This article examines the Dubai International Financial Centre (DIFC) Courts as one of the most striking legal experiments of the last two decades: a fully functioning English-language #common_law judiciary planted inside a jurisdiction whose national legal system is rooted in the #civil_law tradition and influenced by Islamic law. Using the DIFC as a case study, the article traces how a small financial #free_zone in Dubai built its own courts, imported judges and doctrines from England and other common-law states, and then expanded its reach far beyond its original geographic borders. The study analyses the constitutional foundation that made this possible, the institutional design of the Courts, the mechanics of #legal_transplant, and the two great jurisdictional innovations that define the DIFC model: the opt-in jurisdiction that turned a local court into a global forum, and the conduit function that made the DIFC a gateway for the #enforcement of judgments and arbitral awards. It also addresses the friction that autonomy produces, including conflicts with the onshore Dubai Courts, the 2021 restructuring of arbitration in the emirate, and criticism that such courts risk becoming reputational cover for weaker parts of a host legal system. The article concludes that the DIFC Courts have largely succeeded on their own terms, but that their long-term legitimacy depends on maintaining genuine #judicial_independence, managing conflict with domestic courts, and continuing to develop a distinct body of DIFC #jurisprudence rather than simply copying English precedent. Keywords: DIFC Courts; common law; civil law; legal transplant; international commercial courts; jurisdiction; arbitration; enforcement; Dubai; United Arab Emirates 1. Introduction The image in the title of this article is deliberate. The #DIFC is often described as a common-law island in a civil-law sea, and the metaphor captures something real. The United Arab Emirates is a federation whose national courts apply a codified civil law system, conduct proceedings in Arabic, and draw on Islamic legal principles as a source of legislation. Inside the emirate of Dubai, however, sits a financial district of roughly 110 acres in which a different legal world operates. There, an independent judiciary hears #commercial_disputes in English, applies procedures modelled closely on the Civil Procedure Rules of England and Wales, and staffs its bench with judges drawn from England, Australia, Singapore, Malaysia, and other common-law jurisdictions. This is the DIFC Courts, and it is the subject of this case study. The question that animates the article is not simply how such a court came to exist, but why it has worked. Legal history is full of failed transplants, systems of law imported into a foreign environment that either withered or were quietly ignored. The DIFC Courts did neither. They have grown from a handful of cases at inception into a busy commercial judiciary handling disputes worth billions of dirhams, and their model has been copied across the Gulf and into Central Asia and beyond. Understanding that #success, and its limits, tells us something important about the movement of law across legal families, the design of international commercial courts, and the way a single emirate used #jurisdiction as an instrument of economic policy. The article proceeds in nine further parts. Part two sets out the constitutional and legislative architecture that allowed Dubai to carve out a zone exempt from federal civil and commercial law. Part three describes the institutional design of the Courts. Part four analyses the common-law framework and the theory of #legal_transplant, including the unresolved question of how far English doctrine actually applies. Part five examines the two forms of jurisdictional innovation: opt-in and conduit. Part six turns to the recurring conflict between the DIFC Courts and the onshore Dubai Courts, and the successive tribunals created to manage it. Part seven addresses arbitration and the sudden 2021 restructuring that abolished the DIFC-LCIA. Part eight offers an empirical snapshot of the Courts in practice. Part nine places the DIFC in comparative perspective alongside its rivals and imitators. Part ten confronts the main criticisms. The conclusion assesses whether the experiment can be called a success and what conditions sustain it. 2. The Constitutional and Legal Architecture: Building a Common Law Island A court that applies foreign law inside a sovereign state does not appear by accident. It requires legal permission at the highest level, and in the UAE that permission was engineered through a careful sequence of constitutional and statutory steps. The foundation is Article 121 of the UAE Constitution. In its original form, the Constitution allocated legislative competence between the federation and the individual emirates. In 2004, Article 121 was amended to give the federation the power to legislate for the creation of #financial_free_zones and, critically, to exempt those zones from the application of certain federal laws. This amendment is the constitutional hinge on which the entire DIFC project turns. Without it, no emirate could have lawfully displaced the federal Civil Code, the federal Commercial Transactions Law, and the federal law of civil procedure within a piece of its own territory. With the constitutional path cleared, the federal legislature passed Federal Law No. 8 of 2004 on Financial Free Zones. This statute allows a financial free zone to be created in any emirate by federal decree, and it exempts such zones and the financial activities conducted within them from all federal civil and commercial laws. Federal criminal law, including anti-money-laundering legislation, continues to apply. The effect is precise and limited: the zone is freed from federal private law but not from the criminal authority of the state. This is why the DIFC Courts hear only civil and commercial matters, while all criminal cases remain with the Dubai Public Prosecution and the onshore Dubai Courts. Federal Decree No. 35 of 2004 then established the DIFC as a financial free zone in the emirate of Dubai. At the level of the emirate, Dubai Law No. 9 of 2004 established the Centre itself and recognised its financial and administrative independence, while Dubai Law No. 12 of 2004, usually called the Judicial Authority Law, created the DIFC Courts, defined their jurisdiction, and provided for the independent administration of justice within the Centre. A separate instrument, DIFC Law No. 10 of 2004, set out the powers, procedures, and administration of the Courts. Dubai Law No. 9 of 2004 was later replaced by Dubai Law No. 5 of 2021, which restated the Centre's independence and reorganised its constituent bodies. Three bodies emerged from this framework, and it is worth keeping them distinct. The DIFC Authority operates and develops the Centre. The Dubai Financial Services Authority, or #DFSA, is the independent financial regulator. The DIFC Courts are the independent judiciary. The regulator and the judiciary are separate; a firm may be supervised by the DFSA and, quite separately, sue or be sued in the Courts. This separation of the regulatory and adjudicatory functions is a deliberate feature of the design and a point of reassurance for foreign investors accustomed to that division elsewhere. The motivation behind this elaborate construction was economic rather than doctrinal. Dubai holds relatively modest hydrocarbon reserves compared with some of its neighbours, and its development strategy has long emphasised diversification into trade, logistics, tourism, and financial services. A credible, independent, common-law commercial court was a necessary piece of the infrastructure for a regional financial hub, because international banks, funds, and corporations are reluctant to base significant operations in a jurisdiction whose dispute-resolution system they do not understand or trust. The DIFC Courts were, in this sense, built to answer a commercial anxiety, namely the fear of the unfamiliar forum. By offering a court that looked and worked like the courts of the financial centres those institutions already knew, Dubai lowered the perceived legal risk of doing business in the emirate and, with it, the risk premium attached to locating there. The genius, and the risk, of this architecture is that it creates a genuine second legal order inside one emirate. As Dimitropoulos (2021) argues, financial free zones of this kind are best understood as a form of adjudicatory #unilateralism, in which a state domesticates international commercial disputes by building a court that sits in the space between the national and the international and is based on English law. The DIFC did not sign a treaty or join an international tribunal. It simply legislated a common-law jurisdiction into being and invited the world to use it. The autonomy is real, but so is the tension it creates with the surrounding civil-law system, a tension that runs through the rest of this article. 3. Institutional Design of the DIFC Courts The DIFC Courts were established in 2004 and began hearing cases in 2006. Their structure will be familiar to anyone who has practised in a common-law commercial court, which is precisely the point. At the base sits the Small Claims Tribunal, an institution designed for speed and accessibility. It hears claims below a monetary threshold, generally AED 100,000, with a higher ceiling for employment matters and the ability to hear larger claims where the parties agree in writing. Proceedings are informal, and the Tribunal has become a heavily used forum for employment and lower-value commercial claims. Above it, the Court of First Instance hears larger civil and #commercial_disputes, sitting as a single judge. The Court of Appeal, composed of at least three judges including the Chief Justice, hears appeals from the Court of First Instance and may also interpret DIFC legislation on a referred question. Its decisions are final; there is no further appeal, which gives the DIFC system a self-contained quality unusual for a court embedded in a larger federal state. Over time the Courts added specialist divisions to match the disputes they were attracting. A Technology and Construction Division handles complex engineering and infrastructure claims. A Digital Economy Court, and later work on digital assets, positioned the Courts to hear disputes over cryptocurrency and financial technology, an area in which the DIFC has been an early mover. An Arbitration Division deals with applications connected to arbitral proceedings, including recognition, enforcement, and challenges to awards. The most distinctive institutional feature is the #bench. From the outset the Courts recruited senior judges from established common-law jurisdictions. The current Chief Justice, appointed with effect from September 2024, is Justice Wayne Martin, the former Chief Justice of Western Australia. During 2024 the bench was reinforced by the appointment of Justice Thomas Bathurst of Australia, Justice Sapna Jhangiani KC of Singapore, and Justice Roger Stewart KC of England and Wales. The Deputy Chief Justice, Justice Ali Al Madhani, is Emirati, and the Courts have made a point of appointing UAE-national judges alongside the international bench, both to build local capacity and to answer the criticism that the institution is simply an offshore British court. The quality and reputation of these judges is not incidental. Gu (2022) identifies the calibre of the bench as one of the principal reasons the DIFC Court came to dominate the caseload among the three Gulf international commercial courts, ahead of its counterparts in Abu Dhabi and Doha. Procedure follows the English model. The Rules of the DIFC Courts track the English Civil Procedure Rules, with case management, disclosure, witness statements, and costs that shift to the losing party. Legal representation is drawn from a register of practitioners, and the register has grown steadily, reaching well over eight hundred registered practitioners and nearly two hundred registered law firms. English is the language of the Court, of the pleadings, and of the judgments, which are published and reasoned in the common-law style. For a party used to litigating in London, Sydney, or Singapore, the DIFC feels immediately legible, and that legibility is a core part of the product the Centre is selling. 4. The Common Law Framework and the Question of Legal Transplant To call the DIFC a common-law jurisdiction is accurate but incomplete. The more interesting question, and the one that has generated the most #jurisprudence, is exactly how English #common_law enters DIFC law and how far it governs. The theory of legal transplant, associated with comparative lawyers who study the movement of legal rules between systems, warns that borrowed law often behaves differently in its new home. A rule can be copied on paper yet fail in practice because the surrounding institutions, professions, and habits of mind are absent. This is sometimes called the transplant effect. Bantekas (2023), writing about the closely related Abu Dhabi Global Market, argues that the Gulf experiments have largely avoided this fate because the transplant was accompanied by the institutional structures needed to support it: expert specialist courts, a continuing dialogue between judges across common-law jurisdictions, and the direct importation of experienced common-law judges and practitioners. The law did not arrive alone. It arrived with the people and the procedures that make it live. Yet the DIFC statutes themselves are surprisingly quiet about English law. As Bantekas (2023) notes, DIFC legislation refers only fleetingly to the common law or to the law of England and Wales. The Centre operates under its own statutes, its own Contract Law, its own Law of Obligations, its own Arbitration Law, and so on. These statutes were drafted in a common-law idiom and often codify common-law principles, but they are, formally, DIFC statutes. This raised a practical question that took years to resolve: what happens when a party asserts a common-law cause of action, defence, or remedy that the DIFC statutes do not expressly provide? For much of the Courts' early life, practitioners assumed that DIFC statutory law was backstopped by English common law. On this view, if a doctrine or remedy existed at English common law and was not excluded by DIFC statute, a party could rely on it. That assumption was challenged directly in The Industrial Group Ltd v Hamid, decided by the DIFC Court of Appeal in 2022. The Court held that DIFC law is essentially statutory, and that judges and arbitrators cannot assume the existence of a common-law or equitable doctrine, cause of action, defence, or remedy unless a DIFC statute provides for it. On the facts, claims for malicious prosecution and abuse of process failed because no DIFC statute created them. The decision unsettled practitioners who had relied on the backstop theory, and it reframed the DIFC as a codified system that borrows the reasoning style of the common law rather than its entire body of unwritten doctrine. The Court of Appeal was careful about the mechanism. Where DIFC statutes identify principles that originate in other jurisdictions, it is legitimate to look to those jurisdictions to determine the content and appropriate development of those principles. But judges are not free to import any attractive common-law development from anywhere simply because it exists elsewhere, and gaps in the legislation should be filled by statute rather than by what the Court called impermissible judicial legislation. Subsequent amendments to the DIFC's framework on the role of common law refined this position further. The trajectory, as commentators have observed, is that the DIFC is moving from deference to English precedent toward the development of its own distinct #common_law tradition, shaped by the region's commercial needs rather than merely copied from London. Underlying this debate is a mechanism unique to the DIFC known informally as the waterfall. The Law on the Application of Civil and Commercial Laws in the DIFC sets out an ordered cascade that a judge must follow when deciding which law governs a matter. The DIFC's own statutes come first. Failing an applicable DIFC statute, the judge turns to the law of any jurisdiction expressly chosen in the relevant DIFC legislation, then to any law the parties themselves have agreed upon, then to the law of the jurisdiction most closely connected to the facts, and finally, as a residual default, to the laws of England and Wales. English law therefore occupies the bottom of the cascade rather than the top. It is a safety net that catches whatever the higher tiers do not resolve, which is a very different thing from a general presumption that English #common_law fills every gap. The Industrial Group decision can be read as an insistence that the waterfall be respected as written, so that English doctrine enters only through the channels the legislation actually provides. Later amendments clarified how far judges may look to common-law developments when giving content to principles that DIFC statutes have adopted, confirming that the role of the common law is real but structured rather than open-ended. This maturation matters for the transplant story. A transplant that only ever mirrors the donor system remains dependent. A transplant that begins to generate its own doctrine has taken root. The DIFC's willingness to decide, in cases such as Industrial Group, that it is a statutory system with its own logic is evidence that the borrowed law has become genuinely local, even as it remains recognisably common-law in method. 5. Jurisdictional Innovation: From Territorial Enclave to Global Forum If the constitutional architecture explains how the DIFC Courts could exist, the story of their #jurisdiction explains why they became significant. Two innovations stand out, and both involved expanding the reach of the Courts far beyond what their founders originally contemplated. 5.1 The opt-in jurisdiction When the Courts opened, their jurisdiction was tied to the DIFC itself. They heard disputes involving DIFC-registered entities, contracts concluded or performed within the DIFC, and matters otherwise connected to the Centre. This is the classic model of a special-zone court: a court for the zone. Its reach was geographic. That changed on 31 October 2011, with Dubai Law No. 16 of 2011, which amended the Judicial Authority Law. The amendment introduced what is now the defining feature of the DIFC model, the #opt_in_jurisdiction. Under the new Article 5A(2), the Courts gained the power to hear any civil or commercial dispute, whether or not it had any connection to the DIFC, provided the parties agreed in writing to submit to the Courts' jurisdiction. In a single stroke, a court built for a financial district became a court available to the world. Two companies with no presence in Dubai, contracting over goods that would never touch the emirate, could nonetheless choose the DIFC Courts as their forum in exactly the way parties choose the courts of London or New York, or choose to arbitrate. This was a bold act of legal marketing dressed as procedural reform. It repositioned the DIFC as a neutral, English-language, common-law venue for international #commercial_disputes, competing directly with arbitration and with established litigation centres. Consent became the engine of jurisdiction. Where the parties opt in, the DIFC Courts' jurisdiction is exclusive of the Dubai Courts, subject to the parties' freedom to agree otherwise. As Dahdal (2023) observes, the founding instruments drew a clear demarcation of jurisdiction, but the common-law courts of the Gulf have read their empowering provisions expansively, and the opt-in reform is the clearest example of that expansion being written directly into the law. The opt-in power is not unlimited, and the Courts have had to police its boundaries. Because jurisdiction rests on consent, the validity and clarity of the written agreement matters, and the Courts have examined whether particular clauses genuinely conferred DIFC jurisdiction or merely referred to Dubai in a general way. The distinction between the DIFC as a chosen forum and Dubai as a wider geographic location has proved important in practice, since a loose reference to Dubai does not by itself select the DIFC. This policing function is significant because opt-in jurisdiction, if read too generously, would allow the Courts to reach almost any dispute, deepening exactly the #conflict with the onshore courts that later tribunals were created to manage. By insisting on a clear and genuine agreement, the Courts keep the opt-in gateway wide enough to be commercially useful but narrow enough to remain defensible. 5.2 The conduit jurisdiction The second innovation was not written into any statute in so many words. It was developed by the Courts themselves through interpretation, and it became known as the #conduit_jurisdiction. The mechanism rests on Article 7 of the Judicial Authority Law, under which the DIFC Courts and the onshore Dubai Courts are required to enforce each other's judgments and ratified awards without reviewing the merits. The execution judge at the Dubai Courts, in particular, has no power to reopen the substance of a DIFC judgment, order, or award. This mutual-recognition regime was intended to make the two-court system work smoothly within Dubai. The DIFC Courts, however, saw in it a wider possibility. The reasoning runs as follows. Suppose a party holds a foreign judgment, or an arbitral award made outside Dubai, against a debtor with assets in the emirate. Ordinarily that party would seek enforcement directly in the onshore Dubai Courts, where the process is in Arabic, follows civil-law procedure, and may involve review that the creditor would prefer to avoid. Instead, the creditor could ask the DIFC Courts to recognise and enforce the foreign judgment or award first, obtaining a DIFC judgment or enforcement order. That DIFC order could then be passported into the onshore Dubai Courts under Article 7, where the execution judge is barred from reviewing its merits. The DIFC thus functions as a conduit: a common-law gateway through which foreign judgments and awards flow into the wider UAE, and potentially beyond, since DIFC judgments enjoy recognition across the Gulf Cooperation Council under regional enforcement conventions. The Courts confirmed this function in a line of cases. In Banyan Tree Corporate Pte Ltd v Meydan Group LLC, the Courts accepted a role in enforcing arbitral awards even where the connection to the DIFC was the enforcement itself. In DNB Bank ASA v Gulf Eyadah Corporation, the Court of Appeal confirmed that a DIFC judgment recognising a foreign judgment could then be taken onward to the Dubai Courts for execution. Gu (2022) highlights a related and unusual feature of the DIFC's enforcement strategy, namely the historical practice by which a DIFC Court judgment could be converted into an arbitral award, a device that helped move judgments across borders in an era when the UAE was not a party to the Hague judgments framework but was bound by the New York Convention on arbitral awards. The conduit function transformed the commercial value of the DIFC. It made the Courts useful not only to parties who wanted to litigate a dispute from scratch, but to creditors anywhere who needed a reliable, merits-proof route to enforcement against UAE assets. It also, as the next part shows, provoked the sharpest #conflict with the onshore Dubai Courts, which did not universally welcome being turned into an execution arm for judgments they had never reviewed. 6. The Conflict of Jurisdiction: Managing the Two-Court System Autonomy has a cost. When you place two courts inside one emirate, one applying common law in English and the other applying civil law in Arabic, and you give litigants some freedom to choose between them, you create the conditions for overlap, competition, and conflict. Managing that #conflict of jurisdiction has been one of the central problems of the DIFC system, and the solutions have evolved considerably. The problem takes a predictable shape. A claimant sues in the DIFC Courts. The defendant, preferring the onshore forum, or simply wanting to delay, starts parallel proceedings in the Dubai Courts, and both courts accept jurisdiction. Now the same dispute is running in two courts that answer to different legal traditions and could reach conflicting judgments. Something has to decide which court prevails. For the first years, the theoretical answer lay in federal law. Under the Union Supreme Court framework, conflicts of jurisdiction between courts within an emirate could be referred to the federal Union Supreme Court. In practice this was cumbersome and largely untested for DIFC disputes. The Courts themselves developed a robust position, repeatedly holding that the federal rules on inter-emirate jurisdiction did not apply inside the DIFC, given the Centre's exemption from federal civil and commercial law. Opponents argued that jurisdictional rules were constitutional rather than merely commercial and therefore could not be disapplied so easily. The debate was never fully settled at the constitutional level, and Deputy Chief Justice Sir David Steel, in Standard Chartered Bank v Investment Group Private Limited, pointedly raised the question of whether an emirate-level decree could override a federal law. In June 2016, the Ruler of Dubai issued Decree No. 19 of 2016, establishing the Judicial Tribunal for the Dubai Courts and DIFC Courts, often called the Joint Judicial Committee. Its task was to resolve conflicts of jurisdiction and conflicts between judgments, and to determine which court should hear a matter and which judgment should be enforced. The Committee was composed of a majority of onshore Dubai judges, chaired by the President of the Dubai Court of Cassation with a casting vote, together with a smaller number of DIFC judges. Many commentators read the creation and composition of the Committee as a deliberate check on the DIFC Courts' expanding reach, and particularly on the conduit function that allowed foreign awards to bypass onshore review. The Committee's early decisions bore out that reading. In its first case, Daman Real Capital Partners Company LLC v Oger Dubai LLC, the Committee addressed a situation where an award seated in onshore Dubai was subject to annulment proceedings in the Dubai Courts while the successful party sought recognition in the DIFC. The general principle that emerged was that where the onshore courts, as the supervisory courts of the seat, were seised of an annulment application, the DIFC Courts should refrain from enforcing. This narrowed the conduit for domestically seated awards and raised concerns that a mere application to the Committee could be used to freeze DIFC proceedings and frustrate #enforcement. That concern crystallised around the automatic stay. Under Decree No. 19, an application to the Committee stayed proceedings in both courts pending a decision. Parties quickly learned to manufacture jurisdictional disputes purely to trigger the stay and delay enforcement. The DIFC Courts pushed back. In Lakhan v Lamia, decided by the DIFC Court of Appeal in 2021, the Court held that applications to the Committee must not be abusive and that a stay should be granted only where the Court was satisfied that a genuine conflict of jurisdiction had actually arisen. A mere application, the Court warned, does not automatically trigger a stay, because automatic suspension would encourage spurious applications designed to fabricate conflicts. This was a significant assertion of self-protection by the DIFC bench against tactical abuse. The framework was overhauled again in 2024. Decree No. 29 of 2024, effective in April 2024, abolished the Joint Judicial Committee and replaced it with a new body, generally referred to as the Conflict of Jurisdiction Tribunal. The new Tribunal has a wider remit than its predecessor. It resolves conflicts not only between the DIFC Courts and the Dubai Courts, but between the DIFC Courts and a broader category of judicial bodies in the emirate, including the Rental Disputes Centre and various specialist committees. The reform also introduced two notable features. First, a modest security deposit is required with an application, to be forfeited if the Tribunal upholds the opposing forum, a measure aimed at deterring frivolous or tactical referrals. Second, and more striking, the Tribunal's decisions are made binding judicial precedents for both the DIFC and the Dubai Courts. This importation of a doctrine of #precedent, familiar to common lawyers but novel within the civil-law tradition of the onshore courts, is itself a small transplant flowing in the opposite direction, from the DIFC's legal culture into the emirate's wider system. Early practice under the new Tribunal suggests a more balanced approach. Since it began operating, the Tribunal has decided a series of applications, dismissing several for lack of a genuine conflict and dividing the remainder between the two court systems rather than reflexively favouring either. The DIFC Courts, for their part, have continued to insist that a stay is not automatic and have applied the reasoning of Lakhan v Lamia to the new decree. At the same time, the Courts have shown a willingness to defer where deference is genuinely appropriate. In Union Insurance v IPMR, for example, the DIFC Court granted an indefinite stay in favour of an onshore Sharjah judgment even though it had earlier confirmed its own jurisdiction over the dispute, a gesture of comity that suggests the relationship is maturing from open competition toward coordination. The evolution from the 2016 Committee to the 2024 Tribunal is best read not as a defeat for the DIFC but as the slow construction of a workable interface between two legal orders that were never designed to share territory. The friction was real, and at times the conduit was curbed, but the system has not collapsed into paralysis. Dahdal (2023) frames this as the central governance challenge of common-law commercial courts in Arab civil-law states: how to expand jurisdiction attractively while keeping the peace with the domestic judiciary that ultimately shares the same sovereign. 7. The Arbitration Dimension and the 2021 Restructuring No account of the DIFC would be complete without arbitration, which sits alongside litigation as the other pillar of Dubai's dispute-resolution offering and which experienced a dramatic upheaval in 2021. For years the flagship arbitral institution associated with the Centre was the DIFC-LCIA Arbitration Centre, established in 2008 as a joint venture between a DIFC body and the London Court of International Arbitration. The DIFC-LCIA became popular across the Middle East, valued for administering arbitrations under a version of the well-regarded LCIA rules, and many contracts across the region specified DIFC-LCIA #arbitration in their dispute-resolution clauses. Separately, the DIFC could serve as the legal seat of an arbitration, in which case the DIFC Arbitration Law of 2008 governed and the DIFC Courts exercised supervisory jurisdiction over the award. That settled picture was overturned by Dubai Decree No. 34 of 2021, issued in September 2021. The Decree abolished two of Dubai's arbitral institutions, the Emirates Maritime Arbitration Centre and the DIFC Arbitration Institute that operated the DIFC-LCIA, and transferred their assets, cases, and functions to the Dubai International Arbitration Centre, or #DIAC. The change was abrupt. Practitioners described it as arriving with little warning and without a detailed transition plan, and ongoing cases faced a period of uncertainty while the new arrangements were worked out. Article 6 of the Decree declared that existing arbitration agreements referring disputes to the abolished centres remained valid and effective, but provided that DIAC would replace those centres in administering the disputes, unless the parties agreed otherwise. The controversy that followed went to the heart of arbitration theory. Party autonomy, the principle that parties are bound to and by the forum they actually chose, is the foundation of the whole system. Critics argued that unilaterally redirecting a DIFC-LCIA arbitration to DIAC, an institution with different rules and administration, substituted a forum the parties had never selected and therefore undermined that autonomy. The dispute produced a striking divergence between courts. In the United States, litigation arising from the Baker Hughes dispute saw a Louisiana court refuse to compel arbitration before DIAC, reasoning that neither the Dubai government nor a foreign court could unilaterally rewrite the parties' chosen forum, and that DIAC was simply not the same forum as the DIFC-LCIA. That ruling was later revisited on appeal. The Singapore courts expressed similar reservations about the difference between the two institutions' rules. Within the UAE, by contrast, the courts moved to uphold the Decree. The Abu Dhabi courts confirmed that DIFC-LCIA agreements were valid and enforceable under the new regime. Most importantly for this study, the DIFC Court of First Instance addressed the question directly in Narciso v Nash in 2024. Justice Michael Black KC held that, because Decree No. 34 formed part of DIFC law by operation of Dubai Law No. 5 of 2021, the Court was bound to give it effect; that the Decree did not render the arbitration agreement null, void, or inoperative but rather preserved the parties' bargain while allowing them to resort to DIAC or another centre if they wished; and that an anti-suit injunction restraining a party from pursuing parallel court proceedings in breach of the arbitration agreement was justified. The decision provided reassurance within the UAE that legacy DIFC-LCIA clauses would be honoured, even as questions lingered about how foreign courts would treat them. The episode is instructive for the larger argument of this article in two ways. First, it shows the limits of #autonomy. The DIFC's arbitral institution was not beyond the reach of the emirate's executive; a decree could and did dissolve it overnight. The DIFC Courts survived the change with their litigation jurisdiction intact, but the arbitration landscape around them was reordered by a single instrument. Second, it shows the DIFC Courts performing their characteristic function of stabilising expectations. Faced with a disruptive change and conflicting signals from courts abroad, the DIFC bench produced a reasoned, common-law judgment that told the market how the Decree would be applied within the Centre. That capacity to generate predictability, even from turbulence, is precisely the service the DIFC exists to provide. 8. Empirical Case Study: The DIFC Courts in Practice Theory and doctrine describe what the Courts may do. The caseload describes what they actually do, and here the picture is one of steady, and recently sharp, growth. The trajectory is easy to state. In 2017 the Court of First Instance recorded a modest number of cases, on the order of a few dozen. By 2022 the Courts as a whole handled several hundred cases, and 2023 was described by the Courts as a record year, with close to a thousand claims filed across all divisions. The figures for 2024 show both rising volume and rising value. The total value of claims across all divisions reached roughly AED 7.7 billion, with overall case numbers up around ten per cent. The Court of First Instance recorded an average claim value above AED 100 million, a figure that signals the seriousness of the disputes now being brought. The Arbitration Division alone accounted for claims of about AED 6.8 billion. The Small Claims Tribunal processed several hundred matters, up around thirteen per cent, and enforcement claims rose again year on year. By late 2025 the Courts were reporting cumulative case values well above AED 17 billion, and the institution marked its twentieth anniversary in 2024 as an established fixture rather than an experiment. The composition of the caseload is as telling as its size. Claims before the Court of First Instance span banking and finance, real estate, construction, manufacturing, retail, intellectual property, and pharmaceuticals, with disputes over breach of contract, employment, and enforcement recurring throughout. This is the profile of a general commercial court serving a diversified economy, not a narrow regulatory tribunal. The parties come from many nationalities, consistent with the Courts' self-image as a neutral international forum rather than a purely local one. Three features of the Courts' practice deserve particular attention as illustrations of institutional #innovation. The first is the aggressive move to digital operation. The Courts became fully paperless, and by 2024 they were conducting the overwhelming majority of hearings remotely, with figures around ninety-nine per cent, and issuing orders and judgments in digital form in very large numbers. The digital model was accelerated by the pandemic but has outlasted it, and it lowers the cost and friction of using a court located far from many of its users. The Chief Justice has since signalled a rebalancing, with longer trials to be heard in person while shorter matters remain remote. The second is the Courts' early engagement with technology as subject matter, not merely as method. The DIFC established a Digital Economy Court and heard one of the region's first cryptocurrency disputes, in the Huobi litigation, an area where reported decisions remain rare anywhere in the world. The Courts also issued practice guidance on the use of large language models and generative artificial intelligence in proceedings, permitting such tools subject to conditions on accuracy, disclosure, and confidentiality. For a court whose value proposition is predictability, being early and explicit about novel technology is a way of extending that predictability into new commercial frontiers. The third is the layer of ancillary services that has grown around the core judicial function. The Wills Service, created in 2014 and launched in 2015, allows non-Muslims resident in the UAE to register wills and pass on assets under common-law principles of testamentary freedom rather than the default rules that would otherwise apply, and its registrations have risen year on year. More recently the Courts launched a Mediation Centre, giving parties a structured, court-supervised route to settlement before or during litigation, with approved settlement agreements recognised as enforceable instruments under the new Judicial Authority Law of 2025. Alongside this, Practice Direction No. 1 of 2025 addressed access to justice in employment disputes. These additions show an institution behaving less like a static courthouse and more like a full-service #dispute_resolution ecosystem, continually adding products to widen its user base. Two further practical features shape how the Courts are used. The first is the costs regime. Following the English approach, costs generally follow the event, so that an unsuccessful party is ordinarily ordered to contribute to the winner's legal expenses. This concentrates the minds of litigants and discourages weak claims, and it stands in contrast to the more limited costs recovery historically available in some onshore proceedings. The Courts periodically revise their fee guidance and their guideline rates for the assessment of costs. The second feature is the character of the bar that appears before them. Because proceedings are in English and follow common-law procedure, the advocates are frequently international practitioners, and the register of firms and practitioners has grown accordingly. Local advocates are increasingly developing the relevant skills, and the presence of Emirati judges and practitioners has grown over time, but the international composition of the profession before the Courts remains one of their defining, and to some critics one of their more contested, features. Taken together, the empirical record supports the article's central claim of qualified success. The Courts are busy, the disputes are large and varied, the users are international, and the institution keeps building. Basedow (2024) situates this kind of growth within a broader global phenomenon in which elite international law firms actively promote and populate international commercial courts, which helps explain both the DIFC's rapid rise and the international character of its bar. The numbers are not merely vanity metrics; they reflect a market decision by sophisticated parties to entrust serious money to a common-law court sitting in the Gulf. 9. Comparative Perspective: The DIFC Among Its Peers and Imitators The DIFC was not the last word in this field. It was closer to the first, and its model has been studied, adapted, and challenged by a growing family of international commercial courts. Placing the DIFC in that comparative frame sharpens our understanding of what is distinctive about it. The closest comparison is its domestic sibling, the Abu Dhabi Global Market, established in the UAE capital and operational in the following decade. The ADGM took the transplant one step further than the DIFC. Rather than enacting its own suite of statutes drafted in a common-law style, the ADGM adopted English law directly, applying the common law and specified English statutes as they stand through its Application of English Law Regulations. Bantekas (2023) treats the ADGM as an instant and almost complete transplant of English private law, made viable by a bench of experienced common-law judges. The contrast with the DIFC is instructive. The DIFC codified common-law reasoning into local statutes and, as Industrial Group v Hamid showed, must decide case by case how far unwritten English doctrine reaches. The ADGM, by importing English law wholesale, reduces that uncertainty but arguably develops less of its own distinct identity. The DIFC's Chief Justice has expressly framed the DIFC as drawing on a range of business-friendly common-law jurisdictions, including Singapore, Australia, India, and New Zealand, alongside England, rather than tethering itself to any single donor system. In Qatar, the Qatar Financial Centre and its associated court, the Qatar International Court and Dispute Resolution Centre, offer a variant model. As Dahdal (2023) explains, the Qatari financial centre operates on an onshore basis, with participants able to operate throughout the state rather than being confined to a geographic enclave, and its relationship with the local judiciary has been organised differently from the DIFC's more elaborate interface with the Dubai Courts. Gu (2022) records that the DIFC has consistently carried the largest caseload among the three Gulf courts, a lead attributable to its earlier start, its bench, and its aggressive jurisdictional innovations. Beyond the Gulf, the model has travelled. Kazakhstan established the Astana International Financial Centre and its court, drawing explicitly on the experience and support of the DIFC and the Qatari court, and applying common-law principles in an English-language forum in Central Asia. In Asia and Europe, related but distinct institutions emerged, including the Singapore International Commercial Court, the China International Commercial Court, and the Netherlands Commercial Court. The Singapore court is often cited as the contrasting success story on the governance question that has troubled the DIFC. As a specialised division of Singapore's own Supreme Court, it enjoys a harmonious relationship with the surrounding judiciary and does not face the turf wars that the DIFC has had to manage with the onshore Dubai Courts. The comparative literature collected by Brekoulakis and Dimitropoulos (2022) treats these courts as a genuinely new species of adjudication, domestic in legal form but international in design and caseload, and it identifies the management of the relationship with ordinary national courts as a defining variable in whether such courts flourish or stumble. The comparative lesson for the DIFC is double-edged. On one hand, the fact that its model was copied so widely is powerful evidence of its success; institutions are imitated when they work. On the other hand, the Singapore comparison shows that the DIFC chose, or was forced into, a harder path. By sitting outside the national judiciary rather than within it, the DIFC gained autonomy and a clean common-law identity, but it also inherited a permanent problem of coexistence that a court embedded in its host system, like Singapore's, largely avoids. Synthesising the comparative picture, several success factors recur across the institutions that have flourished. A clear constitutional foundation removes doubt about a court's legal right to exist. A bench of respected judges supplies immediate credibility that a new institution cannot otherwise buy. Procedure and language familiar to international users lower the cost of entry. Jurisdiction grounded in consent lets a court grow beyond its borders without displacing anyone against their will. And a functioning, if imperfect, interface with the domestic judiciary keeps the experiment from tearing at the fabric of the host state. The DIFC assembled all five of these ingredients earlier and more completely than most of its peers, which is why it continues to be studied as the reference model even where later courts have refined particular elements of the design. 10. Challenges, Critiques, and the Legitimacy Question A case study that only catalogued achievements would be incomplete and unpersuasive. The DIFC model attracts serious criticism, and taking that criticism seriously is part of assessing whether the experiment truly succeeds. The first challenge is the coexistence problem examined above. The relationship with the onshore Dubai Courts has been, at various points, competitive and unstable. The succession of tribunals, from the 2016 Committee to the 2024 Conflict of Jurisdiction Tribunal, is evidence both that the problem is real and that it required repeated intervention by the Ruler to manage. The conduit function, which did so much to make the DIFC commercially valuable, was precisely the feature that provoked the strongest onshore reaction, and it was curbed for domestically seated awards. A model whose most powerful innovation generates recurring institutional conflict cannot be called an unqualified success; it is a success that must be continually renegotiated. The second challenge is the limit on autonomy exposed by the 2021 arbitration restructuring. The abolition of the DIFC-LCIA by executive decree, without extensive consultation, reminded the market that the Centre's institutions exist at the sufferance of the emirate's executive. For an institution whose entire appeal rests on stability and predictability, an abrupt structural change of that magnitude is a reputational cost, even though the DIFC Courts themselves subsequently worked to restore certainty through reasoned judgments. The third and deepest challenge is the legitimacy critique, sometimes framed in the academic literature under the uncomfortable label of whitewashing. The concern, voiced in the scholarship on international commercial courts, is that a gleaming Western-style, English-language court staffed by respected foreign judges may function as reputational cover, reassuring international investors while the broader national legal system continues to have flaws that those same investors would find troubling. The worry is sharpened by the boundary between civil and criminal jurisdiction. The DIFC Courts hear only civil and commercial matters; criminal jurisdiction remains entirely with the onshore system. A party comfortable litigating a contract in the DIFC could still find that a related allegation, for example of fraud, is dealt with by a very different forum under very different rules. Critics argue that the existence of a first-class commercial court should not lead investors to assume that all of their legal risks are governed by first-class common-law protections. This is a genuine tension in the DIFC model, and it is not fully answerable by pointing to the Courts' commercial competence, because the critique is precisely that commercial competence can obscure other gaps. There are reasonable responses to each critique, and honest analysis should acknowledge them. On coexistence, the trend since Lakhan v Lamia and the 2024 reforms is toward a more balanced and rule-bound interface, with the DIFC Courts asserting anti-abuse principles and, in cases like Union Insurance v IPMR, extending comity to onshore judgments where appropriate. On autonomy, the DIFC Courts weathered the 2021 restructuring and used it as an occasion to demonstrate their stabilising function. On legitimacy, the appointment of Emirati judges, the development of a distinct DIFC jurisprudence rather than mere mimicry of English law, and the sheer transparency of published, reasoned judgments all cut against the whitewashing charge. The willingness of the Courts, in Industrial Group v Hamid, to reach a result that disappointed practitioners rather than simply importing a convenient English remedy is itself a marker of judicial independence. The honest conclusion is that the critiques identify real vulnerabilities rather than fatal flaws. The DIFC's #success is conditional. It depends on the Courts continuing to behave as genuinely independent adjudicators, on the interface with the onshore system remaining workable, and on the emirate continuing to see value in preserving the autonomy that makes the whole thing credible. Remove any of those conditions and the island could be reclaimed by the sea. 11. Conclusion The DIFC Courts began as an audacious proposition: that a small emirate could legislate a common-law jurisdiction into existence inside a civil-law federation, staff it with imported judges, and persuade the international commercial world to use it. Two decades on, the proposition has been substantially vindicated. The Courts are busy, respected, and continually expanding their reach, their services, and their subject-matter competence, from construction and banking to cryptocurrency and artificial intelligence in the courtroom. The model has been copied from Abu Dhabi to Astana, the surest sign that it works. The mechanism of that success was #jurisdictional_innovation. The opt-in reform of 2011 turned a court for a district into a forum for the world, grounding jurisdiction in consent rather than geography. The judicially developed conduit function turned the Courts into a common-law gateway for the enforcement of foreign judgments and awards. Together these innovations converted a modest zone court into a strategic instrument of economic policy, offering global parties a familiar, predictable, English-language venue in a region where such venues had been scarce. Yet the case study also shows the price of autonomy. Building a second legal order inside one emirate created a durable problem of coexistence with the onshore courts, one that has required repeated institutional redesign and that curbed the conduit function at its most ambitious. The abolition of the DIFC-LCIA in 2021 revealed that the Centre's institutions remain subject to executive power. And the legitimacy critique reminds us that a first-rate commercial court does not by itself cure the wider legal environment in which it sits. The most accurate verdict is therefore one of qualified and conditional success. The DIFC Courts have integrated English #common_law into the Middle East more effectively than almost anyone predicted, not by copying England wholesale, but by codifying common-law reasoning into local statutes, recruiting a credible bench, innovating on jurisdiction, and gradually growing a distinct body of DIFC law. Their future depends on holding the qualities that made them credible in the first place: independence, transparency, comity with their civil-law neighbour, and the steady development of their own jurisprudence. If they keep those qualities, the common-law island will remain not a curiosity but a genuine landmark in the comparative study of how law crosses the boundaries between #legal_traditions. #DIFC_Courts #common_law #civil_law #legal_transplant #jurisdictional_innovation #Dubai #UAE #international_commercial_courts #conduit_jurisdiction #opt_in_jurisdiction #arbitration #enforcement #comparative_law #rule_of_law #dispute_resolution References Bantekas, I. (2023). The Abu Dhabi Global Market and its transplant of English private law. The Chinese Journal of Comparative Law, 11(2), cxad007. https://doi.org/10.1093/cjcl/cxad007 Basedow, R. (2024). Pushing the bar: elite law firms and the rise of international commercial courts in the world economy. Review of International Political Economy, 31(6), 1764. Brekoulakis, S., and Dimitropoulos, G. (Eds.). (2022). International commercial courts: The future of transnational adjudication. Cambridge University Press. Chaisse, J., and Dimitropoulos, G. (2021). Special economic zones in international economic law: Towards unilateral economic law. Journal of International Economic Law, 24(2), 229. https://doi.org/10.1093/jiel/jgab025 Dahdal, A. (2023). Navigating judicial conflict amidst jurisdictional expansion: Common law commercial courts in Arab civil law countries. The Chinese Journal of Comparative Law, 11(2), cxad009. https://doi.org/10.1093/cjcl/cxad009 Dimitropoulos, G. (2021). International commercial courts in the modern law of nature: Adjudicatory unilateralism in special economic zones. Journal of International Economic Law, 24(2), 361. Gu, W. (2022). The global rise of international commercial courts: Typology and power dynamics. Chicago Journal of International Law, 22(2). Menon, S., and Reyes, A. (Eds.). (2022). Transnational commercial disputes in an age of globalism and pandemic. Hart Publishing. Antonopoulou, G. (2022). Procedure before international commercial and ordinary courts: A comparative perspective. In S. Brekoulakis and G. Dimitropoulos (Eds.), International commercial courts: The future of transnational adjudication. Cambridge University Press.

