REUTERS | Clodagh Kilcoyne

Reform of the Arbitration Act 1996 and confidentiality: it’s not all about an opt-in

In December 2017, the Law Commission launched its Thirteenth Programme of Law Reform. In it, the Law Commission suggests that it might be time to reform the English Arbitration Act 1996 (AA 1996).

Much has been written about what those reforms might entail. A popular suggestion is that it is time to reverse, through statute, the presumption that arbitration is a confidential process.

It is said by those pressing for reform that commercial arbitration suffers from both a lack of transparency and a legitimacy deficit, and that it is also too unpredictable. The proposed solution is to reform the AA 1996 by introducing an “opt-in” system, so that arbitral proceedings seated in England would only be treated as confidential where the parties have expressly provided as such.

We are unconvinced by this proposal.

In its report on the Arbitration Bill in 1996, the Departmental Advisory Committee on Arbitration Law noted that, “there is… no doubt whatever that users of commercial arbitration in England place much importance on privacy and confidentiality as essential features of English arbitration”.

In his 2016 BAILII Lecture, Lord Thomas (the then Lord Chief Justice) suggested that the importance of confidentiality in arbitration is overrated. However, the evidence suggests that this is not the case, with Queen Mary University of London’s 2015 International Arbitration Survey finding that confidentiality was selected by in-house counsel as being “the second most frequently listed valuable characteristic”.

Given the apparent importance of confidentiality to users of arbitration, it is necessary to test carefully the underlying rationale for the proposed reforms; namely, to provide greater transparency and predictability to the arbitral process. It is also necessary to assess whether an “opt-in” system would have the desired effect.


Some strands of international arbitration are moving towards increased transparency at an exponential pace. In the field of investor-state disputes in particular, the new 2014 UNCITRAL Rules on Transparency in Treaty-Based Investor-State Arbitration and the ensuing Mauritius Convention on Transparency in Investment Arbitration reversed the presumptions of confidentiality and privacy in investment treaty arbitration, in favour of a presumption of openness. In investor-state dispute settlement (ISDS), there is a clear trend towards restricting confidentiality.

Should this increased transparency be applied to commercial arbitration?

Transparency helps alleviate concerns about legitimacy in disputes between investors and sovereign states. Disputes involving public bodies often involve matters of public interest. Taxpayers have a legitimate interest in knowing how their tax dollars, pounds and euros are being spent. However, such factors are, if not absent altogether, usually much more muted in commercial arbitration.

Moreover, mechanisms are already in place that allow for a certain degree of transparency in commercial arbitration.

Parties can already bring arbitration proceedings into the public eye by challenging an award before the English courts, or by commencing enforcement proceedings. There is no express rule applying confidentiality to arbitration proceedings before the courts (which includes documents on the court file). It is at the court’s discretion whether or not details of an underlying arbitration will be made publicly available.

The recent cases of Symbion Power LLC and UMS Holding demonstrate the willingness of the English courts to allow awards and other materials arising from confidential arbitration proceedings to enter the public domain.

Additionally, both the International Chamber of Commerce (ICC) and the London Court of International Arbitration (LCIA) already have policies of publishing anonymised awards and decisions in certain circumstances.

It is, of course, important that the legitimacy of the arbitration process is maintained so that the system continues to operate effectively. At its core, however, arbitration is a consensual mechanism for resolving disputes, which is ultimately designed to facilitate trade and commerce. In order to achieve this objective, it should meet the needs of the parties who use it. Otherwise, they may stop doing so.

Beyond academic circles, there doesn’t seem to be much concern about a lack of transparency amongst users of commercial arbitration, whose numbers are consistently growing (as confirmed by statistics published by the LCIA, the ICC and the Singapore International Arbitration Centre (SIAC)).

It therefore seems unnecessary, for the time being at least, to seek to preserve the legitimacy (and presumably also the popularity) of the arbitral process by undermining one of the features that is most valued by its users.


It has also been argued that reversing the presumption of confidentiality would help make the arbitral process and its outcomes more predictable. It is said that one way of achieving this is to encourage a richer body of precedent, helping the development of the law in fields that traditionally rely heavily on arbitration, such as construction or shipping.

This was Lord Thomas’ point in arguing for the reform of section 69 of the AA 1996 (a subject for another day). However, it is not a reason for reversing the presumption of confidentiality because previous awards do not have precedential value.

In those circumstances, there is a risk that depriving awards of confidentiality could have the paradoxical effect of developing two parallel sources of decisions: awards given by arbitrators (without precedential value) and court judgments (that do have precedential value), potentially leading to uncertainty as to which will be preferred by arbitrators in subsequent cases. No doubt parties and their counsel would feel the need to rely on both in their submissions, which would inevitably increase the time and costs proceedings.

As regards the arbitral process itself, in our view, predictability would be better served through the development of soft law principles and guidelines (such as the International Bar Association (IBA) Rules on the Taking of Evidence in International Arbitration) than by doing away with confidentiality.

Inherent disadvantages of “opt-in” systems

Moreover, an “opt-in” system may not adequately address concerns about transparency and predictability in any event.

A default “opt-in” system is Schrodinger’s confidentiality mechanism: it simultaneously works and it doesn’t. For those parties who do not necessarily have sophisticated legal processes or whose commercial contracts are drafted without arbitration-specific advice, it will “work” in the sense that they might find themselves unwittingly deprived of a feature of arbitration they had taken for granted. The “opt-in” system also doesn’t work, in the sense that those parties who actively want to preserve confidentiality will simply opt-in and keep the “curtain of secrecy” firmly shut.


Reversing the presumption of confidentiality risks seriously undermining one of the most valued characteristics of English seated arbitration. Transparency and predictability are valid concerns and objectives, but they are not so pressing in commercial arbitration that such a radical change is required. Although more can be done in both respects, in our view reversing the presumption of confidentiality is too extreme and risks throwing out the baby with the bathwater.

A measured approach is especially important given the current political and economic environment, particularly in the lead up to Brexit. In post-EU times, London will need to work harder to maintain its reputation as a business-friendly and commercially-oriented arbitration hub. Reversing the presumption of confidentiality at such an uncertain time risks sending the wrong message to the market.

Hogan Lovells Ben Hornan Marie Davoise

Leave a Reply

Your email address will not be published. Required fields are marked *

Share this post on: