts in cement need capture.
Reading. The target is close to the reference (90.9%) and requires almost twice the historical pace. Three quarters of the reduction comes from proven levers. They are not free of risk: clinker substitution depends on product standards and on the availability of substitutes (slag and fly ash become scarcer as blast furnaces and coal plants close), biomass waste is contested between sectors. Capture carries 22.7% of the reduction and almost all of the execution risk: one unit, a storage chain that Castellan does not control, a start date at the very end of the target period. The first industrial-scale capture unit on a cement plant, at Brevik in Norway, was inaugurated in June 2025 with a capacity of about 0.4 Mt a year, so the technology exists; a 0.5 Mt unit by 2030 is credible only with a final investment decision in 2027.
Question. Does the capex budget match the levers, and does the plan make economic sense?
Data. Castellan’s 2026 to 2030 capex plan totals EUR 2,300m, of which EUR 900m is for decarbonisation: EUR 450m for the capture unit and EUR 450m for the other levers. The board has approved EUR 540m (all the conventional levers and EUR 90m of engineering for the capture unit); the remaining EUR 360m awaits the final investment decision. Grants of EUR 180m have been awarded for the capture unit. We annualise the net capex over 20 years at a cost of capital of 6%.
Calculation. Decarbonisation is of capex, and of it is approved. The annuity factor for 20 years at 6% is 11.47, so the net capex of EUR 720m costs EUR 62.8m a year; with EUR 45m of operating cost, the plan costs EUR 107.8m a year for 2.2 Mt abated.
Result. The abatement cost is EUR 49.0 per tonne, against an allowance price of EUR 85. Without the grants it would be EUR 56.1.
INTERACTIVE LAB — Capex credibility and abatement cost of the plan
Check the share of decarbonisation capex that is approved and compare the annualised cost per tonne abated with the carbon price.
| Parameter you can move | Range | Default |
|---|---|---|
| Decarbonisation capex | 300 to 2000 EUR m | 900 EUR m |
| Of which approved by the board | 0 to 2000 EUR m | 540 EUR m |
| Public grants | 0 to 600 EUR m | 180 EUR m |
| Total capex plan | 1000 to 5000 EUR m | 2300 EUR m |
| Additional operating cost | 0 to 150 EUR m | 45 EUR m |
| Cost of capital | 3 to 12 % | 6 % |
| Annual abatement in 2030 | 0.5 to 5 Mt | 2.2 Mt |
| Result at default settings | Value |
|---|---|
| Decarbonisation share of capex | 39.1 % |
| Approved share of decarbonisation capex | 60.0 % |
| Annualised net capex | 62.8 EUR m |
| Annual cost of the plan | 107.8 EUR m |
| Abatement cost | 49.0 EUR/t |
| Abatement cost without grants | 56.1 EUR/t |
Try it. Raise the cost of capital to 10%: the abatement cost rises to EUR 58.9, still under the allowance price. The plan survives a higher financing cost; it would not survive an allowance price of EUR 45.
Reading. The plan passes the test that matters most to a lender: it pays for itself at the current carbon price, so management has a reason to carry it out that does not depend on goodwill. The weak point is the unapproved EUR 360m, which is exactly the capture lever. The plan as approved by the board delivers 505 kg; the plan as announced delivers 480 kg.
Question. How does Meridian summarise the assessment in one class that the committees can use?
Data. Meridian’s transition scorecard (fictional) scores five criteria from 0 to 4 and weights them. The class is “Credible” from 75%, “Developing” from 50% to 75%, “Insufficient” below 50%. The sustainable finance adviser, Tomasz Wrobel, scores Castellan as follows.
| Criterion | Weight | Score | Reason |
|---|---|---|---|
| Targets | 20% | 3 | Validated, 90.9% of reference; no interim 2028 milestone |
| Levers | 25% | 3 | Quantified and mostly proven; 22.7% on capture |
| Capex and financing | 25% | 2 | 60% approved; capture decision pending |
| Governance | 15% | 3 | Board committee; 15% of bonus on intensity |
| Track record and disclosure | 15% | 2 | Required pace 1.95 times the past; limited assurance |
Result. The weighted score is out of 4, or 65.0%. Castellan is “Developing”.
