Corporate lending is the oldest business of Meridian Bank’s Corporate Banking division. The desk lends the bank’s own balance sheet to large and mid-sized companies through two plain products: term loans, drawn once and repaid on a schedule (see B02), and revolving credit facilities (RCFs), which are promises to lend on demand. This module covers the second and its sustainability-linked version, which between 2019 and 2024 became the default format for European investment-grade RCFs and has since been questioned.
The clients are investment-grade and crossover companies: the borrower of our case, Borealis Packaging, is a listed European group rated BBB. Its treasurer does not come to the bank for cash, since Borealis funds itself in the bond and commercial paper markets. He comes for insurance: a committed line that it can draw if those markets close, and that the rating agencies count as liquidity.
How the desk earns money is a short and slightly uncomfortable story. An undrawn investment-grade RCF pays the bank a commitment fee of roughly 15 to 35 bp a year on the unused amount and a margin of roughly 40 to 100 bp over EURIBOR on whatever is drawn, plus small upfront and utilisation fees. Against that the bank holds capital and liquidity for five years. We will see in section 5 that Meridian’s EUR 75m share of the Borealis facility earns about EUR 226,000 a year and returns about 4% on allocated capital, a third of the 12% hurdle. The facility is a loss-leader. It is granted because lenders in the facility form the “relationship group” from which the treasurer chooses the banks for bond issues, hedging, cash management and advisory mandates. The desk’s real profit and loss is the relationship’s, not the loan’s.
WHO DOES WHAT — The coverage banker owns the client and the cross-sell plan. The loan structurer (see B04) prices the facility, negotiates the term sheet and runs the syndicate. The sustainable finance solutions team acts as sustainability coordinator: it challenges the KPIs and targets. The credit analyst writes the credit application; the credit committee approves the hold. Legal and external counsel draft the facility agreement on the Loan Market Association (LMA) form. After signing, the loan agency team (see B17) runs drawings, fees, rate fixings and the annual certificates, and portfolio management decides whether to keep, hedge or distribute the exposure (see B16).
Adjacent functions: Treasury sets the funds transfer price the desk pays for liquidity (see E04); Finance decides the IFRS 9 classification and the sustainability reporting (see E05, E06); Compliance reviews any public claim that the facility is “sustainable” (see E10).
A realistic week in early November 2026. Monday: signing of the Borealis facility. Tuesday: annual review of a borrower whose second KPI test failed in the spring, to decide whether Meridian votes for declassification. Wednesday: credit committee for a EUR 40m increase under a client’s accordion option. Thursday: call with a chemical group whose treasurer asks whether to keep the sustainability-linked format at refinancing or drop it, as several large borrowers did in 2025 and 2026. Friday: quarterly relationship-profitability review, where every facility below the hurdle must show the ancillary revenues that justify it. The numbers of this module are the ones needed in that meeting.
IN ONE SENTENCE — A revolving credit facility is a commitment by a group of banks to lend up to a fixed amount, on demand, for a fixed period, with amounts repaid available to be borrowed again.
INTUITION — An RCF is a credit card for a company, with one difference: the company pays rent on the unused limit. The commitment fee is the rent on an empty flat that the tenant may move into at any time, with a few days’ notice, for the next five years.
The borrower sends the facility agent a utilisation request, usually three business days before the drawing date, stating the amount and the interest period (one, three or six months). Each lender funds its pro-rata share through the agent. At the end of the period the borrower repays or rolls the loan. Interest is EURIBOR for the period, floored at zero, plus the margin.
Investment-grade RCFs are used in two ways, and the difference drives the bank’s economics:
Borealis does both: the facility backs a EUR 500m commercial paper programme and covers a seasonal working-capital peak. The case assumes an average utilisation of 20%: EUR 150m drawn, EUR 600m undrawn.
An RCF has four prices where a term loan has two. The term sheet of the Borealis facility, signed on 2 November 2026, reads as follows (case assumptions, in line with market practice for a BBB borrower):
| Term | Borealis facility | Paid on |
|---|---|---|
| Amount and tenor | EUR 750m, 5 years, two one-year extension options (5+1+1) | n/a |
| Drawn margin | 70 bp over EURIBOR (zero floor) | Drawn amount |
| Commitment fee | 35% of the applicable margin, i.e. 24.5 bp | Undrawn amount |
| Utilisation fee | 10 bp up to one third drawn, 20 bp up to two thirds, 40 bp above | Whole drawn amount |
| Upfront (participation) fee | 25 bp flat at signing | Each lender’s commitment |
FORMULA
is the annual cost excluding the reference rate; the facility size; the average drawn amount; the utilisation; the drawn margin; the utilisation fee, which depends on the utilisation tier; the commitment fee expressed as a share of the margin (35% here); the upfront fee; the tenor in years over which we spread the upfront fee.
Term by term. The drawn margin pays for credit risk and capital on money actually lent. The commitment fee pays for the option to draw; because it is a percentage of the margin, any change in margin (rating grid, sustainability ratchet) flows into it. The utilisation fee compensates lenders when a back-up line is heavily drawn, which usually signals stress; it applies to the whole drawn amount once a tier is crossed. The upfront fee is paid once and rewards the size of the ticket.
CASE — At 20% utilisation Borealis pays EUR 1.05m of margin (EUR 150m at 70 bp), EUR 0.15m of utilisation fee (EUR 150m at 10 bp) and EUR 1.47m of commitment fee (EUR 600m at 24.5 bp): a running cost of EUR 2.67m. Adding one fifth of the EUR 1.875m upfront fee gives EUR 3.045m a year, or 40.6 bp on the whole commitment. Expressed per euro actually borrowed, the facility costs 203 bp over EURIBOR. With three-month EURIBOR assumed at 2.65% (it was close to 2.67% in late October 2026), interest at the reference rate on the EUR 150m adds EUR 3.975m.
INTERACTIVE LAB — What a revolving credit facility costs per year
Decompose the annual cost of the Borealis RCF (excluding EURIBOR) into drawn margin, utilisation fee, commitment fee and amortised upfront fee, and see how it moves with utilisation.
| Parameter you can move | Range | Default |
|---|---|---|
| Facility size | 100 to 2000 EUR m | 750 EUR m |
| Average utilisation | 0 to 100 % | 20 % |
| Drawn margin over EURIBOR | 20 to 300 bp | 70 bp |
| Commitment fee as share of margin | 20 to 50 % | 35 % |
| Upfront (participation) fee, flat | 0 to 100 bp | 25 bp |
| Tenor used to spread the upfront fee | 1 to 7 years | 5 years |
| EURIBOR for the interest period | 0 to 6 % | 2.65 % |
| Result at default settings | Value |
|---|---|
| Margin on drawn amount | 1.050 EUR m |
| Utilisation fee (10 / 20 / 40 bp by tier) | 0.150 EUR m |
| Commitment fee on undrawn amount | 1.470 EUR m |
| Running cost (margin and fees, before upfront) | 2.670 EUR m |
| Upfront fee spread over the tenor | 0.375 EUR m |
| Total annual cost excluding EURIBOR | 3.045 EUR m |
| Total cost on the whole commitment | 40.6 bp |
| Total cost per euro actually drawn | 203.0 bp |
| Interest at EURIBOR on the drawn amount (for reference) | 3.975 EUR m |
Try it. Move utilisation from 30% to 35%: the cost jumps because the utilisation fee tier applies to the whole drawn amount, not only to the slice above one third.
PITFALL — Do not quote the “price” of an RCF as its margin. At 20% utilisation, more than half of what Borealis pays is commitment fee, and a concession on the margin applies in full to one fifth of the facility and at 35% to the rest. The same arithmetic will deflate the sustainability ratchet in section 5.
The “5+1+1” structure gives the borrower two requests, around the first and second anniversaries, to extend the termination date by one year. Each lender decides individually; those that decline are repaid at their original maturity. Consenting lenders receive an extension fee (5 bp here). For the bank this is potentially a seven-year commitment granted on a five-year credit decision, so each extension needs a fresh credit approval. An accordion option lets the borrower ask lenders to increase the facility, here by up to EUR 150m, without obligation for them to accept.
The risk of an RCF is that the line is fully drawn when the borrower defaults: companies in difficulty draw their lines, as many did in March 2020. Regulation captures this with the credit conversion factor (CCF), the share of the undrawn amount that is added to the drawn amount to obtain the exposure at default:
is the exposure at default, the drawn amount, the undrawn commitment. Under CRR3, in force since 1 January 2025, the CCF for a committed facility such as this one is 40% under the standardised approach and also under the foundation IRB approach (details and sources in section 6). For Meridian’s EUR 75m hold at 20% utilisation, .
Protection comes from the documentation: representations, information undertakings, a negative pledge, restrictions on disposals and mergers, cross-default, a change-of-control prepayment right and, for BBB borrowers, often a single leverage covenant or none. The lenders’ obligation to fund is subject to no default continuing and to the repeating representations being true. They cannot otherwise refuse, which is why the commitment is not unconditionally cancellable.
WHY IT MATTERS ON THE DESK — On the desk, three questions are asked about any RCF before its “green” features are mentioned: how drawn will it be, what is our final hold, and what else does the client give us. The answers determine capital, revenue and the relationship return, whatever the quality of the KPIs.
