Accounting firm roll-ups in France and the UK

Because the economics suit a build-up and the supply is fragmented: compliance work recurs every year, client retention is high, and most practices are too small to fund software or offshore delivery on their own. The ageing-owner argument usually offered alongside those is testable, and it does not survive the test -- see below.

The mechanism is familiar. A platform practice is acquired, back-office and technology are centralised, and bolt-ons are added around it, usually smaller firms whose partners are approaching retirement with no internal successor. What is unusual here is not the thesis but the constraint on it: across much of Europe, who may hold the equity of an accounting or audit firm is set by statute, so the deal structure is settled before the price is.

That constraint is also why screening this sector is harder than it looks. A buyer needs to know which firms are large enough to be a platform, which are small enough to be a bolt-on, and which are regulated practices rather than bookkeepers carrying the same activity code. Only the first of those has a clean answer in the public registers, and even that answer differs by country.

Far fewer than the register suggests. France carries 32 383 active companies in the accounting, bookkeeping, audit and tax-consultancy class and the United Kingdom 57 805, but firms reporting twenty employees or more number 1 528 and 880 respectively.

Three cautions before anyone reads a market size into the first row. The activity class is declared at registration, so it holds regulated practices, unregulated bookkeepers and one-person tax advisers in one bucket; it is a population, not a licence register. And the two countries treat cessation differently, France keeping ceased entities on the register where the United Kingdom removes dissolved companies, so only the active counts are comparable at all. And availability here means available for screening, not what a filing physically contains: French accounts include a balance sheet, but the aggregate each country's source exposes is not the same one.

The row that matters to a buyer is the last one, and it is a floor rather than a count. A headcount is on file for only 13 156 of the French firms and 39 243 of the British ones, so a company that files no staff figure cannot be tested and is not in that row. Read it as the firms we can PROVE are of platform size: on that test the addressable market runs to hundreds and low thousands, not tens of thousands. The true number is higher by however many silent firms are large, which is a question for the next pass rather than an adjustment to make here.

Because British small companies never had to file the number. Section 444 of the Companies Act 2006³ lets a small company deliver a balance sheet without a profit and loss account, and almost every accounting practice qualifies as small, so turnover is available for fewer than one in fifteen active British firms in this class.

France is not more generous so much as differently shaped. Turnover is available for not quite one in four active French firms, while a balance-sheet total is not one of the fields carried for France. The consequence is that the same screen cannot be run on both countries.

Rank a European accounting universe by turnover and the United Kingdom all but disappears, not because British practices are small but because they do not publish the number being ranked. This is the most common way a cross-border sector screen produces a confidently wrong shortlist: a missing value is read as a small value, and an entire market drops out of the top decile.

The workable method is to rank each country on what it actually publishes and to keep the metric visible beside the result:

• In the United Kingdom, headcount and balance-sheet size, both available for the majority of active firms.
• In France, turnover, available for a minority but a much larger one, supported by equity.
• In neither, an estimate ranked alongside a reported figure as though the two were the same kind of observation.

In France: small, established, and growing slowly. The median French practice reporting accounts turns over EUR 547 000 with 4 staff at a 9.4% operating margin, growing 4.0% a year. The United Kingdom cannot honestly be described the same way, for the reason above.

The three withheld cells are a refusal rather than a gap in the data. A British median can be computed, and we have computed it; it would rest on the 3 585 firms that file turnover out of 57 805 active ones, and the margin on 729 of them. That is not a sample of the British market, it is the subset that chose to publish, so a median over it describes filing behaviour rather than an industry. Printing it would be the confidently wrong number this article is about.

Three cautions apply to the French column too, and each one moves a number a buyer would otherwise price on:

• Both halves of the margin come from the same fiscal year, the latest each firm discloses both. Taking each from its own latest filing instead lifts the median by roughly three points, because a fifth of these firms disclose the two figures in different years. And partner remuneration sits in payroll in an owner-managed practice, so this is what the business earns AFTER paying its owners to work in it: a floor under an acquirer's margin, not an estimate of it.
• The growth figure is a constant panel: 1 497 firms reporting turnover in both years. Firms that file consistently skew established, so this understates a market where new entrants are growing fastest.
• Incorporation date is not the owner's age. It shows a mature population; the owners' own ages are measured further down, and they say something different.

