Private equity roll-ups of accounting firms to 2028
What the independence rules were written for, what the market has become, and what an owner should negotiate.
The rules that govern an accounting firm taking outside money were adopted in 2000 and assumed a public company would control the non-attest side. The standard-setter now says in its own exposure draft that a private equity investor "may or may not control the nonattest entity" and may hold significant influence over other portfolio companies across different funds with different general partners. That is not a drafting problem. It is a different market wearing the old structure.
This is a forecast rather than a market summary: seven structural shifts with the evidence under each, the loop that drives them, five scenarios with odds at December 2027 and December 2028, what commoditises and what keeps its scarcity, and seventeen dated calls a named public source will settle. Every reading carries a source grade. Every judgment carries a confidence. The odds are judgments too, and the page says so.
Rules written in 2000, a decade of deals, and a shrinking profession
When the current rules were adopted
The AICPA's alternative practice structure interpretations, written when a public company was the assumed investor.
Deals over ten years, per one tracker
Including 473 acquired firms and $49bn of enterprise value, to 31 May 2026. One tracker, no filing.
US accounting-services jobs since the peak
FD estimate: 1,121.2k − 1,160.5k = −39.3k, Sep 2026 against the Jan 2024 peak, on the BLS series.
Comment letters on the first draft
On the AICPA's 29 Dec 2025 exposure draft, before a revised draft was announced on 3 Nov 2026.
By the end of 2028 the rules have caught up with the structures, and what a selling firm negotiates has moved from price to control
The money arrived first and the rulebook is arriving second. Owners selling in between are agreeing terms nobody has tested.
We put the rules-bite shape at 31% by December 2028 against 24% a year earlier, and today's unconstrained consolidation down from 36% to 21%. The second panel is how a $2m to $20m revenue firm actually changes hands. Each column sums to 100, and both are Frontier Desk judgments rather than model output.
The financing market
Which shape compute financing is in at each date. Frontier Desk judgments, not model output
| Scenario | Dec 2027 | Dec 2028 |
|---|---|---|
| Consolidation continuesToday's shape: capital keeps arriving, structures keep being written, rules lag | 36% | 21% |
| Rules biteIndependence, ownership and disclosure requirements constrain what a structure can do | 24% | 31% |
| Second roundPlatforms sell to each other, and the question becomes who owns the owner | 18% | 22% |
| Returns disappointMultiples compress, deal pace slows, and some platforms stop buying | 14% | 18% |
| Audit separatesAttest and advisory become permanently different businesses with different owners | 8% | 8% |
How a new large GPU purchase is funded
The dominant structure for financings signed in each year. Today the secured facility still leads
| Structure | Dec 2027 | Dec 2028 |
|---|---|---|
| Sale to a sponsor-backed platformThe firm joins a consolidator that has already taken outside capital | 44% | 46% |
| Sale or merger into another CPA firmA traditional combination, with no outside investor in the structure | 29% | 23% |
| Internal succession to partnersThe next generation buys the firm, usually with seller or bank financing | 19% | 18% |
| Wind-down or sale of the client bookNo succession, no buyer for the firm, and the clients are sold on | 8% | 13% |
Odds are Frontier Desk judgments at the two horizons, set against the triggers and early markers that decide each one in Scenarios, and the method is in Sources and method.
Bottom line
Eight forward-looking judgments, each with the evidence under it and the confidence we hold it at.
- 01The independence rules were written for a structure that no longer describes the market, and the standard-setter says so in its own words.The AICPA Professional Ethics Executive Committee's exposure draft of 29 December 2025 sets the existing interpretation against what it calls an "APS with PE" in a side-by-side table. The existing interpretation "[p]resents an APS in which a public company controls the nonattest entity"; in an alternative practice structure with private equity, "[t]he investor may or may not control the nonattest entity". The existing rule "[a]ssumes the public investor not only controls the nonattest entity but also controls the other public company entities"; in the new structures "the other portfolio companies in which the PE investor has holdings may or may not be in the same fund as the nonattest entity, and the PE investor may have less than control over them", with "different general partners, fund managers, and investment advisers". PEEC's conclusion, in its own voice: "revisions to the code are necessary."Confidence: high. Every clause is quoted from the exposure draft text as published in a state board's own public meeting packet. What the final rule will say is not settled, which is what the scenarios are about.
- 02A firm selling today is agreeing terms the rulebook has not finished describing.The sequence is on the public record. PEEC appointed its task force in November 2022. The National Association of State Boards of Accountancy's private equity task force issued "Alternative Practice Structures & Private Equity: Considerations and Questions for Boards of Accountancy" on 24 October 2025. The AICPA exposure draft followed on 29 December 2025 and drew 81 comment letters. The California Board of Accountancy sent its comment letter to NASBA on 29 January 2026 and prepared a second for the AICPA draft. A revised exposure draft was announced on 3 November 2026, with a 120-day comment period and an effective date one year after adoption in 2027. An owner signing in 2026 is signing into a structure whose rules take effect in 2028.Confidence: high on the sequence and the dates, which come from the bodies' own documents and announcements. Moderate on the effective date, which comes from named trade press reporting the announcement rather than from the draft itself.
- 03The deal count is high, the capital is large, and both come from one tracker rather than from a filing.CPA Trendlines Research, publishing on 9 June 2026 with a cut-off of 31 May 2026, counts "525 events, including 473 acquired firms" over ten years, "$49 billion in transacted enterprise value", across roughly "72 platforms and 69 capital sources", covering about "$24 billion in annual revenue" and nearly "96,000 working accountants and staff". FD estimate: 525 events ÷ 10 years = about 53 a year; $49bn ÷ $24bn = about 2.0 times revenue; $49bn ÷ 96,000 = about $510,000 per head. We carry those figures with the tracker's name attached and grade them C, because no filing or regulator publishes a count of this market and nobody can check the tracker's method from outside. Both ratios carry a further caveat: the tracker describes the revenue and headcount as what the consolidation represents, which may be the platforms' totals rather than only what was acquired, so the numerator and denominator may not cover the same population.Confidence: moderate. The figures are specific, dated and from a named publication that has tracked this market for a decade. They are also uncheckable, which is why every judgment on this page that depends on them says so.
- 04Headcount in the industry has been falling since January 2024, while the capital has been arriving.Employment in US accounting, tax preparation, bookkeeping and payroll services peaked at 1,160.5 thousand in January 2024 and was 1,121.2 thousand in September 2026, on the Bureau of Labor Statistics' own series. FD estimate: 1,121.2 − 1,160.5 = −39.3 thousand, or −3.4% from the peak, against +10.6% since January 2019. Separately, US schools awarded accounting degrees down 6.6% year on year in the 2023–24 academic year, with master's down about 15% and bachelor's down 3.3%, and 13,070 candidates passed their fourth CPA exam section in 2024 against 19,382 in 2023 — a fall the AICPA itself ties substantially to a change in reporting methods, under which most fourth-quarter results were counted in the first quarter of 2025, and which it expects to moderate. The degree figures carry no such caveat and the exam comparison should not be read as a pipeline measure. A consolidation wave into a shrinking labour pool is not buying growth. It is buying the right to automate.Confidence: high on the employment series, which is a government statistical release we read directly. Moderate on the education figures, which come from a named report quoted by named press rather than from the report itself, which sits behind a download. The inference about automation is ours.
- 05What is actually being sold is the non-attest business, and the attest firm is what is left behind.In the largest transaction on the public record this year, the announcement of 11 June 2026 says that "funds managed by KKR will make a significant equity investment in Crowe Advisory LLC, becoming its first institutional capital partner", while "Crowe LLP will remain a licensed CPA firm and continue to provide all attest services, including audits and reviews", with the capital deployed through "its North America Fund XIV" and closing "expected to close in the third calendar quarter of 2026, subject to customary closing conditions and required regulatory approvals". That is the alternative practice structure in one paragraph: the investor buys the growing half, and the licensed half keeps the licence and the liability.Confidence: high. Every clause is quoted from the parties' own announcement. Whether the split is durable under the revised rules is the question the whole forecast turns on, and no source answers it yet.
- 06State boards, not the AICPA, decide whether any of this is enforceable — and they have said what worries them.The California Board of Accountancy's own meeting materials record that its members, reviewing the AICPA draft on 19 February 2026, were guided by two questions: "Can the CBA (or other state boards of accountancy) effectively regulate and enforce the proposed changes?" and "Do the proposed revisions continue to effectively provide for a strong standard of independence, not just in fact, but in appearance?" Its staff emphasised "the importance that CPAs are the final decision makers with respect to attest services and day-to-day management of the licensed CPA firms", supported "increased transparency in website content and consumer-facing disclosures", and sought responses that would "strengthen understanding of firm ownership". One board, in one state, has named enforceability, control and disclosure as the three live issues.Confidence: high on what that board said, which is quoted from its own published packet. Moderate on it being representative, because fifty-five jurisdictions license CPAs and we read one board's materials in full.
