Analysis

Chelsea have to keep selling players. Manchester City barely need to

Manchester City have been found guilty of the Premier League's financial charges and have appealed. Chelsea reported their former owner's undisclosed payments themselves, admitted them and paid. Both clubs were built on an owner's fortune, and both now play under a Premier League spending rule that gives the money those owners put in no value at all. What counts is what a club earns and what it makes selling players. Both clubs sold heavily this summer. On our central figure Manchester City are on the limit before any of those sales are counted, 1.4 percentage points over it, a margin too small to tell reliably which side they are on. Chelsea need all of theirs, and more.

By Arthur van de Laak27 September 2026Updated 31 October 2026 · View updates38 min read

Figures as of 31 October 2026, Orange Engine release V2CI · Latest figures · What the engine does not capture

The title rests on a central model estimate, not a certainty. Before any sale is counted, City’s squad costs 86.4% of the revenue the rule counts: 1.4 percentage points above the 85% line, which is smaller than the uncertainty in our estimate of last season’s commercial revenue. On that central figure City need about £12m a year from selling players to stay inside the line; on revenue about £12m higher they would need nothing. Chelsea need £164.2m a year.

This piece started with an argument on X. A Chelsea supporter told me that the £1.39bn of debt sitting in the companies above his club had nothing to do with Chelsea, because they keep separate books, and that with Clearlake now holding more of the club there would be more money to spend on players. He is not alone in thinking that, and he is right that the debt is not in the club’s own accounts. The rest of it is where this article goes.

On 25 September The Athletic reported that the independent commission hearing the Premier League’s case against Manchester City had found 114 of the 115 charges proven. The case has run since February 2023, its charges reach back to 2009, and City, who deny any wrongdoing, are expected to appeal once the decision is published.

Six months earlier Chelsea settled with the same league over £47.5m of payments made during Roman Abramovich’s ownership and never disclosed, and in July they accepted a £10m penalty from the Football Association on 74 charges arising from the same era. Those cases reached the regulators by a different route: Chelsea’s new owners found the problems in due diligence in 2022 and reported them themselves.

Both cases go back to the years when an owner’s money was turning a club into a champion, Chelsea from 2003 and City from 2008. Both concern the two numbers the Premier League now uses to limit squad spending: what a squad costs and what a club earns. And under the squad cost ratio, the money that built both clubs counts for nothing at all.

What the rule measures, and what it ignores

The rule is simple in principle. Add up what the squad costs each year: the players’ wages, the agents’ fees, and a slice of every transfer fee, because a £60m signing on a five-year contract is charged at £12m a year rather than all at once. Then compare that with what the club earns from football, plus its profit from selling players in the last six transfer windows, about three years, divided by three. If the squad costs more than 85% of that, the club is over the line and the League examines its actual figures. At 115%, sanctions begin.

An owner who wants to put money into a club has three routes, and each lands in a different place.

• A loan. The club gets cash and owes it back. The money can pay transfer fees and wages, but it is not income, so it does nothing to raise the limit those costs are measured against.
• Shares. The owner buys new shares in the club. Nothing is owed back, but again it is not income, and the limit does not move.
• Income. The money arrives as revenue, typically sponsorship or other commercial deals. This is the only route that counts, and it counts in full, although where the sponsor is connected to the owner the League first checks that the deal is worth what is being paid.

The rules in force while Chelsea and City were being built, UEFA’s break-even requirement and then the Premier League’s profitability and sustainability rules, measured losses, and an owner could cover part of a club’s permitted losses with secure funding. What mattered was the size of the loss. Under the new rule, a £1bn loan or share issue leaves what a club may spend on its squad exactly where it was. Only income and profit from selling players move it.

That is why the two clubs now sit so far apart.

