Source: International Dispute Resolution News (ABA, Nov. 13, 2020). View PDF
In August 2019, the United States and over forty other nations signed the United Nations Convention on International Settlement Agreements Resulting from Mediation (known as the "Singapore Convention").[1] The Singapore Convention requires each signatory to "enforce a settlement agreement (which is defined as one resulting from mediation) in accordance with its rules of procedure and under the conditions laid down in [the] Convention."
In the months following the United Nations General Assembly's adoption of the Singapore Convention,[2]and through the present, much has been written concerning the new convention. Many commentators have characterized the Singapore Convention as the “mediation equivalent” to the Convention on the Recognition and Enforcement of Foreign Arbitral Awards (1958, known as the “New York Convention”).[3]
The announcement of the Singapore Convention has been met generally with a lack of critical analysis. Specifically, little if any attention by dispute resolution ("DR") specialists has been directed to the following related questions: (a) whether there is a genuine need for an international convention that governs the enforcement of settlement agreements arrived at through mediation (such an agreement is referred to herein as a "Mediated Agreement"), and (b) whether some aspects of the Singapore Convention actually do more harm than good.[4]
I address these issues below, and my conclusions are that the Singapore Convention is (a) superfluous and therefore largely a waste of time and effort, and (b) likely, in many cases, to be counterproductive.
In the course of over 27 years of practicing international DR, this author has been involved (as counsel, not as a mediator) in dozens of mediations, almost all of which have had an international element. I have seen good mediators resolve disputes involving (inter alia) (a) real property located in a foreign country, (b) shares of a foreign corporation, (c) intellectual property rights registered in a foreign country, and/or (d) bank accounts located in a foreign country.
In many disputes involving parties from more than one country (or property located/registered in a country other than the home country of all the disputants), issues of foreign law and foreign procedure arise. For a mediator in such a case, the foreign law issues make the case more complicated and challenging than in a purely domestic dispute. In an international case, in order to address the complexities caused by foreign law/procedure, a good mediator avails himself of three primary tools, described below (these tools are also used in domestic mediations, but their importance is enhanced in the international context):
Each Party shall do and perform, or cause to be done and performed, all such further acts and things, and shall execute and deliver all such other agreements, certificates, instruments, and documents that the other party may reasonably request in order to carry out the intent and accomplish the purposes of this Agreement and the consummation of the transactions contemplated hereby.
It is difficult to imagine a good reason for any settlement agreement not to include a “further documents” clause – a fortiori, it is difficult to imagine a good reason for any Mediated Agreement in the international context not to include such a clause.
I do not believe that my experience is unique – I believe that most experienced international DR practitioners have seen mediators use an escrow clause, an arbitration clause, and a "further documents" clause. These three clauses are collectively referred to herein as the "Three Clauses."
The use of these Three Clauses serves as a "check" on the good faith of the disputants to bring their dispute to a final and complete resolution. The inclusion of the Three Clauses in a Mediated Agreement ensures that the parties' good faith is “proven” not merely through the execution of such an agreement but through its implementation.
Put slightly differently, a good mediator understands that, in many cases, (a) the very reason that the disputants got to where they are – namely, around the mediator's conference room table – is the lack of trust between the two sides, and (b) because of that lack of trust, the full implementation of the terms of the Mediated Agreement cannot be taken for granted or left to wishful thinking.
Here is an illustration of how the Three Clauses work together to ensure compliance with a Mediated Agreement:
However, the Transferring Disputant presumably remembers that he covenanted to execute any "further documents reasonably requested" to carry out the purposes of the Mediated Agreement. That disputant also (presumably) realizes that, if he were to resist executing such further documentation, his conduct would constitute a breach of the Mediated Agreement. Therefore, when the Transferring Disputant evaluates whether he should execute the requested documents, it is likely that he understands that, if he resists, (i) he will soon become a defendant in an arbitration, and (ii) the arbitrator will be the same person who already is familiar with the dispute
It is also reasonable to assume that the Transferring Disputant understands that, if there is any arbitrator who is unlikely to need much time to resolve a dispute concerning the "further documents" clause of the Mediated Agreement, that arbitrator is the mediator who facilitated execution of that agreement in the first place.[7]
Because the enforcement of an arbitral award is a relatively straightforward procedure in the more than 150 countries that are parties to the New York Convention, the Transferring Disputant knows that, if he were to drag his feet, it would be just a matter of time before (a) an arbitral award is rendered, requiring him to transfer the property, and (b) a legal action is commenced in a court in his home country (or in the country where the property is located/registered) to enforce that arbitral award against him. Because the Transferring Disputant knows that he will likely lose the legal fight and that the battle will be relatively short, he is likely to sign the requisite further documentation without such a fight.
In summary, under circumstances like those described above, the Transferring Disputant has a great incentive to comply with his obligation under the Mediated Agreement and to execute the requisite additional documents, without delay.
More generally, the same
party considers the pros and cons of possible foot-dragging (such a disputant is sometimes referred to herein as a "Recalcitrant Disputant"), it knows that there could soon be an arbitral award rendered against it and that such award can be enforced in court reasonably promptly.
In summary, when dealing with rational disputants, the inclusion of the Three Clauses in a Mediated Agreement eliminates, in virtually all cases, any concern that the agreement might not be implemented in the relevant jurisdiction(s).
With this lay of the land, we can proceed to evaluate the need (if any) for the Singapore Convention.
In addressing the question of whether an international convention is needed to improve the
As explained below, in the view of this author, the answer to the first question is that generally there should not be a significant need, and the answer to the second question is clearly no.
As indicated above, many of those who advocated for the Singapore Convention see it as the "mediation equivalent” to the New York Convention.[8]Yet few commentators have addressed the precise issue of whether the potential availability of relief under the New York Convention itself alleviates the need for any international convention governing the enforcement of Mediated Agreements.
The heretofore superficial treatment regarding the "need" for the Singapore Convention is perhaps best epitomized in a piece from the European Journal of International Law (the "EJIL"), as follows:
[O]nce a mediated agreement is reached, there is no comprehensive legal framework for the enforcement of international settlement agreements. The result is that parties are forced to attempt to enforce such agreements in domestic courts, typically as an ordinary breach of contract claim.