- 07The binding constraint is not capital or rules. It is whether a consolidated firm can keep the people.An accounting firm's assets go home every night and can leave with the clients. A roll-up changes compensation, partnership, autonomy and brand in the same year, in an industry whose employment has fallen 3.4% from its peak and whose degree pipeline fell 6.6% in a year. The capital is abundant, the structures are solvable, and the thing nobody has demonstrated at scale is that a firm which has been bought retains its people and its clients through the second and third year. 38% (judgment) that a named US accounting platform publicly discloses a material partner or client-retention problem by 31 December 2028.Confidence: moderate. The labour and pipeline readings are solid, and the inference that retention is the binding constraint follows from them rather than from any disclosure. No platform is required to publish retention data, which is why the call is written to be settled by a disclosure that may never come.
- 08For the owner of a firm selling into this, the price is the least negotiable term and the control terms are the most.The structure that is being bought splits one business into two entities, assigns the licence and the attest liability to one of them, and puts the investor in the other. Everything an owner will care about in five years lives in that split: who appoints the leaders of the attest firm, what the non-attest entity can require of it, how the earn-out is measured across both, what happens to the licence on a second sale, and whether independence rules the firm must follow could disqualify it from serving a client it has had for twenty years because a different fund owns a competitor. The standard-setter has now written down that last risk in its own exposure draft.Confidence: high on the mechanism, which follows from the quoted exposure draft text. The ranking of terms is Frontier Desk's judgment, informed by the regulatory record rather than established by it.
The baseline
Where consolidation of US accounting firms stands on 5 November 2026. This band is collection, not analysis: every row is a reading from a named public source, graded A, B or C, with no inference attached. The inferences are everywhere else on the page.
The readings below are the position the rest of this report forecasts from: a rulebook written in 2000 for a structure that no longer exists, three standard-setting and regulatory bodies working on it at once, a deal count nobody but a tracker publishes, a shrinking workforce, and a licensed attest firm left holding the licence in every structure.
When the current rules were adopted
The AICPA's alternative practice structure interpretations, written when a public company was the assumed investor.
Deals over ten years, per one tracker
Including 473 acquired firms and $49bn of enterprise value, to 31 May 2026. One tracker, no filing.
US accounting-services jobs since the peak
FD estimate: 1,121.2k − 1,160.5k = −39.3k, Sep 2026 against the Jan 2024 peak, on the BLS series.
Comment letters on the first draft
On the AICPA's 29 Dec 2025 exposure draft, before a revised draft was announced on 3 Nov 2026.
Every reading, with its source
Publicly observableGrades describe the source, not the conclusion. A — record: a standard-setter's own exposure draft, a state board's own published meeting materials, or a government statistical series. B — party: an organisation's own announcement about itself, including a firm's or an investor's press release. C — crowd: trackers, surveys and named press reporting figures nobody else publishes.
Confidence describes Frontier Desk's inference from those readings: high where two or more independent sources agree and none contradicts, moderate where the direction is corroborated but the magnitude is not, low where a single source or an analogy carries it. The odds on this page are judgments of the same kind, not the output of a model, and they are stated as judgments wherever they appear.
| Reading | Source | Figure | Grade |
|---|---|---|---|
| When the current alternative practice structure rules were adopted | AICPA PEEC exposure draft, 29 Dec 2025 | 2000, into the AICPA Code of Professional Conduct | A · record |
| What the existing interpretation assumes | The same exposure draft | An APS in which a public company controls the nonattest entity | A · record |
| What the standard-setter says is different now | The same exposure draft | The investor may or may not control the nonattest entity, and may hold less than control over other portfolio companies | A · record |
| Why other portfolio companies matter | The same exposure draft | They may sit in different funds with different general partners, fund managers and investment advisers | A · record |
| PEEC's own conclusion | The same exposure draft | Revisions to the code are necessary | A · record |
| What the exposure draft would replace | The same exposure draft | ET sec. 1.220.020 in its entirety, with revisions to ET secs. 0.400.36, 1.810.050, 1.210.010 and 1.000.010 | A · record |
| When the task force was appointed | The same exposure draft | November 2022, with members from APS and traditional firms, an attorney, NASBA and regulators | A · record |
| The state boards' own task force | California Board of Accountancy meeting materials, March 2026 | NASBA PE Task Force white paper issued 24 Oct 2025, with seven members | A · record |
| A state board's comment letter to NASBA | The same meeting materials | Sent 29 Jan 2026, after discussions beginning at the September 2025 board meeting | A · record |
| That board's two guiding questions on the AICPA draft | The same meeting materials | Whether boards can effectively regulate and enforce the changes, and whether independence holds in appearance as well as in fact | A · record |
| What that board's staff emphasised | The same meeting materials | That CPAs are the final decision makers on attest services and day-to-day management of the licensed firm | A · record |
| Comment letters received on the first draft | Journal of Accountancy, 3 Nov 2026 | 81 | C · crowd |
| The revised exposure draft | AICPA announcement, 3 Nov 2026 | Announced 3 Nov 2026, posting later in the month, with a 120-day comment period | A · record |
| What the revised draft adds | The same announcement | Clarified closely aligned entity and network firm definitions, broader covered members, and restricted attest services where the investor has significant influence | A · record |
| Expected effective date | Journal of Accountancy, 3 Nov 2026 | One year after adoption in 2027, with earlier implementation permitted | C · crowd |
| Deals over ten years | CPA Trendlines Research, 9 Jun 2026, cut-off 31 May 2026 | 525 events including 473 acquired firms | C · crowd |
| Capital transacted | The same tracker | $49 billion of enterprise value, across about 72 platforms and 69 capital sources | C · crowd |
| Revenue and headcount covered | The same tracker | About $24 billion of annual revenue and nearly 96,000 accountants and staff | C · crowd |
| Implied pace and pricing | FD estimate from the same tracker | About 53 events a year; $49bn ÷ $24bn = about 2.0 times revenue; $49bn ÷ 96,000 = about $510,000 per head | C · crowd |
| The largest 2026 transaction's structure | Crowe and KKR announcement, 11 Jun 2026 | KKR funds invest in Crowe Advisory LLC; Crowe LLP remains the licensed CPA firm providing all attest services | B · party |
| That transaction's fund and timing | The same announcement | KKR North America Fund XIV; expected to close in the third calendar quarter of 2026 subject to regulatory approvals | B · party |
| US accounting-services employment, latest | Bureau of Labor Statistics series CES6054120001, read 5 Nov 2026 | 1,121.2 thousand in September 2026 | A · record |
| The same series at its peak | The same series | 1,160.5 thousand in January 2024 | A · record |
| Change from the peak and from 2019 | FD estimate from the same series | −39.3 thousand (−3.4%) from the peak; +10.6% from January 2019 | A · record |
| Accounting degrees awarded | AICPA Trends report via CFO Dive, 27 Oct 2025 | Down 6.6% year on year in 2023–24; master's down about 15%, bachelor's down 3.3% | C · crowd |
| CPA exam completions | The same report and article | 13,070 candidates passed a fourth section in 2024, against 19,382 in 2023 — a fall the AICPA ties substantially to a reporting-cycle change it expects to moderate | C · crowd |
| Accounting enrolments, against the degree trend | The same report and article | Up about 12% year on year, suggesting the pipeline may turn | C · crowd |
One document carries more of this forecast than any other, and it is worth reading in the standard-setter's own words rather than in a summary. The AICPA Professional Ethics Executive Committee's exposure draft of 29 December 2025 contains a two-column table setting the rules adopted in 2000 against the structures now being built.
On the left: the existing interpretation "[p]resents an APS in which a public company controls the nonattest entity", and "[a]ssumes the public investor not only controls the nonattest entity but also controls the 'other public company entities'". On the right, under the heading "APS with PE": "[t]he investor may or may not control the nonattest entity", and "the other portfolio companies in which the PE investor has holdings may or may not be in the same fund as the nonattest entity, and the PE investor may have less than control over them. Additionally, the other funds and portfolio companies may be managed and advised by different general partners, fund managers, and investment advisers."