City’s squad costs £103.2m a year more than Chelsea’s, and their ratio is 17.9 percentage points lower. The whole difference comes from the revenue side: City count £273.7m more. The third row is the part of that revenue that depends on where a club finished and whether it plays in Europe: Premier League prize money that rises with league position, UEFA money for European competition, and the extra home gates European matches bring.

Those figures are my own, and they took some getting. I rebuilt the ratio for every Premier League club from the accounts filed at Companies House, player by player, and checked each squad against the Premier League’s own registration lists. Chelsea were the hardest. An early version was missing nine senior players, Moisés Caicedo and Wesley Fofana among them. An audit of contracts found three new deals the reconstruction had not carried, and a rebuild of every departure since June 2025 added thirteen missing sales and corrected two fees. Then, on 26 September, I found ten registrations being written down twice, once inside the book value in the filed accounts and once individually, and correcting that moved Chelsea’s ratio from 101.5% to 87.6%. Every correction is logged on the club’s page, and The Orange Engine sets out how the checks work.

Chelsea: loans that paid for a squad

Abramovich funded Chelsea almost entirely by lending. Chelsea’s accounts for 2016/17 put his investment since 2003 at £1.17bn, all of it in loans, and the figure was reported at more than £1.5bn by the time of the sale in 2022. The loans did not travel with the club. They sat with Fordstam, the holding company he owned Chelsea through, whose accounts still show about £1.4bn owed to Camberley International Investments, a Jersey vehicle of his, while the £2.5bn of sale proceeds remain frozen.

What the money bought was twenty years of transfer fees and wages, which are costs. The trophies helped build the commercial and sporting profile Chelsea still trade on, but the funding itself never became revenue of the kind the current rule counts. In the year to 30 June 2025 Chelsea’s commercial income was £200.9m.

I watched this happen from the north side of London. I became an Arsenal supporter first through the Dutch connection, Bergkamp and Overmars, and later through the way the team played. Abramovich arrived in the summer of 2003, the season Arsenal went unbeaten, and Chelsea won the next two titles. In May 2005 Arsène Wenger called them a financially doped club. Arsenal were by then building a new stadium on borrowed money, and the sponsorship deals signed to secure that financing tied up the club’s commercial income for years afterwards. The stadium needed them, and they held Arsenal back. Wenger kept the club in the Champions League while selling Fàbregas, Van Persie and Nasri to balance the books, and the move of Ca$$$hley Cole to Chelsea in 2006 still hurts. One club was spending owner’s money on a squad, and the other was building revenue the slow way. Today Arsenal’s squad costs £479.8m against £738.2m of the revenue the rule counts. That is what the slow way built.

Some of the owner’s money also went where the accounts did not show it. The Premier League established that between 2011 and 2018, undisclosed payments by third parties associated with the club, totalling £47.5m, were made to players, unregistered agents and others, and that they should have been treated as made by the club itself. The published sanction agreement ties more than £23m of them to the signings of five players including Eden Hazard, and more than £19m to the moves of Willian and Samuel Eto’o from Anzhi Makhachkala in 2013. The FA’s 74 charges, under its agents, intermediaries and third-party investment regulations, came mainly from the seasons 2010/11 to 2015/16. UEFA had already fined the club €10m in 2023 for incomplete financial reporting.

The new owners handled it by disclosure. The League cited Chelsea’s “proactive self-reporting, admissions of breach and exceptional cooperation” as mitigation, fined the club £10.75m across two processes, with a nine-month ban on signing academy players and a suspended one-year first-team transfer ban, and found that even with the payments included in its historical submissions Chelsea would not have breached the profitability and sustainability rules. The FA penalty was financial, with a suspended two-window registration ban. Our Chelsea reconstruction sets out all four proceedings.

Manchester City: equity, and a commercial base

Sheikh Mansour’s route was different. City’s accounts for 2017/18 stated that he had invested more than £1.3bn in the club directly, in return for new shares rather than as loans. Over the same years City built commercial income, first on sponsorship from Etihad and other Abu Dhabi and UAE partners, and later more widely.

