Disclosure Issues in International IP Arbitration

Introduction

When parties choose arbitration to resolve their IP disputes, there are certitudes to guide them on arbitration law and procedure, and how their case will be handled: the courts will help enforce their arbitration clause; an impartial tribunal will be selected to judge the case without a jury; and the final award will be widely enforceable in the signatory nations of the New York Convention.^[2] But the details of what the particular procedure will be like are often not well known in advance, and in no area are these uncertainties greater than in the procedure the tribunal will adopt for the most important phase of fact-finding in IP and other disputes: disclosure and discovery – that is, the exchange among the parties of relevant documentary and testimonial evidence that can be used to build, or confirm, the legal arguments in the case.^[3]

In part, these uncertainties stem from the great cultural variety of arbitrators, as each arbitrator brings his or her own legal culture to the process of establishing an arbitral fact-finding procedure that will apply to all parties. Arbitrators trained in common law will have more familiarity with document discovery and depositions, and will expect parties to use available tools to develop and refine their arguments; arbitrators from civil law traditions will have less familiarity with disclosure tools, and may see a party’s recourse to them as a sign that the party’s claim is as yet unsupported by available evidence, or worse, unfounded.

However, the uncertainties also derive from a historical lack of procedural rules for this phase of the arbitration. In 1999, the IBA adopted the Rules on the Taking of Evidence in International Arbitration,^[4] in a quite successful attempt to bridge the gap between common and civil law traditions in disclosure. Other institutions and actors contributed to rules and guidance that can frame the disclosure procedure.^[5]

Even with this guidance, tribunals and parties are often left to their own devices to adopt the disclosure process that will be best suited to their dispute, and this should be seen as one of the advantages of arbitration: the ability to fashion procedural rules tailored to the needs of the dispute and reflecting the backgrounds of the participants. And nowhere is such flexibility welcome as much as in IP arbitration, where issues of confidentiality, privilege and trade secrecy are often asserted to limit the organised exchange of documents and testimony.

The arbitral tribunal has wide latitude in setting procedural rules for disclosure. In ICC arbitration, for example, the Rules provide that the ‘arbitral tribunal shall proceed within as short a time as possible to establish the facts of the case by all appropriate means’.^[6]

But their discretion is not complete. The rules must adhere to the principles of due process set out in the New York Convention, or else the ultimate arbitration award could risk annulment by a party dissatisfied with the outcome.^[7] Accordingly, setting the rules of disclosure, and ensuring compliance, becomes indissociable from the tribunal’s mission to reach a valid and enforceable award. These uncertainties should be addressed early on by the tribunals to avoid such later risks.

Certain procedural aspects of arbitration are fairly settled now, with the help of guidance and enforcement decisions. It is common, for example, for parties to exchange their requests for documents through the use of ‘Redfern schedules’,^[8] which identify for each request the document sought and its relevance to the dispute. This is a handy and efficient way of narrowing disclosure disputes so that the tribunal may rule on unresolved matters or dispensing with witness depositions in favour of witness hearings before the tribunal.

But in the following areas there is still uncertainty and, therefore, risk, including a risk that the award may not survive in post-award court proceedings. This chapter will focus, in particular, on (1) privilege as a bar to disclosure, (2) recourse to national courts for assistance in obtaining disclosure and (3) compliance with the EU General Data Protection Regulation (GDPR) in arbitration discovery.

Privilege as a bar to disclosure

Whichever rules are adopted to guide disclosure, the process may be blocked if the producing party can validly invoke legal privilege, such as attorney–client privilege, to prevent production. But which privilege? Since 2010, confusion has reigned over which legal counsel is covered by attorney–client privilege. This is particularly troubling in international arbitration, where counsel from diverse bars are allowed to practise, leading to inconsistent and unsatisfactory results. This is particularly so in IP arbitration where one party, but not the other, may be deprived of the confidentiality protections of attorney–client privilege, leading to an uneven playing field. Some institutional rules have attempted to rectify this, such as the ICDR Rules^[9] and the IBA Rules,^[10] but they do not fully resolve the issue.

