The GAR 30

How does one “rank” international arbitration practices?

It sounds like something that ought to be easy.

After all, they compete with each other, don’t they? So, in theory, one could simply look at who’s been winning most often and work it out that way . . .

Alas, life is more complicated.

Take “results”. Skilful practices do win more, that’s undeniable. But spotting the “win” is not always easy with the information available. A party could “win” US$100 million, which may become public knowledge. But if it had been hoping for US$2 billion, and needed no less than US$500 million, then that US$100 million becomes more of a loss, especially if the other side was awarded its costs. Unfortunately, these nuances are not readily available to those who would rank firms.

Couldn’t one go by size of claim – “sexiness”, if big means sexy? Alas, it doesn’t help much because there can be more to the case than meets the eye, as this story from the great Jan Paulsson helps to illustrate.

Paulsson, then a junior lawyer with Coudert Brothers, received a “small case from an Asian client, about vested rights”. Some US$50 million seemed to be at stake. Except – and here’s the rub – on scratching the surface the team found the same disputed terms being cited by the opposing side – a Middle Eastern agent – in a swathe of other contracts.

Worse, the client’s corporate loans all contained acceleration clauses. So, if it lost more than US$5 million, they would all become eligible for early repayment.

A “small” case was, in fact, existential.

After a “pretty bloody fight” (which included unearthing a forgery on the opponent’s side (“my first experience of that”)), Paulsson’s client lost.

But not by much – US$2 million.

“They were overjoyed.”

“A silly summary of the case,” concludes Paulsson, “would have been that we lost a little case.” In reality, they saved the client’s business, and they were so grateful that “we all got an all-expenses trip to the Seoul Olympics”.

The opposite is more common: a small case that someone paints as big. Indeed, for many years it was noticeable how few so-called “billion-dollar investment arbitrations” produced anything like US$1 billion for the winner. To give the ISDS industry its due, of late there have been some big wins – see, for example, Yukos v Russia, Occidental v Ecuador, Tethyan Copper Company v Pakistan, Unión Fenosa Gas v Egypt or ConocoPhillips v Venezuela. Sometimes a “billion-dollar claim” means just that. But most of the time it does not.

So how can one reliably gauge the skill and effectiveness of an international arbitration practice?

Around a decade ago, GAR journalists and the magazine’s editorial board debated that question, leading to the thought: what about a survey built chiefly on the number of hearings conducted by the firm in a two-year period?

For the uninitiated, let’s pause to explain a key term: “hearing”. The hearing in international arbitration (particularly the merits hearing) is the closest thing the two sides have to a day in court. It’s the moment they convene somewhere on the planet – in a hotel room, a law firm office or an arbitration hearing centre – for oral argument.

Up to then, the battle has been at arm’s length – conducted in correspondence and teleconferences – from their home countries. But for the hearing, they all decamp to somewhere pre-agreed, for their day in court.

The hearing is when the rubber meets the road.

Nowadays there are a few sub-genres of hearing. Some are purely about jurisdiction; some purely about merits; and some, both. Some carve out damages.

For the senior counsel (the advocates), a hearing tends to mean a period of isolation, beforehand – with the proverbial cold towel pressed to their head – prepping. For the younger lawyers, it means sorting out all the logistics of the presentation, marshalling everything the team will need – and dealing, no doubt, with a more-jittery-than-usual boss.

If it’s a very important case, the team may have “mocked” their case – that is, held a dress rehearsal in front of real arbitrators hired specially to put them through their paces.

Why is the incidence of hearings an indicator of a practice’s expertise and skill?

As a metric, the number of hearings has several strengths. First, it’s innate. Just as fissile material emits radiation, so a busy arbitration practice cannot help but produce hearings. This survey could be seen as a Geiger counter.

Second, it’s the same for everyone. Though some complain it can be affected by the rate of settlement, there is no evidence to suggest that the rate of settlement is different between firms.

Another benefit: we can check it. International arbitration is often confidential, but firms can reveal descriptive, generic information about each hearing (such as language, dates, claim size, opposing counsel, chair of the tribunal and so on) without breaching that, and so allow its existence to be verified.

Most important of all, perhaps, real experience is gained in merits hearings. Any advocate will tell you there’s no better gymnasium for the advocacy muscle than going into a hearing.

So that became our metric. Measure hearings!

And it remains the heart of how we compare international arbitration firms.

Over the years, we’ve added a few tweaks, at the prompting of learned readers (some of which cautiously reintroduce the element of “size” into the final ranking).

Early editions of the ranking were unstable – firms moved up and down a lot year-on-year.

So, we refined the formula to add some ballast and ensure a more gradual shift from year to year. In particular, we now leave out very small matters and certain forms of arbitration (commodities arbitration, and cases in front of centres such as CIETAC).

And we now count jurisdictional hearings as well as merits hearings.

We also added in points for billable hours and market reputation. For the latter we use Who’s Who Legal: Arbitration and Who’s Who Legal Future Leaders: Arbitration (a companion guide aimed at the under-45s), and we also measure the number of times a firm’s members are selected as arbitrators as the yardsticks of reputation. Those who have done it say that it is enormously beneficial for someone who is mainly an advocate to sit as an arbitrator from time to time (though fewer and fewer firms allow it).

Finally, we began to put extra weight on hearings for large and super-large amounts (labelled “large” and “bet-the-company” cases in the chart). While it is true that the numbers in a claim don’t always mean that much in isolation, high-stakes cases tend to gravitate to the lawyers who command most respect in the market, and to be harder fought.

Methodology

Having collected the data, how do we produce the final GAR 30 “ranking”?

The process is similar to assessing a class of students at the end of the year. Those students have sat various tests throughout the year. But some of those tests were easy in relative terms – with everyone scoring at the top end – and others were hard. If one simply added the percentages earned in all tests together, the easy tests swamp the total. In particular, a situation where one student does well (relative to the average performance) in a very hard (ie, low-scoring) test goes unrecognised. This leads to grading on a bell curve, where what matters is relative performance in each test – not the actual score. (An “A” doesn’t mean “80% or above”. It means it is in the top 10% of performances – or something close.) The more high grades, the better the student’s final score.

We use the same approach here, except the “tests” are criteria recorded in a spreadsheet. Lots of A (or A+ or A-) grades equate to a high final position in the GAR 30; Bs and Cs mean a mid-table score; and so on.

A firm’s GAR 30 score derives from the following:

  • how many hearings of above US$10 million it took part in (merits and jurisdictional) in a two-year period;
  • the value at stake in those hearings;
  • the hours it billed to arbitration in the past two years;
  • its performance in Who's Who Legal: Arbitration and Who's Who Legal Future Leaders: Arbitration; and
  • the number of pending cases in which a firm’s members are sitting as arbitrators.