In the year to 30 June 2025 City’s commercial income was £340.4m, and the club states it would be £61.8m higher if its retail operations were run in-house. That base is what the ratio counts today.

It is also where the Premier League’s case began. The League charged City in February 2023 with breaches over the seasons from 2009/10 to 2017/18 concerning the provision of accurate financial information, “in particular with respect to its revenue (including sponsorship revenue), its related parties and its operating costs”. The charges also alleged that managerial remuneration was not fully disclosed in the four seasons from 2009/10, alleged breaches of UEFA’s financial rules and the League’s profitability rules, and included charges of failing to cooperate with the investigation, counted as 35 when the case was brought and grouped by the commission into four, three of which it upheld.

As of 27 September the League has not published the commission’s decision, and no sanction has been set. Under its rules the award is published, and City then have 14 days to appeal. They have denied the charges throughout, and their chairman said on 26 September that the process still has a long way to run.

Where the £273.7m comes from

The revenue gap between the two clubs has two parts, with different causes.

• £117.0m comes from league position and Europe. City finished second and are in the Champions League, which is worth £147.1m to them in this calculation. Chelsea finished tenth and are out of Europe, which is worth £30.1m, all of it from league position.
• £156.6m is revenue City earn and Chelsea do not. Most of it is commercial: £340.4m against £200.9m in the last filed year, and the rule counts merchandise at gross profit, which takes £19.2m off Chelsea’s commercial revenue for the test and nothing off City’s, whose shops are run by a partner.

It is not the gate. City’s filed matchday income for 2024/25 was £75.1m, and Chelsea’s gate for this test is £82.1m: last season’s estimated receipts per home match, over the 24 home matches the rule counts this season. Stamford Bridge holds about 40,000 and Chelsea’s stadium decision is still pending, while City’s expanded North Stand takes the Etihad from 53,400 to more than 60,000 from this season, so that difference is likely to open up. It has not yet.

The £156.6m of earned revenue is more than the £115.0m a year of averaged player-sale profit and manager compensation that Chelsea rely on. Add that £156.6m to Chelsea’s revenue and, before any player trading, Chelsea’s ratio would be 86.0%, against City’s 86.4%, 0.3 percentage points apart. Chelsea’s trading profit does most, but not all, of the work that City’s larger commercial base does.

What the rule does about either kind of case

The rule compares two things, what the squad costs and what the club earns, and the two cases sit on opposite sides of that comparison. Payments to players and agents are part of what a squad costs. Kept outside the accounts, they make the squad look cheaper than it was, which is what Chelsea admitted for the Abramovich years. Earnings are the other side, and the accuracy of City’s reporting of revenue, including sponsorship, is what the League’s principal charges concerned.

A ratio is only as reliable as the two numbers in it, and the new rule is built with that in mind. Income from parties associated with a club’s owner is tested against fair market value before it counts. And a club over the 85% line has its actual figures for the season checked, not just the estimate it submitted in March.

My own view is that the rule is right to ignore owner money, even though it is hard on clubs that never had a benefactor to build their revenue for them. Owner money that arrives as a loan or as shares can stop whenever the owner decides, and a club that has built its squad on it has nothing to fall back on when it does. Revenue a club earns for itself does not leave with a change of owner. The harder question is the one the City case raises: a rule that counts revenue is only as fair as the checks on where that revenue comes from, which is why the published decision matters well beyond one club.

The Orange Ledger’s City figures rest on revenue in the filed accounts for the year to 30 June 2025, outside the period of the charges, and this piece takes no view on that revenue beyond what the accounts report. The point is structural. A revenue-based rule counts the revenue a club reports now, however the base behind it was built. The money Abramovich put into Chelsea plays no part in today’s calculation, while City’s commercial income does. What follows from the case for sanctions, for claims by other clubs or for the rules themselves depends on the published award and on any appeal.

