ARBITRATION’S PROCEDURAL MATRIX*
In 1891, a Dutch doctor named Eugene Dubois unearthed in Java a skull fragment with a low forehead, buried only a few metres from where a human thigh bone was later found. Concluding (incorrectly) that both fossils came from the same creature, he announced the discovery of evolution’s missing link, an extinct primate called Pithecanthropus erectus that was neither fully ape nor fully human.1
The evolution of ideas also implicates a search for connections and missing links. Sometimes the hunt yields only dead ends, like Dr Dubois’s hypothetical primate. Sometimes, however, the quest supplies clues to advance knowledge about the type of change we call progress.
While change is inevitable (today is not yesterday), not all change is positive. A cannibal who learns to use a knife and a fork remains a cannibal. Moreover, some things are better left alone. The familiar configuration of keyboard letters endures through generations because the costs of innovation (including the time to learn something new) outweigh potential benefits.2
The evolution of business arbitration remains of critical importance to international economic cooperation. Without reliable ways to resolve cross-border disputes, many wealth-creating transactions either will remain unconsummated, or will be concluded at higher costs to reflect the absence of adequate mechanisms to vindicate contract rights. Arbitration of business disputes thus provides net benefits from the perspective of both national welfare and the shared interests of the global commercial community.
Three case studies in change provide springboards for inquiry. The first looks at judicial review of commercial awards under national arbitration statutes. The second addresses the treaty foundations of arbitral power over international investment disputes. The final one examines norms governing the conduct of the arbitral proceedings themselves.
Like prisms, these examples of progress separate the threads of several dilemmas that have nipped at the heels of business arbitration for decades. All three implicate a fine-tuning in the counterpoise of fairness and efficiency. Revealing only fragments of truth (the most one can expect this side of eternity), the tentative conclusions on these issues are presented here as an aperitif to more substantial conversations about the direction of arbitration.
The past half century has brought an embarras de richesse in the evolution of arbitration’s procedural architecture. Much of the change has been directed at enhancing the reliability of the arbitral process. Although arbitration is consensual, enforcement is not. To give global currency to commercial awards, the United Nations in 1958 adopted the New York Arbitration Convention.3 Seven years later, the Washington Convention on Investment Disputes was concluded, providing a structure for arbitration between host states and foreign investors that has since been pressed into service through a multiplicity of free trade agreements and investment treaties.4
Legislatures got involved, promoting arbitral progress through statutory reform. In 1979, England fired the starting gun with a new Act designed to reduce challenges to awards.5 Arbitration reform blossomed during the next two decades, with new laws in France,6 Belgium,7 Italy8 the Netherlands,9 and Switzerland,10 as well as another round of reform in England.11 In 1985, the United Nations sponsored a Model Law on International Commercial Arbitration (the UNCITRAL Model)12 which has been enacted in forty countries.13
A parallel explosion occurred worldwide in the procedural rules which litigants adopt to guide their arbitration. Since 1975, there have been three revisions of arbitration rules by the International Chamber of Commerce,14 two by the London Court of International Arbitration,15 and at least four by the American Arbitration Association.16 The United Nations in 1976 promulgated rules for unsupervised arbitration,17 and in 1994 the World Intellectual Property Organization created its own framework for arbitration of patent, copyright and trademark disputes.18 During the late 1990s, specialized rules were developed to permit arbitration of mass claims related to Holocaust era insurance policies and bank accounts.19 Rules have been promulgated or revised by regional organizations in Sweden,20 China,21 Finland,22 and Switzerland.23
During this time, neither academia nor the bar remained idle. Tenure-hungry professors and eager practitioners churned out treatises (on both domestic24 and international25 arbitration) and casebooks26 at a pace that made arbitral scholarship a veritable industry. Journals, yearbooks, conferences, and courses on the topic have multiplied to the point where few mortals can keep current with the literature.27
Making sense of the changes in arbitration’s rich procedural tapestry is not always easy. As discussed later, however, some springboard for discussion might be taken from analogies in the worlds of culture and biology.
Saris and catsuits
Not too long ago, National Geographic Magazine ran a cover story called “Global Culture” which explored the interaction of people and ideas from different places.28 The lead photo, intended to show the meeting of starkly different styles, featured a mother and daughter in India. The mother wore a traditional red and gold sari, while her daughter sported a skintight polyvinyl black catsuit.