INTERACTIVE LAB — Transition-plan credibility scorecard
Turn five scores into a weighted credibility score and a class, and see which criterion would change the class.
| Parameter you can move | Range | Default |
|---|---|---|
| Targets (weight 20%) | 0 / 1 / 2 / 3 / 4 | 3 |
| Levers (weight 25%) | 0 / 1 / 2 / 3 / 4 | 3 |
| Capex and financing (weight 25%) | 0 / 1 / 2 / 3 / 4 | 2 |
| Governance (weight 15%) | 0 / 1 / 2 / 3 / 4 | 3 |
| Track record and disclosure (weight 15%) | 0 / 1 / 2 / 3 / 4 | 2 |
| Result at default settings | Value |
|---|---|
| Weighted score out of 4 | 2.60 |
| Credibility score | 65.0 % |
| Class (0 Insufficient, 1 Developing, 2 Credible) | 1 |
| Points missing to reach Credible | 0.40 |
Try it. Move ‘Capex and financing’ from 2 to 4, as a final investment decision on the capture unit with financing in place would justify: the class becomes Credible.
Reading. The score is a summary for governance; the judgement sits in the reasons column. Its value for the banker is that it names what would change the class: the investment decision on the capture unit. That happens to be the event that creates the financing need, which gives Ilse the theme of her client meeting. The method’s limits should be stated in the plan: the weights are conventions, the scores are one adviser’s judgement reviewed by a second person, and a different bank using another grid could reach another class from the same documents.
Question. What business does Castellan’s transition create for Meridian, and does the plan reach the hurdle?
Data. The pipeline workshop of 24 November lists five opportunities for 2027.
| Opportunity (desk, module) | Meridian revenue a year | Probability |
|---|---|---|
| Transition capex term loan: EUR 400m club deal, 7 years, margin 130 bp, upfront 60 bp; draft hold EUR 80m (B02, B07) | EUR 0.773m, lending | 60% |
| RCF refinancing as sustainability-linked, coordinator role: fee EUR 0.30m over 5 years (B03, B04) | EUR 0.06m | 80% |
| Active bookrunner on a EUR 600m bond every two years: 0.40% fee shared by four banks (C01, C02) | EUR 0.30m | 50% |
| Allowance purchases, rate and currency hedging (C07, C10) | EUR 0.60m | 50% |
| Cash management in a second country (B14) | EUR 0.35m | 40% |
The seven-year loan pays Treasury a liquidity premium of 42 bp, so its net margin is 88 bp: , plus of upfront fee. On its own, the loan returns 7.38%: less than the relationship does today.
Calculation. Two versions of the plan are computed: the expected case, where each revenue and the loan’s exposure are weighted by probability, and the full-success case.
Result. The draft plan returns 10.95% in the expected case and 11.91% if everything is won. It does not reach the hurdle even on full success. The bookrunner line was, moreover, optimistic: when Castellan executed its sustainability-linked bond, the deal priced with a gross fee of 20 bp shared by five bookrunners, half the fee rate of the plan’s 0.40% and one more bank than the four assumed (Meridian’s actual share and structuring fee are worked out in C02).
INTERACTIVE LAB — Account plan: pipeline, probabilities and relationship RAROC
Weight the opportunities by probability and compute the RAROC of the relationship after the plan, in the expected and in the full-success case.
| Parameter you can move | Range | Default |
|---|---|---|
| Hold in the transition term loan | 0 to 150 EUR m | 80 EUR m |
| Term loan margin | 80 to 250 bp | 130 bp |
| Probability: term loan | 0 to 100 % | 60 % |
| Probability: coordinator role | 0 to 100 % | 80 % |
| Probability: bookrunner | 0 to 100 % | 50 % |
| Probability: hedging | 0 to 100 % | 50 % |
| Probability: cash management | 0 to 100 % | 40 % |
| One-year PD | 0.2 / 1 | 0.2 % |
| Result at default settings | Value |
|---|---|
| Term loan revenue if won | 0.773 EUR m |
| Stand-alone RAROC of the term loan | 7.38 % |
| Expected lending revenue | 1.643 EUR m |
| Expected ancillary revenue | 1.788 EUR m |
| Expected EAD | 197.0 EUR m |
| RAROC, expected case | 10.95 % |
| Lending revenue, full success | 1.952 EUR m |
| Ancillary revenue, full success | 2.46 EUR m |
| RAROC, full success | 11.91 % |
Try it. Reduce the hold from EUR 80m to EUR 60m: the full-success RAROC passes the hurdle. A smaller ticket with the same roles is the banker’s best lever.