Chelsea now: an owner whose money does not count either

Clearlake has agreed to buy out Todd Boehly and Mark Walter and take full control of Chelsea. That changes who owns the club without changing the arithmetic. The reported £950m is being paid to Boehly and Walter for their shares, and none of it reaches the club, as our analysis of the transaction sets out. Any equity Clearlake puts into Chelsea from here counts for nothing in the ratio, as Abramovich’s loans would have counted for nothing had the rule existed while he was lending.

What has changed is the nature of the owner. Clearlake’s investment was led by one of its funds on behalf of affiliated funds and other partners, which means institutional capital that expects a return. Separately, the companies between Clearlake and the club, Blueco 22 Ltd and 22 Holdco Ltd, carry £1.39bn of external debt, and the £794.2m senior facility at Blueco 22 matures in July 2027. Abramovich covered Chelsea’s losses for nineteen years. A fund manager’s investors expect their money back with a return.

The trading that holds Chelsea near the line

That leaves Chelsea’s permitted spending resting on two things: revenue, and profit on selling players.

The rule adds up the profit of the six most recent transfer windows and divides it by three (Rules A.1.13 and A.1.53), so a strong window counts in three successive tests and a weak one does not sink a single season. Chelsea do not have to produce £115m of profit every year, but they have to produce it over time, because every profitable year eventually rolls out of the average and has to be replaced. Holding squad cost and revenue where they are, with no further sales, the ratio would move from 92.3% to 98.6% at the 2028 test, 99.8% in 2029 and 115.4% in 2030. The 115% line applies to the ratio after sale profit, 92.3% now; by 2030, with no sale profit left in the average, the ratio would sit 0.1 points above it. Revenue is held at this season’s figures in that arithmetic; next season’s commercial revenue is not forecast.

Every £10m less of average annual sale profit adds about 1.5 percentage points to Chelsea’s ratio, because it comes straight out of the denominator: at £129.2m, the same squad cost gives 93.9%.

Profit here is the fee less the player’s remaining book value, which is why academy graduates, carried at almost nothing, contribute nearly their whole fee, while a player bought recently for a large sum contributes much less. The largest single contribution to the current pool is Enzo Fernández’s £125m move to Manchester City, which produced about £49m of profit: the fee less a carrying value of about £74m, less 5% for the costs of the sale, roughly an eighth of the pool the average is taken over. Part of the profit supporting Chelsea’s ratio this season was paid by the club on the other side of this comparison.

Accounting profit is not cash, either. Transfer fees are usually paid in instalments, and Chelsea’s own accounts show £391.6m owed to other clubs for players at 30 June 2025, £197.8m of it due within a year. A sale does its full work in the ratio in the season it is made while the money can arrive over three or four years.

Both clubs sold heavily. Only one needs to keep doing it

Manchester City sold heavily this summer too. Sixteen registrations left in 2026, and City’s profit on player sales this season stands at £145.8m, against Chelsea’s £170.2m. The difference is what the selling is for.

Before any sales, City’s squad costs 86.4% of the revenue the rule counts, 1.4 percentage points above the 85% line, a margin too small for the estimate to tell which side City are on before a sale is counted. On that central figure the sale profit City need to reach 85% is about £12m a year, and their £118.4m average covers it many times over. Before any sales, Chelsea’s squad costs 115.4%, and their £115.0m average is what brings them down to 92.3%, which is still above the line. City could sell far less and stay inside the limit. Chelsea would need to sell more, or earn more, to get inside it at all: on the present cost and revenue, about £164m a year of sale profit would bring them to 85%, against the £133.5m they have.

City’s comfort rests on one condition. £147.1m of the revenue counted for them depends on finishing near the top of the league and playing in Europe. Without it, the same squad would cost 107.1% of their revenue before any sales, about eight percentage points below Chelsea’s 115.4%. So City can sell far less than they do for as long as they keep finishing high and qualifying for Europe. A league-points deduction or a European ban, if either were ever imposed, would reach exactly that part of their revenue.

