On May 1, 2024, two century-old brands stopped existing on their own. Allen & Overy, a Magic Circle firm born in London in 1930, and Shearman & Sterling, a Wall Street institution dating to 1873, became a single entity called A&O Shearman. The press release ran the usual vocabulary. Global. Integrated. Client-first. I skipped past it and went straight to the arithmetic, because the arithmetic is the only part of a law firm merger that cannot be edited by a communications team.
The combined figures were large. Reports put the merged firm at roughly 3,900 lawyers, around 800 partners, close to 48 offices across 29 countries, and combined revenue near 3.4 billion dollars. Those are real numbers. They describe scale. They do not, by themselves, describe value. That distinction is the whole subject of this post.
Here is what the announcement did not say out loud. Shearman had spent 2022 and 2023 shrinking. Revenue slid. Partners walked. Earlier in 2023, Shearman had been in merger talks with Hogan Lovells, and those talks collapsed in March 2023, which is roughly the corporate equivalent of getting left at the altar in front of the wedding party. A&O announced its own deal weeks later, in May 2023. When a firm goes from one abandoned merger to a completed one inside a single calendar year, the second suitor usually has pricing power. That matters for the math.
the number that started the wave
Start with profit per equity partner. PEP is the metric that either closes a law firm merger or kills it, and it kills more than it closes.
Allen & Overy going into the deal reported PEP in the region of 1.9 to 2.0 million pounds, call it roughly 2.4 million dollars at the exchange rates of the time. Shearman’s PEP was harder to pin down because it was moving in the wrong direction, but the firm had been a multimillion-dollar-PEP shop that was compressing. When two firms combine, the partners at the higher-profit firm ask one question before any other. Am I going to make less money next year because of people I did not hire and clients I did not win?
That question has ended more mergers than any conflict check. In 2019, when Allen & Overy explored a combination with the American firm O’Melveny & Myers, the talks fell apart after nearly a year and a half. The reasons reported at the time were the familiar ones. Profit gaps. Compensation systems that did not align. A&O ran a modified lockstep. O’Melveny ran an eat-what-you-kill model closer to the American norm. You cannot bolt lockstep onto eat-what-you-kill and expect the partners to smile through it. So that deal died, and A&O waited four more years to find a target where the math and the timing lined up.
Shearman in 2023 was that target. A firm that has just lost a merger and is bleeding partners is a firm whose PEP problem is somebody else’s opportunity. The gap between the two firms was survivable precisely because Shearman was weaker. That is not cynicism. That is how price discovery works.
I want to be blunt about what the merger math actually measures, because the industry keeps describing these deals with words that mean nothing.
what merger math actually measures
There are maybe six numbers that decide whether a law firm combination creates value or destroys it. Everything else is decoration.
- PEP, and the gap between the two firms. A gap under 15 percent is manageable. A gap over 40 percent means the high-profit partners are subsidizing the low-profit partners, and subsidies produce departures.
- Revenue per lawyer. This is the cleaner productivity signal because it strips out how equity is sliced. Two firms with similar RPL can integrate even when their PEP diverges, because the underlying engine runs at the same speed.
- Realization rate. The percentage of billed hours that actually get collected in cash. A firm can post beautiful billing rates and collect 82 cents on the dollar. The merger deck never shows realization. I always ask for it.
- Leverage. The ratio of associates and non-equity lawyers to equity partners. High leverage lifts PEP. When you merge a high-leverage firm into a low-leverage one, the PEP math changes overnight, and not always in the direction the slides promised.
- Overlap and cannibalization. How much of the combined revenue comes from the same clients paying both firms for the same work? Overlap is not synergy. Overlap is double-counting that collapses the moment the client consolidates its panel.
- Partner retention through the first two fiscal years. This is the number that gets measured last and matters most. A combination that loses 12 percent of its partners in year one has not merged. It has laid off senior people and paid a branding agency for the privilege.
Notice what is not on that list. Number of offices. Number of countries. Total headcount. Those figures fill the announcement and impress nobody who reads a balance sheet. A 48-office footprint is a cost center until it proves it is a revenue engine. Every one of those offices carries rent, staff, technology, and partners who expect to be paid. Scale in a professional services firm is a liability until the cross-referrals show up in collected cash. The question is never how big. The question is how profitable per unit, and whether the units talk to each other.
