Press Release
PARCS Market Reaches USD 6.19 Billion by 2033 at 9.03 Per Cent — and Four-Fifths of the Growth Is the Same Bay Earning More
19 November 2026 — The global market for parking access and revenue control systems (PARCS) was worth USD 3.10 billion in 2025, the base year for Pheonix Research’s new report on the segment, reaches USD 3.38 billion in 2026, and is forecast to reach USD 6.19 billion by 2033 — a compound annual growth rate of 9.03 per cent, measured from that 2025 base.
The headline is ordinary. What the research actually finds is not. Almost none of that growth comes from parking. Controlled spaces rise from 40.0 million to 46.5 million across the window — 16.2 per cent in eight years, 1.90 per cent a year. Market value rises 99.7 per cent. The difference, 78.2 per cent of the total increase, is the same bay earning more from whoever controls access to it: revenue per controlled space moves from USD 77.50 to USD 133.12 a year, up 71.8 per cent, on a physical asset count that barely moves.
This is not a volume story wearing a value story’s clothes. It is a change in what a parking supplier sells, and four real procurements make the mechanism visible.
Four procurements, and the falling price of a space
Pheonix’s model is built from awarded contracts rather than from vendor price lists, because vendors do not publish prices and consultants do not publish bids. Four public procurements carry the argument, and they run in one direction.
Houston Airports, 2022. A full replacement of the access and revenue control estate across George Bush Intercontinental and William P. Hobby — roughly 30,000 spaces — was awarded at USD 14,143,015, or USD 471.43 per space. Cameras and licence-plate recognition at every lane, barrier gates throughout, the whole classical PARCS build.
Evanston, Illinois, 2025. Three municipal decks totalling 3,583 spaces were awarded at USD 969,823, or USD 270.67 per space. Seven firms bid. The top-scoring proposal, at USD 1,278,823, was rejected — a 31.9 per cent spread over the winning bid — along with a gateless option the city declined.
Irvine Great Park, California, 2026. 4,419 spaces at USD 1,083,377, or USD 245.16 per space — and this one is gateless, run through kiosk, web, app and QR rather than through a barrier.
Three procurements, three years, and the capital cost of equipping a space falls by roughly half. That fall is the most important input in the model, because it changes what the supplier has to sell in order to grow.
The number the report rests on
San Antonio, 2024. Around 10,000 spaces, awarded at USD 847,000 a year — USD 84.70 per space per year — over a ten-year term, with no upfront equipment cost at all. The vendor installs the cameras and the gates, and the fee moves to the driver: 99 cents to USD 3.99 per transaction, collected at the point of exit.
Over the full term that structure costs the buyer USD 847 per space. That is 1.80 times the highest capital price observed in this research and 3.45 times the lowest. And it is why the gate can be given away. A vendor that is paid per transaction does not need to sell hardware; it needs the transaction, and it will finance the hardware to get it.
The economics do not stop at the equipment line either. Evanston’s own disclosed service history on a nine-year-old system runs at USD 34.61 per space per year — 130 invoices totalling more than USD 125,000 in 2023, 119 totalling more than USD 123,000 in 2024. Maintenance on installed access control is a permanent annuity, and it is the part of the stack that shrinks slowest when the capital ticket falls.
The replacement is triggered by an audit, not by a barrier
Nobody replaces a working gate because it is old. They replace it because the payment path it terminates no longer passes an assessment.
PCI DSS v4.0.1 retired v4.0 on 31 December 2024, and the future-dated requirements — the ones that were announced in 2022 and deferred — took effect on 31 March 2025. A lane controller commissioned in 2016, terminating a card reader built to an earlier standard, fails a compliance assessment years before it fails mechanically. That is the forcing function behind the procurement cycle, and it is why the 2025 and 2026 awards in this dataset are replacements rather than greenfield.
Regulation reaches the money as well as the card data. New Jersey’s parking fee statute, N.J.S.A. 56:8-2.33, defines a parking fee broadly and then carves out exemptions in defined circumstances; operators who read the exemption narrowly and price broadly have a compliance exposure that has nothing to do with parking and everything to do with how the system is configured to charge. Configuration is a PARCS function. That is the sense in which this market is now sold by the transaction rather than by the lane.
Where the money moves inside the stack
The layer split is where the report departs most sharply from how the market is usually described. Hardware is not the growth line. Software is.
| Layer | 2025 share | 2026 share | 2033 share | Change | CAGR 2025–2033 |
|---|---|---|---|---|---|
| Lanes and access hardware | 33.9% | 33.0% | 22.6% | −11.3 pts | 3.64% |
| Software and platform | 20.0% | 20.7% | 29.9% | +9.9 pts | 14.65% |
| Service and maintenance | 25.2% | 24.8% | 21.3% | −3.9 pts | 6.76% |
| Transaction and enforcement | 21.0% | 21.5% | 26.2% | +5.2 pts | 12.09% |
Software compounds at 14.65 per cent and transaction and enforcement at 12.09 per cent, against 9.03 for the market. Lanes and access hardware compounds at 3.64 per cent — a third of the market rate — while service and maintenance runs at 6.76. In plain terms: the box on the post loses share every year, and the thing that reads the plate, prices the stay, collects the money and enforces the outcome gains it.
