Post date 22 Jun 2026 (original LinkedIn date)
Original LinkedIn version· 1,850 words
‘Down by the water’ by Amy MacDonald, released in 2017 on her album Under Stars
It’s London Climate Action Week, or LCAW if you prefer (although I still cannot quite get my head around the parrot-esque running together of the acronym!). As London hosts the world’s biggest week on climate action (give or take, although I’m sure that NY would have something to say on that) in what is likely to be sweltering temperatures this week, imagine three rooms in the city that have been having three different conversations about the same river, with none of them is talking to the others.
In the first room, the Environment Secretary has just written to Ofwat to say that the rescue deal for Thames Water is not good enough. Emma Reynolds called the London & Valley Water proposal a ‘weak plan’ after fifteen years of failure and warned it would put an undue burden on customers (hard to argue she’s not right). In the second room, somewhere inside the Bank of England, two economists are quietly discussing their published model of what happens to British mortgages when flood insurance starts to fail. In the third room, the one we are all about to walk into for LCAW, the talk will be of London becoming the world’s marketplace for resilience and adaptation finance. The first room is about a balance sheet, the second is about an average, while the third is about an ambition. The thing they share is a catchment.
Start with Thames Water, because it is the easiest target and therefore the one most worth trying to be fair to. The London & Valley Water consortium, a group of the company’s own senior creditors, is proposing to inject around £3.35 billion of new equity and up to £6.55 billion of new debt, and to write off something in the order of £12.5 billion of existing value. By the consortium’s own account that is expected to be the largest loss ever taken on a UK infrastructure investment. This is certainly not nothing - it is real capital discipline, and the alternative, special administration, is may not be that attractive an option. I am happy to concede all of that before I proceed.
The next thing is to notice what the deal is actually solving for, and the clearest tell is the distance between what the consortium itself said it was for and what it has asked for in return. London & Valley Water introduced itself in the language of restoration. Its proposed leadership talked about fixing the foundations, reducing pollution, rebuilding public trust and putting the environment first, and the founding announcement promised to clean up the waterways. Built into the same rescue, however, is a request to be spared the accountability that would make any of that binding. The consortium has said it will settle the existing fines, but its plan defers a full return to legal and environmental compliance for years, and the live negotiation now turns on a multi-year reprieve from new pollution penalties. Campaigners have put it more bluntly, calling it a request for permission to keep polluting on a protected timetable. The company that offers to clean the rivers is asking, in the same document, not to be held fully to account for fouling them. That is not a footnote to the deal - it is a term of the deal, and in my view it tells you what the deal is really about ...
... the capital structure. It is recapitalising a company, ranking the claims, writing down the debt, and seeking roughly £750 million to cover creditors, lawyers and advisers. It is a sophisticated negotiation in which every interested party has a seat except the one that matters most. The river is not a creditor, and the catchment is not a class of debt. The physical system that is actually failing, the sewers, the overflows, the lost drainage capacity of a city built on top of its own water, is treated as collateral to the financial problem rather than as the problem itself. The pollution is the variable the deal proposes to flex, precisely because it is the part with no one at the table to defend it. They are repairing the cap table and calling it a recovery of the asset, but not the same asset.
This is where the Bank of England’s economists become so useful, because they have written down, with admirable care, the exact mechanism by which finance is allowed to look away. Their model asks how big the gap could grow between mortgaged homes that have flood insurance and those that do not. Under conservative assumptions, the share of uninsured mortgagors roughly doubles, from about five per cent today to somewhere between seven and ten per cent by 2050, and could reach around sixteen per cent in the year after a severe flood. That is up to 1.39 million households without cover. And yet their headline conclusion is somewhat reassuring: the hit to the banking system as a whole is small, nothing like the financial crisis, nothing like the stress test. Unless of course you are unlucky enough to live in those affected houses!
I believe their numbers, so I want to draw attention to the sentence that makes the reassurance possible. The aggregate impact is modest, they explain, because flooding is a geographically bound risk: large impacts, even across large areas, average out to something smaller at the national level. Pause, it’s worth reading twice (honestly!). The system looks resilient precisely because the damage is concentrated. The national average is calm because the local reality is not. Their own charts show it - as many as eighteen per cent of mortgaged properties could fall more than ten per cent in value after a severe flood, with the worst-hit regions taking roughly four times the national average. The average is not describing the country, rather it is masking the places.