As a result, when a party to a mediated settlement agreement reneges on its obligations or otherwise refuses to uphold [its] terms. . . , the other party has had to commence separate proceedings in court or through arbitration to enforce the agreement. This has essentially meant initiating a new dispute after resolving the underlying one, adding increased costs and delay.[9]
Based on this practitioner's experience, the EJIL’s articulation of the "problem" is greatly exaggerated and misfocused – for at least two reason: (a) there is no reason to believe that a breach of Mediated Agreements is anything other than atypical, and (b) the solution to the "problem" is obvious – namely, including an arbitration clause in the agreement.
The EJIL’s rationale gives the unverified (to say the least) impression that Mediated Agreements are routinely breached in a manner that leaves the aggrieved party with nothing more than a breach of contract claim to assert in court. As a matter of empirical fact, this assertion is questionable.
One European law firm has gone on record stating that breaches of Mediated Agreements are "rare.”[10]If that is correct, then the paucity of breaches – in and of itself – raises the serious question as to whether there is a need for any international convention to "rectify the problem.”
Yet if in fact Mediated Agreements are routinely breached in a manner that leaves one party with no alternative but to commence litigation, then the uncomfortable conclusion for mediators is that they
are (apparently routinely) doing only half a job.[11](A corollary of that conclusion is that it is questionable whether there would or should be an international consensus on the need to enforce agreements that are characterized by inadequate draftsmanship or insufficient detail.)
There is not even a hint in the negotiative history of the Singapore Convention that its draftsmen considered whether the use of an escrow mechanism to carry out the terms of a settlement agreement would decrease the likelihood of breach. As indicated above, the use of an escrow arrangement is intended (inter alia) to minimize the chances that one party to a settlement agreement will perform to its detriment without reciprocal performance by the other party.
A party to a Mediated Agreement will not be able to "enforce" that agreement (whether domestically or internationally, whether under the Singapore Convention or otherwise) unless that party has fully performed. Because the purpose of an escrow arrangement is (inter alia) to reduce the chances that one party might fully perform without there being reciprocal performance, the use of escrow reduces the likelihood that any party to a Mediated Agreement might be placed in a situation in which it has no alternative but to sue to enforce that agreement.
It goes without saying that, the less the incidence of breach of Mediated Agreements, the less the need for any international legal regime to enforce those agreements.
The negotiative history of the Singapore Convention indicates that the "need" for such a regime was simply assumed.
Although the Working Group was aware that Mediated Agreements sometimes include arbitration clauses,[12]it failed to acknowledge the significance – both practical and legal – of the arbitration option.
The common feature of mediation and arbitration is the direct involvement of a "neutral" whose job requires (inter alia) that he familiarize himself with the parties' dispute. In most successful mediations, the mediator plays a major role in formulating the Mediated Agreement. Because of that role, the mediator's knowledge as to the intention of the parties concerning their Mediated Agreement is usually second to none.
Perhaps disputants do not (as a matter of course) give thought prior to the mediation as to the best mechanism for enforcing the Mediated Agreement that they hope to sign. Nonetheless, during the mediation, at the stage where the disputants are engaged with the mediator in crafting such an agreement, the disputants cannot help but realize that the mediator is intimately aware of the mutual understandings that form the basis of their agreement.
When disputants give thought to the most efficient legal mechanism for enforcing their Mediated Agreement (and if the disputants do not devote thought to this issue, then their lawyers should), it should be apparent that the mediator's level of knowledge of the matter is significant. Considering the fact that, in most breach of contract litigation, few issues are more important than determining the intention of the parties, the
In the event of such a dispute, in almost all cases, the amount of time that it would take the
All of these factors suggest that the
And regardless of the identity of the arbitrator, any ensuing arbitral award could (of course) then be enforced in court, in approximately 150 nations, under the international convention that already exists – the New York Convention.
Therefore, in all but the most exceptional cases, the best solution to the "lack" of an international enforcement regime for Mediated Agreements is to include therein an arbitration clause.
This author recommends that, in a Mediated Agreement, the default option for an arbitration clause should be to designate the (former) mediator to serve as arbitrator. Nonetheless, the identity of the arbitrator is secondary to the issue of the redundancy of any international treaty to enforce Mediated Agreements. The advantages of including an arbitration clause in a Mediated Agreement (irrespective of the identity of that arbitrator) in and of themselves show that the Singapore Convention is superfluous.
In light of the widespread belief that the Singapore Convention would be the “mediation equivalent” to the New York Convention, the natural question is: Did those who proposed an international legal regime for the enforcement of Mediated Agreements consider the possibility that, by simply incorporating an arbitration clause in a Mediated Agreement, adequate enforcement relief would already be available for Mediated Agreements
The use of an arbitration clause in a Mediated Agreement was discussed by the draftsmen of the Singapore Convention:
It was also stated that the existence of a dispute resolution clause in the settlement agreement should not be a ground for resisting enforcement in the instrument, as there were existing mechanisms to address those issues. For example, it was mentioned that if there was an arbitration clause in the settlement agreement, the enforcing authority would generally refer the parties to arbitration in accordance with article II(3) of the New York Convention.[15]
The Working Group was aware of the fact that there are disputants that include arbitration clauses in their settlement agreements, yet there is no indication in any of the Working Group Reports that the draftsmen considered whether the possibility of increasing the use of arbitration clauses in Mediated Agreements could render the (draft) Singapore Convention superfluous.
Such an increase could come about simply by having leading DR practitioners and organizations call for such an increase.
The blog of one large law firm indicates that, in deciding on the need for a convention to govern the field of Mediated Agreements, the use of an arbitration clause in such agreements was in fact considered but was deemed "inadequate":
Inspired by the successful New York Convention of 1958, under which many arbitration awards are directly enforceable [internationally], the [Singapore] Convention would ensure thecross-border enforceability of international settlement agreements arising from mediation, through an international framework. Currently, mediated settlement agreements are not enforceable unless the mediation is part of a pending arbitration and they are converted into an arbitral award . . . .
This lack of enforceability carries risks for both parties and mediators. A party wishing to enforce a mediated dispute settlement against a reluctant opposing party is forced to engage (again) in proceedings to obtain a court judgment in a foreign jurisdiction – an often lengthy, difficult and costly endeavour.