“Hearings” are split into four categories according to their value: bet-the-company cases (US$1 billion and upwards); large cases (US$250 million to US$999 million); medium-sized cases (US$10 million to US$249 million); and small cases (these are ignored, except as a tiebreaker).

Of all the criteria, bet-the-company and large hearings have the greatest effect on final ranking.

We also favour firms with a balance of commercial and investment work over those who purely handle investment work.

It’s important to note a few things that don’t count towards a GAR 30 position.

Wins and losses. A firm’s track record matters – enormously. But for GAR 30 purposes, it is not a factor, owing to the difficulty of assessing who won accurately. If winning counted then, for example, Shearman & Sterling would have been a runaway winner a few years ago, after its US$50 billion Yukos award. As would Debevoise & Plimpton, Arnold & Porter Kaye Scholer, Squire Patton Boggs or, recently, Freshfields Bruckhaus Deringer and Three Crowns, to name but a few who’ve brought in the big wins.

Second, a practice’s success as a business. The GAR 30 doesn’t look at whether an international arbitration practice is operating at a profit, or making money for its members. Indeed, the GAR 30 can seem out of touch with current performance (as the hearings we count are up to two years old at the time of assessing). It is not uncommon for a practice to finish well in a GAR 30, even as team members accept offers from competitors and leave. And on the flipside, a firm that has gone out and purchased extra muscle and talent won’t see that extra heft feed through to the GAR 30 for several editions. This lag is unfortunate but hard to eliminate without also losing one of the ranking’s strengths: objectivity.

How to do well in the ranking

What’s the secret to scoring highly in the GAR 30?

There are generally two paths to success.

One – more certain – path is to do better than average across the board. This is the “well-rounded” route to success. If one looks at WilmerHale, DLA Piper, Squire Patton Boggs and some others, this tends to be how they appear year in, year out; and why some other firms are also regularly in the GAR 30, without particularly standing out in any single column.

The second way is to do exceptionally well in one or two important columns and averagely elsewhere. Usually, the column where the firm outperforms is the bet-the-company column; Gibson Dunn is an example of such a strategy. This is a “big-game hunting” approach.

Both approaches rank well, although the first tends to hold its position year-on-year more easily.

The firms at the very top of the ranking – White & Case, King & Spalding and, some years, Freshfields and Herbert Smith have an element of both – they do well across the board, and outperform, usually in the bet-the-company and large hearing columns.

With the formalities out of the way, let’s turn to this year’s ranking.

The GAR 30 is 15 years old. During that time, it’s borne witness to tectonic shifts.

In its early years, it documented an oligopoly. A group of firms that had put the practice on the map (some might say, built the market) – Freshfields, Shearman & Sterling, WilmerHale, White & Case, Clifford Chance and Debevoise & Plimpton – monopolised the top slots. These “founding father” practices had a chasing pack consisting, in part, of firms staffed by their alumni (Allen & Overy, Herbert Smith, Quinn Emanuel and Three Crowns, and one or two “grand” European names. (The era of Coudert Brothers, the first firm known for international arbitration, predated the GAR 100.)

In the years since, we’ve seen:

  • some of those founding fathers wobble;
  • ditto the original chasing pack;
  • newer names supplanting the old order – particularly from the US;
  • the emergence of the dispute resolution (and even international arbitration) boutique as breakaways from some of those founding firms;
  • a gradual decline in the civil law presence in the table;
  • a swelling of the newer chasing pack, to the point where having a cut-off at 30 firms now feels almost unfair, so similar are the firms at the bottom of the table to the next 20 or so contenders.

What does this year’s GAR 30 add to this picture?

This year’s GAR 30 appears below. The official table shows all of the data that controls the ranking – except billable hours. Two columns that have no effect on the ranking – value of current portfolio as counsel; and number of cases settled in two years – are also presented, because they’re interesting. That said, portfolio size correlates well to ranking and is also a leading indicator of where a firm may be in a few years’ time. So, it is helpful to have to hand alongside the ranking. (See “Methodology”, below, for more.)

This year, 283 firms supplied information, of which we now present the top 30 and the nearest runners-up (the GAR 31–40).

View full size table

 

GAR 30, 2022

What is the mood music this year and how does it compare with the recent editions?

Last year, the GAR 30 was “different, yet familiar”, to quote its commentary. That followed one (2020) that was merely familiar, and one that was the opposite (2019): [2019 was] “a year of sturm und drang – [when] “well-known firms jump[ed] in all directions”.  

The year 2022 appears to share 2021’s weather. Its 30 has familiar outlines but a freshness not seen in a while – particularly in the middle order.

What’s causing that?

First, for two-thirds of the firms in the table, hearing figures are pretty similar in key columns (bet-the-company and large hearings). Things have concertinaed.

Second, the order has been surprisingly jigged around by the implosion (explosion?) of a top 10 regular – Shearman & Sterling – in late 2020 and early 2021 (Shearman & Sterling remains in the table – just lower).

The first means the slightest change tends to be magnified – a bit like a counterweight system: the object descending helps a motor that is raising another object up. This is always a problem when putting firms into a sequence rather than grouping them in bands.

The “loss” of a top 10 was prophesied in this article in 2020.

Commenting on the top 10’s settled look – and how the mid-10 was taking on a similar aspect (at least then) – we asked:

[How long will it be before a firm from the mid-10 breaks into the top 10?] A while, possibly. It is not immediately obvious who [in the middle order] has the juice to break out of this section . . . Of course, things may be taking place of which we are unaware, and which will turn this prediction on its head.

The split of the mighty Shearman & Sterling practice into two approximately equal pieces (emphasis on “approximate”) – GBS disputes, led by the luminary Emmanuel Gaillard, and legacy Shearman – is the sort of thing we had in mind.

It created a vacuum and dropped a rock (or rather two rocks) into the millpond of the middle order.

And there’s covid-19, which may lie behind both the above developments.

We speculated on covid’s effect on the table last year. The number of hearings reported for 2021 was lower – by about 4%. Not the hugest impact.

We hypothesised that this might be six months of lost working time – while the world, and international arbitrators, switched to remote working.

But there seemed one catagory – the very biggest disputes, especially in industries such as mining – where hearings, etc, were more absent. We suggested they were on hold until in-person work could resume.

Overall, we decided it wasn’t possible to be sure of covid’s impact. Whether we’d seen the whole iceberg, or just the tip, wouldn’t be clear for a while.

Well, it’s a year on, and the 30 is at its pre-pandemic size. Total hearings are 846 (bet-the-company, large and mid-sized for all 30 firms), which compares favourably to 848 in 2020 (the last pre-covid edition) – and 801 in 2021.

The only category that is “down” is bet-the-company hearings: five hearings or 4% lower than in 2021 (120 hearings this year versus 125), and 11% lower than in 2020 (120 versus 135).

