Everything Is Working Exactly As Designed

Fish Discos, Bat Tunnels
and a Government Spreadsheet.

In Buckinghamshire, HS2 built a kilometre-long mesh tunnel to protect 300 bats. It cost £216 million. At Hinkley Point C, 300 underwater speakers repel fish from intake pipes that suck in 134 cubic metres of seawater per second. The fish protection measures cost £700 million and are expected to save 0.083 salmon per year.

Everyone agreed this was insane. The government agreed too. Environment Secretary Steve Reed called the bat tunnel “ludicrous.” Rachel Reeves said she wanted industry to “stop worrying about the bats and the newts.” The solution wasn’t to fix the planning system. It was to replace it with a market — one that turns nature into a financial instrument. Instead of protecting the actual bats in the actual wood, developers pay into a fund. The fund handles nature. The bat tunnel made that sound reasonable.

This is what the new system does. It doesn’t measure loss. It banks it.

Since February 2024, every new development in England must leave habitats 10% better than before. Developers who can’t deliver the improvement on-site buy credits — tradable financial instruments — from landowners who can. The credits are measured by a government spreadsheet that scores habitats by size, type, condition and location. The difference between a degraded baseline and a restored one is calculated, priced, registered, and sold.

The system does not measure environmental health. It measures the gap between a degraded state and a restored one — and that definition makes degradation a prerequisite, not a problem. Existing stewardship must be removed or must fail so that institutional restoration can begin from a baseline low enough to generate tradable credits. The displacement is administered by institutions that simultaneously set the conditions, create the market, and enforce against anyone who does the work outside the system.

The environmental outcomes can be real. Oysters are genuinely restored. Trees are genuinely planted. The science is genuinely done. But a functioning ecosystem replaced by an instrumented one is not necessarily a gain. The environment may improve, stay the same, or get worse. The data always improves. And data is the product.

The system is not being gamed. The system is the game.

This piece draws together four case studies — a farmer, a barrister, an oysterman, and a piece of woodland — and explains the mechanism behind all of them.

On Dartmoor, semi-wild ponies have grazed the moor for 4,500 years. They are classed as endangered. Peer-reviewed science — which Natural England accepts — shows that pony grazing is the most effective tool for managing the invasive Molinia grass choking the moor’s protected habitats. Natural England has demanded livestock reductions of up to 89%. The ponies will be the first to go: cattle and sheep pay the bills, ponies don’t. The cull isn’t ordered. It’s made financially inevitable. The ponies’ grazing IS the biodiversity management — but pony grazing doesn’t generate Biodiversity Units. What the system can’t measure, it can’t value. What it can’t value, it replaces with something it can.

On the River Roding in East London, a barrister named Paul Powlesland pulled over 200 bags of rubbish out of the water. The Environment Agency is prosecuting him — operating without a permit carries up to two years in prison. A few hundred metres upstream, Thames Water pumps 750 million litres of raw sewage into the same river. Not under investigation. The system generates credits from the gap between a degraded baseline and a restored one. A volunteer who cleans a river for free doesn’t just fail to produce credits. He destroys the baseline that makes the credits valuable. He improved 250 metres of waterway in a way the system cannot monetise, and in doing so, he burned the value the system intended to sell.

On the River Avon in Devon, Tim Edwards is cutting up trestles. His oyster farm has been here for fifty years — 125,000 oysters a year, no feed, no chemicals, sterile stock. A government-funded study conducted on his farm found it enhances biodiversity and improves water quality. The Duchy of Cornwall is removing him. His oysters are classified as “invasive.” Native oysters can’t survive on this river — he tried, they died. Nobody is replacing him with a native restoration project. The estuary will simply lose its oysters. For the estuary, the loss counts now — less filtration, reduced biodiversity, today. For the system, the loss counts later — when the degraded baseline becomes the floor from which future credits are measured. The system doesn’t register the removal as damage. It registers it as potential. Four hundred miles away, a pet food company restores four million native oysters off Norfolk, generating credits for a pension fund targeting 8–12% returns. The metric registers national gain. The Avon has no oysters. The spreadsheet doesn’t know the difference, because ecology is location, and the metric has abstracted location out of ecology.

And across eight million acres of UK woodland — 97% of it unregistered with any credit scheme — the same logic operates without a person in the frame. The woodland sequesters carbon, filters air, stabilises soil, shelters species. Nobody pays it. Nobody measures it. Nobody counts it. The system that claims to value these things cannot see a standing forest, because a standing forest produces no measurable change. Its stewardship is invisible. Its destruction would be an entrance fee.