Mediators, on the other hand, may refuse to serve as sole arbitrators simply to convert a settlement agreement into an arbitral award, since that service is often not covered by their professional insurance.[16]
This author has never encountered a mediator who refuses to serve as an arbitrator in connection with a claim for breach of a Mediated Agreement. Nonetheless, for the sake of argument, let's assume that some mediators refuse to so serve because of a lack of professional insurance.[17]Is this phenomenon a sufficient reason for dozens of nations to sign and then ratify (presumably after going through the process
of enacting implementing legislation) a treaty on the enforceability of Mediated Agreements?
It seems that there are other, simpler ways of addressing the problem (if any) of insurance coverage for a mediator who might become an arbitrator. Here are two suggestions:
In summary, the concern as to a lack of professional insurance coverage appears to be an entirely insufficient reason to embark on the execution and (global) ratification of a treaty governing Mediated Agreements.
The "lack of insurance" argument appears to be artificial.
In at least three respects, the Singapore Convention risks harming the field of the international enforcement of Mediated Agreements. As explained below, two of those risks concern the convention's defense (to enforcement) of "mediator breach" – (a) the Convention overlooks the importance of good faith reliance in instructing national courts as to the enforceability of Mediated Agreements when the defense of "mediator breach" is raised; and (b) the Convention permits challenges on the grounds of "mediator breach" in very inconvenient fora. The third risk stems from the Convention's exception for "employment law," which is drafted in a manner likely to encompass disputes other than those between an employer and an employee.
The Singapore Convention provides, in Article 5, section 1(e), that enforcement of a Mediated Agreement may be refused upon proof that there was "a serious breach by the mediator of standards applicable to the mediator or the mediation without which breach that party would not have entered into the settlement agreement.”[18](Such clause is referred to herein as the "Mediator Breach Provision").[19]
To some it might seem that the appropriateness of the Mediator Breach Provision is
Defenders of the Singapore Convention would likely point out that (a) Article 5, section 1(e) states expressly that the disputant who opposes enforcement on the grounds of mediator breach is required to prove that, absent such a breach, the disputant would not have entered into the Mediated Agreement, and (b) therefore, a mediator's breach of standards is a necessary but insufficient grounds for refusing to enforce a Mediated Agreement.
However, the mere fact that the disputant who invokes the mediator breach defense would need to prove "but for" causation does not overcome the primary substantive flaw inherent in that defense – namely, the failure of the Singapore Convention to instruct the court to take into consideration good faith reliance by the other disputant.
The significance of good faith reliance should be obvious in light of the fact that settlement agreements differ from other agreements (and, for that matter, from court judgments or arbitral awards) in that settlement agreements very often provide for staged performance.[20] Consider the following scenario:
Assume further that, at the enforcement stage in court, Disputant B raises, pursuant to the Mediator Breach Provision, the defense of mediator breach. When that issue is litigated in court, to what extent will the good faith reliance by Disputant A be a factor in the court's decision as to whether to exercise discretion to refuse enforcement?
The Singapore Convention is silent on the issue of good faith reliance. This is surprising – not least because the significance of good faith reliance in international commercial law is universally recognized. Take for example the United Nations Convention on Contracts for the International Sale of Goods (the "CISG").[21] The CISG is a multilateral convention that establishes a uniform framework for contracts relating to the international sale of goods. The CISG has been ratified by more than ninety countries. Article 16 of the CISG governs generally the acceptance of an offer, and it provides that an offer may not be revoked "if it was reasonable for the offeree to rely on the offer as being irrevocable and the offeree has acted in reliance on the offer."[22]
Article 16 of the CISG demonstrates that the significance of good faith reliance is unquestionably accepted as a part of the law governing international commerce.
Yet the Singapore Convention is completely silent on the issue of good faith reliance.[23]
Let's return to our example above. To the extent that the Mediator Breach Provision would be interpreted as requiring a court to disregard good faith reliance by the
The Singapore Convention's silence as to good faith reliance should cause concern among DR practitioners that the Mediator Breach Provision might be interpreted in a manner that precludes a court from considering the issue of reliance.
Defenders of the Singapore Convention might counter by pointing to the first sentence of Article 5, which sets forth various grounds under which a court "may" refuse to enforce a Mediated Agreement. Those defenders would presumably argue that (i) Article 5 does not require a court to refuse enforcement on the grounds of mediator breach but permits a court to refuse enforcement, and (b) any court adjudicating a defense asserted under the Mediator Breach Provision would have the discretion to consider (inter alia) the good faith reliance by the disputant who already performed.
Such an attempted defense of the Singapore Convention would be weak. Set forth below is an outline of the argument that would be made by the disputant who, on grounds of mediator breach, opposes enforcement and whose position is that the court in which enforcement is sought should ignore the issue of good faith reliance:
The disputant who opposes enforcement on the grounds of mediator breach will conclude his argument by observing that (i) there is not even a hint in the convention that good faith reliance should be a factor in the court's consideration of the applicability of the mediator breach defense, and, (ii) therefore, a court has no discretion to consider reliance in the context of the mediator breach defense.
The absence from the Singapore Convention of any express reference to reliance will undoubtedly be used by Recalcitrant Disputants to argue that, when a court is adjudicating the defense of mediator breach, good faith reliance should not be a factor taken into consideration.
The Singapore Convention's failure to address expressly the issue of good faith reliance will make it more difficult for a party that seeks enforcement to convince a court that reliance should be a consideration when adjudicating the mediator breach defense.
Under the Singapore Convention, a disputant who needs to seek enforcement of a Mediated Agreement – after having already performed – will be relegated to hoping that the court in the foreign state will value good faith reliance as much as that disputant does.
International conventions (treaties) are not supposed to be premised on wishful thinking.
In light of the fact that settlement agreements frequently provide for staged performance, and in light of the fact that good faith reliance has long been recognized as a significant factor in international commercial law, the Singapore Convention should have stated that a court that considers the defense of mediator breach needs to take good faith reliance into consideration when adjudicating the issue of "mediator breach."
The convention's failure to so provide is a major flaw.