So, maybe it was the iceberg, not the tip (that said, the table usually grows year-on-year, by about 5% a year). So that portion has been missing – for two years.

That doesn’t mean there are no effects.

The distribution of work across the table has changed. Scores are tighter.

Again, it’s perhaps easiest to see in the bet-the-company column. The scores here are up, year-on-year, at the bottom of the table.

But not at the top. This is partly owing to the arrival in the top 10 of Baker McKenzie (which is more of a volume practice with fewer bet-the-company items, although this may be changing) and because Shearman & Sterling has gone from occupying one slot in the table to occupying two – for GBS and legacy Shearman & Sterling.

But it’s also because other members of the top 10 are not posting the numbers that they were in 2020; whereas the 21 to 30 firms are.

It’s a similar story in the large hearings column.

The ends of the table have metaphorically squeezed together. The top and bottom of the table are nearer, as far as our algorithm is concerned.

This is easiest to see in the mid-section of the table. Read from 11th to 25th in the bet-the-company column. The scores are much of a muchness – between three and four matters all the way down (the spread would have been five to one a year ago, and fives were quite common). In the large hearings column, what was a spread of 11 to two a year ago has become one of eight to three.

The result is a table with new names, more boutiques and more of a civil law component, and where some firms climb or drop a long way.

While it looks like an active year, most of the difference makes sense within the confines of the project and are an occupational hazard of attempting to rank firms in single file, rather than collected into “bands”.

A few other details. Although the total for bet-the-company cases is down, the score for mid-sized hearings is at a three-year high: 545 in this edition versus 519 in 2021, and even ahead of 2020 (532).

Another phenomenon we saw this year seems to be a wave of “mega” cases from the Asia-Pacific region. In part, this was a wave of billion-dollar LNG matters from Australia, but not exclusively. It’s the first time work originating from Asia has been so prominent in the biggest hearings, too.

Peter & Kim, White & Case, Quinn Emanuel, Three Crowns and Herbert Smith have all been making hay with these Asian disputes, to name but some. (Anecdotally, this is why various firms are pivoting towards Asia – with offices and attention – having realised it is already their most productive theatre.)

First, the statistics.

Eleven firms go up; 15 go down; and no firm holds static.

Three firms who weren’t in last year appear – Dentons, Linklaters and Sidley Austin; three who were drop out – Lalive, Chaffetz & Lindsay and Clyde & Co). One firm makes a debut – GBS Disputes , the breakaway from Shearman & Sterling formed in late 2020 and led by the late Emmanuel Gaillard until April 2021.

The number of boutiques remains at an all-time high – (GBS, Peter & Kim and Three Crowns – and Quinn Emmanuel at a pinch).

And Lalive is no longer the sole non-Anglo-Saxon firm in the 30. This time, the civil law world is represented by Peter & Kim (Swiss-Korean) and GBS disputes.

At the same time, much about the ranking evokes previous years.

Ways the table resembles last year:

  • same top two – little sign of gap closing with rest of pack;
  • near identical top 10;
  • very similar mid-10 featuring Curtis Mallet-Prevost, Dechert, Jones Day, DLA Piper and Squire Patton Boggs; and
  • Reed Smith, Norton Rose, Orrick, Arnold & Porter, Gibson Dunn Eversheds and Hogan Lovells are also present.

Ways in which it is less typical:

  • more boutiques;
  • three names not in last year;
  • GBS and Shearman are right next to each other (somewhat poetically);
  • no Lalive or Allen & Overy;
  • Hogan Lovells is now in 21 to 30; and
  • Latham & Watkins and Sidley Austin higher than they’ve ever been.

One should beware of reading too much into a single edition of the GAR 30. But across various editions, patterns do form. And, as the novelist Vladimir Nabokov once noted, at a certain point coincidences, “cease to be coincidences and form, instead, the living organism of a new truth”.

If nothing else, the GAR 30 reminds us how competitive international arbitration has grown. Neither the founding fathers of the discipline, nor the “original” chasing pack, are guaranteed high spots anymore.

In fact, right now, nobody can be certain about preserving their position year-on-year, except maybe the top two. And even there – for how long?

One to 10

In this segment of the table:

  • there’s a new number one;
  • White & Case posts a low score (for them) for bet-the-company matters;
  • Baker McKenzie enters the top 10 and Shearman & Sterling drops out;
  • few firms improve their scores on a year ago;
  • but portfolios seem to have grown for many; and
  • the top two still look a long way ahead of the rest.

At the top of the table, King & Spalding and White & Case switch places for the fourth time in four years.

We said last year there was almost nothing between them. Then, it was White & Case’s greater percentage of commercial versus ISDS work that secured top position.

This year was easier. King & Spalding’s scores in the two highest value columns were well up; White & Case’s have fallen back.

King & Spalding arguably posted its best figures in the hearing columns for four years:

  • 2022 (first): 14 – 20 – 24
  • 2021 (second): 15 – 15 – 16
  • 2020 (first): 12 – 17 – 26
  • 2019 (second): 10 – 14 –26

White & Case, on the other hand, is up in some areas but down in others:

  • 2022 (second): 9 – 17 – 39
  • 2021 (first): 13 – 15 – 25
  • 2020 (second): 11 – 15 – 22
  • 2019 (first): 10 – 25 – 31

It’s still a pretty close contest, even with White & Case down in one column.

The two firms are similar in lots of ways:

  • US: both have their roots in the US;
  • big: King & Spalding had 22 partners take the lead on hearings; while White & Case had over 18, but had certain partners carrying a big hearings load (six in particular – one of the busiest was Anne Véronique Schlaepfer);
  • construction and oil and gas-led: oil and gas, energy (including renewables) and construction for both firms contribute around one-third of the total work; and
  • known for ISDS: both carry plenty of ISDS work, and work for states as well as investors.

But there are differences:

  • White & Case’s biggest matters lean more to the commercial side than ISDS (this year, six of King & Spalding’s 14 largest items were ISDS; at White & Case, the proportion was lower).
  • White & Case has far more of a construction component. At least 27 of White & Case’s hearings are construction-related. It’s been a component of the firm’s practice for years. King & Spalding takes some construction – now – but it’s a newer string to its bow (five items this time out). For White & Case, it is the largest single category.
  • King & Spalding is more male-dominated. King & Spalding has many well-known female partners in international arbitration – for example, Ruth Byrne, Elodie Dulac and Elizabeth Silbert – but White & Case has more, and they’re more senior. (King & Spalding took a step in the right direction, recently poaching Samaa Haridi from Hogan Lovells.)
  • White & Case’s work is less same-y (please excuse the technical term). King & Spalding came to prominence in international arbitration with a wave of ICSID cases against Argentina. At times there still seems to be an element of mass representation, for want of a better phrase, to its work: different clients, similar underlying legal matrix.

