A recent decision from the Paris Court of Appeal raises interesting questions about the role of the adverse inferences doctrine in international arbitration.[1]

Although the Court was clear in stating that the majority of arbitrators on the International Chamber of Commerce (“ICC”) panel based their decision on the exhibits in front of them, the decision has potentially widened the scope and applicability of the adverse inferences doctrine.

BACKGROUND

In 2011 Dresser-Rand Group Inc., a Delaware company, agreed to purchase all shares in Grupo Guascor, a Spanish company, from a group of Spanish companies and shareholders (“Original Shareholders”). The share purchase agreement provided for price adjustment upon completion of the transaction. In 2012, Dresser-Rand Group Inc. transferred its shares under the agreement to Dresser-Rand Holdings Spain SLU. The Original Shareholders and Dresser-Rand Group Inc were not able to agree on the price adjustment and in 2012 the Original Shareholders requested arbitration from the ICC against Dresser-Rand Group Inc. and Dresser-Rand Holdings Spain SLU (“Dresser-Rand Companies”).

In 2015, a majority of the arbitration tribunal found partially in favour of the Original Shareholders. As part of its reasoning the majority drew an adverse inference against the Dresser-Rand Companies under Article 9(5) of the IBA Rules due to the their failure to produce certain audit reports requested by the Original Shareholders. Notably, the tribunal did not order the production of these documents and did not ask the Dresser-Rand Companies for submissions explaining their absence. The majority concluded that these documents would have been prejudicial to the Dresser-Rand Companies’ interests; however, the central basis for the reasoning behind their award came from exhibits that the arbitrators actually reviewed.

THE COURT OF APPEAL’S DECISION

The Dresser-Rand Companies challenged part of the arbitral award at the Paris Court of Appeal. They did so on the basis that: (1) the arbitrators inappropriately relied on the IBA Rules  (the “IBA Rules”) without consulting the parties beforehand, and (2) the arbitrators violated various principles of fairness by failing to invite submissions from the Dresser-Rand Companies explaining the non-production of the documents and because the Original Shareholders had not specifically asked the arbitrators to draw an adverse inference.

The Court rejected these arguments, noting that the IBA Rules were explicitly referred to in a Procedural Order in the matter.

Article 9(5) of the IBA Rules provides as follows:

If a Party fails without satisfactory explanation to produce any Document requested in a Request to Produce to which it has not objected in due time or fails to produce any Document ordered to be produced by the Arbitral Tribunal, the Arbitral Tribunal may infer that such document would be adverse to the interests of that Party.

The Court noted that the Order was a direct result of negotiation between the parties, and the parties were aware that the IBA Rules could be applied.

Additionally, the Court held that the majority of arbitrators had predominately decided the matter based on the audited statements in front of them and the conclusions they drew from those documents. Any adverse inference drawn seemed to have been a secondary conclusion that supported the ultimate decision.

COMMENT

This decision raises a number of important points. First, it indicates that once parties enter into agreements or orders involving the IBA Rules, arbitrators are not required to give further notice that they may be applying the Rules and related principles.

Second, arbitrators may not need to seek further submissions from parties about a failure to produce documents, even if such failure to produce relates to requests that did not come from the arbitrators themselves.  Because the parties had accepted that the arbitrators could apply the IBA Rules, it was open to the arbitral tribunal to apply the adverse inferences doctrine without inviting further submissions on the matter (and without any of the parties expressly requesting that the tribunal draw such an inference).

Third, the decision seems to demonstrate judicial deference to arbitrators’ decisions on their own process.

Finally, the decision appears to emphasize the importance that arbitrators and courts may place on the parties’ choice to be governed by the IBA Rules. The Paris Court of Appeal placed significant weight on the parties’ negotiated choice to use the Rules. This choice is, in part, what allowed the Court to determine that the arbitrators did not need to inform the parties that they were considering drawing an adverse inference without hearing submissions on that point.

Care should be taken, however, not to overstate the decision’s scope. As noted, the Court of Appeal was clear that the majority of arbitrators decided the matter based on exhibits they actually reviewed. The arbitrators did not, according to the Court, refer to the Dresser-Rand Companies’ deficient production in a way that made any adverse inference central to their decision.  This raises the question of how far arbitrators can go in relying on the adverse inferences doctrine and whether an adverse inference could serve as the main (or sole) basis for an arbitral award.

Further, in a case where a party’s consent to or understanding of the applicability of the IBA Rules (or similar rules) was less clear, it is possible that arbitrators might not be able to apply the doctrine in the same fashion.

That said, the Court did not appear to take issue with the majority of arbitrators making or applying an adverse inference in the manner that they did.  This leaves the door open for arbitrators to make use of the doctrine, without notice to or comment from any of the parties, in future disputes.

[1] CA Paris, Pôle 1 – Chambre 1, 28 February 2017, 15/06036.