Assuming arguendo that it is appropriate for a court – in some forum – to refuse to enforce a Mediated Agreement because of "mediator breach" (see supra, subsection A), the question that arises is whether it is wise for the Singapore Convention to impose no limitations as to the judicial forum in which a disputant who wishes to avoid enforcement may raise that defense. As explained below, (a) the answer is no, and (b) the Singapore Convention's failure to impose a limitation as to the forum raises serious questions as to whether the draftsmen of that convention in fact "modelled" it in any significant manner after the New York Convention.[26]
As a practical matter, a proceeding in court to enforce a Mediated Agreement will almost always be brought in a forum in which the Recalcitrant Disputant has assets. In many cases (perhaps the vast majority), that forum is the "home" country of the Recalcitrant Disputant. As a result, the issue of a mediator's alleged breach of applicable standards will usually be raised by the Recalcitrant Disputant in his/her home forum. And even though a Recalcitrant Disputant may have more than one "home," in many cases (again, perhaps the vast majority), the issue of mediator breach will be raised in a forum in which the Recalcitrant Disputant feels "more at home" than does the mediator whose conduct would be under scrutiny.
Allowing a Recalcitrant Disputant to raise the issue of "mediator breach" in that disputant's home forum – or in any forum other than one to which the mediator has a significant nexus – is very unwise.
But that is precisely what the Singapore Convention does. It does not require the disputant who asserts that the mediator breached applicable standards to raise that issue (i) in the country wherein the Mediated Agreement was arrived at, or (ii) in the country that defines the standards “applicable to” the mediation.
To illustrate the absurdity, let's look at a simple example:
At that stage, the German court will be required to adjudicate the allegation of mediator breach – even though the mediator has no jurisdictional nexus to Germany.
Will the
Independent of the inconvenience caused by the travel issue, in those (presumably many) cases in which enforcement is sought in a jurisdiction other than one with which the mediator has a significant nexus, the Singapore Convention virtually guarantees that, if a Recalcitrant Disputant wishes to raise the defense of mediator breach, that disputant may do so in a forum that is inconvenient in at least two respects – (a) it would be inconvenient for the court to receive evidence from the foreign mediator, and (b) in order to understand both the standards applicable to the mediator and the standards applicable to the mediation,[28] it would be necessary for the court to receive evidence as to foreign law and/or foreign practice.
This problem did not completely escape the draftsmen of the Singapore Convention. In one of the Working Group Reports, the inconvenient forum problem was summarized as follows:
. . . subparagraph (e) could lead to many litigations, making the enforcement cumbersome, which would run contrary to the purpose of the instrument; and . . . the court at the place of enforcement might not be best placed to consider issues pertaining to the conciliation process which, in most cases, would have taken place in a different State.[29]
The statement that the court at the place of enforcement "might not be best" suited to consider issues pertaining to the mediation process that took place in a different country – possibly even in a foreign language – is a great understatement.
Even though the Working Group was aware of the inconvenient forum problem, the decision was made, essentially, to ignore that problem or wish it away.
That decision was an irresponsible one. The practical effect of the
Any international legal regime that permits non-enforcement of a Mediated Agreement on the grounds of mediator breach should require that such a defense be brought only before a court in a country with which the mediator has a meaningful nexus.
The failure of the Singapore Convention to so limit the fora for the assertion of mediator misconduct is a major flaw of the convention.
This flaw is surprising because of the prevailing view (as noted above) that the Singapore Convention is intended to be the mediation "counterpart" to the New York Convention, which governs the international enforcement of arbitral awards. According to the official records of the negotiative history of the Singapore Convention, when the draftsmen were crafting Article 5, they looked for guidance to Article V of the New York Convention,[30] which sets forth the circumstances under which the (international) enforcement of an arbitral award may be refused.
Yet the phrase "arbitrator standards" does not appear in the New York Convention.
Rather, Article V, section 1 of the New York Convention permits (in subsections (a) through (e)) a signatory nation to refuse recognition and enforcement of an arbitral award on grounds related to (i) lack of
party capacity, (ii) invalidity of the arbitration agreement, (iii) inadequate notice or opportunity to be heard, (iv) the award exceeded the scope of the arbitration agreement, (v) a defect in composition of the arbitral authority, or (vi) the award is not yet binding.[31]
These six grounds (referred to herein as the "Administrative Grounds") largely do not imply any act of commission by the arbitrator, and, therefore, the conduct at issue with respect to such grounds would normally not be considered to constitute "arbitrator breach" of applicable standards.
With respect to the forum for raising any of the Administrative Grounds, the New York Convention does not impose any restriction – in other words, any of the Administrative Grounds may be asserted as a defense in any court in which enforcement of an arbitral award is sought.
However, Article V, section 1(e) of the New York Convention also sets forth an additional grounds – more accurately, a set of grounds – for nonrecognition. Subsection 1(e) provides that recognition may be refused if "[t]he award has . . . been set aside or suspended by a competent authority of the country in which, or under the law of which, that award was made." (Subsection 1(e) of the New York Convention is referred to herein as the "Set Aside Clause.")
The scope of the Set Aside Clause has been interpreted broadly. In Yusuf Ahmed Alghanim & Sons v. Toys "R" Us,[32] the U.S. Court of Appeals for the Second Circuit held that the Set Aside Clause authorizes a court in the country in which, or under the law of which, the award was made to set aside or suspend an award in accordance with the "full panoply of express and implied grounds for relief" of that country.[33]
In Republic of Argentina v. AWG Grp Ltd.,[34] the U.S. District Court for the District of Columbia cited to (inter alia) Yusuf Ahmed Alghanim & Sons, supra, and observed that the “full panoply” of relief under domestic law includes assertions of corruption, fraud, and evident partiality.[35]
To summarize the distinction under the New York Convention – unlike the Administrative Grounds, which may be asserted in any forum in which recognition is sought, the grounds referred to in the Set Aside Clause may be asserted only in "the country in which, or under the law of which, that award was made."[36]
That distinction is a wise one.
By making such distinction, the New York Convention precludes a losing party (usually the debtor) from forum-shopping as to the best court in which to assert a defense of "arbitrator breach of standards" – such as corruption, fraud, or evident partiality. If such a losing party wishes to raise any grounds available under the Set Aside Clause, it may do so only in a maximum of two possible fora.