The practical consequence is the part an owner should read twice. Independence is not only about who owns your firm. It is about what else your owner owns, across funds you will never see, managed by people you will never meet. A firm that joins a platform inherits the independence problems of every investee its investor significantly influences, and the exposure draft exists because the 2000 rules have no clean way to work out which those are.
Hold those readings together and the next two years are already implied. A rulebook being rewritten will produce terms that bind structures already signed. A deal count nobody can check will stay uncheckable until somebody is required to file. A workforce falling 3.4% from its peak means the growth has to come from price or from software. And a licensed attest firm left holding the licence in every structure is where the liability will land when any of this goes wrong.
Structural shifts underway
Seven shifts are already running. Each separates what the sources say from what we infer from them: the evidence first with its grade, then the reading with its confidence, then how fast, then the specific thing that would reverse it. These are the inputs to the scenarios.
The rulebook is rewritten around the investor's portfolio rather than the firmAccelerating
Speed: Draft now; effective 2028
What the sources say. The AICPA exposure draft of 29 December 2025 would replace the "Alternative Practice Structures" interpretation at ET sec. 1.220.020 "in its entirety", with revisions to the network firm definition at ET sec. 0.400.36, to ET sec. 1.810.050, and to both conceptual framework interpretations. The announcement of 3 November 2026 describes a revised draft clarifying "closely aligned entity" and "network firms", broadening "covered members" to include the closely aligned entity's chief executive, and clarifying that an attest firm cannot provide attest services to organisations over which its closely aligned entity's investor has significant influence.
Evidence grade: A on the exposure draft text and on the standard-setter's own announcement.
What we read into it. The direction of travel is from the firm outward. The 2000 rules asked what the investor owns of your firm; the revised rules ask what else your investor significantly influences, and prohibit attest work there. For a firm with an industry specialism, that is the whole book: if your platform's investor has significant influence over a company in your niche, that client becomes unservable. The negotiation an owner has to run is not about the valuation; it is about which clients the structure will still permit in 2029.
Confidence: high on what is proposed, which is quoted. Moderate on the final shape, because the draft is in a 120-day comment period and the first version drew 81 letters.
What would reverse it. PEEC withdrawing the draft or adopting a materially narrower version confined to control relationships, which would leave the 2000 framework substantially intact.
State boards move from watching to asking enforceable questionsRunning
Speed: 2027 for the first state action
What the sources say. The California Board of Accountancy's March 2026 meeting materials record board discussions beginning in September 2025, a comment letter to the NASBA private equity task force sent 29 January 2026, two members appointed to review the AICPA draft, and two guiding questions from a 19 February 2026 meeting: whether boards "can effectively regulate and enforce the proposed changes", and whether the revisions "continue to effectively provide for a strong standard of independence, not just in fact, but in appearance". Its staff emphasised that "CPAs are the final decision makers with respect to attest services and day-to-day management of the licensed CPA firms". NASBA's own task force white paper was issued 24 October 2025.
Evidence grade: A · record — a state board's own published meeting packet, including the letters it sent.
What we read into it. The AICPA writes the code; the state boards license the firms and bring the discipline. A board asking in public whether it can enforce a proposed rule is the earliest reliable signal that the rule will be supplemented locally. Expect at least one state to propose a firm-ownership disclosure or registration requirement before the AICPA rule takes effect, and expect it to focus on the three things this board named: enforceability, who decides, and what the public is told.
Confidence: moderate. The board's position is quoted and its task force seat is documented. That other boards follow is an inference from one board's materials, and we read one of fifty-five jurisdictions in full.
What would reverse it. NASBA's task force concluding that existing state rules suffice and recommending no changes, with boards accepting it, which would leave enforcement entirely with the code.
The workforce shrinks while the capital arrivesRunning
Speed: Already here
What the sources say. Employment in US accounting, tax preparation, bookkeeping and payroll services was 1,121.2 thousand in September 2026 against a peak of 1,160.5 thousand in January 2024 and 1,013.8 thousand in January 2019, on the Bureau of Labor Statistics' own monthly series. Accounting degrees awarded fell 6.6% year on year in 2023–24, with master's down about 15%, and 13,070 candidates passed a fourth CPA exam section in 2024 against 19,382 in 2023, with enrolments up about 12%. The same article records the AICPA attributing the “substantial decline” in passes substantially to a recent change in reporting methods, with most fourth-quarter results counted in the first quarter of 2025, and saying the impact “will be moderated in 2025”.
Evidence grade: A on the employment series, which we read from the statistical agency's own API; C · crowd on the education figures, which come from a named report quoted by named press.
What we read into it. A consolidation wave into an industry whose headcount is falling cannot be a growth story on volume. It has to be a story about price, about share taken from firms that cannot recruit, or about doing the same work with fewer people. All three are plausible and only the third is new, which is why every platform's thesis now mentions technology. For an owner the implication is uncomfortable: your buyer's model may require your firm to need fewer of your people than you employ today.
Confidence: high on the employment reading. Moderate on the interpretation, because headcount can fall for reasons — offshoring, software, retirement — that a single series cannot separate.
What would reverse it. The employment series recovering past its January 2024 peak for two consecutive quarters, which would mean the industry is growing again and the consolidation thesis can rest on volume.
The attest firm becomes the smaller, licensed, liability-bearing halfRunning
Speed: Standard already
What the sources say. The largest 2026 transaction's own announcement, dated 11 June 2026, states that "funds managed by KKR will make a significant equity investment in Crowe Advisory LLC" and that "Crowe LLP will remain a licensed CPA firm and continue to provide all attest services, including audits and reviews". The AICPA exposure draft describes the same split as the defining feature of an alternative practice structure, and its proposed revisions concentrate on the relationship between the attest firm and the "closely aligned entity".
Evidence grade: B · party on the transaction, from the parties' own announcement; A · record on the structure, from the exposure draft.
What we read into it. Every alternative practice structure makes the same trade. The investor buys the part that can be owned and scaled, and the licensed firm keeps the part that cannot be owned by non-CPAs — along with the professional liability, the peer review, the state licence and the discipline exposure. That asymmetry is stable while everyone is making money. It is the first thing that gets tested when a client sues, and no public source tells us how it has been tested so far.
Confidence: high on the structure, which both the transaction and the standard-setter describe in the same terms. The judgment about where liability concentrates is ours, and we have found no litigated example to cite either way.
What would reverse it. A jurisdiction permitting non-CPA majority ownership of attest firms outright, which would collapse the two-entity structure and with it most of the independence analysis in this forecast.
The market has no public record, and the absence is itself the riskPersistent
Speed: Until a regulator requires filing
What the sources say. The only decade-long count of this market is a tracker: CPA Trendlines Research, publishing 9 June 2026 with a 31 May 2026 cut-off, reporting "525 events, including 473 acquired firms", "$49 billion in transacted enterprise value", about "72 platforms and 69 capital sources", "$24 billion in annual revenue" and nearly "96,000 working accountants and staff". No securities regulator, no standard-setter and no state board publishes a comparable count. The California Board of Accountancy's staff sought responses that would "strengthen understanding of firm ownership".
Evidence grade: C · crowd on the tracker, carried with its attribution; A · record on the board's stated interest in ownership transparency.
What we read into it. A market this size with no filing requirement is one where every participant's view is formed from the same handful of sources. That matters practically: an owner negotiating a price has no public comparable, an investor has no independent denominator, and a regulator cannot see concentration until it is asked to approve something. The pressure to fix this is already visible in a state board's own words, and a disclosure requirement is the single most likely regulatory output of the next two years.
Confidence: moderate. The absence of a public record is easy to establish; the forecast that it gets filled is a judgment, and the mechanism would be a state registration requirement rather than a federal one.
What would reverse it. A state or NASBA adopting a firm ownership register, which would turn this from a structural feature into a transitional one and give the market its first checkable denominator.
The second round begins, and the question becomes who owns the ownerStarting
Speed: 2027 to 2028
What the sources say. One tracker counts about 72 platforms and 69 capital sources over ten years, with 525 events to 31 May 2026. The largest transaction of 2026 is a fund investment in an advisory entity, deployed through a named vintage fund with a defined life. The AICPA exposure draft's proposed revisions turn on whether an investor controls or significantly influences the non-attest entity and other investees, and on the identity of the general partners and advisers behind each fund.
Evidence grade: C · crowd on the platform count; B · party on the transaction structure; A · record on the exposure draft.
What we read into it. Private equity funds have finite lives, and a platform assembled in 2021 and 2022 reaches its exit window inside this forecast. The buyer of a platform is most likely another sponsor, which means a selling firm's independence analysis has to be redone with a new investor's whole portfolio. An owner who sold in 2024 and read the independence terms carefully will discover that the terms were about an investor who is no longer there. That is the single most under-negotiated risk in this market.