A Study of the Treaty Governing the Enforcement of Civil and Commercial Court Judgments Across the EU and EFTA States Abstract When a court in one country decides a dispute, its judgment usually stops at the border. A winning party who wants to collect money or enforce an order in a different country often has to start again, argue the case in a foreign court, and hope that local law agrees. The #Lugano_Convention was built to solve exactly this problem for Europe. This article explains how the 2007 Lugano Convention allocates cross-border jurisdiction and how it makes #civil_and_commercial court judgments travel between the European Union and the three participating states of the #EFTA family, namely Iceland, Norway, and Switzerland. It sets out the treaty's history, its core rules on which court may hear a case, the way it handles two courts hearing the same dispute at once, and the process by which one state's judgment is recognised and enforced in another. It then studies the treaty's most debated recent chapter, the refusal of the European Union to let the United Kingdom rejoin after #Brexit, and compares the Lugano system with the Brussels regime inside the EU and with the newer #Hague_2019 framework now filling part of the gap. The article is written for students, and it uses plain language while keeping the structure of a formal journal article. The central argument is that Lugano remains a strong and coherent model of #judicial_cooperation, but that its future depends on political trust as much as on legal technique. Keywords: Lugano Convention; cross-border jurisdiction; recognition and enforcement of judgments; EFTA; Brussels I regime; private international law; Brexit; Hague Conventions. 1. Introduction Modern life crosses borders far more easily than the law does. A supplier in Zurich sells parts to a factory in Lyon. A family in Oslo buys a holiday home in Spain. A software company in Reykjavik licenses its product to a client in Germany. Each of these ordinary transactions can turn into a dispute, and when it does, two hard questions appear before anyone even argues about the money. The first question is which country's courts are allowed to decide the case. The second is whether a judgment won in one country will actually count for anything in another. Without clear answers, the same fight can be run twice in two different states, with two different results, and the winner in one place may find their victory worthless somewhere else. The #Lugano_Convention exists to give predictable answers to both questions across a defined group of European states. Its full name is the Convention on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, and the version in force today was signed in the Swiss city of Lugano on 30 October 2007. It binds the European Union, Denmark as a separate party, and the EFTA states of Iceland, Norway, and Switzerland. In simple terms, it is a treaty that tells courts in these states when they may take a case, when they must step aside, and how they must treat each other's decisions. What makes the Convention interesting is not only what it does but how closely it mirrors the internal rules of the European Union. For decades the EU has run its own system for cross-border jurisdiction between member states, first through the 1968 Brussels Convention and later through a series of regulations. Lugano is often described as a #parallel_regime, because it extends almost the same rules to a small circle of non-member states that share close economic ties with the Union (Hartley, 2023). This parallel design is the treaty's greatest strength and, as later parts of this article will show, also the source of its deepest political problems. This article has three broad goals. First, it explains the machinery of the Convention in a way that a law student or an interested reader can follow without prior training. Second, it places that machinery in context by comparing it to the two other main frameworks that now sit around it, the EU's own Brussels I Recast Regulation and the growing family of Hague Conventions. Third, it uses the recent story of the United Kingdom's failed attempt to rejoin the Convention after leaving the EU as a case study in how legal design and political judgment interact. The result, it is hoped, is a clear map of one of the most practical and least understood instruments in European #private_international_law. 2. The Road to Lugano: Historical Development The Lugano Convention did not appear out of nowhere. It is the outer ring of a legal system that grew up inside the European Community over more than half a century. Understanding that history makes the treaty far easier to read, because almost every article of Lugano has a twin somewhere in the EU's own law. The story begins with the 1968 Brussels Convention, an agreement among the original members of the European Economic Community. Its aim was to remove a serious obstacle to the common market. Traders were reluctant to do business across borders if they could not be sure that a court judgment obtained in one member state would be honoured in another. The Brussels Convention answered this worry by creating shared rules on jurisdiction and by making judgments move freely between the member states. It worked well, and over time the Court of Justice built a large body of case law explaining what its provisions meant. By the 1980s, several states outside the Community wanted access to the same benefits. These were mainly the members of the #EFTA, the European Free Trade Association, a group of countries that traded closely with the Community but chose not to join it. Because the Brussels Convention was tied to Community membership, a separate treaty was needed. That treaty was the first #Lugano_Convention, signed in 1988. It copied the Brussels Convention almost word for word and opened its rules to the EFTA states. Lawyers began to speak of the two instruments together as the Brussels and Lugano regime, a single approach expressed in two texts. The next change came from inside the Union. In 2001 the Community turned the Brussels Convention into a regulation, known as the Brussels I Regulation or Regulation 44/2001. A regulation is a stronger form of EU law than a convention, and this shift signalled that the field had become a settled part of Union policy. Once the internal rules had been modernised, the parallel Lugano text had to catch up. The result was the revised Convention signed in 2007, the instrument this article studies. The 2007 text was drafted to match the 2001 Regulation, so once again the internal and external systems moved in step (Kohler, 2022). The 2007 Convention entered into force gradually. It began to apply for the European Union, Denmark, and Norway on 1 January 2010, for Switzerland on 1 January 2011, and for Iceland on 1 May 2011. It is worth pausing on the list of parties, because it surprises many newcomers. Among the four EFTA members, only three joined Lugano. Liechtenstein, the fourth EFTA state, never became a party and remains outside the system. Denmark appears in its own right rather than simply as an EU member, because Denmark has a special #opt_out from the Union's area of justice policy and therefore signs certain instruments separately. There is one more twist that shapes everything that follows. In 2012 the European Union modernised its internal rules again, replacing the 2001 Regulation with the Brussels I Recast, formally Regulation 1215/2012, which applied from January 2015. The recast made important improvements, most famously abolishing a slow enforcement step and strengthening the protection of choice of court agreements. Lugano, however, was not updated at the same time. It still reflects the 2001 model. As a result, a gap opened between the EU's internal rules and the parallel Lugano rules, and that gap is now one of the most practically important features of the whole subject (Hartley, 2023). The parties have discussed restoring full parallelism, but they agreed to wait until the EU decides whether to revise the recast itself, so for now the two systems remain slightly out of step. 3. The Legal Architecture: Scope and Structure Before looking at individual rules, it helps to see the shape of the Convention as a whole. Lugano is what specialists call a #double_convention. This means it does two jobs in one treaty. It sets out direct rules on jurisdiction, telling a court whether it may hear a case, and it also sets out rules on #recognition_and_enforcement, telling a court how to treat a judgment that has already been given elsewhere. Many international treaties do only the second job. The fact that Lugano does both is what makes it powerful, because it manages a dispute from beginning to end rather than only at the finish line. The subject matter of the Convention is defined at the very start. It applies to #civil_and_commercial matters, a phrase that covers ordinary disputes about contracts, debts, property, and compensation for harm. It deliberately leaves out several areas that states prefer to keep under their own control or under specialised treaties. The excluded areas include revenue, customs, and administrative matters, along with the status and legal capacity of individuals, family property arising from marriage, wills and succession, insolvency, social security, and arbitration. The exclusion of #arbitration is especially important in practice, because arbitration has its own global framework and the two systems are kept apart on purpose. The body of the Convention is organised into clear parts, usually called titles. The first title fixes the scope, as just described. The second and longest title contains the rules on jurisdiction. The third title governs recognition and enforcement. Later titles deal with authentic instruments and court settlements, with the relationship between Lugano and other treaties, and with final matters such as accession and review. Attached to the Convention are three protocols. The most significant for our purposes is #Protocol_2, which is devoted to the uniform interpretation of the treaty and to a Standing Committee that watches over how it develops. One structural idea deserves special mention because it runs through the entire treaty. The Convention is built around the concept of #domicile. As a general rule, a person or company should be sued in the state where they are based, and that base is called their domicile. The Convention gives its own test for the domicile of a company, which is met if the company has its statutory seat, its central administration, or its principal place of business in a bound state. This choice of #domicile as the anchor is deliberate. It protects defendants, who can usually expect to be sued at home, and it gives claimants a reliable starting point for working out where to bring a claim (Hartley, 2023). Finally, it is worth stressing what kind of instrument Lugano is. It is a treaty in public international law, negotiated between sovereign parties, yet it functions almost like a shared civil procedure code for a region. That hybrid nature explains many of its features. Its interpretation must respect both the ordinary rules for reading treaties and the special goal of keeping the Brussels and Lugano texts aligned. Its future can be changed only by agreement among the parties, which is slow, and its reach can be widened only by admitting new members, which, as we shall see, requires everyone to agree. 4. Jurisdiction Under the Convention The heart of the Convention is its treatment of jurisdiction. Jurisdiction here means the authority of a particular country's courts to hear a particular dispute. The Convention does not leave this to chance or to each state's own preferences. Instead it lays down a ranked set of rules that together decide where a case belongs. These rules can be grouped into five families, and understanding the order between them is as important as understanding each one. 4.1 General jurisdiction: sue the defendant at home The default rule is simple and fair. A defendant who is #domiciled in a bound state must, as a starting point, be sued in the courts of that state. This is often called #general_jurisdiction. Its logic is that a person should not be dragged across Europe to defend a claim in a place they never expected. The rule applies regardless of the defendant's nationality, which means a French national living in Norway is treated the same as a Norwegian national living in Norway. The focus is on where the person actually lives or where the company is actually based, not on the passport they hold. There is an important limit to the reach of the Convention that follows directly from this rule. The treaty is chiefly concerned with defendants who are based inside a bound state. Where a defendant has no domicile in any bound state, the Convention generally hands the question back to the national law of the court that has been asked to decide, subject to a few strong exceptions such as the rules on exclusive jurisdiction and on choice of court, which apply whatever the defendant's domicile. This means that the shared system operates fully among insiders, while cases against outsiders are handled by a patchwork of local rules. The point is easy to overlook, but it explains a great deal about why leaving the Convention has such serious effects. A state that steps outside the circle does not simply lose a convenience. Its residents become the kind of outside defendants whose treatment depends on the varied national rules of every other state, which is exactly the fragmentation that the treaty was designed to remove. 4.2 Special jurisdiction: extra options for the claimant Alongside the home court, the Convention offers the claimant a set of alternative courts for certain types of claim. These are known as #special_jurisdiction rules, and they exist because sometimes another court has a closer link to the dispute. Two of them appear constantly in practice. In matters relating to a contract, the claimant may sue in the courts of the place where the obligation in question was to be performed, for example the place where goods were to be delivered or services provided. In matters relating to #tort, meaning a civil wrong such as negligence or defamation, the claimant may sue in the courts of the place where the harmful event happened. The Court of Justice has read this last phrase broadly, allowing suit both where the wrongful act took place and where the damage was felt. These special rules do not replace the home court. They sit beside it as extra choices. A claimant who has been injured by a defective product made in one state and sold in another may often pick between several courts. This flexibility is useful, but it also creates room for tactical behaviour, a theme we return to when discussing #forum_shopping. 4.3 Protective jurisdiction: shielding the weaker party Not everyone bargains from a position of equal strength. The Convention recognises this and creates a third family of rules that protect parties who are usually weaker in a relationship. These #protective_jurisdiction rules cover three groups. The first is the policyholder, insured person, or beneficiary in insurance contracts. The second is the #consumer who buys goods or services outside their trade or profession. The third is the individual employee in a contract of employment. The pattern for these groups is the same. The weaker party can usually sue the stronger party either at the stronger party's home or, more conveniently, in the weaker party's own home state. At the same time, the stronger party is generally limited to suing the weaker party only in the weaker party's home state. A company cannot easily drag a consumer to a distant court. These rules also limit the power of contract terms that try to remove the protection in advance, so a business cannot simply write the safeguard out of the small print. This part of the Convention shows that #cross_border_jurisdiction is not only about efficiency. It also reflects a policy choice to keep courts accessible to ordinary people. 4.4 Exclusive jurisdiction: cases only one country may decide For a small number of subjects, the Convention says that one and only one state's courts may decide, regardless of where the parties live and regardless of any agreement between them. This is #exclusive_jurisdiction, and it overrides almost everything else. The clearest example concerns rights in immovable property, such as ownership of land or buildings. A dispute about who owns a house in Italy belongs to the Italian courts, because that is where the property sits and where records and enforcement are managed. Other examples include the validity of a company's existence or decisions of its organs, the validity of entries in public registers, the registration or validity of patents and other registered intellectual property, and the enforcement of judgments. In each case the link between the subject and a particular state is so strong that shared decision-making would create chaos. 4.5 Choice of court: letting the parties decide The final family reflects a basic principle of commercial life, which is that businesses often prefer to decide in advance where any dispute will be heard. The Convention allows parties to make a choice of court agreement, sometimes called a prorogation of jurisdiction, in which they name the courts of a particular bound state. If the agreement meets the formal requirements, usually that it is in writing or evidenced in writing, the chosen court has jurisdiction and other courts should not interfere with exclusive subjects aside. This is where the gap between Lugano and the modern EU rules becomes sharp. Under the older model that Lugano still follows, a chosen court did not always get to act first. If one party ignored the agreement and rushed to sue in a different, slower court, the rules on parallel proceedings could force even the chosen court to wait. This tactic became so common that it earned a nickname, the #Italian_torpedo, because slow proceedings were sometimes launched to stall a case. The Court of Justice confirmed this harsh result in the well-known Gasser decision. The EU later fixed the problem in its recast Regulation by giving priority to the chosen court, but Lugano was not amended, so the older and weaker protection still applies within the Lugano circle (Hartley, 2023; Beaumont, 2021). Recent case law shows the rules are still developing. In early 2025 the Court of Justice ruled for the first time on the validity of so-called asymmetric jurisdiction clauses under both the recast Regulation and the 2007 Convention, confirming that such clauses are judged by the treaty's own shared rules rather than by each state's national law. This kind of ruling matters because it keeps the two parallel systems reading their choice of court provisions in the same way. 5. Parallel Proceedings: Lis Pendens and Related Actions One of the hardest problems in #cross_border_jurisdiction is that two courts can start hearing the same dispute at the same time. If both are allowed to run to the end, they may reach opposite conclusions, and the two states would then be stuck with irreconcilable judgments. The Convention tackles this with a mechanism known by its Latin name, #lis_pendens, which simply means a lawsuit already pending. The rule is mechanical and easy to state. When proceedings involving the same cause of action and the same parties are brought in the courts of different bound states, the court that was seised first has priority. Any court seised later must stay its proceedings, meaning it pauses, until the first court decides whether it has jurisdiction. If the first court accepts jurisdiction, the second court must decline. The point of this strict, first come first served approach is certainty. Everyone can work out in advance which court will decide, simply by looking at the order in which the cases were filed. The Convention also deals with a softer version of the same problem through the concept of #related_actions. Sometimes two cases are not identical but are so closely connected that it would be wise to hear them together to avoid conflicting outcomes. Here the later court has a discretion rather than a duty. It may stay its proceedings, and in some situations it may decline jurisdiction, so that the connected matters can be resolved in one place. This gives courts a tool to manage messy, multi-part disputes without the rigidity of the strict rule. The strength of the #lis_pendens rule is its predictability. Its weakness, under the older model that Lugano follows, is the very problem described earlier. Because the first seised court always wins the race, a party who wants to delay can file first in a slow court even where a choice of court agreement points somewhere else. The chosen court then has to wait for the slow court to sort out its own jurisdiction, which can take years. The EU addressed this inside its own system, but within Lugano the older rule survives, and litigants must plan around it (Kohler, 2022). Understanding the timing of proceedings is therefore not a small technicality. It can decide the practical outcome of a large commercial dispute. 6. Recognition and Enforcement of Judgments Deciding which court hears a case is only half of the Convention's work. The other half, and the reason many parties care about it most, is what happens to a judgment once it exists. A judgment is only as good as a party's ability to make it count where the losing side keeps its assets. The third title of the Convention is devoted to this, and it draws a careful line between two related but different ideas. The first idea is #recognition. To recognise a foreign judgment means to accept that it settles the matter it decided. Under the Convention, a judgment given in one bound state is recognised in the others without any special procedure. If a Norwegian court has decided that a debt is owed, a court in Switzerland treats that finding as settled and will not reopen the same question between the same parties. Recognition is what stops the losing party from simply re-running the whole case in a friendlier country. The second idea is #enforcement. To enforce a judgment means to use the machinery of the state, such as seizing assets or freezing bank accounts, to make the losing party comply. Here the Convention, following its 2001 model, requires an extra step. Before a foreign judgment can be enforced, the party seeking enforcement must obtain a declaration that the judgment is enforceable, a step traditionally called exequatur. The application is made in the state where enforcement is sought, and at the first stage it is largely a formality. The debtor can then challenge the declaration on limited grounds, and only at that stage does the court examine whether one of the specific defences applies. It is worth describing the enforcement procedure in a little more detail, because its stages shape how a real case unfolds. The party seeking enforcement first applies to the competent authority in the state where the debtor holds assets, supplying a copy of the judgment and a standard certificate from the original court. At this opening stage the court does not consider any defences and does not hear the debtor. It simply checks that the paperwork is in order and issues the declaration of enforceability. Only once the declaration has been made and served on the debtor can the debtor bring a challenge, and that challenge is decided in ordinary adversarial proceedings with a right of appeal. The design puts the burden of raising objections on the losing party rather than requiring the winning party to prove its case afresh. This tilts the process toward the free movement of judgments while still preserving a controlled opportunity to resist enforcement on the narrow grounds the treaty allows. This is the single largest practical difference between Lugano and the modern EU regime. The #Brussels_I_Recast abolished the exequatur step inside the Union, so a judgment from one member state can be enforced in another almost as if it were local, without a separate declaration. Lugano kept the step. This means enforcing a judgment across the Lugano circle involves more time, cost, and paperwork than enforcing a judgment between two EU member states (Hartley, 2023). It is a good example of how the failure to update Lugano in 2012 left a real, everyday gap. The Convention is strict about the grounds on which a state may refuse to recognise or enforce a foreign judgment, and this strictness is intentional. There is a firm rule against reviewing the substance of the judgment. The requested court is not allowed to check whether the original court got the facts or the law right. If that were permitted, every enforcement would turn into a fresh trial and the whole system would collapse. Instead, refusal is possible only on a short list of defences. The main ones are that recognition would be manifestly contrary to #public_policy in the requested state, that a defendant who did not appear was not served with the documents in time to arrange a defence, that the judgment is irreconcilable with a judgment given between the same parties in the requested state, and that it conflicts with an earlier judgment from another state that meets the conditions for recognition. There is also protection where the original court ignored the Convention's protective or exclusive jurisdiction rules. Outside these narrow grounds, the judgment must be honoured (Stamboulakis, 2022). The overall effect is a system of high mutual trust. Each state agrees to accept the others' judgments with only a light check at the border, on the understanding that all of them apply the same jurisdiction rules and share basic standards of fair procedure. This trust is not unlimited, and the public policy defence provides a safety valve for extreme cases. But the default is acceptance, and that default is what allows judgments to move across the region with confidence. 7. One System, Two Texts: Uniform Interpretation and Protocol 2 A treaty is only useful if everyone reads it the same way. This is a serious risk for Lugano, because the same words are applied by many different courts in states with different legal traditions, and because the twin Brussels text is interpreted separately inside the EU by the Court of Justice. If each court went its own way, the neat parallel system would slowly drift apart into a set of national variations. #Protocol_2 to the Convention exists to prevent that drift. The core idea of Protocol 2 is #uniform_interpretation. It asks the courts of every bound state to pay due account to the relevant decisions of the courts of the other states and to the decisions of the Court of Justice on the equivalent Brussels provisions. In plain terms, a Swiss judge deciding a point under Lugano is expected to look at how the same point has been decided by the Court of Justice under Brussels and by the courts of Norway or Iceland, and to aim for a consistent result. The goal is a single shared understanding stretched across two treaties (Hartley, 2023). To support this, the Protocol sets up institutional tools. There is a #Standing_Committee made up of representatives of the contracting parties, which can be convened when questions arise about the operation or revision of the Convention. There are also meetings of experts, which allow the states to exchange views on new case law and on new legislation that might affect how the treaty works. A system of reporting national case law feeds this exchange. These bodies do not issue binding rulings the way a court does, but they keep the parties talking and help spot problems before they grow. At a recent meeting the parties agreed to hold off on adapting Lugano until the EU decides whether to revise its own recast, while reaffirming that #parallel_regime alignment should be restored in the medium term. The interpretation question also reveals a quiet imbalance in the system. Inside the EU, the final word on the meaning of the Brussels rules belongs to the Court of Justice. The EFTA states that belong to Lugano have no vote in that court and no power over its rulings, yet in practice they are expected to follow its lead in order to keep the texts aligned. This is manageable when trust is high, because the EFTA states have chosen close cooperation with the Union. It becomes a point of tension whenever the relationship between the Union and an outside state grows uneasy, which is precisely the situation the next section examines. 8. The Accession Puzzle: Who Can Join and How For most of its life the question of who could join the Lugano Convention seemed technical and dull. The events of recent years turned it into one of the most contested issues in European #private_international_law. To follow the debate, it is necessary to understand exactly how a new state becomes a party. The Convention describes two routes to membership. The first route is open to states that were members of the EFTA when the Convention was drawn up and to members of the European Community. The second route is the one that matters for outsiders. It provides that #any_other_State may be invited to accede. This is the phrase that later caused so much argument. On its face it appears generous, suggesting that the door is open in principle to states from anywhere. The accession procedure, however, is demanding. The depositary of the treaty, which is Switzerland, may invite a state to accede only after obtaining the #unanimous_agreement of all the contracting parties. The parties are asked to try to give their consent within about a year of an application, but consent is not automatic and any single party can block the application by withholding it. There is a further layer that is easy to miss but crucial. Although the EFTA states are individual parties to the Convention, the European Union speaks with a single voice on the question of accession. This follows from a ruling of the Court of Justice known as Opinion 1/03, given in 2006, in which the Court held that the Union has #exclusive_competence to conclude the Lugano Convention. Because the treaty falls within the Union's external competence, it is the Union, acting through its institutions, that decides whether to consent to a new member, not each EU state separately (Kohler, 2022; Lein, 2024). In practice this means that even if every EFTA state welcomes a newcomer, the decision that counts is the one taken in Brussels. The design of the accession rule therefore builds in a powerful veto. This was probably intended to protect the coherence of the system, since letting in a state that did not share the region's standards could weaken the #mutual_trust on which enforcement depends. But the same design gives the Union enormous leverage over who joins. When the applicant is a country with which the Union has a difficult political relationship, the veto turns a technical question of treaty law into a matter of high policy. That is exactly what happened when the United Kingdom applied. 9. Brexit and the United Kingdom's Application The clearest way to see how the Convention works in the real world is to follow the story of the United Kingdom. For decades the UK was part of the Brussels and Lugano system through its membership of the European Union. British judgments moved easily across Europe, and European judgments moved easily into Britain. The UK's decision to leave the Union, a process known as #Brexit, put all of this at risk, and the treaty rules described above shaped every step of what followed. During the transition period after the UK formally left, the existing arrangements were kept alive. Under the Withdrawal Agreement, the Lugano Convention and the equivalent EU rules continued to apply to and in the United Kingdom until the end of the transition on 31 December 2020. There were also legacy provisions so that proceedings begun before that date would still be handled under the old rules. But once the transition ended, the UK became what the Convention calls a third state, meaning a country outside the club, and the smooth movement of judgments stopped (Beaumont, 2021). The UK tried to fix this by rejoining Lugano in its own name. On 8 April 2020 it formally applied to accede to the 2007 Convention. Here the accession rule became decisive. The three EFTA parties, Switzerland, Norway, and Iceland, all indicated support for the UK's return. Under the unanimity requirement, however, the European Union also had to consent, and because of the Union's #exclusive_competence, the Union's answer was the one that mattered. On 4 May 2021 the European Commission published its assessment recommending against consent, and shortly afterwards the Union formally declined. The Swiss depositary notified the parties in July 2021 that the Union was not in a position to give its consent. The application has remained blocked ever since (Lein, 2024). The reasons the Commission gave are worth studying, because they reveal how the Union understands the treaty. The core argument was that Lugano is not meant for all third countries but only for states that have a particularly close #regulatory_integration with the Union, meaning those that take part, at least partly, in the internal market through the European Economic Area or the EFTA framework. The UK, having chosen to leave both the Union and the single market, no longer fitted that description in the Commission's view. The Commission also argued that the proper tools for cooperation with more distant third countries are the global Hague Conventions rather than Lugano. In other words, the Union treated Lugano as a #flanking_measure of its internal market rather than as an open door to any willing state (Kohler, 2022; European Parliament, 2023). Critics of this position pushed back hard. They pointed out that the treaty text itself speaks of #any_other_State, which reads more broadly than the Commission's narrow interpretation. They noted that the UK's legal system had been fully aligned with the Lugano rules only moments before, so the usual worry about incompatible standards was weak. And they argued that refusing the UK increased cost and uncertainty for ordinary citizens and businesses on both sides, without any clear benefit, turning a technical instrument into a bargaining chip in a wider political contest. On this view the refusal was a political choice dressed in legal language, and the loser was #legal_certainty for people who simply wanted their disputes resolved (Lein, 2024). Supporters of the refusal replied that admitting a state with which the Union has no close market relationship would dilute the system and set an awkward precedent, and that the Union was entitled to reserve its most integrated instrument for its closest partners. Whatever view one takes, the outcome is now settled in practice. The UK has moved on and built its post-Brexit arrangements around the Hague Conventions instead. It rejoined the 2005 Hague Convention on Choice of Court Agreements in its own right, which protects exclusive choice of court agreements and the judgments that flow from them. More importantly, the UK signed the #Hague_2019 Convention on the Recognition and Enforcement of Foreign Judgments in January 2024, ratified it in June 2024, and the Convention entered into force for the UK on 1 July 2025. Because the European Union, covering its member states other than Denmark, had already been bound by Hague 2019 since September 2023, the entry into force for the UK created a new shared framework for the movement of judgments between the UK and most of the EU, the first such agreement in this field since the UK left the Union (Brand, Coffee and Herrup, 2023). 10. Lugano Compared: Brussels Ia and the Hague Conventions The Lugano Convention no longer stands alone. To judge its value, it must be placed next to the two systems that surround it. The comparison shows both why the UK wanted back in and why the alternatives, while useful, do not fully replace what Lugano offers. 10.1 Lugano and the Brussels I Recast The closest relative of Lugano is the EU's own #Brussels_I_Recast Regulation. The two are so alike that most of their provisions can be read side by side. Both allocate jurisdiction using the same families of rules, both protect weaker parties, and both make judgments move between the participating states. The differences, though fewer, are decisive in practice. The recast abolished the #exequatur step, so enforcement inside the EU is faster and cheaper than enforcement across the Lugano circle, which kept the step. The recast also reversed the harsh outcome of the Gasser case by giving priority to a chosen court, while Lugano still follows the older first seised rule. In short, the recast is a modernised version of the same idea, and Lugano is the slightly older model that has not yet been upgraded (Hartley, 2023). Restoring full parallelism between the two is a stated long-term goal of the Lugano parties, but it waits on the EU's own review of the recast. 10.2 Lugano and the Hague Choice of Court Convention 2005 The 2005 Hague Convention is far narrower than Lugano. It deals only with #exclusive_choice_of_court agreements, meaning contracts in which the parties have agreed that one particular country's courts, and no others, will decide their disputes. Where such an agreement exists, the chosen court must generally take the case, other courts must stay out, and the resulting judgment must be recognised and enforced in the other contracting states. This is valuable, and it gave the UK a foothold after Brexit. But it does nothing for the many disputes that lack an exclusive jurisdiction clause, and it does not provide the broad, ranked jurisdiction rules that Lugano supplies. It is a single tool, not a full system (Beaumont, 2021). 10.3 Lugano and the Hague Judgments Convention 2019 The newest instrument is #Hague_2019, which came into force for the UK on 1 July 2025 and was already binding on the EU. It is broader than the 2005 Convention because it is not limited to exclusive jurisdiction clauses. It allows judgments arising from a range of jurisdictional connections to be recognised and enforced between contracting states, including many judgments based on non-exclusive or asymmetric agreements, and it covers both money and non-money judgments. It cannot review the merits of the foreign judgment and allows refusal only on specific grounds such as fraud or a serious conflict with public policy, an approach that echoes Lugano's own strict list (Brand, Coffee and Herrup, 2023). Yet Hague 2019 is not a substitute for Lugano, and the difference is fundamental. Lugano is a #double_convention that governs jurisdiction directly, telling courts when they may hear a case, and then governs enforcement. Hague 2019 governs only the recognition and enforcement side. It sets out filters that a judgment must pass to be enforced, but it does not tell the original court whether it may take the case in the first place, and it does not manage #lis_pendens between competing courts. It also excludes interim measures such as injunctions, and it applies only to proceedings begun after it entered into force for the relevant state. The practical result is that Hague 2019 restores much of the ease of enforcement that was lost, but it leaves the jurisdiction stage and the problem of parallel proceedings to national law. Lugano covered the whole journey, and that completeness is what the newer conventions, useful as they are, cannot fully match (European Parliament, 2023). 11. Practical Consequences for Litigants and Businesses Legal theory becomes real when it changes what a company or a person actually does. The rules described above have concrete effects on how contracts are written, where cases are filed, and how much a dispute costs. A worked example helps make this clear. Imagine a manufacturer based in Switzerland that supplies specialised machinery to a buyer in Germany. Both states are bound by different sides of the wider Brussels and Lugano system, since Switzerland is a Lugano party and Germany applies the EU's internal rules that mirror it. When the two sign their contract, they can include a choice of court clause naming, say, the courts of Zurich. If a dispute later arises and the buyer refuses to pay, the supplier knows in advance where to sue and can rely on the Convention's rules to have a Zurich judgment recognised and enforced against the buyer's assets in Germany. The cost of the enforcement step is higher than it would be between two EU states, because the #exequatur declaration is still required across the Lugano line, but the outcome is predictable and the risk is manageable. Now change one fact. Suppose the buyer is based in the United Kingdom rather than Germany. Before Brexit, the same clause would have worked in the same smooth way. After Brexit and the refusal of the UK's Lugano application, the picture is more complicated. The parties can still rely on the 2005 Hague Convention if their clause is an exclusive one, and since July 2025 they can rely on #Hague_2019 for a wider range of judgments. But they must think carefully about the type of clause they use, the date their proceedings begin, and whether interim relief such as a freezing order will be available, because those newer instruments do not cover everything Lugano did. The lesson for anyone drafting a contract is that the choice of governing rules is not a formality. It shapes the value of any judgment they may one day win (Beaumont, 2021; Brand, Coffee and Herrup, 2023). The example also shows why timing matters so much. Under the Lugano model, the first court seised wins the race for jurisdiction, even against a #choice_of_court agreement. A well-advised party therefore watches for any sign that the other side may try to file first in a slow or unfavourable court, and acts quickly to protect its position. This is not gamesmanship for its own sake. It is a rational response to the way the rules are built, and it is one reason lawyers pay close attention to the exact moment a case is commenced. For consumers and employees, the practical message is more reassuring. The #protective_jurisdiction rules mean that an individual generally cannot be forced to litigate far from home, and that a business cannot easily strip away that protection through a contract term. A consumer in Iceland who buys from a trader elsewhere in the Lugano circle can usually bring a claim in Icelandic courts, and can usually resist being sued anywhere else. For ordinary people, this is perhaps the most valuable feature of the whole system, because it keeps justice within reach. Finally, the wider business community values the Convention for something less tangible than any single rule. It values #predictability. When companies can forecast where disputes will be decided and whether judgments will hold, they price risk more accurately, they trade more confidently across borders, and they spend less on precautionary legal advice. The economic case for instruments like Lugano rests on this quiet benefit. Fragmented rules, by contrast, raise costs and discourage cross-border activity, which is one of the reasons the UK's exclusion has been so widely criticised by practitioners (European Parliament, 2023). 