Not only is this distinction under the New York Convention wise – but in order to simplify the dichotomy between the two types of jurisdiction (and perhaps because the phrase "the country in which, or under the law of which" is awkward), American case law on the subject has adopted the terminology "primary jurisdiction" and "secondary jurisdiction." The term "primary jurisdiction" refers to the courts of the country "in which, or under the law of which," the award was made, while the term "secondary jurisdiction" refers to all other New York Convention courts.
Subsequent to the Second Circuit's decision in Yusuf Ahmed Alghanim, appellate courts in the Fifth Circuit[37] and the District of Columbia Circuit[38] adopted the "primary/secondary" terminology to describe the different sets of jurisdictions under the New York Convention.[39]
The Fifth Circuit reiterated the use of the "primary jurisdiction" and "secondary jurisdiction" distinction, in Gulf Petro Trading Co. v. Nigerian Nat'l Petroleum:[40]
We have characterized the country “in which, or under the [arbitration] law of which,” an award was made as having primary jurisdiction over the award. . . .
All other signatory countries are then said to be secondary jurisdictions. . . .
[T]he Convention does not restrict the grounds on whichprimary-jurisdiction courts may annul an award, thereby leaving to a primary jurisdiction's local law the decision whether to set aside an award.” . . . Such courts are “free to set aside or modify an award in accordance with [the country's] domestic arbitral law and its full panoply of express and implied grounds for relief.” . . . .
In contrast, the Convention significantly limits the review of arbitral awards in courts of a secondary jurisdiction; essentially, “parties can only contest whether that [country] should enforce the arbitral award.” . . .[41]
As observed by the Second Circuit in Yusuf Ahmed Alghanim & Sons, supra, under the New York Convention, there are "very different regimes"[42] for (a) the review of an arbitral award in a primary jurisdiction and (b) the review of an award in any secondary
Put slightly differently, outside of a primary jurisdiction, an arbitrator's "breach of standards" is not one of the grounds available to a court for refusing to recognize an arbitral award.[43]
In summary, case law under the New York Convention has held that the convention means what it says – even though the convention does not use the terminology "primary jurisdiction/secondary jurisdiction," such distinction is clear, and it means that only a court of primary jurisdiction may adjudicate a defense of "breach of standards" by an arbitrator.
In looking to the New York Convention for an analogy to the Singapore Convention's "mediator breach" concept, it seems clear that two conclusions can be drawn regarding the category of "mediator breach": (a) it is similar (perhaps very similar) to the New York Convention's defenses under the Set Aside Clause, including those of "corruption, fraud, and evident partiality," and (b) it is largely dissimilar to the Administrative Grounds. In other words, the defenses under the New York Convention that are the most analogous to the Singapore Convention's category/defense of "mediator breach" may be asserted only in a "primary" jurisdiction.
The draftsmen of the Singapore Convention presumably realized that there is a similarity between an arbitrator's breach of standards and a mediator's breach of standards.
It is also reasonable to assume that those draftsman knew that, for sixty years, the New York Convention has required that any challenge on the grounds of "arbitrator breach" be brought exclusively in a primary jurisdiction. Therefore, one would have expected that, when the draftsmen of the Singapore Convention were formulating the grounds for refusal to enforce a Mediated Agreement, they would have taken note of the New York Convention's rule under which the analogous defenses may be asserted only in a primary jurisdiction.
Yet the draftsmen either ignored that distinction – for reasons that have not been explained – or they made a decision to reject it.
There is nothing in the negotiative history of the Singapore Convention indicating an intention to ignore the primary/secondary distinction. This suggests that no conscious decision was made to ignore it, which in turn suggests that the
In summary concerning the forum for asserting mediator breach, the permission given in the Singapore Convention to any court to refuse to enforce a Mediated Agreement on the grounds of mediator breach (a) ignores the sound policy embodied in the New York Convention, (b) invites Recalcitrant Disputants to raise the defense of mediator breach in an inconvenient forum, and (c) is counterproductive to the goal of encouraging the use of mediation to resolve international disputes.
This is a major flaw of the Singapore Convention.
The Singapore Convention states expressly that it does not apply to settlement agreements "[r]elating to . . . employment law."[44] Such an "exception" would appear to be consistent with the overall purpose of the convention to govern business disputes. However, as explained below, (a) the phrase "relating to employment law" is overly broad to accomplish the desired goal of "protecting" employees, and (b) like the convention's failure to state that a court is required to consider good faith reliance when adjudicating the "mediator breach" defense to enforcement,[45] the flawed drafting of section 2(b) will almost certainly invite superfluous litigation.
The Singapore Convention sets forth a number of limitations on the scope of its applicability. Subsection 2(a) provides that the convention does not apply to settlement agreements that are concluded "to resolve a dispute arising from transactions engaged in by one of the parties (a consumer) for personal, family or household purposes."[46] Subsection 2(b) addresses not only "employment law" but also provides that the Singapore Convention does not apply to agreements relating to "family" law or "inheritance" law.[47]
In light of these exceptions, although the convention does not use the term
The Singapore Convention is not the first international convention designed to regulate contracts the primary subject of which is
Article 2, Section 1(b) of the Hague Convention provides that the convention does "not apply to exclusive choice of courts agreements . . . relating to contracts of employment."
The phrase in the Singapore Convention "settlement agreements . . . [r]elating to . . . employment law" is broader than the phrase in the Hague Convention "choice of court agreements . . . relating to contracts of employment.”[49]
The draftsmen of the Singapore Convention were aware of the Hague Convention,[50] and with respect to certain other provisions of the Singapore Convention, the Working Group attempted to "align" the language of the draft convention with that of the Hague Convention.[51] Yet from the negotiative history of the Singapore Convention, there is no indication that the draftsmen took note of the more narrow language used in the Hague Convention – "relating to contracts of employment."
The practical effect of the use of the very broad language "relating to . . . employment law" is that it affords a Recalcitrant Disputant the ability to argue that a
A simple example illustrates the problem:
In the above example, no employee is a party – no employee had been a party to the dispute premediation, no employee is a party to the Mediated Agreement, and no employee is a party to the proceeding in court to enforce the Mediated Agreement.