That is also the explanation for the falling capital price per space. A gateless deck at USD 245 a space is not a cheaper version of a gated one. It is a different product, in which the supplier’s revenue sits on the transaction and the barrier is an expense it would rather not carry.
Consolidation is repricing the layer above the hardware
The ownership of this market changed in the same window, and the change is instructive about where the margin is believed to be.
Metropolis took SP+ private at USD 1.8 billion — USD 54.00 a share, a 52 per cent premium — and followed it in November 2025 with USD 1.6 billion of financing at a valuation of roughly USD 5 billion. That is a company whose product is checkout, not gates, buying the operator that stands inside the asset. On 12 June 2025 Arrive was formed, consolidating a set of parking technology and payment businesses into a single platform. In the UK, Mansfield District Council awarded a five-year parking contract on 16 June 2025, of the kind that used to be decided on equipment price and is now decided on the service wrapper around it.
None of this is visible in a hardware price index. All of it is visible in revenue per controlled space, which is the number that actually moved.
Key findings
- USD 3.10 billion in 2025, the base year; USD 3.38 billion in 2026; USD 6.19 billion in 2033 — 9.03 per cent, measured from the 2025 base.
- Controlled spaces rise only 40.0 million to 46.5 million at 1.90 per cent a year, while value rises 99.7 per cent. 78.2 per cent of the growth is price per space, not space.
- Revenue per controlled space climbs USD 77.50 to USD 133.12 a year, 71.8 per cent across the window — the single line that explains the forecast.
- Capital prices per space are falling: USD 471.43 (Houston, 2022), USD 270.67 (Evanston, 2025), USD 245.16 (Irvine, 2026). The cheapest of the three is gateless.
- San Antonio’s ten-year structure costs USD 847 per space with no upfront equipment — 1.80 times the highest capital price and 3.45 times the lowest — because the fee moved to the driver at 99 cents to USD 3.99 a transaction.
- Compliance sets the replacement cycle. PCI DSS v4.0.1 retired v4.0 on 31 December 2024 and the future-dated requirements took effect 31 March 2025; a 2016 lane fails an assessment long before it fails mechanically.
- Software is the fastest layer at 14.65 per cent (20.0 to 29.9 per cent of value) and transaction and enforcement second at 12.09 (21.0 to 26.2), against 3.64 for lanes and access hardware.
Competitive layer
The supply side is a set of hardware incumbents being repriced by software and payment entrants, and the report tracks both groups because the forecast depends on which one captures the transaction. SKIDATA and Scheidt & Bachmann hold the classical access and revenue estate across European and airport accounts; TIBA Parking Systems, Amano McGann and Designa supply lane, pay-station and software combinations; Arrive is building the consolidated platform position; Metropolis is the clearest case of a transaction-layer company owning the operator; and Parking Concepts represents the operating layer that actually signs the contract and carries the labour. The strategic question for all of them is the same one the four procurements describe: whether you are paid for the lane or for the stay.
Method and disclosures
This release accompanies the full report and inherits its method. The research is 70 per cent secondary and 30 per cent primary. The secondary strand is public procurement records and award notices, council and airport board papers, statutory and standards documents, and company disclosure. The primary strand is structured interviews with parking operators and airport landside teams, open-ended interviews with access-control, payment and enforcement suppliers, a survey programme covering replacement cycles, compliance-driven upgrades and revenue-control structures, and conference-floor discussion through 2025 and 2026. Primary work here is used to date the record and to test whether what operators say they do matches what their contracts and their accounts show.
No competitor research is cited anywhere in this report or this release. Every forecast number is Pheonix’s own.
Two figures in the model deserve scrutiny rather than confidence, and the report says so on the page. The 40.0 million controlled-space base is a Pheonix build-up — no register of controlled parking spaces exists in any jurisdiction — and the regional sections state, market by market, what is disclosure and what is inference. The USD 650 million transaction layer is fee revenue retained after scheme and acquirer costs, a line almost nobody discloses; a change in take rate moves it, and the forecast, more than anything else in the model. Both are flagged in the same terms in the document.
Related research
The report is the sixth and final title in Pheonix’s Parking, Curb & Airport Mobility coverage, and it joins five siblings that describe adjacent layers of the same journey:
- Global Vertiport & UAM Ground Infrastructure Market 2026–2033 — a count market where the ticket per site falls 5.80 per cent a year, the mirror image of this one, where the count is flat and the ticket rises.
- Global Airport Baggage Handling Systems Market 2026–2033 — the same compliance-forces-replacement logic, applied to a system whose failure is measured in mishandled bags rather than failed transactions.
- Global Curb & Kerbside Management Market 2026–2033 — the allocation layer immediately upstream, where a space is priced by schedule rather than by stay.
- Global Airport Parking & Pre-Booked Reservation Platforms Market 2026–2033 — the demand side of the same bay, where a perishable space becomes a bookable asset.
- Global Parking Occupancy Sensing & Space-Guidance Market 2026–2033 — the measurement layer underneath, which supplies the occupancy data a revenue-control system prices against.
About Pheonix Research
Pheonix Research produces market intelligence on industrial, infrastructure and technology markets, built from primary enquiry layered over disclosure and public record, with every forecast owned by the firm and every source named. The full PARCS report — twelve sections, six analytical chapters, full table of contents, segment and regional splits, and the complete sourced evidence base — is available at pheonixresearch.com.
