Banking risk is no longer hidden

This article is part of In conversation about sustainable finance & emission reduction systems, a new series by Diego Balverde. You're reading volume seven of the The Collateral Crisis series. Here is volume six

Prologue

Banking risk is no longer buried only in regulatory tables, capital ratios or non-performing loan statistics.

It now lives in energy bills, freight rates, industrial margins, sovereign spreads, climate-related haircuts, refinancing costs and the growing distance between the price of money and the stability of cash flow.

That is the real shift.

For years, the European banking system could still pretend that risk was mostly visible inside the balance sheet and manageable through buffers, provisions and supervision. That era is ending. In late 2025, banks in the EU and EEA still reported what looked like solid aggregate indicators: a CET1 ratio of 16.3%, an LCR of 160.7%, an NSFR of 126.8%, total assets of 29.1 trillion euros, an NPL stock of 373 billion euros and a return on equity of 10.7%. At first glance, that sounds reassuring. But the problem is no longer whether banks still look stable on paper. The problem is whether the economic system beneath those numbers is still stable enough to keep those numbers credible. That is where the hidden risk has stopped being hidden. It has moved from the balance sheet into the structure of the economy itself.

The balance sheet still looks strong, but the economy underneath it does not

A bank can report comfortable capital and still be sitting on borrowers whose real resilience is deteriorating month after month.

That is the danger of this stage.

European banks have not entered 2026 as weak institutions in a classic sense. They entered it as institutions with apparently solid balance sheets exposed to an increasingly unstable operating environment.

Energy remains structurally more expensive than in the old normal. Freight costs remain more sensitive to geopolitical shocks. Industrial margins remain under pressure. Consumer demand is more fragile. Credit standards have tightened. Working capital is more expensive to fund. This means the quality of bank assets is no longer determined only by the historical creditworthiness of borrowers. It is increasingly determined by the ability of those borrowers to survive a system with permanently higher friction. That distinction is critical. A firm that was financeable when electricity, gas, transport and debt were cheap may stop being financeable in a world where all four remain elevated or unstable. A bank does not need an immediate wave of default to face pressure. It only needs a sufficient number of borrowers to become weaker at the same time. If a large industrial borrower sees its EBITDA fall 15% to 20% because of energy, logistics and refinancing costs, its probability of default rises even before it misses a payment. If enough firms go through that process simultaneously, banks do not see collapse first. They see migration. Stage 2 loans remain one of the clearest warning lights here. In the third quarter of 2025, Stage 2 loans still represented 9.3% of total loans in the EU and EEA banking system. That means nearly one in every ten loans was already under reinforced surveillance. A system with that amount of latent deterioration cannot honestly claim that banking risk is still hidden only in the back pages of a supervisory report.

Credit risk has moved out of the loan file and into the real economy 

The classical view of banking risk starts with the loan book. The real one now starts with the economy. When energy costs remain high, when transport corridors become more fragile, when insurance becomes more expensive, when sovereign borrowing needs rise and when collateral becomes more sensitive to climate and transition risk, the bank is no longer the origin of the problem. It becomes the transmission channel. That is why banking risk is no longer hidden. It is visible in industrial hesitation, in weaker investment plans, in more cautious treasury decisions and in a loan market that has become more selective even before a visible credit crisis.

The euro area bank lending survey made that point brutally clear. In the fourth quarter of 2025, banks tightened credit standards for firms by a net 7%, after a 4% tightening in the previous quarter, bringing the cumulative tightening since the third quarter of 2024 to 19%.

For the first quarter of 2026, they expected a further 6% tightening. This is not a minor technical adjustment. It means the system is already changing behavior because it perceives a broader vulnerability in borrowers and in the economy.

Banks themselves said the main drivers were higher perceived risk and lower risk tolerance. In other words, the credit system is already reacting to stress that has not yet fully materialized as default. That is exactly what makes the present phase so important. The risk is no longer invisible because it is no longer waiting to become delinquency. It is already changing the flow of money.

This has a second implication. Credit risk is no longer mostly idiosyncratic. It is increasingly correlated. If one company fails because of bad management, that is a company problem. If thousands of companies face higher energy costs, more expensive transport, larger working-capital requirements and higher refinancing costs at the same time, that is a banking problem. Correlated weakness destroys the illusion of diversification. A bank may think it is diversified across industries, but if those industries are all exposed to the same macro-frictions, then correlation rises precisely when the system needs diversification the most. That is why hidden banking risk is now embedded in energy markets, in freight routes and in industrial cash flow, not just in the legal wording of a loan agreement.

Collateral is losing neutrality

The next layer of hidden risk sits in collateral. Banks lend against future cash flow, but they also lend against the assumption that if the borrower fails, the collateral will preserve value. That assumption is changing. A commercial building that is energy-inefficient, an industrial asset exposed to volatile utility costs, a company tied to emissions-heavy processes or a borrower operating in a sector under transition pressure is no longer the same type of collateral it was five years ago. The value may still be there on paper. The credibility is not as automatic.

That is why the ECB’s decision to introduce a climate factor in the Eurosystem collateral framework matters far beyond technical risk management. From 15 June 2026, certain marketable assets pledged as collateral may receive a lower assigned value depending on their exposure to climate-related uncertainty. That is a major signal. It means that climate and transition risk have moved from being a debate about disclosure and strategy into the mechanics of monetary operations and collateral treatment. Once central bank collateral is no longer neutral, commercial bank collateral cannot remain neutral either. Banking risk is no longer hidden when the very asset used to support liquidity is being re-read through a different lens.