In business arbitration, however, different cultures often meet in a way similar to a combination of the top half of the sari and the bottom of the catsuit. Rather than stark contrasts, procedural evolution in arbitration has led to homogenization, hybrids, and amalgam.29
The resulting legal culture may appear more non-national than inter-national. For example, written witness statements30 are often combined with some measure of oral testimony.31 As lawyers from different countries find themselves on opposite sides of the same case, a shared legal culture tends to develop on at least some of the cross-border procedural questions that arise during the arbitration.
Change frequently arrives through intellectual cross-pollination, a metaphor evoking images of lawyers as worker bees, buzzing with news of different ways to resolve disputes.32 Exchanges occur among divergent legal traditions (exemplified by the international cooperation in drafting the UNCITRAL Model Arbitration Law33) and between different types of dispute resolution, as when commercial practices influence investment arbitration.34
Cross-pollination is not always well received. In particular, one frequently hears complaints about the “Americanization” of arbitration,35 usually related to aggressive litigation tactics that include hefty boxes of unmanageable exhibits, costly pre-trial discovery, and disruptive objections to evidence.36
Procedural change involves not only the Americanization of international arbitration, but also the internationalization of American dispute resolution, as reflected in greater use of written testimony and reasoned awards.37 Evolution in ethical standards for arbitrators presents another striking example of internationalization.38 Traditionally, American practice presumed party-nominated arbitrators to be partisan, and thus permitted ex parte communication with their appointers.39 By 2004, however, most arbitration in the United States was brought into line with global standards,40 requiring independence for all arbitrators.41 While doctrinal predisposition may still be relevant (investors claiming for asset expropriation would hardly appoint a left-wing professor who supported uncompensated confiscation), bias is no longer permitted.42
Accident or design?
Progress in arbitration law sometimes happens by accident, and sometimes through purposeful attempts by determined goal-oriented individuals. Without pushing the comparison too far, one might suggest analogies (albeit imperfect) to two different explanations for biological evolution. One model holds that random variations survive simply by accident, while the competing paradigm rejects chance as a suitable explanation for how life developed.
According to Darwinist43 and neo-Darwinist44 views, if a billion monkeys banging on keyboards for a billion years might write Hamlet, then billions of mutations might arguably create an eye, ear, or brain by accident. Evolution combines both chance and necessity, with chance evident in genetic variations, and necessity present in the selection of the mutations that survive.45
By contrast, a competing school of thought sees evidence that evolution involves some purposeful design.46 Accident is not adequate, say these scholars,47 to explain the irreducible complexity of life.48 Something more than blind chance is required, even if the nature of that “something more” remains uncertain and challenges scientific orthodoxy.49
Explaining arbitration by the first model posits a business community that accidentally stumbles across laws that seem to work. The alternative view sees design as the leitmotiv, emphasizing purposeful efforts of scholars and legislators in instruments such as the UNCITRAL Model Arbitration Law, the New York Convention, and the various professional guidelines for arbitral procedure. As the following discussion hopes to show, the more robust explanations of arbitral progress rely on both models.
Competing models of judicial review
Today, most arbitration statutes recognize that commercial arbitration should neither be subject to a system of full judicial appeal, nor free from all recourse against aberrant decisions. Rather, court scrutiny exists to promote the integrity of arbitration by ensuring that arbitrators follow a modicum of procedural fairness and remain within the limits of their mission.50
Such has not always been the case, however. Until well into the twentieth century in many countries, an unhappy litigant could challenge an arbitrator’s decision under an “appellate model” of judicial review. Judges were permitted to consider not only the integrity of the arbitral process, but also alleged mistakes on points of law.51
By contrast, consensus has emerged on what might be called the laissez-faire model of judicial review. Under this paradigm, courts intervene only to monitor arbitration’s basic procedural integrity, by assuring (i) the basic procedural fairness of an arbitration (lack of bias, right to be heard and equality of arms) and (ii) respect for limits of arbitral jurisdiction.
The laissez-faire model means that arbitration sometimes results in a looser approximation of legal rights than found in court litigation, tolerating occasional uncorrected mistakes in the interpretation of law and contract language. However, the laissez-faire standards usually comport with commercial expectations.52 Normally, the arbitrator’s award is supposed to be the end rather than the beginning of dispute resolution—something other than mere foreplay to litigation.