Why Chelsea fans expect more, and why the numbers disagree

For twenty years a Chelsea supporter could assume that any player was within reach. Abramovich would fund him, and the rules of the day allowed it as long as the owner covered the losses. Manchester City then arrived as bigger spenders still, and they put money into the squad and into revenue at the same time, although how some of that revenue was accounted for is exactly what the Premier League case disputes.

Now the ownership has narrowed. Clearlake is buying out Boehly and Walter, leaving Hansjörg Wyss as the one other shareholder, and to many fans that looks like the Abramovich model returning: one decisive owner, deep pockets, more new players.

I think that expectation runs into four things at once.

• The rule. Chelsea’s squad already costs 115.4% of the revenue the rule counts before any sales, and owner money cannot change that.
• The stadium. Stamford Bridge holds about 40,000, and there is still no decision on a new ground.
• The revenue. Commercial income of £200.9m against Manchester City’s £340.4m, and no European football this season.
• The cost of money. The £794.2m senior loan above the club is charged at 3.25 percentage points over SONIA, the sterling benchmark rate that follows the Bank of England’s, and falls due in July 2027. The £595.9m of notes above that is charged at 7.5 percentage points over the same rate, and the interest is not paid in cash but added to the debt. Group interest came to £136.4m in 2024/25, about a quarter of the group’s turnover.

The debt is outside the club’s own accounts, but it is not outside its reality. It has to be serviced or refinanced from a business whose squad stays near the line only because it makes about £139m a year of profit selling players. Keeping that going while also buying the next wave of expensive signings looks extremely difficult to me, because every new signing adds to the squad’s cost, while the sales needed to make room for it must also produce an accounting profit.

What this doesn’t say

• That Manchester City’s current revenue is misstated. The figures used here come from City’s accounts for the year to 30 June 2025, outside the period of the charges, and nothing in this piece questions them.
• Anything about the outcome of the case. The commission’s decision was published on 29 September; City appealed on 2 October (Premier League statement of that day); an Appeal Board hears the appeal in private, and no sanction has been set.
• That the two cases are equivalent. Chelsea’s concerned undisclosed payments that the club admitted; City’s charges were broader, concerned revenue as well as costs, and are denied.
• That Chelsea cannot spend. They can, within what their revenue and their player trading support, and Clearlake can fund whatever the club spends.
• That the squad cost ratio measures financial health. Chelsea’s £1.39bn of external debt above the club, and the interest on it, sit entirely outside it.

What it does say

• Under the squad cost ratio, owner loans and owner equity enlarge nothing. Only revenue and player-sale profit raise what a club may spend on its squad.
• City’s squad costs £103.2m more than Chelsea’s and their ratio is 17.9 percentage points lower, because City count £273.7m more revenue: £117.0m from league position and Europe, and £156.6m earned, most of it commercial.
• That £156.6m is more than the £115.0m a year of averaged player-sale profit and manager compensation that brings Chelsea from 115.4% to 92.3%.
• Each £10m less of that average adds about 1.5 percentage points, and on unchanged costs and revenue the current pool has rolled out entirely by the 2030 test.
• On the central figure Manchester City need about £12m a year of sale profit to reach 85%, against a £118.4m average: before any sale they are 1.4 percentage points above the line, too small a margin to tell reliably which side they are on, and that holds only as long as the £147.1m that comes from league position and Europe does.

Chelsea helped create the era of the owner who covered the losses. Twenty years on, the clubs best placed under the rule are the ones that built recurring revenue the slow way. Liverpool own their stadium, make a profit and sit at 73.6% on my figures. Arsenal borrowed to build theirs and paid for it themselves. The question for Chelsea, and for the fan who started this argument, is whether financial prudence finally wins. If it does, Arsène Wenger may yet get his final revenge.