The facility agreement follows the LMA’s recommended form for investment-grade revolving facilities. Around it sit the mandate letter, the term sheet and the fee letters (upfront, agency, coordination). The Borealis facility is a club deal: twelve relationship banks, six at EUR 75m and six at EUR 50m, all paid the same 25 bp participation fee. Meridian is coordinator, sustainability coordinator, facility agent and one of the EUR 75m lenders. Syndication and fee waterfalls belong to B04; agency operations to B17.
IN ONE SENTENCE — A sustainability-linked loan (SLL) is any loan whose economic or structural terms change according to whether the borrower meets predefined sustainability performance targets, while the money can be used for anything.
A green loan (see B02) is defined by its use of proceeds: the money goes to an identified green asset. An SLL is defined by the behaviour of the borrower: the money is general-purpose, and performance is measured once a year against targets written into the agreement. An RCF that backs commercial paper has no identifiable use of proceeds, so the linked format is the only way to attach a sustainability feature to it. This is why SLLs took over the investment-grade RCF market, and it is the root of most criticism: the label says nothing about what is financed.
INTUITION — A green loan is a mortgage that can only buy an energy-efficient house. A sustainability-linked loan is an ordinary overdraft whose rate goes down a little if you lose weight this year and up a little if you do not. The bank does not care what you buy; it reads the scale once a year.
Vocabulary. A KPI (key performance indicator) is the metric, for example absolute scope 1 and 2 emissions. An SPT (sustainability performance target) is the level the KPI must reach in a given year. The margin ratchet is the grid that converts SPT outcomes into a discount or a premium on the margin.
The governing standard is voluntary: the Sustainability-Linked Loan Principles (SLLP), published jointly by the Loan Market Association (EMEA), the Asia Pacific Loan Market Association and the Loan Syndications and Trading Association (Americas). First issued in March 2019, they were tightened in 2022 and February 2023 and apply today in the version of 26 March 2025. That revision was mostly clarificatory: a formal reading of “shall” (mandatory), “should” (recommended) and “may” (optional), the end of grandfathering for older transactions, and more precision on KPIs and targets. The Guidance was updated at the same time and again on 3 November 2026, when the three associations also published their Transition Loan Principles (see A04).
The SLLP rest on five core components.
| Component | What the SLLP require | What the desk checks |
|---|---|---|
| 1. Selection of KPIs | Material to the core business, consistent with the sustainability strategy, measurable, externally verifiable where feasible | Does the KPI cover the main impact? Is its scope written down? |
| 2. Calibration of SPTs | Ambitious: beyond business as usual and beyond regulation; benchmarked against history, peers and science | Yearly trajectory? What happens without extra effort? |
| 3. Loan characteristics | An economic or structural outcome linked to the SPTs | The ratchet grid |
| 4. Reporting | Information to lenders at least once a year | Certificate, deadline |
| 5. Verification | External verification for each period that can trigger an adjustment | Who verifies, assurance level, timing |
Three points deserve attention. First, verification after signing is mandatory, whereas a pre-signing review such as a second-party opinion is only recommended; information already assured in the borrower’s annual sustainability report can be reused. Second, materiality is read in the double sense of European reporting: impact of the company on people and the environment, and financial effect on the company. Third, the SLLP deal explicitly with sleeping SLLs: facilities signed with the sustainability mechanics agreed but the KPIs or SPTs still to be set. The borrower must make every reasonable effort to set them before signing, explain why it cannot, and agree a backstop date at most twelve months after origination; until then the facility should not be presented as sustainability-linked.
PITFALL — “SLLP-aligned” is not a regulated label and no authority certifies it. Each bank applies its own classification policy to decide whether the loan counts towards its sustainable finance target, so two banks in the same syndicate can classify the same facility differently, a point the UK Financial Conduct Authority raised in 2023 (section 8).
Principles say what a good SLL looks like; the contract says what happens when things go wrong. In May 2023 the LMA published draft provisions for sustainability-linked loans, and in August 2026 it released their first comprehensive update. The update makes the provisions usable with any LMA form, adds a Part B for sleeping SLLs, lets existing reports serve as reporting and verification deliverables, and widens the declassification triggers. The architecture is the following.
No European regulation defines a sustainability-linked loan: the Taxonomy, the European Green Bond Standard and the Green Asset Ratio are built on use of proceeds or on the counterparty’s activities, not on KPI-linked pricing. Regulation reaches the product through four doors.
Structures have converged on two or three core KPIs with annual SPTs. Two-way pricing is the norm in European investment-grade lending, with total adjustments of about 2.5 to 5 bp either way. And the product is in retreat from its 2024 peak: some large borrowers have removed the feature at refinancing, and lenders decline to label facilities whose targets they cannot defend (dated figures in section 8).
INTERACTIVE LAB — Life of a sustainability-linked RCF, from request for proposals to annual test
Follow the Borealis facility through the people who touch it and the documents they produce.
| # | Who | Does what | Deliverable |
|---|---|---|---|
| 1 | Group treasurer, Borealis | Sends a request for proposals to the relationship banks to refinance the RCF maturing in 2027 (early September 2026) | Request for proposals |
| 2 | Coverage banker and loan structurer, Meridian | Pitch size, tenor, pricing grid, hold and a view on KPIs, after a pre-screening by credit risk and a first RAROC | Pricing proposal and indicative term sheet |
| 3 | Group treasurer, Borealis | Appoints Meridian as coordinator and sustainability coordinator | Mandate letter and sustainability coordinator letter |
| 4 | Sustainable finance solutions, Meridian, with Borealis sustainability team | Select KPIs, set baselines and annual SPTs, document benchmarks and methodology | KPI and SPT annex, rationale memo for lenders |
| 5 | External reviewer | Assesses materiality of KPIs and ambition of SPTs before signing (optional under the SLLP) | Second-party opinion |
| 6 | Credit analyst and credit committee, Meridian | Analyse the credit, the RAROC and the classification; approve the final hold | Credit application and approval for EUR 75m |
| 7 | Loan structurer, Meridian | Invites the relationship banks, answers lender questions on KPIs with the sustainability coordinator, allocates | Commitment letters, allocation of EUR 750m to twelve lenders |
| 8 | External counsel and Legal | Draft the facility agreement on the LMA form with the sustainability-linked provisions; signing on 2 November 2026 | Facility agreement, fee letters |
| 9 | Facility agent (Meridian loan agency) | Checks conditions precedent, distributes upfront fees, records KPIs, SPTs, grid and due dates in the loan system | Conditions precedent letter, facility static data |
| 10 | Borealis treasury and facility agent | Drawings, rollovers, repayments; quarterly commitment and utilisation fees | Utilisation requests, rate fixing and fee notices |
| 11 | Borealis and external verifier | After year-end, compute the three KPIs; verifier gives limited assurance | Verification report |
| 12 | CFO, Borealis | Signs and delivers the annual certificate within 120 days of year-end (by 30 April) | Sustainability compliance certificate |
| 13 | Facility agent | Checks completeness, distributes to lenders, applies the new margin and commitment fee from 1 May | Margin adjustment notice |
| 14 | Lenders, sustainability coordinator, agent | On an acquisition, methodology change or repeated failure: negotiate amendments or vote on declassification | Amendment agreement or declassification notice |
Mandate (steps 1 to 3). The request for proposals asks each bank for a margin grid, fees, a view on the sustainability structure and its hold appetite. The mandate letter appoints the coordinator; a separate sustainability coordinator letter defines that role narrowly. The control at this stage is internal: a pre-screening from credit risk and a first RAROC before any number leaves the bank.
Structuring the KPIs (steps 4 and 5). The sustainable finance team works with the borrower’s sustainability and treasury teams for three to five weeks. The deliverable is an annex to the term sheet: definition and scope of each KPI, baseline, annual SPTs, calculation methodology, and a rationale memo for lenders with the benchmarks used. Meridian’s internal rule (a case assumption, but typical) is that its sustainable finance classification committee must approve the structure before the facility can be recorded as sustainability-linked in the bank’s systems. A second-party opinion is optional under the SLLP; Borealis commissions one because several lenders in the group require it for their own classification.
Credit approval (step 6). The credit application adds three items to the usual ground: the stand-alone and relationship RAROC, the ESG risk assessment of the borrower (see E01), and the classification decision. The committee approves a final hold of EUR 75m.
Syndication and documentation (steps 7 and 8). In a club deal, syndication is an invitation to relationship banks at pre-agreed tickets. Lenders’ questions on the KPIs go to the sustainability coordinator; each lender has its own classification policy and may make its commitment conditional on a change. External counsel then inserts the LMA sustainability provisions in the facility agreement.
Closing and life of the facility (steps 9 and 10). The agent checks the conditions precedent, distributes the upfront fee, and records the sustainability attributes in the loan system: KPI definitions, SPTs, certificate due date, grid. An error in this static data means a wrong margin on every interest period (see B17).
Annual test (steps 11 to 13). Section 5.7 replays one test in detail.
Events (step 14). Acquisitions, methodology changes, missed deadlines and repeated failures go through the amendment and declassification provisions, with majority-lender votes organised by the agent.
Question. What does Borealis need, and what does the facility look like before the sustainability layer?
Data (case assumptions). Borealis Packaging is a listed European group producing fibre-based, rigid plastic and flexible packaging for food, beverage and consumer goods companies.
| Item | Value |
|---|---|
| Revenue / EBITDA (FY2025) | EUR 5.2bn / EUR 780m |
| Net debt / EBITDA | 2.5x |
| Rating | BBB stable |
| Enterprise value including cash (EVIC) | EUR 7.0bn |
| Scope 1 + 2 emissions FY2025 (market-based) | 880 ktCO2e |
| Scope 3 emissions FY2025 | 3,400 ktCO2e |
| Climate target | Near-term target validated by the SBTi: -42% scope 1 + 2 by FY2032 from FY2022 |
| Existing facility | EUR 600m RCF maturing October 2027, sustainability-linked, one KPI |
In September 2026 the treasurer launches the refinancing: EUR 750m, five years with two extension options, to back the EUR 500m commercial paper programme and seasonal working capital. Pricing was given in section 2.2. Meridian is mandated as coordinator and sustainability coordinator, with a final hold of EUR 75m (10%).
Result. From section 2: at 20% utilisation the facility costs Borealis EUR 2.67m a year in margin and fees, EUR 3.045m including the spread upfront fee. This is the base on which the ratchet will act.
Reading. The old facility had a single KPI, an ESG rating. The lenders’ first message is that this would no longer pass their classification committees: a rating is an opinion of a third party about many things, not a measure of what the company does. The 2025 SLLP guidance tolerates ESG ratings as KPIs, but current practice has moved to two or three operational KPIs.
Question. Which KPIs are material for a packaging group, and where should the SPTs be set so that they are ambitious but reachable?
KPI 1: absolute scope 1 and 2 greenhouse gas emissions (ktCO2e, market-based). Paper mills and plastic converting plants burn gas and use a lot of electricity; operational emissions are the first impact a lender expects to see. Borealis has a science-based target validated under the SBTi’s version 1 criteria, whose cross-sector pathway for 1.5°C requires a linear reduction of 4.2% of base-year emissions per year. From a base of 1,000 kt in FY2022 this gives a straight line losing 42 kt a year.
FORMULA
is the target for financial year ; the base-year emissions (1,000 ktCO2e). The reduction is linear in tonnes, not compound: each year removes 4.2% of the base-year figure, which is a growing percentage of what is left.
The SPTs are set exactly on this line: 832, 790, 748, 706 and 664 kt for FY2026 to FY2030, the last being 33.6% below the base year. Borealis reported 880 kt in FY2025, slightly above the line (874 kt). Reaching 664 kt in FY2030 therefore requires a compound fall of 5.48% a year for five years. A business-as-usual path of 1.5% a year (efficiency gains only, a case assumption discussed with the borrower’s engineers) would leave emissions at 816 kt in FY2030, 152 kt above the target. That gap is the measure of ambition: the SPT cannot be met without the planned biomass boiler conversions and renewable power contracts.
INTERACTIVE LAB — KPI 1 trajectory: scope 1 and 2 emissions against the SPTs
Build the linear SPT path from the base year, compare it with a business-as-usual path and with the case outcome, and test what an acquisition does to the comparison.
| Parameter you can move | Range | Default |
|---|---|---|
| Base-year emissions (FY2022) | 500 to 2000 ktCO2e | 1000 ktCO2e |
| Linear annual reduction, share of base year | 1 to 7 % per year | 4.2 % per year |
| Business-as-usual annual reduction after FY2025 | 0 to 5 % per year | 1.5 % per year |
| Emissions of a business acquired in FY2028 (no rebasing) | 0 to 300 ktCO2e | 0 ktCO2e |
| Result at default settings | Value |
|---|---|
| Trajectory value FY2025 (reported: 880) | 874 ktCO2e |
| SPT FY2026 | 832 ktCO2e |
| SPT FY2027 | 790 ktCO2e |
| SPT FY2028 | 748 ktCO2e |
| SPT FY2029 | 706 ktCO2e |
| SPT FY2030 | 664 ktCO2e |
| SPT FY2030 versus base year | -33.6 % |
| Compound annual change needed from FY2025 actual to SPT FY2030 | -5.48 % per year |
| Business-as-usual emissions FY2030 | 816 ktCO2e |
| Effort beyond business as usual in FY2030 | 152 ktCO2e |
| Years in which the case outcome meets the SPT (out of 5) | 3 |
Try it. Add an acquisition of 120 kt in FY2028 without rebasing: Borealis fails every remaining test although its own plants performed exactly as before. This is why the facility contains a sustainability amendment event.
KPI 2: recycled content in plastic packaging placed on the market (% by weight). This is the core circularity metric for the plastics division and the one customers ask for. The baseline is 18% in FY2024. SPTs rise by four points a year: 22, 26, 30, 34 and 38% for FY2026 to FY2030. The SLLP test is “beyond regulation”: applying the 2030 legal minima of the Packaging and Packaging Waste Regulation to Borealis’s product mix (40% contact-sensitive non-PET at 10%, 25% contact-sensitive PET at 30%, 35% other plastic packaging at 35%) gives a weighted floor of 23.75%. The FY2030 SPT of 38% is 14 points above what the law will require.
KPI 3: share of purchased-goods emissions covered by suppliers with validated science-based targets (%). Scope 3 is almost four times scope 1 and 2 for Borealis, mostly resin, pulp and aluminium. An absolute scope 3 KPI was considered and rejected by the lenders: the data rely largely on spend-based estimates that move with every change of emission factors, which makes an annual test fragile. The supplier-engagement metric is measurable and verifiable. Baseline 28% in FY2024; SPTs 35, 42, 50, 58 and 67%.
| Financial year | KPI 1 (ktCO2e, max) | KPI 2 (%, min) | KPI 3 (%, min) |
|---|---|---|---|
| Baseline | 1,000 (FY2022) | 18 (FY2024) | 28 (FY2024) |
| FY2026 | 832 | 22 | 35 |
| FY2027 | 790 | 26 | 42 |
| FY2028 | 748 | 30 | 50 |
| FY2029 | 706 | 34 | 58 |
| FY2030 | 664 | 38 | 67 |
Reading against the five components. KPI 1 and KPI 2 are outcome metrics, material and benchmarked against an external reference (the SBTi pathway, the regulation). KPI 3 is weaker and the sustainability coordinator says so in the rationale memo: it measures an effort (suppliers setting targets), not an outcome (suppliers’ emissions falling). It was accepted as the least bad way to bring scope 3 into the structure. A safety KPI (injury frequency rate) was proposed by the borrower and declined: material for employees, but several lenders exclude from their classification any structure in which a social KPI can offset a missed climate KPI. Two further weaknesses are recorded honestly. The FY2026 SPT is set in November 2026, when ten months of the year are already known; it sits on the trajectory, but it is hardly a forward-looking target. And the SBTi published version 2.0 of its Corporate Net-Zero Standard in June 2026, which will require separate scope 1 and scope 2 targets for new validations; the combined KPI may need to be restated during the life of the facility, which is one purpose of the rendez-vous clause.
Question. How should three test results be converted into a margin?
Four design choices have to be made, and each has a market convention and a critique.
Two-way or one-way. A one-way ratchet only gives a discount; the borrower cannot lose. A two-way ratchet adds a premium when targets are missed. The lenders insist on two-way: with a discount-only grid, a borrower that abandons its targets pays exactly what it would have paid without the label. The premium raises an awkward question, since the bank earns more when the client fails. Some facilities direct the premium to a charity or a ring-fenced sustainability budget to remove that incentive; Borealis’s lenders keep it, which is the more common choice.
Per-KPI or aggregate scoring. An aggregate grid (“discount if at least two of three are met”) lets a strong KPI hide a weak one. A per-KPI grid prices each target separately. Borealis’s grid is per-KPI and symmetrical: each SPT met reduces the margin by 1 bp, each SPT missed increases it by 1 bp.
| SPTs met | Adjustment | Applicable margin | Commitment fee |
|---|---|---|---|
| 3 of 3 | -3 bp | 67 bp | 23.45 bp |
| 2 of 3 | -1 bp | 69 bp | 24.15 bp |
| 1 of 3 | +1 bp | 71 bp | 24.85 bp |
| 0 of 3, or no certificate | +3 bp | 73 bp | 25.55 bp |
FORMULA
is the margin adjustment in bp; the maximum adjustment (3 bp); the number of SPTs met (0 to 3). With the adjustment is bp; with it is bp.
Size of the step. Three basis points on a 70 bp margin is 4.3% of the margin. Market practice for investment-grade borrowers is 2.5 to 5 bp in total. Lenders would accept more; borrowers resist a larger premium, and lenders resist a larger discount on a facility that already misses their hurdle. The step stays small because both sides want it small.
Margin only, or commitment fee too. Because the commitment fee is 35% of the applicable margin, the adjustment flows into it: 1.05 bp at the maximum. On a facility that is 80% undrawn this choice matters more than the headline step, as the next stage shows. Some facilities define the commitment fee on the unadjusted margin; the ratchet then pays only on drawings.
INTERACTIVE LAB — Annual KPI test: from three results to one margin adjustment
Replay the annual test of the three KPIs for each financial year of the case and read the resulting adjustment on the per-KPI grid.
| Parameter you can move | Range | Default |
|---|---|---|
| Financial year tested | 2026 / 2027 / 2028 / 2029 / 2030 | 2027 |
| Maximum adjustment (three KPIs) | 0 to 7.5 bp | 3 bp |
| Compliance certificate and verification report delivered on time | yes / no | yes |
| Result at default settings | Value |
|---|---|
| KPI 1 SPT: scope 1 + 2 emissions (lower is better) | 790 ktCO2e |
| KPI 1 verified result | 795 ktCO2e |
| KPI 2 SPT: recycled content in plastic packaging | 26.0 % |
| KPI 2 verified result | 26.5 % |
| KPI 3 SPT: supplier emissions under science-based targets | 42.0 % |
| KPI 3 verified result | 41.0 % |
| SPTs met | 1 of 3 |
| Margin adjustment for the next 12 months | 1.00 bp |
| Applicable margin | 71.00 bp |
| Applicable commitment fee (35% of margin) | 24.850 bp |
| Maximum adjustment as share of the 70 bp margin | 4.3 % |
| KPI 2: weighted legal minimum for the Borealis mix in 2030 | 23.75 % |
Try it. Pick FY2029 and switch the certificate to ‘no’: three SPTs were met but the margin goes up by the full 3 bp, because an undelivered certificate is treated as three misses.
PITFALL — The grid is tested on verified figures, but the trigger is a document. If the compliance certificate and the verification report are not delivered by 30 April, the margin steps up by the full 3 bp whatever the real performance, until they are delivered. Treasurers lose discounts through a late assurance report more often than through a missed target.
Three requests from the treasurer, and the answers. During the negotiation Borealis tested three alternatives, each of which a structurer should be able to argue.
A sleeping structure. The treasurer first proposed to sign in November and agree the KPIs by mid-2027, once the group’s new climate plan was published. The SLLP allow this with a backstop of twelve months, and the August 2026 LMA provisions contain the drafting. Meridian advised against it: the three KPIs already existed in the published sustainability statement, so the SLLP’s condition that the borrower has made every reasonable effort to set them before signing would not have been met, and the facility could not have been presented as sustainability-linked in the meantime.
Targets for the extension years. If both extension options are exercised the facility runs to November 2033, two financial years beyond the last agreed SPT. The agreement therefore contains a rendez-vous: SPTs for FY2031 and FY2032 must be agreed with the majority lenders before the first extension takes effect, no less ambitious than the linear trajectory. If they are not, the facility is declassified at the original maturity and continues without ratchet. The August 2026 provisions include an amendment process of this kind with a longstop date.
A larger step. One lender proposed 10% of the margin, that is 7 bp. The treasurer accepted the discount and refused the premium; the lenders refused an asymmetric grid. The step stayed at 3 bp, in line with the FCA’s observation that adjustments have not grown since the market began.
Question. How many euros does the grid move, at realistic utilisation?
Calculation. One basis point of margin adjustment is worth 1 bp on the drawn amount and 0.35 bp on the undrawn amount:
is the annual value of 1 bp of adjustment; the facility size; the utilisation; the commitment fee share (35%). At 20% utilisation: .
Result. With all three SPTs met, Borealis saves 3 bp, that is EUR 108,000 a year: 4.0% of the running cost of the facility and 1.4 bp of group EBITDA. With none met it pays EUR 108,000 more. The whole swing between a perfect and a failed sustainability year is EUR 216,000.
| Average utilisation | No adjustment | All three met | None met |
|---|---|---|---|
| 0% (pure back-up) | EUR 1.838m | EUR 1.759m | EUR 1.916m |
| 20% (base case) | EUR 2.670m | EUR 2.562m | EUR 2.778m |
| 50% (acquisition bridge) | EUR 4.294m | EUR 4.142m | EUR 4.446m |
Intermediate outcomes at 20%: EUR 2.634m with two SPTs met, EUR 2.706m with one. The table below widens the view to the range of steps seen in the market.
| Maximum adjustment | 0% drawn | 20% drawn | 50% drawn | 100% drawn |
|---|---|---|---|---|
| 2.5 bp | EUR 66k | EUR 90k | EUR 127k | EUR 188k |
| 3 bp (Borealis) | EUR 79k | EUR 108k | EUR 152k | EUR 225k |
| 5 bp | EUR 131k | EUR 180k | EUR 253k | EUR 375k |
INTERACTIVE LAB — What the margin ratchet is worth in euros per year
Translate a sustainability margin adjustment of a few basis points into euros for Borealis, for any utilisation and any number of SPTs met.
| Parameter you can move | Range | Default |
|---|---|---|
| Facility size | 100 to 2000 EUR m | 750 EUR m |
| Average utilisation | 0 to 100 % | 20 % |
| Drawn margin before adjustment | 20 to 300 bp | 70 bp |
| Commitment fee as share of adjusted margin | 0 to 50 % | 35 % |
| Maximum adjustment when all three SPTs are met or missed | 0 to 7.5 bp | 3 bp |
| Number of SPTs met this year | 0 / 1 / 2 / 3 | 3 |
| Borrower EBITDA | 100 to 3000 EUR m | 780 EUR m |
| Result at default settings | Value |
|---|---|
| Sustainability margin adjustment | -3.00 bp |
| Running cost without adjustment | 2.6700 EUR m |
| Running cost with adjustment | 2.5620 EUR m |
| Annual saving for the borrower (negative = extra cost) | 108.00 EUR k |
| Value of 1 bp of adjustment | 36.00 EUR k per year |
| Saving as share of running cost | 4.04 % |
| Saving as share of EBITDA | 1.38 bp |
| Swing between all met and all missed | 216.00 EUR k |
Try it. Set the commitment fee share to 0% to mimic a ratchet that applies to the drawn margin only: at 20% utilisation the saving falls from EUR 108k to EUR 45k.
Reading. Read the second table by column: the value of the ratchet depends as much on how drawn the facility is as on the size of the step, and a 5 bp step on an undrawn line is worth little more than a 3 bp step at 20% utilisation. A EUR 750m facility with a best-in-class ratchet moves about one hundred thousand euros a year for a group that earns EUR 780m of EBITDA. The biomass boiler conversion that delivers KPI 1 costs tens of millions. Nobody at Borealis will decide that investment because of the facility. If the ratchet applied to the drawn margin only, the saving at 20% utilisation would fall to EUR 45,000. The honest description of the mechanism is a signal with a price tag, not an incentive.
WHY IT MATTERS ON THE DESK — When a treasurer asks “what do I gain?”, the structurer must answer in euros, not in basis points, and must put the answer next to the cost of assurance and internal reporting. Banks that sold SLLs on the discount alone created the disappointment that shows in the 2025 and 2026 volumes.
Question. Over five years, with a plausible sequence of hits and misses, what is the ratchet worth after the cost of running it?
Data. Timing first. The financial year ends on 31 December; the certificate is due by 30 April; the adjustment applies from 1 May for twelve months. Signed on 2 November 2026, the facility therefore has no adjustment for its first six months, four full adjustment years (results of FY2026 to FY2029) and a final half-year from May to November 2031 (result of FY2030): 4.5 years of adjustment in a five-year life. The case outcome is the one replayed in the scorecard above: 3, 1, 2, 3 and 2 SPTs met, giving adjustments of -3, +1, -1, -3 and -1 bp. Costs for Borealis (case assumptions): EUR 60,000 at signing for the second-party opinion and additional legal work, and EUR 30,000 a year for KPI reporting and the verifier’s limited assurance on three KPIs. Discount rate 3.5%, close to Borealis’s marginal cost of drawn debt.
FORMULA
is the adjustment in period (negative for a discount); the annual value of 1 bp (EUR 36,000 at 20% utilisation); the length of the period in years (1, 1, 1, 1, 0.5); the time of the period’s mid-point from signing (1, 2, 3, 4 and 4.75 years); the discount rate; the one-off structuring cost; the annual verification cost, paid each spring.
Result. Savings and penalties add up to EUR 234,000 undiscounted (an average adjustment of -1.3 bp over the five years), with a present value of EUR 212,600, or 2.8 bp of the facility amount in total over five years. The present value of the structuring and verification costs is EUR 197,800. Net present value for Borealis: EUR 14,800. If every SPT is met every year the present value of savings is EUR 442,600 and the net EUR 244,800; if every SPT is missed the facility costs EUR 442,600 more than a plain RCF, before costs. Two sensitivities frame the result. On a facility that stays undrawn, the case path yields a present value of savings of EUR 155,000 and a net value of EUR -42,800: the structure costs more than it brings. With a 5 bp maximum step and a perfect record, the present value of savings reaches EUR 737,600, still less than 10 bp of the facility amount over five years.
INTERACTIVE LAB — Present value of the ratchet over the life of the facility
Discount the yearly savings and penalties of the ratchet over the five-year tenor, net of the cost of running the sustainability-linked structure.
| Parameter you can move | Range | Default |
|---|---|---|
| Average utilisation | 0 to 100 % | 20 % |
| Maximum adjustment (three KPIs) | 0 to 7.5 bp | 3 bp |
| KPI outcome | Case path / All SPTs met / All SPTs missed | Case path |
| Discount rate | 0 to 8 % | 3.5 % |
| Annual cost of KPI reporting and external verification | 0 to 150 EUR k | 30 EUR k |
| One-off structuring cost (second-party opinion, legal) | 0 to 300 EUR k | 60 EUR k |
| Result at default settings | Value |
|---|---|
| Value of 1 bp for one year | 36.00 EUR k |
| Average adjustment over the 5 years, time-weighted | -1.30 bp |
| Sum of savings, undiscounted | 234.0 EUR k |
| Present value of savings | 212.6 EUR k |
| Present value of structuring and verification costs | 197.8 EUR k |
| Net present value for the borrower | 14.8 EUR k |
| Present value of savings as share of facility size | 2.83 bp |
Try it. Keep the case path and set utilisation to 0%: on a pure back-up line the discount no longer covers the verifier, and the net present value turns negative (about EUR -43k).