12. Criticisms, Tensions, and the Road Ahead No legal instrument is perfect, and Lugano has attracted thoughtful criticism from many directions. Setting these criticisms out honestly gives a fuller picture than praise alone, and it points to where the system may go next. The first criticism concerns the frozen text. Because Lugano still follows the 2001 model while the EU has moved on to the 2015 recast, the two systems no longer match on important points. The surviving #exequatur step and the outdated approach to #choice_of_court agreements make Lugano slower and less protective than it needs to be. The parties have discussed updating the treaty and have signalled that they wish to restore #parallel_regime alignment, but they have chosen to wait until the EU decides whether to revise its own recast. This caution is understandable, since it avoids the risk of updating Lugano twice, but it leaves users stuck with a known and fixable gap in the meantime (Hartley, 2023). The second criticism concerns the accession rule and the events surrounding the UK. Many scholars argue that reading #any_other_State as narrowly as the Commission did is hard to square with the plain words of the treaty, and that the refusal of the UK's application put politics ahead of the practical interests of citizens and businesses (Lein, 2024). Defenders of the Union's position reply that the strength of the system depends on shared standards and #mutual_trust, and that the Union is entitled to reserve its most integrated instrument for states that participate in the internal market. This debate is unlikely to be settled soon, but it has already changed how outsiders view the treaty. A state considering close judicial cooperation with the Union now knows that access to Lugano may depend as much on the state of its political relationship with Brussels as on the quality of its legal system. A third and deeper tension concerns interpretation and sovereignty. The EFTA states that belong to Lugano are expected to keep their reading of the treaty aligned with the rulings of the Court of Justice, a court in which they have no say. This works while relations are warm and the EFTA states willingly choose alignment, but it is an uneasy arrangement in principle, because it asks sovereign states to follow a court they do not control. #Protocol_2 softens the tension by encouraging dialogue rather than command, yet the underlying imbalance remains a structural feature of the whole Brussels and Lugano model. Looking ahead, several futures are possible. The most likely near-term development is that the Hague Conventions continue to grow in importance, gradually building a global framework for the movement of judgments that reaches far beyond Europe. If #Hague_2019 attracts many more members, it could one day play a role for court judgments similar to the one the New York Convention plays for arbitration, although it remains untested and its reach today is still modest (Brand, Coffee and Herrup, 2023; Stamboulakis, 2022). Within Europe, the Lugano parties may eventually update the Convention to match the recast, restoring the neat parallelism that once defined the system. And the door to the UK, while firmly shut for now, is not welded closed, since a future change in the political relationship could reopen the question. What seems clear is that the technical design of Lugano will continue to be shaped by forces well outside the courtroom. 13. Conclusion The Lugano Convention is a quiet instrument that does an enormous amount of work. It answers two of the most stubborn questions in any cross-border dispute, which court may decide and whether the decision will count elsewhere, and it answers them for a defined circle of European states with a clarity that few other frameworks match. Its rules on #general_jurisdiction, #special_jurisdiction, #protective_jurisdiction, #exclusive_jurisdiction, and #choice_of_court fit together into a single ranked system. Its handling of #lis_pendens keeps two courts from deciding the same case twice. Its rules on #recognition_and_enforcement, built on #mutual_trust and a short list of defences, let judgments move across borders with a light check rather than a fresh trial. The Convention's greatest strength is its close link to the EU's own rules, which lets a small group of non-member states share in a mature and well-tested system of #judicial_cooperation. That same link is also its greatest vulnerability, because it makes membership a matter of the Union's judgment and ties the treaty's fate to the shifting politics of European integration. The refusal to let the United Kingdom rejoin, whatever one thinks of its merits, shows that a treaty designed as a technical tool can become an instrument of policy, and that #legal_certainty for ordinary people can be caught in the middle. For students of #private_international_law, the Convention is worth studying not only for its content but for what it teaches about how law travels across borders at all. Enforcing a judgment abroad is never automatic. It depends on states agreeing, in advance and in general terms, to trust one another's courts. Lugano is one of the most successful expressions of that trust, and the current search for alternatives through the Hague Conventions is really a search for the same trust on a wider scale. The judicial path across borders, in the end, is paved less by clever drafting than by the willingness of states to honour each other's decisions. The Lugano Convention shows both how much that willingness can achieve and how quickly it can be withdrawn. References Beaumont, P. (2021). Some reflections on the way ahead for UK private international law after Brexit. Journal of Private International Law, 17(1), 1-17. https://doi.org/10.1080/17441048.2021.1894757 Brand, R. A., Coffee, M. S., and Herrup, P. (2023). The 2019 Hague Judgments Convention. Oxford: Oxford University Press. https://doi.org/10.1093/law/9780192889836.001.0001 European Parliament, Policy Department for Citizens' Rights and Constitutional Affairs. (2023). Ensuring Efficient Cooperation with the UK in Civil Law Matters: Status Quo and Ways Forward (Study PE 743.340). Brussels: European Parliament. European Parliamentary Research Service. (2021). The United Kingdom's Possible Re-Joining of the 2007 Lugano Convention (Briefing PE 698.797). Brussels: European Parliament. Hartley, T. C. (2023). Civil Jurisdiction and Judgments in Europe: The Brussels I Regulation, the Lugano Convention, and the Hague Choice of Court Convention (2nd ed.). Oxford: Oxford University Press. Kohler, C. (2022). Judicial Cooperation in Civil Matters after Brexit: The Case of the British Application to Rejoin the Lugano Convention of 2007. In A. Bonomi, I. Pretelli, and G. P. Romano (Eds.), Yearbook of Private International Law, Volume XXIII (2021/2022) (pp. 81-102). Cologne: Verlag Dr. Otto Schmidt. https://doi.org/10.9785/9783504388072-006 Lein, E. (2024). An "English Good-bye": The United Kingdom and the Lugano Convention. New York University Journal of International Law and Politics, 56(1), 269-289. Stamboulakis, D. (2022). Comparative Recognition and Enforcement. Cambridge: Cambridge University Press. #Lugano_Convention #Cross_Border_Jurisdiction #Recognition_And_Enforcement #Civil_And_Commercial_Matters #EFTA_States #Brussels_I_Recast #Private_International_Law #Judicial_Cooperation #Enforcement_Of_Judgments #Brexit_And_Lugano #Hague_2019 #Choice_Of_Court #Lis_Pendens #Exequatur #Legal_Certainty

This paper studies the United Nations Convention on International Settlement Agreements Resulting from Mediation, better known as the #Singapore_Convention on mediation, which opened for signature in August 2019 and came into force in September 2020. The Convention was built to solve one clear problem. For many years, a settlement agreement reached through international #mediation could not be enforced easily across borders, so a party that broke the deal often faced no quick penalty. The Convention gives such agreements a direct route to #enforcement in the courts of member states. The central argument of this article is that the drafters copied the working method of the 1958 #New_York_Convention on the recognition and enforcement of foreign arbitral awards, which is one of the most successful commercial treaties in history. The paper places the two treaties side by side, showing where they match and where they part ways. It explains the core articles of the Singapore instrument, compares the grounds for refusing enforcement in both treaties, and reviews the reservation options that states may use. It then looks at early ratification data and at how a few countries have put the Convention into their own law. The paper closes with the live academic debate over whether the new treaty is a genuine breakthrough or mostly a repeat of tools that already exist. The study is written for students of #international_trade law, #private_international_law, and #dispute_resolution, and it uses plain language while keeping to the structure of a peer reviewed article. The conclusion is that the treaty faithfully copies the enforcement engine of its model while rebuilding the parts that assume an adjudicative process. Keywords: Singapore Convention; New York Convention; international mediation; cross-border enforcement; settlement agreements; UNCITRAL; alternative dispute resolution; private international law 1. Introduction Business today crosses borders as a matter of routine. A supplier in one country ships goods to a buyer in another, a builder in a third country signs the contract, and the money moves through banks in a fourth. When something goes wrong, the parties need a way to fix the problem that is fast, fair, and not tied to the courts of a single nation. For a long time the two main answers were litigation and #arbitration. Mediation, in which a neutral third person helps the sides reach their own deal, sat in an awkward third place. It was cheaper, quicker, and kinder to business relationships, yet it carried one serious weakness. If a party signed a mediated deal and then refused to honour it, the other side usually had to start a fresh court case to enforce the promise as an ordinary contract. That extra step made mediation look risky next to arbitration, where an award could be enforced in more than 170 countries under a single treaty. The #Singapore_Convention was designed to remove that weakness. Its formal name is the United Nations Convention on International Settlement Agreements Resulting from Mediation. It lets a company take a written #settlement_agreement straight to the competent authority of a member state and ask for direct enforcement, without first having to sue for breach of contract. In this way it aims to put mediated deals on a footing much closer to arbitral awards. The treaty does not turn a mediated deal into an award, and it does not force anyone to mediate. It simply gives the results of mediation the strong travel rights that only arbitration had before. The choice of design was not an accident. The people who drafted the Convention at the United Nations Commission on International Trade Law, known as #UNCITRAL, looked hard at the 1958 New York Convention and borrowed its structure on purpose. That older treaty took a simple idea, that a party should present the award and a limited list of defences should apply, and turned it into a global standard for arbitration. The new treaty tries the same approach for mediation. This is why many commentators quickly nicknamed it the New York Convention for mediation, and why the question of how closely the two instruments mirror each other has become central to the scholarship. This article has three goals. First, it explains the Convention in clear terms so that a student can understand what it does and how it works. Second, it compares the treaty article by article with the New York Convention, drawing out both the strong likeness and the real differences that follow from the gap between an award and a consensual deal. Third, it weighs the early evidence and the academic debate about whether the treaty will match the success of its model. The paper does not treat the two conventions as identical twins. It treats the newer one as a careful student of the older one, copying what worked while adapting to a different kind of legal document. The reader should keep one theme in mind throughout. Almost every similarity and every difference between the two treaties can be traced back to a single fact. An arbitral award is the product of a private form of judging, while a #settlement_agreement is a bargain the parties reached themselves. Where that difference does not matter, the treaties look alike. Where it does matter, they part company. Holding on to this idea makes the whole comparison easier to follow. The rest of the paper is organised as follows. Section 2 sets out the enforcement gap that the treaty was meant to close. Section 3 explains why the New York Convention became the natural template and looks briefly at the drafting history. Section 4 walks through the structure and main articles of the Singapore instrument. Sections 5 and 6 present the points of convergence and divergence between the two treaties. Section 7 compares the grounds for refusing enforcement. Section 8 examines scope and reservations. Section 9 reviews early state practice and ratification figures. Section 10 sets out the critical debate. Section 11 draws out the practical meaning for students, lawyers, and traders, and Section 12 concludes. 2. Background: The Enforcement Gap in International Mediation To see why the treaty matters, it helps to picture the old position. Suppose a manufacturer in one state and a distributor in another fall into a dispute over a supply contract. They agree to try #mediation. After two days of talks a skilled #mediator helps them reach a deal, which they write down and sign. The distributor promises to pay a reduced sum, and the manufacturer agrees to drop its wider claims. For a short while everyone is satisfied. Then the distributor changes its mind and refuses to pay. Under the traditional rules of most legal systems, that signed deal was simply a new contract. The manufacturer could not walk into a court and demand payment on the strength of the mediated document alone. Instead it had to sue the distributor for breaking the settlement contract, prove the contract, win a judgment, and only then try to enforce that judgment, often in a foreign country where the distributor kept its assets. Each of those steps cost time and money, and the final step, enforcing a foreign judgment, ran into a patchwork of national rules that offered no guarantee of success. This is the enforcement gap. The heart of the problem was that a mediated settlement had weaker travel rights than an arbitral award. An award could move across borders under the New York Convention. A court judgment had a partial network of treaties, though a less complete one. A mediated deal had almost nothing beyond the ordinary law of contract in each country. As several writers have observed, this situation was less than ideal for the promotion of mediation, because rational businesses will hesitate to choose a process whose results are hard to enforce (Zhao, 2021). The gap mattered more as global commerce grew. Companies value mediation because it is quick, it keeps costs down, it protects #confidentiality, and it can save a working relationship that a bitter court fight would destroy. In many disputes both sides would prefer to keep trading with each other after the argument is over, and mediation makes that far more likely than a public court battle. Yet the fear that a deal might unravel pushed many parties back toward arbitration, or toward a hybrid path in which a mediated deal was converted into a consent award so that it could ride on the older treaty. That workaround was clever but clumsy, and it was not always available. A treaty that made mediated deals enforceable in their own right promised to close the gap directly and to let mediation stand on its own feet as a form of #alternative_dispute_resolution. It is worth pausing on why the contract route was so weak in practice. When a party sues on a settlement contract, the defendant can raise all the ordinary defences to a contract claim, can dispute the terms, and can drag the matter through the full length of a trial. The very act of settling, which was supposed to end the dispute, becomes the start of a new one. Worse, the winner of that second case still holds only a judgment, which then has to be carried into whatever country holds the assets. For a small or medium sized business without deep pockets, this chain of steps could make enforcement so slow and expensive that the original deal was, in effect, unenforceable. The commercial value of mediation was therefore blunted by a legal weakness that had nothing to do with the quality of the process itself. The idea of a treaty did not appear overnight. Discussions about a possible instrument on the enforcement of mediated settlements took shape within UNCITRAL after a proposal was brought forward through the United States, and the matter was handed to the body's Working Group II for detailed drafting. The impetus is often traced to a 2014 public meeting of a United States advisory committee on public international law, where the case for a new instrument was set out. Over several sessions the Working Group turned the concept into treaty text, weighing how far it could copy the arbitration model and how far the special nature of mediation demanded fresh solutions. The debates were not easy, because states held different views on how much control a court should keep over a mediated deal, and on whether the treaty should apply automatically or only when parties chose it. The final text is a compromise that leans toward automatic application while leaving room for cautious states, a balance examined later in this paper. 3. The New York Convention 1958 as a Model Any discussion of the Singapore Convention has to begin with its ancestor. The Convention on the Recognition and Enforcement of Foreign Arbitral Awards, signed in New York in 1958, is widely seen as the most successful private law treaty ever made. More than 170 jurisdictions have joined it. Its reach is so broad that a party who wins an arbitral award in one member state can usually enforce it in the courts of almost any trading nation on earth. This near universal acceptance is the main reason international business trusts #arbitration and often prefers it to going to court. The genius of the #New_York_Convention lies in its simplicity. It does not try to control how arbitration is run in every detail. Instead it does two focused things. First, under its Article II, courts of member states must respect written arbitration agreements and send parties to arbitration rather than hearing the case themselves. Second, and more famously, under Articles III to V, courts must recognise and enforce foreign awards, subject only to a short, closed list of defences. The party seeking enforcement supplies the award and the agreement. The burden then shifts to the losing party to prove that one of the narrow grounds for refusal applies. Those grounds, set out in Article V, are the heart of the system. They include a party's incapacity or an invalid arbitration agreement, a failure to give proper notice or a denial of the chance to present a case, an award that goes beyond the scope of what was submitted, an irregular composition of the tribunal or an irregular procedure, and an award that is not yet binding or has been set aside at its place of origin. On top of these, a court may refuse enforcement if the subject matter cannot be settled by arbitration under its law, or if enforcement would be contrary to #public_policy. The list is exhaustive. A court cannot review the merits of the dispute or refuse an award simply because it disagrees with the outcome. This design has three qualities that made it attractive as a template. It is pro-enforcement, meaning the default is that the award will be honoured. It is predictable, because the defences are known and limited. And it respects #party_autonomy, since it enforces what the parties themselves agreed to submit. The result is a treaty that reduces the risk of cross-border dealing and lowers the cost of doing business. Over more than sixty years, courts around the world have built a large body of case law applying the Convention, and while they do not always agree on the details, the core promise has held. That track record is exactly what a young treaty on mediation would want to inherit. When UNCITRAL turned to mediation, it faced a natural question. Could the same recipe work for a different dish? Arbitral awards and mediated settlements are not the same kind of legal animal, a point developed later in this paper. Yet the underlying business need was identical. Give the winning document strong travel rights and a short list of defences. The drafters decided that the New York Convention offered the best available blueprint, and they set out to adapt rather than reinvent it. This borrowing is why the two treaties share so much of their skeleton, and why a comparative reading is the most useful way to understand the newer one (Wen and Finder, 2023). There is also a strategic reason for copying the model. A new treaty has to persuade states to join, and states are more comfortable adopting a design they already know and trust. By echoing the familiar shape of the New York Convention, the drafters made the new instrument easier to explain to governments, easier to fit into existing court procedures, and easier for lawyers to advise on. The nickname the New York Convention for mediation was therefore not only a description but also a selling point. It told the world that the new treaty was built on proven foundations rather than on an untested theory. 4. Overview of the Singapore Convention: Structure and Core Provisions The Singapore Convention is a short instrument of sixteen articles, but the working core sits in the first eight. It was adopted by the United Nations General Assembly in December 2018, opened for signature in Singapore in August 2019, and entered into force in September 2020 after the third ratification was deposited. A brief tour of the main articles shows how the machine is meant to run. Article 1 sets the scope. The Convention applies to a #settlement_agreement that is international, that results from #mediation, and that resolves a commercial dispute. An agreement is international, in broad terms, when the parties have their places of business in different states, or when the state where they carry out a large part of their obligations, or with which the subject matter is most closely connected, differs from the state of their business. Article 1 also carves out several categories. It does not cover deals made by a consumer for personal or family use, or matters of family, inheritance, or employment law. Just as important, it excludes agreements that a court has already approved or that can be enforced as a judgment, and those that have been recorded and can be enforced as an arbitral award. Those carve-outs are deliberate. They stop the treaty from clashing with the New York Convention on arbitral awards or with treaties on court judgments, so that each instrument keeps to its own lane. This tidy division of labour is one of the quiet strengths of the treaty, because it avoids the confusion that would follow if two instruments claimed the same document. Article 2 supplies key definitions, including what counts as a written agreement and, importantly, a broad definition of mediation as a process where a third person helps the parties reach a deal without the power to impose one. The definition is flexible on purpose. It does not depend on what the process is called, so conciliation and similar labels are covered as long as the substance matches. The written requirement is also generous, since an agreement can meet it even in electronic form, which reflects the reality of modern commerce where deals are often recorded and signed online. Article 3 is the engine. It states the two duties of a member state. First, each member must enforce a covered settlement in line with its own procedural rules and the conditions of the Convention. Second, each member must allow a party to invoke the agreement to prove that a matter has already been resolved. Scholars often describe these as the sword and the shield. The sword lets a party press for enforcement of the deal. The shield lets a party block a fresh claim over a dispute that the deal already settled. This dual function is important, because in practice many disputes about a settlement are not about forcing payment but about stopping the other side from reopening a closed matter. Article 4 lists what a party must supply to rely on the Convention. It must produce the settlement signed by the parties and evidence that the agreement resulted from mediation, such as the #mediator's signature, a document signed by the mediator confirming that mediation took place, or an attestation from the institution that ran the process. The competent authority may ask for any further document needed to check that the conditions are met. This mirrors the document-based approach of the older treaty, where the party simply files the award and the agreement, and it keeps the court's task narrow and quick. Article 5 sets out the grounds on which a court may refuse to grant relief. Because this article is the direct counterpart of Article V of the New York Convention, it is examined in detail in Section 7. For now it is enough to note that the grounds are limited and that some of them are unique to mediation, in particular the grounds dealing with mediator misconduct. Articles 6 to 8 handle procedural and policy matters. Article 6 deals with the situation where a related claim or application about the same agreement is running elsewhere, allowing a court to pause its decision if that is proper. Article 7 preserves any more favourable law or treaty, so the Convention sets a floor rather than a ceiling, and a party may still rely on a friendlier national rule if one exists. Article 8 sets out the two permitted #reservations, discussed in Section 8. The remaining articles deal with signature, ratification, entry into force, and similar formal matters. Taken together, the eight core articles create a compact, document-driven route to cross-border enforcement that clearly echoes the logic of 1958 while adjusting the details for a consensual process (Palao, 2022). 5. Mirroring the New York Framework: Points of Convergence The likeness between the two treaties is not a coincidence of style. It is a deliberate design choice, and it runs deep. Several points of #convergence stand out, and setting them out in order shows just how closely the newer treaty follows its model. The first is the shared aim. Both treaties exist to support #international_trade by making a favoured form of #dispute_resolution reliable across borders. The New York Convention did this for arbitration; the Singapore Convention tries to do it for mediation. Both start from the belief that commerce flows more freely when parties can trust that the outcome of their chosen process will be honoured in foreign courts. Trust is the real product both treaties sell. The second is the pro-enforcement default. In each treaty the basic rule is that the relevant document, an award in one case and a settlement in the other, will be given effect. The party resisting enforcement carries the burden of proving that a defence applies. Courts are not invited to reopen the merits. This tilt toward #recognition is the feature that gives both treaties their commercial value, because it means the successful party does not have to fight the whole battle over again. The third is the document-based method. Article 4 of the Singapore instrument closely follows Article IV of the New York instrument. In both, the party seeking relief presents a defined set of papers, and the process then turns on whether a narrow list of objections is made out. This keeps the court's task simple and quick, and it reduces the chance that a debtor can drag out proceedings with side issues. The fourth is the closed list of defences. Article 5 of the Singapore Convention plays the same role as Article V of the New York Convention. Each provides an exhaustive set of grounds for refusal. A court cannot invent new reasons to decline enforcement. This shared feature is what gives each treaty its predictability, because parties can read the list and know in advance what can go wrong. The fifth is the presence of the same two general safety valves. Both treaties allow a court to refuse enforcement if the subject matter cannot be resolved by the chosen process under the forum's law, and both allow refusal where enforcement would breach public policy. These twin exceptions let a state protect its most basic legal values while keeping the exceptions narrow enough that they do not swallow the rule. Courts applying the New York Convention have generally read the public policy exception narrowly, and there is good reason to expect that the same restraint will be urged on courts applying the Singapore Convention. The sixth is respect for party autonomy. Each treaty enforces what the parties themselves chose, an award from an arbitration they agreed to, or a deal they reached in mediation. Neither treaty forces an outcome on unwilling parties; both give strength to bargains freely made. This respect for choice is the moral centre of both instruments, and it explains why courts are told not to second guess the outcome. These parallels explain why the shorthand New York Convention for mediation caught on so fast. The Singapore Convention takes the trusted architecture of 1958, the direct route, the shifted burden, the short defence list, and the two safety valves, and applies it to a new document type. For students, the clearest way to learn the treaty is to hold it up against its model and note how faithfully the frame has been copied. Yet the copy is not perfect, and the differences matter just as much as the likeness, as the next section shows (Zhao, 2021; Wen and Finder, 2023). 6. Where the Two Conventions Diverge: Key Differences The two treaties share a skeleton, but the flesh differs, because an arbitral award and a mediated deal are not the same thing. An award is the product of arbitration, a private form of adjudication in which a tribunal hears evidence and imposes a decision. A settlement is a consensual contract, a bargain the parties reached themselves. This basic gap drives most of the #divergence between the instruments. The first and most striking difference is the absence of a seat. In arbitration, every award has a legal home, the seat, which is the country whose courts supervise the process and can set the award aside. The New York Convention is built around this idea, so an award has a nationality and a place of origin. Mediation has no seat. A mediated deal is not anchored to any single legal system in the same way. The Singapore Convention therefore treats settlements as effectively a-national, meaning a party can seek enforcement wherever the debtor's assets are without pointing to a country of origin. This makes the newer treaty in one sense broader, because it is not tied to a place of origin, but it also removes the supervising court that arbitration relies on for a first check on fairness. The second difference follows from the first. Under the New York Convention, one ground for refusing an award is that it has been set aside or suspended at the seat. There is no equivalent ground in the Singapore Convention, because there is no seat and no supervisory court that could annul a mediated deal. A settlement can be attacked on contract grounds, such as fraud or incapacity, but there is no separate annulment system layered on top. This makes the mediated route simpler in structure, though it also means the safeguards must all be built into the #enforcement stage itself rather than being split between a supervising court and an enforcing court. The third difference is the set of grounds tied to the conduct of the mediator. Article 5 of the Singapore Convention allows a court to refuse relief where there was a serious breach by the #mediator of standards that apply to the mediator or the mediation, without which breach a party would not have entered the agreement, and where the mediator failed to disclose circumstances raising justifiable doubts about impartiality or independence, again where that failure had a material effect. These grounds have no parallel in the New York Convention, and they exist precisely because mediation depends on the neutrality and fairness of the facilitator rather than on a formal tribunal. They are also among the hardest grounds to apply, since #confidentiality can make mediator conduct difficult to prove, and standards for mediators vary widely from one country and institution to another. The fourth difference concerns the nature of the document being enforced. An award carries adjudicative finality, since a tribunal has ruled. A mediated deal carries only the finality of a contract that the parties agreed to. This affects the questions a court may ask. Under the newer treaty a court may refuse relief where the agreement is not binding or not final, where its obligations are unclear, or where enforcement would run against the terms of the deal itself. These merits-adjacent checks reflect the contractual nature of the instrument and have no direct match in the arbitration model, where the tribunal has already resolved the merits. The fifth difference lies in the tension with confidentiality. Mediation is prized for privacy, yet Article 4 requires a party to prove that the deal resulted from mediation, which can force the disclosure of information the parties expected to keep private. Arbitration has no equal problem, because an award is a public-facing legal document by design. This tension is a genuine cost of importing an arbitration-shaped enforcement model into a confidential process, and careful drafting of the settlement and of the evidence of mediation is one way parties can reduce it (Tan, 2023). Taken together, these differences show that the Singapore Convention is not a photocopy. It is an adaptation. The drafters kept the enforcement engine but rebuilt the parts that assumed an adjudicative process, replacing seat-based supervision with mediator-conduct safeguards and contract-based checks. Understanding this trade-off is the key to reading the treaty well, and it is the point where a student can most easily go wrong by assuming the two treaties are the same. 7. Grounds for Refusing Enforcement: Article 5 Compared With Article V Because the defence lists are the beating heart of both treaties, they deserve a closer look. A student who can map Article 5 of the Singapore Convention onto Article V of the New York Convention will grasp the relationship between the instruments better than through any general summary. Start with the grounds that plainly match. Both treaties allow refusal where a party was under some incapacity. Both allow refusal where the underlying agreement is invalid, null, void, or inoperative. Both allow refusal on the two public grounds already noted, that the subject matter cannot be resolved by the chosen process, and that enforcement would breach #public_policy. In these areas the newer treaty simply reuses the tested language of 1958, adjusted from award to #settlement_agreement. A lawyer familiar with the arbitration defences will feel at home, and courts will be able to draw on decades of case law about what these shared terms mean. Next come the grounds that adapt the older model to the contractual setting. The New York Convention lets a court refuse an award that goes beyond the scope of the submission. The Singapore Convention has a related idea in a different form. A court may refuse relief where the obligations in the agreement have been performed, or where the obligations are not clear or comprehensible, or where granting relief would be contrary to the terms of the settlement. These are checks that make sense for a contract rather than an award. They ask whether the deal still needs enforcing and whether its terms can actually be given effect. There is no exact arbitration equivalent, but the family resemblance to the scope ground is clear, and both share the idea that a court should not enforce something the parties did not truly agree to. Then come the grounds with no counterpart at all, the two mediator-conduct grounds described in Section 6. A serious breach of applicable standards by the mediator, and a failure by the mediator to disclose facts casting justifiable doubt on independence or impartiality, may each defeat enforcement, but only where the breach or failure was material to the party's decision to settle. These grounds import into the treaty the ethical expectations that surround mediation. They are the clearest sign that the drafters did not simply cut and paste, and commentators have noted that they may prove among the most litigated and least predictable parts of the treaty, precisely because judging a mediator's conduct is far less mechanical than checking whether notice was given in an arbitration (Alexander, Chong and Giorgadze, 2022). Finally, note what the newer treaty leaves out. There is no ground based on the setting aside of the instrument at a seat, for reasons already given. There is also no ground based on the composition of a tribunal or the regularity of a tribunal's procedure, because mediation has no tribunal. The removal of these arbitration-specific grounds is as telling as the addition of the mediator grounds. Each change tracks the shift from an adjudicative process to a consensual one. The overall picture is a defence list that keeps the shared public safeguards, reworks the contract-facing checks, adds new safeguards for mediator conduct, and drops the grounds that only make sense for an award. This selective mirroring is the most precise way to state how the Singapore Convention relates to the New York Convention. It borrows the structure of Article V while rewriting the specific defences to fit a different kind of document. A useful classroom exercise is to place the two articles in two columns and mark each ground as shared, adapted, added, or dropped, because that simple table captures the whole relationship at a glance. 8. Scope, Reservations, and the Opt-In Mechanism Every multilateral treaty must balance uniform rules against the differing needs of states, and both conventions use #reservations to strike that balance. Here again the Singapore Convention borrows the idea of reservations from the older treaty while offering a different menu. The New York Convention permits two well-known reservations. A state may declare that it will apply the treaty only to awards made in the territory of another member state, known as the reciprocity reservation, and it may declare that it will apply the treaty only to disputes that are commercial under its own law, known as the commercial reservation. These options let states join without fully opening their courts to every foreign award, and many states used them, especially in the early years when trust in the system was still being built. The Singapore Convention offers two reservations of its own, set out in Article 8. The first lets a state declare that the Convention will not apply to a settlement to which that state, or a government body, or a person acting for a government body, is a party. This shields public entities from automatic enforcement and reflects the caution some states feel about binding their own organs. Several early parties, including Brazil, made use of this reservation on joining, keeping the treaty away from deals involving the state itself. The second reservation is more significant for the treaty's ambition, and it has no clean equivalent in the arbitration model. Under it, a state may declare that the Convention applies only to the extent that the parties to the settlement have agreed to its application. In effect this turns the treaty into an opt-in system within that state, so that a mediated deal benefits from #direct_enforcement only where the parties expressly chose the Convention. This is the reverse of the treaty's normal default, under which the Convention applies unless a listed defence is made out. Critics worry that wide use of the opt-in reservation could dilute the very automatic quality that makes the treaty valuable, because parties who never think to opt in would lose the benefit. Supporters reply that the option makes the treaty easier for cautious states to accept, which may matter more in the long run than perfect uniformity. The design is a bet that a slightly weaker rule adopted by many states is better than a stronger rule adopted by few. Scope also differs in an important way that connects to reservations. The New York Convention allows a reciprocity reservation, so enforcement can depend on the award coming from another member state. The Singapore Convention contains no reciprocity requirement. A covered settlement may be enforced in a member state regardless of where it was concluded or which law governs it, and regardless of whether the state connected to the deal is itself a party. This makes the newer treaty potentially broader in reach, since it does not gatekeep by nationality of the document. It also means that the treaty's benefits can flow to parties whose home states have not joined, which may encourage wider participation over time as businesses in non-member states see the advantage of choosing the Convention. The reservation design therefore shows the same pattern seen throughout this comparison. The drafters kept the general tool, reservations as a way to ease acceptance, but reshaped the specific options to fit a consensual process and to tempt hesitant states with an opt-in path. Whether that path helps or hurts the treaty's long-term strength is one of the open questions the scholarship continues to debate (Xiong, 2025). 9. Early Implementation and State Practice A treaty lives or dies by adoption, so the record of #ratification and domestic implementation is central to any honest assessment. Here the story is one of steady but modest growth, and it is best read against the slow early history of the treaty it copies. The Singapore Convention opened for signature in August 2019 and drew a striking show of early support. A record number of states, forty-six, signed on the opening day, and the list of signatories has since grown to around sixty. Actual ratification, which is the step that makes a state legally bound, has moved more slowly. The treaty entered into force in September 2020 once three states, Singapore, Fiji, and Qatar, had deposited their instruments. By the second half of 2025 the number of #contracting_states had reached nineteen, with Brazil becoming the nineteenth party after ratifying in August 2025, its obligations due to take effect in early 2026. Other parties over the years have included Bahrain, Belarus, Costa Rica, Ecuador, Georgia, Honduras, Israel, Japan, Kazakhstan, Nigeria, Paraguay, Saudi Arabia, Sri Lanka, Turkey, and Uruguay. The gap between signature and ratification is itself instructive. Signing shows political goodwill, but ratification requires a state to align its domestic law with the treaty, and that legislative work takes time and care. The pattern is not unusual for a young treaty, and it is worth remembering that the New York Convention also grew gradually before reaching near-universal acceptance over several decades. Still, the slow pace, and the continued absence of major economies such as the United States, the members of the European Union, and China, has drawn comment from scholars who question how quickly the treaty can deliver on its promise. A treaty draws much of its power from the size of the network it creates, and the network is still forming. Domestic implementation offers more encouraging detail. Singapore, the treaty's namesake, passed the Singapore Convention on Mediation Act in 2020, which came into force on the same day as the treaty. That Act lets a party to an international settlement apply to the Singapore High Court to record the agreement as an order of court, giving it the force of a judgment and a clear domestic route to enforcement. Singapore has paired this legal step with heavy investment in mediation institutions, so that it can offer parties both a place to mediate and a reliable way to enforce the result. Japan, the largest economy to join so far, ratified through an implementing Act that took effect in April 2024 and created a special jurisdiction, channelling enforcement applications to designated district courts in Tokyo and Osaka. Japan's arbitration institution also updated its commercial mediation rules in the same period to fit the new framework. These examples show the treaty being woven into national systems rather than left as a paper promise. The United Kingdom's path illustrates the deliberate pace at which serious states move. The United Kingdom signed the Convention in 2023 and, in 2025, ran a public consultation on how to implement it, confirming a commitment to ratify once the necessary domestic measures are in place. The consultation asked practical questions about how mediated settlements should be registered, how the grounds for refusal should be handled, and how the treaty should be given effect across the whole of the United Kingdom. This careful approach reflects the fact that joining a treaty is not just a signature but a promise to build the machinery that makes it work. Discussion has been just as lively in states that have not yet joined. In the United States, a major bar association passed a resolution urging the country to become a party, and the case for ratification has been debated at length among practitioners and academics, though no ratification has followed. Within the European Union, the question of whether the Union or its member states should join, and on what legal basis, has been the subject of expert roundtables and study. In China, scholars have examined in detail how the treaty would fit with the country's own commercial mediation system and what changes to domestic law ratification would require (Hu and Jiang, 2025; Xiong, 2025). This mix of firm early adopters, careful consulters, and interested but hesitant giants captures the treaty's present position. It is established, growing, and still working to reach the critical mass that made its model so powerful. 