In 2010, the European Court of Justice (ECJ) declared in the Azko Nobel case^[11] that the scope of legal privilege in Europe does not extend to in-house lawyers, based on established precedent that in-house counsel are not independent as they are bound to their client by an employment relationship.^[12] The court held that legal privilege only protects independent attorneys – ‘that is to say lawyers who are not bound to the client by a relationship of employment’,^[13] though each EU Member State may adopt more generous privilege rules.^[14] This is the case in Spain, for example, where the rules do not distinguish between external and internal counsel.^[15] Some states, such as Belgium, have adopted new statutes to provide in-house lawyers with the same degree of privilege and confidentiality as independent attorneys.^[16]

Under the rules of the Dutch Bar, all admitted lawyers benefit from legal privilege, regardless of their position as in-house or external counsel. English courts also grant legal privilege to in-house lawyers, specifically with regard to their communications. During the 2010 appeal in Akzo Nobel Chemicals Ltd and Akcros Chemicals Ltd v. European Commission, the United Kingdom and the Netherlands endorsed the order sought by the appellants Akzo and Akcros,^[17] as did the International Bar Association, the Association of Corporate Counsel Association, the European Company Lawyers Association and the Council of Bars and Law Societies of Europe.^[18]

The United States does not view attorney independence as a cornerstone criterion for the existence of legal privilege. Accordingly, legal privilege extends to in-house counsel and accordingly communications between an in-house lawyer and his or her client are privileged, even if that client is the employer.^[19]

Further, the Akzo ruling sheds doubt on whether the broad legal privilege that may apply to counsel outside the European Union, such as for in-house counsel in the United States, would even be recognised in the European Union or an arbitral forum. This leads to an increased risk of unwanted or unforeseen disclosures when non-EU parties produce legally privileged information in an arbitration the seat of which is in the European Union, creating further uncertainty for the parties.^[20]

By way of example, imagine an arbitration seated in Paris over a failed joint venture involving a patent dispute, with US and French parties, represented respectively by US and French counsel each admitted to their respective bars. The US party asks for the documents and email addresses of the French company’s in-house counsel during the time when the French company’s inventors were preparing their patent, while the French party does the same, each seeking to show the other venturer’s lack of good faith performance of the joint venture agreement. As French in-house counsel do not benefit from any recognised legal privilege, they can be required to hand over these documents; but as US in-house counsel are covered by privilege, they can avoid this obligation unless the tribunal decides to alter, one way or another, the scope of applicable privilege.

However, assuming the tribunal applies each legal privilege according to its terms, the French party will be at a disadvantage as it is required to disclose this type of communication. If the tribunal rules in favour of the US party, basing its decision in whole or in part on the evidence produced by the French party’s in-house counsel, this could be a basis for the French party to seek the annulment of the award for violating its due process rights under the New York Convention.^[21]

What can a tribunal do to accommodate the varied privilege claims presented by parties subject to different national laws, and organise a fair hearing? The answer is to avoid applying rigid institutional rules that may not take into account the circumstances, and to address these issues early in the proceedings. Indeed, this is the type of issue that can and should be raised at the start of proceedings,^[22] in the terms of reference or another initial procedural order agreed by the parties, once it is clear based on the composition of the parties and their counsel that issues could arise from unequal treatment of legal privilege.

After consulting with the parties, the tribunal can choose to apply:

• each party’s privilege rules despite potential unfairness, but obtain, if possible, the parties’ consent or waiver to avoid later attacks on the final award;
• the most limiting of the parties’ privilege rules (in this case, the French rule) – however, the US party could complain that the tribunal’s rule would require that it waive valid legal privilege and may have grounds to seek annulment or non-recognition of the award; or
• the most expansive of the parties’ privilege rules (in the case, the US rule), rendering undisclosable by tribunal order the French parties’ otherwise non-privileged documents that, if produced, could affect the outcome of the case.

