For most of the last decade, ESG in commercial real estate was a marketing layer. Landlords published sustainability reports. Consultancies scored buildings on frameworks with limited bearing on rents or valuations. Institutional investors bolted ESG language onto fund documentation while still deciding on financial metrics that treated environmental performance as a nice-to-have.

That gap has largely closed — but not for the tidy reason the sector’s marketing suggests. ESG became an underwriting variable in European commercial real estate over the last 18 months because of three forces that are real, measurable and, crucially, still in motion. Anyone who tells you the timeline is fixed hasn’t been reading the regulators this year.

The first force is the EU Taxonomy moving from paper to pricing. In force since 2020, its practical effect on lending was minimal until 2024, when major European banks — BNP Paribas, Santander, ING, BBVA, ABN AMRO — began applying Taxonomy alignment criteria to their commercial real estate loan books. A building that fails the technical screening criteria for energy performance is increasingly locked out of the “green” capital that prices the best terms. The result is a two-tier credit market: aligned assets in one financing pool, non-aligned in another, at materially different spreads. This one is not reversing — it’s a lending-desk behaviour now, not a policy timetable, and behaviour doesn’t get repealed.

The second force is national energy-performance regulation — and this is exactly where the “fixed deadline” story falls apart. The conventional line, repeated in pitch decks all year, was that UK MEES tightens to EPC C in 2027 and EPC B by 2030. As of June 2026, both of those dates are wrong. The UK government’s interim response to its 2019 and 2021 consultations dropped the EPC C 2027 interim milestone entirely and pushed the EPC B requirement for larger commercial buildings to 2031, subject to secondary legislation and a cost-effectiveness test. The direction of travel — higher minimum standards, unlettable stock below the threshold — is intact. The dates that underwrote the urgency are not. The Netherlands’ 2023 office rules, Germany’s GEG ratchet, and France’s Décret Tertiaire remain in force; but the UK slippage is the tell: when compliance cost collides with political reality, regulators blink on timing, not on direction.

The third force is the one least dependent on any government: tenant demand. Procurement teams at the largest corporate tenants — banks, tech firms, professional-services groups — now screen buildings for sustainability credentials before signing. This isn’t virtue signalling; it’s the flow-through of the tenant’s own Scope 3 obligations under CSRD. A tenant reporting emissions publicly cannot afford a building that drags its numbers down. BREEAM Excellent, LEED Gold and DGNB Platinum are becoming the entry price for prime office lettings across Europe. This force has no deadline to slip because no legislature controls it — which is precisely why it is the most durable of the three.

Together, financing pressure, regulation and tenant demand have produced what asset managers call the “brown discount”: the price gap between compliant and non-compliant assets. Underwriting teams at the major managers are working with 15–30% below headline valuation for materially non-compliant assets, wider still for UK offices below the old MEES thresholds.

Here is where the honest analyst has to pause, because the sharpest objection to this whole thesis lives inside that number. A meaningful part of the “brown discount” is not an ESG penalty at all — it is deferred-capex arithmetic that has always existed. A building needing €600/m² of refurbishment has always been worth less than one that doesn’t; that is not a new environmental variable, it is old-fashioned real estate maths wearing a green label. Conflating the two lets a lot of consultants sell a story. The part of the discount that is genuinely new — the portion driven by financing exclusion and tenant screening rather than by the repair bill — is real, but it is smaller and harder to isolate than the headline 15–30% implies. An owner who treats the entire discount as ESG-driven will over-spend on certification; one who treats none of it that way will under-invest and strand the asset. The skill is telling the two apart.

Which reframes the owner’s options more usefully than “refurbish, divest, or lose.” The genuine decision is about timing against a moving regulatory line. Refurbish to standard — typically €200–800/m² — makes sense where tenant demand and financing access, not just a deadline, justify it. Divesting to a specialist repositioner transfers the capex risk at a price that already reflects it. And holding is not automatically the bad decision the marketing claims: with the UK having just bought its market three extra years, an owner of a well-located asset with a credit-worthy sitting tenant may rationally wait, refinance around the gap, and let the retrofit-cost curve fall before committing. “Doing nothing” is sometimes discipline, not denial — the error is doing nothing without knowing which.

Where the effect is sharpest is unambiguous: secondary office stock in tier-one cities — 1980s and 1990s buildings that met their era’s standards and need substantial work to meet today’s. In Madrid, Barcelona, Milan and Amsterdam, the ratio of transacting to non-transacting assets in this segment has diverged hard. Compliant buildings trade at compressed yields. The worst non-compliant ones don’t trade at all — not because ESG forbids it, but because buyer and seller can’t agree on who eats the capex.

The proptech ecosystem is scaling into the gap — Measurabl, Deepki and Sensorfact on building performance, retrofit finance emerging from specialist and incumbent lenders, digital twins becoming standard on major refurbishments because certification and lending both now demand granular energy modelling.

The owners who come out ahead are not the ones who panicked into certification, nor the ones who dismissed ESG as a communications exercise. They are the ones who learned to separate the part of the discount that is a repair bill from the part that is a structural repricing — and who read the regulators closely enough to know that the deadlines move, but the direction doesn’t.