The Ghost Tonnes

The Certificate

Here's what most people think happens when they put plastic in the recycling bin:

The council collects it. Someone sorts it. It gets recycled. The packaging companies that made all that plastic pay for the process, because something called "producer responsibility" means the people who create the waste are responsible for dealing with it.

Here's what actually happens:

The council collects it. Someone sorts it. Then a piece of paper enters the story.

A PRN — Packaging Recovery Note — is a certificate issued when packaging waste is recycled in the UK. Its cousin, the PERN — Packaging Export Recovery Note — is issued when waste is shipped abroad for recycling. Packaging companies buy these certificates to meet their legal obligations under the Producer Responsibility Obligations (Packaging Waste) Regulations 2007.⁵

The crucial detail: a PERN is issued when waste is shipped, not when it's recycled.

The British Plastics Federation, which represents the industry, notes a crucial distinction: PRNs are issued only after waste has been recycled, whereas PERNs can be claimed when material is exported, potentially including contaminated and non-recyclable content.⁴

A PRN proves recycling happened. A PERN proves a ship left.

The distinction matters because of what counts. When a company buys a PERN, it is counted as recycled for compliance purposes — even though no one has verified it was recycled in practice.² In 2017, according to the National Audit Office, roughly half of UK packaging waste was exported.² Half of our "recycling" was measured at the dock, not at the destination.

The Database

I followed the paper trail.

The National Packaging Waste Database is where accredited reprocessors and exporters register the tonnes they've handled. It's administered by the Environment Agency. Companies enter their figures. The figures become the national recycling rate.

The system relies on self-reporting. Accredited operators enter their own tonnage figures into the database. The figures are supposed to be accurate. The word "supposed" is doing a lot of work in that sentence.

The EIA's investigation identified 13 distinct methods of committing fraud within this system.¹ They fall into three categories:

Data manipulation — falsifying the tonnage figures entered into the database.

Fictitious tonnage — claiming certificates on material that doesn't exist or isn't packaging waste. In one case documented by the EIA, wheelie bins were relabeled as "crates" to qualify as packaging.¹

Illegal shipments — double-counting the same bale twice, claiming material that was landfilled or incinerated, or shipping waste that never reaches its declared destination.

The EIA sent undercover investigators into the market. They received an open offer of fraud from a company director willing to disguise non-packaging waste as packaging material for fraudulent certificate claims.¹

An industry source told them: "It's a big scam."¹

The Inspectors

Someone is supposed to check.

The Environment Agency is responsible for monitoring the PRN and PERN system, ensuring that accredited operators are legitimate and that the figures in the database correspond to reality.

In 2016-17, according to the National Audit Office, the Environment Agency completed 124 compliance visits against a target of 346.²

That's 36%. Less than 40% of planned inspections actually happened.²

In 2017-18, the Agency carried out 3 unannounced visits to accredited reprocessors and exporters.² Three. Across the entire country. That's 1.4% of accredited sites receiving an unannounced inspection in a year.²

The Agency's own internal audit had raised concerns about the number of unannounced visits in 2015.² Three years later, they managed three.

But here's the detail that stops me: the Environment Agency conducted its own risk analysis of exporters, rating them as high-risk or low-risk. You would expect high-risk exporters to receive more scrutiny.

The NAO found the opposite: "Exporters it rated as high risk were less likely to receive a compliance visit than those rated low risk."²

Only one of four high-risk exporters received a compliance visit in 2017.² A lower proportion than low-risk operators.

The inspection regime wasn't underfunded. It was inverted. The riskiest actors received the least scrutiny.

The Price

The market was doing exactly what markets do.

A PRN and a PERN have the same face value to the producer who buys them. Both count equally toward meeting legal obligations. But the cost of generating them is different.

A domestic PRN requires actually recycling the material — sorting, processing, removing contaminants, producing usable output. All of that costs money.

A PERN requires putting material on a ship.

Viridor, a major waste management company, told Parliament that processing waste abroad was around 20% cheaper for producers than domestic recycling.⁶

The British Plastics Federation confirmed the structural incentive: "The costs incurred in issuing a PRN are much higher than the costs relating to earning a PERN. Therefore, there is currently a financial incentive to export waste rather than process it domestically."⁴

The incentive worked. By 2017, half of UK packaging recycling was exported.² Domestic recycling capacity has been shrinking — estimates suggest around 260,000 tonnes of capacity lost per year since 2022.¹¹

The system told the market: ship it, don't process it. The market listened.

The Methods

The EIA documented 13 ways to commit fraud within the PRN system.¹ I won't list them all. The point isn't the specifics. The point is that all 13 methods are enabled by the same structural gap: the system certifies that someone claimed recycling happened, not that recycling happened.

Double-counting: claim the same bale of waste twice under different certificate numbers. The database doesn't automatically cross-reference.

Fictitious tonnage: enter figures for material that doesn't exist. The database accepts what you type.

Misclassification: label non-packaging waste as packaging. A wheelie bin becomes a "crate." The certificate is issued.

Lost paperwork: ship material, lose the documentation, claim it was recycled at the destination. No one can prove otherwise because no one checks the destination.

Energy recovery: send material to be incinerated, claim it was recycled. The word "recovery" in the regulations covers both.