This is the argument I keep coming back to, and London is the cleanest case of it I have seen. A catchment does not average - one street floods to the ceiling while the postcode in its entirety might look fine. One aquifer empties while the country’s water balance reads as merely stressed. The instruments we use to reassure ourselves (the sector recapitalisation, the national stress test, the systemic-resilience headline) are averaging instruments. They are designed to net concentrated physical loss down to a manageable abstraction, and they do their job beautifully ... which is the danger. Capital can read the average and walk straight past the catchment.
It helps to ask who is standing inside the average while it is being computed. Fellow researchers at Bayes have a phrase for them: climate prisoners. Low-income homeowners in flood-prone areas already face an annual expected loss several points higher than their better-off neighbours, a gap projected to widen toward 2050, fed straight into their premiums as insurers move to risk-reflective pricing and reinsurers pull back. With Flood Re scheduled to end in 2039 and around four million British homes already in medium or high flood-risk areas, the people with the least capacity to move are the ones most likely to be left in homes that are becoming uninsurable and, in time, unmortgageable. The four-year fines holiday for the creditors and the rising premium for the homeowner who cannot leave are not two unrelated stories. They are the same decision read from opposite ends of the catchment. Finance is extraordinarily good at protecting the senior claim and extraordinarily bad at pricing the place.
Now stand on the place itself. London is a city built on top of its own rivers, in case we’ve forgotten (look at the map at the top). The Fleet, the Tyburn, the Westbourne, the Effra, the Walbrook and a dozen more were culverted out of sight over two centuries ago, and most of them now do their work as combined sewer overflows, the very channels that tip storm water and sewage into the Thames when the rain comes hard. We did not lose those rivers, we chose to demote them. We took a free, distributed, nature-based stormwater network and buried it, and we have spent the years since rebuilding a worse version of it in concrete, most expensively as a single super-sewer beneath the tideway. Daylighting these rivers, bringing them back to the surface, is usually filed under amenity and nostalgia. It could just as easily be filed under capacity. When sixty per cent of England’s flood-risk properties are exposed to surface water, when the Environment Agency’s own chief executive has told MPs that surface water flooding is the thing that keeps him awake at night, when two hours of rain in July 2021 flooded more than two thousand London properties with storm water and sewage, the buried river is not a heritage feature. It is decommissioned infrastructure that we are paying not to use.
The London Climate Resilience Review from back in 2024 understood this better than the deal-making has. It put nature-based solutions among its guiding principles, not its decorations, and it pointed out the awkward, instructive fact that East London’s water supply already depends on nature-based measures upstream in Hertfordshire. The catchment was always the asset. We simply stopped seeing it once we covered it over, and we have organised our balance sheets around the things we can still see.
So upon re-reading, I ask you to take the Review’s ambition seriously, that London should become the world’s marketplace for resilience. But a marketplace needs something to buy, and right now the only instrument on offer is recapitalising a monopoly and granting it a pass on its fines. That is not a resilience market; it is a restructuring, shuffling the deckchairs on the Titanic. A real one requires resilience to become a priced, contractable thing at the scale at which it physically exists, which is the catchment, not the company and not the country. It requires the parties who actually lose when a catchment fails, the insurers facing the loss, the lenders watching their collateral, the infrastructure operators downstream, and the public balance sheet behind all of them, to pay, in advance, for measured resilience outcomes upstream. Nature treated as infrastructure, bought under something closer to a purchase agreement than a grant, settled where the water is, not where the headquarters are. If we are serious about it, the principle worth arguing for over the next week is a simple one: resilience net gain, owned at the catchment.
That is my work for this week, that is if we actually want it to be work and not a conference. The Thames Water creditors, the Bank of England modellers and the resilience-finance panels are all describing one catchment, in three vocabularies, in three rooms, and the value is sitting in the space between them that nobody is being paid to occupy. London is a catchment, not a cap table. The river was here first. It is still here, out of sight under our streets, quietly doing the accounting we so far keep refusing to do.