In international arbitration, such laissez-faire standards clearly represent the trend for review at the “seat of arbitration,”53 an expression that increasingly serves to designate the country to which the parties have pegged their international arbitration, notwithstanding that the hearings and deliberations take place elsewhere.54 Laissez-faire review standards have found their way into the law of countries that have adopted the UNCITRAL Model Arbitration Law, as well as most major arbitral centers (including England, France and Switzerland) whose laws have evolved more independently.55 In contrast to domestic arbitration, where appellate review on matters of law has sometimes been considered appropriate, international arbitration has long been driven by an understandable need for greater autonomy from national court supervision.56 For cross-border transactions, arbitration serves above all as a means to reduce the prospect of litigation in courts that apply unfamiliar procedures in a foreign language. For international disputes, arbitration justifies itself not so much by considerations of cost and speed, but rather as a politically and procedurally neutral forum that helps foreclose access to the disruption of competing lawsuits in multiple litigation venues.57
Prior to 1979, London arbitration meant judicial review of substantive legal issues under the so-called “case stated” procedure.58 The perceived result was that arbitrations were crossing the Channel to the more hospitable environment of Paris, with a loss of what were sometimes called “invisible exports,” a euphemism for fees paid to arbitrators, lawyers, and expert witnesses.59 It seemed that the users of arbitration services no longer wanted high-priced barristers arguing lengthy appeals.
So England adapted to the new environment by permitting parties to international contracts to exclude review on questions of law.60 The effect was felt principally in commercial cases subject to the rules of institutions such as the International Chamber of Commerce (ICC), which incorporated automatic exclusion of appeal.61
The reforms did not go all the way, however. Exclusion of appeal was generally not allowed in so-called “special category” disputes related to admiralty, commodities, and insurance.62 In these areas, the pre-eminence of English law was thought to require fertilization by judgments covering novel controversies. Not until the 1996 Act did England subject all commercial arbitration to the same level playing field.
Independently, off in Vienna, a United Nations Working Group was drafting its model arbitration statute, which would incarnate a similar approach to judicial review (inspired by Article V of the New York Arbitration Convention), but with a very different motivation. This UNCITRAL Model Arbitration Law represents a careful attempt to promote more harmonious economic relations through a uniform law that minimizes transaction costs from idiosyncratic national statutes.
Some countries have shown additional levels of intelligent design by filling gaps in the UNCITRAL Model Arbitration Law in a way consistent with its goals.63 For example, in Australia64 and Scotland65 awards may be set aside in the event of fraud, a ground for vacatur not explicit in the Model Law.66
These changes in the framework for judicial review reflect a growing deference to party autonomy. For international transactions, New Zealand permits parties to elect review on the substantive merits of an award, with a more laissez-faire regime applying as the default rule.67 England, Switzerland and Belgium recognize, in differing measures, liberty of contract with respect to award finality.68 In Australia and Singapore, parties can by contract exclude application of the UNCITRAL Model Arbitration Law altogether,69 although the results of exclusion vary between the two countries.70
Some statutes (such as the UNCITRAL Model Arbitration Law and the arbitration acts in France and Switzerland) also include “public policy” as a ground for judicial review,71 albeit with a more limited scope (ordre public international) when applied to international transactions.72 Public policy (which should rightly be considered a plural rather than singular noun) implicates a cluster of chameleon-like notions whose unifying essence lies in overriding societal interests that constrain how arbitrators decide cases.73 Some relate to procedure (minimum fairness in the conduct of an arbitration) while others relate to the content of the award as it reflects substantive merits of the dispute.74
The malleability of public policy notions makes them problematic in their implementation,75 principally because such a chameleon-like concept risks misapplication when refracted through parochial cultural lenses.76 This does not mean that international public policy has no role to play in the resolution of cross-border disputes, but simply that the concept must be approached humbly and cautiously.77
To a large extent, laissez-faire judicial review remains the legacy of two great French thinkers, Professors Berthold Goldman and Philippe Fouchard, who during the 1960s began to explore notions of “delocalized” arbitration autonomous from national law.78 Their ideas have since been invoked (sometimes thoughtfully and sometimes less so) with respect to both procedural79and substantive80 legal norms.