Reading. On the case path the sustainability-linked structure is financially neutral for the borrower: the discount pays for the verifier and the second-party opinion and leaves about EUR 15,000 in five years. Because Borealis already publishes assured sustainability data, part of the EUR 30,000 would be spent anyway; a smaller, unlisted borrower would face a clearly negative value, the cost barrier that the FCA reported for smaller companies in 2025 (the LMA published a practice note on applying the SLLP to SMEs in September 2026). The reason to sign must lie elsewhere: discipline of annual external targets, consistency with what the company tells bond investors, and the lenders’ own preference. This is the same finding as the greenium of the Helios bond (see C01): real, small, and not the main reason.
Question. What does Meridian earn on EUR 75m, what capital does it hold, and what does the ratchet do to its return?
Data. Meridian’s share is 10% of every flow. Assumptions stated once: the desk pays Treasury an FTP liquidity premium of 30 bp on drawn amounts and a contingent liquidity charge of 5.6 bp on undrawn amounts (derived in section 6); operating costs are 40% of revenues before ratchet; Borealis’s internal rating corresponds to a one-year PD of 0.20%; Meridian uses the foundation IRB approach with the regulatory LGD of 40% and the CCF of 40%; capital is 13% of RWA; tax 25%.
Calculation.
| Line | Computation | EUR k |
|---|---|---|
| Net margin and utilisation fee on drawn | 15m x (70 + 10 - 30) bp | 75.0 |
| Commitment fee net of liquidity charge | 60m x (24.5 - 5.6) bp | 113.4 |
| Upfront fee spread over 5 years | 75m x 25 bp / 5 | 37.5 |
| Revenues | 225.9 | |
| Operating costs | 40% | -90.4 |
| Expected loss | 0.20% x 40% x 39m | -31.2 |
| Pre-tax result | 104.3 | |
| Tax at 25% | -26.1 | |
| Net result | 78.3 |
Capital: ; the F-IRB risk weight for a PD of 0.20% and an LGD of 40% is 39.02%; RWA are EUR 15.22m and allocated capital EUR 1.978m.
Result.
The amounts are in EUR thousand. Stand-alone, the facility returns a third of the 12% hurdle. With the full 3 bp discount Meridian gives up EUR 10,800 a year, 10% of what Borealis saves, and RAROC falls to 3.55%. With the full premium it rises to 4.37%. On the case path (average adjustment of -1.3 bp over the life) it is 3.78%.
INTERACTIVE LAB — Meridian’s RAROC on its EUR 75m hold: stand-alone, with ratchet, with cross-sell
Rebuild the lender’s revenue, expected loss, capital and RAROC line by line, then see what the ratchet removes and what ancillary business adds.
| Parameter you can move | Range | Default |
|---|---|---|
| Meridian final hold | 10 to 150 EUR m | 75 EUR m |
| Average utilisation | 0 to 100 % | 20 % |
| Sustainability margin adjustment | -7.5 to 7.5 bp | -3 bp |
| Ancillary revenues from the relationship | 0 to 1500 EUR k per year | 620 EUR k per year |
| One-year PD of the borrower | 0.05 to 2 % | 0.2 % |
| Risk weight used for capital | F-IRB / SA 75% / Floor-bound (72.5% of SA) | F-IRB |
| FTP liquidity premium on drawn amounts | 0 to 80 bp | 30 bp |
| Contingent liquidity charge on undrawn amounts | 0 to 15 bp | 5.6 bp |
| Result at default settings | Value |
|---|---|
| Net margin and utilisation fee on drawn amount, after FTP | 75.0 EUR k |
| Commitment fee after contingent liquidity charge | 113.4 EUR k |
| Upfront fee spread over 5 years | 37.5 EUR k |
| Revenues without ratchet | 225.9 EUR k |
| Effect of the ratchet on revenues | -10.8 EUR k |
| Operating costs (40% of revenues without ratchet) | 90.4 EUR k |
| Expected loss | 31.2 EUR k |
| Pre-tax result, no ratchet | 104.3 EUR k |
| Net result after 25% tax, no ratchet | 78.3 EUR k |
| Exposure at default | 39.00 EUR m |
| Risk weight | 39.02 % |
| Risk-weighted assets | 15.217 EUR m |
| Allocated capital (13% of RWA) | 1.978 EUR m |
| RAROC stand-alone, no ratchet | 3.96 % |
| RAROC stand-alone, with ratchet | 3.55 % |
| Relationship RAROC, with ratchet and cross-sell | 11.75 % |
| Relationship RAROC, no ratchet | 12.12 % |
| Ancillary revenues needed to reach 12% with ratchet | 640.7 EUR k |
Try it. Switch the risk weight to ‘SA 75%’: the same facility consumes almost twice the capital. Then raise utilisation to 100% and watch the stand-alone RAROC improve while staying below the hurdle.
Sensitivities. The credit committee asks for the same calculation under other capital regimes and utilisations, all from the same lab.
| Scenario | RAROC, no ratchet | RAROC, full discount |
|---|---|---|
| F-IRB, 20% drawn (base) | 3.96% | 3.55% |
| Standardised approach, 20% drawn | 2.06% | 1.84% |
| Output-floor bound, 20% drawn | 2.84% | 2.55% |
| F-IRB, undrawn | 4.12% | 3.73% |
| F-IRB, fully drawn | 6.36% | 5.92% |
A lender on the standardised approach holds almost twice the capital for the same ticket and earns half the return, which makes IRB banks the natural holders of such facilities. Full drawing improves the stand-alone return, thanks to the 40 bp utilisation fee, without bringing it near the hurdle; and a fully drawn back-up line usually comes with a deteriorating rating.
Cross-sell. The coverage banker’s account plan (see B01) lists the ancillary revenues that Meridian expects from Borealis each year as a member of the relationship group: cash management and payments EUR 150,000, foreign exchange and interest-rate hedging EUR 180,000, a share of bond bookrunner fees averaging EUR 220,000, agency fee EUR 40,000, guarantees EUR 30,000. Total: EUR 620,000. These businesses run at a 60% cost-income ratio and consume little capital (the case assumes EUR 2.5 of RWA per euro of annual revenue, for operational and counterparty risk). That ratio suits payments, cash management and DCM-type fees, not derivatives, which consume far more capital: the hedging package of C07 needs about EUR 28 of RWA per euro of revenue, so applying 2.5 to the EUR 180,000 of hedging revenue flatters the relationship figures below. Adding them gives a relationship RAROC of 12.12% without ratchet and 11.75% with the full discount. To reach exactly 12% with the full discount Meridian needs EUR 641,000 of ancillary revenues, EUR 21,000 more than the plan.
Reading. Three lessons, which apply to every relationship facility on the desk. The RCF does not pay for itself, with or without the ratchet: it is an entry ticket. The 3 bp that were negligible for the borrower move the relationship from just above to just below the hurdle, because the lender’s return is a thin residual; this is why credit committees ask who “pays” for the discount. And the approval rests on revenues that are not contractual. The Friday review exists to check, a year later, whether the EUR 620,000 materialised.
PITFALL — Do not present the premium for missed SPTs as upside in the credit application. A lender that budgets for its client’s failure to meet climate targets has a conflict of interest that classification committees and supervisors look for. Meridian’s RAROC is approved on the full-discount case.
Question. What exactly happens when a target is missed, and when the group changes shape?
The FY2027 test. In February 2028 Borealis closes its accounts and its sustainability statement. The statutory auditor, who already gives limited assurance on the sustainability statement, is engaged as external verifier for the three KPIs. Results: scope 1 and 2 emissions 795 kt against an SPT of 790 kt (missed, a boiler conversion slipped by five months); recycled content 26.5% against 26% (met); supplier coverage 41% against 42% (missed). On 20 April 2028 the CFO signs the sustainability compliance certificate and the treasurer sends it to the agent with the verification report. Meridian’s agency team checks the form, not the substance: the certificate is signed by an authorised officer, covers each KPI, and the verifier is one permitted by the agreement. The agent posts both documents to the lenders and issues a margin notice: one SPT met, adjustment +1 bp, margin 71 bp and commitment fee 24.85 bp from 1 May 2028.
Consequence. At 20% utilisation the miss costs Borealis EUR 36,000 over the next twelve months, of which Meridian receives EUR 3,600 (10%). In the agent’s loan system the change is one line of static data, applied to every interest period and to the quarterly commitment fee from 1 May. There is no default, no waiver request, no drawstop. The facility remains sustainability-linked: a missed ambitious target is not a failure of the structure, as the FCA noted in 2025. What lenders watch is the pattern. Two consecutive misses on KPI 1 would prompt a remediation discussion (see B15); the August 2026 LMA provisions allow the parties to define a “prolonged” breach as a declassification trigger.
Fallbacks. If the verifier cannot sign in time, the certificate is late and the +3 bp default adjustment applies until delivery. If a figure is later restated and a discount proves undue, the borrower repays the difference. If a KPI can no longer be calculated (a methodology is withdrawn, a data source disappears), the amendment process starts and, failing agreement, the KPI is treated as missed or the facility is declassified, depending on the drafting.
An acquisition. Suppose that in FY2028 Borealis buys a flexible-packaging business that emits 120 kt a year. Without any change to the documents, group emissions jump to 865 kt against an SPT of 748 kt, and KPI 1 fails in FY2028, FY2029 and FY2030: over the five years only one test out of five is passed although the original plants perform exactly as planned. The acquisition is a sustainability amendment event. Following GHG Protocol practice, the base year is recalculated to include the acquired business (150 kt in FY2022), giving a base of 1,150 kt, and the same 4.2% a year yields an SPT of 860 kt for FY2028. Borealis still misses, by 5 kt instead of 117 kt, because the business it acquired was decarbonising more slowly than the group. The rebasing does not rescue the borrower; it restores a fair test. The amendment needs majority-lender consent; if the parties cannot agree within the negotiation period, lenders may declassify.