10. Critical Perspectives: Breakthrough or Redundancy? Scholars do not agree on how much the Singapore Convention changes. The debate is worth setting out plainly, because it goes to the value of the whole project. One camp sees the treaty as a genuine breakthrough. On this view, the enforcement gap was real and damaging, and closing it removes the single biggest reason businesses distrusted international mediation. By giving mediated deals #direct_enforcement without a fresh lawsuit, the treaty raises mediation toward the status that arbitration has long enjoyed, and it does so using a proven design. Supporters point to the strong early signature numbers and to the growing list of implementing statutes as signs that the treaty answers a real need. For them, the Singapore Convention is to mediation what the New York Convention was to arbitration, a foundational instrument whose full effect will only be felt over decades rather than years. A second camp is more cautious and asks whether the treaty is partly redundant. The argument runs like this. Parties who want a strongly enforceable outcome already have a workaround, the hybrid #Arb_Med_Arb process, in which a dispute is placed into arbitration, settled through mediation, and then recorded as a consent award that rides on the New York Convention. Since that route already produces a globally enforceable document, some writers ask what the new treaty adds. On this reading, the Singapore Convention may compete with existing hybrid tools rather than replace them, and its real gains may be limited to pure mediated settlements that were never converted into awards (Tan, 2023). The reply from the first camp is that the workaround is neither simple nor universal. Converting a deal into a consent award requires an arbitration to be on foot, which adds cost and formality and defeats part of the point of choosing mediation in the first place. Many mediated deals are reached outside any arbitration, and for those the older treaty offers nothing. The Singapore Convention fills exactly that space, giving stand-alone mediated settlements the travel rights they previously lacked. So even if the treaty overlaps with hybrid tools at the edges, it covers a large field those tools cannot reach. A third strand of criticism focuses on doubts about whether the New York success can be repeated at all. Arbitration produces an award, an adjudicative decision with a clear structure that courts are used to enforcing. Mediation produces a consensual contract, which invites more merits-adjacent questions and carries the confidentiality tension discussed earlier. Because the mediated document is softer and more variable than an award, some scholars argue that the achievements of the New York Convention do not automatically promise the same success for the newer treaty, and that enforcement may prove less predictable in practice (Zhao, 2021). Much will depend on how courts handle the mediator-conduct grounds, which have no settled case law behind them. There is also a practical worry about reach. A treaty's power depends on how many states join and on whether the world's largest markets are inside it. The continued absence of several major economies, and the availability of the opt-in reservation, could slow the treaty's rise and weaken the uniformity that made its model so valuable. Against this, defenders note that the New York Convention itself started small and that the Singapore Convention's early trajectory is broadly comparable at the same age. A treaty should not be judged a failure simply because it has not conquered the world within its first five years. A further point in the debate concerns culture and habit. In some legal systems, lawyers and judges are deeply comfortable with arbitration but less familiar with mediation, and habits change slowly. Even a perfect treaty cannot force parties to choose mediation, and its success will partly depend on whether the business and legal communities come to see mediation as a serious first choice rather than a soft alternative. Institutions that promote mediation, train mediators, and publish clear rules therefore play a large part in whether the treaty realises its promise. The honest conclusion is that the debate is unresolved and depends on evidence that is still being gathered. Whether the treaty proves a breakthrough or a partial repeat will turn on ratification numbers, on how courts apply the mediator-conduct grounds, and on whether business habits shift toward stand-alone mediation now that its results can travel. What is clear is that the treaty has changed the strategic landscape of #dispute_resolution and forced a serious rethinking of where mediation sits among the options. 11. Implications for Students, Practitioners, and Cross-border Trade The comparison between the two treaties is more than an academic exercise. It carries lessons for several audiences, and drawing them out shows why the topic deserves close study. For students of law, the Singapore Convention is a model case study in how treaties are designed. It shows that lawmakers rarely start from a blank page. Instead they find a working instrument, in this case the New York Convention, and adapt it to a new problem, keeping what fits and reshaping what does not. Learning to read the newer treaty against the older one teaches the comparative method that lies at the centre of #private_international_law. It also shows how legal categories matter, because the whole pattern of convergence and divergence flows from the single fact that an award and a settlement are different kinds of document. A student who understands that one distinction can predict most of the differences between the treaties, which is a powerful demonstration of how a small conceptual point can shape a whole legal design. For practitioners, the treaty changes the advice they give. When drafting a mediation clause or a mediated deal, lawyers now need to think about #enforcement from the start. They should check whether the relevant states are parties, whether any of those states used the opt-in reservation so that an express choice of the Convention is needed, and whether the agreement is drafted so clearly that its obligations can be enforced without the ambiguity that Article 5 treats as a defence. They must also keep in mind the evidence requirement of Article 4, and plan how they will prove that a deal resulted from mediation without giving away more than confidentiality allows. A well-drafted settlement will state plainly that it resulted from mediation, will identify the mediator, and will set out obligations in clear and measurable terms. In short, the treaty rewards careful drafting and punishes sloppiness. For business, the treaty lowers the risk of choosing mediation and therefore should make it a more attractive option for #cross_border disputes. Mediation is faster and cheaper than litigation or full arbitration, and it can preserve a commercial relationship that a court fight would end. The old fear, that a mediated deal was hard to enforce abroad, was the main brake on its use. As more states join the treaty and as courts build a track record of enforcing mediated settlements, that brake should ease, and companies may increasingly place mediation on an equal footing with arbitration when they plan how to handle disputes. For a smaller company in particular, a cheaper process whose results can now travel is an attractive prospect. For legal systems as a whole, the treaty pushes states to think about how mediated settlements fit within their courts. Implementation is not automatic. As the Singapore and Japan examples show, a state usually needs new legislation to create a clear domestic route for #recognition and enforcement, and it must decide which courts will handle applications and how the treaty's grounds for refusal map onto local procedure. The treaty thus acts as an engine of #harmonization, nudging national systems toward a shared approach while leaving room for local design. This is how private international law often advances, not by a sudden global rule but by many states quietly adopting the same shape. Finally, for the wider system of international trade, the Singapore Convention adds a third pillar of cross-border enforcement alongside the New York Convention on awards and the developing framework on court judgments. Together these instruments aim to give parties confidence whichever process they choose, whether a court, an arbitration, or a mediation. The completion of this trio, if the newer treaty succeeds, would mark a real step toward a more predictable global market for dispute resolution, in which the choice of process is driven by what suits the dispute rather than by which outcome happens to be enforceable. 12. Conclusion The Singapore Convention on mediation set out to close a stubborn gap. For decades a mediated #settlement_agreement was the weakest of the major dispute outcomes when it came to crossing borders, and that weakness held mediation back despite its clear advantages in cost, speed, and the preservation of relationships. The treaty answers the problem by giving such agreements a direct route to enforcement in the courts of member states, subject only to a short and closed list of defences. The central claim of this paper is that the treaty achieves this by mirroring the framework of the 1958 #New_York_Convention. The likeness is deliberate and deep. Both instruments share a pro-enforcement default, a document-based method, a closed list of grounds for refusal, and the twin safety valves of non-eligibility of the subject matter and #public_policy. Both respect #party_autonomy, and both aim to make a favoured form of dispute resolution travel reliably in the service of international trade. It is this shared architecture that earned the newer treaty its nickname as the New York Convention for mediation. Yet the mirror is not flawless, and the differences are as important as the similarities. Because a mediated deal is a consensual contract rather than an adjudicative award, the Singapore Convention has no seat, no supervising court, and no set-aside ground. In their place it adds contract-facing checks and, above all, new grounds tied to #mediator conduct that have no parallel in the arbitration model. It also offers a different set of reservations, including an opt-in option that could either ease acceptance or dilute the treaty's automatic reach. These adaptations show that the drafters were careful students of the older treaty rather than mere copyists. The early record is one of promise and patience. Signature numbers were strong, ratification has grown steadily to nineteen states, and several countries, including Singapore and Japan, have built real domestic machinery to give the treaty effect. At the same time, the pace is measured, major economies remain outside, and scholars still debate whether the treaty is a true breakthrough or partly a repeat of tools that already exist through hybrid practice. That debate cannot yet be settled, because much depends on evidence still to come, on how many states join, and on how courts apply the treaty's more novel grounds. For now, the fairest assessment is that the Singapore Convention has done for mediation what its model did for arbitration in structure, if not yet in scale. It has taken a trusted design and applied it to a new document, and in doing so it has changed the strategic calculus of #cross_border dispute resolution. Whether it grows into a foundational treaty on the scale of 1958 will depend on the years ahead. What is already clear is that mediation now has, for the first time, a serious international instrument built to make its results enforceable across the world, and that alone marks a turning point in the story of #alternative_dispute_resolution. #Singapore_Convention_on_Mediation #international_mediation #cross_border_enforcement #New_York_Convention_comparison #settlement_agreements #UNCITRAL_treaty #alternative_dispute_resolution #private_international_law #commercial_mediation #enforcement_of_mediated_settlements #Singapore_Convention_2019 #mirroring_the_New_York_framework #international_commercial_disputes #mediation_law #global_dispute_resolution References Alexander, N. M., & Chong, S. (2022). Leading the way for the recognition and enforcement of international mediated settlement agreements: The Singapore Convention on Mediation Act 2020. Singapore Academy of Law Journal, 34, 1-50. Alexander, N. M., Chong, S., & Giorgadze, V. (2022). The Singapore Convention on Mediation: A Commentary (2nd ed.). Alphen aan den Rijn: Wolters Kluwer. Chahine, J. H., Lombardi, E. M., Lutran, D., & Peulve, C. (2021). The acceleration of the development of international business mediation after the Singapore Convention. European Business Law Review, 32(4), 769-800. Hu, K., & Jiang, Q. (2025). Research on the integration of the Singapore Mediation Convention with China's commercial mediation system. Journal of International Economy and Global Governance, 2(2), 65-83. https://doi.org/10.12414/jiegg.250452 Palao, G. (Ed.). (2022). The Singapore Convention on Mediation: A Commentary on the United Nations Convention on International Settlement Agreements Resulting from Mediation. Cheltenham: Edward Elgar Publishing. Tan, D. (2023). The Singapore Convention on Mediation to reinforce the status of international mediated settlement agreement: Breakthrough or redundancy? Conflict Resolution Quarterly, 40(4), 467-482. https://doi.org/10.1002/crq.21377 Wen, X., & Finder, S. (2023). Comparative analysis of the Singapore Convention in light of the New York and Hague Choice of Court Conventions. In S. Ali (Ed.), Comparative and Transnational Dispute Resolution (p. 170). London: Routledge. https://doi.org/10.4324/9780429354625-13 Xiong, H. (2025). The legislative adjustment of Chinese enforcement regulation of the international commercial settlement agreement in the context of the Singapore Mediation Convention. Journal of International Dispute Settlement, 16(1), idae025. https://doi.org/10.1093/jnlids/idae025 Zhao, Y. (2021). The Singapore Mediation Convention: A version of the New York Convention for mediation? Journal of Private International Law, 17(3), 538-566. https://doi.org/10.1080/17441048.2021.2006399

For a long time, the global fight against dirty money focused almost entirely on banks. Regulators assumed that if they watched the doors of the formal banking system closely enough, criminals would struggle to hide their profits. That assumption turned out to be too narrow. As banks tightened their controls, offenders moved their money through other channels: buying property, hiring lawyers to build corporate structures, and turning cash into gold, diamonds, and luxury watches. This article examines how anti-money laundering (AML) rules have expanded beyond the banking sector to reach a group of businesses and professionals known as Designated Non-Financial Businesses and Professions (DNFBPs). It gives special attention to three high-risk groups named in the title: real estate agents, lawyers, and dealers in precious metals and stones (including jewelers). The article explains the international standards that drive this expansion, mainly the Financial Action Task Force (FATF) Recommendations, describes the core obligations these professionals now face, and analyses the practical and ethical tensions the expansion creates, such as the clash between reporting duties and legal professional privilege. Drawing on recent scholarship and policy developments in the United States, the United Kingdom, the European Union, Australia, and elsewhere, the article argues that bringing DNFBPs into the AML system is necessary but far from sufficient. Without proportionate design, adequate supervision, and honest measurement of results, the expansion risks producing paperwork rather than protection. The article closes with recommendations for a smarter, risk-based, and evidence-led approach to regulating non-financial gatekeepers. Keywords: designated non-financial businesses and professions; anti-money laundering; gatekeepers; real estate; legal professionals; precious metals and stones; FATF; beneficial ownership; suspicious transaction reporting; risk-based approach. Introduction Money laundering is the process of making the proceeds of crime look clean and lawful so that criminals can spend, invest, or move their wealth without attracting attention. For decades, the international response treated this as mainly a banking problem. Banks were required to know their customers, keep records, and report anything that looked suspicious. This made sense, because banks sit at the centre of the payment system. But criminals are adaptive. When one route becomes risky, they look for another. As banking controls improved, illicit funds increasingly flowed through professionals and businesses that were never designed to act as financial watchdogs. To see why gatekeepers matter, it helps to recall how laundering usually works. Analysts often describe it in three stages. In placement, criminal cash first enters the economy, for example through a business or a purchase. In layering, the money is moved through a series of transactions and structures to break the trail back to the crime. In integration, it re-emerges as apparently clean wealth, ready to be spent or invested. Non-financial professionals can appear at every stage. A dealer accepts cash at placement, a lawyer builds the layering structure, and a property purchase completes integration. This is exactly why the AML system could not stop at the bank. This shift produced a category that now sits at the heart of AML policy: #DNFBPs, or Designated Non-Financial Businesses and Professions. The term was introduced in the FATF Forty Recommendations and covers casinos, real estate agents, dealers in precious metals and precious stones, lawyers, notaries and other independent legal professionals, accountants, and trust and company service providers. What unites this varied group is a shared feature: they handle large sums, high-value assets, or the legal structures that hold wealth, and they enjoy the trust of clients and society. That combination makes them attractive to anyone who needs to disguise the origin of criminal money. These professionals are often described as #gatekeepers. The word captures the idea that they control access points to the legitimate economy. A lawyer can create the company that owns an asset. A real estate agent can help move a large sum into a stable, respectable form. A jeweler can turn suspicious cash into a portable, high-value object that crosses borders easily. If gatekeepers ask the right questions, they can slow or block the flow of illicit funds. If they do not, they become the very conduits that let dirty money enter the mainstream. This article analyses the regulatory expansion that has pulled these gatekeepers into the AML system. It focuses on the three sectors named in the title. Real estate is one of the oldest and largest laundering channels in the world. Legal professionals raise the most difficult questions because their duties to clients can collide with duties to report. Dealers in precious metals and stones, including jewelers, deal in goods that behave almost like cash but are far easier to hide and move. Studying these three together shows both the logic of the expansion and its limits. The purpose of the article is threefold. First, it explains why regulators concluded that watching banks alone was not enough. Second, it sets out the concrete obligations that DNFBPs now carry, including customer due diligence, beneficial ownership checks, suspicious transaction reporting, record-keeping, and a #risk_based_approach to compliance. Third, it examines the tensions and criticisms that surround the expansion, from the burden on small firms to the debate about whether these rules actually reduce crime. The article is written for students and newcomers to financial crime studies, so it keeps the language plain while still engaging with current academic and policy debate. Conceptual and legal framework 2.1 What DNFBPs are and why they matter A useful way to understand DNFBPs is by contrast. Financial institutions, such as banks, insurers, and money transfer firms, are directly in the business of moving and storing money. DNFBPs are not. They sell property, give legal advice, prepare accounts, or trade in valuable goods. Yet in the course of that ordinary work, they touch money and assets in ways that criminals can exploit. A conveyancing lawyer moves client funds through a trust account. A real estate agency channels a purchase price. A jeweler accepts payment for a high-value item. Each of these ordinary transactions can be used to place, layer, or integrate the proceeds of crime. Nduka and Sechap (2021) describe DNFBPs as important actors in both the formal and informal economy, precisely because the services they offer are central to daily commercial life. That centrality is also their weakness. Their study of West Africa found repeated cases of DNFBPs being misused to launder criminal proceeds, and it linked weak compliance in this sector to poor results in FATF mutual evaluations. Their broader point applies well beyond West Africa: any AML system that ignores non-financial gatekeepers leaves a large door open. 2.2 The FATF architecture The engine of the global system is the #Financial_Action_Task_Force, an inter-governmental body founded in 1989. FATF does not pass laws. Instead, it issues standards, known as the Forty Recommendations, and then assesses how well countries meet them through peer reviews called #mutual_evaluations. Although the Recommendations are soft law, they carry real weight, because countries that fall short can be placed on public monitoring lists that damage their reputation and raise the cost of doing business (Benson, 2025). Three Recommendations matter most for DNFBPs. #Recommendation_22 requires DNFBPs to apply customer due diligence and record-keeping when they engage in specified activities. #Recommendation_23 extends further preventive duties to them, including the reporting of suspicious transactions and the application of internal controls and higher scrutiny for high-risk countries. Recommendation 28 requires countries to regulate and supervise DNFBPs for AML purposes. Together these provisions convert private professionals into part of the public defence against financial crime. The activities that trigger obligations differ by sector, which reflects a deliberate policy choice. Real estate agents are covered when they are involved in buying and selling property for clients. Lawyers, notaries, and accountants are covered when they carry out particular transactions on behalf of clients: buying and selling real estate, managing client money or assets, managing accounts, organising contributions for companies, and creating or managing companies, trusts, and other legal arrangements. Dealers in precious metals and stones are covered when they engage in cash transactions above a set threshold, generally the equivalent of fifteen thousand United States dollars or euros. This activity-based design tries to focus attention on the moments of highest risk rather than on the profession as a whole. 2.3 The gatekeeper concept The idea of the professional gatekeeper did not begin with FATF. It grew out of policy work in the late 1990s and early 2000s, when governments recognised that grand corruption and organised crime often depended on skilled intermediaries. A corrupt official who steals public money still needs someone to move it, hide it, and make it usable. That someone is frequently a lawyer, an accountant, or a company formation agent. Levi (2022) traces how this recognition turned into a sustained policy campaign to bring the professions into the AML net, and how the label of enabler has since attached to lawyers in particular. The gatekeeper logic is powerful, but it also carries a warning. Treating an entire profession as a suspect class can be unfair and counterproductive. Most lawyers, agents, and dealers never knowingly touch criminal money. The challenge for policy is to design rules that catch the small number of wrongdoers and negligent firms without smothering honest practitioners in cost and suspicion. That balance runs through every part of the debate examined in this article. Approach and scope This article uses a doctrinal and analytical method. It combines a reading of the relevant international standards and national frameworks with recent peer-reviewed scholarship and official risk assessments. It does not present new empirical data. Instead, it synthesises existing evidence to explain how and why AML obligations expanded to DNFBPs, and to assess the strengths and weaknesses of that expansion in three focus sectors. The scope is deliberately comparative but not exhaustive. The article draws examples mainly from the United States, the United Kingdom, the European Union, and Australia, because these jurisdictions illustrate the full range of positions, from early adoption to long resistance and recent reform. Where useful, it refers to developments in other regions. The three focus sectors, real estate agents, legal professionals, and dealers in precious metals and stones, were chosen because they represent, respectively, the largest laundering channel by value, the most ethically complex profession, and the goods most similar to cash. Casinos, accountants, and trust and company service providers are mentioned where relevant but are not analysed in depth. From banks to gatekeepers: why the system expanded 4.1 The displacement problem The clearest reason for the expansion is displacement. When one channel is closed, activity moves to another. As banks built stronger #customer_due_diligence and monitoring systems, the relative attractiveness of non-bank routes rose. Recent criminological work supports this picture of adaptive behaviour. Riccardi and Reuter (2024) show that offenders choose laundering methods based on cost, risk, and the amount and type of proceeds they hold, which means that tightening one route predictably pushes activity toward softer targets. Nazzari and Riccardi (2024), studying the behaviour of thousands of Italian offenders, found that criminals often prefer familiar, low-tech methods, including cash and real estate, close to home. If the softest targets sit outside the AML system, the system leaks. 4.2 The evidence of misuse Policy did not expand on theory alone. It responded to a steady stream of evidence that gatekeepers were being used. Investigative journalism, leaked document sets, and law enforcement cases repeatedly showed lawyers forming anonymous companies, agents processing suspect property deals, and dealers accepting large cash payments with few questions. Levi (2022) notes that until fairly recently the study of money laundering concentrated on criminal organisations and the financial sector, and that the role of professionals was surprisingly neglected given how central they are to concealment. The rise of the anti-corruption agenda changed that. Once policymakers accepted that hidden ownership and professional facilitation were core to grand corruption, extending AML duties to gatekeepers became an obvious step. 4.3 The professional launderer A further driver is the emergence of specialists who sell laundering as a service. Kramer, Blokland, Kleemans, and Soudijn (2023) describe money laundering networks in the Netherlands that behave like businesses, offering their skills to multiple criminal clients for a fee. These specialists often rely on legitimate professionals and structures to do their work. Matanky-Becker (2024) draws a helpful distinction between high-end laundering, which uses professionals and complex structures, and cash-based laundering, which relies on volume and simpler methods. The high-end variety is exactly the type that gatekeepers enable, whether knowingly or through wilful blindness. Bringing DNFBPs into the system is an attempt to disrupt the supply of professional facilitation on which high-end laundering depends. 4.4 The limits of a bank-only model Finally, regulators came to accept a structural point. A #compliance system that stops at the bank teller cannot see the transaction that never reaches a bank in an obvious form. When a house is bought with layered funds through a shell company, or when cash becomes a diamond that is later sold abroad, the bank may see only a clean-looking wire or nothing at all. The gatekeeper, by contrast, sits closer to the underlying deal and can see the red flags the bank cannot. This is the core rationale for expansion: place a set of eyes at the point where the real economy and criminal money actually meet. 4.5 A perimeter that keeps widening The expansion has not stopped with the classic gatekeepers. As criminals adopt new tools, the regulatory perimeter keeps stretching. The clearest recent example is the treatment of #virtual_asset_service_providers, such as cryptocurrency exchanges, which have been drawn into the AML framework and made subject to due diligence and reporting duties that mirror those applied to older sectors. Art and antiquities dealers are a second frontier, because high-value art shares many features with precious stones: it is portable, hard to value objectively, and often traded with limited transparency. Several jurisdictions have moved to bring high-value art dealers within their AML rules, and researchers continue to debate how far the perimeter should go. This constant widening is both a strength and a warning. It shows that the system is learning and adapting rather than standing still. But each extension also adds cost, complexity, and new groups of reluctant participants who must be trained and supervised. A perimeter that grows without matching growth in supervision risks becoming wide but shallow. The lesson from the older DNFBP sectors, examined in the rest of this article, is that adding a sector to the list is the easy part. Making the obligations meaningful within that sector is the hard part, and it is where success or failure is actually decided. Real estate agents 5.1 Why property is attractive to launderers Real estate is, in many ways, the ideal laundering vehicle. Property holds and grows value, absorbs very large sums in a single transaction, and carries social respectability. A luxury apartment does not look like crime. It looks like success. #Real_estate also offers several distinct laundering techniques. Buyers can use anonymous companies or nominees to hide who really owns an asset. They can manipulate prices, paying above or below market value to move value quietly. They can launder ongoing criminal income by collecting rent on a property bought with dirty money. And in some markets they can pay in cash or through informal channels that leave little trace. The scale is significant. Analyses of the United States market have estimated billions of dollars laundered through property over recent years, and civil society groups have documented how anonymous ownership makes it very hard to identify the true beneficiaries of high-value purchases. In Australia, the national risk assessment produced by the financial intelligence agency AUSTRAC rated the real estate market as a persistent laundering channel, and law enforcement restrained hundreds of millions of dollars in criminal assets, much of it property, over a four-year period. In the United Kingdom, transparency campaigners identified more than a billion pounds worth of property linked to individuals accused of corruption, a large share of it in London. 5.2 The regulatory response Different jurisdictions have travelled very different distances. The United Kingdom was an early adopter. Estate agents were brought into the AML regime through the Money Laundering Regulations in 2003, and later reforms extended coverage to letting agents handling higher-value rentals. British agents must register with a supervisor, carry out due diligence, and report suspicions. The European Union tightened its rules through successive Anti-Money Laundering Directives, extending obligations across property and high-value goods and setting cash payment limits, and it has since moved toward a single rulebook and a new central supervisory authority. The United States tells a more complicated story. Its foundational AML law, the Bank Secrecy Act, historically did not cover real estate professionals, who received a temporary exemption that lasted for many years. To fill part of the gap, the Financial Crimes Enforcement Network used #Geographic_Targeting_Orders from 2016, requiring title insurance companies to identify the beneficial owners behind certain high-value, all-cash purchases in named locations. In 2024, the Treasury announced a broader residential real estate reporting rule requiring certain professionals to report the beneficial owners behind non-financed transfers of residential property to legal entities and trusts. Even so, civil society analysts have noted that the United States long remained an outlier among major economies in the limited AML duties placed on real estate professionals. Australia illustrates the politics of delay. Despite being a founding FATF member, it left real estate agents, lawyers, and accountants outside its AML regime for years. Benson (2025) and Nicolas (2024) both trace this long resistance and the eventual passage of reform legislation in 2024, often called the #Tranche_2 reforms, which finally extends obligations to these sectors. The Australian case shows that even when the risk is well documented, professional lobbying and concerns about cost can hold back reform for a long time. 5.3 What the obligations look like in practice For a real estate professional, compliance means several concrete steps. Before acting, the agent should identify and verify the client and, where a company or trust is involved, identify the #beneficial_owner behind it. The agent should understand the purpose of the transaction and consider whether the source of funds is plausible given what is known about the client. Higher-risk situations, such as a politically exposed client or an opaque offshore structure, call for enhanced checks. If something looks wrong and cannot be explained, the agent should file a suspicious transaction report to the national financial intelligence unit and must avoid tipping off the client that a report has been made. Records of these checks must be kept for a set period, usually five years. The practical difficulty is that agents are salespeople, not investigators. Their business depends on closing deals, which creates a natural tension with the duty to slow down and question a lucrative buyer. This is why supervision and clear guidance matter so much in this sector. Without them, the obligations risk becoming a box-ticking exercise that criminals learn to satisfy on paper. 5.4 Red flags in property transactions Effective #compliance in real estate depends on recognising warning signs. Several patterns recur across cases and typology studies. A buyer who insists on paying wholly or partly in cash, or who is willing to complete a purchase without financing, removes the natural scrutiny that a mortgage lender would apply. A price that does not match the true value of a property, whether inflated or deflated, can be a way of moving value between parties without an obvious transfer. Purchases made through #shell_companies, trusts, or a chain of intermediaries, especially those registered in secrecy-friendly jurisdictions, can hide the real buyer. Sudden changes in the parties to a deal, a buyer who shows no interest in the property itself, funds arriving from unrelated third parties, and a client who is reluctant to explain the source of wealth are all signals that something may be wrong. None of these signs proves wrongdoing on its own. Many have innocent explanations, and treating every cash buyer as a criminal would be both unfair and impractical. The professional judgement required is to weigh the whole picture and to ask questions when the story does not add up. Nazzari and Riccardi (2024) note that offenders often prefer familiar methods close to home, which means that many laundering purchases look ordinary on the surface. That is precisely why trained human judgement, rather than a simple checklist, remains the frontline defence in property markets. Legal professionals 6.1 The special role of lawyers Lawyers occupy a unique position in the AML debate because they do more than move money. They design the structures through which money moves. A lawyer can form a company, create a trust, arrange the purchase of property, and manage client funds through a pooled account. These are legitimate and valuable services. They are also, in the wrong hands, the building blocks of concealment. The same skill that protects a client's lawful interests can, if misused, hide the true owner of an asset or launder the proceeds of crime. Because of this, FATF Recommendation 22 covers legal professionals when they carry out specific high-risk activities, such as buying and selling real estate, managing client assets and accounts, organising company contributions, and creating or managing companies and trusts. When engaged in these activities, lawyers are expected to perform #customer_due_diligence, keep records, and, in most jurisdictions, report suspicious transactions. 6.2 The privilege problem The obligation to report is where the deepest conflict arises. Lawyers owe their clients duties of confidentiality and loyalty, and in many systems certain communications are protected by #legal_professional_privilege or professional secrecy. These protections are not mere professional courtesies. They support the right to legal advice and the proper functioning of the justice system. A client who fears that a lawyer will report every conversation to the state may not seek advice at all, which harms both the client and the rule of law. The AML system tries to manage this tension rather than resolve it completely. Privilege generally protects communications connected to the giving of legal advice or to litigation, but it does not protect a lawyer who is being used, knowingly or not, as an instrument of crime. Most systems recognise an exception where the professional relationship is being abused to further a criminal purpose. In Australia, Nicolas (2024) describes a mechanism under recent reform legislation that requires a lawyer to make a positive assessment of whether privilege applies and, where it does, to file a special form disclosing that privilege is claimed rather than the underlying information. This is an attempt to preserve privilege while preventing it from becoming a blanket shield. 6.3 The enabler debate Perhaps no question in this field is more contested than whether lawyers should be seen as enablers of laundering. Levi (2022) offers a careful analysis and warns against crude generalisation. He argues that the profession is highly diverse, ranging from sole practitioners to large international firms, and that the evidence rarely shows lawyers doing anything beyond their ordinary business of setting up lawful structures that happen to reduce outside scrutiny. He also notes how hard it is to tell apart the naive, the negligent, the wilfully blind, and the genuinely criminal lawyer. Calling the whole profession enablers, he suggests, overstates the known facts and may distract from the specific conduct that truly deserves sanction. At the same time, the record contains real cases of legal professionals who crossed the line, and official assessments in several countries continue to rate the misuse of lawyers as a significant threat. The honest conclusion is that both extremes are wrong. Most lawyers are not launderers, but the small number who are, together with those who look away, can cause serious harm because of the powerful tools they command. 6.4 Resistance and reform Legal professions have often resisted AML duties, and their reasons vary in quality. Some concerns are serious, such as the protection of privilege and the independence of the profession. Others are weaker, such as the claim that lawyers rarely appear in laundering schemes, which sits uneasily with decades of evidence. In the United States, a legislative proposal known as the #ENABLERS_Act sought to extend AML obligations to lawyers and other gatekeepers. It passed the House of Representatives in 2022 but was blocked in the Senate later that year, in part because of opposition from the organised bar citing privilege. This episode shows how strong professional lobbying can be, and how far some major economies still are from full coverage of the legal sector. 6.5 The problem of intent A final complication concerns the mental element of the offence. In some jurisdictions, notably the United Kingdom, a professional can face liability for money laundering offences on the basis of objective suspicion, meaning that there were reasonable grounds to suspect the origin of funds, even without proof of actual knowledge. This lowers the bar for liability and pushes professionals toward caution, but it also raises fairness concerns and can encourage defensive over-reporting. The result is a system in which large numbers of reports are filed, but their quality and usefulness are open to question, a point that reappears in the discussion of effectiveness below. 