Can the ECB set a global benchmark for credible climate finance?

Key points:

• The collapse of voluntary alliances like the NZBA shows that market-led climate commitments fail under political pressure without robust regulatory backing.

• Unreliable scope 3 reporting – where over 60% of year-on-year changes in reported emissions cannot be explained by operational factors – makes genuine decarbonisation hard to identify and undermines green finance.

• Tackling “greenwashing” requires an NGFS-aligned framework featuring open emission databases, industry-specific pathways and standardised progress attribution.

---

“This is not about mitigation or adaptation. We need both,” Ursula von der Leyen, European Commission president, told the European Parliament in her State of the Union address this month, after a summer of record heat losses.

Climate science is clear: delaying mitigation raises losses and can narrow the options for adaptation. But mitigation policy runs on short political cycles, vulnerable to war and energy shocks. Mark Carney, then governor of the Bank of England, called climate change the “tragedy of the horizon”. Finance faces the same challenge.

The Net-Zero Banking Alliance (NZBA), the voluntary coalition Carney helped launch in 2021, collapsed in 2025, when major US banks exited under political pressure and remaining members voted to dissolve it that October.

A European Central Bank (ECB) study found NZBA banks cut lending to polluters by 20% between 2018 and 2022 — no faster than banks with no pledge at all. Voluntary commitments can only go so far when political conditions change.

In its 2022 climate stress test, the ECB found extensive reliance on proxies for scope 1, 2 and 3 emissions. Scope 1 covers emissions from a company’s own operations, scope 2 those from the energy it buys and scope 3 those generated across its wider value chain. Different methods produced very different estimates for the same company. In June 2026, the ECB introduced a climate factor in its collateral framework to account for transition-related uncertainty. Green bonds made up 33% of the ECB’s own funds portfolio at the end of 2025, up from 28% a year earlier, with a target of 35% in 2026. But green finance gets harder when you look beyond fossil-fuel exposure.

Lending portfolios can be heavily exposed to scope 3 emissions, i.e. emissions embedded in the value chain from suppliers, upstream and downstream transport and the use of sold products. Banks then have to work out whether a reported year-on-year reduction in emissions reflects genuine decarbonisation, a change in measurement methodology or something else entirely. Our peer-reviewed research shows how large that gap still is. Across 401 European companies, 62.5% of year-on-year changes in reported scope 3 emissions could not be explained by operational factors.

Moreover, current scope 3 guidance relies on sector-wide averages and is not tailored to each industry’s unique supply chains. This hinders alignment assessments across lending portfolios and the integrity of corporate climate action.

The problem is not so much about measurement, as it is about giving finance a reliable way to evaluate the alignment of transition plans with temperature targets while simultaneously distinguishing genuine climate action from methodological changes. 

That is what closing the “tragedy of the horizon” requires.

Regulation is crucial, but it cannot do it alone

The EU has spent years building a framework for non-financial disclosures. But regulation alone cannot solve this problem since the underlying numbers are not comparable.

The ultimate goal of disclosure is to make performance comparable. That only works if the underlying data is coherent. Our research found that identical purchasing data, run through different emission-factor databases produced scope 3 estimates that varied by 2.09 times for the same company.

If the same activity can produce materially different emissions estimates, banks cannot reliably tell whether reported changes reflect real decarbonisation or a change in measurement. And if they cannot make that distinction at company level, they cannot reliably aggregate progress across a portfolio.

The same gap appears in sustainability-linked bonds, where the World Bank has flagged structural loopholes that let issuers avoid penalties even after missing their targets. Bank transition planning shows the same pattern. The ECB found about 70% of 95 banks could face elevated litigation risk where portfolios were misaligned with Paris commitments.

A shared infrastructure for green finance

Taking the Network for Greening the Financial System’s (NGFS) own diagnosis of the problem, our research lays a scientific foundation in three practical areas to strengthen green finance. Together they form a shared scientific infrastructure for green finance.

First, accounting. The NGFS has called for harmonised emissions metrics. An open, scientifically harmonised emission-factor database, tailored to industry value chains, could improve comparability and benchmarking. Such a database could play a role similar to the Intergovernmental Panel on Climate Change’s emissions factor database, but focused on corporate value chains.

Second, alignment. The NGFS notes that scenario providers differ in sectoral and geographic detail. Developing industry-specific net-zero benchmarks would lay a scientific foundation for aligning corporate targets in lending portfolios with temperature pathways, while limiting greenwashing risk in how residual emissions are defined. They could give banks a more defensible basis for assessing borrower and portfolio trajectories.

Third, attribution. The NGFS identifies year-on-year changes in reported emissions as an open attribution analysis problem. These reported emissions reflect a combination of genuine decarbonisation, business growth, changes in methodology and organisational changes such as mergers. We propose a decomposition method that isolates genuine climate action from other effects. It could help banks distinguish genuine progress from changes driven by other factors.

From a European fix to a global benchmark

Addressing Carney’s horizon problem requires more than climate scenarios.

Banks sit at the point where climate ambition becomes capital allocation. Central banks therefore need credible evidence of how companies are actually decarbonising: data that can distinguish real emissions reductions from other accounting effects, compare companies against industry-specific pathways and aggregate those trajectories into portfolio-level risk.

That is what the proposed shared scientific infrastructure could provide.

Von der Leyen is right that climate action needs both mitigation and adaptation. The same research sketches how this infrastructure could extend to physical risk and adaptation costs at company level, helping supervisors build a holistic picture on the impacts of climate change on financial institutions.

The proposed shared scientific infrastructure complements the ECB’s climate factor while addressing key gaps the NGFS already identified and its 40-plus members spanning six continents. If the ECB builds this infrastructure, other supervisors already using NGFS scenarios could apply the same industry-level lens to their banks’ books. A  European initiative could then become a common reference point for assessing climate risk in financial markets well beyond Europe.

This piece draws on the authors’ peer-reviewed perspective, From Corporate Net-Zero Pledges to Credible Climate Action, co-authored with Luis Costa, Ponraj Arumugam, Matthew Agarwala, Felix Creutzig and Jürgen P. Kropp, published in Nature Sustainability.

Written by