England: Many attempts, not many successes - what the data says about challenging arbitral awards

The 2023/2024 court year saw a record 1  number of challenges to arbitral awards filed with the English Commercial Court (113), a 59% increase in applications compared to the previous year.

Does that suggest that the English courts are becoming more interventionist, or that it is becoming easier or more common to challenge or attempt to challenge arbitral awards in the English courts? If true, that could have significant implications on the overall perception of England and Wales as an arbitral centre and the English courts’ approach to dealing with arbitral challenges.

However, as is so often the case, those numbers in isolation do not tell the whole story: although the number of applications may be at a high, that says nothing about the number of successful challenges or the courts’ approach to those applications.

So, does the increase in the number of attempted challenges form part of a wider trend? If so, has that translated into an increase in the number of challenges granted by the courts? And to what extent does the number of arbitrations more generally have an impact?

In this article, we consider six years’ worth of statistics from the Commercial Court and various arbitral institutions in order to consider those questions.

By way of introduction, the English law rules on challenges to arbitral awards are contained within the Arbitration Act 1996 (the “Act”), as amended by the Arbitration Act 2025 (the “2025 Act”), which applies to arbitrations seated in England and Wales (and Northern Ireland).

The Act provides for three grounds on which an arbitral award can be challenged:

• Section 67 provides that an award can be challenged on the ground that the arbitral tribunal lacked substantive jurisdiction.
• Section 68 provides that an award can be challenged on the basis that there was a serious irregularity affecting the (i) arbitral tribunal, (ii) the proceedings or (iii) the award, which has caused or will cause substantial injustice to the applicant.
• Section 69 provides for a right of parties to appeal to the English court on a point of law.

The first two grounds are consistent with the Model Law (and the approach adopted in the arbitral laws of most major jurisdictions). 

However, in providing for a right of appeal to the national courts, per s.69 of the Act (which is not provided for in the Model Law), English law is something of an outlier. Although some commentators have raised concerns about the impact of this right of appeal on the principle of arbitral finality, s. 69 of the Act seeks to mitigate this risk by providing that:

• a party can only appeal on a point of law where it has permission to appeal from the court (or the agreement of all other parties); and
• a high threshold must be met before permission to appeal will be granted (including a requirement that the court considering the permission to appeal application should be satisfied either that (i) the decision of the tribunal was “obviously wrong”, or (ii) there is a question of general public importance and the decision of the tribunal is “at least open to serious doubt.”

So, what lessons can we learn from by digging into the data?

1. No clear trend in the overall number of challenges

First, there is no clear trend in terms of the number of applications under ss.67, 68, and 69 of the Act (“Challenge Applications”) brought in the Commercial Court.

Whilst the 2023/2024 court year saw the highest number of challenge applications across the six-year period, that does not appear to be illustrative of a general increase in the number of Challenge Applications.

Indeed, the previous year saw the lowest number of applications during the relevant period and, to the extent that any pattern can be discerned, the data suggests an overall downward trend in the number of Challenge Applications.

2. The number of Challenge Applications is likely to be driven by trends in the number of overall arbitral awards, rather than due to any change in approach by the English courts

In order to provide further context to the number of Challenge Applications we attempted to calculate the number of overall awards issued by the major institutions in relation to English-seated arbitrations (the latter of which would fall within the scope of the Act).

Unfortunately, although the majority of the major institutions publish statistics relating to their caseloads, most of them do not provide details of the numbers of awards handed down in a given year.

However, the ICC International Court of Arbitration (ICC) and SCC Arbitration Institute (SCC) have confirmed the numbers of their English-seated arbitral awards for each year from 2017 – 2022 and the Hong Kong International Arbitration Centre (HKIAC) has provided information that enabled us to estimate similar numbers.

The numbers collected represent a far from complete picture of the number of English-seated arbitral awards issued on a year-by-year basis (in particular, the above data does not include data from the LCIA, which appears to produce the most English-seated awards) and, given that the institutions report on a calendar year (rather than court year) basis, like-for-like comparisons are not possible.