It could be that the tribunal’s ruling, by limiting otherwise available documents in discovery (in the case of the French party), or by imposing the waiver of a valid legal privilege (in the case of the US party), could create grounds to challenge the validity of the award if the tribunal has not sought and obtained the consent of the parties.

And so we return to the central point of the parties’ consent, and the recommendation, if there are competing legal cultures present before the tribunal, to establish scope of privilege rules early in the proceedings before any dispute has arisen, and to obtain the parties’ consent to such rules.^[23]

Recourse to national courts for assistance in obtaining disclosure

Tribunals and parties should also consider early in the proceedings whether national court discovery in aid of arbitration is appropriate, and if so, how it should be regulated by the tribunal. Fact-finding is at the centre of an arbitration, and the outcome can be greatly influenced by documents or witness testimony exchanged during the proceedings. Parties often commence arbitrations without having marshalled all of the facts that support their claim, or to better understand the position of their adversary, and the exchange of information through the discovery process can help narrow disputed factual and legal claims, and even lead to settlement.

The IBA rules on disclosure provide an invaluable framework for assessing discovery claims in arbitration, and have become a widely used resource for tribunals and parties alike.^[24] But they do not address how an arbitral tribunal may regulate – or even allow – recourse by the parties to national courts for fact discovery. This is particularly significant in IP arbitration, where access to an adversary’s internal documents, such as laboratory testing and meeting notes, engineer or accident reports, emails with patent agents, insurance claims and other internal documents, can often reveal whether a claim is well founded.

Below, we discuss two examples of national laws that may allow for such disclosure, followed by a recommendation on how a tribunal might address this issue.

Federal law in the United States permits any party or other interested person involved in proceedings before a foreign or international tribunal, or the tribunal itself, to make a request to a US federal district court to compel discovery from a person or entity found in the district in which the court sits. For years, however, there have been disagreements as to whether international arbitration proceedings constitute foreign or international tribunals under the statute.

Prior to 2004, the Second and Fifth Circuits had ruled that Section 1782 could not be used to obtain discovery in aid of private arbitration. This year, the Second Circuit had an opportunity to revisit the issue in its decision of In re: Application and Petition of Hanwei Guo,^[25] in which it declined a request for discovery assistance in connection with a commercial arbitration, consistent with its past precedent.^[26]

The US Supreme Court ruled in 2004, in Intel v. AMD,^[27] that a complaining party that had filed an antitrust complaint before the EU competition authorities could obtain discovery from Intel in California of documents that had been compiled by Intel for use in that European investigation. The Supreme Court found that the Directorate General for Competition of the European Commission was a tribunal under Section 1782.

Since Intel, US courts of appeal have differed on whether private arbitrations fall within the scope of Section 1782. In 2009, the Fifth Circuit reaffirmed its previous narrow interpretation that private foreign arbitrations do not fall within this scope,^[28] while two circuits considering the question for the first time, the Sixth^[29] and Fourth Circuits,^[30] reached the opposite conclusion (in 2019 and 2020, respectively).

The Seventh Circuit^[31] has now weighed in on the issue in a case that concerns the same parties and arbitration as the Fourth Circuit decision. Notably, the Seventh Circuit reached the opposite conclusion to the Fourth Circuit, thereby aligning itself with the Second and Fifth Circuits in determining that Section 1782 does not apply to private arbitral tribunals.

The uncertainty created by this law requires parties to international commercial arbitrations to carefully consider their discovery options in the United States. This situation of clear splits among circuit courts of appeal can only be resolved by the Supreme Court. However, this split has existed for years without once attracting the attention of the Supreme Court, and so it is likely that this discovery tool will not be uniformly available in all US jurisdictions to parties in international arbitration for the foreseeable future.

