Climate risk, in plain terms: a starter guide for people working in banking

Ask three people at a bank to define climate risk and you'll likely get three different answers. One might mention flooding. Another might talk about EU regulation. A third might reach for "sustainability" and stop there, unsure if that's even the right word. The term gets used more and more, in credit notes, in board decks, in supervisory letters, and it rarely arrives with a definition attached.

At its simplest, climate risk is the financial exposure a specific asset or loan carries because of climate-related hazards. In everyday use at a bank the term stretches wider than that, and it usually points at three different things. This guide walks through all three, written to be useful without a climate science background, whether you advise customers directly or own the risk framework behind them.

Start with what climate risk is not

Climate risk is not the same as sustainability. Sustainability is a broad, often voluntary set of ambitions around emissions, social impact, and governance. Climate risk is narrower and more mechanical: it is about how climate-related events and trends create financial exposure for a specific asset or a specific loan. It is financial risk, not a reporting exercise, and that distinction turns out to matter.

Climate risk is also not a prediction. Nobody is claiming to know exactly which building will flood next year. It is closer to insurance logic: understanding exposure and likelihood well enough to price it, structure around it, or decide against it, the same way a bank already prices the risk that a borrower might default without claiming to know which borrower will.

And climate risk is not new in kind, only in cause. Physical damage to collateral, rising operating costs for a borrower, shifting market demand: banks have always priced these things. What's changed is that climate-related hazards are now a significant and growing source of all three.

The three things people actually mean when they say "climate risk"

In practice, the term covers three distinct categories, and most confusion comes from treating them as one thing.

Physical risk is the most intuitive: the weather actually damaging something. This splits further into acute physical risk, a single event like a flood, storm, or landslide, and chronic physical risk, a slower-moving trend like rising average temperatures, changing precipitation patterns, or gradual coastal erosion. A warehouse damaged by a hundred-year flood is acute physical risk. A vineyard slowly becoming unviable as growing seasons shift is chronic physical risk. Both are physical, and neither needs a change in regulation to matter.

Transition risk is the cost of a shift already underway: tightening energy standards, changing carbon pricing, shifting demand from tenants and buyers who increasingly factor sustainability into their decisions. A building with a poor energy label isn't damaged today. It faces transition risk because, in a market with tighter energy standards, that building becomes more expensive to keep competitive and less attractive to occupy over time. Transition risk is about direction, not a single event.

Nature and biodiversity risk is the newest and least understood of the three, and it sits slightly apart from the other two. Strictly, it belongs to its own discipline rather than to climate risk proper, and the formal frameworks treat it separately. It earns its place here because bankers meet it in the same decisions, at the same desks, often on the same property. It concerns how a property or a piece of land interacts with protected species, wetlands, or otherwise vulnerable ecosystems nearby. In Norway, this shows up concretely in things like hollow oaks, which are legally protected and can trigger buffer zones restricting nearby construction, and peatlands, which combine biodiversity value, carbon storage, and flood regulation in a single habitat type. Nature risk often surfaces late, during permitting or detailed site assessment, well after a loan has already been approved.

A single property can carry all three types of risk at once, and each one calls for a different kind of response. That distinction matters more than most conversations about "climate risk" tend to acknowledge.

Why this ended up in a bank's vocabulary at all

The mechanism connecting a weather event to a bank's balance sheet is short, and worth tracing once.

A flood damages a building. That building becomes harder or more expensive to insure. Reduced insurability affects the value and liquidity of that property as collateral. A change in collateral quality affects the credit risk of the loan secured against it, specifically the assumptions a bank makes about probability of default and loss given default. And changes in credit risk across a portfolio eventually show up in how much capital a bank needs to hold against it.

None of these five steps require a new regulation to be true. They are ordinary financial logic. What's changed is the frequency and scale of the event that starts the chain.

Why this isn't the same conversation as ten years ago

Three things have shifted at once. Physical events that used to be described as unusual, a once-rare flood, an unprecedented storm, are becoming more frequent in specific, mappable locations. Data that used to be expensive or unavailable, address-level flood modelling, energy performance records, protected species mapping, has become far more accessible, which means exposure can now be assessed for a single property rather than estimated for a whole region. And supervisors and regulators across Europe have started asking banks to demonstrate, rather than assert, that they understand this exposure across their loan books.

Any one of these on its own might not have changed much. Together, they've moved climate risk from a topic mentioned in passing to one that shows up in ordinary credit conversations.

Where this shows up in an ordinary working week

A loan application comes in for a property in a flood-mapped area. The question isn't whether to decline it automatically. It's whether the exposure is understood and reflected in the terms.

A customer asks their advisor whether a planned renovation qualifies for a green loan. Answering that well means checking more than the energy label: physical exposure and nearby nature constraints matter too, even if the advisor hasn't been trained to think of them as part of the same question.

A quarterly portfolio report shows that lending has quietly clustered in one coastal region over the past few years, without any single loan looking unusual on its own. Catching that requires looking at the portfolio, not just the individual file, and it's the risk function's question more than the advisor's.

None of these moments require deep climate expertise. They require knowing which question to ask, and where to look for the answer.

The difference between a box ticked and a decision made

There's a real gap between a credit note that says "climate risk was considered" and one that says "this property carries elevated flood exposure at the address level, with an estimated effect on collateral value reflected in the loan-to-value calculation." The first satisfies a disclosure line. The second changes what the bank actually does.

That gap is the whole game. Credible use of the term is always specific: a named hazard, a named location, a stated financial consequence. Anything vaguer than that can be reported, but it can't be acted on, not in a credit committee, not in a portfolio review, not in a conversation with a supervisor who wants to see the reasoning. If a sentence about climate risk could be dropped into any credit note without changing a word, it isn't yet saying anything.

The bottom line

Climate risk isn't a new discipline invented by regulators or consultants. It's an old one, financial risk, with a newer and faster-growing set of causes. Physical, transition, and nature risk are three different things that happen to share a name, and treating them separately is the first step toward managing them.

The banks that treat climate risk as a real input, specific enough to change a number, will be a step ahead of the ones still treating it as a paragraph to include. Once that vocabulary is shared, from the advisor at the branch to the risk team reporting to the board, the rest of the conversation gets considerably easier.