Crystal Palace, rebuilt from the accounts: the club that files its player wages

Five years of the club's accounts, the group's, and a debenture registered in August 2025. Palace's reconstructed squad cost ratio ran at about 90% for three seasons with almost no player trading, and sits at about 74% this season because the trading pipeline turned on. The accounts also do something no other club rebuilt so far does: they state the players' wages, which changes how the earlier three should be read. And they show a £150m refinancing arranged eleven days after the ownership changed.

Crystal Palace is the fourth club in this series and the one that has changed the method. PSRwatch’s post-window review put the club’s squad cost ratio at 90%, five points above the League’s green threshold, on the same all-staff basis as the others. Palace’s accounts give a different number, and they also give something the other three clubs’ don’t: the players’ wages, stated as a figure. That single line is worth reading before the ratio.

What the accounts say

CPFC Limited files full accounts to 30 June; CPFC 2010 Limited files the group’s. The year to 30 June 2025:

• Turnover £195.3m: broadcasting £142.3m, sponsorship and advertising £18.8m, gate receipts £15.5m, other commercial £11.3m, other income £7.4m including £4.4m of FA Cup prize money.
• Player wage costs £110.8m, stated in the strategic report as a KPI, 56% of turnover. Wages and salaries for all staff £127.3m; social security £17.8m. Headcount: 131 players, managers and coaches; 247 administration and commercial.
• Amortisation of player registrations £54.0m; profit on player sales £66.1m; profit for the year £7.4m after a £33.6m loss.
• Registrations on the books: £317m cost, £160.4m net.
• Football transfer creditors £78.4m (£36.2m due after a year), up from £10.8m in 2021; transfer debtors £13.2m. Cash £13.7m. Net liabilities £40.3m.
• Interest payable £16.7m, of which £7.5m is “unwinding discount on deferred payments for player registrations”.
• Contingent transfer fees payable £35.8m, not provided.

The line the other clubs don’t file

For Fulham, Brentford and Bournemouth the accounts give one payroll figure for every employee, and the squad’s share has to be estimated; those pieces used 73–76% of wages and salaries, the range Kieran Maguire has given, and said so. Palace’s “player wage costs” makes the share observable for one club: on the face of the disclosures, £110.8m of £127.3m, 87%, and 84–93% in each of the four years before, leaving roughly £16.5m attributable to the rest of the payroll. The KPI’s accounting perimeter is not disclosed.

What it covers may include signing-on fees, bonuses and the coaching staff, which would make it close to the rule’s perimeter; it may not. And Palace may simply have a leaner structure than most. What the number establishes is narrower and more useful: the squad share is a club-specific variable, not a constant, with Maguire’s 73–76% as one benchmark and Palace’s filed 84–93% as another. If the earlier three clubs’ squads took the same proportion of payroll as Palace’s, their ratios would be roughly seven to ten points higher than published this morning; whether they do is unknown. Those three pieces now carry that caution, their bands should be read as wider on the upside, and this one uses the filed figure with the Maguire share as the low case.

Rebuilding the ratio

• Wages. £110.8m of player wages plus employer charges pro rata, £126.3m for FY25. Rolled forward for the summer of 2025 (Eze to Arsenal, Guéhi to City in January) and 2026 (about £70m of signings; Lacroix to Chelsea, Muñoz, Brennan Johnson out), with a European squad and a new head coach: about £128m for 2026/27, band £112m (Maguire share) to £136m.
• Amortisation. £54.0m in FY25 on a £160m book; about £62m for FY26; about £72m for 2026/27.
• Agents. The FA’s cash figure is £16.8m; about £11m enters the numerator as expensed, band £8–15m.
• Revenue. £195.3m rolled to about £214m: a fifteenth-place finish feeding this season’s distributions, a Europa League campaign after winning the Conference League, and a full Selhurst Park. Band £200–226m.

Profit on player sales, and this is where Palace differs from the other three. Because it does so much work in the result, it gets a ledger:

Fees from Transfermarkt and Sky, converted at 0.85; book values from purchase fee, date and contract; 5% direct costs; sell-ons (QPR’s on Eze is reported) unquantified and covered by the bands.

The result:

Band on the central figure: 66–84%. Confidence Medium: the filed wage line removes the largest uncertainty, but the FY26 and FY27 profits are estimated from headline fees and book values, and the European revenue is a guess. What survives the band: Palace is inside the green line on either profit treatment, its underlying cost base is about 99% of revenue, the same structural picture as the other three, and the difference is that Palace now sells every year. The trading effect is about 25 points, Brentford-sized, and it makes the point this whole series is about: for some clubs player trading is no longer peripheral income, it is part of the operating model, and it is what creates the regulatory room.

