A recent issue of the Japan Commercial Arbitration Association (JCAA) Newsletter is largely devoted to these topics (No 22, July 2009). Sydney Law School and ANJeL are privileged to host not only one of Japan’s doyens in ICA (and other cross-border dispute resolution, especially WTO procedures), Professor Yasuhei Taniguchi (over July-August 2009). We also welcome (over September – March 2010) Kokushikan University Professor Tatsuya Nakamura, a leader of Japan’s ‘new generation’ of arbitration specialists who heads JCAA’s Arbitration Department.
They have already got me thinking further about Arb-Med (arbitrators encouraging parties to settle their dispute), in the context also of interesting new JCAA Rules focused more specifically on Mediation. Both developments are important for Australia, presently reviewing its legislative and institutional framework for international commercial arbitration (ICA), as well as for many other Asia-Pacific countries intensely interested nowadays in efficient mechanisms to resolve cross-border disputes.
]]>
In the Newsletter (pp 10-12), Professor Nakamura presents an intriguing preliminary study of Arb-Med, which parties agreed to try in 48 (40%) out of 121 JCAA arbitrations that proceeded over 1999-2008. About half (25 or 52%) of these attempts resulted in a settlement agreement (with 18 then recorded as a ‘consent award’), and involved terms proposed by the arbitrator(s) (28 or 58%). In most cases (37 or 77%), the arbitrators attempted settlement after all evidence had been presented, suggesting that they adopted an ‘evaluative’ rather than ‘facilitative’ style to mediation. Attempts were particularly common among arbitrators with civil law rather than common law backgrounds, with the former almost being mostly Japanese. Professor Nakamura suggests that the background of the parties themselves did not seem so important, however, because in the 7 out of 48 cases (15%) where at least one party was from a common law background, those parties did not oppose the arbitrators’ attempt to settlement.
I do have some doubts first about that last generalisation, as the parties in those 7 cases may be unrepresentative. Specifically, we should examine the background of the 73 cases where Arb-Med was not even attempted. If there is a considerable number of cases where parties are from a common law background, it could be that arbitrators did not suggest Arb-Med because they felt (rightly or wrongly) that they might be rebuffed. So we could infer that party background can be significant after all.
Secondly, I would be interested to know more about the background of the parties’ legal advisors (especially outside counsel). For example, in the 7 cases where the parties didn’t oppose Arb-Med overtures, did the common law background parties engage civil law (especially Japanese) lawyers? I would expect English lawyers to be the most reticent about Arb-Med. Even more so than American lawyers, who not only have longer familiarity with ADR even in connection with court proceedings (since the 1980s), but also are trained in a less formalist variant of the common law (reflected also in less objection, for example, to the lex mercatoria).
Thirdly, have Arb-Med attempts have been increasing over this ten-year period? Professor Taniguchi, in private conversation, has a sense that instead they may have been decreasing. Indeed, Professor Nakamura himself downplayed the tendency for Arb-Med to occur in JCAA arbitrations when writing in an article published in 2001 (18 JoIA p641). This was partly in justifiable reaction to some continued critiques about various allegedly distinctive features of JCAA arbitrations (including by one of my former USydney students: Thirgood, ibid p177). Leading Japanese arbitrators like Professor Taniguchi had also publically questioned whether Arb-Med was appropriate for international arbitrations in Japan, at least at this stage of its development, despite renewed advocacy for this hybrid procedure advanced by other leaders such as Professor Toshio Sawada (see my 2004 article, with further references). But there is growing evidence that costs - even more so than delays, perhaps - are re-emerging as a major problem in ICA world-wide. And so the winds may be shifting in Japan too – towards more positive assessments and usage of measures like Arb-Med that promise greater efficiencies in cross-border dispute resolution.
Fourthly, however, it would be helpful to know more about a key issue in current debates world-wide about Arb-Med: whether or how the arbitrators might meet separately which each party (‘caucusing’). As I have previously mentioned, experts from European civil law jurisdictions (let alone the Anglo-Australian variant of the common law tradition) generally remain skeptical about this, due to concerns about natural justice and (apparent or actual) bias tainting the arbitrators. By contrast, Professor Nakamura tells me that caucusing is very widely used in JCAA Arb-Med. (Similarly, this tends to occur in civil litigation before Japanese judges – unlike for example German judges, who actively encourage settlement but in open session.)
I don’t have so much problem with caucusing when combined with a more facilitative approach to mediation, eg rephrasing and relaying one party’s argument to the other where parties have become too emotional or otherwise incapable of understanding what the other side is saying or proposing. I am also open to caucusing when, for instance, the parties have basically resolved major issues and only want Arb-Med to settle some remaining minor points. But I do have reservations when major issues and amounts are at stake. There can be a temptation for the arbitrator to indicate privately to one party that the other will settle within a particular range (even if the latter hasn’t really made that clear), and vice versa, in achieve a settlement. At least, a suspicion that this has occurred can arise, and it cannot easily be tested afterwards because of confidentiality obligations during the Arb-Med process.