Yet when Company B sues to enforce the Mediated Agreement – pursuant to the Singapore Convention – Company A asserts that, under Article I, section 2(b), the convention does not apply. That argument is based on the (very broad) definition of an agreement "[r]elating to . . . employment law." Company A argues that (a) because it covenanted in the Mediated Agreement not to employ certain employees, that agreement is one "relating to" employment law, and (b) because that agreement "relates to" employment law within the meaning of section 2(b) of the Singapore Convention, the agreement is outside the scope of this convention.
Based on the broad language in section 2(b) of the Singapore Convention ("[r]elating to . . . employment law"), Company A would have a reasonable argument that the agreement falls outside the scope of the convention and that, therefore, the convention may not be used to facilitate its enforcement. Such an argument could be made even though most readers of the convention would not assume that it was the intention of the convention's draftsmen to exclude such a
The above argument would not be possible if the Singapore Convention had used more narrow language, such as that used in the Hague Convention. As indicated above, the Hague Convention excludes from its ambit "contracts of employment." In our example above, there is no "contract of employment" between Companies A and B. Therefore, had the Singapore Convention used (in section 2(b)) the more narrow language "contracts of employment," Company A would not be able to argue that the Mediated Agreement between it and Company B falls within any exception to the scope of the convention.
Defenders of the Singapore Convention might argue that it is "clear" that the intention of the convention is not to exclude a
It is also possible that defenders of the Singapore Convention might argue that there is a good explanation – that the nature of the Hague Convention is such that it would have been inappropriate to use its "contracts of employment" language as a guide for a convention governing Mediated Agreements. Such an argument (presumably) could be based on differences in the purposes of the two conventions
terminate disputes, that convention needs exclusionary language that is broader than "contracts of employment."
But such an argument only goes so far. Even assuming that the draftsmen of the Singapore Convention were of the view that the language from the Hague Convention – "contracts of employment" – is too narrow, that does not explain the overly broad language in section 2(b) of the Singapore Convention. Had the draftmen of that convention wanted the exclusion to cover Mediated Agreements that relate to termination of the
Doing so would have ensured that the kind of
As explained in parts I and II above, the provisions of the Singapore Convention do not go any further in ensuring
In addition, and as explained in part III, (a) because the convention essentially invites Recalcitrant Disputants to assert the defense of "mediator breach," the convention risks causing harm to the international enforcement of Mediated Agreements, and (b) because of the overly broad exclusion of agreements relating to "employment" law, the convention will likely be interpreted (by at least some courts) to exclude certain
The web sites of the Department of Justice and the Department of State provide almost no information concerning the status (if any) of the ratification process of the Singapore Convention.
In light of the flaws described above, it is respectfully submitted that the administration and the senate take no action in furtherance of ratification of the Singapore Convention.
[1]. Reuters, U.N. members sign mediation convention to settle trade disputes, (last visited May 26, 2020).
The text of the Singapore Convention is available here (last visited May 26, 2020).
The web site of the United Nations states that a Working Group of the United Nations Commission on International Trade Law (UNCITRAL) considered the draft convention from its
The web site of UNCITRAL identifies the negotiative history of the convention as being embodied in (a) six Reports that the Working Group issued and (b) two Notes that the Secretariat issued.
Appendix A hereto sets forth the list and URLs of (a) the six Reports of the Working Group and (b) the two Notes of the Secretariat.
The defined terms set forth in Appendix A are used herein to refer to those Reports and Notes.
This article also refers to a third Note of the Secretariat, which appears elsewhere in the UN's web site. See infra, n. 26.
[2]. The UN General Assembly adopted the Singapore Convention on Dec. 20, 2018. See Press Release, General Assembly, General Assembly Adopts the United Nations Convention on International Settlement Agreements Resulting from Mediation, UNIS/L/271 (Dec. 21, 2018, the "UN Press Release"), available at https://unis.unvienna.org/unis/en/pressrels/2018/unisl271.html (last visited May 26, 2020).
[3]. Convention on the Recognition and Enforcement of Foreign Arbitral Awards (New York, 1958), 21 U.S.T. 2517, 330 U.N.T.S. 38; see UN Press Release (“[t]he [Singapore] Convention provides a uniform and efficient international framework for mediation, akin to the framework that the New York Convention has successfully provided over the past 60 years for the recognition and enforcement of foreign arbitral awards”); see also The "Singapore Convention": The way forward in international mediation?, formerly available here (last visited Sept. 9, 2019, hereinafter the "Linklaters Post"), available here (it is "hoped that the Convention will . . . quickly become as widely accepted as the New York Convention").
[4]. Contemporaneously with the adoption of the Singapore Convention, the United Nations adopted the UNCITRAL Model Law on International Commercial Mediation and International Settlement Agreements Resulting from Mediation, 2018 (the "New Model Law"). See here (last visited July 26, 2020).
The New Model Law amended certain provisions of the Model Law on International Commercial Conciliation, 2002 (the "First Model Law"). Very few of the major states (those that are very active in international commerce) of the United States adopted the First Model Law – specifically, New York, California, Texas, Florida, Pennsylvania, and Michigan have not adopted it. In addition, several of the major trading partners of the United States – such as the UK, Germany, Japan, and South Korea – have not adopted the First Model Law.
Substantially all of the criticisms voiced herein concerning the Singapore Convention apply as well to the New Model Law.
[5]. In many jurisdictions, even an arbitrator who served previously as a mediator and is known to the disputants would be required to make disclosures regarding possible conflicts of interest. See e.g. CA Civ Pro Code § 1281.9 (2016) (requiring arbitrator to disclose "[a]ny matters required to be disclosed by the ethics standards for neutral arbitrators adopted by the Judicial Council"); Honeycutt v. JPMorgan Chase Bank, N.A., 25 Cal. App. 5th 909, 922, 236 Cal. Rptr. 3d 255, 263, 2018 Cal. App. LEXIS 679 at *15; 2018 WL 3654895 at *5 (2d Dist. 2018) (an "arbitrator's duty of disclosure is a continuing one"). Presumably even in the jurisdictions that apply the most strict standards regarding arbitrator disclosure, the fact that the disputants already consented, in writing, to that person's acting as arbitrator would (in all but the rarest cases) constitute a valid and enforceable waiver of any possible conflict.