That said, even based on the limited data-set available, it is possible to discern a broad correlation between (i) the number of Challenge Applications, and (ii) the number of English-seated awards issued by the ICC, HKIAC and SCC.

3. This could suggest that, rather than increased judicial interventionism, trends in the number of Challenge Applications are driven by the numbers of arbitrations and arbitral awards more generally. Section 69 is by far the most popular ground of challenge.

When we look at the types of challenges that parties are bringing, it is clear that challenges under s.69 of the Act (i.e. the right of appeal on a point of law) are consistently the most common ground of challenges, whilst s.67 (substantive jurisdiction) is consistently the least common ground.

That is perhaps unsurprising: it is far from unusual for an unsuccessful party to disagree with a tribunal’s decision on a point of law, whereas the circumstances in which a jurisdiction challenge may arise are considerably more limited.

However, given that English law is something of an anomaly in having the right of appeal under s.69 of the Act, does the fact that the right of appeal is such a popular ground suggest that the English courts are more interventionist, and England and Wales less arbitration-friendly as a jurisdiction, as some commentators may suggest? 

4. Number of successful challenges remains low, with no correlation to the number of overall applications

To answer the above question, it’s not enough to look at the number of overall challenges: we need to look at the number and breakdown of successful Challenge Applications. What then do those statistics show?

First, we can see that, in real terms, the number of instances where parties successfully challenge arbitral awards is consistently low, with an average of just under four successful Challenge Applications per year (across each of ss. 67, 68 and 69).

Second, when we compare the number of successful challenges against the number of overall applications (i.e. including those that were dismissed, withdrawn etc), we can see that there is no correlation between the two: even in years where the number of Challenge Applications spikes significantly, there is little impact on the number of successful challenges, which remains broadly consistent. By way of illustration:

• 2021/2022 saw the second highest number of Challenge Applications during the relevant period, but none of those were ultimately successful.
• Similarly, although 2023/2024 had the highest number of overall applications, just two were successful (the second lowest success rate during the period).

• Similarly, when we look specifically at s.69 applications (which, as noted above, involved a permission to appeal stage), we can see that there is little relationship or correlation between the overall number of applications and the number of cases granted permission to appeal: the latter figure is relatively consistent across the relevant period, ranging from 7 – 13 cases a year (or a conversion rate of 17 – 33%). 

 5. Success rates are highest for s.69 applications (though still low).

The rate of success for each of the different grounds of Challenge Applications is also consistently low.

When we look at the success rate for applications under ss. 67, 68 and 69, respectively, we see that across the relevant period, none of the three grounds of challenge have had a success rate higher than 11%, with several instances where there have been no successful applications under a particular ground. The data indicates that, in general, applications under s.69 of the Act have had the most success but the mean and median success rate of such applications is still just 5%, underscoring how difficult it is to convert an application to a successful challenge. Successful challenges under s.69, therefore, remain rare.

6. The success rates for applications under ss.67 and 68 are much more erratic but appear to be trending downwards

The mean rate of success for applications under ss.67 and 68 is 3% and 4% respectively, which suggests, at first glance, that the success rate is roughly the same as under s.69.

However, looking at the year-to-year success rates for ss.67 and 68 applications (as shown in the above chart), we can see that those figures are somewhat skewed by the somewhat anomalous results in 2019/2020. Indeed, the success rate of s.68 applications has trended downwards since then such that no party has succeeded with one since the 2020/2021 court year.

Thus, it is clear that applications under sections 67 and 68 succeed only in exceptional cases such as where there is a clear lack of jurisdiction (see Tecnicas Reunidas Saudia for Services & Contracting Co Ltd v Petroleum Chemicals and Mining Co Ltd [2025] EWHC 1785 (Comm)), serious departures from due process (The Republic of Kazakhstan v World Wide Minerals Ltd [2020] EWHC 3068 (Comm)) or admitted procedural mistakes (Doglemor Trade Ltd v Caledor Consulting Ltd [2020] EWHC 3342 (Comm)).