Old enough, and no older than anyone else. The median officer of a French accounting firm is 51; the median officer of every company on the French register is 51. The ageing-owner story this sector is sold on turns out to be the national base rate.

That does not remove the succession pipeline. In absolute terms it is substantial: 14 065 of the 35 292 French accounting officers we can date are fifty-five or over, and the United Kingdom looks much the same, with a median age of 52 and 34 268 officers at fifty-five or over. Plenty of firms will change hands. What the comparison removes is the reason to prefer this sector to any other on demographics.

This is the step most likely to change a decision and the cheapest to skip. Every fragmented professional-services market is marketed on retiring owners. A register can say whether that is true of the sector or merely true of the country, and the answer costs one extra query.

Consolidation has started and it is early. 3 025 French accounting firms already sit under a parent company, and the largest single owner holds 37 of them. Against an active population in the tens of thousands, the biggest visible platform is a rounding error.

That shape -- a long tail of small groups rather than three national champions -- is what a build-up wants to find, and the register shows it directly instead of by reputation. It also names the competition at auction.

On price we publish nothing, deliberately. No source we are willing to cite quotes an acquisition multiple for French or British accounting firms. The ranges in circulation come from advisers marketing their own deal flow, largely on American comparables, and a multiple carried across a border and an ocean is a guess wearing a decimal point.

What a register does give you is the number a multiple applies to. The median French practice disclosing both figures earned EUR 44 300 of operating profit in its latest year. Bring your own multiple to that, remembering the margin is struck after paying the owners to work in the business.

Experts-comptables, in the main. Article 7² of the 1945 ordinance governing the profession requires qualified professionals to hold more than two thirds of the voting rights in an accounting firm, directly or through another firm registered with the Ordre. Outside capital is possible below that line; control is not.

Statutory audit carries a separate European constraint. Under the Statutory Audit Directive¹, a majority of the voting rights in an audit firm must be held by approved auditors or approved audit firms. It applies across the Union, so a pan-European audit platform meets the same shape of restriction in every member state whatever the local company-law wrapper.

The United Kingdom draws the line elsewhere. Eligibility for appointment as a statutory auditor is restricted under Part 42 of the Companies Act 2006⁵, through the rules of a recognised supervisory body. That restriction attaches to the audit appointment rather than to the rest of a firm's work, and compliance, bookkeeping, payroll and tax advice, where most of a mid-sized practice's fee income sits, fall outside it.

The French constraint is visible in the register itself. France created a dedicated holding vehicle for accounting practices, the societe de participations d'expertise comptable, and 47 of them are active across all activity classes. That is the structure a French build-up is obliged to use, and how few there are is a fair measure of how early this market still is.

Six questions in this order, and a public register answers five of them. Accounting is the worked example; the sequence is the point, and it transfers to any fragmented sector.

• How many are ACTIVE, not how many have ever registered. France keeps ceased entities on the register, the United Kingdom removes them.
• What does each register actually disclose? Rank every country on what IT publishes, never on another country's absence.
• What is the size, the margin and the growth, with both halves of any ratio taken from one fiscal year?
• Are the owners older than the register's own baseline, or only as old as everyone else?
• How much consolidation has already happened, and how large is the biggest holder?
• What earnings would a multiple apply to? Bring the multiple yourself.

The question a register cannot answer is the legal one, and in this sector it is decisive: who may own the equity. In a trade with no licence it may cost nothing; here it settles the structure before the price.

One date is worth carrying forward. From 1 April 2028⁴ British small companies and micro-entities must deliver a profit and loss account to Companies House and abridged accounts are withdrawn, though a company may opt out of publishing it. Coverage improves; it does not become complete.

Gamma Insights holds company identity and the available financial fields across national registers in one schema, with the source country and the missing fields kept visible rather than filled in. That is less convenient than a database returning a number for every row, and considerably more useful when the question is which of two hundred firms to approach first.