Reading. The transition creates a real pipeline: the full-success ancillary revenue of EUR 2.46m would give Meridian 10.25% of the non-credit wallet, in line with its lending share at last. But the loan that opens the door is itself dilutive, at 7.38%. Every additional euro lent must be carried by fee business, and the draft plan asks for too much balance sheet for the fees it can credibly claim. Ilse knows the committee will say so.
Question. What do Meridian’s steering rules do to the draft, and what is the plan that goes to the client?
Data. Meridian’s ESG steering for high-emitting sectors (fictional rules) has three elements.
Calculation. Growth cap: ; the EUR 80m ticket breaches it, and Ilse reduces the hold to EUR 60m. Sector intensity today: kg; after the loan, with EUR 135m drawn on Castellan, 624.4 kg; in 2030, if Castellan delivers 480 kg and nothing else changes, 608.9 kg. Financed emissions today: of 12.7 Mt, 158,750 tCO2e; after the loan, 2.25%, or 285,750 tCO2e.
Result. With a EUR 60m hold, the plan returns 11.16% in the expected case and 12.31% on full success. If the plan failed and Castellan were downgraded to BB, the same relationship would return 3.91%.
INTERACTIVE LAB — ESG steering: growth cap, sector intensity and financed emissions
Apply the growth cap of the transition class, then measure what the new loan does to the cement book’s intensity and to financed emissions.
| Parameter you can move | Range | Default |
|---|---|---|
| Transition class (0 Insufficient, 1 Developing, 2 Credible) | 0 / 1 / 2 | 1 |
| Requested hold in the new loan | 0 to 150 EUR m | 80 EUR m |
| Current committed exposure | 100 to 500 EUR m | 250 EUR m |
| Current drawn exposure | 0 to 300 EUR m | 75 EUR m |
| Castellan intensity in 2030 | 400 to 590 kg/t | 480 kg/t |
| Rest of the cement book, drawn | 200 to 2000 EUR m | 825 EUR m |
| Intensity of the rest of the book | 500 to 750 kg/t | 630 kg/t |
| Result at default settings | Value |
|---|---|
| Maximum increase allowed | 62.5 EUR m |
| Hold retained (rounded down to EUR 5m) | 60 EUR m |
| Book intensity today | 626.7 kg/t |
| Book intensity after the loan | 624.4 kg/t |
| Book intensity in 2030 if the client delivers | 608.9 kg/t |
| Financed emissions today | 158,750 tCO2e |
| Financed emissions after the loan | 285,750 tCO2e |
| Financed emissions in 2030 if the client delivers | 231,750 tCO2e |
| Rise in financed emissions at drawdown | 80 % |
| Financed emissions in 2030 against today | 46 % |
Try it. Set the class to Credible: the cap disappears and the EUR 80m ticket passes. The class is worth EUR 20m of balance sheet to this client.
PITFALL — Financing a transition raises the bank’s financed emissions today. Meridian’s attributed emissions from Castellan rise by 80% the day the loan is drawn, and are still 46% above today’s level in 2030 even if the client delivers (231,750 tCO2e against 158,750). A bank steered only on absolute financed emissions would refuse the loan that pays for the capture unit. This is why sector steering in cement uses physical intensity, and why the two metrics must be read together.
The plan presented on 3 December. Ilse’s final account plan contains five decisions.
Reading. ESG steering changed three things in the plan: the size of the ticket (60 instead of 80), the structure of the new money (ring-fenced, with conditions tied to the weakest part of the scorecard), and the calendar of the relationship (a review tied to the client’s own investment decision). It did not change the price: Meridian grants no discount on its hurdle for a transition asset, and the 130 bp margin is set by the club of lenders. The downgrade scenario shows where the money is. One notch of plan failure costs the relationship more than eight points of RAROC; the entire pipeline adds little more than two.
Question. What does Ilse say to Castellan, and what does she ask for?
The strategic dialogue meeting takes place on 15 December 2026 at Castellan’s head office, with the CFO, the group treasurer and the chief sustainability officer. Ilse comes with the sector head and with Tomasz Wrobel. The document is eight pages long and has three parts.
