England’s flood defences: what the NAO found, and what happened next

England's flood defences were the subject of a National Audit Office report that found the country spending more on flood protection than at any point in its history, and still on course to protect far fewer homes than it had promised. Three years on, the outturn figures are in. The Environment Agency now reports better protection for more than 150,000 properties since April 2021, against an original promise of 336,000 by 2027 and the revised forecast of roughly 200,000 that the auditors settled on. On delivery, the warning was broadly right. On maintenance, the story is stranger: the condition target was met, but partly by moving it.

The programme promised 336,000 better-protected properties by 2027. The auditors forecast 200,000. Five years into a six-year programme, the running total is a little over 150,000.

For anyone buying, building or insuring property, that distance between the promise and the delivery is the part that matters. A defence drawn on a map is not the same as a defence that holds when the river comes up, which is why a site-specific flood risk assessment for planning still does work that no national flood map can do for you.

Did the auditors' warnings come true?

Broadly, yes on delivery and only partly on maintenance. The shortfall the National Audit Office forecast in 2023 has largely materialised, and the capital programme is now tracking its revised figure rather than its original promise. The maintenance warning was answered, but by shifting money out of new construction and by measuring performance against a lower annual target than the 98 per cent the auditors held the Environment Agency to.

The last row is the one worth sitting with. The 98 per cent figure was the long-term condition goal, and the number the Public Accounts Committee measured the Agency against when it found it wanting. The 93 per cent met in 2026 was the annual target set for that year. Halfway through 2025/26 the Environment Agency's corporate scorecard recorded 92.9 per cent against an in-year target of 92 per cent, and marked it green. Both statements are true, and only one of them sounds like success.

What the NAO and the Public Accounts Committee actually found

The National Audit Office's report on resilience to flooding, published in November 2023, examined the six-year capital programme running to 2027 and concluded that it would protect substantially fewer properties than the figure it had been sold on. The Public Accounts Committee took evidence and reported in January 2024, and was blunter still.

• The headline was cut by more than 40 per cent. Government committed £5.2 billion to around 2,000 schemes between 2021 and 2027, targeting 336,000 better-protected properties and about £32 billion of avoided economic damage. By the time the auditors reported, the forecast had fallen to 200,000 and roughly 500 schemes had been dropped.
• Inflation explains some of it, not all. Rising costs accounted for between half and two-thirds of the shortfall. They do not account for the £310 million the Environment Agency spent below plan in the programme's first two years, when the money was available and the risk was climbing.
• The maintenance shortfall was small and its consequences were not. For want of £34 million a year in maintenance funding in 2022/23, the NAO calculated that more than 200,000 properties sat at increased risk behind defences allowed to deteriorate.
• Nobody could say whether the country was getting safer. The Public Accounts Committee's finding was that government "does not know if it is making the UK more resilient to flooding", having set no overall target for the level of resilience it was trying to reach.
• The pattern is old. An NAO report from 2007 found only 57 per cent of asset systems at target condition and 46 per cent of the high-risk ones, with the Agency estimating it needed another £150 million a year. The same argument, in the same terms, nearly two decades earlier.

The recurrence is the uncomfortable part. Reviews arrive, recommendations are accepted, and the maintenance question returns. It is the through-line of the Pitt Review and its unfinished business as well. It also helps to be clear about who answers for what, because flood risk in England is shared out: the Environment Agency's role in flood risk and planning sits alongside lead local flood authorities, water companies, internal drainage boards and councils, and Defra's role is different again. We set out the full picture in our guide to who is responsible for managing flood risk.

The maintenance warning, and how the target was met

Asset condition has recovered, and the recovery was paid for out of the new-build budget. The Environment Agency reported 92.1 per cent of high-consequence assets at required condition in the second quarter of 2024/25, rising to 93.0 per cent by 1 April 2026 after £108 million was reprioritised into maintenance. In 2025/26 alone, £262 million went on maintaining flood and coastal erosion assets, of which £72 million was moved across from the investment programme.

Every pound moved from construction into repair is a pound not building the schemes that were meant to deliver the 336,000. The two halves of the auditors' criticism are not separate problems; they are the same budget.

That is not a criticism of the decision. Repairing a deteriorating embankment that already protects a town is almost always better value than starting a new scheme elsewhere, and the Agency was under explicit pressure from both committees to do exactly this. But it does mean the delivery shortfall and the maintenance recovery are connected, and that reporting them as separate successes and failures obscures the trade-off being made.

What a condition rating means in practice is a separate question, and a widely misread one. A defence recorded below its required condition has not failed and will usually still perform close to its design intent; we unpack that in below-standard flood defences: protected on paper, vulnerable in practice. It is also why so many owners ask why they are still in a flood zone after new defences were built. A defence reduces risk. It very rarely removes it.

The funding gap the Environmental Audit Committee put a number on

In October 2025 the Environmental Audit Committee reported on flood resilience in England and did something the earlier reviews had not: it set the committed spending against what the responsible bodies say is actually needed.

The ten-year £7.9 billion capital programme works out at roughly £790 million a year. The Environment Agency's own stated minimum, to stop risk getting worse, is £1 billion a year. The committee's recommendation, drawing on National Infrastructure Commission work, was that the budget should rise to at least £1.5 billion a year by 2030.