Natural England demanded the Dartmoor livestock reductions. Natural England runs the Biodiversity Gain Sites Register — the national registry where restoration credits are listed for sale. Natural England administers the nutrient mitigation scheme. Natural England advised the Duchy of Cornwall on the oyster farm. Natural England is building POSEIDON, a marine mapping tool launching summer 2026 to give offshore wind developers access to marine biodiversity data.

The Environment Agency celebrates river cleanup partnerships on one webpage and prosecutes the same work on another. The EA’s own Catchment Data Explorer lists Powlesland’s charity as a catchment partnership contributor. The same website celebrates “rubbish clearance” and “clearing of invasive Himalayan Balsam” along the River Roding as success highlights. The partnership arm celebrates the work. The enforcement arm prosecutes it.

They set the conditions that remove existing practice. They administer the market that the replacement enters. And they prosecute anyone who does the work outside the system.

The mechanism behind all four case studies is the same, and it extends beyond biodiversity into carbon, nutrients, and ecosystem services.

The system measures deltas — the gap between a degraded state and a restored one. A degraded field restored to wildflower meadow: measurable. A scrubby riverbank enhanced to functioning wetland: measurable. A thousand acres of woodland standing quietly, doing what it’s done for a century: not measurable. There is no change. There is no unit to generate, register, or sell.

The Woodland Carbon Code — the UK’s standard for generating tradable carbon credits from trees — applies only to new woodland creation on land that has not been wooded in the last 25 years. Existing woodland is ineligible. The same woodland can receive Countryside Stewardship payments for management — but what qualifies as “good management” under the Biodiversity Metric includes retaining standing and fallen deadwood. Deadwood is a habitat indicator. It scores positively. Dead material is also fuel.

No scheme, no grant, no credit, and no payment exists for actively reducing wildfire risk on your own woodland. Every funded activity accumulates fuel. The one activity that would reduce it is unfunded. In a country that just recorded its worst wildfire year in history, the absence of any programme that pays for fuel management is not an oversight in any of the published frameworks. It is a gap that every framework leaves in exactly the same place. The full mechanism — from carbon accounting to national climate targets to the institutional investors assembling behind degraded land — is documented in When Trees Aren’t Carbon.

Edwards’s oyster farm. Powlesland’s stretch of river. The ponies on Dartmoor. The woodland. Each one was stewardship producing no delta. The system couldn’t see them. What it could see — and what it could sell — was the gap that their removal would create.

The system doesn’t measure the health of the river, the moor, the estuary, or the forest. It measures the gap between a degraded state and a restored one. Stewardship closes that gap for free, which makes it invisible. Degradation opens it, which makes it valuable. The mechanism doesn’t require malice. It requires measurement — and the measurement only works in one direction.

The architecture documented here — genuine environmental concern converted into centralised data collection, algorithmic classification, and tradable compliance instruments — is not unique to nature. It runs simultaneously across every sector of public and economic life. A genuine concern. An engineered response requiring centralised data, algorithmic classification, surveillance, and tradable compliance. The substrate changes. The architecture doesn’t.

In packaging, Extended Producer Responsibility modulates fees by a Recyclability Assessment Methodology that penalises glass — infinitely recyclable, inert, found in no living tissue — and advantages plastic, which is found in brain tissue, placentas, and bloodstreams. The metric scores by weight, not by health impact.

In housing, the Energy Performance Certificate algorithm scores inputs, not outcomes. A cold, expensive, recently retrofitted house outranks a warm, cheap one with lower bills. The certificate measures what was installed, not whether anyone is warmer.

In transport, the Zero Emission Vehicle mandate creates a credit market in which manufacturers trade compliance rather than reduce emissions. The Ultra Low Emission Zone classifies vehicles by age, not by actual emissions — a well-maintained 2014 diesel producing less particulate than a worn 2016 petrol fails the test. The metric classifies the vehicle, not the air.

In farming, digital animal identification applies the same compliance burden to twelve sheep as to ten thousand head. Environmental Land Management payments reward environmental data production rather than food production. The farmer is paid to generate legible outcomes, not to farm.

In waste, your own rubble on your own land requires a digital waste transfer note. The tracking system doesn’t reduce fly-tipping — it creates a digital record of compliant disposal and a financial incentive for non-compliant disposal by anyone who can’t afford the compliance.

In water, tradable abstraction rights turn a 200-year user into a market participant. In energy, time-of-use tariffs generate granular consumption data that flows to retailers and demand aggregators. The meter produces data. Whether it produces savings depends on who can afford to shift their usage.

One row in this table behaves differently from the others. In every other sector, the metric measures the wrong thing — habitat type instead of animals, inputs instead of outcomes, weight instead of health impact. In planning, the metric measures the right thing — cumulative impact — and nobody is required to add it up. Environmental impact assessment requires in-combination effects to be considered, but the assessment is scoped by the applicant, commissioned by the applicant, paid for by the applicant, and bounded to the scheme in front of it. No body holds the running total. The sum is required of the party with the strongest interest in it being small — and the absence of anyone else who might produce it is not an oversight. An oversight would imply someone was overseeing.