The author of Section 1782 stated that when the statute was revised in 1964, the hope was that, by allowing foreign tribunals to experience the advantages of US discovery through this tool, they would be inspired to adopt in their own countries broader availability of pretrial discovery; the objective was to lead by example, even if foreign countries did not then offer reciprocal discovery in aid of US litigation.^[32]

In practice, many countries have efficient discovery tools that can be used in aid of international arbitration, even where the seat of the arbitration is located in a foreign country. In France, Article 145 of the Code of Civil Procedure provides the framework for court-ordered experts that can be used prior to any litigation or arbitration proceeding with a view to obtaining relevant documents related to potential claims. Article 145 enables a party to request from the French judge the production of documents that are held by another party and located in France – for use in French or foreign proceedings – without needing to prove ‘urgency’, which would generally be required in most other forms of summary court proceedings.

The procedure is typically started on motion by a party before the French judge, who will consider the types of discovery requested and hear the parties, and may then issue an order for production of documentary and testimonial evidence. The requests must be narrowly tailored but courts regularly accept requests for categories of documents provided such requests are limited in scope and relevant to the underlying claim. The court will appoint an expert who will gather the evidence ordered to be produced, and will make the evidence available to the parties.

Filing an Article 145 motion is subject to two general conditions:

•the requesting party must demonstrate that its request is legitimate (that the evidence sought is necessary to protect its rights or prove its claims); and

• the proceeding must be filed in contemplation of future legal proceedings.

The Article 145 order may also be requested ex parte, but only if the applicant can show that circumstances – such as the need for surprise – require an ex parte solution.^[33] The discovery is then maintained by the bailiff; if no application is made by the seized party within 30 days, the documents will be released to the applicant.^[34]

Note that there is no obligation that the collected evidence be used in later legal proceedings, as it may be that the results of the discovery show there to be no basis for a claim. Nor is there any prohibition on such evidence being used in an arbitration, whether in France or abroad. And importantly, the French Supreme Court has confirmed that this procedure may be used to obtain evidence from non-parties.^[35]

Accordingly, despite all the attention paid to the application of Section 1782 in the United States, and the uncertainty created by the post-Intel split among the circuit courts of appeal, in practice there are discovery solutions in other countries that are often more efficient and less uncertain than the procedures under US law.

The way parties resort to national judicial discovery techniques, however, should be organised and planned with the arbitral tribunal, so that the tribunal is notified and has the possibility to oversee, or potentially to prohibit, recourse to national courts to obtain discovery. As with privilege, this is an issue that can and should be raised with the parties early in the proceedings, at the time of the terms of reference or the initial procedural order. Even if the parties have not contemplated at that early stage of the proceeding that they might use such discovery tools, it is recommended that the tribunal include a reference to this type of discovery request if such requests are not addressed in any of the applicable institutional rules or guidance.

For example, there is no reference to discovery proceedings brought in national courts by the parties in the ICC Rules. However, these rules do provide that:

• ‘the arbitral tribunal and the parties shall make every effort to conduct the arbitration in an expeditious and cost-effective manner’ (Article 22); and
• ‘the arbitral tribunal shall proceed within as short a time a possible to establish the facts of the case by all appropriate means’ (Article 25).

On the strength of this institutional delegation to the tribunal to organise the proceedings, and the duty of the tribunal to reach an enforceable award, the terms of reference should require that recourse by any party to national courts for purposes of discovery must be notified in advance to the tribunal, failing which such evidence may not be used in the arbitration. Indeed, it would give the tribunal even more control of its proceedings if the parties were required not only to notify the tribunal of recourse to national courts, but also that such recourse be approved by the tribunal (though in practice this is rarely done). Although this might seem a severe restriction on the parties’ rights to use available litigation tools, allowing parties free reign to seek discovery from national courts could be a recipe for procedural disaster as parties may be tempted to use such tools as much for purposes of delay – or to impose unnecessary costs on an adversary – as for legitimate fact-finding purposes.