7. Obtaining permission to appeal is not even an indicator, let alone a guarantee, of success for s.69 applications.

The process of obtaining permission to appeal for s.69 applications  requires parties to show (inter alia) that a tribunal’s decision was either “obviously wrong” or “at least open to serious doubt.”

One might assume, given the difficulty of clearing those high hurdles, that obtaining permission to appeal would at least be a strong indicator that the s.69 application was likely to be successful. However, that is not borne out by the data.

When we compare the rates of success for cases that obtain permission to appeal against applications overall, we see that:

• applications that obtain permission to appeal do, indeed, have a higher rate of success; but
• there is no discernible correlation between the PTA grant rate and the success rate of appeals that follow permission. Indeed, 2021/2022 saw the highest number of successful PTA applications (both in real terms and in terms of rate of success) but also had the lowest number of successful appeals across the relevant period (with zero successful appeals); and
• although obtaining permission to appeal is a major hurdle, the post-PTA success rate is still relatively low, with an average of just over 22%.

We also see that there is no discernible correlation between the PTA grant rate (the light blue line in the above chart) and the post-PTA success rate (the brown line).

Despite the fact that the test for permission to appeal requires (i) a judge to carry out an initial assessment of the merits of an appeal, and (ii) to apply a strict threshold when doing so, we see that more than 75% of applicants across the relevant period that obtained permission to appeal were still ultimately unsuccessful with their s.69 appeals. This suggests that obtaining permission to appeal, therefore, is not a strong predictor an appeal will succeed (and is certainly no guarantee of success).

It is clear from the data that there is no indication that the English courts are becoming more active in the arbitral process.

Instead, statistically speaking, successfully challenging an arbitral award is difficult and successful challenges remains rare, with English courts adopting a very measured approach to the exercise of their powers.

There appears to be no correlation between, on the one hand, the number of ultimately successful challenges and, on the other, any of the number of arbitral awards, Challenge Applications or (where applicable) grants of permission to appeal.

This perhaps illustrates both the difficulty of meeting the high threshold for challenging awards under English law, and that the English courts are meticulous in considering each case on its own merits, without being influenced by external factors or the vicissitudes in the number of challenges or arbitrations more generally.

We can also expect, with the introduction of the recent amendments to the Court’s powers under s. 67 introduced by the 2025 Act (which limit the ability of (i) parties to raise new grounds of objection and adduce new evidence, and (ii) the courts to rehear evidence), the number of successful challenges may reduce further.

The starting point for this analysis was the annual reports published by the English Commercial Court, which is one of three specialist courts within the High Court of England and Wales that deals with challenges to arbitral awards (as well as other types of claims relating to arbitration), and which handles a significant majority of such cases.

Those annual reports provide granular data regarding the arbitration work carried out by the Commercial Court in the relevant year broken down by application type, including the number of:

1.  applications received;
2.  applications for which permission to appeal was granted (where applicable); and
3. challenges that were ultimately successful.

For the purposes of this report, we looked at the Commercial Court statistics for each year going back to the 2018/2019 court year (i.e. the earliest date from which granular data is available).

The work of the Commercial Court is the focus of this article given that it deals with a significant majority of the arbitral challenges that reach the English courts and produces the most detailed statistics. 2

Finally, in order to provide further context to those figures, and as discussed above, we reviewed statistics relating to the number of arbitral proceedings and/or awards from the following arbitral institutions:

• the ICC;
• the London Court of International Arbitration;
• the HKIAC;
• the SCC; and
• the Singapore International Arbitration Centre.

In addition to reviewing the relevant institutions’ own published reports, we were also able to obtain additional data that is not typically publicly available from the records of the ICC, SCC, and HKIAC. We are grateful to those institutions for their assistance.