[6]. This author has previously argued (in an article published by the Business Law Section of the ABA) that every international arbitration agreement should include a provision for costs to be paid by the losing party. See Eric Sherby, A Checklist for Drafting an International Arbitration Clause, BUSINESS LAW TODAY (Sept. 20, 2010), available here (last visited Aug. 2, 2020).
At the same time, the author acknowledges that many mediators discourage counsel from including a costs clause in Mediated Agreements – on the grounds that "anything that suggests a future dispute goes against the spirit" of the Mediated Agreement.
This author tends to reject that advice from mediators.
[7]. When the mediator who facilitated a Mediated Agreement later serves as the arbitrator to adjudicate disputes under that agreement, such an arbitration is in the category of "ad hoc" arbitration. In contrast to institutional arbitration, agreements that provide for ad hoc arbitration frequently are silent regarding the mechanism for replacing the arbitrator if such a need arises. In this context, the International Bar Association Guidelines for Drafting International Arbitration Clauses (2010, the "Guideline") offers the following wise advice:
The need to designate an appointing authority in the context of ad hoc arbitration constitutes a significant difference between [institutional arbitration] and [ad hoc arbitration]. In institutional arbitration, the institution is available to select or replace arbitrators when the parties fail to do so. There is no such institution in ad hoc arbitration. It is, therefore, critical that the parties designate an ‘appointing authority’ in the ad hoc context, to select or replace arbitrators in the event the parties fail to do so. Absent such a choice, the courts at the place of arbitration may be willing to make the necessary appointments and replacement. . . .
The appointing authority may be an arbitral institution, a court, a trade or professional association, or another neutral entity. The parties should select an office or title . . . rather than an individual (as such individual may be unable to act when called upon to do so). . . .
See the Guideline ¶¶
[8]. See Linklaters Post, supra note 3.
[9]. A New Legal Framework for the Enforcement of Settlement Agreements Reached through International Mediation: UNCITRAL Concludes Negotiations on Convention and Draft Model Law, EJIL: Talk!, Posting of Christina Hioureas and Shrutih Tewarie to https://www.ejiltalk.org/a-new-legal-framework-for-the-enforcement-of-settlement-agreements-reached-through-international-mediation-uncitral-concludes-negotiations-on-convention-and-draft-model-law/ (last visited Jun. 22, 2020).
[10]. Mediation on the world stage, posting of Eilidh Smith to
[11]. If Mediated Agreements are routinely breached, it is possible (perhaps likely) that the reason for routine noncompliance is the failure to include one or more of the Three Clauses.
[12]. See infra text accompanying note 15.
[13]. Nothing in this article is intended to express a view concerning the
Med/arb is sometimes criticized for the failure to appreciate the inability of the arbitrator to "forget" that which he was told in ex parte communications with the disputants. See Brian Pappas
Yet even the most vehement critics of "med/arb" should not be able to find anything objectionable in designating the (former) mediator to serve as the arbitrator in the event of a dispute concerning the Mediated Agreement. In such a situation, the myriad of issues that had been in dispute,
[14]. Exceptions do exist – such as when a disputant decides, during the course of the mediation, that s/he does not have confidence in the ability of the mediator to be impartial in an adjudicative role. In such a situation, it is possible (a) that the disputant will be of the view that the lack of confidence in the mediator is insufficient to terminate participation in the mediation process yet sufficient to preclude designating that mediator to serve as arbitrator, and (b) the two disputants will sign a Mediated Agreement that includes an arbitration clause that provides for someone other than the mediator to serve as arbitrator.
Presumably such situations are exceptional. Mediators are usually selected by counsel, and a lawyer who anticipates that the Mediated Agreement will provide for the mediator to serve as arbitrator (in the event of a future dispute) will try to steer clear of mediators who have a reputation for displaying bias or the like.
[15]. Report 896 ¶ 95.
[16]. See the Linklaters Post, supra, n.3 emphasis added). There is no indication in any of the Working Group Reports that the issue of any "lack" of insurance coverage was discussed.
[17]. This author's lack of familiarity with mediators who refuse to act as arbitrator because of an insurance coverage issue is not the only reason for assuming that this phenomenon is (at most) rare. Major arbitral institutions – such as the International Chamber of Commerce, the American Arbitration Association, and JAMS (Judicial Arbitration and Mediation Services) – offer mediation services, and most such institutions hold out their "neutrals" as being able to serve in both the capacity of mediator and of arbitrator. See web site of AAA, referring collectively to "Arbitrators & Mediators," last visited May 26, 2020; web site of JAMS, list of neutrals does not distinguish between mediators and arbitrators; last visited May 26, 2020.
From the published materials of those ADR organizations, there is no indication that they charge more for mediations in which the mediator agrees in the settlement agreement to act as arbitration in connection with future disputes.
[18]. Art. 5, section 1(e).
The plain language of section 1(e) of the Singapore Convention (quoted above) makes clear that there might be a difference between (i) the standards applicable to the mediator, and (ii) the standards applicable to the mediation. Because section 1(e) permits refusal to enforce if the mediator committed a serious breach of either body of standards, implicitly the convention holds the mediator to the more stringent set of standards.
The "more stringent rule" is not the erroneous aspect of section 1(e). By way of example, a mediator from the UK who travels to South Africa to conduct a mediation there would presumably understand that he is subject to the laws of South Africa – at least for the period of time that he is physically located in South Africa.
It is possible that the “standards applicable [to] the mediation” derive solely from the law of the place where the mediation takes place. But those standards could also derive from the rules of an organization or institute that appointed the mediator. For example, it is not uncommon for the London Court of International Arbitration to appoint a mediator to conduct a mediation in a country other than the UK. See
[19]. As indicated above, this author believes that most treatment to date of the Singapore Convention has been superficial. But see F. Peter Phillips, Concerns on the New Singapore Convention, BUSINESS CONFERENCE MANAGEMENT LLC (2018), (last visited Jun. 22, 2020; criticizing the Singapore Convention for opening the door to motion practice concerning alleged mediator misconduct). Perhaps because Phillips does not support any motion practice regarding mediator breach, his analysis is silent as to whether a challenge should take place in a country other than the one with which the mediator has a significant nexus.
