Editorial Article
The UK audited 57 car park operators and rated 23 red: 40.35 per cent failed a test with no statutory damages attached, and 17 million keeper requests a year made the gate worth closing
The United Kingdom does not price a mistake. It prices the purpose, and then it audits the operator — and in 2025/26 that audit rated 23 of 57 car park management companies red.
40.35 per cent. Five more were amber, 8.77 per cent, and 29 green at 50.88. A red rating typically means significant weaknesses or serious compliance concerns, including possible misuse of keeper data or inadequate safeguards. Across a decade of assessments the findings include one company rated red in five of its six audits, and another rated red in all three of its audits including the most recent.
No statutory damages attach to any of that. There is no liquidated sum, no per-person cause of action, no class. One wrong record in Britain costs the operator nothing in law. What it costs is a rating.
That is a completely different regulatory mechanism from the one that governs the same business in the United States, where 18 U.S.C. §2724 sets a floor of USD 2,500 on a single improper lookup and requires nobody to prove they were harmed. The two regimes regulate the same act — a plate resolved into an accountable person — and they locate cost in opposite places. Understanding which cost lands where is the difference between building a compliant product and building a product that is compliant in one jurisdiction and unsellable in the next.
The mechanism: reasonable cause, at cost, audited
Registered keeper information is released by the Driver and Vehicle Licensing Agency where an applicant can show “reasonable cause” under the Road Vehicles (Registration and Licensing) Regulations 2002, and private parking enforcement is accepted as reasonable cause. The DVLA’s own request route is the front door for it.
Three design choices in that arrangement matter commercially.
The first is the fee: £2.50 per record, set to recover administrative cost. The agency is not permitted to profit from it. This is not a market price and it is not a deterrent — it is a cost-recovery charge, which means it cannot be used to ration demand. Volume can grow without limit and the charge stays at cost.
The second is that access requires membership of an accredited trade body — the British Parking Association or the International Parking Community. The permission to look up a keeper is bundled with membership of a club that can expel you.
The third is oversight by audit rather than litigation. The Government Internal Audit Agency assessed those 57 companies. Nobody sued anybody. The output was a rating.
| Regime | Primary instrument | What it prices | Where the cost lands |
|---|---|---|---|
| United States, federal | Driver’s Privacy Protection Act, 18 U.S.C. §§2721–2725 | The act of obtaining a record, at a USD 2,500 statutory floor per person | On the exception, per person, with fees |
| United States, states | 28 state ALPR statutes; retention, sharing and audit rules | Storage duration and access logging | Up front, in architecture |
| United Kingdom | Keeper-at-date-of-event regime; £2.50 per record; accredited trade body; government audit | The permitted purpose, continuously | In the audit rating, which gates access |
| European Union | GDPR purpose limitation, minimisation and storage limitation | Lawfulness of basis and duration | In the decision architecture |
Read the last column twice. Three of the four regimes move the cost forward, into how the system is built. Only the American federal statute moves it backward, into what happens after the system made a decision.
Seventeen million lookups and what they cost
Volume is the other half of the British picture, and it is the half that became uncomfortable this year. In the twelve months to the end of March 2026, private parking companies issued a record 17 million requests for keeper details. At £2.50 a record that is a bill of £42.5 million — roughly 46,575 requests a day, every day, for a year.
Because the fee is cost-recovery, that £42.5m is a cost paid by operators to identify drivers, and revenue to nobody. From the agency’s side it is a volume of administrative work to be funded at cost. From the operator’s side it is the price of being able to send an invoice at all.
Set that against the American arithmetic and the two regimes stop looking like variations on a theme. Seventeen million UK keeper requests, at £2.50 each, is £42.5 million. Seventeen thousand records at the United States statutory floor of USD 2,500 is also £42.5 million, at the same notional conversion. A thousandth of the traffic, the same money — because one regime charges per query and the other charges per failure to justify it.
Which regime is cheaper depends entirely on an operator’s accuracy, and that is the point of the design. A careful operator prefers the American model: lookups are effectively free and errors are rare. A careless one prefers the British model, because in Britain a careless operator pays £2.50 whether the record is used correctly or not, and in America it pays USD 2,500 when it is not.
Why a rating is not yet a penalty
Here is the weakness in the British arrangement, and it is structural rather than a matter of enforcement effort. A red rating is, today, a reputational signal. It does not remove the right to query the database. The 23 companies rated red in 2025/26 could, on the published record, keep requesting keeper details at the same rate.
A Private Parking Code of Practice is expected to change that by imposing operating standards as a condition of database access. If it does, the rating stops being a signal and becomes a gate — and the £42.5m of annual keeper-request volume acquires a distribution that is currently irrelevant.