Confidence: moderate. Fund lives and exit behaviour are well understood generally, and the structural consequence follows from the quoted exposure draft. That the second round lands inside the window is a judgment, and no public source lists platform vintages.
What would reverse it. Platforms refinancing with continuation vehicles or permanent capital rather than selling, which would postpone the second round past 2028 without resolving it.
The thing being bought is a client base that can leavePersistent
Speed: Tested in 2027 and 2028
What the sources say. A tracker's figures imply about $510,000 of enterprise value per accountant and staff member and about 2.0 times annual revenue — FD estimate: $49bn ÷ 96,000 and $49bn ÷ $24bn. Industry employment is 3.4% below its January 2024 peak. The California Board of Accountancy's staff emphasised consumer-facing disclosure and public understanding, which in practice is about whether a client knows who owns the firm it hired.
Evidence grade: C · crowd on the per-head and revenue multiples, which derive from the tracker; A on the employment series and the board's statements.
What we read into it. Paying about two times revenue for a business whose producers can resign and whose clients can follow is an underwriting bet on retention, in an industry where the people are scarcer every year. The structures address this with earn-outs, rollover equity and non-competes, all of which bind the partners who sign them and none of which binds the partner who joins afterwards. The first platform to publish a retention figure will reset how this market is valued, and that is why no platform will publish one voluntarily.
Confidence: moderate. The multiples are derived from a tracker we cannot check. The retention argument is structural and widely understood, but no public source quantifies it in this industry, which is a gap the report states rather than fills.
What would reverse it. A platform disclosing strong multi-year partner and client retention, which would convert the central unknown of this market into a measured and apparently manageable risk.
The causal model
The loop that produced everything above, in six steps, with the figure that shows each one is running. Three arms reinforce it and three brake it; the scenarios are what happens when one arm or one brake wins.
- 1
Retire
Partners reach retirement with no internal successor
−39.3k jobs from peak - 2
Capitalise
A sponsor funds a platform to buy them
72 platforms, 69 sources - 3
Split
The firm divides into attest and non-attest entities
Licence stays with the CPA firm - 4
Buy
The platform acquires more firms at scale
473 firms, $49bn - 5
Automate
Margin comes from doing the work with fewer people
Degrees −6.6% a year - 6
Re-rule
Standard-setters and boards rewrite what the structure may do
ET 1.220.020 replaced
↻ Every firm bought removes a competitor and a succession option for the next seller, which raises the price the next platform will pay and shortens the list of buyers a firm can go to. The loop tightens on the seller as it runs.
The three arms, and the three brakes
Read the loop and the scenarios stop being moods. Consolidation continues is R1 and R2 running with nothing engaging. Rules bite is B1 and B3 together. Returns disappoint is B2 engaging. Audit separates is B1 engaging so hard that the two halves stop being one business. The odds below are our judgment about which gets there first.
How it evolves, 2027 and 2028
Five phases to the end of 2028, each named for what changes in it, with the Frontier Desk judgments that attach to that window. Every figure here is one of the calls in band 13, carried unchanged.
The revised draft posts, and the comment period runs
The AICPA announced its revised exposure draft on 3 November 2026 for posting later that month, with a 120-day comment period. Expect state boards to file again — one has already sent two letters in a year — expect the large platforms and their counsel to file in volume, and expect the argument to concentrate on the definition of "significant influence" and on which other investees it reaches. For an owner mid-process, this is the window in which the terms you are being offered are being argued about in public.
The revised draft is posted and the comment period opens: 91% · state boards file comment letters: 83%
The code is adopted, and the states start to diverge
On the central path PEEC adopts a version of the revised interpretation in 2027, effective a year later. The code binds CPAs because state boards adopt it; where a board has already questioned enforceability, expect a supplementary state rule rather than a straightforward adoption. Expect at least one jurisdiction to propose a firm-ownership disclosure or registration requirement, and expect the first visible divergence between what the code permits and what a state will license.
PEEC adopts a revised APS interpretation: 72% · a state proposes a firm-ownership disclosure or registration rule: 54%
The first platform exits, and the independence analysis restarts
Platforms assembled in 2021 and 2022 reach their exit windows. A sale to a second sponsor means every firm inside the platform has a new investor whose whole portfolio has to be assessed against the independence rules, and some clients become unservable. Expect this to surface as resignations from engagements rather than as announcements, which makes it hard to see and easy to under-weight.
A US accounting platform is sold to a second sponsor: 58% · a firm publicly resigns an engagement citing investor independence: 26%
The code takes effect, and the structures have to comply
A rule adopted in 2027 and effective a year later lands here, on structures signed in 2024, 2025 and 2026 whose documents were drafted against the 2000 interpretation. Expect restructuring work, expect some client transitions, and expect the gap between what owners were told at signing and what the structure can do to be the main source of disputes in this market.
A revised APS interpretation is in effect: 66% · a platform discloses restructuring to meet the new rules: 47%
Rules bite, or the money has already moved on
On the central path the rules have caught up, the structures have been adjusted, the deal pace has slowed from its ten-year average, and what a selling firm negotiates has moved decisively from price to control. The alternative is that returns disappointed first: retention proved harder than modelled, multiples compressed, and the platforms that bought at two times revenue in 2025 are the sellers rather than the buyers.
Rules bite: 31% · returns disappoint: 18% · a platform discloses a retention problem: 38%
Scenarios
Five ways this consolidation can go. Each carries Frontier Desk odds at 31 December 2027 and 31 December 2028, the trigger that puts you on it, the early markers that tell you first, and what it means on each side of a deal. Each column sums to 100, and every signpost in band 12 names the scenario it moves.
Consolidation continues
Today's shape continues. Capital keeps arriving, firms keep selling, the structures keep being documented from the same template, and the rules keep lagging. The reason this falls from 36% to 21% is not that the capital stops. It is that three bodies — a standard-setter, a state board association and the state boards themselves — are all working on this at once, and their output lands inside the window.
Trigger: None. This is the no-trigger path: the revised interpretation is adopted in a narrow form, state boards adopt it without supplement, and nothing constrains the structures.
Early markers: Deal pace holding near 53 events a year; no state proposing an ownership rule; no platform restructuring publicly.
For an owner: Your leverage is competition between buyers, and it is highest now. Run a real process rather than responding to the first approach, because the number of independent firms left to compare you against falls every year.
For buyers: The window stays open and the prices stay full. The risk is paying 2026 prices for a structure that 2028 rules will make less useful.
Rules bite
The modal destination by the end of 2028. A revised interpretation takes effect, state boards supplement it, and what a platform-owned firm may do narrows — particularly on attest work for companies its investor significantly influences. Structures signed against the 2000 rules get restructured. Deal pace slows and prices separate: a firm with a clean, portable client base is worth more, and a firm whose book overlaps an investor's portfolio is worth less.
Trigger: PEEC adopting a revised interpretation that reaches significant influence rather than control, and at least one state board supplementing it.
Early markers: The revised draft adopted; a state proposing an ownership disclosure rule; a platform disclosing restructuring; engagements resigned on independence grounds.
For an owner: The independence analysis is now a diligence item you should run on the buyer, not only one they run on you. Ask what their investor's portfolio contains in your sector, and get the answer in writing before you sign.
For buyers: Underwriting has to include the investor's whole portfolio and its likely future contents. The platform that can demonstrate a clean independence position becomes the one firms prefer to join.
Second round
Platforms assembled in 2021 and 2022 are sold, mostly to other sponsors, and every firm inside them acquires a new investor whose portfolio has to be assessed from scratch. The economics work — scale is real — but the independence position resets with each sale, and the terms an owner negotiated against one investor turn out not to have been about the investor at all.
Trigger: A first platform sale to a second sponsor, at a price that makes the rest of the cohort move.
Early markers: Platform sales announced; continuation vehicles raised; firms inside platforms changing which clients they serve after an ownership change.
For an owner: Negotiate for the second sale now. Change-of-control consent, an independence representation that survives a transfer, and a right to be released if a new investor makes your book unservable are worth more than the last turn of price.
For buyers: The exit is real and the independence reset is a cost of it. The platforms that sell well are the ones whose client bases are not entangled with their investor's portfolio.
Returns disappoint
Retention proves harder than modelled, the automation thesis delivers more slowly than underwritten, multiples compress, and the platforms that paid about two times revenue stop buying. This is the second-likeliest 2028 outcome and the one least represented in the terms firms are being offered today, because nobody in the market has an incentive to model it out loud.