The bank’s reading of the client. The first pages show Castellan its own carbon-cost curve as Meridian computes it (stage 5.4), with the bank’s assumptions on price and pass-through, and the lever waterfall (stage 5.5). Clients value this more than any product page: it tells the CFO how a lender, and by extension the credit market, reads the plan. Ilse states the conclusion plainly. Meridian finds the plan economically sound and the target close to its reference; its reservation is the unapproved capture investment, which carries a quarter of the reduction. She does not disclose the internal score or class, which are internal risk tools, but she does say what would improve the bank’s appetite, because the client can act on it.
The financing options. For the EUR 720m of net capex, the document compares three routes: a ring-fenced transition term loan from the core banks; a bond, green if the capture-equipped plant meets the Taxonomy threshold or sustainability-linked on group intensity; and a mix, with the loan bridging to a later bond once the unit is under construction. The comparison is honest about cost: the loan is cheaper to arrange and flexible on drawdown, the bond gives tenor beyond seven years and diversifies funding. Meridian’s interest is the mix, where it would lend, coordinate and place.
What Meridian commits and what it asks. Ilse offers EUR 60m and a structuring role, subject to credit approval. She then raises reciprocity, in the terms the treasurer himself uses: Meridian provides 10% of the group’s facilities and receives under 5% of its other business. She names the mandates: the coordinator role on the RCF, an active role on the next bond, a share of the allowance purchases that the free-allocation schedule makes unavoidable, the second cash-management country. She asks for no commitment in writing and gets none. The CFO answers that roles follow ideas and that this meeting counts as one.
Reading. The call report filed that evening records what was said on each side, the committee conditions that were communicated (use of proceeds excluding the quarry extension) and the follow-ups by desk. The plan is now a list of dated actions. Whether it reaches 12.31% will be known in a year; what the bank can control is that the capital goes in at the pace the mandates come in.
| Stage | Key figure | Value |
|---|---|---|
| 5.1 | Paid to Meridian; share of wallet | EUR 2.608m; 7.2% |
| 5.1 | Share of facilities; of non-credit wallet | 10.0%; 4.8% |
| 5.2 | Relationship RAROC today | 9.91% (capital EUR 7.93m) |
| 5.2 | Ancillary needed for 12%; gap | EUR 1.78m; EUR 0.63m |
| 5.3 | Reputational risk committee | Approved with three conditions |
| 5.4 | Carbon cost 2026; 2030 no plan; 2030 plan | EUR 198.4m; 578.0m; 391.0m |
| 5.4 | 2030 leverage, no plan; with plan | 2.68x; 3.04x |
| 5.4 | Annual gain of the plan; payback | EUR 142.0m; 5.1 years |
| 5.5 | Target coverage; required pace | 90.9%; 1.95x historical |
| 5.5 | Levers: clinker; fuel; capture | -47.8; -37.0; -25.0 kg |
| 5.6 | Approved capex share; abatement cost | 60.0%; EUR 49.0/t |
| 5.7 | Credibility score; class | 65.0%; Developing |
| 5.8 | Draft plan RAROC, expected; full | 10.95%; 11.91% |
| 5.9 | Growth cap; hold retained | EUR 62.5m; EUR 60m |
| 5.9 | Final plan RAROC, expected; full | 11.16%; 12.31% |
| 5.9 | Final plan RAROC if downgraded to BB | 3.91% |
| 5.9 | Financed emissions today; after loan | 158,750; 285,750 tCO2e |
A relationship is a bundle of exposure types, each with its own conversion into exposure at default. Under CRR3 (Regulation (EU) 2024/1623, applicable since 1 January 2025), drawn loans count in full; undrawn committed facilities at a credit conversion factor of 40%; performance guarantees and other transaction-related contingent items at 50%; financial guarantees at 100%. Meridian applies the foundation IRB approach, which uses these same factors, a regulatory loss given default of 40% and the bank’s own estimate of the probability of default. After the plan, Castellan’s exposure at default is and the credit RWA EUR 81.55m at a risk weight of 39.02%. Under the standardised approach, a BBB corporate is weighted 75%: RWA would be EUR 156.75m, and the output floor at its final level of 72.5% counts EUR 113.64m. A bank bound by the floor sees this relationship with 39% more capital than Meridian’s model shows.