[20]. See Scott M. Himes, A Best Practices Primer For Drafting Settlement Agreements, NYLJ (Jul. 15, 2013) (available here; last visited Jun. 7, 2020; "many settlements involve payments over time, or 'installment' payments. In this scenario, the party to be paid does not want to give a release and dismiss the lawsuit before receiving full payment, which might be years away").
The UNCITRAL Working Group itself made a similar observation:
It was noted that the flexible nature of settlement agreements, which allowed for conditional obligations, was a key feature of conciliation that made it attractive to parties and thus, the need to preserve such a feature was highlighted.
Report 861 ¶ 48; see also Report 896 (¶ 89) (referring to the Working Group's understanding that there are "complex settlements [in which] parties would settle parts of their dispute over time"); see also id. ¶ 160 ("the place where a substantial part of the obligation under the settlement agreement was to be performed . . . might not be known at the time of conclusion of the settlement agreement").
[21]. United Nations Convention on Contracts for the International Sale of Goods, U.N. Doc. A/CONF. 97/18, Annex 1, S. Treaty Doc. No.
[22]. Id. art. 16.
[23]. There is no question that the UNCITRAL Working Group took into consideration other aspects of the CISG. See Report 896 ¶¶ 58, 132; Report 929 ¶ 22.
[24]. See note supra, n. 20.
[25]. If the draftsmen of the Singapore Convention had been of the view that the convention should specify good faith reliance as a factor to be taken into consideration in determining whether to refuse enforcement on the grounds of mediator breach, it would not have been difficult for the convention to so provide expressly, such as by adding:
When considering whether to exercise its discretion under Article V, Section 1(e), the court should take into consideration any good faith reliance by the party that seeks to enforce the Mediated Agreement.
[26]. See supra text accompanying note 3; see also Note by the Secretariat, UNCITRAL Working Group, A/CN.9/WG.II/WP.198, ¶ 34 (available here (Jul. 26, 2016) (the Working Group generally was of the view that the defenses "should not be less favourable than [those] provided for . . . under the New York Convention").
[27]. For purposes of this discussion, it is not necessary to define the jurisdictional nexus that could or should be applied to a mediator. That nexus need not be nearly as intensive as the kind that would subject a foreign national to personal jurisdiction.
[28]. The Disputant who asserts mediator breach will likely assert that the mediator violated at least one set of standards, if not both.
[29]. Report 896 ¶ 106.
[30]. See Report 861 ¶¶ 86, 92; Report 867 ¶ 171. As indicated above, a proceeding in court to enforce a Mediated Agreement will almost always be brought in a forum in which the Recalcitrant Disputant has assets. The same (obviously) can be said with respect to enforcing an arbitral award – a proceeding in court to enforce an award will almost always be brought in a forum in which the recalcitrant debtor (from the arbitration) has assets.
[31]. Section 2 of Article V of the New York Convention sets forth two additional grounds for refusal to enforce – (a) the subject matter is not arbitrable in the state in which enforcement is sought, and (b) recognition would be contrary to public policy. These two grounds are of little relevance to the issues discussed herein.
[32]. 126 F.3d 15 (2d. Cir. 1997).
[33]. Id. at 23.
[34]. 211 F. Supp.3d 335 (D. D.C. 2016).
[35]. Id. at 346 (citing grounds enumerated under the Federal Arbitration Act, section 10).
[36]. For example, if a Canadian company and a French company agree to arbitrate in the UK under an agreement governed by New York law, then the judicial act of setting aside the award could be within the jurisdiction only of a court in the UK or of a court in New York.
[37]. It appears that the first court to have used the "primary/secondary" terminology was the Fifth Circuit in Karaha Bodas Co. v. Perusahaan Pertambangan Minyak Dan Gas Bumi Negara, 335 F.3d 357, 364, 368 (5th Cir. 2003) ("Karaha Bodas "). The Fifth Circuit reiterated the use of such terminology in Karaha Bodas Co. v. Perusahaan Pertambangan Minyak Dan Gas Bumi Negara, 364 F.3d 274,
[38]. The Court of Appeals for the District of Columbia adopted the use of the terminology "primary/secondary" in TermoRio S.A. E.S.P. v. Electranta S.P., 487 F.3d 928, 935 (D.C. 2007).
[39]. The "primary/secondary" nomenclature has been criticized to the extent that it is anything beyond a "shorthand that . . . identifies the nationality of courts that have the power of annulment." Marc Goldstein, Annulled Awards in the U.S. Courts: How Primary Is "Primary Jurisdiction?", 25 AMER. REV. INT. ARBITRATION 19 (2014).
[40]. 512 F.3d 742, 746 (5th Cir. 2008).
[41]. Id. at 746 (citations and quotations marks omitted).
[42]. 126 F.3d at 23.
[43]. See generally Lea Haber Kuck and Amanda Raymond KalantirskyVacating An International Arbitration Award Rendered in the United States, 6 NYSBA (2017) (previously available here).
[44]. Singapore Convention, art. I, section 2(b).
[45]. See supra, text accompanying notes
[46]. Id. section 2(a).
[47]. Id. section 2(b).
[48]. https://www.hcch.net/en/instruments/conventions/full-text/?cid=98.
[49]. In an online article published a few months after the Singapore convention was signed, mediate.com erroneously stated that the convention's list of exclusions includes "employment disputes." See New Horizon for International Commercial Mediation: The Singapore Convention, posting of Stanley Santire (last visited June 4, 2020). (The last name of the author of the cited article is "Santire," even though the URL contains the word "satire." Nothing in the article indicates that it is meant as satire.) As indicated in the text, the language of the convention is broader than "employee disputes."
[50]. Report 896 ¶¶ 49, 170.
[51]. Id. ¶ 52; see also Report 929 ¶ 62 (certain provisions of the Hague Convention could serve as "a useful model").
Source: Click here (last visited Jun. 7, 2020)
The author specializes in international litigation and arbitration at the Israeli law firm that he founded in 2004, Sherby & Co., Advs., www.sherby.co.il. He serves as a Vice Chair of the International Litigation Committee.
The author expresses his deep gratitude to Jeff Kichaven – a seasoned mediator – for serving as a resource for many of the issues addressed in this article. Notwithstanding Mr. Kichaven's input and assistance, the views set forth in this article are solely those of the author.