WHO DOES WHAT — Borealis treasury prepares the certificate; the external verifier signs the verification report; the facility agent checks completeness, distributes, and reprices; each lender’s portfolio manager updates the internal sustainable-finance flag; the sustainability coordinator leads any amendment discussion; the agent runs the vote. Nobody at the lenders re-audits the KPI figures.
| Figure | Value | Source stage |
|---|---|---|
| Facility | EUR 750m, 5+1+1, margin 70 bp, commitment fee 24.5 bp | 2.2 |
| Annual running cost at 20% utilisation | EUR 2.67m (EUR 3.045m with upfront fee) | 2.2 |
| Cost on commitment / per euro drawn | 40.6 bp / 203 bp | 2.2 |
| KPI 1 SPTs FY2026 to FY2030 | 832, 790, 748, 706, 664 ktCO2e | 5.2 |
| KPI 1 FY2030 SPT versus base year / versus business as usual | -33.6% / -152 kt | 5.2 |
| KPI 2 FY2030 SPT versus weighted legal minimum | 38% versus 23.75% | 5.2 |
| Ratchet | +/-1 bp per KPI, maximum +/-3 bp, applies to commitment fee | 5.3 |
| Value of 1 bp at 20% utilisation | EUR 36,000 a year | 5.4 |
| All SPTs met / swing best to worst | EUR 108,000 / EUR 216,000 a year | 5.4 |
| Saving as share of running cost / of EBITDA | 4.0% / 1.4 bp | 5.4 |
| Case path: adjustments | -3, +1, -1, -3, -1 bp; average -1.3 bp | 5.5 |
| PV of savings / of costs / net | EUR 212,600 / EUR 197,800 / EUR 14,800 | 5.5 |
| Meridian revenues on EUR 75m hold | EUR 225,900 a year | 5.6 |
| EAD / RWA / capital | EUR 39m / EUR 15.22m / EUR 1.978m | 5.6 |
| RAROC stand-alone: no ratchet / full discount / full premium | 3.96% / 3.55% / 4.37% | 5.6 |
| Relationship RAROC with EUR 620k cross-sell: no ratchet / full discount | 12.12% / 11.75% | 5.6 |
| Ancillary revenues needed for 12% with full discount | EUR 641,000 | 5.6 |
| FY2027 test | 1 of 3 met, +1 bp, EUR 36,000 extra cost | 5.7 |
IN ONE SENTENCE — An undrawn commitment consumes capital because regulation assumes that a fixed share of it, the credit conversion factor, will have been drawn when the borrower defaults.
CRR3 (Regulation (EU) 2024/1623) has applied since 1 January 2025. Under the standardised approach a committed corporate RCF falls in the bucket of “other commitments”, with a CCF of 40% whatever its maturity; the previous rules gave 50% above one year and 20% up to one year, so the advantage of 364-day facilities has gone. Only unconditionally cancellable commitments receive 10%, with a transitional arrangement that keeps the effective factor at zero until the end of 2029 and phases it in by 2033; a syndicated committed facility is not one of them.
The foundation IRB approach now uses the standardised CCFs: 40% as well, where the previous F-IRB factor for credit lines was 75%. Own estimates of LGD and CCF are no longer allowed for corporates in groups with consolidated revenues above EUR 500m, so Borealis is an F-IRB exposure by construction. The regulatory LGD for senior unsecured exposures to non-financial corporates is 40% (45% before), the PD input floor is 0.05%, and the 1.06 scaling factor has been removed.
FORMULA
is the capital requirement per euro of exposure; the standard normal distribution function and its inverse; the one-year probability of default; the asset correlation, which falls from 24% to 12% as PD rises; the maturity adjustment; the maturity, fixed at 2.5 years in Meridian’s F-IRB set-up: under Article 162(1) CRR as amended by Regulation (EU) 2024/1623, a bank without own-LGD permission applies M consistently, either fixed at 2.5 years or calculated under Article 162(2) for its whole book, and Meridian has chosen the fixed value (were M calculated, this five-year facility would weigh about 60%).
CASE — For Meridian’s hold, EAD is EUR 39m. F-IRB: risk weight 39.02%, RWA EUR 15.22m, capital EUR 1.978m, or about 203 bp of capital per euro of undrawn commitment. Standardised approach: BBB corporates are weighted 75%, RWA EUR 29.25m, capital EUR 3.80m. A bank bound by the output floor (72.5% of standardised RWA when fully phased in 2030; 55% in 2026) counts EUR 21.21m at the margin. With the pre-2025 F-IRB factor of 75%, EAD would have been EUR 60m and capital EUR 3.04m.
INTERACTIVE LAB — Capital for an RCF under CRR3: standardised approach against F-IRB
Compute exposure at default with the credit conversion factor, then RWA and capital under the standardised approach and under the foundation IRB formula.
| Parameter you can move | Range | Default |
|---|---|---|
| Commitment of the bank | 10 to 150 EUR m | 75 EUR m |
| Utilisation | 0 to 100 % | 20 % |
| Credit conversion factor on the undrawn amount | 10 / 40 / 50 / 75 / 100 | 40 |
| One-year PD (F-IRB) | 0.05 to 2 % | 0.2 % |
| LGD (F-IRB senior unsecured corporate: 40%) | 25 to 45 % | 40 % |
| Maturity M in the IRB formula | 1 to 5 years | 2.5 years |
| Standardised risk weight (BBB corporate: 75%) | 20 to 150 % | 75 % |
| Result at default settings | Value |
|---|---|
| Exposure at default | 39.00 EUR m |
| F-IRB risk weight | 39.02 % |
| RWA under F-IRB | 15.217 EUR m |
| RWA under the standardised approach | 29.250 EUR m |
| 72.5% of standardised RWA (fully phased output floor) | 21.206 EUR m |
| Allocated capital under F-IRB (13%) | 1.978 EUR m |
| Allocated capital under the standardised approach (13%) | 3.803 EUR m |
| F-IRB capital held per euro of undrawn commitment | 202.9 bp |
Try it. Set the CCF to 75%, the F-IRB value that applied to this kind of commitment before 2025, and compare the capital with the 40% of CRR3.
Nothing here depends on the sustainability-linked label: CRR3 has no green supporting factor (see E02), and the evidence on PD points to selection, not causation (section 8).
The table and sensitivities of section 5.6 appear as such in the credit application. The sustainable feature changes two lines. Revenues fall by up to EUR 10,800 a year. Costs rise slightly (coordinator’s time, annual certificate processing), although Meridian, like most banks, does not charge them to the transaction. Capital, expected loss and funding do not move. The discount is therefore financed out of the lender’s return, unless internal pricing grants an allowance such as a lower hurdle or an FTP rebate for classified assets (see E03). Meridian grants none.
INTUITION — A committed line is a promise to hand over cash on three days’ notice. To keep it in a crisis, Treasury holds a cushion of liquid assets today, and liquid assets yield less than they cost to fund. The desk that sold the promise pays for the cushion.
The Liquidity Coverage Ratio (Delegated Regulation (EU) 2015/61, Article 31) assumes that, in a 30-day stress, non-financial corporates draw 10% of undrawn committed credit facilities and 30% of undrawn committed liquidity facilities. A liquidity facility is a back-up line for debt the client cannot roll over in the market; its amount is the backed debt maturing within 30 days. The Borealis facility is both. The case assumes EUR 120m of commercial paper maturing within 30 days on average, 20% of the EUR 600m undrawn, so the blended outflow rate is . On Meridian’s EUR 60m undrawn share, Treasury holds EUR 8.4m of high-quality liquid assets. With a net carry cost of 40 bp (a case assumption), the annual cost is EUR 33,600, or 5.6 bp on the undrawn amount: almost a quarter of the 24.5 bp commitment fee. The 5% stable-funding requirement of the NSFR on undrawn amounts is included in the same charge.
INTERACTIVE LAB — Contingent liquidity cost of the undrawn line
Convert the LCR outflow rates on undrawn committed facilities into a liquidity buffer and an annual cost charged by Treasury through FTP.
| Parameter you can move | Range | Default |
|---|---|---|
| Commitment of the bank | 10 to 150 EUR m | 75 EUR m |
| Utilisation | 0 to 100 % | 20 % |
| Share of the undrawn amount classified as liquidity facility (30% outflow) | 0 to 100 % | 20 % |
| Net cost of carrying HQLA | 0 to 100 bp | 40 bp |
| FTP liquidity premium on drawn amounts | 0 to 80 bp | 30 bp |
| Result at default settings | Value |
|---|---|
| Undrawn commitment | 60.00 EUR m |
| Blended LCR outflow rate on the undrawn amount | 14.0 % |
| HQLA buffer required | 8.40 EUR m |
| Annual contingent liquidity cost | 33.6 EUR k |
| Contingent liquidity charge on the undrawn amount | 5.60 bp |
| Share of the 24.5 bp commitment fee absorbed | 22.9 % |
| FTP liquidity premium paid on drawn amounts | 45.0 EUR k |
Try it. Push the liquidity-facility share to 100%: a pure commercial paper back-up line costs three times more liquidity than a working-capital line of the same size.