This matters for more than corporate bonds. It changes the way banks look at real estate, industrial equipment, ports, logistics assets and any operating platform whose long-term profitability depends on cheap energy, stable transport and a weak carbon penalty. The question is no longer simply “what is this asset worth today?” The question is “will the market still trust this asset to support debt tomorrow?” That is a much harder question. And once markets begin asking it with discipline, banking risk becomes less about accounting and more about credibility. When collateral loses neutrality, balance sheets lose invisibility.

The sovereign and the industrial loan are getting closer again

One of the most underestimated aspects of this new phase is that sovereign risk and corporate risk are starting to converge again through the banking channel. Higher debt levels, slower growth, tighter fiscal space and more expensive refinancing increase the sensitivity of sovereign spreads. Banks still hold very large portfolios of sovereign debt because that remains central to liquidity management, regulation and portfolio structure. At the same time, those same banks are lending to industrial, commercial and household sectors that are exposed to energy costs, weak growth and tighter credit conditions. This means the banking system is again becoming the point where state risk and real-economy risk meet.

That does not automatically mean a euro area sovereign-bank doom loop in the old form. But it does mean the old separation is weakening. If states need more issuance, growth weakens and industrial borrowers become more fragile, the same banking system absorbs pressure from both directions. The EBA’s 2025 stress test was useful here. It showed that under the adverse scenario, aggregate CET1 depletion was 370 basis points, bringing the ratio down to around 12.1%. That is not a collapse. But it is a reminder that resilience exists under stress only because the starting point was relatively strong. A system that loses 370 basis points of capital under adverse assumptions is not invulnerable. It is simply better prepared than before. The question is what happens if geopolitical tensions, climate transition costs, sovereign funding pressure and industrial weakness overlap in the real world more persistently than in the test horizon.

That is where hidden banking risk becomes historical banking risk again.

Implementation and value capture

The only serious way to respond to this environment is to reduce risk where it is actually generated: in operations, in energy use, in logistics friction, in idle assets and in unstable cash flow. That is why the future of banking resilience will not depend only on supervisors, buffers and capital plans. It will also depend on whether the real economy can become more financeable. This is where efficiency stops being a technical conversation and becomes a banking conversation.

If an industrial operation reduces energy consumption by 15%, lowers idle time by 20%, improves throughput by 10% to 15% and stabilizes cost volatility, the improvement is not just operational. It changes cash flow quality. And once cash flow quality improves, banks price the borrower differently. A reduction of 50 to 150 basis points in financing cost on a 100-million-euro debt structure means 500,000 to 1.5 million euros per year. That is not marginal. It is the difference between surviving under tighter credit and slowly losing access to it.

That is where models like BalGreen matter in practical terms. Not because they sound sustainable, but because they identify where money leaks, where energy is wasted, where friction destroys margin and where process redesign can recover cash flow. A port that cuts idle time, energy use and coordination failures becomes more productive, but also more financeable. A logistics chain that reduces fuel intensity, empty routes and waiting time becomes not only cleaner but less risky. In a world where banking risk is no longer hidden in loan files alone, every measurable reduction in friction becomes part of the borrower’s defense. The bank of the next cycle will not just lend against assets. It will lend against disciplined systems.

Analysis

The real debate is no longer whether the banking system still looks stable. It does, at least superficially.

The real debate is whether that stability still rests on an economy capable of carrying it.

How many apparently healthy banking indicators are still the delayed reflection of an environment that has already changed underneath them?

How much of current bank profitability still comes from the rate cycle rather than from durable borrower strength?

How many corporate borrowers remain alive only because deterioration has migrated into Stage 2 instead of open default?

How much collateral still looks acceptable because markets have not yet fully repriced energy, climate and industrial fragility into valuation?

Can the banking system still call itself diversified if the same cost shocks hit multiple sectors at once? Are we sure that sovereign exposures and industrial exposures can still be treated as separate conversations?

And what happens if credit standards keep tightening not because regulators order it, but because the system simply stops trusting future cash flow as much as it used to?

The hardest question is the simplest one: are banks still funding productive expansion, or are they increasingly funding only what looks least likely to go wrong? Because if the answer is the second, then banking risk is no longer hidden in a balance sheet. It has already moved into the architecture of growth itself. At that point, the issue is no longer only how strong the banks are. It is whether the economy is still strong enough to deserve the balance sheets it currently has.

Conclusion

My conclusion is direct. Banking risk is no longer hidden because it is no longer confined to accounting ratios, supervisory templates or bad-loan statistics. It is now visible in energy, in logistics, in collateral, in sovereign exposure, in industrial fragility and in the way banks have already begun to change their lending behavior. That is the key transition. The banking system may still look resilient, but the economy beneath it is carrying more friction, more cost and more instability than the old framework assumed.

This means the next phase of European banking will not be defined only by how much capital banks hold. It will be defined by the quality of the system they are financing. If borrowers can reduce friction, stabilize cash flow and defend collateral, banks will remain strong. If borrowers cannot, then capital buffers will eventually be forced to absorb what operations failed to correct. That is why the hidden risk is no longer hidden. It is already circulating through the economy. The question now is not whether it exists. The question is who will be disciplined enough to reduce it before the balance sheet does the disciplining by force.

illuminem Voices is a democratic space presenting the thoughts and opinions of leading Sustainability & Energy writers, their opinions do not necessarily represent those of illuminem.

Track the real‑world impact behind the sustainability headlines. illuminem’s Data Hub™ offers transparent performance data and climate targets of companies driving the transition.