For such reasons and others, like CEDR I still prefer ‘no caucusing’ as the starting principle. And perhaps, if parties do get invited to caucus, they should be required to get separate professional legal (and practical) advice about such additional risks, as well as its possible benefits. They might then decide, for example, to allow caucusing but only in a more ‘facilitative’ style by the arbitrators, requiring the latter to try more ‘evaluative’ Arb-Med (eg proposing specific settlement terms or ranges) only in open session. Another possibility might be to allow one arbitrator to caucus in a more evaluative style, but retain a panel of three arbitrators to issue an award if settlement did not eventuate.
Interestingly, a commentary in the Newsletter by Mark Goodrich and Christopher Hunt regarding JCAA’s ‘New International Arbitration Commercial Mediation Rules’ quotes Professor Nakamura as believing ‘that lawyers be ideal candidates to be evaluative mediators but that more focus should be placed on training facilitative mediators, as the skills required for that approach are more subtle’ (p10). On the other hand, they report that the first mediation under these Rules (in mid-March) involved an evaluative approach, with ‘the two days of the mediation meeting … filled by oral submissions and questions from the panel of three mediators’ (p7). This suggests that little or no caucusing took place, or at least that it could be a realistic option for parties to cross-border disputes even in ‘pure’ mediation settings.
In other words, these new Mediation Rules could be used either as a more facilitative manner, leaving subsequent arbitrations to allow the possibility of more evaluative Arb-Med (with or even without caucusing). Or instead the new Rules could be used to have an evaluative-style mediation, possibly not even involving caucusing – although Rule 9.5 creates a default rule allowing it (subject to any contrary party agreement, however, as mediation naturally is rooted in consent).
Either way, however, it is important to note that Rule 8 allows a mediator subsequently to sit as an arbitrator simply ‘if the parties so agree’. Goodrich and Hunt argue that this differs from mediation rules for the ICC and SCC (Stockholm Chamber of Commerce), ‘where the starting assumption is that this should not happen’ (p8). They also contrast HKIAC Mediation Rule 14, prohibiting the mediator from later becoming the arbitrator even if parties so agree. (Presumably, if parties wish this they should instead commence arbitration, then allow Arb-Med but subject to the provisions of the Hong Kong Arbitration Ordinance – although in practice those Arb-Med provisions apparently are hardly ever used.)
I would also add that JCAA Mediation Rule 8 seems to allow even oral agreement. But if Japan’s Arbitration Act 2003 applies then agreement triggering the subsequent arbitration does have to be in writing (Japan has not adopted the revised UNCITRAL ICA Model Law variant that permits purely oral agreements) and arbitrator selection is anyway usually done by writing. By contrast, if the arbitration commences first, the Act requires any Arb-Med to be recorded in writing (unless – perhaps even orally! – the parties dispense with that writing requirement).
An even more important feature rightly highlighted by Goodrich and Hunt is JCAA Mediation Rule 11: if parties agree to settle, they can appoint the mediator as arbitrator to issue a consent award. They question whether the 1958 New York Convention regime will enforce ‘the somewhat artificial “award” as in this case’ (p8). However, they note that SCC Mediation Rule 11 has a similar provision. But I would point out that the latter’s effectiveness has been criticised too (Newmark and Hill, ‘Can A Mediated Settlement Become An Enforceable Arbitration Award?’ (2000) 16 Arb Int’l 81). For this reason too, it may be better to begin with an arbitration, then allow (in writing) Arb-Med under specified conditions (such as ‘no caucusing’). Indeed, parties could require the arbitrator to attempt settlement well before the evidentiary phase. This would also maximize the potential for time and cost savings, a major attraction of the new JCAA Mediation Rules.
That alternative would also avoid a third interesting feature: Rule 7.3 requires the mediator to ‘be independent in principle’ but remain always ‘impartial’ (although Rules 7.4 and 7.7 refer to ‘independence or impartiality’ in connection with the mediator’s disclosure requirements). By contrast, JCAA Arbitration Rule 28.1 demands both independence and impartiality of arbitrators. (Further, Article 5.4 of the 2002 UNCITRAL Model Law on International Commercial Conciliation encourages appointment of mediators who are both independent and impartial.)
A final caveat highlighted by Goodrich and Hunt (p8) is that the confidentiality obligation in Mediation Rule 12.3 does not expressly cover all information disclosed in the mediation, and so may not extend to all documents produced during those proceedings. I would contrast Arbitration Rule 40.2, which refers to all ‘facts’ relating to the arbitration). That formulation could be another reason to chose arbitration but with additional consent to Arb-Med (pursuant to Arbitration Rule 47, and as allowed by Article 38.4 of the Japanese Act).
However, the alternative is to use the new JCAA Mediation Rules, trying to circumvent these issues as much as possible through further party agreement varying relevant rules or clarifying more precisely what is expected (and allowed) regarding the mediator. The Rules will also be useful for other arbitral institutions interested in updating their own rules on international mediation. For example, ACICA has an official ‘panel of mediators’ – albeit currently with only one person listed – but its mediation rules are still based quite closely on the 1980 UNCITRAL Conciliation Rules. JCAA’s new Rules are also very welcome, especially when considered alongside the innovative study on Arb-Med by Professor Nakamura, for the growing number of other dispute resolution organisations and experts world-wide interested in developing best practice for arbitrators facilitating settlement.
]]>
