Sustainable loans fell 18% because the label stopped pricing risk

Sustainable loan volume is down 18% year over year, and the decline is being read as a retreat from transition finance; follow the money to eastern Washington instead, where Twelve has repriced $45 million of debt against its Moses Lake power-to-liquid plant, and the drop reads more like a relocation. The construction loan became debt against an operating asset, the point at which an e-fuels project stops being a technology story and becomes a facility with a known output, a power bill and a spread. The price agreed there was set by plant economics rather than by a margin ratchet tied to a corporate target, and it is the number the next developer will be measured against.

The two developments describe one movement: borrowers are routing around the KPI scaffolding that defines the sustainability-linked loan, moving transition assets into private credit and unlabeled project finance, where credit is underwritten against the asset in front of the lender rather than against the issuer's published commitments. A labeled loan carries obligations — a target, verification that it was met, disclosure when it was not — and what those obligations return is a corporate-level promise that says very little about whether one specific plant will service its debt. On a conversion like Twelve's, that trade has stopped clearing.

This is a change in who holds the pen on price: an SLL is negotiated between a borrower's treasury team and a bank's coverage group, and the grid that comes out of it reflects the issuer's ambition as much as its engineering. Repricing a plant is a conversation between lenders who have read an operating history and a sponsor who has to show what the facility consumes, produces and earns, and those two conversations produce different numbers — the second is the one an allocator can underwrite.

None of this means the banks have walked away. The plausible read is that the transition relationship now sits with the project finance desk rather than the sustainable finance group, because the questions that decide the credit — construction schedule, operating history, the durability of what the plant sells — are the questions that desk already asks. A KPI-linked corporate facility has to be negotiated with a sustainability team and reported on annually to investors who may care more about the marker than the margin.

Watch the price on the way out. A labeled loan may have carried a concession for the marker, with issuers willing to publish a target and accept a step-up negotiating tighter spreads than the same credit without the framework, and the move to asset-level debt will test whether that concession was real. If it was, first-of-a-kind plants and pre-scale process companies will pay more for their money than they would have under a label — uncomfortable for developers and clarifying for everyone else, because the assets that clear their hurdle rate without the marker were never relying on it.

Construction-to-operating conversion rarely makes headlines, yet it is the point at which a project's economics become checkable: while a plant is being built, lenders are underwriting completion — whether the contractor delivers, whether the equipment arrives, whether the process holds at commercial scale — and once it runs, those questions retire and the credit becomes a wager on operations. Twelve's repricing is where e-fuels lending acquires a genuine price, and whether that price turns out generous or punishing matters less than the fact that it now exists. Every developer approaching a financial close has a comparable where it previously had a model, and sponsors raising the next equity round will find their valuation argued from it.

LOIM is working the same problem a stage earlier, with a plastics fund that pairs a mid-teens return target with a $10 million seed check from the Alliance to End Plastic Waste and has anchored two pre-scale process companies. The charitable money is doing specific work — absorbing the earliest risk so the fund can underwrite process technology before a plant is demonstrably commercial — and the measure that matters is plant economics rather than the portfolio announcement, because the seed check buys the time to find out whether the plant runs. Neither the fund nor its anchor needs a sustainability-linked structure to make that bet.

In Brussels, the ECON committee wants companies in SFDR's new Transition category to spend more on sustainable activities than on new fossil fuel projects, and it has set up a fight with member states over who qualifies; read that as a capital test and the squeeze becomes visible from the other end. A category gated on a spending ratio admits the issuers with the largest build programs and turns away the ones still running legacy assets down, often the companies whose transition is most expensive and most dependent on outside capital. Tightening the entry test is defensible on its own terms, but expecting it to increase the supply of labeled borrowing is a different bet, and the loan numbers are already answering it: assets that cannot clear a spending test, or whose owners will not carry the compliance cost, keep migrating to structures where nobody asks for a KPI report.

None of that makes the framework worthless: for a diversified industrial with a decade-long decarbonization plan and a balance sheet to match, an issuer-level target binds management to something public and hands lenders a number to monitor quarter to quarter. The difficulty is scope: the label was built for corporate borrowers with broad footprints and long horizons, and the pipeline now moving through the market is full of single-asset projects with short histories. When a structure suits only part of the market it was designed for, the rest of the market builds something else.

Eurazeo's €150 million circular-IT platform shows the pattern with no linkage to a sustainability framework at all: the acquisition of Flex IT joins corporate collection of used equipment to a reseller network of roughly 13,000 across Europe, and the return will come from collection and resale economics rather than from a marker on the debt. ARC Ride's $33.3 million for battery-swapping across Africa, backed by development-finance institutions, a Japanese auto supplier and lenders, treats the electric two-wheeler shift as infrastructure rather than as a consumer product, and Bluecore's $50 million seed for floating nuclear plants is a bet on a different regulatory clock entirely. Revaia's debut energy fund arrives behind a €500 million deployment record, which does the opening work of an investor's due diligence before the first meeting.

The aftermarket is the tell: PWD's deal log this week carried a €120 million secondaries vehicle aimed at transition assets, and secondaries trade only once buyers and sellers agree on what the underlying projects are worth, which takes operating history, comparable financings and a view on the residual. Twelve's repricing, Eurazeo's platform purchase and ARC Ride's station economics all feed the same thing — a body of transactions from which marks can be derived — and the labeled loan market never produced anything comparable, because the label priced the borrower rather than the asset, and borrowers are not what secondaries buyers underwrite.

For allocators the practical consequence is a diligence question: direct lending and infrastructure mandates are where this pipeline is being financed, and the skill that earns the fee sits close to engineering — reading a conversion, a plant's operating history, a construction schedule that has already slipped once. Funds whose mandates require a labeled instrument will screen out much of what is now being underwritten, and that constraint belongs in the fee conversation with the LP rather than in a marketing appendix. M&G's new global impact head for private markets reports where the products are built, which reads as one large manager moving measurement toward underwriting; the managers worth watching are the ones staffing for that work rather than for the framework.

That leaves the 18% as a verdict on the label, not on the capital. Sustainability-linked loans made sense when lenders had no way to underwrite a transition asset directly; the format supplied governance in place of analysis. That gap has been closing for a while, and this week it closed a little further. Two things will settle it. Watch whether the ECON capital test survives into the final SFDR text, and watch how the next e-fuels financial close comes to market; if it arrives as straight project finance, the margin ratchet is a legacy instrument, and the treasury teams still bargaining over one are optimizing a number the market has stopped paying.