The scale of that gate is straightforward to illustrate, though not to state precisely, because request volume is not published per operator. Pro rata across 57 assessed companies the £42.5m is about £745,598 each, which is an average of a distribution nobody has disclosed. If the red-rated 40.35 per cent carried their pro rata share, it would be £17.15 million of annual keeper-request volume sitting with operators who currently cannot pass an audit — and who would have to remediate or leave the business.
That is a Pheonix illustration on published totals, not a disclosed figure. It is also the most likely single change to the structure of the British market, and it requires no legislation at all: a code of practice attaching a condition to a database permission.
The third regime, which nobody prices but everybody pays for
Retention is where the American states, not the federal statute, determine cost — and the spread is extreme. At least 28 states have statutes expressly addressing ALPR use or retention, and at least 19 impose retention requirements, with three more requiring a local policy. The range runs from three minutes to five years.
| Jurisdiction | Maximum retention | In minutes | Multiple of the shortest |
|---|---|---|---|
| New Hampshire | 3 minutes | 3 | 1x |
| Maine | 21 days | 30,240 | 10,080x |
| Kentucky | 90 days | 129,600 | 43,200x |
| Montana | 18 months | Approx. 788,400 | Approx. 262,800x |
| Colorado | 3 years | Approx. 1,576,800 | Approx. 525,600x |
| Alabama | 5 years | 2,628,000 | 876,000x |
Three minutes and five years are not two settings of the same product. They are two products. A three-minute cap imposes ephemeral matching: the plate is compared at the point of decision and the history is never written. A five-year cap makes the archive the asset — searchable, licensable, subpoenable, and the thing a plaintiff’s lawyer asks for.
Two states illustrate how differently the same intent can be drafted. California imposes a full operational code: reasonable security, a public usage and privacy policy, an access record carrying date, time, plate or query, username and purpose, a designated custodian, a public comment opportunity before a public agency implements a system, a prohibition on sale or transfer except as authorised by law, and a civil cause of action for breach with actual and punitive damages and fees. A vendor complies with California by improving process.
Connecticut’s 2026 statute drafts from the other end: a short list of permitted purposes that expressly includes parking enforcement and access control, a 21-day deletion default, and a 30-day carve-out for highway usage fee collection. Against Alabama’s five years, Connecticut’s 21 days is a factor of 86.95. A vendor complies with Connecticut only by changing architecture — and one that has already built for Connecticut can sell into California, while one built for Alabama cannot sell into either without rebuilding.
Retention caps are, in other words, a competitive filter that operates at the design stage and costs nothing to administer. They are the reason a hardware company’s roadmap is a legal document as much as a technical one — the same point that applies to the sensing layer, where the occupancy sensing market has been forced into edge processing for identical reasons of retention and minimisation.
What converges, and what it means for the forecast
All four regimes are converging on the same architecture from different motives. Under the GDPR the constraints are purpose limitation, data minimisation, storage limitation and a lawful basis, and applied to an access reader they produce the instruction to match at the point of decision, retain the decision rather than the history, and document the purpose. Europe arrives there for reasons of principle. The United States arrives there for reasons of statutory damages. Britain arrives there because an audit trail is the only evidence that the purpose existed.
What does not converge is who pays. In the United Kingdom the lookup is cheap by policy and the compliance is expensive by audit. In the United States the lookup is cheap in practice and catastrophic in exception. In retention-capped jurisdictions the cost is borne up front, in architecture. Three mechanisms, three balance sheets, one market.
And a market that is repricing, because the arm carrying the legal risk is the arm growing fastest. The global market for automated license plate recognition and vehicle identity for access runs from USD 4,180.0m in 2025 to USD 12,150.0m in 2033 at 14.27 per cent, on readers in service growing from 1,860 thousand to 5,240 thousand at 13.82. Revenue per reader is nearly flat across those eight years — USD 2,247.31 to USD 2,318.70, up 3.18 per cent — and the flatness hides the reallocation underneath it: hardware per reader down 29.72 per cent, identity per reader up 61.10 per cent.
Identity revenue rises from 18.66 per cent of the market to 29.14 at 20.81 per cent a year, contributing 34.64 per cent of the USD 7,970.0m the market adds. Europe grows at 13.89 per cent, slower than Asia-Pacific’s 16.18, and the reason is not technology adoption — it is that Europe’s retention and purpose rules were settled before the market scaled, and the compliance has already been capitalised into the product. Britain’s audit regime is a variant on that: expensive to satisfy, cheap to operate once satisfied.