Trigger: A named platform disclosing a material partner or client-retention problem, or a platform sale at a price below its cost.
Early markers: Deal pace falling below the ten-year average; platforms pausing acquisitions; partner departures reported at a named platform.
For an owner: An earn-out in this scenario pays less than the headline, and rollover equity pays less again. If you can be paid in cash now rather than in the platform's future, the case for doing so is stronger than it looks.
For buyers: The firms bought in 2025 and 2026 are marked against a slower market. Expect the gap between the platforms that integrated well and the ones that only acquired to become visible and permanent.
Audit separates
Independence requirements tighten enough that attest and advisory stop being one business: audit firms shrink to audit, advisory platforms drop attest work entirely, and the alternative practice structure stops being worth the complexity. The odds do not rise with the horizon because this requires a rule change more decisive than anything currently drafted; it is a tail, not a trend.
Trigger: A final rule, or a state licensing decision, that makes attest work inside an investor-backed structure impractical rather than merely constrained.
Early markers: Platforms divesting attest practices; firms deregistering from audit; a state refusing to license a structure the code permits.
For an owner: Which half of your firm you are selling becomes the first question rather than a structural detail. An attest-heavy firm and an advisory-heavy firm would face entirely different markets.
For buyers: Advisory platforms get simpler and audit gets harder to fund. Whoever owns the remaining independent audit capacity in the middle market has something scarce.
Winners and losers
Who gains and who loses on the central path, and why. This band is analysis throughout: direction is stated against today, not against each other, and nothing here is a reading.
Firm owners and partners
| Segment | Direction | Why |
|---|---|---|
| Owners of advisory-heavy firms with portable, diversified client bases | Clear winner | This is what the capital is buying and what the independence rules leave alone. A book with no concentration and little attest work is the cleanest asset in this market, and it will stay clean through a rule change and an investor change. |
| Owners of attest-heavy firms with a sector specialism | Loser on the central path | A rule reaching companies over which an investor has significant influence hits specialists hardest, because a specialist's clients cluster where an investor's portfolio clusters. The thing that makes the firm valuable is the thing that makes it conflicted. |
| Partners within five years of retirement at a firm with no internal successor | Winner, with a deadline | The buyers are present, funded and competing, and the number of independent firms to compare against falls every year. This is the best market such a partner will see, and the odds say it is better now than in 2028. |
| Younger partners taking rollover equity rather than cash | Loser in two of five scenarios | Rollover equity pays in the second sale, and the second sale is where the independence position resets and the returns-disappoint scenario lands. It is the right trade in a continuing consolidation and the wrong one in the other outcomes. |
| Owners who have already sold and read only the price | Clear loser | The terms that will matter — who appoints attest leadership, what the investor's portfolio does to the client list, what happens on a second sale — were negotiable once. The standard-setter has now written down the risk they did not price. |
| Firms choosing to stay independent with a credible succession plan | Winner, selectively | Scarcity works in their favour on both sides: clients that want an unconflicted auditor, and staff that want a partnership track. The condition is a real succession plan, which is exactly what most of the sellers did not have. |
Platforms, investors and the profession
| Who | Direction | Why |
|---|---|---|
| Platforms with real integration and a clean independence position | Clear winner | Scale in software, offshore delivery and specialism is genuine, and a platform whose investor's portfolio does not collide with its clients' sectors can keep buying through a rule change. Both halves are rare together. |
| Platforms assembled quickly without integration | Loser | An industry whose headcount is 3.4% below its peak does not reward acquirers who cannot retain. The returns-disappoint scenario is mostly about this group, and so is the second round's downside. |
| Large sponsors with wide portfolios | Loser on independence, winner on capital | The breadth that makes a sponsor a good owner in most industries is a liability here: the wider the portfolio, the more clients an attest firm inside it cannot serve. Expect structures designed to ring-fence funds from each other. |
| State boards and NASBA | Winner in authority, loser in workload | They gain the decisive role and acquire the obligation to supervise structures they did not design, in an industry with no ownership register. One board has already said in public that it is unsure it can enforce what is proposed. |
| Independent mid-market audit firms | Winner, quietly | Every firm that joins a platform narrows the pool of auditors a given company can use without an independence question. Scarcity accrues to whoever stays out, and it accrues faster as the consolidation succeeds. |
Frontier Desk is the advisor to the firm's owner, never a broker. The first conversation covers which buyers would want your firm, what each would check, and which terms in a structure like this will matter in five years.
Commoditised or scarce
What becomes a commodity in this market and what keeps its scarcity value, with the date we think each crosses. This band is a forecast, and the band after it argues against itself.
Commoditising, and by when
| What | By when | Why |
|---|---|---|
| The alternative practice structure itself | Already | About 72 platforms and 69 capital sources over ten years means the legal work is a form. Nobody pays for the structure in 2028; they pay for what it is allowed to do. |
| Compliance tax preparation and bookkeeping | 2028 | A shrinking workforce and a consolidation thesis that depends on margin both point the same way. This is the work the automation argument is actually about. |
| Capital for accounting platforms | Already | $49 billion across 69 capital sources, per one tracker, in an industry with no filing requirement. Money is the input this market has most of. |
| Generic advisory services inside a platform | 2028 | Once several platforms offer the same specialisms at the same scale, the differentiator stops being capability and becomes which clients the structure permits. |
| Headline deal multiples at the small end | 2027 | About two times revenue is now a known number in a market with 525 events behind it. Owners quoting it at each other is what a commoditised price looks like. |
Keeping its scarcity value, and why
| What | How long | Why |
|---|---|---|
| A clean independence position across an investor's whole portfolio | Through 2028 and beyond | A rule reaching significant influence makes this a property of the owner rather than the firm. The wider the portfolio, the scarcer it gets, and it cannot be bought back once lost. |
| Partners and staff who stay after an acquisition | Through 2028 and beyond | Industry employment is 3.4% below its January 2024 peak and degrees fell 6.6% in a year. The one input that decides whether about two times revenue was a good price. |
| Independent mid-market attest capacity | Through 2028 | Every firm absorbed narrows the pool of unconflicted auditors. Scarcity accrues to firms that stay out, and it accrues faster the better consolidation works. |
| A licence, and the people who may hold it | Through 2028 and beyond | Accounting degrees awarded fell 6.6% year on year in 2023–24, with master's down about 15%. The 13,070 fourth exam sections passed in 2024 against 19,382 in 2023 look worse than the pipeline is, because the AICPA ties most of that fall to a reporting-cycle change. Capital cannot create a CPA, and in every structure the licence stays with the CPAs. |
| Client relationships that belong to the firm rather than a partner | Through 2028 | This is the difference between buying a business and renting a partner. It is the single most valuable thing an owner can establish before a sale, and it takes years. |
The case against this table.
The strongest argument against this table is that we have built a market view on one unverifiable tracker. The deal count, the capital, the platform count, the revenue and the headcount all come from a single publication, and no regulator, filing or statistical agency publishes anything comparable. If that tracker over-counts — by including minority stakes, add-ons and recapitalisations as separate events — then the pace is slower than we show, about two times revenue is not a market price, and consolidation is a smaller phenomenon than the regulatory response implies.
The second argument is that we have treated a proposed rule as though it were an adopted one. The AICPA exposure draft is in a comment period, the first version drew 81 letters, and standard-setters routinely narrow proposals under pressure from the people they regulate — who, in this case, now include some very large platforms with counsel. If the adopted interpretation ends up confined to control relationships, the independence constraint largely disappears, rules-bite should be well under 31%, and consolidation-continues deserves most of what we took off it.
What to do about it
What the forecast implies for the two readers this report is written for: the owner of an accounting firm, and the adviser or investor underwriting one. Frontier Desk is an advisor, never a broker, and nothing here is legal, ethics or accounting advice.