6.6 Client accounts, notaries, and the tools of concealment Two features of legal practice deserve special attention because they appear again and again in laundering cases. The first is the client account. Lawyers routinely hold client money in pooled accounts as part of ordinary transactions, such as conveyancing or the settlement of a deal. Because funds passing through a respected law firm's account inherit the firm's good reputation, criminals have used these accounts to give dirty money a clean history. A payment that would raise questions if it came directly from an unknown person looks far less suspicious when it arrives from a solicitor's account. This is why regulators pay close attention to how legal professionals handle client funds and why misuse of client accounts is a well-documented red flag. The second feature is the role of notaries and company formation. In civil law countries, notaries authenticate important transactions, including property transfers and company creation, which places them at the same sensitive points as lawyers in common law systems. The creation and management of companies and trusts is central to concealment, because these structures can separate the visible owner from the real controller. When a legal professional forms an entity, identifies its directors, and never asks who truly benefits, the resulting structure becomes a ready-made tool for hiding #beneficial_ownership. Bringing legal professionals into the AML system is, in large part, an attempt to place a check at exactly these moments, when otherwise legitimate legal machinery is being built and could be turned to unlawful ends. Dealers in precious metals and stones, including jewelers 7.1 Goods that behave like cash Gold, diamonds, and other precious stones share a set of features that make them almost perfect laundering instruments. They pack enormous value into a small physical space, they hold value across time and across borders, and they can be sold in many markets around the world. A diamond is a liquid and transferable store of value, which is exactly what a launderer needs to move wealth quietly. Because much of the trade has historically involved cash, opaque supply chains, and lightly regulated free-trade zones, it has offered criminals both anonymity and mobility. FATF has named dealers in #precious_metals and stones as DNFBPs and issued specific risk-based guidance for the sector. The typologies are well documented. Criminal groups have used diamond parcels to move value across borders, purchased high-value jewellery at auction using structured cash payments to avoid reporting thresholds, and run gold smuggling operations that route illicitly sourced metal through intermediary hubs before it re-enters the legitimate market. Jewelers sit at the retail end of this chain, where dirty cash can be turned into a wearable, portable, high-value object in a single transaction. 7.2 The regulatory response Coverage of this sector is uneven. In the United States, the Financial Crimes Enforcement Network issued a rule requiring dealers in precious metals, stones, and jewels to establish AML programs once their relevant purchases exceed a set annual threshold, with an exception for smaller retailers. Such a program must include internal policies and controls based on a risk assessment, a designated compliance officer, ongoing staff training, and independent testing. In the European Union, dealers in high-value goods are drawn into the AML regime where they accept large cash payments, with a reporting trigger set at ten thousand euros. The United Kingdom maintains broadly equivalent requirements for high-value dealers, and Switzerland, home to major watch and jewellery centres, applies AML duties to dealers in precious metals and stones under its own legislation. The common thread is the focus on cash. Because the sector's main vulnerability is the conversion of large amounts of physical currency into portable goods, most regimes concentrate their obligations on cash transactions above a threshold. This targeting reflects the risk-based logic of the whole DNFBP system, but it also creates a well-known weakness: launderers can break a single large purchase into several smaller ones to stay below the trigger, a technique known as structuring. 7.3 What compliance requires A responsible dealer or jeweler should apply due diligence to customers involved in large or unusual transactions, verify identity, and question the source of funds where the pattern looks suspicious. Dealers should refuse or report large cash payments that lack documentation, apply enhanced scrutiny to politically exposed persons and high-risk clients, and keep proper records. Where the pattern suggests structuring or where a customer is unwilling to explain the source of wealth, the transaction should be treated as a red flag. As with real estate, the practical barrier is commercial. A jeweler focused on making a sale may be reluctant to interrogate a wealthy customer paying promptly, which is why supervision, training, and a genuine compliance culture are decisive in this sector. 7.4 Free-trade zones, borders, and opaque supply chains The precious goods sector carries an extra layer of risk that the property and legal sectors do not share to the same degree: physical mobility. A house cannot be carried across a border, but a parcel of diamonds or a quantity of gold can. This mobility interacts with two structural weaknesses. The first is the free-trade zone, an area where goods can be stored and traded with reduced customs and documentation requirements. These zones serve legitimate commercial purposes, but the very features that make them efficient, light paperwork and limited oversight, also make them attractive for moving value with few questions asked. The second weakness is the opaque supply chain. Metals and stones pass through miners, brokers, cutters, refiners, and manufacturers before reaching a retail jeweler, and at each hand-off the origin of the goods can become harder to trace. Together these features allow criminals to convert dirty cash into a portable asset, move it across #terrorist_financing and money laundering hotspots, and reintroduce it into the legitimate market far from where the crime occurred. Documented typologies include gold sourced from conflict or illegal mining that is smuggled through intermediary hubs before entering mainstream refineries, and gemstones used to transfer value between jurisdictions with little transparency. For dealers, this means that #customer_due_diligence cannot stop at the person standing in front of them. Responsible sourcing, questions about the origin of goods, and awareness of high-risk supply routes are all part of managing the sector's true risk. A jeweler who checks the buyer but never questions where the stock came from has covered only half of the danger. The common toolkit: obligations that cut across all DNFBPs Although the sectors differ, the obligations placed on them share a common structure. Understanding this shared toolkit is the key to understanding the whole system. 8.1 The risk-based approach At the centre of modern AML policy is the #risk_based_approach. Rather than applying identical checks to every client and transaction, a business is expected to assess where its real risks lie and to concentrate its effort accordingly. A small local business with low-value, local clients faces different risks from a firm handling large cross-border deals for offshore companies. The risk-based approach allows resources to flow to the areas of greatest danger, but it also demands judgement, which is harder to supervise than a simple rule and can be gamed by firms that prefer to see little risk in their most profitable clients. 8.2 Customer due diligence and beneficial ownership The most basic duty is customer due diligence, sometimes called #know_your_customer. This means identifying the client, verifying that identity with reliable information, understanding the purpose of the relationship, and monitoring it over time. Gaviyau and Sibindi (2023), drawing on evidence from South Africa, underline that due diligence is the frontline control on which the rest of the system depends, and that weaknesses here undermine everything built on top of it. A central and difficult part of due diligence is identifying the #beneficial_owner, meaning the real human being who ultimately owns or controls a client or an asset. Criminals hide behind layers of companies and trusts precisely to obscure this. Countries have responded by building beneficial ownership registers, though their design remains contested. In the European Union, a major court ruling found that giving the general public unrestricted access to such registers went too far in interfering with privacy rights, which forced a rethink of how access should be granted. The episode shows that transparency and privacy must be balanced rather than treated as opposites. 8.3 Suspicious transaction reporting When a professional cannot explain away a red flag, the duty is to file a suspicious transaction or suspicious activity report to the national financial intelligence unit. These reports feed intelligence to law enforcement. A related rule, the prohibition on tipping off, forbids the professional from warning the client that a report has been made, so that investigations are not compromised. Reporting is the point where a private gatekeeper becomes an active part of public defence, and it is also where the system's effectiveness is most sharply debated, because large volumes of low-quality reports can overwhelm the very agencies they are meant to help. 8.4 Internal controls, records, and training Finally, DNFBPs are expected to build internal systems that make compliance routine rather than accidental. This typically includes written policies and controls, a designated compliance officer or money laundering reporting officer, ongoing training so that staff can recognise red flags, independent testing of the program, and record-keeping for a defined period. For a large firm these duties are manageable. For a sole practitioner or a small shop they can feel heavy, which brings the analysis to the tensions that surround the whole expansion. 8.5 Politically exposed persons and sanctions screening Two further duties have grown in importance and now cut across all DNFBP sectors. The first concerns #politically_exposed_persons, often shortened to PEPs. These are individuals who hold or have held prominent public functions, together with their close family and associates. Because such people can be exposed to bribery and corruption, they are treated as higher risk, and dealing with them triggers enhanced #customer_due_diligence, senior sign-off, and closer questioning of the source of their wealth. The point is not that every PEP is corrupt, which is plainly untrue, but that the category deserves extra care given what history has shown about the laundering of the proceeds of corruption. The second duty is #sanctions_screening. Global events, including the response to the invasion of Ukraine, have pushed financial sanctions to the centre of policy, and professionals are now expected to check that they are not helping a sanctioned person move or hide assets. Benson (2025) describes how the shift in language from money laundering to the broader idea of illicit finance has widened what gatekeepers are asked to watch for, folding sanctions evasion, corruption, and proliferation financing into the same set of duties. For a real estate agent, a lawyer, or a jeweler, this means the compliance task is no longer only about the origin of funds. It is also about the identity and status of the people involved, which adds both protective value and further burden. Tensions, criticisms, and unintended effects 9.1 Cost and proportionality The first and most practical objection is cost. AML compliance requires time, systems, and expertise, and the burden falls hardest on small firms and sole practitioners who lack dedicated compliance staff. If the rules are not proportionate, they can push honest small operators out of business or into superficial compliance, while doing little to stop determined criminals who can afford sophisticated concealment. A well-designed system must scale its demands to the real risk each business poses, which is easier to state than to achieve. 9.2 The effectiveness question A deeper criticism concerns results. After decades of AML expansion, the evidence that these controls actually reduce crime remains surprisingly thin. Levi (2022) warns that much anti-money laundering activity may amount to what he calls #control_mimicry, producing the appearance of action, political rhetoric, and paperwork rather than measurable reductions in serious crime. Recent empirical work continues to probe this question. Andiojaya (2025) examines whether stronger AML measures actually reduce offences such as corruption, bribery, and environmental crime, and Bramantyo (2025) explores why convictions for money laundering remain so low despite extensive rules. Morgan (2024), using a large data linkage study, connects money laundering to the wider harms of organised crime, which strengthens the case for action while also highlighting how hard it is to measure success. The uncomfortable truth is that we impose large compliance costs on the basis of a plausible theory that remains only partly tested. 9.3 Reporting without intelligence Related to effectiveness is the problem of report quality. When professionals face liability for failing to report and can be sanctioned on the basis of objective suspicion, the rational response is to report defensively. This produces high volumes of reports of uneven value. Financial intelligence units can drown in filings, and the genuinely important signals can be lost in the noise. Bociga, Lord, and Bellotti (2024) study information and intelligence sharing within the United Kingdom's AML regime and highlight how much depends on cooperation and trust between the public and private sectors. Reports only add value if the system can process them and feed insight back to the professionals who file them. Without that feedback loop, reporting becomes an end in itself. 9.4 The privilege and independence dilemma For lawyers, as discussed, the tension between reporting duties and privilege is not merely procedural. It touches the independence of the legal profession and the client's ability to obtain candid advice. Nicolas (2024) frames this as a genuine balance to be struck rather than a conflict to be won by one side. Push too far toward reporting, and the profession's core function is compromised. Push too far toward privilege, and the profession becomes a safe harbour for concealment. The mechanisms now emerging, such as narrowly defined exceptions and privilege declaration forms, are attempts to hold this balance, but they remain untested over the long term. 9.5 De-risking and exclusion A further unintended effect is #de_risking, where firms avoid whole categories of client rather than manage the risk they pose. Faced with heavy obligations and severe penalties, a business may simply refuse to serve clients from certain countries, sectors, or backgrounds. This can push legitimate customers out of the formal economy and into less transparent channels, which is the opposite of what AML policy intends. De-risking shows how controls aimed at criminals can harm the innocent if they are applied bluntly. 9.6 Uneven global implementation Finally, the system is only as strong as its weakest link, and implementation across countries is deeply uneven. Nduka and Sechap (2021) document weak DNFBP compliance and poor mutual evaluation outcomes across a range of African jurisdictions, and similar gaps exist elsewhere. Criminals exploit these differences, routing money through jurisdictions where gatekeepers face little real scrutiny. Jofre, Bosisio, and Riccardi (2024) show how complex ownership structures in secrecy-friendly environments can be analysed with modern tools, but the underlying problem is political and legal rather than technical. Until coverage and enforcement are consistent, the expansion of duties in strict jurisdictions can simply divert activity to lax ones. A comparative snapshot Comparing jurisdictions helps show that the DNFBP expansion is a spectrum rather than a single model. The United Kingdom is a comparatively early and comprehensive adopter. It brought estate agents, legal and accountancy professionals, high-value dealers, and trust and company service providers into its regime from 2003, with later refinements, and it operates a network of sector supervisors. Its lawyers file suspicious activity reports and appoint reporting officers, though the volume and quality of reporting, and the burden on smaller firms, remain live issues. The #European_Union has progressively tightened its rules through successive directives and is now moving toward a harmonised single rulebook and a central #Anti_Money_Laundering_Authority to reduce the gaps between member states. Its experience with beneficial ownership registers, including the court ruling on public access, shows the difficulty of balancing transparency against privacy. The United States presents a patchwork. Its banking-focused #Bank_Secrecy_Act long left major gatekeeper sectors, especially real estate and the legal profession, lightly regulated. Targeted measures such as Geographic Targeting Orders and the newer residential real estate reporting rule have begun to close the property gap, but efforts to bring lawyers fully into the system have met strong resistance, as the fate of the ENABLERS Act shows. Australia is the classic late reformer. Long criticised for leaving real estate agents, lawyers, and accountants outside its regime, it enacted reform legislation in 2024 that extends obligations to these sectors, a change analysed in detail by Benson (2025) and Nicolas (2024). Its journey shows both the strength of professional resistance and the eventual pull of international expectations. Other jurisdictions illustrate the reputational stakes. Some economies with major trading centres for property, precious metals, or company formation have faced international pressure to strengthen DNFBP supervision, and improvements in this area have been central to their efforts to satisfy FATF reviewers. The lesson across all these cases is consistent: the direction of travel is toward broader coverage, but the pace, design, and enforcement vary widely, and those variations are exactly what sophisticated launderers exploit. Discussion: what the expansion can and cannot achieve Pulling these threads together, several conclusions emerge. First, the expansion is justified in principle. The displacement of laundering from banks to gatekeepers is real, the evidence of misuse across real estate, legal services, and precious goods is substantial, and a bank-only system leaves obvious gaps. Placing observers at the point where criminal money meets the real economy is a sound idea. The theory behind the DNFBP regime is stronger than its critics sometimes allow. Second, the expansion is not self-executing. Rules on paper mean little without supervision, guidance, and a genuine compliance culture within each sector. The recurring pattern across countries is that formal coverage arrives long before effective supervision does. A country can claim to regulate its estate agents or jewelers while providing almost no meaningful oversight, in which case the obligations become symbolic. Effectiveness depends less on the existence of duties than on whether anyone checks that they are met. Third, measurement must improve. The field has expanded faster than its evidence base. As Levi (2022), Morgan (2024), Andiojaya (2025), and Bramantyo (2025) each suggest in different ways, we still lack clear proof of how much crime these controls prevent, and low conviction rates raise hard questions. Continuing to add obligations without measuring their impact risks entrenching an expensive system that satisfies auditors more than it protects society. Honest measurement, including the willingness to find that some measures do not work, is essential to good policy. Fourth, proportionality and fairness are not optional extras. The burden on small firms, the risk of de-risking, and the tension with legal privilege are not merely complaints from regulated sectors. They are signs that a blunt system can cause real harm, excluding legitimate people from the economy and weakening the independence of professions that matter to the rule of law. A mature AML regime treats these concerns as design constraints, not as obstacles to be dismissed. Finally, the global dimension is decisive. Because money moves across borders, the value of strict rules in one country is limited by weak rules in another. The uneven implementation documented in the literature is the system's central vulnerability. Progress therefore depends as much on raising the floor in weak jurisdictions as on adding new duties in strong ones. Recommendations Several practical directions follow from this analysis. The first is to match effort to risk. Supervisors should hold firms to genuine, risk-based standards rather than uniform checklists, and should focus scrutiny on the transactions and clients that carry the highest danger, such as opaque offshore structures, politically exposed persons, and large cash or non-financed deals. This concentrates protection where it matters and reduces wasteful burden elsewhere. The second is to invest in supervision and feedback. Coverage on paper should be matched by real oversight, clear sector-specific guidance, and a working feedback loop so that professionals learn which reports were useful and why. Bociga, Lord, and Bellotti (2024) show that intelligence sharing is central to a functioning system, and feedback is what turns reporting from a compliance ritual into a source of usable intelligence. The third is to protect the legitimate core of each profession. For lawyers in particular, reforms should preserve genuine legal professional privilege and professional independence while closing the abuse of that privilege for concealment. Narrow, well-defined exceptions and transparent privilege declaration mechanisms, of the kind analysed by Nicolas (2024), are more sustainable than blunt reporting mandates that treat every client interaction as suspect. The fourth is to attack anonymity at its root. Much laundering through gatekeepers depends on hidden ownership. Reliable, well-verified beneficial ownership information, made available to competent authorities and, within privacy limits, to obliged professionals, would make the gatekeeper's job far easier. The goal should be accurate registers with strong verification, designed to respect the privacy limits identified by courts while still defeating anonymous concealment. The fifth is to build the evidence base. Governments and researchers should treat effectiveness as an open empirical question and fund the studies needed to answer it. If a measure cannot be shown to reduce harm, it should be revised or removed. Policy should follow evidence rather than assumption. The sixth is to work on the global floor. Technical assistance, honest mutual evaluation, and cooperation should aim to raise standards in weak jurisdictions, because the whole system is only as strong as its most permissive gate. Strengthening one country while ignoring another simply relocates the problem. Conclusion The story of DNFBP regulation is the story of a system learning that the front door is not the only way in. For years, the fight against dirty money watched the banks and assumed the rest of the economy could look after itself. Criminals proved otherwise, moving their wealth through property, legal structures, and high-value goods with the help, whether knowing or not, of trusted professionals. The regulatory response, driven by the FATF standards and adopted unevenly around the world, has been to turn these gatekeepers into part of the defence. Real estate agents, lawyers, and dealers in precious metals and stones now carry duties that would have seemed unthinkable a generation ago: to know their customers, to identify hidden owners, to report suspicion, and to keep records for the state. This expansion is necessary. The gaps it addresses are real, and the evidence of misuse is too strong to ignore. But necessity is not the same as success. The obligations only protect society if they are proportionate, properly supervised, fairly applied, and honestly measured. Where they become paperwork, they impose cost without delivering safety. Where they ignore privilege and fairness, they can damage the professions and people they touch. And where they stop at national borders, they merely move the problem elsewhere. The task for the next phase of policy is not simply to add more gatekeepers to the list, but to make the gatekeeping real. That means turning the promise of the DNFBP regime, that trusted professionals will help keep dirty money out of the legitimate economy, into a result that can actually be seen and proven, rather than a hope that lives only on the compliance file. Hashtags #DNFBPs #Anti_Money_Laundering #Designated_Non_Financial_Businesses_and_Professions #FATF #Gatekeepers #Real_Estate_Money_Laundering #Legal_Professional_Privilege #Precious_Metals_and_Stones #Jewelers_AML #Customer_Due_Diligence #Beneficial_Ownership #Suspicious_Transaction_Reporting #Risk_Based_Approach #Terrorist_Financing #Financial_Crime_Compliance References Andiojaya, A. (2025). Do stronger anti-money laundering (AML) measures reduce crime? An empirical study on corruption, bribery, and environmental crime. Journal of Economic Criminology, 8, 100157. https://doi.org/10.1016/j.jeconc.2025.100157 Benson, K. (2025). From money laundering to illicit finance? The evolving AML regulatory regimes for legal professionals in the UK and Australia. Current Issues in Criminal Justice, 37(2), 245-263. https://doi.org/10.1080/10345329.2024.2443702 Bociga, D., Lord, N., & Bellotti, E. (2024). Dare to share: Information and intelligence sharing within the UK's anti-money laundering regime. Policing and Society. Advance online publication. https://doi.org/10.1080/10439463.2024.2427317 Bramantyo, M. (2025). Where do you hide your money? The explanation of low conviction in money laundering. Journal of Economic Criminology, 8, 100161. https://doi.org/10.1016/j.jeconc.2025.100161 Bullough, O. (2022). Butler to the world: How Britain became the servant of tycoons, tax dodgers, kleptocrats and criminals. Profile Books. Financial Action Task Force. (2022). Horizontal review of gatekeepers' technical compliance related to corruption. FATF. Financial Action Task Force. (2023). International standards on combating money laundering and the financing of terrorism and proliferation: The FATF Recommendations (updated 2023). FATF. Gaviyau, W., & Sibindi, A. B. (2023). Anti-money laundering and customer due diligence: Empirical evidence from South Africa. Journal of Money Laundering Control, 26(7), 224-238. https://doi.org/10.1108/JMLC-06-2023-0103 Jofre, M., Bosisio, A., & Riccardi, M. (2024). Financial crime risk assessment: Machine learning insights into ownership structures in secrecy firms. Global Crime, 25(3-4), 242-267. https://doi.org/10.1080/17440572.2024.2416229 Kramer, J.-A., Blokland, A. A. J., Kleemans, E. R., & Soudijn, M. R. J. (2023). Money laundering as a service: Investigating business-like behavior in money laundering networks in the Netherlands. Trends in Organized Crime, 27(3), 314-341. https://doi.org/10.1007/s12117-022-09475-w Levi, M. (2022). Lawyers as money laundering enablers? An evolving and contentious relationship. Global Crime, 23(2), 126-147. https://doi.org/10.1080/17440572.2022.2089122 Matanky-Becker, R. (2024). High-end and cash-based money laundering: Defining and disaggregating complex phenomena. European Journal on Criminal Policy and Research, 30(3), 421-433. https://doi.org/10.1007/s10610-023-09562-w Morgan, A. (2024). Money laundering and the harm from organised crime: Results from a data linkage study. Australian Institute of Criminology. https://doi.org/10.52922/sp77628 Nazzari, M., & Riccardi, M. (2024). Cleaning mafia cash: An empirical analysis of the money laundering behaviour of 2800 Italian criminals. European Journal of Criminology, 21(4), 583-608. https://doi.org/10.1177/14773708231207877 Nduka, B. E., & Sechap, G. (2021). Refocusing designated non-financial businesses and professions on the path of anti-money laundering and combating the financing of terrorism compliance. Journal of Money Laundering Control, 24(4), 693-711. https://doi.org/10.1108/JMLC-11-2020-0125 Nicolas, I. (2024). Striking the balance: Anti-money laundering goals and legal privilege in Australia. Current Issues in Criminal Justice, 37(2), 208-227. https://doi.org/10.1080/10345329.2024.2420146 Riccardi, M., & Reuter, P. (2024). The varieties of money laundering and the determinants of offender choices. European Journal on Criminal Policy and Research, 30(3), 333-358. https://doi.org/10.1007/s10610-024-09573-8

The Convention on the Recognition and Enforcement of Foreign Arbitral Awards, signed in New York on 10 June 1958 and widely known as the New York Convention, is the single most important treaty behind the modern practice of international #arbitration. This article explains, in plain language but with academic depth, why the Convention is often called the engine that powers cross border commerce. It traces the treaty from its origins in the weak Geneva instruments of the 1920s, describes its structure and its famous pro enforcement design, and analyses the limited grounds on which a court may refuse to enforce a foreign award. The study pays particular attention to Article V, the heart of the Convention, and to the two grounds that courts may raise on their own motion, namely #arbitrability and #public_policy. It then compares how leading jurisdictions, including the United States, England, France, India, Singapore and China, have applied the treaty, and it considers contemporary pressures such as awards set aside at the seat, sovereign immunity, sanctions, and the rise of technology in dispute resolution. The article argues that the durability of the #New_York_Convention rests on a simple but powerful bargain: states surrender a measure of judicial control in exchange for a predictable global system in which an #arbitral_award earned in one country can be turned into money almost anywhere. For students of international commercial law, understanding this bargain is the key to understanding why arbitration has become the default method for resolving cross border disputes. Keywords New York Convention 1958, international arbitration, recognition and enforcement, foreign arbitral awards, Article V, public policy, arbitrability, pro enforcement bias, seat of arbitration, UNCITRAL Model Law, party autonomy, cross border dispute resolution. 1. Introduction When two businesses from different countries sign a contract, they face a quiet but serious problem. If a dispute arises, whose courts will decide it, and will the winner ever collect what is owed? Litigation in a foreign court can feel like playing an away game against the home team. Judgments from national courts have historically been difficult to enforce abroad, because there was no broad global treaty that required one country to respect the court decisions of another. This is the practical gap that #arbitration was built to fill, and it is the gap that the #New_York_Convention closed. The genius of international #arbitration is not the hearing itself. Private tribunals have existed for centuries. The genius is #enforceability. A commercial party will only agree to arbitrate if it can trust that the resulting decision, the #arbitral_award, will be honoured. Without a reliable way to convert an award into cash or seized assets, arbitration would be little more than an expensive form of private advice. The 1958 Convention supplies that reliability by binding its member states to a common promise: they will recognise and enforce foreign awards, and they will refuse to do so only on a short, closed list of grounds. This is why the treaty is so often described as the engine of the whole system. An engine converts fuel into motion. The Convention converts an arbitration clause, a few lines of text buried in a contract, into a legally powerful instrument that can travel across borders. The #arbitration_agreement is the fuel, the arbitral proceeding is the mechanism, and #enforcement under the Convention is the drive shaft that puts the result on the road. Remove the Convention and the machine still turns, but it goes nowhere useful. The consequences of this arrangement reach far beyond the courtroom. Surveys of international businesses consistently show that arbitration is the preferred method for resolving cross border commercial disputes, and the reason most often given is the ease of enforcing the resulting award compared with a court judgment. That preference has fed a steady rise in the caseloads of the major arbitral institutions and has turned international arbitration into a substantial global industry of arbitrators, counsel, and specialist institutions. None of this ecosystem would exist without the underlying guarantee of enforceability. In this sense the Convention is a piece of legal infrastructure, as important to global trade as shipping lanes or payment systems, even though most people who benefit from it will never read a word of its text. It quietly underwrites the confidence that allows a company in one country to do business with a stranger in another, knowing that if things go wrong there is a neutral process whose result can actually be enforced. The scale of the treaty's success is remarkable. From a small group of original signatories, membership has grown to more than 170 states, making the Convention one of the most widely adopted commercial treaties in history (ICCA, 2024). Very few instruments of private international law can claim such near universal reach. That breadth is precisely what gives an award its value, because an #award_creditor can pursue a losing party wherever that party keeps its money. This article is written for students who need a clear map of how the Convention works and why it matters. It moves through the treaty in a logical order. It begins with history, because the Convention only makes sense once you see the failures it was designed to fix. It then examines the structure of the treaty, the machinery of #recognition and #enforcement, and the defences that a resisting party may raise. After that it turns to comparative practice, showing how the same words have been read differently in different legal cultures, before closing with the modern challenges that will shape the treaty's second century. Throughout, the aim is to keep the language simple while respecting the seriousness of the subject. 2. Historical Evolution: From Geneva to New York The Convention did not appear from nowhere. It was the third and most successful attempt by the international community to make foreign #arbitral_awards enforceable. To appreciate what the drafters achieved, we must first look at what came before. The first serious effort was the Geneva Protocol on Arbitration Clauses of 1923. This instrument, developed under the League of Nations, did something modest but important. It required contracting states to recognise the validity of #arbitration_agreements in commercial matters and to allow arbitrations to proceed. The problem was that recognising the agreement to arbitrate is only half the battle. A party still needed a way to enforce the award that came out of the process. The second attempt tried to fill that hole. The Geneva Convention on the Execution of Foreign Arbitral Awards of 1927 addressed #enforcement directly, but it was burdened by two crippling defects. The first was a requirement often called double exequatur. Before an award could be enforced abroad, the winning party had to prove that the award had become final in the country where it was made, which usually meant obtaining a confirmation from the courts of that country first, and then seeking enforcement in the second country. In effect the #award_creditor had to win twice, once at home and once abroad. This doubled the cost, doubled the delay, and multiplied the opportunities for a determined loser to obstruct the process. The second defect was the burden of proof. Under the Geneva system, the party seeking #enforcement carried the heavy load of proving that all the conditions for enforcement were satisfied. The resisting party could simply sit back and demand proof. Together, these two features made the 1927 Convention slow and unpredictable, and international business grew frustrated. By the early 1950s the appetite for reform was strong. The International Chamber of Commerce prepared an ambitious draft in 1953 that would have created a genuinely international award, largely detached from any national legal order. The United Nations Economic and Social Council took a more cautious line and reshaped the draft to focus on foreign awards rather than truly stateless ones. The revised text became the working document for a diplomatic conference held at United Nations Headquarters from 20 May to 10 June 1958. One delegate famously called the result a bold innovation, and history has proved him right (ICCA, 2024). The 1958 conference made two decisive changes that transformed the field. First, it abolished double exequatur. Under the new treaty an award need only be binding on the parties, not final in the old sense, and there is no need to obtain confirmation at the seat before seeking enforcement elsewhere. Second, it reversed the burden of proof. The party seeking #enforcement now enjoys a presumption in its favour and need only present a limited set of documents. It is the party resisting enforcement who must prove that one of the narrow grounds for refusal applies. These two shifts, small on paper, produced an enormous change in practice, and they are the foundation of what scholars call the pro enforcement bias of the Convention (Born, 2021). The treaty was adopted on 10 June 1958 and entered into force on 7 June 1959. Its promotion later became part of the work of the United Nations Commission on International Trade Law, and its interpretation has been reinforced by the widely adopted #UNCITRAL Model Law on International Commercial Arbitration, first issued in 1985 and revised in 2006. The Model Law and the Convention work as partners. The Model Law shapes national arbitration statutes so that the rules for making, challenging and enforcing awards look broadly similar across borders, while the Convention guarantees the international travel of the award itself (Blackaby, Partasides and Redfern, 2022). 3. Architecture and Scope of the Convention For such an influential instrument, the Convention is strikingly short. It contains only sixteen articles, and the operative core sits in the first seven. Understanding its scope means understanding exactly which awards and which agreements the treaty covers. Article I sets the boundaries. The Convention applies to the #recognition and #enforcement of arbitral awards made in the territory of a state other than the state where enforcement is sought. It also applies to awards that are not considered domestic in the enforcing state, a category that gives some jurisdictions flexibility to treat certain locally made but internationally flavoured awards as covered. Article I also confirms that awards include not only those made by arbitrators appointed for a single case but also those made by permanent arbitral bodies, which captures the awards of established institutions. Article I also permits two reservations. Under the #reciprocity reservation, a state may declare that it will apply the Convention only to awards made in the territory of another contracting state. Under the commercial reservation, a state may declare that it will apply the treaty only to disputes considered commercial under its own law. Many states have made one or both reservations, and a careful practitioner always checks which reservations the relevant country has entered, because they affect whether the Convention route is available at all (ICCA, 2024). Article II shifts attention to the #arbitration_agreement itself. It requires each contracting state to recognise a written agreement to arbitrate, and it obliges a court, when a party brings a lawsuit in a matter covered by such an agreement, to refer the parties to arbitration unless it finds the agreement null and void, inoperative, or incapable of being performed. This provision is the front door of the whole system. Article V may govern the exit, but Article II governs the entrance, because it forces disputing parties out of national courts and into the arbitral process they chose. The written form requirement in Article II has been read generously over time, and in 2006 UNCITRAL issued a recommendation encouraging courts to interpret it in a flexible, modern way that reflects how commercial parties actually communicate (Blackaby, Partasides and Redfern, 2022). Two principles that make Article II work deserve mention, even though they are not spelled out in the text. The first is separability, the idea that an arbitration clause is treated as an agreement independent of the main contract, so that an attack on the validity of the contract does not automatically destroy the clause. The second is competence competence, the principle that an arbitral tribunal has the power to rule on its own #jurisdiction in the first instance. Together these doctrines protect #party_autonomy by preventing a reluctant party from stalling arbitration simply by alleging that the underlying contract is flawed. Article III states the central obligation in a single sentence: each contracting state shall recognise arbitral awards as binding and enforce them in accordance with its own rules of procedure, and it must not impose substantially more onerous conditions on the enforcement of Convention awards than it imposes on domestic awards. This is a promise of both #enforcement and equal treatment. The Convention does not create a uniform global procedure for collecting on an award. Instead it tells each country to use its own machinery, but to keep the door at least as wide open for foreign awards as for local ones. It is worth pausing on how the Convention fits together with the UNCITRAL Model Law, because the two instruments are frequently confused. The Convention is a treaty that governs the international life of an award, its recognition and enforcement across borders. The Model Law is not a treaty at all. It is a template that individual countries can copy into their own arbitration statutes. When a country adopts the Model Law, its domestic rules for how an arbitration is conducted, and the grounds on which its own courts may annul an award made on its soil, come to mirror the grounds in Article V of the Convention almost word for word. The effect is a neat symmetry. The reasons a court at the #seat_of_arbitration may set an award aside line up with the reasons a court abroad may refuse to enforce it. A party therefore faces broadly the same test whether it attacks the award at home or resists it abroad. This alignment, achieved without amending the Convention, is one of the main reasons the modern system feels so coherent, and it explains why so many national arbitration laws now read like close cousins of one another (Blackaby, Partasides and Redfern, 2022). 