Compliance with GDPR in arbitration discovery

Much has been written since 2018, when the GDPR^[36] came into effect, about the impact of the new rules on arbitration. Of course, data protection regulation is not new, dating back in France, for example, to 1978,^[37] and in Europe to the 1995 Directive.^[38] But as this was a new and evolving area of the law, with little guidance at the start and even fewer sanctions for non-compliance, few in the arbitration world paid much attention and indeed arbitration practitioners suffered no penalties during this time.

The French^[39] and EU^[40] privacy regulators, in 2008 and 2009 respectively, started to pay attention to data protection in international litigation, offering guidance for national court litigants engaged in cross-border discovery on compliance with the European system of data protection, including the use in such litigation of appropriate protective orders. At least one US court (before the GDPR even came into effect) appointed a privacy monitor in the context of international discovery to oversee compliance by the parties with the EU privacy rules, showing judicial respect for the EU rules.^[41]

The GDPR brought new protective requirements and new sanctions relating to the management of personal data within the broad territorial scope of the Regulation.^[42] The Regulation applies to all covered personal data; it is not industry- or sector-dependent. As a result, anyone who manages or processes personal data is potentially subject to the Regulation, including in international litigation (as had been indicated in the 2008 EU guidance) and in international arbitration.^[43]

There is not yet any detailed GDPR guidance specific to international arbitrations. However, it is clear from the detailed landmark analysis by Kathleen Paisley,^[44] and by the guidance offered by the ICCA Working Group in 2020,^[45] that arbitration practitioners must now consider the data privacy consequences of the fact-finding phase of each arbitral proceeding to account for the new data protection rules.

Several institutions have begun to provide guidance to practitioners on how to account for the new rules,^[46] but there remains great uncertainty as to who should be taking the lead on this issue, who will be ultimately responsible should there be any sanctions and what consequences non-compliance might have on the final arbitration award.^[47]

It is inarguable that the parties that were originally involved in data collection and processing for their businesses remain responsible for the proper handling of that data in all circumstances; the mere existence of an arbitral proceeding does not shield them from potential liability. Accordingly, parties should insist from the start of the proceedings (or even in their arbitration clause) with the tribunal and with any organising institution that the arbitration procedure not increase their risk under applicable privacy rules.^[48]

And yet both arbitration institutions and tribunals have their role to play in compliance, though both seem equally unfit, at least for now, to do so. Arbitrators are selected for their many different qualities of analysis, reasoning, management and legal acumen, but rarely for their detailed knowledge of privacy law and the practical management of data protection under the GDPR. Institutions may be able to assemble more expertise and provide guidance in individual matters, but typically do not seek to lead on such matters, particularly when staking out positions of data protection oversight could turn them into co-controllers of the personal data or otherwise create liability under the data protection laws.

It will fall, then, to the tribunal to address the issue on a case-by-case basis, upon consultation with the parties as to the nature and extent of the personal data that they expect will be exchanged and communicated with the tribunal. The responsibility for raising and resolving the data privacy issues also falls to the tribunal because it is a more active participant in setting the rules of disclosure, and in overseeing in the process, than a national court judge would be. Unlike discovery in a US civil matter, where the parties exchange documents and witness testimony among themselves without submitting any of the exchanged information to the court unless there is either a discovery dispute or the information is put into evidence, arbitration panels can be involved from the start in reviewing document requests, defining the allowable scope of discovery, reviewing and narrowing Redfern schedules and even – if deemed necessary by the tribunal for oversight – receiving copies of each side’s documents before individual documents have been selected by the parties for submission as evidence. As a more active participant in the discovery process, there is accordingly a greater risk that the tribunal could be considered a co-controller of the data under the GDPR because it is determining with the parties, ‘the purposes and means of the processing of personal data’.^[49]

As with the other subjects addressed in this chapter, it is recommended that the tribunal raise this issue early on with the parties and the institution, and if the institutional rules do not adequately address the management of the data protection and the allocation of potential liability, then the tribunal should address this in the terms of reference or another organisational document agreed by the parties.