If you own the firm
| Decision | What the forecast says |
|---|---|
| Which half are you selling? | Every structure splits the firm: the investor buys the non-attest entity and the licensed CPA firm keeps the attest work, the licence and the liability. Work out which half carries your revenue, your people and your reputation before you read a single number, because the two halves are now different businesses with different futures. |
| What does your buyer's investor own? | The revised interpretation would stop an attest firm serving organisations over which its closely aligned entity's investor has significant influence. Ask for a list of the investor's holdings in your sectors, ask what happens if it acquires one of your clients' competitors, and get both answers in the agreement rather than in a meeting. |
| What happens on the second sale? | Platforms assembled in 2021 and 2022 reach their exit windows inside this forecast, and a new investor means a new independence assessment. Ask what the agreement says about change-of-control consent, whether any independence representation survives a transfer, and what the exit is if a new owner makes your book unservable. Those are questions for your own counsel, not positions this page takes. |
| Who decides, after you sign? | A state board has said in public that it expects CPAs to be "the final decision makers with respect to attest services and day-to-day management of the licensed CPA firms". Write into the documents who appoints the attest firm's leadership, what the non-attest entity may require of it, and what happens when the two disagree. |
| Cash or rollover? | Rollover equity pays at the second sale, which is also where the independence position resets and where the returns-disappoint scenario lands. We put consolidation continuing at 21% by the end of 2028. Size the rollover against that, not against the pitch. |
| Should you sell at all? | Scarcity is working for independent firms too: clients that want an unconflicted auditor and staff that want a partnership track. The condition is a real succession plan. If you have one, staying independent is a live option; if you do not, the market for your firm is better now than it is likely to be in 2028. |
If you underwrite or advise on these deals
Diligence the investor's portfolio, not only the firm. The proposed interpretation reaches organisations over which the investor has significant influence, across funds with different general partners. A platform's independence position is a property of its owner's whole book and changes without the firm doing anything.
Model retention explicitly, and say what you assumed. About two times revenue and about $510,000 per head — both derived from one tracker, and both possibly dividing transacted value by a wider population than was transacted — are prices for a business whose producers can resign. No public source quantifies retention in this industry, so the assumption is doing the work and should be visible.
Treat the effective date as a deal term. A rule adopted in 2027 and effective a year later lands on structures being signed now against the 2000 interpretation. Who bears the cost of restructuring to comply is a question that can be answered in the agreement and usually is not.
Frontier Desk is the advisor to the firm’s owner, never a broker. The first conversation covers which buyers would want your firm, what each would check, and which terms in a structure like this will matter in five years: Get your company ready.
See who buys firms like yours, and what each one checks.
A platform, a neighbouring firm and your own partners will value the same firm very differently, and only one of those routes involves an investor's portfolio. A short conversation covers which is which.
The evidence base
The present-state record the forecast rests on, as a reference band. Everything here is a reading from a named public source, with its date; the inferences are in the bands above.
What the standard-setters and boards have actually published
Publicly observable| What | Who published it | When | What it says |
|---|---|---|---|
| Alternative practice structure interpretations adopted | AICPA Professional Ethics Executive Committee | 2000 | Into the Code of Professional Conduct, for a structure in which a public company controls the non-attest entity |
| Task force appointed on private equity in APS | AICPA PEEC | November 2022 | To evaluate whether PE investment in the non-attest entity needs code revisions or non-authoritative guidance |
| Private equity task force established | National Association of State Boards of Accountancy | 2025 | To evaluate the implications of PE investment in CPA firms and support state boards in their oversight responsibilities |
| Considerations and Questions for Boards of Accountancy | NASBA PE Task Force | 24 October 2025 | A white paper for state boards on alternative practice structures and private equity |
| First exposure draft on APS revisions | AICPA PEEC | 29 December 2025 | Would replace ET sec. 1.220.020 in its entirety, with revisions to ET secs. 0.400.36, 1.810.050, 1.210.010 and 1.000.010 |
| Comment letter to the NASBA task force | California Board of Accountancy | 29 January 2026 | On independence and professional standards, disclosure and public understanding, and regulatory oversight and enforcement |
| Member review of the AICPA draft | California Board of Accountancy | 19 February 2026 | Guided by whether boards can enforce the changes, and whether independence holds in appearance as well as in fact |
| Comment letter on the AICPA exposure draft | California Board of Accountancy | March 2026 meeting | Prepared by staff with two appointed board members, published in the board's own meeting materials |
| Revised exposure draft announced | AICPA PEEC | 3 November 2026 | Clarifying closely aligned entity and network firm, broadening covered members, and restricting attest work where the investor has significant influence |
| Comment period and expected effective date | Journal of Accountancy, reporting the announcement | 3 November 2026 | 120 days from posting; effective one year after adoption in 2027, with earlier implementation permitted |
Every row is quoted or summarised from the body's own published document, read on 5 November 2026. The AICPA exposure draft text was read as published in the California Board of Accountancy's own public meeting packet, which reproduces it in full as an attachment; the two rows attributed to Journal of Accountancy are named trade press reporting an announcement and are graded accordingly. Frontier Desk is not a CPA firm, does not provide attest, tax or accounting services, and takes no position on what the final rule should say.
What the market looks like, as far as anyone publishes it
Publicly observable| Reading | Source | Figure | Grade |
|---|---|---|---|
| Transactions over ten years | CPA Trendlines Research, 9 Jun 2026 | 525 events, including 473 acquired firms, to 31 May 2026 | C · crowd — one tracker, method not published |
| Capital transacted | The same tracker | $49 billion of enterprise value | C · crowd |
| Platforms and capital sources | The same tracker | About 72 platforms and 69 capital sources | C · crowd |
| Revenue and people covered | The same tracker | About $24 billion of annual revenue, nearly 96,000 accountants and staff | C · crowd |
| Implied pace | FD estimate from the same tracker | 525 ÷ 10 = about 53 events a year | C · crowd |
| Implied revenue multiple | FD estimate from the same tracker | $49bn ÷ $24bn = about 2.0 times annual revenue | C · crowd |
| Implied value per person | FD estimate from the same tracker | $49bn ÷ 96,000 = about $510,000 per accountant and staff member | C · crowd |
| The largest 2026 transaction | Crowe and KKR announcement, 11 Jun 2026 | KKR funds invest in Crowe Advisory LLC through North America Fund XIV; Crowe LLP remains the licensed CPA firm | B · party |
| US accounting-services employment | Bureau of Labor Statistics series CES6054120001 | 1,121.2 thousand in Sep 2026; peak 1,160.5 thousand in Jan 2024; 1,013.8 thousand in Jan 2019 | A · record |
| Change from peak and from 2019 | FD estimate from the same series | −39.3 thousand (−3.4%) from the peak; +10.6% since January 2019 | A · record |
| Accounting degrees awarded | AICPA Trends report via CFO Dive, 27 Oct 2025 | Down 6.6% year on year in 2023–24; master's about −15%, bachelor's −3.3% | C · crowd |
| CPA exam completions and enrolments | The same report and article | 13,070 passed a fourth section in 2024 against 19,382 in 2023 — a fall the AICPA ties substantially to a reporting-cycle change — with enrolments up about 12% | C · crowd |
The workforce the capital is buying
Publicly observableAll employees in US accounting, tax preparation, bookkeeping and payroll services, in thousands, from the statistical agency's own series
Source: Bureau of Labor Statistics series CES6054120001, read 5 Nov 2026
The pipeline behind it, indexed to the prior year
Publicly observableEach bar is that year's figure as a percentage of the one before, so 100 is flat. The underlying counts are in the labels
Source: AICPA Trends report as reported by CFO Dive, 27 Oct 2025, read 5 Nov 2026
Degrees and exam figures are from the AICPA Trends report as reported by named press on 27 October 2025, because the report itself sits behind a download we could not retrieve. They are graded C and attributed wherever they appear. The same article records the AICPA tying the fall in exam passes substantially to a change in reporting methods it expects to moderate, so the exam bar is not a pipeline measure.
Three gaps in this evidence base are worth stating plainly, because the forecast above is weaker where they are.
First, there is no public record of this market. Every deal figure on this page comes from one tracker whose method is not published, and no securities regulator, state board or statistical agency publishes a count of private equity investment in accounting firms. Second, we read one state board's materials in full and fifty-four jurisdictions not at all, so every statement about what boards will do generalises from one. Third, the revised exposure draft announced on 3 November 2026 had not been posted when this report was written, so its text is described from the standard-setter's own announcement and from named trade press rather than read directly; the December 2025 draft, which we did read in full, is what the quotations come from.