4. The Enforcement Machinery: Articles III and IV Once a party holds a favourable #arbitral_award, how does it actually turn that paper into a court order that can seize assets? The Convention makes this process deliberately simple, and Article IV is where the simplicity lives. Article IV lists the documents the applicant must supply. The party seeking #enforcement need only produce the authenticated original award or a certified copy, and the original #arbitration_agreement or a certified copy. If the award or agreement is not in the official language of the enforcing country, a certified translation may be required. That is the entire evidentiary burden. The applicant does not have to prove that the tribunal acted correctly, that the procedure was fair, or that the award is sound. The presumption runs the other way. This design is the practical face of the pro enforcement bias. By keeping the applicant's task to a short checklist, the Convention front loads the process in favour of the #award_creditor. The burden then passes to the #award_debtor, who must actively prove that one of the limited grounds in Article V applies. If the debtor cannot do so, and if the court finds no problem with arbitrability or public policy, the court must enforce. The word must matters. The general rule is enforcement, and refusal is the exception that must be justified (Born, 2021). It helps to picture the two stages. Recognition is the court's acceptance that the award is valid and binding, which may be used defensively, for example to stop a losing party from relitigating the same dispute. Enforcement is the further step of using the state's coercive power to make the award effective, by attaching bank accounts, seizing property, or issuing a judgment in the terms of the award. In many systems a foreign award, once recognised, is treated much like a domestic judgment for the purpose of collection. This is what gives an award its real world bite, and it explains why enforceability, rather than the private hearing, is the true product that the Convention delivers. One practical point deserves emphasis for students. Because Article III routes #enforcement through each country's own procedural law, the exact steps differ from place to place. Some countries require a short summary application, others a fuller proceeding. Time limits, court fees, and the level of court that hears the application all vary. The Convention harmonises the grounds for refusal, not the mechanics of collection, and a well advised party plans its enforcement strategy around the assets and procedures of the specific jurisdiction where the debtor's money sits (Tran, 2025). A simple worked example makes the machinery concrete. Suppose a manufacturer in Germany and a distributor in Country X sign a supply contract containing a clause that sends any dispute to arbitration seated in Singapore. The distributor fails to pay, the German company starts arbitration, and the tribunal issues an award ordering the distributor to pay ten million dollars. The award is now binding, but the distributor refuses to pay and keeps no assets in Singapore. Its money and its warehouses are in Country X and in a third country where it holds a bank account. The German company does not need to enforce the award in Singapore at all. It takes the authenticated award and the arbitration agreement to the courts of Country X and of the third country, both of which are parties to the Convention, and asks each to recognise and enforce the award against the local assets. In each forum the distributor may resist, but only by proving one of the Article V grounds, for example that it was denied a fair hearing. If it cannot, the courts must enforce, and the German company can attach the bank account and seize the warehouse. This is the everyday reality that the treaty enables. The award, made in one place, becomes collectible in several others, which is exactly what a commercial party bargained for when it agreed to arbitrate. Without the Convention, the German company might have had to sue afresh in each country, gambling on unfamiliar courts with no guarantee that its Singapore award would count for anything (Born, 2021). 5. The Exhaustive Grounds for Refusal: Article V Article V is the heart of the treaty and the part every student must know cold. It lists the only grounds on which a court may refuse to recognise or enforce a foreign #arbitral_award. The list is exhaustive, which means a court may not invent additional reasons to refuse. It is also narrow, and courts across the world have generally read it narrowly, in keeping with the treaty's purpose (ICCA, 2024). Article V has two parts, and the distinction between them is fundamental. The grounds in Article V(1) must be raised and proved by the party resisting #enforcement. The court will not consider them on its own. The grounds in Article V(2), by contrast, concern the enforcing state's own fundamental interests, and a court may raise them on its own motion, even if the resisting party never mentions them. This structure reflects a careful balance between respecting #party_autonomy and protecting the enforcing state's legal order. Crucially, Article V uses the word may. It says a court may refuse enforcement if a ground is established, not that it must. This wording has generated a lively debate about whether a court retains a residual discretion to enforce an award even when a ground for refusal technically exists, a question examined further below. Two further features of Article V are easy to overlook but important for students. The first is that the burden of proof on the Article V(1) grounds is heavy and rests squarely on the resisting party. It is not enough to raise a doubt. The #award_debtor must actually establish the defect, and vague or unsupported allegations will fail. The second is that the standard applied to most of the grounds is functional rather than technical. Courts repeatedly ask whether the alleged problem was serious enough to matter, not merely whether some rule was breached in a formal sense. A trivial irregularity that made no difference to the fairness or the outcome of the arbitration will almost never defeat enforcement. This practical, results oriented approach is a large part of why refusals under Article V remain uncommon even though the grounds are argued frequently (Baltag, 2022). 5.1 Incapacity and invalidity of the agreement: Article V(1)(a) The first ground covers two related defects. Enforcement may be refused if a party to the #arbitration_agreement was under some incapacity, or if the agreement itself is not valid under the law the parties chose or, failing any choice, the law of the country where the award was made. In plain terms, if there was never a proper agreement to arbitrate, or if a party lacked the legal capacity to enter it, the whole basis of the arbitration falls away. This ground protects consent, which is the bedrock of arbitration. No one should be forced to honour an award produced by a tribunal it never agreed to. 5.2 Due process and the right to be heard: Article V(1)(b) The second ground protects fairness. Enforcement may be refused if the resisting party was not given proper notice of the appointment of the arbitrator or of the proceedings, or was otherwise unable to present its case. This is the #due_process ground, and it guards the minimum standards of natural justice. It does not allow a court to review whether the tribunal reached the right result. It asks only whether the losing party had a fair chance to participate. Courts apply this ground carefully, refusing enforcement only where there was a real and serious denial of the opportunity to be heard, not merely a procedural imperfection that made no difference to the outcome (Born, 2021). 5.3 Excess of mandate: Article V(1)(c) The third ground addresses a tribunal that went beyond its authority. Enforcement may be refused if the award deals with a dispute not contemplated by, or not falling within, the terms of the submission to arbitration, or contains decisions on matters beyond the scope of what was submitted. Arbitrators derive all their power from the parties' agreement, so if they decide something the parties never asked them to decide, that part of the award lacks a foundation. Importantly, this ground includes a rule of severability. If the parts of the award that exceed the mandate can be separated from the parts that fall within it, the valid parts may still be enforced. Courts thus prefer to salvage as much of an award as possible rather than strike it down entirely. 5.4 Irregular composition or procedure: Article V(1)(d) The fourth ground concerns the shape of the process itself. Enforcement may be refused if the composition of the tribunal or the arbitral procedure was not in accordance with the parties' agreement, or, where the agreement is silent, was not in accordance with the law of the country where the arbitration took place. This ground reinforces #party_autonomy over procedure. If the parties agreed on three arbitrators and got one, or agreed on a specific set of rules that were ignored, the resulting award may be vulnerable. Again, minor departures that caused no prejudice will rarely defeat #enforcement, because courts focus on whether the irregularity actually mattered. 5.5 Award not binding, or set aside at the seat: Article V(1)(e) The fifth ground is the most debated in modern practice. Enforcement may be refused if the award has not yet become binding on the parties, or if it has been set aside or suspended by a competent authority of the country in which, or under the law of which, the award was made. The country whose courts have this special supervisory power is usually the #seat_of_arbitration. The seat is not merely the physical place of hearings. It is the legal home of the arbitration, the jurisdiction whose courts may annul the award and whose arbitration law governs the proceeding. This ground raises a profound question. If the courts at the seat annul an award, should courts everywhere else automatically refuse to enforce it, or can an award survive its own annulment and still be enforced elsewhere? This is the #set_aside puzzle, and it is discussed in detail in section 8. What matters here is the structure: Article V(1)(e) treats the seat as the primary place of control, while leaving room, through the word may, for enforcing courts to reach their own view. Empirical study of court decisions suggests that this ground, while among the more successfully argued grounds under Article V, still fails in the majority of cases, which shows how reluctant courts are to refuse enforcement even here (Baltag, 2022). The five grounds in Article V(1) share a common thread. Each protects a basic feature of a legitimate arbitration: real consent, a fair hearing, a tribunal acting within its mandate, an agreed process, and a binding result from a supervising legal order. None of them permits a review of the merits. A court asked to enforce a foreign award is not sitting as a court of appeal. It may not refuse enforcement simply because it thinks the arbitrators made a mistake of fact or law. That restraint is the whole point of the system, because it is what makes an award final and therefore valuable. 6. Public Policy and Arbitrability: Article V(2) Article V(2) contains the two grounds that a court may raise on its own initiative. They protect the enforcing state itself rather than the individual resisting party, and they mark the outer limit of a country's willingness to lend its coercive power to a foreign #arbitral_award. 6.1 Arbitrability: Article V(2)(a) The first of these grounds is #arbitrability. Enforcement may be refused if the subject matter of the dispute is not capable of settlement by arbitration under the law of the enforcing state. Every legal system reserves certain categories of dispute for its own courts, on the view that they involve public interests too important to be left to private judges. Traditional examples have included some questions of criminal law, family status, insolvency priorities, and certain aspects of competition or antitrust law. Over the decades the scope of what is considered arbitrable has steadily widened, as states have grown more comfortable allowing arbitrators to handle commercial matters that were once thought off limits. Still, the ground remains a genuine safeguard, and a party seeking enforcement must be alert to whether the underlying subject matter is arbitrable where it wants to collect (Born, 2021). 6.2 The public policy exception: Article V(2)(b) The second and more famous ground is #public_policy. Enforcement may be refused if recognising or enforcing the award would be contrary to the public policy of the enforcing state. Public policy is the ultimate safety valve of the Convention. It allows a court to protect the most fundamental values of its legal system, for example by refusing to enforce an award that would give effect to bribery, fraud, or a serious violation of natural justice. The danger with this ground is obvious. If public policy were read broadly, it could swallow the whole treaty. Any losing party could argue that enforcing an unfavourable award somehow offends local values, and courts inclined to protect domestic interests could refuse enforcement whenever they disliked the result. The drafters and the courts have avoided this trap by insisting on a narrow, international conception of public policy. The exception is reserved for violations of the most basic notions of morality and justice, not for ordinary disagreements about how the arbitrators applied the law. This narrow reading has deep roots. In an early and influential American decision, a court refused to allow a defeated party to escape a foreign award merely because enforcement was said to clash with national interests, holding that the public policy defence should be applied only where enforcement would violate the forum state's most basic notions of morality and justice. Courts around the world have echoed this language ever since, and the leading interpretive guides urge judges to keep the exception exceptional (ICCA, 2024). The distinction that scholars draw is between domestic public policy, which is broad and governs purely internal matters, and international public policy, which is much narrower and is the only version that should defeat a foreign award. Recent commentary continues to stress that a public policy objection is not a shortcut to reviewing the merits of an award, and courts that treat it as such undermine the certainty the Convention was built to provide. Taken together, the two grounds in Article V(2) show the treaty's careful architecture. They give each state a defence for its own fundamental interests, but they are drawn so tightly that they rarely defeat #enforcement in practice. The overall picture that emerges from decades of case law is one of striking consistency: contracting states have interpreted the grounds for refusal in broadly the same restrained way, and outright refusals remain the exception rather than the rule (ICCA, 2024). 7. The Pro Enforcement Bias and Judicial Discretion The phrase pro enforcement bias appears constantly in the literature, and it is worth pausing to understand exactly what it means and where it comes from. It is not a vague attitude. It is built into the very structure of the treaty through a series of concrete design choices. The bias comes first from the abolition of double exequatur, which removed a whole layer of obstruction. It comes second from the reversal of the burden of proof, which places the load on the party resisting #enforcement. It comes third from the exhaustive and narrow list of refusal grounds in Article V, which prevents courts from inventing new excuses. It comes fourth from the rule against reviewing the merits, which keeps enforcement courts from behaving like appeal courts. And it comes fifth from the equal treatment guarantee in Article III, which forbids treating foreign awards worse than domestic ones. Each choice tilts the field toward giving effect to the award (Born, 2021). The concept has become so central that scholars now debate whether it is being used too loosely. Some argue that labels such as arbitration friendly, pro arbitration, and pro enforcement are often run together as if they meant the same thing, when in fact they describe different values that can sometimes pull in different directions. A court can be strongly pro enforcement, in the sense of interpreting the Convention to favour giving effect to awards, while still policing the basic integrity of the process. Being pro enforcement, properly understood, means reading the treaty in line with its object and purpose, not enforcing every award regardless of its flaws. This brings us to the discretion question raised by the word may in Article V. Because the treaty says a court may refuse enforcement when a ground is established, rather than must, several jurisdictions have concluded that an enforcing court retains a limited discretion to enforce an award even where a technical ground for refusal exists. The typical example is a minor procedural irregularity that caused no real prejudice. A court exercising this residual discretion may decide that refusing enforcement would be disproportionate and enforce anyway. Other jurisdictions read the discretion more cautiously. The debate matters because it shows that the Convention, for all its uniform text, still leaves genuine room for national courts to shape outcomes, and it is one reason why comparative study is so important (Blackaby, Partasides and Redfern, 2022). 8. The Seat, Set Aside Awards, and the Delocalization Debate No topic captures the intellectual tension inside the Convention better than the treatment of awards that have been annulled at the #seat_of_arbitration. To understand it, we must first understand the role of the seat and the two remedies that surround an award. There are two distinct proceedings that a losing party may pursue. The first is an application to set aside, also called annulment or vacatur, which can only be brought before the courts of the seat. If successful, it wipes out the award in its home jurisdiction. The second is resistance to #enforcement, which can be raised in any country where the winner seeks to collect, on the grounds listed in Article V. The seat is therefore the primary place of control, and every other country is a secondary place of enforcement. Choosing the seat is one of the most consequential decisions in drafting an #arbitration_agreement, because it determines which courts may annul the award and which arbitration law governs the process. The puzzle arises from Article V(1)(e). Because a court may refuse #enforcement of an award that has been set aside at the seat, but is not required to, two schools of thought have developed. The territorial view treats the seat as decisive. On this view, once the seat annuls an award the award ceases to exist, and there is nothing left to enforce anywhere. The delocalized view treats the award as detached from any single national legal order. On this view, an annulment at the seat is just one factor, and an enforcing court in another country may still give effect to the award if its own law permits. France has become the leading home of the #delocalization view. French courts have repeatedly enforced awards that were annulled at their seat, reasoning that an international award is not anchored in the legal order of the seat and that French law, being more favourable to enforcement, may be applied under the more favourable right provision discussed below. The line of French decisions on this point, culminating in the well known Putrabali ruling, treats the international award as a decision of international justice whose validity is judged under the enforcement forum's own standards rather than the seat's. The United States has taken a more cautious and mixed path. In one early case an American court enforced an award that had been set aside abroad, but later decisions have generally shown strong deference to annulments at the seat, enforcing set aside awards only in unusual circumstances. England and many other common law jurisdictions likewise tend to respect a foreign annulment while reserving a narrow residual power (Born, 2021). The bridge between these positions is Article VII, one of the treaty's most quietly important provisions. Article VII(1) contains the more favourable right rule. It provides that the Convention does not deprive any interested party of any right it may have to rely on an award under the law or treaties of the country where enforcement is sought. In plain terms, if a country's own domestic law is more generous to #enforcement than the Convention, a party may use that more generous law instead. This is the legal hook that allows France to enforce annulled awards, because French domestic law does not treat annulment at the seat as a bar. One respected commentator called this provision the treasure and the ingenious idea of the Convention, because it sets only a floor of enforceability and lets states build higher (ICCA, 2024). Article VII(2) performs a tidy housekeeping function by providing that, between contracting states, the New York Convention supersedes the old Geneva instruments of 1923 and 1927. Article VI completes the picture. It provides that if an application to set aside or suspend the award is pending at the seat, the enforcing court may adjourn its decision, and it may also order the resisting party to provide suitable security. This gives enforcing courts a sensible middle path. Rather than rushing to enforce or refuse while the seat is still deciding, a court can pause, protect the winner by requiring security, and wait for clarity. The set aside question remains one of the most researched issues in the field, precisely because it exposes the deep disagreement about whether arbitration ultimately belongs to a single national legal order or floats above all of them (Baltag, 2022). 9. Comparative Perspectives Across Jurisdictions The true test of any treaty is not its text but its life in the courts. Because Article III channels #enforcement through national procedures, and because the word may leaves room for national attitudes, the same Convention has been applied with different accents in different places. A comparative tour is the best way to see both the treaty's uniformity and its variation. 9.1 The United States The United States implements the Convention through a dedicated chapter of its Federal Arbitration Act. American courts are firmly pro enforcement and read the Article V grounds narrowly. The classic statement of the #public_policy defence, that it applies only where enforcement would violate the forum's most basic notions of morality and justice, comes from American case law and has been widely borrowed abroad. On the set aside question, American practice is cautious. Courts generally defer to annulments at the #seat_of_arbitration and will enforce a set aside award only in rare situations, reflecting a preference for respecting the primary jurisdiction while keeping a narrow safety valve (Born, 2021). 9.2 England and Wales England enforces Convention awards through its Arbitration Act, and its courts are known for a commercial, pro enforcement outlook combined with careful attention to jurisdiction and consent. A leading decision concerned whether a party that had not signed an #arbitration_agreement could be bound by it. The English Supreme Court refused to enforce an award against a government body that it found was not a party to the arbitration agreement under the law of the seat, while courts in the seat jurisdiction later took the opposite view of the same facts. The episode is a vivid reminder that even careful courts can diverge on the threshold question of who agreed to arbitrate. England has continued to modernise its framework, with a further reform of its arbitration legislation reaching the statute book in 2025, aimed at keeping the jurisdiction attractive and its law clear (Blackaby, Partasides and Redfern, 2022). 9.3 France France is the intellectual champion of the #delocalization approach. French law and its courts treat an international award as a decision that is not tied to the legal order of the seat, and they will enforce awards under French standards even where the seat has annulled them. France also gives the more favourable right provision of Article VII a central role, applying its own liberal enforcement rules in preference to the Convention's minimum. For students, France is the clearest example of how Article VII can transform the treaty from a floor into a launch pad for enforcement (Born, 2021). 9.4 India India offers the most instructive story of evolution. For years Indian courts were seen as unpredictable, partly because of an expansive reading of #public_policy. Decisions in the early 2000s widened the concept, including a notion of patent illegality that allowed courts to look closely at the merits of awards, which worried the international community. The trend began to reverse. A landmark ruling clarified that the general part of the Indian arbitration statute did not apply to foreign seated arbitrations, reducing the scope for Indian court interference in awards made abroad. Legislative amendments then narrowed the public policy ground, and later Supreme Court decisions confirmed that the narrow ground applies to foreign awards and that patent illegality is confined to purely domestic awards. More recent rulings have continued this pro enforcement direction, reinforcing that courts should not use enforcement proceedings to reopen the merits. India's journey shows that a jurisdiction can move, through a mix of legislation and judicial discipline, from suspicion toward alignment with global standards (Tran, 2025). 9.5 Singapore Singapore has built a reputation as one of the most reliable enforcement forums in the world. Its courts apply a consistently narrow, process focused review and are reluctant to disturb awards on public policy grounds. Singapore has also contributed important refinements to enforcement doctrine. In a notable decision its Court of Appeal recognised that where the courts of the seat have already decided a point that a party later tries to raise again at the enforcement stage, a form of transnational issue estoppel may prevent the party from relitigating the same argument. This kind of doctrine promotes #finality and discourages award debtors from taking a second bite at the same cherry in a different country. Singapore's predictability is a large part of why it is so often chosen as a #seat_of_arbitration and an enforcement venue. 9.6 China China acceded to the Convention with both the reciprocity and commercial reservations, and its courts have developed a distinctive internal control mechanism to promote consistency. Under a reporting or verification system, lower courts that are inclined to refuse enforcement of a foreign or foreign related award must generally report the matter up through the court hierarchy, in some situations reaching the Supreme People's Court, before a refusal can be finalised. This centralises decisions about refusal and reduces the risk that local courts will block enforcement for parochial reasons. The result has been a steady strengthening of China's pro enforcement record over time, which matters enormously given the scale of Chinese trade (ICCA, 2024). 9.7 The United Arab Emirates and the Gulf The United Arab Emirates provides a fast moving example of a jurisdiction that has worked hard to become an enforcement friendly hub. The UAE acceded to the Convention in the mid 2000s, and for a period its courts were seen as cautious, sometimes refusing enforcement on procedural technicalities drawn from domestic litigation rules. That reputation has changed. A modern federal arbitration law based closely on the UNCITRAL Model Law brought the country's framework into line with international practice, and the courts have grown markedly more supportive of foreign #arbitral_awards. Dubai in particular has positioned itself as a regional centre for dispute resolution, with dedicated arbitration institutions and specialised courts, including within its financial free zones, that apply common law style procedures. The region has not been free of turbulence, as shown by a restructuring of Dubai's arbitration institutions that raised questions for parties who had chosen a now discontinued forum. Even so, the overall direction has been toward greater predictability and stronger enforcement, which matters for the many cross border contracts that now select a Gulf seat. For students, the UAE illustrates a broader pattern seen across emerging arbitration centres: a deliberate strategy of modernising legislation, training judges, and signalling a pro enforcement stance in order to attract international commerce (ICCA, 2024). What this tour reveals is a treaty that is uniform in its core and flexible at its edges. Every jurisdiction shares the same short list of refusal grounds and the same rule against reviewing the merits. Yet each brings its own legal culture to bear on hard questions such as annulled awards, consent by non signatories, and the reach of #public_policy. For a student, the lesson is that mastering the text of Article V is necessary but not sufficient. Real competence requires knowing how the courts of the specific enforcement forum actually behave. 10. Contemporary Challenges A treaty drafted in 1958 now operates in a world its authors could not have imagined. Several modern pressures are testing the Convention, and how the system responds will define its next chapter. The first challenge is #sovereign_immunity. When the losing party is a state or a state entity, an #award_creditor may win the arbitration and still struggle to collect, because states can invoke immunity from enforcement against their assets. Courts must then decide when a state has waived its immunity and which assets are used for commercial rather than sovereign purposes and therefore open to attachment. A recent English decision addressed whether a state's ratification of the Convention itself can amount to a submission to the jurisdiction of enforcing courts, a question with large consequences for the enforcement of awards against governments. This intersection of arbitration and public international law is one of the most active frontiers in the field. The second challenge comes from #investment_arbitration and regional legal orders. Disputes between investors and states raise distinctive enforcement issues, and tensions have emerged where regional law conflicts with treaty based awards, particularly within the European Union following decisions that questioned the validity of arbitration between EU investors and EU member states. Award creditors have responded by seeking enforcement outside the region, relying on the Convention in more welcoming jurisdictions. These conflicts show how the Convention can become a battleground where different systems of law collide. The third challenge is economic and political, in the form of #sanctions. As states use economic sanctions more aggressively, arbitrators and enforcing courts face difficult questions about whether enforcing an award would require a party to do something prohibited by a sanctions regime, and whether such a conflict engages the #public_policy exception. Because sanctions differ from country to country, an award that is straightforward to enforce in one place may be blocked in another, adding a new layer of complexity to enforcement strategy. The fourth challenge is technological. Remote hearings, electronic signatures, and the growing use of artificial intelligence in the arbitral process all raise fresh questions about #due_process and the integrity of awards. If parts of an award are influenced by automated tools, or if a hearing conducted entirely online is said to have deprived a party of a fair opportunity to be heard, a resisting party may test these arguments under Article V(1)(b). Scholars are already proposing frameworks for using new technology in ways that preserve enforceability under the Convention, precisely because the value of any innovation depends on whether the resulting award will still be enforced (Blackaby, Partasides and Redfern, 2022). A final, quieter challenge is consistency itself. With more than 170 contracting states and no single court sitting at the top to give the treaty an authoritative meaning, uniform interpretation depends on courts voluntarily learning from one another. Interpretive guides, judicial training programmes, and scholarly commentary all help, but the risk of drift is permanent. The remarkable degree of consistency achieved so far is a genuine achievement, not a guarantee, and preserving it requires ongoing effort (ICCA, 2024). 11. Reform, Criticism, and the Road Ahead Given its age, some have asked whether the Convention needs to be rewritten. The dominant view is that it does not, and understanding why is instructive. The most powerful argument against reopening the treaty is practical. It took decades and two failed predecessors to produce the 1958 text, and any attempt to renegotiate it would risk unravelling the near universal consensus that gives the Convention its value. A new treaty might attract fewer members, or it might introduce fresh ambiguities that would take another generation of case law to resolve. The Convention's brevity, once seen as a weakness, has proved to be a strength, because short and general language has allowed courts to adapt the treaty to new circumstances without formal amendment (Born, 2021). Reform, therefore, has largely happened around the Convention rather than to it. The #UNCITRAL Model Law harmonised national arbitration statutes so that the grounds for annulling an award at the seat mirror the grounds for refusing enforcement under Article V, producing a coherent global system without touching the treaty text. Interpretive guides have promoted uniform reading. National legislatures have modernised their own arbitration laws to be more enforcement friendly. And courts have used the more favourable right provision of Article VII to move ahead of the treaty's minimum standard. In this way the system evolves continuously while its foundation stays fixed. Legitimate criticisms remain. The set aside question continues to divide jurisdictions, leaving award creditors uncertain whether an annulled award can be salvaged elsewhere. The #public_policy exception, though narrowly drawn in theory, is still occasionally stretched by courts reluctant to enforce a particular result. Enforcement against states remains difficult because of #sovereign_immunity. And the lack of a central interpretive body means that inconsistency can never be fully eliminated. These are real problems, but they are problems of application rather than of basic design, and they are best addressed through better judicial practice and clearer national law rather than through renegotiation. Looking ahead, the Convention's role as the engine of #arbitration seems secure. As global trade continues to grow and to shift toward new economic centres, the demand for a neutral, enforceable method of resolving cross border disputes will only increase. Newer economies continue to accede to the treaty and to align their courts with its pro enforcement spirit, and established arbitration hubs continue to refine their doctrines to keep enforcement predictable. The direction of travel, across very different legal cultures, has been toward greater respect for #arbitral_awards, not less (Tran, 2025). 12. Conclusion The New York Convention of 1958 is short, old, and quietly revolutionary. It solved a problem that had defeated two earlier attempts by making a simple promise workable on a global scale: an #arbitral_award earned in one country can be recognised and enforced in almost any other, subject only to a narrow, closed list of defences. By abolishing double exequatur, reversing the burden of proof, limiting the grounds for refusal, and forbidding review of the merits, the treaty transformed a private hearing into a globally portable right. That transformation is why the Convention is rightly called the engine of the entire field. For students, three ideas capture the essence of the treaty. First, #enforceability is the product. Arbitration is attractive not because the hearing is superior to a trial, but because the outcome can travel across borders in a way that court judgments historically could not. Second, the pro enforcement bias is a structural choice, not a mood. It is written into the design through the burden of proof, the exhaustive grounds in Article V, and the equal treatment guarantee in Article III. Third, uniformity has limits. The same text lives differently in different courts, especially on hard questions such as annulled awards, consent by non parties, and the outer reach of #public_policy, so real understanding requires attention to how specific jurisdictions actually behave. The genius of the Convention lies in its restraint. It does not try to control how arbitrations are run, nor does it impose a single global procedure for collection. It simply guarantees that the door to enforcement will be open, and it defines the few narrow circumstances in which a court may close it. That modest design has supported the growth of international #arbitration into the default method of resolving cross border commercial disputes and has strengthened the #rule_of_law in international commerce. More than six decades on, the engine still runs, and the world's trade still moves because of it. References Baltag, C. (2022). Article V(1)(e) of the New York Convention: To enforce or not to enforce set aside arbitral awards? Journal of International Arbitration, 39(3). Kluwer Law International. Blackaby, N., Partasides, C., and Redfern, A. (2022). Redfern and Hunter on International Arbitration (7th ed.). Oxford University Press. https://doi.org/10.1093/law/9780192869906.001.0001 Born, G. B. (2021). International Arbitration: Law and Practice (3rd ed.). Kluwer Law International. Born, G. B. (2021). International Commercial Arbitration (3rd ed., 3 vols.). Kluwer Law International. Chauhan, S. (2024). Review of New York Convention: Convention on the Recognition and Enforcement of Foreign Arbitral Awards of 10 June 1958, Article by Article Commentary (R. Wolff, Ed.). Arbitration International, 40(4), 556. International Council for Commercial Arbitration. (2024). ICCA's Guide to the Interpretation of the 1958 New York Convention (2nd ed.). International Council for Commercial Arbitration. Tran, H. T. L. (2025). Vietnam's thirty year journey with the New York Convention: Emerging paradigms in international arbitration. Arbitration International. https://doi.org/10.1093/arbint/aiaf024 Hashtags #NewYorkConvention1958 #InternationalArbitration #ForeignArbitralAwards #EnforcementOfAwards #Article_V #ProEnforcement #PublicPolicyException #Arbitrability #SeatOfArbitration #UNCITRAL_ModelLaw #CrossBorderDisputes #DisputeResolution #CommercialArbitration #GlobalJurisdictions #RuleOfLaw