And that brings us to the main difference in scores between the firms this year: high-value mining disputes.

King & Spalding worked on a series mining disputes – seven, of which five were “bet-the-company”.

White & Case by contrast had one mining dispute – also a billion-dollar affair. All the cases for both firms were ISDS.

So that gave King & Spalding’s scores the edge.

Given the firm’s roots in the wave of cases against Argentina, it’s unlikely this is accidental. Looking back, it had just three mining hearings in 2020 (of about US$300million each). In 2021 another three, but more valuable (two were bet-the-company items). Now it’s up to seven, five very valuable. It looks a lot like a strategy . . .

White & Case’s practice, in some ways, appears more eclectic and looks more impressive on the page. The range of the work is astounding.

But under the GAR 30 methodology this year, King & Spalding is our winner.

Congratulations – to both firms.

What of the gap between the top two and the rest of the top 10, which was identified last year?

Regular readers from last year will know there’s a metric we keep an eye on – bet-the-company and large scores for the top 10 – but that is not part of our algorithm.

A year ago it read like this (“[ ]” indicates gaps):

  • 28 (White & Case)
  • 30 (King & Spalding)
    [ ]
  • 16 (Freshfields)
    [ ]
  • 12 (Debevoise)
  • 12 (Quinn Emanuel)
  • 12 (Clifford Chance)
  • 12 (Three Crowns)
  • 11 (Shearman & Sterling)
  • 11 (Herbert Smith)
  • 8 (WilmerHale)

Viewed like this, the top two were indeed far ahead. We concluded this was a case of White & Case having been out ahead a while, and King & Spalding bridging across.

How about this year?

This year, the same list looks like so:

  • 34 (King & Spalding)
  • 26 (White & Case)
  • 22 (Quinn Emmanuel)
    [ ]
  • 15 (Freshfields)
  • 13 (Debevoise)
  • 12 (HerbertSmith)
  • 10 (Clifford Chance)
  • 7 (Three Crowns)
  • 12 (Baker McKenzie)

So now it looks like Quinn Emmanuel is catching up.

A couple of caveats. This is a very specific view of the ranking and not relevant to the actual method.

Second, Quinn’s combined score grew because of more large, rather than bet-the-company, hearings. The firm posted 15 in 2022 compared to five in 2021. Otherwise, its picture isn’t all that changed. And it is still a long way behind the top two in some other important metrics.

Ergo, this may not presage a challenge from Quinn for either of the top spots all that soon.

All the same, it’s interesting that one more firm in the top 10 appears capable of such a high combined score.

Why has Quinn done so much better?

Well, it appears to have benefited from three seams of work in particular: oil and gas; Asian disputes with a nexus to London; and Geneva-seated cases. Philippe Pinsolle, Ted Greeno, Michael Young and John Rhie all appear multiple times on items in those categories within its 26 largest hearings – with Anthony Sinclair not far behind.

Could Quinn Emmanuel go onwards and upwards from here?

There’s certainly a possible path. A talking point in Paris is, with Emmanuel Gaillard no longer in the francophone market, where will certain elite French corporates take their bigger work – for example, EDF and Thales? GBS is of course a strong possibility, but these clients are creatures of habit and, without the maître, the habit may be broken. Quinn Emmanuel could benefit. Pinsolle is there, a Gaillard protégé with independent renown, backed with the semi-limitless resources of a much bigger firm – something to keep an eye on.

In the short term, it seems more likely that Quinn Emmanuel will slip back. Its portfolio has shrunk – from US$68.9 billion to US$56.4 billion between editions. This suggests the pipeline needs some refilling.

Elsewhere, the top 10 is a little unremarkable.

The big change is Baker McKenzie’s arrival (rising seven places)

Its year-on-year figures are better, but not to an extent one can claim as in character as a practice that has fundamentally changed. We’ve noted in the past that Baker McKenzie is a firm that’s strong where others are weak, and weak where others are strong, and as a result small changes in certain columns can cause it to jump or fall – yo-yo – more than most.

It arrives more because there is a space. Shearman & Sterling, which was previously a charter member of the top 10, split in 2020. A firm producing six to eight bet-the-company hearings per year (and lots more) became two smaller parts.

That created a vacuum and, according to our algorithm, this time Baker McKenzie is the firm to fill it. That said, there are signs the firm is starting to target more top-end work. So it may be that it becomes a permanent sight in the top 10.

In other top 10 news, Herbert Smith and WilmerHale both rise (three and one places, respectively). And Clifford Chance, Debevoise and Freshfields descend – one place each.

Again, there’s not much sign of deep change in any of these practices. Herbert Smith year-on-year performed about the same, and the same for Wilmer. Debevoise, despite its slip, performed better; Clifford Chance slightly worse.

Incidentally, this is the last year Debevoise & Plimpton will appear in the ranking on the back of any contribution from Donald Donovan. He has recently retired from the firm (as of February 2022).

Quinn Emanuel and King & Spalding, then, are the only two firms in this year’s top 10 showing much change.

Herbert Smith and Clifford Chance, in particular, are some ways off their 2020 (pre-covid) levels for the largest hearings. Then, they both had 11 bet-the-company items apiece.

They haven’t got close since: Herbert Smith is on six this year, and Clifford Chance is on four (they were five and six in this count respectively last year). It looks as if 2020 was an anomaly for them, rather than the norm.

Will this be the story of the top 10 next year?

There are a couple of leading indicators; portfolio value, retirements and investment in oneself are arguably the most useful.

The portfolio values for the top 10 show some noteworthy changes. In particular, the total case values for the following firms increased:

  • Freshfields: by 34% on a year ago (from US$86.7 billion to US$120 billion);
  • Herbert Smith: by 22.7% (from US$99.4 billion to US$122.4 billion); and
  • Clifford Chance: by 31% (from US$80 billion to US$105 billion).

And King & Spalding’s and White & Case’s have also increased.

So, one could see interesting things from those.

On the other hand, Quinn Emanuel, Debevoise, Baker McKenzie and Three Crowns have all shrunk a bit in this metric.

Then there’s opening offices and adding partners.

Three Crowns stands out here. Away from the GAR 30, in the real world it’s been strengthening itself and investing – opening a new office in Singapore (with two founding partners – one a lateral hire), and adding two further partners in Europe and an of counsel in DC. Its figures in the table have yet to change, but business must be good. So, one might expect to see that feedthrough as improved figures in due course. It’s also a salutary reminder that “success” in the GAR 30 is very different from success in the real world.

And that’s the top 10 this year.

11 to 20

The middle of the table sees more action:

  • five firms join this section and six depart;
  • new names include Shearman & Sterling, Peter & Kim and GBS;
  • Hogan Lovells and Baker McKenzie are absent;
  • both parts of the “original” Shearman & Sterling are right next to each other; and
  • the number of firms capable of finishing in the mid-10 occasionally, and number capable of finishing in it consistently, are starting to be very different.