I have watched too many managing partners present a merger as if adding two revenue lines together were the same as creating value. It is not. Revenue adds. Costs add faster, at least at first. Integration expense, technology consolidation, real estate rationalization, and the compensation guarantees needed to keep the good partners from leaving all hit the P&L before the promised synergies show up. In the first eighteen months, most combinations are cash-flow worse, not better. The good ones price that in. The bad ones pretend it does not exist.
the transatlantic logic, priced out
The strategic story behind A&O Shearman is genuinely coherent, which is more than I can say for a lot of these deals. The two firms did not overlap much. That is the point.
Allen & Overy was strong where Shearman was thin, and thin where Shearman was strong. A&O had a deep English-law finance practice, a large European network, and a global debt and derivatives franchise. What it did not have was a genuinely elite New York-law capability at the top of the US market. That gap is expensive. The highest-margin work in the world runs on New York or Delaware law: leveraged finance, US securities offerings, big-ticket M&A, and the disputes that follow. A London firm without a real Wall Street practice is renting access to the most profitable legal market on earth.
Shearman had that Wall Street DNA. Weakened, yes. Diminished, yes. But the New York-law capital markets and M&A heritage was real, and it was the specific asset A&O could not build organically at speed. Building a top-tier US practice from scratch through lateral hiring costs a fortune and takes a decade, and the failure rate on that strategy is high because the best US partners cost several million dollars a year each and do not move for firms without an established US platform. A merger buys the platform in one transaction.
So the transatlantic logic reads clean. A&O gets US-law depth. Shearman gets a lifeline, a global network, and a balance sheet that was no longer shrinking under it. On a whiteboard, it works.
The trouble with transatlantic logic is that it is always sound on the whiteboard. The transatlantic merger graveyard is full of deals that read clean and executed badly. What determines the outcome is not the strategic thesis. It is whether the two firms can operate a single compensation system without the top American partners deciding that lockstep is a pay cut wearing a tie.
American equity partners at the top of the market can earn four, five, six million dollars and up. A rigid lockstep caps that. If your best rainmaker generates 40 million dollars of business and the compensation system pays them the same as a partner generating 8 million, the rainmaker leaves, and they take the 40 million with them. Every English firm that has pushed into the US has hit this wall. Some adapted their pay systems. Some lost the exact partners they merged to acquire.
A&O Shearman knew this. The firm signaled from the start that it would run a more flexible compensation approach in the US than pure lockstep, which is the only sane answer. Whether it holds is a multi-year question, and the honest answer in late 2025 is that we do not yet know. The retention numbers are still being written. What I can say is that the deal was structured by people who understood the failure modes, which already puts it ahead of most.
Herbert Smith Freehills, Kramer Levin, and the shape of the deal
Then came the next one. In November 2024, Herbert Smith Freehills, the Anglo-Australian firm, announced a combination with Kramer Levin Naftalis & Frankel, a New York firm of roughly 350 lawyers. The deal completed on June 1, 2025, producing Herbert Smith Freehills Kramer, quickly shortened in the trade press to HSF Kramer.
The strategic logic rhymes with A&O Shearman, which is not a coincidence. It is a pattern. HSF was a large international firm, strong in London, Asia, Australia, and the Middle East, with disputes and energy franchises that travel well. What it lacked was a meaningful US presence. HSF’s American footprint before the deal was small relative to the firm’s global scale. Kramer Levin gave it a New York anchor with a respected restructuring, litigation, corporate, and real estate practice.
Look at the relative sizes and you see the same asymmetry. HSF revenue was reported around 1.3 billion pounds. Kramer Levin was a firm generating something on the order of 500 million dollars, New York-centric, well regarded, and comfortably profitable. The combined entity landed near 2,700 lawyers across roughly 26 offices. Once again the smaller American firm was the specific missing asset, and once again the larger international firm was buying US-law capability it could not grow fast enough on its own.
The structural question is where these two deals differ from an older generation of combinations. A&O Shearman and HSF Kramer were both presented as genuine combinations aimed at operating as unified firms rather than as loose federations. That distinction matters more than any single revenue figure, because the alternative structure has a track record, and the track record is mixed.
the verein hedge, and what it hides
For fifteen years, the dominant tool for cross-border law firm expansion was the Swiss verein. Understand the verein and you understand why a lot of headline mergers are quieter than they look.