This article examines one of the most debated compensation decisions in the history of Arab investment arbitration: the award rendered on 22 March 2013 in Mohamed Abdulmohsen Al-Kharafi and Sons Co. v. State of Libya and others. A Kuwaiti company that had spent only a few million dollars on a tourism project that was never built walked away with an award of roughly USD 935 million, of which USD 900 million represented projected future profits over an eighty-three year horizon. The size of the sum, and the gap between money actually spent and money awarded, turned the case into a reference point for every discussion about #lost_profits, #speculative_damages, and the limits of the #full_reparation principle. The study sets out the facts, the legal framework of the Unified Agreement for the Investment of Arab Capital in the Arab States, and the tribunal's damages reasoning. It then measures that reasoning against the accepted doctrine on causation, certainty, the going concern requirement, and the discounted cash flow method. Finally, it follows the long annulment and enforcement saga through the Egyptian and French courts, ending with the Egyptian Court of Cassation decision of 24 June 2021 that ultimately let the award stand. The article argues that Al-Kharafi is best read not as a rejection of lost profits as a head of damages, but as a warning about what happens when tribunals apply that head without discipline. Keywords: investment arbitration; lost profits; damages; discounted cash flow; full reparation; proportionality; Al-Kharafi; Libya; Unified Agreement; public policy. 1. Introduction Money is the point of most #investment_arbitration. States and investors argue for years about jurisdiction, treaty standards, and breach, but for the party that wins, the number at the end of the award is what matters. That number is also where some of the least examined reasoning in the whole field tends to hide. Tribunals spend hundreds of paragraphs on liability and then, in a comparatively thin section, decide how much a lost opportunity was worth. When the opportunity was real and running, that exercise is difficult but grounded. When the opportunity never existed except on paper, the exercise starts to look like fortune telling. The award in Al-Kharafi v. Libya is the clearest modern example of the second situation. The claimant, a large Kuwaiti conglomerate, had obtained a licence and a long lease to build a tourism complex near Tripoli. Construction never began. The company had put a small amount of money into the venture, and when the project collapsed it asked, at first, for around five million dollars. By the end of the proceedings the claim had grown past two billion dollars, and the tribunal awarded roughly USD 935 million, the overwhelming majority of it for profits that the unbuilt resort might have earned across more than eight decades. With interest, the figure later climbed toward one billion dollars. That result made the case famous, and infamous. Supporters saw a state that had blocked a legitimate investor and then tried to escape the consequences. Critics saw a quantum decision untethered from any recognisable standard of proof, a windfall dressed up as compensation. Both camps agree on one thing: the case forces a hard question about what #lost_profits are for and when they may be awarded. This article uses Al-Kharafi to work through that question. It is written for students, so it explains the underlying doctrine as it goes, but it treats the case with the seriousness that a #Scopus level analysis demands, weighing the tribunal's reasoning against the wider law of #State_responsibility and against recent scholarship on compensation. The argument proceeds in eight further parts. Part 2 sets out the facts. Part 3 explains the unusual legal instrument under which the dispute was heard. Part 4 describes the award and the way the damages were built. Part 5 lays out the doctrine on lost profits, certainty, and valuation. Part 6 applies that doctrine critically to the award. Part 7 traces the post-award litigation in Egypt and France. Part 8 draws lessons for tribunals, states, and the drafters of investment instruments. Part 9 concludes. 2. Background and facts of the dispute The story begins in 2006. On 7 June of that year, by Decision No. 135/2006, the Libyan authority responsible for tourism granted Mohamed Abdulmohsen Al-Kharafi and Sons Co. approval and a licence to establish a large touristic investment project in the Tajura district on the outskirts of Tripoli. A day later, on 8 June 2006, the company signed a land lease with the Tourism Development Authority. The plot covered around 240,000 square metres of state owned land. The plan was ambitious: a five star hotel, serviced apartments, a commercial centre, restaurants, and recreational areas, the kind of coastal development that oil rich Libya hoped would attract regional visitors. The commercial terms were long dated. The #lease ran for ninety years, with an annual rent of 730,000 Libyan Dinars. Construction was expected to take about seven and a half years and to cost in the region of USD 130 million. The contract said that any dispute would be settled first amicably and, failing that, by arbitration under the Unified Agreement for the Investment of Arab Capital in the Arab States. That clause, Article 29 of the lease, later became the doorway to the whole case. Almost immediately the project ran into trouble that had nothing to do with the investor. The land was not delivered clean. Between 2006 and 2007 Al-Kharafi faced third party claims to the site from other Libyan public entities, and it was eventually asked to vacate the plot while those competing claims were sorted out. The investor's central complaint, later accepted by the tribunal, was that the authority had not handed over the property free of occupancies and persons as the contract required, and that various arms of the Libyan state then prevented the company from starting work. Attempts to fix the problem failed. On 21 January 2009 the government offered Al-Kharafi a different plot of land for the same project. The company refused, taking the position that it was entitled to the site it had contracted for, not a substitute. In 2010 the government terminated the licence. For the investor, the sequence of events was a straightforward breach: it had been promised a site, it had been kept off that site, it had been offered an inadequate swap, and then its rights had been cancelled outright. For Libya, the picture was different. The state argued that the company itself was responsible for the delays because it had failed to submit final designs, to deposit ten per cent of the project value in a Libyan bank account, and to apply for the necessary permits. Libya also argued, more bluntly, that the investor had suffered little because it had barely started, and it questioned the honesty of a compensation claim that had swollen so dramatically. On 3 March 2011 Al-Kharafi launched arbitration against the Libyan state and several of its authorities, including the Ministry of Economy, the General Authority for Investment Promotion and Privatization Affairs, the Ministry of Finance, and the Libyan Investment Authority. The timing matters. Libya in 2011 was sliding into the civil conflict that ended the Gaddafi government. A tribunal asked to project the profits of a tourist resort in that country, across more than eighty years, was being asked to see a very long way into a very uncertain future. The way the claim itself changed shape during the proceedings tells its own story. When the dispute began, the sums in play were small and concrete. The company had spent a limited amount getting ready to build, and its early demands reflected that. As the case matured, the figure climbed. At one stage the claim was pitched above two billion dollars, built around the idea that the investor was entitled to the profits of the resort for the full remaining term of the ninety year lease. A claim that starts at a few million and ends in the billions is not automatically dishonest, because a claimant is entitled to argue its best case, but such a steep rise should invite scrutiny of the theory driving it. Here the driver was simple: the longer the profit horizon and the more optimistic the assumptions, the larger the number, and there was little in the way of hard operating data to hold either in check. Libya, for its part, did not only deny liability. It argued that any losses the company suffered flowed from the investor's own conduct, and it went further, describing the compensation claim in strong terms as tainted by corruption. It stressed that the investor had never deposited the ten per cent of project value that the arrangements contemplated, that it had paid only a tiny fraction of the project's stated worth, and that no construction had ever taken place. Those points went both to jurisdiction, whether real Arab capital had been transferred at all, and to damages, whether a company that had barely committed funds could claim the profits of a completed and thriving enterprise. The tribunal was unpersuaded on liability, but the arguments about the thinness of the actual investment would echo, years later, in the reasoning of the Egyptian courts. 3. The legal framework: the Unified Agreement Most readers meet #investment_arbitration through bilateral investment treaties or through the ICSID Convention. Al-Kharafi arose under neither. It was decided under the Unified Agreement for the Investment of Arab Capital in the Arab States, a regional instrument adopted on 26 November 1980 and in force from September 1981, which both Libya and Kuwait had ratified. The #Unified_Agreement is a treaty among Arab League members designed to protect capital moving between them and to give an Arab national a route to relief when a host Arab state harms its investment. Two features of this instrument shaped everything that followed. First, the agreement contemplated its own standing court, the Arab Investment Court, as a special judicial body for disputes arising directly from the treaty. That court was slow to become operational, and, importantly for later events, it has taken the view on more than one occasion that it lacks jurisdiction to review awards issued by ad hoc tribunals set up under the agreement. In other words, the treaty regime offered no reliable internal appeal or annulment mechanism. Second, the agreement allowed disputes to go to ad hoc arbitration, which is what happened here through the reference in Article 29 of the lease. The tribunal was constituted with three members. Dr. Abdel Hamid El-Ahdab served as president, Dr. Ibrahim Fawzi was nominated by the claimant, and Justice Mohamed El-Kamoudi El-Hafi, a judge of the Libyan Supreme Court, was nominated by Libya. The seat was Cairo, and the hearings used the premises and rules of the Cairo Regional Centre for International Commercial Arbitration, or #CRCICA, although the arbitration was not administered by the centre and remained an ad hoc proceeding. The applicable law was Libyan law together with the Unified Agreement. That choice of seat later proved decisive, because it opened the award to review by the Egyptian courts under Egyptian arbitration law, a possibility the treaty regime itself did not clearly provide. Libya challenged jurisdiction. It argued that the project did not involve any real transfer of Arab capital from Kuwait into Libya, so the substantive conditions of the #Unified_Agreement were not met. It pointed out that construction never began and that the investor had failed to deposit ten per cent of the project value as required. The claimant could show only that it had paid a fraction of one per cent of the value to the authority. The tribunal rejected these objections. It read the wording of Article 29 as consent to arbitration under the agreement and found that the requirements for treaty protection were satisfied. Whatever one thinks of that ruling, it set the stage for the part of the award that made history, the assessment of damages. 4. The arbitral award and the damages calculation On 22 March 2013 the tribunal issued its final award in favour of the claimant. It held that Libya had breached a primary obligation of the lease by failing to deliver the land free of occupancies and by preventing the company from beginning the project. It found that the investor had not created its own obstacles, and it treated Libya's offer of alternative land as further proof of the state's failure rather than as a genuine cure. Liability, in short, rested on the host state. The damages were divided into several heads. The tribunal awarded about USD 5 million for the value of the losses and expenses the company had actually suffered, the money genuinely sunk into the venture. It awarded USD 30 million for #moral_damages, on the theory that the affair had harmed the claimant's reputation. It awarded USD 900 million for #lost_profits, described in the award as profits resulting from real and certain lost opportunities. It added roughly USD 1.94 million for the costs of the arbitration. And it ordered four per cent interest on all sums from the date of the award until full payment. The heads together came to approximately USD 935 to 936 million, and the running interest pushed the practical value of the award toward one billion dollars within a few years. The single number that defines the case is the USD 900 million for future profits. To reach it, the tribunal accepted that the resort, had it been built and operated, would have generated profit across the remaining life of the lease, a period the award treated as roughly eighty-three years. It is worth pausing on the shape of the claim to see how far it travelled. Al-Kharafi had originally quantified its loss at around five million dollars, essentially its #sunk_costs. During the proceedings the claim was amended upward, at one point to more than one billion dollars, and the claimant argued for future profits stretching across the whole remaining term of the lease, a theory that could support a figure above two billion dollars. The tribunal did not go that far, but it went very far indeed, moving from a real out of pocket loss of a few million to a projected profit award of nine hundred million. The mechanics of that leap are the heart of the controversy. A profit projection for a business that has never traded rests entirely on assumptions: how many guests, at what room rates, with what occupancy, at what operating margin, growing at what rate, discounted at what rate to reflect risk and the time value of money, over how many years. Each assumption compounds the next. Stretch the horizon to eighty-three years and small changes in the inputs produce enormous swings in the output. This is the classic profile of a #discounted_cash_flow style projection applied to a venture with no operating history, in a country entering armed conflict. The award called the lost opportunities real and certain. Whether they deserved those words is the question the rest of this article tries to answer. 5. Lost profits in international investment law: the governing doctrine Before judging the award, one has to be fair about what the law actually permits. #Lost_profits are not forbidden. They are a recognised and often legitimate head of recovery. The problem is that they come with conditions, and those conditions exist precisely to keep compensation honest. This part sets out the doctrine in plain terms. 5.1 Full reparation and the Chorzow standard The foundation of damages in international law is the principle of #full_reparation, most famously stated by the Permanent Court of International Justice in the Factory at #Chorzow case in 1928. The court said that reparation must, so far as possible, wipe out all the consequences of the illegal act and re-establish the situation that would, in all probability, have existed if the wrong had not been committed. This is the origin of what practitioners call the but for test: compare the claimant's actual position with the position it would have occupied but for the breach, and award the difference. That formulation was later codified in the International Law Commission's Articles on #State_responsibility, known as #ARSIWA, adopted in 2001. Article 31 requires full reparation for injury caused by an internationally wrongful act. Article 34 lists compensation as a form of reparation, and Article 36 says compensation shall cover any financially assessable damage, including loss of profits, insofar as it is established. The phrase to notice is insofar as it is established. Full reparation is not a licence to award every profit a claimant can imagine. It is a duty to restore a claimant to the position it can prove it would have reached. The Chorzow standard cuts both ways, and this is often forgotten. It aims to restore, not to enrich. Awarding a claimant far more than the breach cost it does not wipe out the consequences of the wrong; it creates a new and different situation, a #windfall. Recent scholarship has stressed exactly this discipline. Writers on #proportionality and on the outer limits of reparation have argued that the Chorzow formula contains its own internal brake, because a sum that overshoots the real loss is not reparation at all. The point returns with force when we look at Al-Kharafi, where the money awarded dwarfed the money lost. 5.2 Certainty against speculation The second condition is #certainty. Lost profits are recoverable only when the loss is reasonably certain, not merely possible. The commentary to Article 36 of ARSIWA puts it carefully: lost profits are inherently #speculative and are increasingly vulnerable to commercial and political risk the further into the future a projection runs. They are compensable only where an anticipated income stream has attained enough of the attributes of a legally protected interest to be considered sufficiently certain. Arbitral practice draws a useful distinction here. There is certainty about whether a loss of profit occurred at all, and there is certainty about the exact amount. Tribunals generally demand a fairly firm answer to the first question and accept a looser, reasonable estimate on the second. Assessing lost profits, as tribunals often say, is not a precise science, and some approximation in the amount is tolerated. But that tolerance on quantum does not remove the threshold requirement that the profit be reasonably certain to have arisen in the first place. A claimant who cannot show that its venture would probably have earned profits at all fails at the first hurdle, and no amount of tolerance on the exact figure can rescue the claim. The prohibition on #speculative_damages is the practical expression of this rule. Where projected profits depend on a long chain of unproven assumptions about an untested business, tribunals are supposed to treat them as speculative and to decline them, or at least to fall back on a more grounded measure. One vivid image from the literature compares long range profit forecasts to the prophecies of Nostradamus, a reminder that a projection is only as reliable as the evidence behind it. Bound up with certainty is #causation, the requirement that the loss actually flow from the breach. A claimant must show not only that profits were reasonably certain but that the state's wrongful act, rather than market forces, mismanagement, or the claimant's own choices, was what deprived it of those profits. In a project that never operated, causation and certainty blur together, because both depend on a hypothetical world in which the resort was built, filled with guests, and run profitably for decades. The tribunal has to construct that entire imaginary history and then attribute its collapse to the respondent. The more elaborate the imagined world, the weaker the causal chain becomes, since each additional assumption is one more thing that might have gone wrong for reasons having nothing to do with the state. This is why grounded measures such as recovering what was actually spent are attractive: they ask the tribunal to compensate a loss that certainly happened rather than to referee a contest of forecasts about a business that never was. 5.3 The going concern requirement and the new business problem The third condition follows from the second and is the most directly relevant to Al-Kharafi. Tribunals have long been reluctant to award #lost_profits calculated by forward looking methods unless the damaged business was a #going_concern, meaning an enterprise that was actually operating and had a track record of earning. Some tribunals have put a rough figure on it, looking for an earnings history of two to three years before they will trust a forward projection. The reason is common sense dressed as doctrine. A business that has traded for years produces real data, actual customers, actual revenue, actual costs, and a projection built on that data can be tested. A business that has never opened produces nothing but hopes. Projecting decades of profit from hopes is the definition of speculation. This is sometimes called the #new_business_rule. Under it, when the venture is new or, worse, never built, tribunals typically limit recovery to #sunk_costs, the money actually invested, or to some other grounded measure such as the market value of the assets, rather than to imagined future earnings. Investment jurisprudence is full of examples of tribunals refusing forward looking valuations for early stage projects. In disputes over mining concessions that had reached only the exploration stage, tribunals have declined to value the project by projected production profits because there was no proof the project would ever have produced at a profit and no reason to think a buyer would have paid a price based on such projections. The recurring lesson is that the absence of an operating history is not a technicality. It goes to whether the profits were ever reasonably certain. A resort on which not a single foundation had been poured sits at the far end of this spectrum. 5.4 The discounted cash flow method and its abuse The tool most often used to turn future profits into a present sum is the #discounted_cash_flow method, or #DCF. The idea is respectable and comes from ordinary finance. You forecast the cash a business will throw off in each future year, then discount those amounts back to today using a rate that reflects both the time value of money and the risk that the cash will not appear. Add up the discounted amounts and you have a present value. DCF works well for stable, operating assets with predictable cash flows, a toll road, a power plant with a signed offtake contract, a mature utility. It works badly for ventures whose cash flows are guesses. The method's great weakness is that it launders uncertainty into a single confident looking figure. Change the growth rate by a point or two, shorten or lengthen the forecast period, nudge the discount rate, and the answer moves by hundreds of millions. Because the output looks precise, it can hide how shaky the inputs are. Critics have documented how DCF has drifted from a careful valuation tool into an engine for inflated claims. Commentators writing on the reform of #ISDS have described the ease with which claimants now craft billion dollar claims created, in effect, out of thin air, and the frequency with which tribunals endorse calculations that no real world business would ever accept. The core objection is not that DCF is always wrong. It is that DCF should not be applied to assets that are not going concerns, and that when it is, the numbers stop tracking reality. As we will see, Al-Kharafi is a textbook illustration of DCF style reasoning pushed well past the point where the method can bear the weight. 6. A critical analysis of the Al-Kharafi damages With the doctrine on the table, the award can be tested against it. The tribunal was not wrong to think that lost profits are recoverable in principle, or that Libya's breach was serious. The difficulty lies in how the tribunal moved from a proven small loss to an enormous projected one. Four problems stand out. 6.1 Capital deployed against profits awarded Start with the raw proportion. The money the claimant actually put into the venture, its #sunk_costs, was on the order of five million dollars, and the tribunal recognised roughly that amount as real, out of pocket loss. The projected profit award was nine hundred million. The award for imagined future earnings was therefore something like one hundred and eighty times the capital actually deployed. Set beside the moral damages and costs, the total came to about USD 935 million against a genuine investment measured in single digit millions. A gap of that scale should have been a red flag rather than a result. The principle of #full_reparation restores a claimant to where it would have been; it is not a mechanism for converting a modest, largely undeployed commitment into a fortune. When the compensation is orders of magnitude larger than anything the claimant ever risked, the award has stopped restoring and started transferring. This is the concern that later drove the Egyptian courts, in the annulment phase, to speak of a lost #proportionality between the compensation granted and the damage actually incurred. Whatever the technical merits of that annulment reasoning, which we come to below, the underlying instinct was sound. A number so far removed from the real loss demands an unusually strong justification, and the award did not supply one. Recent writing on compensation has given this instinct a firmer footing. Scholars examining what has been called crippling compensation have questioned whether tribunals should be able to impose sums that bear no sensible relation to the harm and that can fall with real weight on a state and its people. Others, approaching from the economics of investment treaties, have proposed compensation rules designed to avoid exactly the kind of one sided, over generous transfers that Al-Kharafi represents. The common thread is that quantum is not a free zone. It is subject to legal and structural limits, and an award that ignores them is vulnerable, both intellectually and, as it turned out, in court. 6.2 Certainty and the missing operating history The second problem is #certainty. The award described the lost opportunities as real and certain, but the facts point the other way. The resort was never built. There was no hotel, no occupancy data, no revenue history, no customer base, nothing on which a reliable forecast of eighty-three years of profit could rest. Under the #going_concern requirement discussed above, this is precisely the situation in which tribunals are supposed to be most cautious, and in which recovery is usually confined to sunk costs or asset value rather than projected earnings. The tribunal effectively treated an unbuilt project as though it were an operating business with a dependable income stream. It is one thing to accept some approximation in the amount of a loss that is certain to have occurred. It is another to treat the very existence of eight decades of profit as certain when the enterprise never opened its doors. The commentary to #ARSIWA warns that lost profits become more vulnerable to commercial and political risk the further the projection runs, and few projections run further than eighty-three years. By labelling the opportunities certain rather than confronting their speculative character, the award skipped the analytical step that the doctrine exists to force. 6.3 The eighty-three year horizon and Libyan country risk The third problem compounds the second. Even a genuine going concern would be hard to project across eighty-three years. Doing so for an unbuilt resort in Libya, at the exact moment the country was descending into civil war, strains credibility. Sound valuation is supposed to price #country_risk, the political, security, and economic instability that can wipe out returns, through a higher discount rate or an explicit risk adjustment. A serious #DCF exercise for a Libyan tourism asset in 2011 to 2013 would have applied a steep discount for the real possibility that the project would never operate profitably, or at all, in a conflict environment. A projection that produces nine hundred million dollars of profit from such an asset, over such a period, has not taken that risk seriously. The longer the horizon and the higher the instability, the more the forecast becomes an act of imagination rather than measurement. This is where the method's tendency to launder uncertainty into a confident figure does its most damage. The award presents a single, large, definite sum, but behind it lies a tower of assumptions about tourism demand, pricing, costs, and political stability across most of a century in one of the least predictable settings imaginable. Strip away the veneer of precision and what remains is a guess with a decimal point. 6.4 Moral damages The USD 30 million for #moral_damages deserves a brief word because it shows the same tendency in miniature. Moral or reputational damages for a corporate claimant are exceptional in investment arbitration and are usually reserved for egregious conduct, such as physical mistreatment, harassment, or a targeted campaign against the investor. Awarding a substantial sum for reputational harm to a company that had not yet built or operated anything is difficult to square with that restrictive practice. It reinforces the impression that the tribunal approached quantum generously across the board rather than testing each head against its proper legal standard. 6.5 What a disciplined award looks like It helps to see the award against the pattern of decisions that handled similar problems with more care, because the contrast shows that the doctrine is workable and that Al-Kharafi was a departure rather than the norm. Across investment jurisprudence a fairly stable divide has emerged between assets that support forward looking valuation and assets that do not, and tribunals on the disciplined side of that divide tend to reason in a recognisable way. On one side sit operating businesses with real cash flows. A power station that is already generating electricity under a known regulatory regime, for example, has a relatively simple and testable business model. Its demand can be studied, its long run pricing can be modelled from available data, and its costs are known. When such a plant is harmed, tribunals have been willing to award lost future profits over the remaining useful life of the asset, because the projection rests on an operating record rather than on hope. The forecast can be checked against how the business actually performed, and the assumptions can be defended. On the other side sit early stage and pre operational ventures. A mining concession that has reached only the exploration phase, with a permit applied for but production never begun, is the classic example. Faced with such facts, disciplined tribunals have refused to value the project by projected production profits. Their reasoning is consistent: there was no proof the project would ever have produced at a profit, no operating history to test the forecast, and no reason to believe a willing buyer would have paid a price based on speculative #discounted_cash_flow numbers. In those cases the tribunals fell back on what could be shown, the money actually invested, rather than compensating an income stream that had never existed. The measure of #fair_market_value, where it is used, is anchored to what a real purchaser would pay on the valuation date, not to a claimant's most hopeful internal model. Al-Kharafi sits at the extreme end of the second category and yet was treated as though it belonged in the first. The resort had not reached exploration, permitting, or any comparable milestone. It had not been built at all. There was less operating evidence here than in the mining cases where tribunals declined forward looking valuation, and vastly less than in the power plant cases where they allowed it. A faithful application of the pattern would have pointed toward recovery of #sunk_costs and perhaps a measured allowance for the wasted opportunity, not toward nine hundred million dollars of projected profit across eighty-three years. The gap between how the case was decided and how comparable cases were decided is the clearest evidence that the award misapplied the settled approach rather than developing it. Taken together, these four problems do not show that Libya behaved well or that the claimant deserved nothing. They show that the route from breach to nine hundred million ran through every doctrinal checkpoint, certainty, going concern, realistic risk, proportionality, without stopping at any of them. That is why the case became a byword for #speculative_damages, and why it still appears in almost every serious discussion of how not to quantify #lost_profits. 7. The post-award saga: annulment and enforcement An award is only as strong as its ability to survive review and to be enforced. Al-Kharafi turned into one of the longest running post-award battles in the region, and the twists in that battle are as instructive as the award itself. 7.1 The Egyptian annulment proceedings Because the seat of the arbitration was Cairo, Libya sought to set the award aside under Egyptian arbitration law, specifically Law No. 27 of 1994. The first question was whether the Egyptian courts even had the power to hear such a challenge, given that the award had been rendered under the #Unified_Agreement, which pointed disputes toward the Arab Investment Court. This jurisdictional puzzle sent the case bouncing between two levels of the Egyptian judiciary for years. At first the Cairo Court of Appeal declined to annul, reasoning that an award under the treaty fell outside the Egyptian annulment system and belonged to the treaty's own regime. The Egyptian Court of Cassation disagreed. In a decision of 4 November 2015 it reversed and sent the matter back, holding that even if the treaty barred ordinary and extraordinary appeals, it had not clearly barred an original annulment action, and that any award seated in Egypt is subject to the annulment provisions of the Egyptian law. The case returned to the Court of Appeal, was decided again, and went up once more. On 10 December 2019 the Court of Cassation again reversed and remanded, insisting that the annulment court had to examine the challenges to the award on their merits. 7.2 The proportionality annulment and its reversal On the third pass, the Cairo Court of Appeal delivered the decision that made the annulment phase famous. On 3 June 2020, in Judgment No. 39 of Judicial Year 130, it set the award aside on grounds of #public_policy. The court held that the award violated Egyptian public policy because it departed from the principle of equality or #proportionality of compensation, a principle it described as one that cannot be derogated from. It found that the tribunal had failed to apply the rules governing an award of #lost_profits and that the damages were a clear and serious breach of the requirement that compensation be proportionate to the harm actually suffered. The court pointed out, tellingly, that Al-Kharafi had asked for only about five million dollars at the outset. For a moment, then, the controversial award was dead, struck down for the very disproportion this article has described. The victory did not last. Al-Kharafi appealed to the Court of Cassation, and on 24 June 2021, in Case No. 12262 of Judicial Year 90, that court quashed the Court of Appeal's annulment and dismissed the annulment action. Its reasoning was procedural but important. The grounds for annulment under Article 53 of the Egyptian arbitration law are listed exhaustively, and an allegation that damages were grossly overestimated or disproportionate is not among them. Assessing the amount of compensation, the court said, is a matter left to the discretion of the arbitral tribunal, and it falls outside the limited review that an annulment court may conduct. The Court of Appeal had therefore erred by using proportionality as a public policy ground to reopen the merits. In the end the annulment claim failed, and the award survived the challenge at its seat. This outcome is easy to misread, so it is worth stating plainly. The Court of Cassation did not endorse the tribunal's damages. It did not say the nine hundred million was correct or well reasoned. It said that a national court reviewing an award may not second guess the quantum simply because it thinks the number is too high, because doing so would turn annulment into an appeal on the merits. The controversial award stood not because it was right, but because the system of limited judicial review was designed to leave quantum to arbitrators. For some this is reassuring proof that arbitration respects finality. For others it is exactly the danger: if tribunals can produce a figure like this and courts cannot correct it, then the discipline on quantum has to come from the tribunals themselves, and Al-Kharafi shows that it does not always arrive. 7.3 Enforcement in France and beyond While the Egyptian courts wrestled with annulment, the claimant pursued the money. #France was a key battleground. The award was declared enforceable by an order of the President of the Paris first instance court on 13 May 2013, and that order was confirmed by the Paris Court of Appeal on 28 October 2014. Libya's further challenge before the French Court of Cassation was dismissed. Under French arbitration law, an award can be recognised and enforced in France even if it has been set aside at its seat, so the Egyptian annulment proceedings, whichever way they went, were never going to be decisive for the French exequatur. Enforcement then became a long campaign against Libyan assets across several jurisdictions. The claimant pursued Libyan property abroad, including a well publicised attempt to seize an aircraft associated with the former Libyan leadership that was undergoing maintenance in France, an attempt that did not succeed. A large part of the later litigation shifted to the #Libyan_Investment_Authority and its holdings, and disputes over attachment and #enforcement continued through the French courts and elsewhere well after the original award, with proceedings recorded into the mid 2020s. The practical lesson is sobering. A single award, once rendered, can generate a decade or more of satellite litigation on three continents, and the quality of the original damages reasoning determines how much of that follow on conflict was avoidable. 7.4 The numbers in perspective It is worth holding the figures side by side one more time, because the scale is easy to lose in the legal detail. The claimant's genuine, provable outlay was on the order of five million dollars. The projected profit award was nine hundred million. The reputational award was thirty million, six times the real investment. The whole award, before interest, came to roughly nine hundred and thirty-five million dollars, and running interest at four per cent carried it toward one billion within a few years. A dispute over a plot of coastal land, on which nothing had been built, produced a liability approaching the annual budget of a small government. Those numbers explain why the case matters far beyond the parties. For a state emerging from conflict, an award of this size is not an abstract accounting entry; it is a claim on public resources that could otherwise fund reconstruction, and it exposes national assets held abroad to seizure. This is the concern that scholarship on crippling compensation has raised: at some point the size of an award stops being a private remedy and becomes a public burden with consequences for people who were never parties to the contract. Al-Kharafi does not prove that every large award is illegitimate, but it does show how quickly a modest commercial disappointment can be converted, through a chain of optimistic assumptions, into a sum with national significance. The distance between the harm and the remedy is the whole problem in a single line. 8. Lessons and implications Al-Kharafi is more than a curiosity. It carries concrete lessons for the different actors in the system. 8.1 For tribunals The first lesson is for arbitrators, and it is about discipline in the damages phase. The doctrine already contains the tools needed to avoid an Al-Kharafi style result: the #certainty requirement, the #going_concern test, the #new_business_rule, realistic pricing of #country_risk, and the internal restraint built into the #full_reparation principle. What the case shows is that these tools only work if tribunals actually use them and explain their use. Labelling speculative profits real and certain is not analysis; it is assertion. A well reasoned award confronts the weakness of a long range projection openly, either declining forward looking #lost_profits where the venture never operated, or discounting them heavily and saying why. Because national courts often cannot review quantum, as the Egyptian Court of Cassation confirmed, the burden of getting quantum right rests overwhelmingly on the tribunal. Finality is a privilege that assumes rigour. The second lesson concerns method. #DCF and similar forward looking techniques should be matched to the asset. For a going concern with a track record, they can be applied with care. For an unbuilt project, they invite exactly the inflation that critics of #ISDS have documented. Where the only honest measure is #sunk_costs or asset value, tribunals should use it and resist the pull of a bigger, shinier number. A modest, defensible award does more for the legitimacy of the field than a spectacular, fragile one. 8.2 For states and treaty drafters States drew their own conclusions from cases like this. One response has been to write clearer limits on compensation and on the use of speculative valuation directly into investment instruments, so that tribunals are not left to import an open ended reading of #full_reparation. Provisions that restrict #lost_profits for pre operational projects, that require an operating history before forward looking valuation is used, or that cap or structure damages, all respond to the fear of an Al-Kharafi outcome. Scholars proposing reformed compensation rules are, in effect, trying to move the discipline from the goodwill of individual arbitrators into the text of the instruments themselves. A second lesson for states is defensive and practical. The choice of #seat matters enormously. Because Libya's award was seated in Cairo, it became reviewable under Egyptian law, which is why the annulment saga was possible at all. Had the treaty regime provided a genuine internal review mechanism, or had the seat been chosen differently, the post-award story would have unfolded in another way. States negotiating and drafting dispute clauses should think hard about where awards will be seated and what review will be available, because those decisions shape their options long after any breach. 8.3 For the Unified Agreement regime The case also exposed a structural gap in the #Unified_Agreement itself. The treaty pointed disputes toward the Arab Investment Court, but that court has disclaimed jurisdiction to scrutinise ad hoc awards, leaving no reliable internal check on decisions like Al-Kharafi. The result was a regime capable of producing a near billion dollar award with no built in appeal or annulment, which is precisely why the parties ended up in the Egyptian and French courts instead. Commentators analysing the aftermath have argued that both the original agreement and its later amended version suffer from serious shortcomings in judicial oversight. A protection regime that can generate very large awards but offers no dependable way to correct a defective one is not stable. The long Egyptian litigation was, in a sense, the system improvising a review function that the treaty had failed to supply. 8.4 The wider debate on quantum Finally, Al-Kharafi feeds the larger debate about whether #investment_arbitration awards have grown too large and too speculative. The rise of billion dollar claims, the routine use of #DCF for assets that are far from going concerns, and the worry that compensation can be crippling for states have all become central themes in the reform conversation now unfolding in international bodies. Al-Kharafi is one of the sharpest data points in that conversation. It is the case people reach for when they want to show, in a single figure, how far a quantum decision can travel from the money a claimant actually risked. Whether the field tightens its own standards or has them imposed by treaty text, the direction of travel is toward more scrutiny of damages, and this case is one of the reasons why. 8.5 A note on legitimacy There is a final, less technical implication. Systems of dispute resolution survive on trust, and trust in arbitration depends on the sense that outcomes are proportionate to what happened. When a claimant who spent a few million recovers close to a billion for a project that never opened, the result is hard to explain to anyone outside the specialist bar, and impossible to explain to a public asked to pay for it. Every award of that kind gives ammunition to those who argue that the whole system tilts toward capital and against states, and it makes the political case for reform easier to make. The irony is that the finality which protected the Al-Kharafi award, the very rule that kept courts from second guessing quantum, is also what makes disciplined damages reasoning so important. If the arbitrators are effectively the last word on the number, then the legitimacy of the number rests entirely on them. Awards that respect the limits described in this article protect not only the losing state but the credibility of arbitration as a whole. Awards that ignore those limits spend that credibility, and it is not easily replaced. 9. Conclusion The Al-Kharafi award is not a simple morality tale. Libya did breach its obligations. The investor was kept off the land it had been promised and then had its licence cancelled, and it was entitled to be made whole. The trouble was in the meaning of whole. Being made whole should have meant recovering the money actually lost and, at most, profits that could be shown with reasonable #certainty to have followed from a genuine, operating venture. Instead, an unbuilt resort in a country sliding into war was valued as though it were a settled, profitable business with eighty-three guaranteed years ahead of it, and a real loss of a few million dollars became a projected award of nine hundred million. The doctrine had answers to this. The #full_reparation principle restores rather than enriches. #Lost_profits require certainty, not hope. The #going_concern rule warns against projecting earnings for businesses that never traded. Realistic valuation prices #country_risk rather than wishing it away. The award moved past each of these without pausing, and the outcome was a sum so disproportionate to the harm that even a national court, reaching for #public_policy, tried to strike it down, only to be told that quantum is the arbitrators' domain and beyond ordinary review. That last point is the enduring one. The Egyptian Court of Cassation let the award stand in 2021 not because the damages were sound but because the system trusts tribunals with quantum and gives courts little room to interfere. If that trust is to be justified, the discipline has to live inside the awards themselves. Al-Kharafi is the case that shows what happens when it does not. It endures in the literature and the classroom as a warning: that the most dangerous paragraph in an award is often the one that turns a modest, provable loss into an imagined fortune, and that the credibility of #investment_arbitration depends on tribunals treating #lost_profits as a claim to be proved, not a prize to be granted. #Al_Kharafi_v_Libya #Lost_Profits_in_Arbitration #930_Million_Award #Investment_Arbitration #Projected_Future_Profits #Capital_Deployed_vs_Damages #Unified_Arab_Agreement #Discounted_Cash_Flow #Speculative_Damages #Proportionality_of_Compensation #Full_Reparation_Principle #Damages_Quantum #ISDS_Reform #Arbitration_Law #Libya_Arbitration References Aisbett, E., and Bonnitcha, J. (2021). A Pareto-improving compensation rule for investment treaties. Journal of International Economic Law, 24(1), 181-202. Badr, Y. I. (2021). Resolving Arab capital investment disputes: The aftermath of the Al-Kharafi award's annulment by the Egyptian courts. Journal of Law in the Middle East, 1. Bonnitcha, J., and Aisbett, E. (2021). Against balancing: Revisiting the use and regulation distinction to reform liability and compensation under investment treaties. Michigan Journal of International Law, 42(2), 231. Dolzer, R., Kriebaum, U., and Schreuer, C. (2022). Principles of international investment law (3rd ed.). Oxford: Oxford University Press. Hailes, O. (2023). Unjust enrichment in investor-State arbitration: A principled limit on compensation for future income from fossil fuels. Review of European, Comparative and International Environmental Law, 32. https://doi.org/10.1111/reel.12483 Kumar, V., and Sharma, R. (2023). Valuation of damages in cases of unlawful expropriation of data assets: Pitfalls of the DCF method. SSRN Working Paper Series. Marboe, I. (2017). Calculation of compensation and damages in international investment law (2nd ed.). Oxford: Oxford University Press. Paparinskis, M. (2022). Crippling compensation in the International Law Commission and investor-State arbitration. ICSID Review - Foreign Investment Law Journal, 37(1-2), 289-312. https://doi.org/10.1093/icsidreview/siab029 Shehata, I., Rasekh, A., and Duggal, K. (2022). All's well that ends well? Looking at the future of the Unified Arab Agreement in light of the Al-Kharafi v Libya decisions by the Egyptian courts. ICSID Review - Foreign Investment Law Journal, 37(3), 654. https://doi.org/10.1093/icsidreview/siac007 The guide to damages in international arbitration (6th ed.). (2024). London: Global Arbitration Review.