In every case: a compliance industry that doesn’t make, grow, or restore anything — it makes the system legible and charges for the service. Behind the compliance industry, institutional capital: pension funds, habitat banks, credit brokers, converting landowners, CCUS operators. They invest in the market the architecture created. They gain when the market scales. Behind them, the architecture itself: the centralised data infrastructure, the mandatory digital accounts, the audit trails, the algorithmic classification systems, the tokenisation rails. This is the permanent product of the entire exercise.

Without this top-tier demand — the requirement that everything be measured, classified, registered, tracked, tokenised, and made legible to centralised systems — the first two tiers wouldn’t exist. There would be no compliance industry because there would be nothing to comply with. There would be no credit market because there would be nothing to trade. The ponies would just graze. The man would just clean the river. The oysterman would just farm oysters.

Teesworks, on the south bank of the Tees. £560 million of public money spent regenerating contaminated industrial land. That land sold at £1 per acre to private interests who now own 90%. A freeport with tax breaks, relaxed planning rules, and reduced regulation. A Carbon Capture, Utilisation and Storage facility capturing 10 million tonnes of CO₂ per year — not stopping the emissions, but capturing them and generating tradable carbon credits. An AI Growth Zone, with discussions for what is slated to become Europe’s largest data centre. 5,600 jobs delivered against 18,000 promised. The government blocked independent audit.

Freeport tax breaks. Public subsidy. Carbon credits. AI infrastructure. Tokenisation rails. Every layer of the architecture — from the livestock formula on Dartmoor to the oyster credits off Norfolk — converges on this site. The ponies, the river, and the oysters are at the edge of the system. Teesworks is at the centre.

Biodiversity Units are already registered financial instruments. Nutrient credits are already tradable. Carbon credits already have a market. Ecosystem service payments are already being stacked on land.

On 18 May 2026, the Financial Conduct Authority and the Bank of England jointly published a call for input on the future of tokenisation in UK wholesale markets. Sixteen firms are already in the FCA’s Digital Securities Sandbox — a live, regulated environment for testing tokenised issuance, trading, and settlement. In April, PS 26/7 moved fund tokenisation from experimentation to adoption. The FCA and the Bank of England explicitly stated they “expect to look beyond” securities. The rails exist. The assets are already on them. The only question is speed.

One parcel of land generating biodiversity credits, nutrient credits, and carbon credits simultaneously. Three revenue streams, all registered, all tradable, all locked under conservation covenants at the Land Registry. The land isn’t confiscated. It’s tokenised — its ecological value quantified, registered, and entered into financial markets as a tradable asset.

Natural England published its strategy. The FCA published its vision. The Duchy published its policy. The Rivers Trust published its funding model. None of them hid. The plans are in public documents with reference numbers and consultation deadlines. The absence of anyone whose job it is to add them up is not an oversight. An oversight would imply someone was overseeing.

No electorate would vote for mandatory digital tracking of every product supply chain, every farm, every waste movement, every vehicle, every building, every water abstraction. But wrap each one in a genuine environmental concern, and each one arrives with a consultation document, a statutory metric, and a three-letter acronym.

The metrics aren’t failing. They’re succeeding — at building a permanent governance infrastructure that would never have been politically achievable if proposed on its own terms.

A farmer at an October roundup, choosing between ponies and cattle. A barrister with a bin bag and a hired digger. An oysterman cutting up trestles on a Devon mudflat. A thousand acres of woodland, doing everything the system says it values, invisibly. Four places. The same spreadsheet.

Everything is working exactly as designed.

Sources: Environment Act 2021; Gov.uk — Planning and Infrastructure Act 2025; FCA (18 May 2026) — “The Future of Tokenisation”; FCA — PS 26/7; The Times (15 Jun 2026) — Dartmoor ponies; The Guardian (17 Jun 2026) — Powlesland prosecution; BBC News (Jun 2026) — Edwards / River Avon; Envision Marine & University of Essex (2024); Purina UK / Nestlé (Sep 2025); Rebalance Earth / IPE Real Assets (Sep 2025); RWE (Jul 2025); EDP24 (Jun 2026) — Equinor / Norfolk fishermen; Teesworks.co.uk; Natural England — Biodiversity Gain Sites Register; UK Woodland Carbon Code — Statistics (Mar 2026); Forest Research — Provisional Woodland Statistics (Jun 2025); Gov.uk (Dec 2025) — Marine Recovery Fund.