The transition assessment has no direct effect on the regulatory formula: there is no supporting factor for transition loans and no penalising factor for cement (see E02). It enters through the probability of default. Meridian’s rating model takes the financial projections, which now include the carbon-cost scenarios of stage 5.4, and allows a qualitative overlay of one notch. The step from BBB to BB multiplies the probability of default by five and takes the risk weight from 39.02% to 82.06%.
| Line | Computation | EUR m |
|---|---|---|
| Lending revenue today | Section 5.2 | 1.179 |
| Transition loan, net margin | 60 x (130 - 42) bp | 0.528 |
| Transition loan, upfront fee | 60 x 60 bp / 7 | 0.051 |
| Ancillary revenue | 1.15 + 1.31 | 2.460 |
| Revenues | 4.218 | |
| Operating costs | 40% x 1.758 + 60% x 2.460 | -2.179 |
| Expected loss | 0.20% x 40% x 209 | -0.167 |
| Pre-tax result | 1.872 | |
| Net result after tax at 25% | 1.404 | |
| RWA | 209 x 39.02% + 2.5 x 2.46 | 87.70 |
| Allocated capital | 13% | 11.40 |
| RAROC | 1.404 / 11.40 | 12.31% |
The sustainable features move this result at the margin. If the refinanced RCF carries a two-way ratchet of 2.5 bp and Castellan earns the full discount, Meridian gives up EUR 0.016m a year (2.5 bp on EUR 15m drawn and 0.875 bp on EUR 135m undrawn) and RAROC falls to 12.25%. If the 15 bp step-up on the transition loan is triggered, revenue rises by EUR 0.09m and RAROC to 12.67%, which is uncomfortable: the bank earns more when the client’s project is late, and at that moment the rating is probably under review. The steering variable is the rating: at BB the plan returns 3.91%.
INTERACTIVE LAB — RAROC of the final account plan, line by line
Rebuild the plan’s RAROC from its lines and test the effect of the sustainability-linked ratchet, the step-up, the rating and the capital approach.
| Parameter you can move | Range | Default |
|---|---|---|
| Hold in the transition loan | 0 to 150 EUR m | 60 EUR m |
| Ancillary revenue | 1 to 4 EUR m | 2.46 EUR m |
| Ratchet on the RCF margin (negative: discount) | -5 to 5 bp | 0 bp |
| Step-up on the transition loan | 0 / 15 | 0 bp |
| One-year PD | 0.07 / 0.2 / 1 / 4 | 0.2 % |
| Credit RWA approach | F-IRB / Standardised 75% | F-IRB |
| Result at default settings | Value |
|---|---|
| Lending revenue | 1.758 EUR m |
| Revenues | 4.218 EUR m |
| Operating costs | 2.179 EUR m |
| Expected loss | 0.167 EUR m |
| Pre-tax result | 1.872 EUR m |
| Net result | 1.404 EUR m |
| Credit risk weight | 39.02 % |
| Credit RWA | 81.55 EUR m |
| Total RWA | 87.70 EUR m |
| Allocated capital | 11.40 EUR m |
| RAROC | 12.31 % |
| Credit RWA at 72.5% of standardised | 113.64 EUR m |
Try it. Switch to the standardised risk weight: the same plan returns 6.63%. Banks bound by the output floor see hard-to-abate lending very differently from IRB banks.
Treasury charges the coverage perimeter for the funding it uses. The seven-year loan costs a liquidity premium of 42 bp (6 bp per year of tenor), EUR 0.252m a year on EUR 60m: almost a third of the 130 bp margin goes to Treasury before any risk is paid. Long tenors are what transition capex needs and what the funds transfer price penalises most; Meridian’s Treasury gives no rebate for transition assets, although the bank’s own green or transition bond issuance may in time fund them more cheaply (see C15, E04). On the undrawn RCF, the liquidity coverage ratio assumes a 10% drawdown: Treasury holds EUR 13.5m of liquid assets against Meridian’s EUR 135m and charges the 40 bp carry, EUR 0.054m a year or 4 bp of the undrawn amount. During its availability period the transition loan is an undrawn commitment too, and its commitment fee must cover the same charge.