First and foremost, “delocalization” suggests minimal judicial intervention by judges at the arbitral situs. The aim is to safeguard an arbitration’s procedural integrity, but not to engage in what Americans might call “Monday morning quarterbacking” on the correctness of an arbitrator’s decision. For international transactions, this is an idea whose time has clearly arrived in most major arbitral centers.
Second, theories of “non-national” arbitration have sometimes been invoked to justify enforcement of awards vacated at the place of arbitration. According to one view, the country where a losing party has assets might recognize an award even if it had been set aside where made.81 In one well-known case an award annulled in Geneva was nevertheless recognized in Paris.82
Outside of France, such resurrection of dead awards has received a less enthusiastic reception, again illustrating intelligent design. The English judiciary quite sensibly refused to enforce vacated awards,83 and American courts are divided.84 In this regard one sees the shadow of another great internationalist, Dr Francis Mann, who argued forcefully for some control of awards by the law of the arbitral situs, which he called the lex loci arbitri.85 Given the complexity of the topic, it is likely that the debate’s final round is yet to be heard.86
On occasion the evolutionary pendulum has swung wildly out of control, producing self-destructive mutants. Twenty years ago, Belgium conducted a failed experiment in arbitral anarchy by eliminating all right to have awards vacated in disputes between foreign parties.87 Contrary to what was expected, this created more anxiety than comfort. There were no custodians to guard the custodians.88 Few business managers wanted an arbitral pig-in-a-poke, with no recourse at the place of proceedings against gross procedural irregularity. While arbitration might mean assuming the risk that arbitrators get it wrong on the facts or the law, it would not normally constitute a bargain for fraud, bias or excess of authority. So the Belgians had to try again, and ultimately amended their law to provide a safety net of judicial review unless the parties expressly chose otherwise.89
Manifest disregard of the law
Not all countries accept that courts should review international business awards only on matters of procedural fairness. In the United States, most awards may be vacated for “manifest disregard of the law,” a vague term subject to varying interpretations, some of which permit judicial tinkering in the substantive merits of a case,90 and can imply a subjective element in the arbitrator’s intent to deviate from (disregard) his or her duty.91
In giving substance “manifest disregard” some courts take a restrictive view. Building on notions of excess of authority, they apply the principle to awards that ignore contract language92 or violate fundamental public policy.93 Other courts, however, take a more expansive view, including mistakes of law94 and moving well beyond a consumer or employment context.95 Yet another approach has been taken by the Fifth Circuit:96 a two-prong inquiry determining first whether it was manifest that the arbitrators disregarded applicable law, and then whether the award would result in “significant injustice” under the circumstances.97
While judicial review for legal error98 exists in other countries,99 its effect is usually restricted to domestic cases100 or subject to waiver by the parties in non-consumer transactions.101 By contrast, in the United States most courts have determined that “manifest disregard” cannot be excluded by contract.102 Some courts,103 but not all,104 apply the notion even in international arbitrations. From the perspective of both statutory text and arbitral policy the better view is that “manifest disregard” has no place in the review of foreign awards enforceable under the New York Convention.105
The problematic nature of “manifest disregard” lies in its potential for mischief in large international cases, when zealous litigators may be tempted to press the doctrine into service as a proxy for attack on the substantive merits of an award. It is hard to imagine meaningful review of legal error in a complex case without substantial time to understand the facts.106 Even if costly appellate briefing and argument yields a happy ending for the arbitration’s prevailing party, the very existence of “manifest disregard” hangs like the sword of Damocles to be grasped by award debtors who understandably seek relief from costly damages. Outside consumer disputes, “manifest disregard” is often the right doctrine aimed at the wrong target, creating the prospect of judicial meddling that can only alarm foreign enterprises that have expressly agreed to arbitration in the hope of keeping out of national courts.
