An EPC used to be paperwork. Now it's a business decision.

For years, the humble EPC (energy performance certificate) was something you sorted out for a sale, stapled to the deal, and, well, kind of forgot about. It was a box ticked. A document filed. Job done.

That era is ending, but it isn't because tenants have become greener or because boards have suddenly discovered the climate crisis; it's because the rules governing bank lending have changed – and your building sits on the wrong side of them.

Here's what’s changed, and – because the market is full of confident numbers that don't survive checking – which parts of the business case hold up.

What changed on the lending side

Two things.

Banks now have to manage ESG risk as a governance requirement. CRD VI inserted a new article into the EU's banking directive requiring institutions to have "robust strategies, policies, processes and systems for the identification, measurement, management and monitoring of ESG risks over the short, medium and long term" – and to consider a long-term horizon of at least ten years.

In Denmark, those requirements took effect on 1 January 2026, through changes to lov om finansiel virksomhed and ledelsesbekendtgørelsen. Finanstilsynet published the notice in December.

And the supervisor has told banks where to look. The European Banking Authority's guidelines on managing ESG risks apply from 11 January 2026. They ask institutions to make sure ESG factors are reflected in their internal risk classification, in credit scoring and rating models, and – where material – in the valuation of collateral. They single out, for real-estate exposures, "the level of energy efficiency in the case of residential or commercial real estate exposures together with the debt service capacity". They also ask banks to report their real-estate portfolios broken down by the energy efficiency of the collateral.

While these are guidelines, not regulation, the direction isn't ambiguous. Your bank is being asked to sort its book by the energy efficiency of the buildings behind it, and you're in that book. Smaller institutions have until January 2027 — everyone else is already there.

The lenders are already writing the requirements

When Energistyrelsen consulted on the Danish implementation over the winter, Finans Danmark filed a response. They support the direction, but want renovation requirements imposed only where they're technically and economically feasible – and they asked the authorities for a dialogue about how the data foundation and documentation standards should be structured.

The financial sector is asking to help design what owners will have to prove, and in what form. If you want a preview of the evidence your bank will eventually want from you, that's the document to keep your eyes on.

Which returns actually hold up?

Word on the street is that improving a rating pays back on four fronts: valuation, financing, rent and energy cost. That's the standard pitch – one we've made as well.

Two of them are grounded in rules with dates attached. Two rest on opinion surveys that don't hold up as well as they're usually presented, and the one quoted most often didn't ask anyone in the Nordics. But the more useful split isn't solid versus shaky – it's whether each one actually reaches you, which depends on your leases and on how your lender works.

Financing. Banks are required to manage ESG risk, and they're being asked to reflect energy efficiency in how they classify, score and value real estate exposures. It’s important to note that while the guidelines govern how a bank identifies and manages risk, they don’t control what it charges. So it affects whether a bank lends, not what it charges, which matters most at refinancing, where the risk is the loan not being there on the terms you'd planned for.

Energy cost. Measurable, immediate, and the only one of the four you can verify yourself, from your own consumption data. With one caveat that matters more than it sounds: in a commercial lease, whoever pays the utility bill gets the savings. If that's your tenant, the retrofit lowers their costs, not yours, and your route to capturing any of it runs through rent, which puts it back in the directional column. Worth knowing which of your leases put you on which side of that line before you build a business case on this one.

Valuation. If you’ve been shown a “green premium” or “brown discount” figure with RICS attached, that figure isn’t in the RICS standard.

RICS is the international reference point for valuation practice. It doesn’t govern Danish valuations, but if a figure like that existed anywhere, it would be there.

Its current standard on ESG in commercial property valuation mentions the “potential for ‘brown discount’ on value” exactly once – as a possibility, in an appendix.

What RICS does have is survey data. Its 2024 sustainability report, drawing on more than 4,000 responses across 32 countries, found 44% of professionals reporting a rent premium for green buildings, 31% reporting a brown discount but no premium, and 25% reporting neither. That last 25% never makes it into anyone’s deck. The 2025 edition dropped the question altogether, and reported slowing green-building demand instead.

So, it’s hardly accelerating the way the sector’s marketing suggests.

The more useful way to think about valuation is mechanical rather than reputational. A building doesn't become more valuable because it's greener; it becomes more valuable if the rent rises or the yield moves. When INREV modelled one real retrofit four different ways, the energy saving contributed 0.33% of the change in the building's value in every version – while rent and exit value did almost all the work. So the question to ask about a retrofit isn't whether it earns a green premium. It's whether it changes what you can charge, or what a buyer will pay for that income.

Rent: The most-quoted figure – that 79% of corporate occupiers expect sustainability to be the most important factor in rental decisions by 2030 – comes from research commissioned by the law firm BCLP. It's a genuine finding, quoted accurately. But the fieldwork was done in 2023; it's a prediction about 2030; it's self-reported sentiment rather than transactions; and the sample covered the US, UK, France, Germany, the Middle East, China, Hong Kong and Singapore. No Denmark, no Nordics at all.

That’s not to say that it’s completely irrelevant, but if a supplier quotes it at you as though it describes the Danish market, now you know where it comes from

What this adds up to

There’s still a business case. It's just an honest one, and a better basis for a board paper than four confident numbers that don't survive a follow-up question.