In particular, the tribunal should restate the obligation of each party to:

• comply with the data protection rules that apply to its data;
• raise any concerns with the tribunal about the potential processing of the discovery in the arbitral process so that the rules regarding onward transfer of personal data are complied with;
• consider whether a particular confidentiality or protective order should be adopted by the tribunal;
• consider whether to use standard contractual clauses (SCCs) to ensure protection of personal data in discovery, or if the tribunal should do this;
• consider whether any special forms of data security should be used if the documents are of an especially sensitive nature;
• consider limiting access to certain documents to counsel and experts; and
• consider steps to minimise the personal data contained in the discovery (through pseudonymisation or anonymisation).

Until July 2020, many international personal data transfers, including in arbitration, were made by companies that had registered under the Privacy Shield, a mechanism developed in 2016 by the European Union and the US Department of Commerce after the original Safe Harbor framework was annulled by the ECJ.^[50]. On 16 July 2020, the ECJ ruled that the Privacy Shield framework also violated EU law, owing to what the court considered overly intrusive US intelligence agencies, and accordingly for transfers between the European Union and the United States, or between Switzerland and the United States, other safeguards will need to be considered by a tribunal.

The means of ensuring safeguards under the GDPR are set out in Articles 46 to 47 and are principally SCCs and binding corporate rules. Importantly, the exception identified in Article 49(1)(e) of the GDPR for a transfer ‘necessary for the establishment, exercise or defence of legal claims’, is likely to be insufficient to justify personal data transfers in an arbitration, at least in the first instance, as the European Data Protection Board has provided guidance that the exceptions in Article 49 should only be invoked if the appropriate safeguards identified in Articles 46 to 47 are not available.^[51]

The adoption of SCCs by the participants in an arbitration is relatively simple, and ensures compliance with the GDPR for transfers of data outside the European Union to a jurisdiction that does not provide adequate safeguards for the data, such as the United States (e.g., a tribunal seated in New York with European parties).^[52] SCCs may be put in place for data transfers from the foreign data controller (e.g., a party) to another data controller (e.g., US counsel) or a data processor (e.g., a US-based e-discovery vendor). It should be noted that SCCs are regularly used in US litigation to ensure compliance with the GDPR, and that many US courts in litigation are open to accommodating foreign data privacy concerns as recommended notably by the Sedona Conference,^[53] though certain courts have refused to accept the GDPR as an excuse for non-compliance with the ordered discovery.^[54]

In light of the Privacy Shield being invalidated in 2020, increased use by parties and tribunals of SCCs is to be expected. However, the acceptance of this practice in any given arbitration by the tribunal and all the parties will need to be accepted at the start of the proceedings, and accordingly the tribunal must ensure at an early stage that the data protection issue is raised, that potential risks are identified and that appropriate safeguards are adopted.

Conclusion

The arbitration of IP disputes can be discovery-intensive, though paradoxically there is a dearth of guidance for practitioners on how to proceed in these cases. The procedures adopted in arbitration for fact-finding will reflect the parties, their counsel, the tribunal composition and their combined legal cultures. However, there is no set of hard and fast rules, nor even concrete guidance that will provide an answer for every situation.

Arbitration participants should expect, however, that the guidance and precedents that do exist will inform the fact-finding process that will be adopted by the tribunal, with a view to being both efficient and fair. The objective early in the arbitration should be to discuss fully with the parties the tribunal’s plan for the fact-finding process and to tailor the rules to their particular matter to manage with transparency any risk and uncertainty that may be identified. Obtaining parties’ consent to the arbitration procedure in general, and to the fact-finding process in particular, will go far in avoiding later disputes and immunising the award against annulment. But while the parties and arbitral institutions have important roles to play, this mission is one for which the tribunal must take primary responsibility.

Notes