Five firms join this section of the table:

  • Shearman & Sterling (-5)
  • GBS (new)
  • Squire Patton Boggs (+2)
  • Latham & Watkins (+13)
  • Peter & Kim (+10)

Latham & Watkins and Shearman & Sterling have been in the 11 to 20 section before. GBS and Peter & Kim are newcomers.

This means the following leave:

  • Baker McKenzie (+7)*
  • Gibson Dunn (-7)
  • Eversheds (-9)
  • Arnold & Porter (-8)
  • Cleary Gotlieb (-11)
  • Hogan Lovells (-17)

*but exiting upwards.

Jones Day, DLA Piper, Dechert and Curtis Mallet-Prevost “survive” from this segment last year, but they jump around a lot. Jones Day and DLA Piper rise five places; Dechert and Curtis Mallet-Prevost fall five and six, respectively.

Why so much change? And to what extent are the rises and falls deserved?

Well, deserved tends to mean earned – as in, connected to improved underlying performance. In that sense, the answer is mostly “no”.

Jones Day for example – which climbs five places – looks much like the practice it was year ago. It records near identical underlying figures (hearings scores of 3 – 8 – 29 compared with 3 – 11 – 10).

This itself is to its credit. It is emerging as the mid-10’s most consistent performer – three or so bet-the-company items and eight to 10 large items most years, without fail, even if where those cases come from sometimes seems a bit random (this year it had bet-the-company items out of Austin, TX, and Australia).

Likewise, DLA Piper looks very much like the DLA Piper of a year ago. It climbs five on hearing figures of 3 – 4 – 39 versus 2 – 5 – 47 a year ago.

Curtis Mallet-Prevost, meanwhile, falls six because it’s gone from 5 – 7 – 9 a year ago to 4 – 5 – 16 , which may look harsh. Curtis Mallet-Prevost isn’t quite back to its pomp of a few years ago: 2018 and 2019 was the high point whereafter it suffered some hiccups on the personnel side. But it is again looking at least consistent year in, year out.

And, as mentioned, a number of firms completely exit this section because they’ve done “worse”. The reality is, they didn’t do very different from a year ago. It’s just a different table this year – even more competitive.

A couple of things are contributing to that.

First, the middle order is having to bend around two new arrivals: GBS and Shearman & Sterling.

Shearman & Sterling – the residual part of Shearman’s IA practice – drops in from above, having been eighth last year. (It’s eighth place finish last year predated the split with GBS Disputes.)

GBS comes in as a debutant, entering at 13. By way of a yardstick, when Three Crowns made its debut in the table, after separating from Freshfields (mostly), it entered the table at 15.

The GBS and legacy Shearman & Sterling juxtaposition is poetic if undeliberate.

It’s also not a great surprise.

If you have a firm habitually posting six to eight bet-the-company hearings per edition and with a US$48 billion portfolio, and split it down the middle, this is approximately where you end up:

  • original Shearman & Sterling hearing figures: 7 – 4 – 5 (portfolio US$48 billion);
  • legacy Shearman & Sterling hearing figures: 4 – 3 – 5 (portfolio US$22 billion); and
  • GBS hearing figures: 3 – 7 – 6 (portfolio US$32.2 billion).

However, GBS’s scores ignore all of the serious litigation required to defend the Yukos awards that’s been underway in the Dutch court system, and a number of very large settlements.

Of course, 50:50 may not be the correct apportionment. Things remain inchoate. Our invigilators have done their best to assign hearings where they most belong, but inevitably it’s difficult.

The mid-10 doesn’t often have to contend with firms dropping in from above. Most years there seems to be a glass ceiling in place between the top 10 and the rest.

So, when it happens, things shake up even more.

The second reason for all the change is a curious phenomenon.

The set of firms capable of finishing in the middle 10 is growing; but the set of firms capable of doing so regularly is not. It’s shrinking. Really, it’s just a handful of firms – Dechert, Curtis Mallet-Prevost, Jones Day, DLA Piper and (though not this year) Hogan Lovells – who appear regularly at this point. Otherwise, the middle order is composed of a rotating cast of characters who join for a while then fall back.

This cast seems to be expanding. At this point, around 35 firms appear to have the potential to compile the necessary sort of figures based on recent past performances. The list runs through the GAR 30, the 31 to 40 and beyond. It includes network firms and boutiques and names from all over the world.

(It's because of this phenomenon you’ll sometimes see us refer to the mid-table as “oddly same-y” even when it contains many new names. Because it does look the same! The new names have all been in before, just not for a while).

Sure enough, four of the five firms dropping out of the 30 this year had arrived in it not long ago (a year or two years ago, maximum).

Third, as mentioned, covid has had an effect. The spread in certain columns is much tighter, magnifying variations.

It is little wonder, then, why this part of the table has bounced around.

It would be unfair, though, to say no firms in this section ascend on merit – ie, through improved underlying performance.

Dentons, Latham & Watkins and Peter & Kim all posted significantly better figures.

  • Dentons’ line of hearing numbers reads 3 – 4 – 27, compared to 1 – 3 – 14 in 2021. These came from multiple matters in mining and energy, including ICSID cases against Colombia and Bangladesh. However, two out of the seven hearings valued over US$250 million were led by Barton Legum or Jean-Christophe Honlet, who recently left the firm. So, it’s one of the highest risers in the 30 this year, but one to watch next year for that reason.
  • Latham’s figures read 3 – 8 – 8, compared to 2 – 3 – 8 last year. The burst of larger hearings grew in part from defending Colombia, Croatia and Ukraine in investment treaty cases, and from a very healthy commercial arbitration practice in its Paris office (led by another Shearman & Sterling alumnus, Fernando Mantilla-Serrano).
  • Peter & Kim also improves in the underlyings category: 4 – 4 – 3 compared to 2 – 3 – 6 in 2021. A half-Asian firm, its work portfolio is a testament to some of the mega work now underway in, or from, Asia. Three of its bet-the-company items were very big Asian disputes (the business in question was conducted in Asia exclusively), all very different items (ie, not interlinked). The other was a big European gas case. The firm also has some unusually strong metrics in the reputational columns.

Dentons climbs 16, Latham & Watkins 13 and Peter & Kim 10.

And Squire Patton Boggs climbs two. It’s also starting to be one of this section’s most consistent performers.

While those shifts were certainly amplified by the wider atmospheric conditions of the table, it would be hard to say they weren’t deserved.

And if they can sustain those numbers, then they too might join the elite group known as “the mid-table regulars”.

