Lower Marsham · part 6 of 7
The Deadline That Hinges
Lower Marsham, Episode Six. On the year 2039, the banks’ revealed opinion of Parliament, and a prediction registered in public
Every scheme in this series’ acquaintance has claimed to be temporary, so let us begin the final episode with the paperwork. Flood Re’s expiry is not a press release. It is primary legislation: the Water Act 2014 ends the scheme in 2039, and moving that date requires Parliament to pass a new Act in daylight, on the record, with a division bell. This is a genuinely stronger safeguard than anything the American programme ever possessed, and the case for optimism rests on it entirely. This essay is about the weight it is being asked to bear.
A transition is a measurable thing. A scheme thirteen years from handing its risks back to the market should be shrinking, and the set of numbers that was supposed to be falling are rising instead, every one. Policies ceded to Flood Re grew 20 per cent in the last reported year. The scheme’s own reinsurance costs have risen by more than £100 million a year since 2022. The levy was raised this April from £135 million to £160 million. The properties at risk in England were recounted at 6.3 million, one home in five, en route to a projected 8 million by mid-century as the climate delivers what it has long been advertising. Against that current, the exit requires household premiums, national defences, and property-level resilience all improving fast enough that a third of a million households can absorb true prices in a single season. The scheme’s chair assessed the position in the current annual report with a frankness his American counterparts never managed: if Flood Re left the market today, many of the households it supports would face steep premium increases, “if they were able to obtain cover at all.”
Watch the language moult
Institutions signal their intentions in the drift of their prepositions long before they announce them, and Flood Re’s public framing of 2039 has already completed a small but decisive migration. The exit is now said to “hinge on faster risk reduction.” Read that construction the way an option trader would. A date is a commitment; a date that hinges is a contingency, and a contingency whose trigger conditions (defence spending, planning reform, household behaviour) sit outside the scheme’s control is an option held by whoever gets to declare the conditions unmet. The rhetorical scaffolding for extension does not need building in 2038. It stands complete now, painted and load-tested, awaiting a housing minister to cut the ribbon.
The tenth-anniversary package announced this July supplies the scaffolding’s latest course. Its individual merits are weighed elsewhere in this series, and some are real; what concerns this episode is its shape. The statutory loss limit was raised two and a half times over, from £100 million to £250 million. The levy was reset under a fresh three-year funding arrangement with government. A claims cap and a wholly revised premium structure are to be negotiated with the industry and implemented “from 2028,” eligibility “kept under review.” Every verb in the announcement is the verb of an institution building out, not winding down. A scheme thirteen years from its own funeral does not ordinarily enlarge its loss limit, sign new multi-year funding, and commission a premium architecture whose first full decade of operation would end after the death date — unless, somewhere in the building, the date has quietly stopped being treated as one. The package does not mention 2039. That is the most eloquent sentence in it.
If you doubt that reading, consult the people who price Parliament’s promises for a living. Mortgage lenders hold, as collateral, the flood-zone housing stock whose values Episode One showed to be inflated by the scheme’s existence; a statutory sunset that actually executed would mark that collateral down sharply, and a prudent lender would be preparing. Flood Re’s chief executive describes the banks’ actual posture: their working assumption is that “they don’t need to do anything.” The most sophisticated credit institutions in the country have examined a commitment written into primary legislation and priced it at approximately nothing. Their reasons are not published. Their reasons are the previous episode’s Congress, twice around the same circuit, and the arithmetic of marginal constituencies.
So state the three doors plainly, because 2039 offers no fourth. Door one: honesty. The sunset executes, prices tell the truth, and a sharp, concentrated correction lands on whoever holds the deeds that year, many of whom bought in good faith at prices the state’s own scheme inflated; the American attempt at this door lasted twenty-two months. Door two: extension. A new Act, dressed as compassion, converts the transition into an institution, and Britain acquires a permanent NFIP with better manners, compounding liabilities, and a filling floodplain. Door three, the native option: the fudge. A “reformed” successor scheme, a taper that hinges, a review that becomes a fixture, the sunset moved politely over the horizon at the last mile while everyone concerned is praised for pragmatism.
The prediction
Commentary that risks nothing is worth what it risks, so here is a falsifiable claim, dated July 2026, from a motor actuary with no position in the outcome beyond his levy. Before the end of 2035, the government of the day will announce that Flood Re, in something like its current form, will continue past 2039. It will not be called an extension; the working titles will feature transition, reformed, or successor, and the announcement will follow a wet winter. The July package has already normalised the vocabulary; the winter will supply the occasion. The mechanism will be the Samaritan’s dilemma running exactly to specification: concentrated, sympathetic, organised losers on one side, and on the other a diffuse £10.50 that has never once been noticed on a renewal schedule. I commit to returning to this essay publicly when events settle the matter, and to eating the relevant portion of it, in public, if door one opens. No forecaster in this field offers that clause, which tells you something about the field.
I would add only this, for the minister who eventually holds the file. The choice in 2039 will be presented as arithmetic versus mercy, and that framing will be false, because the merciful option was always available and remains so: help the households openly, on a budget line, with money Parliament can see, while letting the price of water speak. What is scheduled to expire in 2039 is not the country’s compassion. It is the falsehood the compassion was wrapped in, and the falsehood, unlike the compassion, has been compounding.
Thirteen more winters, then. The village will follow the announcements with interest, except for one resident, who follows nothing. She held her commitments through the Statement of Principles, the Water Act, the transition plan, and every annual report; hers is the only forward guidance in this story that has never once required revision. The river has not read the Water Act 2014. She has no need of the amending legislation either. Among all the parties to the year 2039, she is the only one who will certainly keep the appointment.
Principal sources: Water Act 2014; Flood Re Annual Report 2025/26 (chair’s statement, ceding and cost figures); Flood Re Transition Plan and CEO remarks; Flood Re/Defra reform announcement (1 July 2026); Environment Agency NaFRA2 (2024); Climate Change Committee (2025); GAO and Congressional record on the NFIP, per Episode Four. Epilogue to follow: How to Help Without Lying.