On the drawn part the desk pays an FTP liquidity premium of 30 bp, or EUR 45,000 a year on EUR 15m (see E04). The sustainability-linked format brings no funding advantage: Meridian’s own green bonds require identified green assets, and a general-purpose SLL is not eligible (see C15).
PITFALL — A structurer who prices the commitment fee against capital only overstates the revenue on the undrawn line by a quarter. A pure commercial paper back-up line, classified entirely as a liquidity facility, would carry a charge of 12 bp instead of 4 bp.
Classification. A loan is measured at amortised cost only if it is held to collect and its cash flows are solely payments of principal and interest (SPPI). For years auditors debated whether a margin that moves with emissions is “interest”. The IASB settled the matter with its Amendments to the Classification and Measurement of Financial Instruments, issued on 30 May 2024, effective for annual periods beginning on or after 1 January 2026 and endorsed in the EU by Regulation (EU) 2025/1047 of 27 May 2025. Under the new paragraph B4.1.10A, a contingent feature unrelated to basic lending risks, such as an emissions target, does not prevent SPPI if, in all contractually possible scenarios, the cash flows would not be significantly different from those of an identical instrument without the feature.
CASE — The Borealis ratchet moves the margin by at most 3 bp on EURIBOR plus 70 bp: the drawn loans pass SPPI and are measured at amortised cost. The amendments add disclosures (IFRS 7 paragraphs 20B to 20D): the contingent event, the range of possible changes in cash flows, and the gross carrying amount concerned. Finance therefore needs the sustainability flag and the grid from the loan system.
Commitment, provisioning and fees. The undrawn commitment is off balance sheet but within the impairment rules: Meridian books a 12-month expected credit loss (Stage 1) on the amount it expects to be drawn at default, presented as a provision, and moves to lifetime losses (Stage 2) if credit risk increases significantly. The EUR 31,200 expected loss of the RAROC is a proxy for the Stage 1 amount; a missed SPT is not in itself a significant increase in credit risk. The commitment fee is recognised over time; the upfront fee, when drawing is not probable, is spread on a straight line over the commitment period, as in the RAROC (see E05).
Green Asset Ratio. For a loan with known use of proceeds, the bank looks at the financed activity. For a general-purpose loan, it applies the counterparty’s published Taxonomy KPIs (aligned share of turnover, and of capex) to the gross carrying amount. The sustainability-linked label plays no role. With Borealis reporting 9% aligned turnover and 21% aligned capex (case assumptions), Meridian’s EUR 15m drawn exposure contributes EUR 1.35m to the turnover-based numerator and EUR 3.15m to the capex-based one; the EUR 60m undrawn commitment contributes nothing. Delegated Regulation (EU) 2026/73, applicable from 1 January 2026, removed exposures to undertakings outside the CSRD from the denominator and lets financial undertakings that claim no alignment omit detailed Taxonomy KPIs until the end of 2027. The EBA’s final draft Pillar 3 standards of June 2026 delete the GAR and BTAR templates, with a first reference date of 31 December 2026 (see E06).
Financed emissions. Under the PCAF standard (Part A, third edition, December 2025), a corporate loan is a “business loan”: the bank is attributed a share of the borrower’s emissions equal to its outstanding amount divided by the enterprise value including cash of a listed borrower. Meridian’s attribution factor is , giving 1,886 tCO2e of scope 1 and 2 financed emissions and 7,286 tCO2e of scope 3. Only amounts on the balance sheet at year-end count (the third edition adds an optional method for undrawn commitments), so a facility repaid in December disappears from the inventory (see E07).
INTERACTIVE LAB — What the facility adds to Meridian’s Green Asset Ratio and financed emissions
Apply the general-purpose rule of the Green Asset Ratio and the PCAF attribution factor to the drawn part of Meridian’s hold.
| Parameter you can move | Range | Default |
|---|---|---|
| Meridian final hold | 10 to 150 EUR m | 75 EUR m |
| Utilisation at the reporting date | 0 to 100 % | 20 % |
| Borealis enterprise value including cash | 2000 to 15000 EUR m | 7000 EUR m |
| Borealis scope 1 + 2 emissions | 100 to 2000 ktCO2e | 880 ktCO2e |
| Borealis scope 3 emissions | 0 to 10000 ktCO2e | 3400 ktCO2e |
| Taxonomy-aligned share of turnover | 0 to 100 % | 9 % |
| Taxonomy-aligned share of capex | 0 to 100 % | 21 % |
| Result at default settings | Value |
|---|---|
| On-balance-sheet exposure | 15.00 EUR m |
| Taxonomy-aligned amount, turnover basis | 1.350 EUR m |
| Taxonomy-aligned amount, capex basis | 3.150 EUR m |
| PCAF attribution factor | 0.2143 % |
| Financed emissions, scope 1 + 2 | 1,886 tCO2e |
| Financed emissions, scope 3 | 7,286 tCO2e |
| Scope 1 + 2 financed emissions per EUR m lent | 125.7 tCO2e |
Try it. Set utilisation to 0% at the reporting date: a EUR 75m sustainability-linked commitment contributes nothing to the Green Asset Ratio and nothing to financed emissions.
Sustainable finance targets. Banks decide by internal policy whether SLLs count towards their published “sustainable finance mobilised” targets; the FCA observed in 2025 that some do not count them. Meridian counts an SLL only if its classification committee has validated the KPIs against the SLLP, and removes it on declassification. Inside the bank, this is the real stake of the label.
| Risk | What can go wrong | Control | Owner |
|---|---|---|---|
| Credit and drawdown | Line fully drawn before default | Rating, covenants, CCF in capital, annual review | Credit risk |
| Relationship return | Cross-sell does not materialise | Annual profitability review, wallet tracking | Coverage |
| Weak KPIs or SPTs | Targets already achieved or immaterial | Classification committee, benchmarks, second-party opinion | Sustainable finance solutions |
| Conflict of interest | Volume target or premium income drives lax structuring | Classification independent of origination; RAROC approved on discount case | Risk and Compliance |
| Operational | Wrong margin after the test, late certificate unnoticed | Static data four-eyes check, diary of due dates | Loan agency |
| Structural change | Acquisition or methodology change makes KPI meaningless | Amendment event, rendez-vous clause, rebasing rule | Structurer and Legal |
| Disclosure | Bank or borrower publicises a declassified or sleeping facility as sustainable | Publicity clause, marketing sign-off | Compliance |
| Accounting | Feature fails SPPI or disclosures missing | B4.1.10A assessment, flag in loan system | Finance |
| Liability | Lenders rely on the coordinator’s work | Narrow mandate, exculpation, no verification duty | Legal |
Red-flag checklist for a structurer, a credit analyst or a lender invited into a deal:
Volumes: peak, contraction, signs of stabilisation. According to Environmental Finance Data, published in the LMA’s Horizons in March 2026, global sustainable loan volume was USD 749bn in 2025, 22% below the record of 2024, a fall of USD 209bn. Sustainability-linked loans were still 52% of the value, but they explain the whole decline: SLL volume fell by USD 279bn and the number of SLL deals by 42%, to 486, while green loans grew. The first half of 2026 was weaker again, at about USD 235bn globally, with Asia’s share rising to about a quarter (Environmental Finance, September 2026). In EMEA, LSEG LPC data quoted in Horizons in June 2026 show sustainable loan volume of USD 62.35bn in the first quarter of 2026, against USD 77.48bn a year earlier and a peak of USD 136.9bn in the first quarter of 2024. The same article reports that in April 2026 Ford removed the sustainability-linked pricing from its USD 18bn of revolving facilities.
Why the contraction. The sources give three reasons: the benefit does not cover the cost of structuring and reporting; lenders scrutinise ambition and decline to label weak structures; and US companies have become cautious about labelled finance.
Supervisory statements. The UK Financial Conduct Authority wrote to heads of sustainable finance on 29 June 2023 after reviewing the market. It reported weak incentives (adjustments of the order of 2.5 bp for investment-grade borrowers), potential conflicts of interest where bankers are rewarded on sustainable finance volumes, and doubts about the robustness of KPIs; one firm had told the FCA that, of some 250 SLL transactions completed in 2022, it considered only about 30% fit for purpose. Its follow-up letter of 14 August 2025 found a more mature market (two or three core KPIs, more scrutiny, occasional declassifications) but noted that margin adjustments remained de minimis and that higher standards might reduce volumes for a time.
Quality as seen by reviewers. Moody’s, quoted by Environmental Finance in September 2026, reports that just over half of the sustainability-linked loans it has reviewed received one of its two top sustainability quality scores, against about 90% for use-of-proceeds loans.
Academic evidence. The literature is young and mostly sceptical about incentive effects.
Open debates in November 2026. Should the step be scaled to the margin? Should premiums be ring-fenced? Should SLLs count in banks’ sustainable finance targets? Will transition loans, now with their own Principles, replace SLLs for high-emitting borrowers? The defensible answer from the desk is that the SLL is a monitoring and dialogue device: it puts three verified numbers in front of twelve banks every spring. Its price effect is small by design.
INTERACTIVE LAB — Self-assessment: sustainability-linked RCFs
Fourteen questions on mechanics, standards, economics and the bank’s internal treatment.
Q1. At 20% utilisation, which component is the largest part of Borealis’s running cost of EUR 2.67m?