A Swiss verein is an umbrella structure. The member firms share a brand, some governance, some systems, and coordinated strategy. They do not, in the classic version, share profits. Each member keeps its own P&L. Baker McKenzie has run as a verein for years. DLA Piper used a verein structure in its transatlantic construction. Norton Rose Fulbright, formed through the 2013 combination of Norton Rose and the Texas firm Fulbright & Jaworski, is a verein. Dentons, the largest firm in the world by headcount, is a polycentric verein that expanded through a rapid series of combinations, including its 2015 tie-up with the Chinese firm Dacheng. Hogan Lovells, formed in 2010 from Hogan & Hartson and Lovells, used a verein. Squire Patton Boggs, formed in 2014, is another.
The verein solves the exact problem that killed the A&O and O’Melveny talks in 2019. If the two firms never share a profit pool, the PEP gap stops being a fight. Nobody subsidizes anybody. The high-profit American partners keep their economics. The lower-profit international partners keep theirs. Everybody keeps the brand and shares the referral network. It is an elegant piece of financial engineering, and it let a generation of firms claim global scale without the pain of a real profit merger.
Here is what the verein hides. A firm that does not share profits does not fully share incentives. If I am a partner in the New York member and I refer a matter to the London member, I have handed revenue to a P&L I do not participate in. The referral culture has to be manufactured through internal credit systems and goodwill, because the money does not automatically flow. Clients notice this. A client who hires a verein and expects a single integrated firm sometimes finds two firms in a trench coat, coordinating politely, billing separately, and occasionally disagreeing about who owns the relationship.
That is why the framing of A&O Shearman and HSF Kramer as unified combinations rather than pure vereins is the interesting part. It is the harder path. A real profit merger forces the PEP conversation to happen instead of deferring it forever. It is also the only structure that produces genuine, durable integration, because shared money is the only incentive that never sleeps. The verein defers the hard conversation. The full combination has it up front and hopes the firm survives the answer.
why the graveyard is crowded
Now the part the announcements skip. Law firm mergers fail quietly, and they fail often, and the reasons are almost always the same three or four things wearing different suits.
The loudest failure in living memory is Dewey & LeBoeuf. The firm was itself the product of a 2007 merger between Dewey Ballantine and LeBoeuf, Lamb, Greene & MacRae. It grew fast, took on debt, and guaranteed compensation to a long list of star partners to keep them from leaving. When revenue could not cover the guarantees, the structure buckled. Dewey filed for bankruptcy in 2012, in what remains the largest law firm collapse in US history. See In re Dewey & LeBoeuf LLP, Bankr. S.D.N.Y. 2012. The lesson is not exotic. A firm built on fixed compensation promises and variable revenue is a firm with a mismatch between its liabilities and its cash flows, and that mismatch does not care how prestigious the letterhead is.
Guaranteed pay is the accelerant in most of these fires. You merge, you need to keep the good partners, so you guarantee their compensation for two or three years. That guarantee is a fixed cost. Legal revenue is not fixed. It moves with the deal cycle, the litigation calendar, and the departures of the very partners whose leaving you were trying to prevent. When a downturn arrives, the guarantees become a noose.
Then there is the departure litigation that follows every collapse and many mergers, which turns on an old and unglamorous doctrine. When partners leave a dissolving firm and take client matters with them, who owns the profit on the unfinished work? California answered one version of this in Jewel v. Boxer, 156 Cal. App. 3d 171 (1984), holding that profits from unfinished business belonged to the old partnership. Decades later, when Thelen and Coudert Brothers dissolved, the estates sued the firms that absorbed the departed partners, arguing those firms owed the value of the pending matters. The Second Circuit sent the question to New York’s highest court, which rejected the unfinished business theory for hourly matters, and the federal court followed suit in Geron v. Robinson & Cole LLP and the related In re Thelen LLP litigation, 736 F.3d 213 (2d Cir. 2013). Coudert Brothers, which had itself collapsed in 2005, generated years of parallel fighting over the same doctrine. The point for a merger analyst is simple. When a combination goes wrong, the cleanup runs through the courts for years, and the recovered value is a fraction of what walked out the door.