Loans and drawn amounts are measured at amortised cost under IFRS 9 if their cash flows are solely payments of principal and interest. A margin that steps up if a capture unit is not commissioned, or a ratchet linked to emission intensity, is a contingent feature; the amendments to IFRS 9 issued by the IASB in May 2024, effective from 1 January 2026, clarify that such features can be consistent with a basic lending arrangement when the resulting cash flows are not significantly different from those of an identical loan without the feature, with additional disclosures under IFRS 7 (see E05). Upfront fees on the loan enter the effective interest rate and are spread over its life; the coordinator and bookrunner fees are service revenue under IFRS 15, recognised when the service is performed. Expected credit losses are booked from day one on a twelve-month basis (stage 1), on drawn and undrawn amounts and on the guarantees. A downgrade of Castellan that qualifies as a significant increase in credit risk moves the exposures to stage 2 and to lifetime losses, a cost that the RAROC above, built on a one-year expected loss, understates. Banks are expected to reflect climate-related information in the forward-looking scenarios of their models; the coverage file is one of its sources.
Green Asset Ratio. For general-purpose exposures the bank applies the client’s reported Taxonomy ratios. With Castellan reporting 4% of aligned turnover and 22% of aligned capex (case assumptions), the EUR 75m drawn today contributes EUR 3.0m to the turnover-based numerator and EUR 16.5m to the capex-based one. The transition loan is a use-of-proceeds exposure: it counts in full if the financed activity meets the Taxonomy criteria, that is if the plant with its capture unit emits less than 0.469 tonne per tonne of cement and the do-no-significant-harm and minimum-safeguard tests are documented. If it does, the capex-based contribution rises to EUR 76.5m. Undrawn commitments and guarantees do not count. Delegated Regulation (EU) 2026/73 simplified the ratio from 1 January 2026, and the Commission’s wider review of bank Taxonomy indicators is still open in November 2026 (see E06).
Financed emissions. Section 5.9 computed them: 158,750 tCO2e today, 285,750 after the loan. Under the PCAF standard a loan with known use of proceeds can be attributed the emissions of the financed project instead of a share of the company’s; the third edition of December 2025 adds guidance on use-of-proceeds structures. The choice changes the reported figure and should follow the bank’s stated method, not the result (see E07).
Pillar 3 and the prudential plan. Castellan appears in Meridian’s Pillar 3 ESG templates among exposures to carbon-intensive sectors, by maturity bucket: a seven-year loan to a cement producer lengthens the profile that supervisors read as transition risk. In the prudential transition plan required by CRD6, the scorecard class is one of the indicators by which Meridian shows how it manages that risk.
INTERACTIVE LAB — What Castellan adds to Meridian’s Green Asset Ratio numerators
Apply the client’s Taxonomy ratios to general-purpose exposures and the use-of-proceeds rule to the transition loan.
| Parameter you can move | Range | Default |
|---|---|---|
| General-purpose drawn exposure | 0 to 300 EUR m | 75 EUR m |
| Client aligned turnover | 0 to 50 % | 4 % |
| Client aligned capex | 0 to 80 % | 22 % |
| Use-of-proceeds loan drawn | 0 to 150 EUR m | 60 EUR m |
| Financed activity Taxonomy-aligned (1 yes, 0 no) | 0 / 1 | 1 |
| Result at default settings | Value |
|---|---|
| Turnover-based numerator, general purpose | 3.0 EUR m |
| Capex-based numerator, general purpose | 16.5 EUR m |
| Capex-based numerator with the loan | 76.5 EUR m |
| Aligned share of drawn exposure, capex basis | 56.7 % |
Try it. Set alignment to 0, the outcome if the plant stays above the threshold or the evidence is missing: the loan adds nothing, although it finances the same equipment.
| Risk | Control | Owner |
|---|---|---|
| Wallet and pipeline overstated to justify balance sheet | Probabilities by opportunity; quarterly tracking against plan | Head of coverage |
| Transition score inflated by the business | Second-line review of method and samples; four-eyes scoring | ESG risk (second line) |
| Sector-policy breach or missed controversy | Automated screening, escalation thresholds, committee minutes | ESG risk, compliance |
| Inconsistent assumptions between credit and ESG views | Shared carbon-price and pass-through scenarios | Credit risk |
| Tying and conflicts of interest | Reciprocity never contractual; conflicts register | Compliance |
| Misuse of confidential information | Wall-crossing procedure; call-report discipline | Compliance |
| Label on a loan that does not deserve it | Use-of-proceeds ring-fencing; external review; declassification rule | Sustainable finance advisory |
| Bank’s own claims about “financing the transition” | Claims tied to documented scorecards and volumes | Communications, legal |
Red flags in a client’s transition plan.