Key assumptions
The assumptions the odds depend on, and the evidence that would break each one. Each is an analytical commitment, not a reading.
| Assumption | Breaks if |
|---|---|
| The adopted interpretation reaches significant influence rather than only control. | A final interpretation confined to control relationships, which would leave most existing structures unaffected and collapse the rules-bite scenario. |
| State boards supplement the code rather than simply adopting it. | NASBA recommending uniform adoption with no state-specific additions, and boards following it without proposing ownership or disclosure rules. |
| The tracker's deal figures are approximately right about the scale of this market. | A regulator, a filing requirement or an independent count producing materially different numbers, which would reprice everything derived from them. |
| Partner and client retention is the binding constraint on platform returns. | A platform disclosing strong multi-year retention, or sustained organic growth at platforms that would be impossible with the attrition we assume. |
| Industry employment does not recover past its January 2024 peak inside the window. | Two consecutive quarters above 1,160.5 thousand on the Bureau of Labor Statistics series. |
| Platforms assembled in 2021 and 2022 reach exit inside the forecast window. | Widespread use of continuation vehicles or permanent capital, which would postpone the second round past 2028. |
| The attest and non-attest split survives as the standard structure. | Any jurisdiction permitting non-CPA majority ownership of attest firms, which would remove the reason the split exists. |
| Liability continues to concentrate in the licensed attest firm. | A litigated or settled case allocating professional liability to the non-attest entity or its investor, which would change how every structure is documented. |
| No federal regulator takes jurisdiction over firm ownership inside the window. | The SEC or PCAOB proposing ownership or governance requirements for registered firms, which would move the whole question to a federal forum. |
| Automation delivers margin slowly enough that retention still decides returns. | A platform demonstrating a large, durable fall in cost per engagement attributable to software rather than to offshoring or headcount cuts. |
Signposts
A signpost is a reading that moves the odds if it crosses the threshold shown. Each one names the scenario or scenarios it moves, so no reading on this page floats free of the forecast it is supposed to update.
| Indicator | Now | Threshold | Scenario it moves | What it would mean | Checked |
|---|---|---|---|---|---|
| Scope of the proposed APS interpretation | Reaches significant influence, per the 3 Nov 2026 announcement | Narrowed to control, or adopted as drafted | Rules bite · Consolidation continues | Narrowed to control and the constraint largely disappears; adopted as drafted and every platform's client list is in play | AICPA PEEC publications |
| State board action on firm ownership | Comment letters; one board questioning enforceability | Any state proposing a disclosure or registration rule | Rules bite | Enforcement moves from the code to the licence, where it actually has teeth | State board meeting materials |
| Deal pace | About 53 events a year over ten years, per one tracker | Below 35 a year, or above 70 | Consolidation continues · Returns disappoint | Below 35 says the capital has paused; above 70 says the window is still wide open | The tracker's own updates |
| Platform sales to second sponsors | None named on the public record | Any one announced | Second round | The exit cycle has started and every firm inside acquires a new investor's independence position | Company announcements and named press |
| Engagements resigned on investor-independence grounds | None on the public record | Any disclosed instance | Rules bite · Audit separates | The constraint has moved from drafting to practice, and specialists are the ones it reaches first | SEC filings on auditor changes, named press |
| US accounting-services employment | 1,121.2 thousand, 3.4% below the Jan 2024 peak | Above 1,160.5 thousand, or below 1,090 | Returns disappoint · Consolidation continues | Recovery supports a volume thesis; a further fall makes retention the whole question | BLS monthly series |
| Accounting degrees awarded | Down 6.6% year on year in 2023–24, with master's down about 15% | Two consecutive years of increase in degrees awarded | Returns disappoint | The people constraint is easing, and the automation thesis becomes optional rather than necessary. Exam completions are excluded from this threshold because the AICPA ties the 2024 fall to a reporting-cycle change | AICPA Trends report, annual |
| Platform retention disclosure | None published | Any platform disclosing partner or client retention | Returns disappoint · Consolidation continues | The market gets its first number for the assumption every valuation rests on | Company announcements and lender presentations |
| Platforms divesting attest practices | None on the public record | Any one | Audit separates | The two halves have stopped being worth holding together, and the structure's whole rationale weakens | Company announcements and state board records |
| Federal regulator action on firm ownership | None in the current window | Any SEC or PCAOB proposal on firm ownership or governance | Rules bite · Audit separates | The question moves to a forum with subpoena power and a filing requirement | SEC and PCAOB rulemaking dockets |
| Comment letters on the revised draft | 81 on the December 2025 draft | Above 120, or below 40 | Rules bite · Consolidation continues | A large response means the proposal is contested and likely to narrow; a small one means it is settled | AICPA PEEC comment files |
| A litigated or settled liability case in an APS | None on the public record | Any one allocating liability outside the attest firm | Audit separates · Rules bite | The assumption every structure is documented against has been tested, in whichever direction | Court dockets and named press |
Calls
Each call is dated, carries a Frontier Desk judgment as a percentage, and names the public source that settles it. No call on this page is settled by a social post, an unnamed source or a private record. We score each one when it resolves, with the Brier score: the squared gap between our odds and what happened, where 0 is perfect.
| Call | Resolves | Odds | Settled by |
|---|---|---|---|
| The revised AICPA exposure draft is posted and its comment period opens | 31 Mar 2027 | 91% | AICPA PEEC publications |
| One or more state boards of accountancy file comment letters on the revised draft | 31 Mar 2027 | 83% | State board meeting materials and the AICPA comment file |
| PEEC adopts a revised Alternative Practice Structures interpretation | 31 Dec 2027 | 72% | AICPA PEEC publications |
| A US jurisdiction proposes a CPA firm ownership disclosure or registration requirement | 31 Dec 2027 | 54% | State legislatures and boards of accountancy |
| A US accounting platform is sold or recapitalised with a second financial sponsor | 31 Dec 2027 | 58% | Company announcements and named press |
| A revised Alternative Practice Structures interpretation is in effect | 31 Dec 2028 | 66% | AICPA PEEC publications |
| A platform discloses restructuring undertaken to meet the revised independence rules | 31 Dec 2028 | 47% | Company announcements and state board records |
| A named US accounting platform publicly discloses a material partner or client-retention problem | 31 Dec 2028 | 38% | Company announcements, lender presentations and named press |
| A firm publicly resigns an engagement citing its investor's independence position | 31 Dec 2028 | 26% | SEC filings on auditor changes and named press |
| US accounting-services employment exceeds its January 2024 peak of 1,160.5 thousand | 31 Dec 2028 | 22% | Bureau of Labor Statistics series CES6054120001 |
| Reported deal pace falls below 35 events a year | 31 Dec 2028 | 41% | The tracker's own published updates |
| A platform divests or shuts an attest practice | 31 Dec 2028 | 19% | Company announcements and state board records |
| The SEC or PCAOB proposes a requirement on registered firm ownership or governance | 31 Dec 2028 | 34% | SEC and PCAOB rulemaking dockets |
| Accounting degrees awarded rise in two consecutive reported years | 31 Dec 2028 | 43% | AICPA Trends report |
| A litigated or settled case allocates professional liability outside the licensed attest firm | 31 Dec 2028 | 16% | Court dockets and named press |
| Market: consolidation continues / rules bite / second round / returns disappoint / audit separates | Jan 2028 / Jan 2029 | 36/24/18/14/8 · 21/31/22/18/8 | Triggers and markers in Scenarios |
| How a $2m–$20m revenue firm changes hands: platform / CPA firm / internal / wind-down | Jan 2028 / Jan 2029 | 44/29/19/8 · 46/23/18/13 | Published deal records, state board filings and the tracker |
Questions to test this
Questions a sharp reader would use to break the argument.
- 01Every deal figure on this page comes from one tracker nobody can audit. If a state adopted an ownership register tomorrow, how different would the real numbers be — and would anyone's thesis survive the difference?
- 02The proposed rule reaches companies over which an investor has significant influence, across funds with different general partners. Can any attest firm inside a large platform actually know its own client eligibility, and if not, who is responsible when it gets it wrong?
- 03A state board asked in public whether it can enforce what is proposed. If the honest answer is no, what is the code for — and does the enforcement gap make the rules a disclosure exercise rather than a constraint?
- 04Paying about two times revenue for a business whose producers can resign is a retention bet. Why has no platform published a retention figure, and what would it mean if the first one to do so were a seller rather than a buyer?
- 05Industry employment is 3.4% below its 2024 peak while capital floods in. Is this consolidation a growth story, a margin story, or a managed decline with a growth story in front of it?
- 06When a platform sells to a second sponsor, every firm inside it inherits a new independence position it did not choose. Is that a term anyone is negotiating today, and what would a provision that actually protected a seller look like?
- 07The licence, the attest work and the liability all stay with the CPA firm while the capital goes to the other entity. Is that split stable under stress, and has any court been asked?
- 08If the adopted rule is narrowed to control relationships — which is what regulated parties usually achieve — does anything in this forecast survive, or is the whole rules-bite scenario an artefact of reading a draft as though it were a rule?
Sources and method
Where the direction comes from
Direction comes from Frontier Desk's standing coverage of the services companies that get bought and the capital that buys them — who acquires, what each checks first, and what changes in the terms. The public surface of that work is at /markets and in the guides at /guides.