Culture is the failure mode that never shows up in the model and causes half the damage anyway. I dislike the word because managing partners use it to explain away deals that failed on the numbers. But there is a hard version of culture that is really about compensation philosophy, and that version is measurable. A lockstep firm and an eat-what-you-kill firm do not merely have different vibes. They have opposite answers to the most consequential question a partnership ever asks, which is how we divide the money. When those answers collide, the partners who lose economics under the new system leave, and they are usually the rainmakers, because rainmakers have the most options.
Look at what happened to King & Wood Mallesons. The firm was assembled through vereins linking a Chinese firm, an Australian firm, and the European arm built on the old SJ Berwin. The European arm ran out of money and collapsed into administration in early 2017, one of the biggest failures in the London market, while the Asia-Pacific operations carried on. That is the verein paradox in one story. The structure let the firm expand without a profit merger, and the same structure meant one member could fail while the brand survived elsewhere, having done real damage to clients, staff, and the name.
what I watch after the press release
So when a combination like A&O Shearman or HSF Kramer completes, I ignore the launch coverage and I build a scorecard. It has a small number of lines, and I check them at 12, 24, and 36 months.
Partner headcount, tracked by practice and office, net of laterals. Not gross departures, net. A firm can lose 60 partners and hire 90 and call it growth, but if the 60 who left were the profit center and the 90 who joined are building, the PEP is going the wrong way regardless of the count.
PEP trajectory in constant currency. Cross-border mergers create foreign exchange noise that flatters or punishes the reported number depending on the year. Strip the currency move and look at the underlying earnings per equity partner. If it is flat or rising two years in, the integration is real. If it is falling and the firm is blaming the market, the integration is not paying for itself.
Cross-referral revenue, which is the entire justification for these deals. The whole thesis of a transatlantic merger is that London clients now buy New York-law work from the same firm, and vice versa. If that number is not visibly climbing by year two, the merger produced a bigger brand and nothing else. This is the figure the firms are most reluctant to disclose, which tells you how much it matters.
Realization and collected cash. Merged firms often chase revenue at the expense of collection, and a rising top line with a falling realization rate is a firm working harder to keep less. I would rather see 5 percent revenue growth at 91 percent realization than 12 percent growth at 84.
And the compensation system, watched for the quiet retreat. If a firm announces a unified pay model at launch and then, eighteen months in, starts carving out special arrangements for its most productive US partners, that is the lockstep wall doing its work. The carve-outs are not failure by themselves. They are honesty. But they signal that the clean integration story in the announcement was aspirational, and the real firm is negotiating with its own economics.
the part nobody puts in the deck
I will land where the math lands, not where the optimism does.
A&O Shearman and HSF Kramer are not the same bet, even though the trade press files them under the same trend. A&O Shearman is the larger, riskier, more consequential wager, because A&O paid to fix a genuine gap in its business by absorbing a weakened but pedigreed New York firm, and the entire return depends on retaining and rebuilding the US practice under a compensation system that American rainmakers will tolerate. If that retention holds through 2026 and the cross-referral revenue climbs, it will be remembered as one of the smartest deals of the decade. If the US partners drift out over three years, it will be remembered as an expensive way to buy a diminished brand, and the number that proves it either way is net partner retention in New York, nothing else.
HSF Kramer is the smaller, cleaner bet. A 350-lawyer New York firm folded into a large international platform is a more contained integration problem than a full-scale combination of near-equals. Fewer overlapping partners means fewer economics to reconcile. That is a feature. It also means the upside is more modest, because a New York anchor of that size gives HSF a foothold, not the top of the US market. It is a down payment on a US strategy, not the strategy itself.
What I reject is the idea that the wave itself proves anything. It does not. The 2007 merger that built Dewey & LeBoeuf was celebrated too, right up until 2012. Combination is not a strategy. It is a transaction with a specific and knowable set of failure modes, and the firms that survive are the ones that price those failure modes honestly instead of burying them under a headcount figure and a map with a lot of pins in it.
The transatlantic logic is real. The profit gap is also real, and it is patient, and it does not read the press release. Two firms merged. Whether one firm emerges is a question the spreadsheets will answer in about three years, and I would not trust anyone who claims to know sooner.