Red flags inside the bank.
Dated facts (verified in November 2026).
What changed recently. The centre of gravity has moved from voluntary collective commitments to supervised individual plans: banks now explain their client transition assessments to their supervisor more than to an alliance. At the same time the Omnibus has reduced the amount of standardised client data, so banks rely more on bilateral questionnaires, which puts the workload on coverage. And transition finance has gained its own loan-market text, which gives hard-to-abate clients a label that the green loan could not offer.
Open debates.
INTERACTIVE LAB — Self-assessment: coverage, transition plans and account planning
Check the concepts and orders of magnitude of the module.
Q1. What is a client’s wallet?
Answer: B. The wallet is the total paid to all banks; the bank’s portion of it is the share of wallet.
Q2. Meridian holds 10% of Castellan’s facilities and earns 4.8% of its non-credit wallet. What does this indicate?
Answer: A. Reciprocity would give a lender a share of fee business in line with its share of the lending.
Q3. With Meridian’s parameters, what is the marginal RAROC of one euro of ancillary revenue?
Answer: D. 0.40 x 0.75 of profit for 0.13 x 2.5 of capital: 0.30 / 0.325 = 92%.
Q4. Which body decides on an escalated ESG controversy before new financing?
Answer: C. The reputational risk committee approves, approves with conditions or refuses; its decision binds the business.
Q5. Under the law in force in November 2026, what is the CBAM factor applied to free allocation in 2030?
Answer: C. 97.5% in 2026, 95% in 2027, 90% in 2028, 77.5% in 2029, 51.5% in 2030, zero in 2034.
Q6. Why can cement not be decarbonised by fuel switching alone?
Answer: B. Process emissions are released by the chemistry of clinker, whatever the fuel; only less clinker or capture reduces them.
Q7. In the case, why is Castellan more leveraged in 2030 with its plan than without?
Answer: B. EBITDA is higher with the plan (EUR 947.1m against 805.1m) but net debt rises to EUR 2,880m: 3.04x against 2.68x.
Q8. Which change would move Castellan from ‘Developing’ to ‘Credible’ in the scorecard?
Answer: C. The capex criterion moves from 2 to 4 and the score from 65.0% to 77.5%.
Q9. The transition term loan returns 7.38% stand-alone. What follows for the account plan?
Answer: B. The loan dilutes the relationship RAROC; only ancillary revenue brings the plan to the hurdle. Meridian grants no hurdle discount.
Q10. What happens to Meridian’s financed emissions when the EUR 60m loan is drawn?
Answer: C. Attribution follows the drawn amount over EVIC: 75 to 135 is +80%, whatever the purpose under the business-loan method.
Q11. What did Omnibus I (Directive (EU) 2026/470) change for transition plans?
Answer: B. The CSDDD obligation was removed; disclosure of an existing plan under the ESRS remains for companies in the narrowed CSRD scope.
Q12. Which statement about reciprocity is correct?
Answer: C. Making a loan conditional on other business is restricted (tying); relationship banking rests on an uncontracted expectation.
Q13. Under CRR3, what credit conversion factor applies to performance guarantees?
Answer: C. Transaction-related contingent items take 50%; undrawn committed facilities 40%; financial guarantees 100%.
Meridian lends EUR 200m for five years to a BBB client at a margin of 90 bp. Compute the stand-alone RAROC and the ancillary revenue needed for the relationship to reach 12%.
Solution. Net margin: 90 - 30 = 60 bp, lending revenue EUR 1.20m. Costs: 40%, EUR 0.48m. Expected loss: . RWA: ; capital EUR 10.14m. RAROC: . Each euro of ancillary revenue adds of net profit and of capital. Solve : . The relationship needs two and a half times the lending revenue in fees.
A producer makes 10 Mt of cement with a clinker ratio of 80% and a scope 1 intensity of 640 kg per tonne. Benchmark 0.656, allowance price EUR 90. Compute the gross carbon cost in 2028 (CBAM factor 90%) and in 2034 at unchanged intensity.