What the fieldwork found
Two findings from our own programme sit under this and they are ours rather than published statistics. The programme behind this research comprised 31 semi-structured interviews across seven markets, nine conference-floor interviews at a European traffic-technology exhibition, one closed roundtable with five operators and one toll authority, and 47 written survey responses.
The first: operators rarely price recognition error into a contract at all. The tolerance, where it exists, is an assurance in a service level agreement about uptime, not about correctness — and correctness is the only dimension the law prices. The second: indemnity is negotiated as a substitute for accuracy rather than as a consequence of failing to achieve it. A clause that transfers the cost of a wrong record is treated, in negotiation, as having discharged the obligation to read correctly. It has not; it has merely moved who pays.
Both findings read differently in each jurisdiction. In the United States a supplier that indemnifies its customer has taken on a liability with a statutory floor. In Britain it has taken on an audit rating it cannot see and a membership it can lose.
Who is exposed
The suppliers in this market are the ones whose products carry the resolution step, which is why the consolidation has been toward databases rather than optics. Motorola Solutions paid USD 445m for VaaS on roughly USD 100m of revenue — 4.45 times — and Metropolis acquired SP Plus to convert around 4,000 locations. Flock Safety reports over 100,000 readers across about 12,000 customers. In Europe the fixed hardware sits with Kapsch TrafficCom, Jenoptik and Vitronic, with Genetec and Conduent in integration and back office. Verra Mobility is the clearest listed comparable for the operating layer and carries no Wikidata entity — no identifier survives verification — so it is named here without a link.
The adjacent markets sit one step away and share the same regulatory edge: the on-street enforcement and digital permitting market, where the citation is the product and the plate is the evidence; the curb and kerbside management market, where the same plate resolves into a permit or a schedule; the concessions market, where the operator holds the contract and outsources the liability; and the mobile payment platforms where the session record becomes the evidence of an event. Every one of them depends on the same act of resolution, and every one of them is exposed to whichever of the three mechanisms its jurisdiction chose.
Three things that would move it
First, a federal amendment narrowing or widening the permitted purposes, or altering the liquidated damages provision, would reprice the entire United States identity arm inside a single legislative cycle. Second, a state retention floor of 24 hours or less in a large market would strand archive-based products and accelerate the shift to query-time resolution. Third, a certified class in the Florida litigation would convert a theoretical exposure into a claims estimate and depress valuations across private-parking assets. None of the three is in the model, and all three sit inside the forecast window.
The British equivalent is smaller and closer: a code of practice that attaches operating standards to database access. It would cost the Treasury nothing, require no new statute, and reprice 40.35 per cent of an industry by making a rating mean something.
The audit is the sentence
Britain’s answer to the DPPA is not a smaller penalty. It is a different instrument, applied at a different moment, by a different party. The American statute asks after the fact whether the purpose existed and charges USD 2,500 per person for the answer. The British regime asks before and during, through an audit that rates the operator, and holds the database over them.
Seventeen million requests a year at £2.50 is what the second model costs to operate, and 40.35 per cent of operators failing the test is what it currently finds. The gate exists. It has simply not been closed yet — and when it closes, the penalty for a wrong record in Britain will not be USD 2,500. It will be the loss of the right to look the record up at all.
Frequently Asked Questions
How many UK car park operators failed their data-protection audit?
In 2025/26 the Government Internal Audit Agency assessed 57 car park management companies. Twenty-three were rated red — 40.35 per cent — typically indicating significant weaknesses or serious compliance concerns, including possible misuse of keeper data or inadequate safeguards. Five were amber (8.77 per cent) and 29 were green (50.88 per cent).
What penalty applies to a UK operator that misuses keeper data?
There is no statutory damages regime. Keeper information is released by the DVLA where reasonable cause is shown, at £2.50 per record — a fee set to recover administrative cost, which the agency is not permitted to profit from. Misconduct is addressed by audit rating, and access requires membership of an accredited trade body such as the BPA or IPC.
How many keeper requests do UK private parking companies make, and what do they cost?
In the year to the end of March 2026, private parking companies issued a record 17 million requests for registered keeper details. At £2.50 a record that is £42.5 million, roughly 46,575 requests a day — and because the fee recovers cost rather than raising revenue, it is a cost to operators and revenue to nobody.
How does the UK system differ from the US Driver's Privacy Protection Act?
They locate cost in opposite places. The US prices the exception — USD 2,500 per improper record at the federal floor, per person, with fees. The UK prices the purpose continuously and audits the operator, so the penalty is a rating that can eventually remove database access rather than a payment per event.