The system's defenders say fraud is the exception. The EIA estimates it may be 10-15% of the market — GBP 30-50 million annually.¹ But the estimate almost doesn't matter. When your system is designed to measure claims rather than verify outcomes, the difference between fraud and error becomes philosophical.

The Bill

In 2017, UK producers paid GBP 73 million toward the cost of recycling their packaging through the PRN system.²

In the same year, English local authorities spent GBP 700 million on collecting and treating packaging waste.²

GBP 73 million from the companies that created the packaging. GBP 700 million from taxpayers who sorted it into bins. That's roughly 10% producer, 90% public.²

The NAO put it in continental perspective: UK producers paid 13 euros per tonne. In Germany, Belgium, and the Netherlands, producers paid more than 48 euros per tonne.²

UK producers paid 73% less than their European counterparts.²

The difference wasn't an accident. In those countries, packaging schemes require companies to fully fund collection of household packaging waste. In the UK system, companies "only contribute indirectly to collection costs through payments for recovery notes."²

The cost transfer wasn't a failure of the PRN system. It was its central achievement.

For 28 years, UK packaging producers met their legal obligations while paying a fraction of actual costs. The gap was filled by council tax. By your money.

The Regulation

I read the Producer Responsibility Obligations (Packaging Waste) Regulations 2007.⁵ Three times.

The regulations establish who is obligated — companies with turnover above GBP 2 million handling more than 50 tonnes of packaging annually. They define how the obligation is split across the supply chain: 6% for raw material manufacturers, 9% for converters, 37% for packers and fillers, 48% for sellers.⁵

They require obligated companies to obtain PRNs and PERNs as evidence they've met their recycling targets.

What they don't require: verification that the recycling actually happened.

The NAO summarized it: "The quality of exported waste is not monitored, and high-risk exporters were less likely to be inspected."² The regulations focus on accreditation — are you allowed to issue certificates? — and on records — did you submit your paperwork? They focus on inputs, not outcomes.

The word "verify" appears in the regulations. It refers to verifying that an operator is properly accredited. Not to verifying that material was actually recycled.

A PERN certifies that someone shipped waste and claimed it would be recycled. It does not certify that recycling occurred. It cannot certify that, because no one is required to check.

The Success

Here's the twist.

The PRN system met its targets every year. The UK's packaging recycling rate climbed from roughly 25% in 1997 to 64.8% in 2023.⁸ Plastic packaging hit 52.5% — the highest recorded.⁸ Statutory obligations fulfilled. Compliance achieved.

Those rates are calculated from the same PRN and PERN tonnages reported by accredited reprocessors and exporters — the same system I've just described.¹² Even the official success metrics rest on the same evidence structure.

By its own metrics, the system succeeded.

The fraud, the ghost tonnes, the closed domestic facilities, the 90% taxpayer subsidy, the high-risk exporters who were less likely to be inspected — none of that shows up in the success metric. Because the success metric measures certificates counted as recycled for compliance purposes, not material verified as recycled in practice.

The NAO called it "a comfortable way of meeting targets without addressing fundamental issues."² Mary Creagh, then chair of the Environmental Audit Committee, was blunter: "a tick-box exercise."²

I almost admire the precision. Design a measure. Optimize for the measure. Declare success.

This is interpretation, but it's interpretation bolted to facts: a weak verification regime, an inverted inspection system, and an incentive structure that rewarded export over domestic processing. The system didn't fail. The system did exactly what it was designed to do: generate paper compliance at minimal producer cost. The 43,575 unaccounted tonnes aren't a bug. They're the system's signature.

This concern is not new. In 2018, Eunomia Research published a methodological critique suggesting the actual plastic packaging recycling rate might be 9-10 percentage points lower than official figures — accounting for moisture, contamination, and the fundamental measurement problem of counting material at the dock rather than at the destination.⁷ Critics may say that analysis is dated, but it identified the same structural issue: when you count certificates, you count what was claimed, not what was recycled.

The question isn't whether 52.5% or 42.5% of plastic packaging is really recycled. The question is whether a system that measures certificates can ever answer that question.

The Inheritance

Extended Producer Responsibility — EPR — came into force on 1 January 2025, with the first local authority disposal cost invoices due from October 2025.¹⁰ It's supposed to be different.

PackUK, the new scheme administrator, will charge producers modulated fees based on the recyclability of their packaging.⁹ Plastic costs are set at GBP 423 per tonne. The estimated cost shift: approximately GBP 1.2 billion annually from taxpayers to producers.¹⁰

EPR includes tighter evidence requirements. Exporters will face new obligations including proof of destination and recycling.⁹ The direction of travel is toward verification — but PRNs remain "the practical mechanism" during the transition, and the first formal scheme review isn't scheduled until late 2028.⁹

I don't know if EPR will work. I know who designed it: Defra. I know who will administer it: the same agencies. I know what the measure will be: compliance with targets.

The measure doesn't know whether your yogurt pot became a new yogurt pot or became smoke. The measure knows whether someone bought a certificate that said it did.

The first step to closing the gap is seeing where it is. The gap is between the certificate and the material. Between what the system counts as recycled for compliance and what is verified as recycled in practice. Between producer "responsibility" and the taxpayer who pays.

Forty-three thousand, five hundred and seventy-five tonnes of plastic recycling, unaccounted for.

Now you see it.