Solution. Emissions: Mt. Clinker: 8 Mt. Free allocation in 2028: Mt. Allowances to buy: 1.677 Mt, costing EUR 150.9m. In 2034 the factor is zero: 6.4 Mt at EUR 90 is EUR 576m, almost four times the 2028 cost with no change in the plant.
A steel client scores 4 on targets, 2 on levers, 3 on capex, 2 on governance and 3 on track record. Meridian’s committed exposure is EUR 400m, of which EUR 90m is drawn. The client’s EVIC is EUR 4,500m and its scope 1 and 2 emissions are 8 Mt. It asks for a new drawn loan of EUR 120m. Compute the score and class, the maximum new exposure, and the financed emissions before and after.
Solution. Score: , or 70%: “Developing”. Growth cap: 25% of EUR 400m, EUR 100m; the request is cut to EUR 100m. Financed emissions before: of 8 Mt, 160,000 tCO2e. After: , 337,778 tCO2e, more than double.
1. What does a coverage banker do, and how is the performance measured? The banker owns the relationship with a portfolio of client groups and originates business for all product lines. Performance is measured on the relationship: revenues across products, net of costs and expected loss, against the capital committed, with a RAROC compared with the hurdle. Share of wallet against share of lending shows whether the bank is paid for its balance sheet. Qualitative measures matter too: the quality of the credit and KYC files, and the mandates won on the banker’s ideas.
2. Why do banks lend to investment-grade corporates at returns below their hurdle? Because lenders form the group from which the client picks its banks for fee business. A back-up facility returns a few per cent on capital, while one euro of ancillary revenue returns far more because it consumes little capital. The bank therefore judges the relationship as a whole. The discipline is to check each year that the fee business has in fact come, and to cut the commitment when it has not.
3. How would you assess the credibility of a transition plan? I would look at five things: targets on the material scopes in the sector’s physical unit and against a recognised pathway; levers quantified in tonnes with their technological risk; capex in the budget and approved by the board; governance and pay; and the track record with assured data. Then I would test the economics: does the plan pay at the carbon price the company faces? A plan that pays is more believable than one that depends on goodwill. I would concentrate on the years that overlap with our exposure.
4. A client in a hard-to-abate sector asks for a seven-year loan. What do you check before taking it to credit committee? The sector policy and any controversy, with escalation closed. The transition class and what it allows in tenor and growth. The carbon-cost scenarios in the financial projections, shared with the credit analyst. The use of proceeds and whether a label is justified. The relationship RAROC with and without the loan, and the fee business that makes it acceptable. Finally the effect on our own sector metrics.
5. Financing a client’s decarbonisation increases our financed emissions. How do you handle that? It is true under the standard attribution method: emissions follow the drawn amount. I would not hide it. I would report the exposure under the sector’s physical-intensity metric, where a client that improves lowers the book’s average over time, and document the use of proceeds and the expected abatement. If the bank has an absolute target, the trade-off goes to the committee that owns it. Refusing transition loans to protect a metric defeats its purpose.
6. What is the difference between KYC and the ESG screen? KYC is a legal obligation under anti-money-laundering rules: identity, ownership, sanctions, source of funds. The ESG screen rests on prudential expectations about risk management and on the bank’s own policies: sector criteria, exclusions, controversies. They share clients, timing and many sources, so they are run together, but the legal basis, the owner and the consequences of a failure differ.
7. The client’s CFO says your bank’s transition assessment is harsher than a competitor’s. What do you answer? That methods differ between banks and that I can explain ours: which criteria, what evidence, and what would change our view. I would not disclose the internal score, but I would be specific about the weak point, for instance an unapproved investment, because the client can act on it. I would also say what the assessment unlocks: tenor, amount, labels. A clear answer usually helps the relationship more than a generous one.
8. How does a carbon price reach a bank’s RAROC? Through the client’s cash flows and therefore its rating. Rising carbon costs that are not passed on reduce EBITDA and raise leverage; the internal rating falls, the probability of default and the risk weight rise, and expected loss and capital increase. In the case, one notch from BBB to BB takes the relationship from above 12% to under 4%. Pricing features such as ratchets move RAROC by a few hundredths of a point; the rating moves it by points.
Regulation and official texts
Supervisors
Standards and market guidance
Market and sector information
Academic and reference