21 to 30

In this section:

  • Linklaters returns;
  • Sidley Austin hits a high point;
  • four firms survive;
  • four drop in from last year’s mid-10;
  • for the first time in a while there’s no debutant; and
  • the 21 to 30 looks like a continuation of the mid-10.

In many ways, the 21 to 30 is the other half of the 11 to 20 equation.

It’s where the firms dropping out, and the firms heading up, meet. Half the firms in this year’s 21 to 30 were in the section above last year.

More so now, the margins in the table are so fine. There’s almost no difference in the hearing figures for Eversheds, Gibson Dunn, Linklaters, Reed Smith, Sidley Austin and Hogan Lovells and the tier above.

Against that backdrop, who seems to be moving in which direction?

Linklaters seems to be moving up, after a slump.

It bounces back into the table, having fallen out (badly) in 2021. It produced the highest climb in this year’s ranking, 20 places, to land in 21st.

Last year, the firm dropped like a stone after a pair of mega cases completed. These now seem to have been replaced by a London-seated oil and gas arbitration, a Portuguese healthcare matter and an infrastructure dispute in Singapore, among other cases. Its portfolio of late has consistently been between US$80 billion and US$95 billion, suggesting it may put together another run of mid-table finishes soon.

Sidley Austin is also back into the 30. It climbs 12 places to finish in 25th place – its highest ever finish (it’s previous highest was 28th after a long run in the GAR 31 to 30).

Sidley’s portfolio value is commensurate with this area of the table; that value also hasn’t climbed in a few years, suggesting that when the current crop of cases works their way through the system, it may be in for another dip. We’ll see.

Reed Smith, who debuted in the ranking three years ago, also achieves a personal best – finishing 24th because of stronger hearing numbers. It reported 2 – 7 – 16 compared to 0 – 7 – 16 a year ago. It has also notably improved in the Who’s Who Legal metric and for arbitrator appointments. José Astigarraga, whose arrival as head of arbitration five years ago broadly coincided with Reed Smith’s improved GAR 30 performance, handed the reins to Peter Rosher in Paris this year.

A number of firms have travelled the other way.

Gibson Dunn in 22nd, Eversheds Sutherland in 23rd and Arnold & Porter at 26th were all in the mid-10 last year.

Gibson Dunn is always something of an anomaly in the table – seeming to do only mega cases. When those are active, it’s a recipe for success in our ranking. Often, those large cases are interlinked (something our invigilators keep an eye on). It also means the practice goes through cycles in GAR 30-terms, which it doesn’t seem to have quite the throughput to smooth out. This year, Gibson Dunn reported around half of the bet-the-company items it was reporting between 2018 and 2019. So, it’s back to yo-yoing: after two years below the middle order, then three years in it – at 15th, 17th and 15th, it is on the way down again (now 22nd).

Still, the practice clearly remains in good health, with multi-billion-dollar investment treaty instructions against India and, apparently, a threatened claim against Qatar.

Eversheds also falls – from 14th a year ago to 23rd. In 2021, it posted a best-ever five bet-the-company hearings; this is now at three, which is more typical of it over the time we’ve monitored. There’s nothing to suggest that its fundamentals have changed in that time. And like Linklaters, it has a portfolio – US$80 billion – that suggests it could head up the table any time. It may even have what it takes to become a mid-table regular.

Arnold & Porter is down eight this year (from 18th to 26th). The firm commands great respect in the market and is beloved by clients, but its work is slanted to ISDS matters, and the throughput and timing of its cases have always been too slow for GAR 30 success. It seldom accumulates enough big hearings in a particular year. That changed three years ago, though, leading to a series of top 30 finishes, including an all-time high of 18th position last year when it posted figures of 3 – 6 – 3 for bet-the-company, large and mid-sized hearings.

This year, the figures have slipped a little (to 2 – 2 – 9), hence the slide. It is notable, though, that mid-sized items rose, which may be a conscious decision by the firm to build the practice out. It was certainly busy enough to make a lateral hire – adding the respected Mélida Hodgson, previously of Jenner & Block.

Twenty-eighth is Hogan Lovells’ lowest appearance in the ranking.

For the first 11 editions of the 30, Hogan Lovells was a top 10 name, with the likes of Shearman, Clifford Chance and WilmerHale. In 2019, it dropped down a level (as did WilmerHale).

Since then, it has been a mid-table regular™, near the top (11th, 13th and 11th) and we have been wondering when it would bounce back (as WilmerHale did). Now, it’s in 28th. Has something fundamental changed?

In all fairness, no; there’s no evidence of that. Its figures are lower this year – from 3 – 7 – 32 a year ago to 1 – 5 – 32 this year. But the hours being billed to arbitration are as large as ever (and commensurate with a higher finish in the table), and the dip in high-end figures is explainable with timing (and even decisive success – aka winning). And, although it did recently lose a highly respected partner in the US (Samaa Haridi) there is no reason to believe there’s any discontent within the IA group or the wider firm. So, it is probably just one of those years; in a year where small differences led to a big climb or drop.

The firm itself is proud of its performance in the past 24 months, and it’s enjoyed real world success. Notably, it won a hotly contested PCA case over an oil block for Georgia, and affiliated entities (defeating Frontera), obtaining costs and seeing off a billion-dollar counterclaim.

Orrick Herrington & Sutcliffe is down eight places, to 29th. It’s another firm for whom meagre change in its figures has resulted in a large change of position. Year-on-year it is almost unchanged – 2 – 6 – 14 for hearings last edition versus 2 – 4 – 13 now. Despite this “fall”, in an era when everyone else in this part of the table is yo-yoing and falling out completely, this is Orrick’s seventh top 30 finish in a row. It’s clearly doing something right.

Cleary Gottlieb also falls this year – from 19th to 30th. It had been up to 13th in 2020. In many ways, it typifies the mobility in this part of the table. Since 2012, it’s been in the middle order six times, the 21 to 30 three times and outside the 30 (in the 31 to 40) twice. At no point was there any reason to think its practice had fundamentally changed (its portfolio value has consistently been one of the highest in the lower half of the table – currently at US$70billion). With numbers like that, the firm is always going to produce billion-dollar hearings. It’s just a question of phase, as it doesn’t handle the highest volume of cases. Currently, the practice seems to be in more of an in-between hearings phase. It posted six bet-the-company hearings two years ago, which became 4 – 2 – 3 last year, and is 2 – 4 – 4 this year. Presumably, the pipeline will be restocked soon. That said, the firm did say farewell to ISDS and public international law specialist Claudia Annacker recently (she joined Dechert). It remains to be seen if that will have a deep impact. At Cleary, the client relationships tend to be with the firm, so it may be limited.

That leaves one firm in the 21 to 30 that moved only slightly.

Norton Rose Fulbright slips three, to 27th.

