Rafal Morek’s post last month, Investor-state disputes: how arbitration and mediation can intertwine to provide more resonant solutions, emphasized the increasing use of mediation to resolve investor-state disputes, albeit still confined to a small number of cases under the International Centre for Settlement of Investment Disputes (ICSID) Convention.

As the late Professor Derek Roebuck noted in The Myth of Modern Mediation in 2007:

“Everywhere in the Ancient Greek world, including Ptolemaic Egypt, arbitration was normal and in arbitration the mediation element was primary.”

Switching hats

So why is it still uncommon, after so many centuries, for arbitrators and mediators to “switch hats” in order to achieve the outcome best suited to the disputants’ needs in the most efficient and cost-effective manner?

Apart from relative speed and economy, hybrid processes such as Med-Arb and Arb-Med give the parties certainty that, either by agreement or by award, their dispute will be resolved. The parties are at liberty to put a time limit on that in their Med-Arb or Arb-Med agreement. If they use only mediation, they run the risk of not settling all the issues in dispute. If they use only arbitration, they know that all the issues will be resolved but they deprive themselves of the creative options their own negotiated settlement agreement might provide.

In a 2002 article in the Journal of Applied Psychology, Putting the cart before the horse: the benefits of arbitrating before mediating,   the authors examined the impact of Med-Arb and Arb-Med on various dispute outcomes involving three disputant structures – individual v. individual, individual v. team, and team v. team. The authors found that disputants in the Arb-Med procedure settled in the mediation phase more frequently and achieved settlements of higher joint benefit than did disputants in the Med-Arb procedure. They concluded that Arb-Med may have broader applicability than originally imagined.


The Arb-Med process involves the arbitrator conducting a hearing and producing an award which is concealed from the disputants while the arbitrator mediates. If the dispute is not settled in the mediation, the award is revealed and is (usually) binding on the parties. However, a common criticism of the Arb-Med process is that, if the dispute is settled in the mediation phase, the time and cost of the arbitration will have been wasted. Another criticism is that where suggestions by the mediator in the mediation phase are taken as hints as to the content of the already sealed arbitral award, the parties will be inappropriately coerced into settlement. However, Arb-Med does have the advantage that it avoids the criticisms of the Med-Arb process mentioned below.


The Med-Arb process, as its name implies, involves the mediator seeking to help the parties resolve their dispute and, if unsuccessful, proceeding to arbitrate. Common criticisms of this process include:

* the disputants may be reluctant to disclose in confidence to the mediator information which could be used against them in the arbitral award, thereby thwarting the mediator’s ability to identify common interests that could potentially lead to creative solutions;

* because the parties know that the mediator will arbitrate if the mediation does not produce agreement, they may not participate actively in the mediation, preferring to fast-forward to an arbitral award; and

* suggestions by the mediator in private sessions with a disputant could be interpreted as implied threats to make an arbitral decision adverse to the disputant unless they are accepted;

* allowing an arbitrator to be privy to private representations made during the mediation phase creates an appearance of bias and may actually bias the arbitrator when determining the dispute; and

* procedural fairness requires that arguments be made in the presence of the opposing party and be subject to rebuttal. In Med-Arb, the mediator-turned-arbitrator is usually bound to keep strictly confidential all private disclosures made in the mediation phase.

Procedural fairness

One approach to resolve these procedural fairness issues has been for  legislation in Australia  to require the written consent of the parties to the arbitrator acting as mediator, (including a conciliator or other non-arbitral intermediary) and their further written consent, given after the termination of the mediation, to the arbitrator resuming the arbitration, in which case the arbitrator must disclose to the parties any confidential information obtained during the mediation that the arbitrator considers material to the arbitration.

In Ku-ring-gai Council v Ichor Constructions Pty Ltd [2018] NSWSC 610 (8 May 2018),   after arbitrating between the parties for twelve days, the arbitrator put forward to the parties’ lawyers in a breakout room during the lunch break a proposal for settlement “under the cloak of mediation”, having obtained their written consent to mediate. The arbitrator said he was very busy and it would take him months to decide the case. His proposal was that each party drop its claim, walk away and ‘stomach its cost’. The parties did not accept the proposal and the arbitration resumed after the lunch break without the parties’ written consent. The court was then called upon to decide whether what happened in the breakout room amounted to mediation and, if so, whether the arbitrator, having acted as mediator, could lawfully resume acting as arbitrator. The court found that the proposal did amount to mediation, so the only participant to benefit from the mediation was the arbitrator, who was relieved of the burden of having to continue to arbitrate.

Where does this leave us?

If we are truly committed to party-centric processes, it is time for the mediation and arbitration communities to get together and have a constructive conversation.


To make sure you do not miss out on regular updates from the Kluwer Mediation Blog, please subscribe here.

Kluwer Arbitration
This page as PDF

One comment

  1. The Global Mediation Exchange Center, GMXC, has a solution now that recognizes the strong need for an enforceable mediation solution for cross-border disputes and reduces or eliminates many of the concerns expressed here about mixed mode processes.

    GMXC is authorized by a specific New Jersey Statute, N.J.S.A. 2A:23E-1, the New Jersey International Arbitration, Mediation and Conciliation Act. If the parties sign the submission agreement, the process begins in Arbitration, the parties agree to proceed to Mediation with the arbitrator changing hats to become the mediator. If a mediated settlement is achieved, the mediator becomes an arbitrator and enters a Consent Award enforceable under the New York Convention in over 160 countries. If a mediated settlement cannot be achieved, the matter proceeds as an arbitration before a different administrator that the parties’ selected in the submission agreement. This process avoids most of the potential pitfalls detailed in this article. Instead, it provides the opportunity for an expeditious and highly cost-effective solution for international disputes.

Leave a Reply

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