The standard every claim had to clear
A figure, a quotation or a market fact is carried only if an independent primary or reputable named source states it — a standard-setter's own exposure draft, a state board's own published meeting materials, a government statistical series, a party's own announcement about itself, or named trade press reporting a named body — and is dropped otherwise. Every source on this page was opened and read on 5 November 2026. No figure on this page comes from a private record, from a company Frontier Desk advises, from any piece of work it is doing, or from an unnamed source.
How a reading was graded
Processing graded every reading A for a record, B for a party's own published statement about itself, and C for a crowd source. The deal figures are all graded C and attributed to the tracker that publishes them in every sentence that uses them, because no filing, regulator or statistical agency publishes a comparable count and the tracker's method is not public. The education figures are graded C for the same reason: they come from a named report quoted by named press, because the report itself sits behind a download we could not retrieve.
What Frontier Desk is not
Frontier Desk is not a CPA firm. It does not provide attest, tax, accounting, legal or ethics advice, and it takes no position on what the final interpretation should say. This report describes what the published record contains and what we judge is likely to follow from it, for an owner deciding what to do.
How the readings became a forecast
Analysis turned the readings into structural shifts, a causal model, phases, scenarios and odds, and every judgment on this page carries its own confidence and the reason for it, kept visibly separate from the grade on the evidence under it. Where a figure is derived rather than quoted it is marked FD estimate and its arithmetic is shown beside it.
What the odds are, and what they are not
The odds on this page are Frontier Desk judgments at two fixed horizons, not the output of a model, and the page says so wherever they appear. They are internally disciplined rather than simulated: each column sums to 100, a call that is a scenario arriving is held at or below that scenario's odds, and a call that is a necessary marker of a scenario is held at or above it. They are intended to be scored and to be wrong in public.
What comes back, and when
Dissemination is this page. Feedback is the signposts, each of which names the scenario it moves, and the Brier-scored calls. The employment series is re-read from the statistical agency before every update. The next update is 5 December 2026.
What we dropped, and why
Several claims that would have strengthened this report did not clear the standard above, and are not on the page in any form.
Counts of how many top-30 or top-100 firms have taken investment
Several widely repeated figures circulate for this. We could not trace any of them to a published, dated list we could check, and the firm rankings they depend on are themselves proprietary. The report carries the tracker's platform and firm counts instead, with its name attached.
Valuations and multiples for individual transactions
Enterprise values attributed to specific deals appear widely in trade press and are rarely confirmed by the parties. The only transaction figures on this page are the aggregate ones the tracker publishes, and the structural details the parties announced themselves.
Practitioner accounts of post-deal experience
Partner accounts of what happened after their firm was bought are the most interesting material available on this subject and the least checkable. None is on this page, which is why the retention question is stated as an open one rather than answered.
The text of the revised exposure draft
The revised draft was announced on 3 November 2026 for posting later that month and had not been posted when this report was written. Everything attributed to it comes from the standard-setter's own announcement and from named trade press; every direct quotation of draft text on this page is from the December 2025 version, which we read in full.
What we still cannot verify
Nobody publishes a count of this market that an outsider can check. Nobody publishes retention data for acquired firms. And we read one state board's materials in full, out of fifty-five jurisdictions that license CPAs.
- 01The standard-setter's own exposure draft of 29 Dec 2025, read in full as reproduced in a state board's public meeting packet: a new “Alternative Practice Structures” interpretation (ET sec. 1.220.020) “to replace the current interpretation in its entirety”, with revisions to ET secs. 0.400.36, 1.810.050, 1.210.010 and 1.000.010; the finding that “revisions to the code are necessary”; and the table setting the existing rule, which “[p]resents an APS in which a public company controls the nonattest entity”, against an APS with PE, in which “[t]he investor may or may not control the nonattest entity” and “the other portfolio companies … may or may not be in the same fund … and the PE investor may have less than control over them”; read 5 Nov 2026: California Board of Accountancy meeting packet, March 2026
- 02The same packet, for what a state regulator actually said: board discussions from September 2025, a comment letter sent to the NASBA PE Task Force on 29 Jan 2026, two members appointed to review the AICPA draft, the two questions that guided their 19 Feb 2026 review — whether boards “can effectively regulate and enforce the proposed changes” and whether independence holds “not just in fact, but in appearance” — and staff emphasis on “the importance that CPAs are the final decision makers with respect to attest services and day-to-day management of the licensed CPA firms”; read 5 Nov 2026: California Board of Accountancy, agenda item and attachments
- 03The revised exposure draft, announced 3 Nov 2026 for posting later in the month with a 120-day comment period: clarified definitions of “closely aligned entity” and “network firms”, broadened “covered members”, and a clarification “that an attest firm cannot provide attest services to other businesses and organizations over which its closely aligned entity's investor has significant influence”; read 5 Nov 2026: AICPA announcement
- 04The same announcement reported with the procedural detail: 81 comment letters on the December 2025 draft, a 120-day comment period, and an effective date one year after adoption in 2027 with earlier implementation permitted; published 3 Nov 2026, read 5 Nov 2026: Journal of Accountancy
- 05The deal record, such as it is: “525 events, including 473 acquired firms”, “$49 billion in transacted enterprise value”, about “72 platforms and 69 capital sources”, “$24 billion in annual revenue” and nearly “96,000 working accountants and staff”, over ten years to a 31 May 2026 cut-off — the only decade-long count of this market, published by a trade research service whose method is not public; published 9 Jun 2026, read 5 Nov 2026: CPA Trendlines Research
- 06The largest 2026 transaction in the parties' own words: “funds managed by KKR will make a significant equity investment in Crowe Advisory LLC, becoming its first institutional capital partner”, while “Crowe LLP will remain a licensed CPA firm and continue to provide all attest services, including audits and reviews”, deployed through “its North America Fund XIV” and “expected to close in the third calendar quarter of 2026, subject to customary closing conditions and required regulatory approvals”; announced 11 Jun 2026, read 5 Nov 2026: The announcement
- 07The workforce this consolidation is buying: all employees in accounting, tax preparation, bookkeeping and payroll services, 1,121.2 thousand in September 2026 against a peak of 1,160.5 thousand in January 2024 and 1,013.8 thousand in January 2019, read from the statistical agency's own series through its public v2 timeseries API on 5 Nov 2026: Bureau of Labor Statistics, series CES6054120001
- 08The pipeline behind it: US accounting degrees down 6.6% year on year in 2023–24 with master's down about 15% and bachelor's down 3.3%, 13,070 candidates passing a fourth CPA exam section in 2024 against 19,382 in 2023 — a “substantial decline” the same article reports the AICPA tying to “a recent change in reporting methods”, with most fourth-quarter results counted in Q1 2025 and the impact to “be moderated in 2025” — and enrolments up about 12% — figures from the AICPA Trends report as reported by named press on 27 Oct 2025, because the report itself sits behind a download we could not retrieve; read 5 Nov 2026: CFO Dive
- 09Our own public work on who buys services firms and what each checks first, which this forecast is the two-year view behind; each page checked 5 Nov 2026: The markets we map; The exit roadmap; How AI changes what a services firm is worth; Selling a services firm AI is changing
- 10Our companion forecast on the capital that funds these platforms, published 5 Nov 2026 and read the same day: Private credit and direct lending to smaller services firms to 2028
Find out what your firm is worth to each kind of buyer, and what each would change.
Frontier Desk is the advisor to the firm's owner. We read your firm the way a platform, another CPA firm and your own partners would read it — the client base, the attest mix, the people, the succession — and tell you which route fits and which terms will matter in five years. Answer what you have and the first conversation covers the rest.
If you have already taken outside investment, the same review runs on what you signed: what the investor's portfolio does to your client list, and what happens to your position when it sells.
The rules were written in 2000 for an investor who controlled everything. Almost nobody selling today has read what replaced that assumption.
Frontier Desk outlook, 5 November 2026. Frontier Desk acts as the advisor to the owner of the firm, through its relationships with buyers, lenders and investors, and a cybernetic system of its own — which is why the forecast stays current. Frontier Desk is not a broker-dealer, lender, investment adviser or CPA firm, and provides no attest, tax, accounting, legal or ethics advice. Every figure is quoted from a named public source read on the date stated, or marked as a Frontier Desk estimate with its arithmetic shown. The odds are Frontier Desk judgments, not model output, and may be wrong. Nothing here is an offer, a recommendation about any transaction, or advice on any professional standard. Firms, investors and regulators are named because their published documents are the record this market has, not because they are connected to Frontier Desk in any way.