Answer: B. The commitment fee is EUR 1.47m (EUR 600m at 24.5 bp) against EUR 1.05m of drawn margin and EUR 0.15m of utilisation fee.
Q2. Which statement defines a sustainability-linked loan?
Answer: B. An SLL is general-purpose; what is linked to sustainability is the pricing or structure, through KPIs and SPTs.
Q3. Under the SLLP of March 2025, which of these is mandatory?
Answer: C. Post-signing verification is a ‘shall’. Pre-signing external review is recommended only, and there is no minimum number of KPIs.
Q4. What is the credit conversion factor for a committed, not unconditionally cancellable corporate RCF under CRR3, in both the standardised and the F-IRB approaches?
Answer: C. CRR3 sets 40% for other commitments whatever the maturity and aligns F-IRB on the standardised CCFs; 75% was the old F-IRB factor.
Q5. Meridian holds EUR 75m, 20% drawn. What is its exposure at default?
Answer: B. EAD = 15 + 40% x 60 = EUR 39m.
Q6. With all three SPTs met and 20% utilisation, how much does Borealis save per year?
Answer: B. 3 bp on EUR 150m drawn (EUR 45,000) plus 1.05 bp on EUR 600m undrawn (EUR 63,000).
Q7. Why do lenders prefer a per-KPI grid to a ‘two out of three’ aggregate grid?
Answer: C. Aggregate scoring allows offsetting between KPIs; per-KPI scoring prices each target.
Q8. The compliance certificate is not delivered by the due date. Under the Borealis documentation, what happens?
Answer: C. Non-delivery is treated as all SPTs missed. A sustainability breach is not an event of default under the LMA provisions.
Q9. What is a sleeping SLL?
Answer: B. The 2025 SLLP allow it with a rationale and a backstop date of at most twelve months after origination.
Q10. How does Meridian’s drawn exposure to Borealis enter the Green Asset Ratio numerator?
Answer: C. A general-purpose exposure is weighted by the counterparty’s own Taxonomy KPIs; the SLL label is irrelevant.
Q11. Under the 2024 amendments to IFRS 9 (effective 2026), an ESG-linked margin adjustment is compatible with SPPI if…
Answer: B. This is the test of paragraph B4.1.10A, endorsed in the EU by Regulation 2025/1047.
Q12. Which LCR outflow rate applies to the part of an undrawn corporate line that backs commercial paper maturing within 30 days?
Answer: C. That part is a liquidity facility (30%); the rest is a credit facility (10%).
Q13. Meridian’s stand-alone RAROC on the hold is about 4%. Why does the credit committee approve?
Answer: B. The RCF is a loss-leader; EUR 620,000 of expected cross-sell brings the relationship to about 12%.
Q14. According to Environmental Finance Data, what happened to global sustainability-linked loan volume in 2025?
Answer: C. Sustainable loans fell 22% to USD 749bn; SLLs, still 52% of the total, accounted for the whole decline.
Tessera Glass has a EUR 400m RCF: margin 110 bp, commitment fee 40% of margin, utilisation fee 10 / 20 / 40 bp on the same tiers as in the case. Average utilisation is 45%. The ratchet has two KPIs, each worth 2.5 bp either way, and applies to the commitment fee. Compute the running cost, and the saving if both SPTs are met.
Solution. Drawn EUR 180m, undrawn EUR 220m. Margin: 180m x 110 bp = EUR 1.980m. Utilisation fee (second tier, 20 bp): EUR 0.360m. Commitment fee: 44 bp x 220m = EUR 0.968m. Running cost: EUR 3.308m. Value of 1 bp: 400m x (0.45 + 0.55 x 0.40) x 0.0001 = EUR 26,800. With both SPTs met (-5 bp) the saving is EUR 134,000, or 4.05% of the running cost. With one met and one missed the adjustment is zero.
A bank holds EUR 50m of the Tessera facility (12.5%). FTP liquidity premium on drawings 45 bp; contingent liquidity charge on undrawn 6 bp; upfront fee 35 bp spread over 5 years; cost-income 45%; PD 0.50%; LGD 40%; CCF 40%; F-IRB risk weight 62% (given); capital 13% of RWA; tax 25%. Compute the stand-alone RAROC without ratchet and with both SPTs met, and the ancillary revenues (cost-income 60%, capital ignored) needed to reach 12% in the second case.
Solution. Drawn EUR 22.5m, undrawn EUR 27.5m. Revenue on drawn: 22.5m x (110 + 20 - 45) bp = EUR 191,250. Commitment fee net: 27.5m x (44 - 6) bp = EUR 104,500. Upfront: EUR 35,000. Revenues EUR 330,750; costs EUR 148,838. EAD = 22.5 + 0.4 x 27.5 = EUR 33.5m; expected loss EUR 67,000; RWA EUR 20.77m; capital EUR 2.700m. RAROC = (330,750 - 148,838 - 67,000) x 0.75 / 2,700,100 = 3.19%. The bank’s share of the discount is 12.5% x 134,000 = EUR 16,750, and RAROC falls to 2.73%. To reach 12% the pre-tax result must be 0.12 x 2,700,100 / 0.75 = EUR 432,016: ancillary business must add EUR 333,854 pre-tax, that is EUR 834,634 of revenues.
A EUR 300m five-year bullet term loan, fully drawn, has a discount-only ratchet of 4 bp if both of two SPTs are met, tested for years 2 to 5 (cash flows at the end of each of those years). The treasurer estimates a 70% probability of meeting both in any year. Costs: EUR 50,000 at signing and EUR 25,000 at the end of each of the five years. Discount rate 4%. Compute the expected net present value and the break-even probability.
Solution. Full discount: 300m x 4 bp = EUR 120,000 a year. Annuity factor for years 2 to 5 at 4%: 0.9246 + 0.8890 + 0.8548 + 0.8219 = 3.4903. PV of the discount if always earned: EUR 418,834; expected PV at 70%: EUR 293,184. Costs: 50,000 + 25,000 x 4.4518 = EUR 161,296. Expected NPV: EUR 131,888. Break-even probability: 161,296 / 418,834 = 38.5%. On a fully drawn loan each basis point is worth far more than on an RCF; and with a one-way ratchet the lender expects to lose 4 bp most years with no compensation, which is why lenders resist this structure.
1. Explain a revolving credit facility to a non-banker, and say how the bank is paid. It is a commitment by banks to lend up to a limit whenever the company asks, for a set number of years, with repaid amounts available again. The company pays a margin over EURIBOR on what it borrows and a commitment fee, usually 35% of the margin, on what it does not, plus small upfront and utilisation fees. For an investment-grade company the line is liquidity insurance and stays largely undrawn, so the bank earns little while holding capital and liquidity against the promise.
2. What are the five components of the Sustainability-Linked Loan Principles? Selection of KPIs, calibration of SPTs, loan characteristics, reporting and verification. KPIs must be material and measurable; SPTs must be ambitious, beyond business as usual and beyond regulation; the loan terms must change with performance; the borrower reports at least annually; and an external party verifies each period that can change the pricing. The current version is dated 26 March 2025.
3. A treasurer asks what a 3 bp ratchet is worth on a EUR 750m RCF. What do you answer? It depends on utilisation and on whether the commitment fee follows the margin. At 20% drawn with a 35% commitment fee, 1 bp is worth EUR 36,000 a year, so the full discount is EUR 108,000. Against this the borrower pays for verification and reporting. The structure is roughly neutral in cash terms; the reasons to do it are discipline, consistency with public commitments and lender appetite.
4. Why is a relationship RCF a loss-leader, and how do you defend it in committee? Because margins on investment-grade lines are set by competition between relationship banks, not by the cost of capital: in our case the stand-alone RAROC is about 4% against a 12% hurdle. The defence is the relationship return: lenders get access to hedging, cash management, bond and advisory mandates. I would show the account plan with named ancillary revenues, the relationship RAROC and last year’s track record.
5. How does CRR3 treat the undrawn part of the facility? The undrawn commitment becomes exposure through a credit conversion factor of 40%, under the standardised approach and under foundation IRB alike. The exposure then receives the corporate risk weight: 75% for a BBB under the standardised approach, or the IRB formula with a 40% regulatory LGD. Only unconditionally cancellable commitments get 10%, and a syndicated committed facility is not one.
6. Does a sustainability-linked loan improve the bank’s Green Asset Ratio? Not as such. The GAR looks at use of proceeds or, for general-purpose lending, at the borrower’s Taxonomy-aligned turnover and capex shares applied to the drawn amount. An SLL to a company with 9% aligned turnover counts for 9% of the outstanding, like a plain loan. The label matters only for the bank’s own sustainable finance target, an internal definition.
7. A borrower misses its emissions SPT two years running. What do you do as sustainability coordinator? A miss triggers the premium, not a default. I would ask the borrower for an explanation and a remediation plan, and check whether the cause is performance or a change of perimeter or methodology, in which case the amendment mechanism applies. If the facility defines a prolonged breach as a declassification trigger, lenders will have to vote. At refinancing I would not roll the structure over unchanged.
8. What is the main criticism of SLLs and how much of it do you accept? That the incentive is too small to change behaviour and that targets were often unambitious, so the label rewarded what companies would have done anyway; the FCA’s 2023 letter and several academic papers support this. I accept most of it. What remains is a governance tool, annual verified targets in a contract with a dozen banks, whose credibility depends on per-KPI two-way grids, science-based trajectories and lenders willing to declassify.
Standards and market documentation
Regulation and accounting
Supervisory statements
Market data
Academic papers