The committee also traced the condition decline in plain terms, from 98 per cent of high-consequence assets in required condition in 2018/19 down to 93 per cent by 2024, affecting around 200,000 properties. And it set out the scale of what the money is being asked to cover: 6.3 million properties in England at risk from rivers, the sea or surface water, of which 4.6 million face surface water risk, a 43 per cent increase on the 2018 assessment. Properties at the highest river and coastal risk are projected to rise by 90 per cent by the 2080s, and surface water risk by 200 per cent.

£790 million a year committed. £1 billion a year is the Environment Agency's stated minimum to stop risk rising. £1.5 billion a year by 2030 is what the committee recommended.

The committee's chair, Toby Perkins MP, made the accountability point more sharply than any of the funding arithmetic does. Announcing the report, he observed that when a house is on fire you know exactly who to call and what to expect from them, and that flooding offers no such clarity. Two-thirds of England, the committee noted, is classified as floodable.

The government's response, published in December 2025, accepted a good deal of it: national flood resilience standards, statutory duties for catchment-scale planning, at least £300 million for natural flood management over ten years, and a unified flood reporting service. It also conceded that not every flood risk asset in England is currently tracked, which is a plainer admission than it first appears. We covered the committee's findings when they landed, in flood resilience in England: a wake-up call.

Bigger budgets, unanswered delivery questions

The funding picture has improved markedly, and credit where it is due. The Spending Review reset the numbers upward: government will invest £10.5 billion in flood defences by 2035/36, the largest flood programme in the country's history, combining a two-year £2.65 billion settlement to March 2026 with a £7.9 billion ten-year capital programme that began in April 2026.

The recent outturn supports the optimism. In the two years to March 2026 nearly 62,000 properties were better protected, against a two-year target of 52,000. That target was beaten, and the first year of the settlement alone accounted for more than 24,000 of them. The detail sits in Defra's £10.5bn funding overhaul, the £7.9bn infrastructure commitment and the £1.4bn confirmed for 2026/27.

How the money is shared out has changed too. From April 2026 a new FCERM funding framework fully funds prioritised projects worth £3 million or less and puts natural flood management on the same footing as concrete and steel, the biggest reform to floods funding in fifteen years.

It would be easy to read all of that as the problem solved. The auditors' own caution was that pressure to hit a headline figure pushes a programme towards rushed decisions and cost overruns, and a larger budget makes that risk bigger rather than smaller. Delivery risk has not gone away either: the delays to the River Thames Scheme are a live example of a funded, approved defence that is not yet protecting anybody. A defence that is funded protects nobody until it is built, and then kept in repair.

There is a structural limit as well. The largest single source of flood risk in England is surface water, and it is the least suited to capital defence spending, because the risk is dispersed across millions of individual sites rather than concentrated behind a line on a map. Most of it will be managed site by site, through drainage design and the planning system, not by a national programme. That is the subject of our guide to surface water flooding and planning.

What the defence gap means for your site

The practical consequence is that a nearby defence tells you much less about your site than people assume, and the audits of England's flood defences are the reason why. National mapping shows broad probability across an area; it does not tell you what happens on your plot when a defence is overtopped, breached, or simply waiting its turn in a maintenance queue. Five things follow from that.

1. Identify the defence and who owns it. Environment Agency, lead local flood authority, internal drainage board, water company or riparian owner. Third-party assets fall below required condition more often, and they are not covered by the national condition target.
2. Find its standard of protection and its condition rating. A wall built to a 1 in 100 standard in 1985 is not offering that today once climate change allowances are applied.
3. Do not assume the defence changes your Flood Zone. Flood Zones on the Flood Map for Planning are undefended extents by design, so the sequential and exception test still applies to the mapped zone regardless of what has been built nearby. Areas benefiting from defences are mapped separately.
4. Test the failure case, not the design case. Overtopping and breach scenarios, finished floor levels, and safe access and egress. Applications on defended sites are usually won or lost here rather than on the mapped zone, and the case often needs site-specific flood modelling rather than a reading off the published map.
5. Design around the residual risk that remains. What is left after the defence has done its job is residual flood risk, and demonstrating you have managed it is a requirement, not a courtesy.

For buyers, a defended location is not a risk-free one, and an automated search will not tell you the difference. An independent flood risk survey for a property purchase tells you what the broad datasets cannot: how flooding would actually reach that building, and what the defence in front of it is rated to withstand. Where an insurer has flagged a property, a property-specific flood risk assessment for insurance can support a better underwriting decision than screening alone produces. And for developers, the wider debate about new homes in flood risk areas is exactly the argument these audit findings feed.

Frequently asked questions

England's flood defences are improving, and the honest position is still that no household or developer should rely on them alone while protection lags behind the risk. If you are weighing up a purchase, a planning application or an insurance question, our flood risk assessments give you the site-specific picture rather than the reassuring average, and our chartered flood risk consultants can tell you within a call whether the defence in front of your site changes anything. Call +44 (0) 1293 214444 or email enquiries@unda.co.uk.