Like Cleary, Norton Rose has travelled all over the 30. Since 2013 (when it took on its current guise through a series of law firm mergers), it’s been in the top 10 (twice), mid-table (once) and 21 to 30 (three times), and outside the 30 three times. Its figures this year are slightly down on a year ago in the large and mid-sized columns (2 – 3 – 13 this year versus 2 – 4 – 25 last year), but its portfolio value has increased to US$35 billion, suggesting it has the raw material to be in the 30 a good few years.

In contrast to a year ago, every firm in the 21 to 30 this year posts at least one bet-the-company hearing. Most had two. Last year, the firms in 28th, 19th and 30th had zero of these between them.

It seems that, in this part of the table at least, covid’s effects have lifted.

That, then, is this year’s 30.

It should be obvious by now that some names from last year are missing – chiefly, Lalive, Clyde & Co and Allen & Overy. Chaffetz Lindsey, which made its GAR 30 debut last year, also doesn’t reappear.

Lalive fell outside the 30 for the first time. In a call with GAR, members of the firm said some of the largest matters in this research period had settled. They also noted it wasn’t a disaster for their associates’ quality of life and professional development for the firm to have a diet with a little more medium-sized work in. (In full candour, there was also a research mix up – meaning one very substantial matter that should have qualified for the table was missed; had this been spotted in better time Lalive would probably have made 30th place.) While still high, this is not where it was once upon a time.

To give it its due, as with (that other example of IA royalty) Three Crowns, Lalive’s absence says a lot about the gulf between GAR 30 success and “real” success. Lalive opened an office in London in 2018 and, by all appearances, the project is going extraordinarily well. From two partners it is now four, having made its first lateral hire (the respected Genevieve Poirier from Skadden Arps) recently.

Clyde & Co failed to make the cut in the high-end work columns. It recorded one bet-the-company hearing this year, but slightly fewer large cases than a year ago, which in the current table wasn’t a winning hand. It’s in the 31 to 40.

Chaffetz & Lindsay made its debut in the table last year on the back of a sequence of billion-dollar cases. Those are no longer on its books, and this year’s scores put it just below the 30.

Allen & Overy declined to participate this year after gradually declining in the table over the past four years. The timing appears unfortunate, as triangulating Allen & Overy’s work from other firm’s information, it appears they would have finished around 20th, and possibly higher – which would have been their best position in three editions. They probably would have done better, as our estimate of their work – 3 bet-the-company hearings, 3 large and 10 mid-sized – is derived from an incomplete source.

(Skadden Arps, another firm that declined to submit and that we triangulated the same way, didn’t do nearly as impressively.)

Are there any firms in the wider chasing pack that look like they could bounce into the 30?

CMS and Mayer Brown both look like contenders. Each has volume of work plus above average scores in the arbitrator appointments column. At CMS those derive from one or two individuals, but at Mayer Brown things are more dispersed. They’re each a little light on the higher-value items.

On that note, CMS’s instructions are seemingly getting larger – its portfolio is now at US$50 billion, and it is reporting nearly double the billable hours of a few years ago.

Derains & Gharavi, which has been in the 30 numerous times, including as high as 19th and 24th, is also posting impressive scores again, often on investment treaty instructions linked to the Middle East and North Africa. It too has impressive arbitrator appointments; this year it had its highest ever figure for large hearings.

WongPartnership looks like another candidate. Its presence in the 31 to 40 shows the 30’s ever-bigger chasing-chasing pack is starting to include Asian firms too. In fairness, it’s knocked on the door in the past – twice. But its practice has clearly strengthened since then. This year, it posted as many billion-dollar matters as those at the bottom of the 30. And its portfolio has grown starkly: from US$6 billion to US$16 billion currently. More evidence that much larger cases are starting to be seen in Asia?

Ashurst, meanwhile, has been advising on a string of billion-dollar cases, many of which it settled, very favourably by all accounts, just before the main hearings. Another one to watch.

Our favourite leading indicator, however, is portfolio size. Although not perfect, it correlates well with the overall ranking.

It is noteworthy that Australia’s Corrs Chambers Westgarth, which makes its debut in the GAR 100 this year, has a portfolio value of (US$101.6 billion). This reflects a series of natural resources and, in particular, LNG and construction disputes that have been unfolding since 2017.

Vinson & Elkins – on US$51.9 billion – is another firm to watch on the basis of potential implied by its portfolio.

And that’s the main GAR 30 this year!

31 to 40

In ranking order, the next 10 firms in this year’s survey are:

  1. Clyde & Co
  2. CMS
  3. Lalive
  4. Wong Partnership
  5. Mayer Brown
  6. BonelliErede
  7. De Brauw Blackstone Westbroek
  8. Derains & Gharavi
  9. Kim & Chang
  10. Winston & Strawn

What can the GAR 30 tell us about the health of international arbitration more generally?

Ordinarily, this is where we modulate to discussing the numerical side of the table for any trends we can detect. What does it tell us about the number of cases taking place? Is the “pie” firms are competing for expanding?

Thanks to covid, we’ve already dwelled on the statistical weather a fair amount.

But let’s look, briefly, in a bit more depth.

First, the total number of hearings recorded in the table has recovered, almost exactly, to its pre-pandemic amount (see Graph A). It recorded 848 hearings in 2020 (the last full pre-pandemic edition), 801 last year and back to 846 this year. This suggests that, after a blip, arbitrators adapted their case management and pressed ahead. Of course, what is missing still is the usual 5% or so annual growth.

Graph A

 

Is there any category of case that is not back to pre-pandemic levels?

Graph B shows the picture for bet-the-company hearings. They’re still trending down but less steeply than a year ago – five fewer hearings in all or a 4% decline, versus 7.4% fewer the year before.

But, as we noted higher up, bet-the-company hearings returned more visibly at the bottom of the table than at the top. A graph for just the top 10 would show slowing down more steeply.

Graph B

 

By contrast, large hearings are back exactly where they were – at 181 – in 2020, after dipping by 4.4% to 173 hearings in 2021 (see Graph C). Again, there hasn’t been any growth over and above the level in 2020 (which one would ordinarily expect).

Graph C

 

Next is the top 10’s share of bet-the-company hearings in Graph D. This has declined. As mentioned, numbers in this column were up year-on-year at the bottom of the table, but not at the top – King & Spalding aside. So this makes sense. Furthermore, the top 10 has “lost” Shearman & Sterling, which tended to post seven such matters a year, and gained Baker McKenzie, which posts as a rule fewer – this year, two.

Graph D

 

But, thanks to Baker McKenzie’s arrival and Quinn’s boom in large hearings, the top 10’s share of large hearings is up (see Graph E).

Graph E

 

The 11 to 20, on the other hand, is neither up nor down in the share of the biggest hearings (see Graph F) and is clearly down in its share of “large” hearings (see Graph G).

Graph F

 

Graph G

Again, that’s largely the disappearance of Baker McKenzie from the mid-tier.