Six seasons on one method

Orange Ledger reconstructed ratios, the same formula every season, on the filed player-wage line.

Two things the series says:

• For three seasons Palace ran at about 90% and sold almost nothing: £0, £0.3m and £1.3m of profit. Player wages barely moved (£101m in each of FY22, FY23 and FY24) while revenue grew, so the ratio held rather than fell. It was a club living just above the new green line on cost discipline alone.
• Then the pipeline turned on: Olise in July 2024, Eze in August 2025, Guéhi in January 2026, Lacroix and Muñoz this summer. The cost base didn’t fall; the denominator grew. On the Maguire share instead of the filed figure every season in the table is seven to nine points lower, which is the size of the question the KPI raises for the other clubs.

The refinancing

At 30 June 2025 Palace was funded by a standing advance against its Premier League central distributions (£63.5m, up from £35.3m in 2021, growing every year), a £27.5m term loan in the group, and £54.6m of loans from its parent, Palace Holdco UK, of which £50m had been waived during the year and taken to reserves. On 23 June 2025 Woody Johnson agreed to buy John Textor’s stake in Palace Holdco; the purchase completed on 24 July.

On 4 August 2025, both existing facilities were repaid in full and the group entered “a new loan agreement for £150,000,000”. The debenture registered at Companies House three days later names the parties: CPFC Limited as borrower, CPFC 2010 and CPFC Selhurst Park as guarantors, Goldman Sachs International as arranger, Kroll Agency Services as agent and Kroll Trustee Services as security trustee. The security is the group’s: Selhurst Park itself, its revenue from tickets and hospitality, “all present and future book debts and all other amounts at any time recoverable or receivable by any Chargor”, the insurances and the shares. On the wording of the debenture that appears broad enough to capture League distributions and transfer receivables, although neither is separately identified by name. The earlier charges were released a fortnight later.

Two things the filings say about that. A standing advance against TV money, of the kind Brentford ran with Barclays, was replaced by a much larger arranged facility with a security trustee, in the same month the majority owner changed. And the accounts book £7.5m of imputed interest on deferred transfer payments, up from £4.8m, the same line Bournemouth (£13.3m) and Brentford (£6.0m) carry: the cost of buying on instalments, quantified.

The second test

Palace are in the Europa League, so UEFA’s 70% applies on the calendar year, with UEFA’s own numerator and a 36-month transfer result prorated to twelve. The bridge, so the figure can be reproduced:

The transaction-dated ledger behind the £213m: Olise and the other FY25 disposals (July 2024 onward, £66.1m filed, all inside the window); Eze (August 2025, ≈£54m), Guéhi (January 2026, ≈£17m) and the smaller FY26 sales (≈£3m); Lacroix (≈£40m), Muñoz (≈£19m), Brennan Johnson (≈nil) and the three academy sales (≈£14m), all August 2026. Every material disposal since January 2024 falls inside the 36 months, so the fiscal-year and calendar-window totals coincide.

About three points over the cap on the central figure, with the band straddling the line. A first breach of up to ten percentage points falls in UEFA’s lowest grid, with a financial measure of 10–25% of the calculated squad-cost excess, the money amount by which the numerator exceeds what 70% would allow. The transfer component is set; revenue and costs run to 31 December. Palace’s sales are what keep it near the line at all: without them the UEFA ratio would be near 100%.

What this doesn’t say

• That Palace’s ratio is 74%. The FY26 and FY27 profits are estimates until the accounts for those years are filed, and the League’s inputs aren’t public.
• That the club is in difficulty. It has a new facility to work with, an owner who has just paid for control, a trading record of three profitable summers, and a European campaign.

What it does say: Palace’s costs sit at about revenue before player trading, as at the other three clubs; its Premier League position has substantial estimated room because its trading pipeline works, while its UEFA position sits much closer to the separate 70% line, about eleven points inside one and three points over the other on the same summer’s business; and its credit position, a £150m facility secured over the ground, specified revenues and broadly defined receivables, depends on that pipeline continuing. And it says one thing about method that outlasts this article: where a club files its player wages, the ratio can be built on a number rather than a share, and the one club that does file them puts the share higher than the estimate the rest of this series has used.

crystal-palacesquad-cost-ratioaccountscompanies-housetransfer-credituefa