Consumers and employees
Why the persistence of “manifest disregard” as a ground for vacatur in the United States? At least one explanation might lie in the absence of any comprehensive statutory regime to safeguard the procedural integrity of consumer and employment arbitration. The value of even-handed adjudication, which commends arbitration among sophisticated business managers, sometimes gets lost when form contracts are imposed on ill-informed individuals by manufacturers or employers with grossly disproportionate bargaining power. The result may be a distant forum of uncertain integrity.107
The special status of consumer and employment disputes is well established in Europe, where statutes of general application safeguard against abusive arbitration108 and other “unfair” contract terms.109 In England consumer protection regulations are expressly applicable to arbitration agreements,110 regardless of whether they cover present or future disputes and regardless of what law is applicable to the arbitration agreement.111 Arbitration agreements will be considered unfair (thus presumptively invalid) if they relate to claims below an administratively fixed amount.112 France has long made a distinction between the predispute clause compromissoire and the post-dispute compromis, the former being valid only in contracts between merchants (commerçants) or persons contracting with respect to a professional activity.113 Moreover, French employment disputes are subject to the exclusive jurisdiction of a special labor court (conseil de prud’hommes)114 except in international cases.115 In Germany, both parties must sign consumer arbitration agreements, which must either be contained in a separate document or have notarial certification.116
Such protective regimes have been generally neglected in American arbitration, at least on the federal level, where a “one-size-fits-all” federal arbitration statute pre-empts state law that tries to protect weaker members of society.117 The Federal Arbitration Act (FAA) can apply equally to a billion-dollar construction contract and a consumers purchase of fifty dollars’ worth of software. The statutory scope can cover both disputes over a CEO’s multi-million dollar golden parachute or an action for unfair dismissal of a filing clerk.118
The problematic nature of American arbitration in consumer and employment contexts has not gone unnoticed.119 In such an environment, award vacatur for “manifest disregard” serves as a less than optimum safety valve against abuse, as does invalidation of arbitration clauses on the basis of “unconscionability,”120 whether labeled procedural (weaker party has no meaningful opportunity to bargain)121 or substantive (one-sided duty to arbitrate, unilateral discovery rights, and oppressive fees).122
While one can understand the impulse to protect weaker members of society through the use of “manifest disregard,” better alternatives do exist. These include protocols on procedural fairness applied specifically to consumer and employment disputes, whether imposed by government regulation or by private institutions.123 Another option might construct a separate legal regime for consumer and employment cases, perhaps enforcing only post-dispute arbitration agreements that give individuals time to consider what is at stake, or subject such arbitration to a heightened degree of judicial supervision.
At least in part, the root of this American exceptionalism lies in the nature of the litigation process in the United States. In particular, claimants in ordinary contract actions possess a right to be heard by a civil jury and in some cases to claim punitive damages. These factors give industry groups a special incentive to rely on arbitration clauses in consumer contracts, and to resist reform that might limit that practice.124 Within the business community, a widespread perception exists that arbitrators will be more reasonable than jurors in hearing claims and awarding punitive damages in actions that seem to pit the “little guy” against a large manufacturer or financial institution.
Rightly or wrongly, a fear exists that even a moderate attempt to fix things would spiral out of control.125 Reform is seen as opening a Pandora’s box of unbridled upheaval, led by an unholy alliance of consumer advocates and plaintiffs’ lawyers bent on destroying an arbitration system they see as protecting crooked finance companies and abusive bosses from sympathetic juries prone to award damages to the likes of dear Aunt Millie and old Uncle Fred.126
The consequence is that American arbitration has taken a special path. As Australia has its own types of animals (kangaroos, wombats and koala bears), so the United States has hosted evolution of its own varieties of arbitration, including judicial review on mistakes of law not limited to consumer and employment arbitration.127
Courts often get involved in arbitration when one side challenges the arbitrators’ jurisdiction, either at the beginning of the process (if a court action is filed in disregard of the arbitration clause) or at the end (if award enforcement is resisted due to an alleged excess of arbitral authority).128 The alleged limits on arbitral power may derive either from the scope of the contract or from public policy. On both counts courts have shown an increasingly benevolent attitude toward arbitral jurisdiction.129
Arbitrators were once considered a bit like foxes guarding the chicken coop: not to be trusted. The resulting judicial hostility toward arbitration meant limited arbitral power to hear disputes implicating sensitive public interests such as competition law, patents, and securities transactions, which were considered “non-arbitrable” subjects beyond the competence of arbitrators regardless of what the contract said.130
One by one, like uncomfortable garments on a hot summer day, such subject matter arbitrability restrictions are being removed. Competition law, patents, and securities regulation are now generally arbitrable.