 

These graphs, then, confirm that the 21–30 closed the gap, statistically speaking, on the upper sections. In other words, the space inside the table shrank.

While the number of overall hearings is on a par with 2020, billable hours are well up on both previous years (see Graph H).

Graph H

 

This chart in and of itself is only so helpful. When one standardises it, however, to a core group (to create year-on-year consistency), it shows that on average firms billed more in 2022 than in 2021.

The average hours billed over two years by members of this core group was 315,545 hours. In the previous edition it was 310,937 hours, and before that 299,929 hours and 294,851.

It’s noticeable, though, that this increase is concentrated on a few firms in the core group. When asked for reasons, some said it was simply better tracking of the totals. So, for now, we have to take these changes with a pinch of salt. But it’s very noticeable that for some firms, hours are way up.

Finally, for the first time, we present year-on-year portfolio values.

Graph I

 

Portfolio value is an okay leading indicator for performance in the table. On this basis, Herbert Smith, Freshfields, Clifford Chance, Jones Day and Peter & Kim look like the firms to watch.

And the best place to watch them? This table – next year.

Until then, thank you for reading.


Words David Samuels and Jack Ballantyne

Data Jack Ballantyne


GAR 30 methodology

The GAR 30 ranks firms according to a “score” built by adding up several T-scores. What’s a T-score? In brief, it’s a way of standardising performance across a series of “tests” so those performances can be more easily compared with performance in a different test. Thus, law firms – or students, or football teams – can be compared by aggregating a series of different performances without fear that any single test will come to dominate the final score. It’s the statistical equivalent of grading on a curve.

The GAR 30 T-scores cover:

  • the number of merits and jurisdictional hearings during the research period (two years);
  • the number of arbitral appointments members are handling;
  • the number of lawyers who qualified for the latest edition of Who’s Who Legal: Arbitration and Who’s Who Legal Future Leaders: Arbitration; and
  • the number of hours billed to arbitration in two years.

Furthermore, we subdivided “hearings” into four categories: bet-the-company cases (more than US$1 billion); large cases; medium-sized cases; and small cases (which don’t count towards the ranking).

The following goes into more detail about the columns in the spreadsheet that count towards the final score (and those that don’t), and the “accounting” principles our researchers use.

Caveats and notes on the ranking process

The GAR 30 chart

People in Who’s Who Legal – shows how many members of a firm won entry to the 2022 edition of Who’s Who Legal: Arbitration.*

People named Future Leaders in Who’s Who Legal – shows how many members of a firm won entry to the 2022 edition of Who’s Who Legal Future Leaders: Arbitration.*

Pending cases (as arbitrator) – shows the number of cases in which a lawyer from the firm has been asked to sit as an arbitrator (snapshot on 1 August 2021). It ignores sports and maritime cases, as well as cases at certain sector-specific arbitral bodies or mainland Chinese institutions.*

Merits hearings completed in two years – shows how many merits hearings the firm participated in as counsel or co-counsel during the past two years.*

Jurisdictional hearings completed in two years – shows the same, but for jurisdictional matters. Hearings under a certain duration in commercial arbitrations are excluded.*

Bet-the-company hearings – shows how many hearings were in the US$1 billion-plus range.*

Large hearings – shows how many hearings were in the US$250 million to US$999 million range.*

Medium-sized hearings – shows how many hearings were in the US$10 million to US$249 million range.*

Cases settled – shows the number of arbitrations that ended in a settlement in two years.

Value of current portfolio as counsel – shows the value of all claims the firm is now handling as counsel.

Billable hours (unpublished) – shows the number of hours billed to international arbitration in a two-year period (excluding the work of paralegals, support staff and trainees).*

*These categories count towards the ranking.

“Accounting” policies

When deciding whether to include a particular reported matter and what value to assign it, we use the following rules:

  • “If in doubt, leave it out” – if a matter is insufficiently described (e.g., there is information missing on the date and duration of the hearing, the name of opposing counsel, the value at stake or the name of the presiding arbitrator), it can’t be included in the score.
  • Jurisdictional hearings in commercial cases are only counted if they exceed a certain duration.
  • “One dispute/multiple panels” (different arbitrators) – each hearing is counted.
  • “One dispute/several merits hearings” – count it once.
  • “Disagreement over the amount at stake” (usually affects higher-value disputes only) – standardise the value, based on the highest total, and apply that to all firms.
  • “Large claim/small award” – if arbitrators have ruled on the value of a dispute, that award becomes the amount.

The research period was 1 August 2019 to 1 August 2021.

Q&A on methodology

How do you get this data?

Law firms provide it, subject to us agreeing to keep it confidential.

Why do you include jurisdictional hearings as well as merits hearings?

At the request of some firms that mainly do investment treaty work. We count jurisdictional hearings in commercial cases too, but only if they exceed a certain duration – to prevent essentially “procedural” hearings being included.

Why do you include billable hours?

At the request of firms that felt they were disadvantaged by the exclusive focus on hearings.

Do you accept the claim sizes asserted at face value?

Yes, but where we know that an award has been handed down, the size of the award becomes the value of the case.

What do you do in the following scenarios?

  • Requests for declarative relief: the firm usually provides (on background) an estimate of the value of the relief sought, so we use that.
  • Emergency arbitrations, hearings on interim measures, ICSID annulment hearings or summary dismissal procedures: we don’t count these.
  • Baskets of public international law matters before unique claims tribunals: we tend not to aggregate these into a bet-the-company case.
  • Test cases: again, the firm usually explains the value at stake.
  • Hearings “happening soon”: these aren’t counted. They can go in next year’s survey.

Overall, we try to use common sense and caution.

Aren’t these figures entirely self-reported?

Yes, but we do cross-reference the cases mentioned with the other side, where possible. The exception is billable hours, but there it becomes obvious pretty quickly who the outliers are, whereupon we go back to the firm seeking further explanation.

Are there any weightings in the formula?

The scores for the different categories of hearing (particularly the bet-the-company and large range) count a little more. Who’s Who Legal names and portfolio value are moderately important, while appointments as arbitrator count a little less.

Large firms now discourage arbitrator appointments. Why do you continue to put weight on those in the score?

Some large firms do indeed now discourage members from sitting as arbitrators, but our observation is that it is more for their own business purposes than any nobler reason (such as antipathy for double-hatting). For many years, the same firms trumpeted the advantages of advocates taking appointments, when they could – the rationale being that there is no better training for advocacy skills than being on the receiving end of some. We continue to regard the value of time spent as an arbitrator proven and thus think firms that perform well on this score should be allowed to benefit from it. Many smaller firms are better able to combine the two sides of the profession, and it would be unfair to tilt the formula even more in the largest firms’ favour.

Get unlimited access to all Global Arbitration Review content