Europe is pricing its clean energy future with the wrong risk data

By Professor Subhamoy (Suby) Bhattacharya, Co-Founder and Chief Scientific Officer at Renew Risk, a leading provider of risk analytics for renewable energy assets

Europe’s clean energy transition is one of the largest capital deployment stories of our time. Hundreds of billions of pounds are flowing into offshore wind, the technology expected to anchor the continent’s energy security for decades. 

Yet a basic problem lies beneath all that investment, and not enough of the decision-makers backing these projects are talking about it. The assets at the centre of the transition, currently worth more than £100 billion in insured value (and growing at around 19 per cent a year), are being assessed, priced and financed using risk models calibrated to a climate that no longer exists.

This is not a technical footnote; it’s a gap between the risk that investors, insurers and lenders think they are taking on and the risk they actually carry. As the climate continues to change and grow more volatile, it is widening.

Built on a history of data that no longer holds true

Risk models are only as good as the historical data they are built on. Pricing risk depends on a long run of weather observations telling you how often an event is likely to occur and how much damage it could do, all on the assumption that the past is a fair guide to the future.

For offshore wind, that assumption is breaking down on two fronts at once.

The first is the climate itself. This summer’s record-breaking heatwave across the UK and Europe is one visible symptom of a wider shift in the climate. Storm behaviour, wind patterns and the interaction between wind and waves are all moving away from their historical baselines. When the climate stops repeating itself, the historical record ceases to be a reliable predictor, and every model built on it inherits that flaw.

The second is technology. Europe has the world’s oldest offshore wind fleet, and it has changed almost beyond recognition in two decades. From turbines of around 70 meters to today’s machines, which can exceed 230 metres in height, taller than ‘The Cheesegrater’ in London. There’s little historical data on how structures this large behave under extreme conditions because they haven’t existed long enough.

Put those two problems together and you have a fleet with no meaningful precedent, exposed to a climate that is itself leaving precedent behind. To put things into perspective, in 1999, when Storm Lothar struck Europe and caused billions in insured losses, the offshore wind fleet was effectively zero. Today, around 39 gigawatts of installed capacity sit directly in the path of potential storms like it.

Damage estimates off by 90 per cent, either way

Start with the most crucial issue: much of the market doesn’t model this risk at all. The part of the market that does model this risk usually utilises onshore property models as a proxy for offshore assets. Those models understand wind, but they miss what really drives offshore losses, above all, the way wind and waves act together.

In one North Sea case study, approaches that used onshore proxies produced physical damage estimates wrong by up to 90 per cent and business interruption estimates by up to 100 per cent, with errors running in both directions depending on the proxies used.

That is the part that should trouble any investor. The numbers are not inaccurate. They are unpredictable. An error that runs in either direction depending on your assumptions cannot simply be adjusted for.

This is a financing problem, not only an insurance one

It would be convenient to file this under insurance, but mispriced risk does not stay where it starts.

Insurers and reinsurers bear the financial impacts first, and they are already raising premiums and tightening the capacity they will commit to these projects. The financial burden flows through to developers as higher project costs and to lenders as financing friction. From there, it reaches the numbers that matter most to investors: the cost of capital, the valuation of operating assets, and the confidence with which a project can be underwritten. A wind farm financed on the assumption of one risk profile, and later found to carry another, is worth something different from what its backers paid.

The danger is also deferred. A serious claim can land five or six years after a policy is written, so today’s mispricing does not surface until well into an asset’s life. The risk is not absent, it is just hidden for now.

The physical realities make this concrete. Offshore assets are expensive to build and hard to maintain, dependent on a small pool of highly specialised vessels for construction and repair. The repair of a £22,000 blade can escalate into a claim of £380,000 or more once vessel mobilisation and weather delays are factored in. Because that mobilisation is so costly, faults are often left until enough accumulate to justify sending a vessel, so an asset can run in a degraded state for months.

Modelling the asset, not the average

The answer is not to slow the transition down, but to assess it honestly. That means moving from proxy models to purpose-built ones that evaluate exposure and vulnerability at the level of the individual asset, draw on current climate data rather than a fading historical average, and reflect the real hazards and repair economics of the offshore environment.

Done properly, asset-level modelling gives investors, insurers and lenders what they currently lack: a clear view of where the vulnerabilities sit, how losses could accumulate across a portfolio, and what a project is genuinely worth. That is the difference between allocating capital on certainty and allocating it on hope. Reducing that uncertainty does more than sharpen a price: it draws insurers back in, gives lenders the confidence to lend competitively, and unblocks the next phase of build-out.

Europe has made extraordinary progress in scaling this technology. Whether the next phase succeeds will depend less on how many turbines the continent can build and more on how honestly it can price the risk of owning them. The capital is ready. The risk map needs to catch up.