The phrase "persona non grata" sits at the meeting point of law, power, and everyday statecraft. It allows a host country to tell a foreign government that one of its diplomats is no longer welcome, and it allows the host to do this without offering any reason. This article studies the legal shape of that power. It explains where the rule comes from, how it is written into Article 9 of the #Vienna_Convention on #Diplomatic_Relations of 1961, and why the drafters chose to free the #receiving_state from any duty to justify its decision. The study then looks at how #persona_non_grata works in practice: the notice, the reasonable time to leave, the recall of the diplomat, and the loss of status that follows if the #sending_state refuses to act. It connects the rule to the wider system of #diplomatic_immunity, showing that expulsion is often the only lawful tool a host has against a diplomat who cannot be arrested or put on trial. The article also treats the consular parallel in Article 23 of the 1963 Convention. Most of all, it examines how states have turned this quiet legal device into a public instrument of pressure, from Cold War spy swaps to the mass #expulsion of Russian officials after 2018 and 2022. The article argues that the "no reasons" rule protects both #sovereignty and stability, yet it also invites abuse, reputational harm, and cycles of retaliation. It ends by weighing these competing values and by suggesting how students and future practitioners should read any expulsion headline through the lens of #international_law. Keywords: persona non grata; Vienna Convention on Diplomatic Relations; Article 9; diplomatic immunity; diplomatic expulsion; sovereignty; receiving state; sending state; reciprocity; coercive diplomacy; consular relations; international law Introduction Every so often the news carries a short, formal sentence: a government has declared a foreign diplomat "persona non grata" and given the person a few days to leave the country. The language is old and Latin, and the tone is cold and polite at the same time. Behind that sentence lies one of the most striking powers in all of #international_law. A #host_state may remove a foreign #diplomatic_agent from its territory, and it does not have to say why. There is no trial, no public evidence, and no court that can overturn the decision. The diplomat simply has to go home. This power is not a loophole or an accident. It is a settled rule, written down in Article 9 of the #Vienna_Convention on Diplomatic Relations of 1961, and accepted by almost every country on earth. The rule reflects a very old idea in the law of nations: no state can be forced to keep a foreign representative it no longer trusts. At the same time, the rule has become something more than a legal safety valve. In recent years it has grown into a #diplomatic_weapon, a way for governments to punish, to signal anger, and to answer hostile acts without firing a shot. The mass removals of Russian officials after the 2018 Salisbury poisoning and after the 2022 invasion of Ukraine show how a quiet legal device can be used at scale and in public. This article has three aims. The first is to explain the legal contours of #persona_non_grata in plain language, so that a student can understand exactly what the rule says, where it comes from, and how it fits inside the larger system of #diplomatic_immunity. The second aim is to look closely at the most unusual feature of the rule, the freedom of the #receiving_state to act "without having to explain its decision," and to ask why the law was built this way. The third aim is to study the rule in action, as a tool of pressure between states, and to weigh its benefits against its costs. The argument developed here is simple to state. The "no reasons" rule is not a defect. It is a deliberate design choice that protects #sovereignty, shields sensitive intelligence and evidence, and keeps disputes out of courts that would only make them worse. Yet the same freedom that makes the rule useful also makes it open to misuse. It can be used to harass, to retaliate for reasons that have nothing to do with the diplomat's conduct, and to damage a person's reputation without any chance to reply. The task for a modern reader is to hold both truths at once: the rule is necessary, and the rule is dangerous. Understanding that tension is the key to reading any #diplomatic_expulsion clearly. The article proceeds in stages. It first defines the term and its scope. It then traces the history from ancient custom to modern codification. It sets out the text of Article 9 and the meaning of each part. It studies the "no reasons" rule and the reasons behind it. It describes the procedure and its consequences. It links the rule to immunity and to the consular system. It surveys state practice and the rise of #coercive_diplomacy. Finally it examines the main criticisms and offers a balanced conclusion. Defining Persona Non Grata: Meaning and Scope The Latin phrase persona non grata means, roughly, "an unwelcome person." In ordinary speech people sometimes use it loosely, as when a town declares a celebrity unwelcome, or when a club bars a former member. In #diplomatic_law the phrase has a precise and technical meaning. It describes the formal act by which a #receiving_state notifies a #sending_state that a named member of a diplomatic mission is no longer acceptable, with the result that the person must be recalled or must leave. Three points give the term its shape. First, the concept is about consent rather than proof. A diplomat serves in a foreign country only because that country agrees to accept the person as a representative. When the host withdraws that acceptance, the legal basis for the diplomat's presence falls away. The host is not accusing the person of a crime in a courtroom sense. It is simply saying that it no longer consents to the person's continued presence in a diplomatic role. Second, the concept can be used before a diplomat ever arrives. A government does not have to wait until the person is on its soil. It may signal in advance that a proposed appointee will not be received. This early use is quieter and less dramatic, but it rests on the same principle: the host controls who may serve as a foreign envoy within its borders. Third, the term has a companion phrase. Article 9 speaks of declaring a diplomatic staff member #persona_non_grata, but it uses a slightly different label, "not acceptable," for other members of the mission staff, such as administrative and technical employees. The legal effect is the same. The distinction simply tracks the different ranks and roles inside a #diplomatic_mission. For the purposes of this article, both labels describe the same underlying power: the right of the host to refuse or end the presence of foreign mission personnel. It is worth separating persona non grata from three neighbouring ideas that are often confused with it. Expulsion is the broad act of forcing someone to leave a country, and it can apply to ordinary foreigners under immigration law. Persona non grata is the specific diplomatic version of that idea, aimed at accredited mission members. Breaking off #diplomatic_relations is a far larger step, in which two states end their formal ties entirely and close their embassies. Recalling an ambassador "for consultations" is a milder signal of displeasure, in which the sending state itself brings its envoy home for a time. Persona non grata sits between these. It removes a specific person without necessarily ending the relationship, and it is driven by the host rather than the sending state. Historical Foundations: From Custom to Codification The idea that a ruler may refuse an unwelcome envoy is much older than any treaty. Long before modern international law took shape, societies that exchanged messengers understood two linked customs. The first was that envoys should be protected and allowed to travel safely, because killing or mistreating a messenger made all future contact impossible. The second was that a host was never obliged to keep an envoy it distrusted. These two customs balanced each other. The messenger was safe, but the messenger could be sent home. Writers who shaped the early law of nations recognised this balance. Classical scholars of diplomacy accepted that a receiving prince could expel a diplomat who plotted against him or interfered in his affairs, while still respecting the envoy's person and safe departure. The right to remove an unwelcome representative was treated as a natural feature of independence. A state that could not choose whom to admit as a foreign envoy would not really be sovereign at all. For centuries this power lived in #customary_international_law, which is law built from the consistent practice of states acting out of a sense of legal duty. The custom was clear enough in principle but messy in operation. There was no agreed procedure, no fixed idea of how much time a diplomat should be given, and no shared vocabulary. Because of this, expulsions often turned into open quarrels, and one removal could trigger an escalating exchange that damaged relations far beyond the single diplomat involved. The nineteenth century brought the first steps toward order. The Congress of Vienna in 1815 produced a common understanding on the ranking of diplomats and on certain points of protocol, which reduced some sources of friction. Still, the core power to expel remained uncodified and uneven. Through two world wars and the collapse of empires, diplomacy grew more complex, travel and communication became routine, and the number of states multiplied. The old customs could no longer carry the weight placed on them. The decisive moment came after the Second World War. The founding of the United Nations created both the machinery and the political will to write down the rules of diplomacy in a single, widely accepted treaty. The International Law Commission, the expert body charged with developing and codifying #international_law, prepared draft articles over several years. Its work led to a conference in the Austrian capital, and on 18 April 1961 states adopted the #Vienna_Convention on Diplomatic Relations. The Convention entered into force in 1964 and has since been ratified by the overwhelming majority of the world's states, making it one of the most successful treaties ever concluded. The Convention did not invent the power to declare a diplomat #persona_non_grata. It captured an old custom and gave it a clear, predictable, and unilateral form. As the leading commentary on the treaty explains, the drafters placed each provision in the context of prior practice, so that the Convention reads as a living instrument grounded in history rather than a set of fresh inventions (Denza, 2025). Article 9 is the article that turned the ancient right of refusal into a modern legal mechanism. It made the declaration one sided, immune from outside challenge, and subject only to a duty of reasonable timing. In doing so, it gave every bilateral relationship a built in exit door that does not depend on the goodwill of the other side. The Legal Framework: Article 9 of the Vienna Convention The heart of the subject is a single article. Article 9 of the #Vienna_Convention on Diplomatic Relations has two short paragraphs, and almost everything in this field flows from them. The first paragraph gives the #receiving_state the right, at any time and without having to explain its decision, to notify the #sending_state that the head of the mission or any member of the diplomatic staff is #persona_non_grata, or that any other member of the mission staff is not acceptable. In that case the sending state must either recall the person or end the person's functions with the mission. The paragraph adds that a person may be declared non grata or not acceptable even before arriving in the territory of the receiving state. The second paragraph deals with what happens if the sending state does not cooperate. If the sending state refuses, or fails within a reasonable period, to recall the person or end their functions, the receiving state may refuse to recognise the person concerned as a member of the mission. This is the enforcement backstop. The host cannot itself arrest the diplomat while immunity lasts, but it can strip away the recognition on which the diplomat's status depends. Several features of this text deserve close reading. The phrase "at any time" means the power has no fixed trigger and no expiry. A host may act on the first day of a posting or years later. There is no requirement that the diplomat have committed any identifiable act. The phrase "without having to explain its decision" is the most important clause in the whole field, and the next section is devoted to it. The choice offered to the sending state, to recall or to terminate functions, shows that the immediate aim is to end the person's diplomatic role, not to punish the individual in a criminal sense. The words "as appropriate" and "within a reasonable period" introduce a measure of flexibility and good faith. The receiving state cannot demand that a diplomat vanish within minutes, and the sending state cannot drag its feet forever. In serious cases, especially those involving #espionage or grave misconduct, host states in practice set very short deadlines, sometimes twenty four to seventy two hours. In less urgent cases the timing may stretch to weeks, allowing an orderly handover and departure. It is important to keep Article 9 apart from two neighbouring articles that are easy to confuse with it. Article 11 concerns the size of a mission. It allows the receiving state to require that a mission be kept within limits it considers reasonable and normal. A #persona_non_grata notice is not the right tool for merely trimming staff numbers; that is Article 11's job. Article 43 lists the ways in which the function of a diplomatic agent comes to an end, including notification by the receiving state under Article 9. In other words, when a sending state fails to comply and the host refuses recognition, Article 43 describes the formal end of that person's diplomatic function. Read together, these provisions form a coherent system: Article 9 creates the power, Article 11 handles mission size, and Article 43 records the legal endpoint. Two further articles supply the moral and legal background against which expulsions are judged. Article 41 places a duty on all persons enjoying privileges and immunities to respect the laws and regulations of the receiving state and not to interfere in its internal affairs. Host states often point to Article 41, at least implicitly, when they act, because interference and lawbreaking are the classic grounds for removal. Article 29 guarantees the inviolability of the diplomatic agent, meaning the person may not be arrested or detained. Because of that #inviolability, expulsion under Article 9 is usually the strongest lawful response a host can make. The rule and the immunity are two halves of one design. The "No Reasons" Rule: Sovereignty, Discretion, and Its Justifications The defining feature of #persona_non_grata is that the #receiving_state does not have to give reasons. This is not a small technical point. It is the axis on which the whole doctrine turns, and it explains why the tool is so powerful and so open to debate. Start with what the rule removes. In most areas of public life, when a state takes a serious action against a person, it must usually offer some justification and often some process. A person facing punishment expects to know the charge and to have a chance to respond. Article 9 sweeps all of that away in the diplomatic setting. The host does not have to prove misconduct, produce evidence, name an offence, or defend its choice in any forum. It simply gives notice, and the legal consequences follow. Why would the drafters build a rule this way? The first reason is #sovereignty. A state's right to decide who represents foreign powers within its borders is treated as a basic attribute of independence. To force a host to justify each refusal would be to place its judgment under the review of the very state whose diplomat is being removed, or of some outside body. That would turn a matter of national self protection into a contest the host might lose. The "no reasons" rule keeps the final say where the drafters thought it belonged, with the host. The second reason is the protection of sensitive information. Many expulsions arise from suspected #espionage or from covert activity that the host has detected through intelligence work. If the host had to explain its decision, it would often have to reveal how it knew what it knew. That could expose sources, methods, ongoing investigations, or the identities of informants. By allowing the host to act in silence, the rule lets governments defend themselves without burning their own tools. This is why the tool is so common in spy cases, where the evidence is classified and can never be shown in open court. The third reason is the avoidance of escalation. Requiring reasons would invite argument. The sending state would dispute the facts, demand proof, and treat every removal as an accusation to be rebutted. Disputes would drag on and poison relations. By making the declaration unilateral and unexplained, the rule allows a diplomat to be quietly removed while both sides save face. In many real cases the host uses a soft formula, describing the diplomat's conduct as "incompatible with diplomatic status," the sending state issues a routine denial, and the relationship survives the incident. The fourth reason is the absence of a suitable court. There is no standing tribunal with power to review whether a host's suspicion was well founded. Even if reasons were required, no neutral body could reliably weigh secret intelligence across borders. Rather than create an unworkable review system, the drafters chose a clean rule that needs no adjudication. A short example makes the logic concrete. Suppose a host's security service watches a foreign diplomat meet, again and again, with a local official who has access to defence secrets, and suppose the pattern strongly suggests recruitment of a spy. The host cannot arrest the diplomat, because of #inviolability. If it had to prove its case in public before removing the person, it would have to reveal the surveillance, the informants, and perhaps the identity of the compromised official. That disclosure could destroy an ongoing investigation and endanger real people. Under Article 9, the host can simply declare the diplomat #persona_non_grata, describe the conduct in a bland phrase, and send the person home. The threat is neutralised, the secrets are kept, and the relationship survives with minimal damage. This is the everyday scenario the "no reasons" rule was built for, and it explains why intelligence services around the world treat expulsion as a normal, almost routine instrument. The drafting history of the Convention supports this reading. During the work of the International Law Commission and at the Vienna conference, some delegations pressed for tighter controls on the power to expel, out of concern that it might be abused against innocent envoys. Others insisted that any such controls would cripple the tool and invade the host's #sovereignty. The compromise that survived was the one in the final text: a broad, unexplained power, tempered only by the duties of good faith and reasonable timing. The balance struck in 1961 has held ever since, which is itself evidence that states, whatever their complaints, find the arrangement acceptable. None of this means the rule is beyond principle. The receiving state is still expected to act in good faith and within a reasonable time, and the freedom from giving reasons does not license physical mistreatment of the diplomat, who remains inviolable until departure. Scholars who study the balance between #diplomatic_immunity and accountability stress that the freedom to expel is meant to be the safety valve of the system, the lawful answer to abuse, not a licence for the host to behave lawlessly in turn (Ahmad, Lilienthal, and Asmad, 2024). The "no reasons" rule, in short, is a considered trade. It sacrifices transparency and individual process in order to protect #sovereignty, secrecy, and stability. Whether that trade is worth it in every case is a question this article returns to later. Procedure and Consequences: How a Persona Non Grata Declaration Works in Practice Although Article 9 is short, the practice around it follows a fairly regular pattern. Understanding the steps helps to demystify the headlines. The first step is internal. Before anything is said publicly, the host government decides that a diplomat should go. That decision may follow a security service report, a criminal incident, a pattern of interference, or a purely political judgment. The decision is usually coordinated among the foreign ministry, security agencies, and the political leadership, because an #expulsion carries diplomatic weight and will often provoke a response. The second step is notification. The host formally tells the #sending_state, usually by summoning the ambassador or a senior diplomat to the foreign ministry and delivering a note. This meeting is sometimes called a demarche. The note names the person, states that the person is #persona_non_grata or not acceptable, and typically sets a deadline for departure. The tone can range from restrained to openly hostile, depending on the political message the host wishes to send. The third step belongs to the sending state. Under Article 9 it must recall the person or end their functions. In the great majority of cases the sending state complies, because the alternative is worse. Compliance is not an admission of guilt. States routinely recall their people while denying any wrongdoing, treating the removal as an unfriendly act rather than a proven offence. The fourth step is departure and the winding down of status. The diplomat leaves, usually within the stated period. Immunity and #inviolability continue to protect the person up to and during departure, so the host cannot use the expulsion as a chance to detain or prosecute. The duration of privileges is governed by Article 39, under which they generally cease when the person leaves the country or when a reasonable period to do so expires. The fifth step arises only if the sending state does not cooperate. If it refuses or fails to act within a reasonable time, Article 9(2) allows the host to refuse to recognise the person as a member of the mission. Once recognition is withdrawn, the legal shield begins to fall away. The person is no longer treated as a diplomat, which means the protections attached to that status can lapse and the host may then deal with the person under its ordinary law. This backstop is rarely reached, precisely because sending states know it exists and prefer to recall their people first. The consequences of the whole process are mainly relational rather than personal in the criminal sense. The individual diplomat is not punished by a court; they are sent home. The real effects land on the relationship between the two states. An expulsion signals distrust, often invites a matching response, and can shrink the practical ability of the affected embassy to operate. In large or repeated actions, missions can be left short staffed, and everyday functions such as visas, consular help, and normal contact can suffer. In this way a single legal act ripples outward into the broader ties between nations. It is also worth noting what a #persona_non_grata declaration is not. It is not a criminal verdict, and it creates no formal finding of guilt. It is not, by itself, a rupture of relations, although a wave of expulsions can push two states toward that point. And it is not a tool for reducing staff numbers, which the Convention handles separately. Keeping these boundaries in mind prevents the common mistake of reading every expulsion as proof that a crime occurred. Persona Non Grata and Diplomatic Immunity: Why Expulsion Is the Main Remedy To see why persona non grata matters so much, one has to understand the problem it solves. That problem is #diplomatic_immunity. A diplomatic agent enjoys strong protection under the #Vienna_Convention. Article 29 makes the person inviolable, so the host may not arrest or detain the diplomat. Article 31 grants immunity from the criminal jurisdiction of the receiving state, and broad immunity from its civil and administrative jurisdiction as well. The purpose of these protections is not to place diplomats above the law for their own comfort. As the Convention's own preamble makes clear, the purpose is functional: immunities exist to ensure that missions can perform their duties efficiently and without pressure or intimidation from the host. A diplomat who feared arrest on political pretexts could not do the job. The price of this design is obvious. If a diplomat breaks the law, the host usually cannot prosecute. It can ask the sending state to waive immunity under Article 32, but waiver is entirely at the sending state's discretion and is often refused, especially in serious or embarrassing cases. So the host is left with a diplomat it cannot arrest, cannot try, and cannot easily hold accountable through its courts. This is exactly where #persona_non_grata does its work. When the ordinary tools of the criminal law are blocked by immunity, expulsion becomes the host's main lawful remedy. The host cannot put the diplomat in the dock, but it can end the diplomat's presence. In this sense the power to expel is the necessary counterweight to the power of immunity. The two doctrines are designed to fit together. Immunity keeps the host's hands off the diplomat's body and freedom; persona non grata keeps the host's ultimate control over who stays. The International Court of Justice recognised this relationship in the case brought by the United States against Iran after the 1979 seizure of the American embassy and its staff in Tehran. The Court underscored that the diplomatic law regime is a self contained system that both grants immunities and provides its own remedy for abuse. If a host believes a diplomat has acted improperly, the Convention's answer is to declare the person #persona_non_grata, not to seize the mission or its people. The remedy for abuse of immunity is expulsion, not the destruction of the protections themselves. That reasoning has become a cornerstone of the modern understanding of the field. Recent practice keeps testing the edges of immunity, and this in turn keeps persona non grata relevant. Courts in various countries have wrestled with claims that fall outside the core diplomatic function, such as exploitation of domestic workers in a diplomat's household. In one prominent decision, a national supreme court examined allegations that a domestic worker had been held in conditions amounting to modern slavery and considered whether the commercial activity exception to immunity applied (Basfar v Wong, 2022). Such cases show that immunity is not always absolute and that its boundaries are contested. Yet even where a court finds a way through immunity, the diplomatic relationship usually still ends with the quiet or not so quiet departure of the person, because the underlying rule remains: a host that no longer wants a foreign representative can always send that person home. The Consular Parallel: Article 23 of the Vienna Convention on Consular Relations Diplomats are not the only foreign officials who serve abroad. Consular officers handle practical matters such as passports, visas, trade contacts, and help for citizens in trouble. They are governed by a separate but closely related treaty, the #Vienna_Convention on Consular Relations of 1963. That treaty contains its own version of the persona non grata rule in Article 23. The consular provision mirrors the diplomatic one. It allows the receiving state to notify the sending state that a consular officer is #persona_non_grata, or that any other member of the consular staff is not acceptable. As with diplomats, the sending state must then recall the person or end their functions, and if it fails to act within a reasonable time, the receiving state may withdraw recognition. The consular treaty also states that the host is not obliged to give reasons, and it allows a person to be declared unacceptable before arrival. The parallel is deliberate. It reflects the settled view that the presence of any foreign official rests on the continuing #consent of the host. There are two useful differences to keep in mind. First, consular immunity is generally narrower than diplomatic immunity. Consular officers are usually protected mainly for acts performed in the exercise of their consular functions, rather than enjoying the broad personal immunity of a diplomat. This means that for consular staff, ordinary criminal process is sometimes available for private conduct, and expulsion is not always the only option. Second, the consular system relies on a specific host authorisation, the exequatur, which is the formal permission a host grants for a consular officer to carry out duties. Because that permission is explicit, its withdrawal is a clean and visible legal step. Where a host wants to remove a consular officer, it can withdraw the exequatur or declare the person not acceptable, and the effect is much the same as a diplomatic expulsion. For a student, the key lesson is that the persona non grata principle is not limited to ambassadors and embassy staff. It runs through the whole architecture of official foreign presence. Wherever one state accepts the officials of another, that acceptance can be withdrawn. The consular parallel simply confirms how basic the underlying idea is. Persona Non Grata as a Diplomatic Weapon: State Practice and Coercive Diplomacy So far the article has treated persona non grata as a legal mechanism. In the real world it is also a tool of pressure, and this dual character is what makes it so interesting. The same rule that quietly removes a single misbehaving envoy can be used loudly and in bulk to punish a foreign government. When it is used this way, it becomes part of what scholars call #coercive_diplomacy, the use of measured pressure to change or answer another state's behaviour. During the Cold War, expulsions were routine currency between the superpowers. The United States and the Soviet Union removed each other's diplomats in waves, often on suspicion of #espionage, and often in matched numbers. An expulsion by one side would draw a mirror expulsion by the other, a pattern so regular that it became a familiar rhythm of the era. These exchanges rarely broke relations. They were understood as a controlled way to register displeasure and to thin out suspected intelligence officers working under diplomatic cover. The tit for tat pattern outlived the Cold War. When one state expels another's diplomats for political reasons, a matching response is now almost automatic. This #reciprocity is not required by Article 9, which imposes no duty to retaliate. It is a feature of state practice, driven by the need to appear strong and to avoid the impression of weakness. The result is that politically motivated expulsions tend to come in pairs, and sometimes in escalating rounds. Several modern episodes show the range of the tool. In December 2016 the United States declared thirty five Russian officials #persona_non_grata as part of its response to interference in the American election, describing their conduct as inconsistent with their diplomatic status. That action signalled a shift: cyber operations and election meddling were now treated as grounds for mass diplomatic expulsion, not merely as matters for private complaint. The Salisbury poisoning of 2018 pushed the tool to a new scale. After the attack on a former Russian intelligence officer on British soil, the United Kingdom expelled twenty three Russian diplomats. What made the episode remarkable was the coordinated allied response. More than twenty partner countries joined in, expelling well over a hundred Russian officials in a matched, collective action. Legally, each expulsion remained a separate bilateral act by each host under its own Article 9 power. Politically, the effect was a single, unified message of condemnation. This showed that #persona_non_grata could serve not only as a bilateral tool but as a way for a group of states to speak with one voice. The largest wave came after the Russian invasion of Ukraine in 2022. European and other states expelled Russian diplomats in numbers that mounted through the year, and by early 2023 the running total approached six hundred. Security officials in some of those states later assessed that a large share of the expelled officials were in fact intelligence operatives working under diplomatic cover, and one senior intelligence chief described the removals as a serious blow to hostile intelligence networks across the continent. Russia responded with its own expulsions of Western diplomats, and the two sides shrank their diplomatic footprints in each other's capitals. Researchers who study this episode have argued that the surge is best understood through the lens of #securitisation, the process by which states frame an issue as an existential security threat and then justify extraordinary measures in response (Moritani and Akiyama, 2023). Seen this way, the expulsions were not only punishments; they were public declarations that the presence of these officials had become a matter of national survival. Beyond the Russia related cases, the tool appears constantly in regional and bilateral disputes. States in South Asia have expelled each other's officials over alleged spying and interference, often in matched pairs, with each removal answered by another. In 2023 a serious rift between Canada and India led to reciprocal expulsions after Canada linked Indian agents to a killing on its soil, a case that showed how quickly persona non grata can escalate a broader political quarrel. Governments have also used the tool to protest another state's foreign policy, expelling an ambassador or senior diplomat to register anger over a war, an offensive, or a diplomatic insult, even when the diplomat personally did nothing wrong. A recurring feature of this practice is the careful language that surrounds it. Because the host is not required to give reasons, and because a public accusation of espionage can inflame relations, governments have developed a set of polite formulas. The most common is the phrase "activities incompatible with diplomatic status," which everyone understands to mean spying or covert interference, but which stops short of a direct charge. This wording lets the host act firmly while giving the #sending_state room to deny wrongdoing and avoid open humiliation. In calmer disputes both governments use the formula, the diplomat departs, and the incident closes without a public quarrel. In sharper confrontations the host drops the euphemism and states the accusation openly, which raises the temperature and signals that the relationship itself is in trouble. Reading the choice of words is therefore one of the best ways to judge how serious an #expulsion really is. The tool is also used before a diplomat ever takes up a post. A host that objects to a proposed appointee, perhaps because of the person's past conduct or known intelligence links, can quietly signal that the person will not be received. This pre arrival use rarely reaches the news, precisely because it is designed to avoid a public clash. It shows that #persona_non_grata is not only a punishment applied after the fact but also a filter that a #receiving_state can use to shape the mission it is willing to host. An older episode illustrates how the power interacts with the hard limits of immunity. In 1984 shots were fired from the Libyan diplomatic premises in central London, and a British police officer, Yvonne Fletcher, was killed. The people inside the mission were protected by #inviolability, so British authorities could not storm the building or arrest those responsible while immunity lasted. The United Kingdom broke off #diplomatic_relations with Libya, and those inside eventually left the country under the protection their status still gave them. The case is remembered as a stark demonstration of the system's design. Even in the face of a killing, the lawful response ran through expulsion and the severing of relations rather than through force against protected persons. It also fed a long running debate about whether the immunity regime grants too much shelter to serious wrongdoing, a debate that continues in the scholarship today. These examples reveal a spectrum of uses. At one end sits the classic case: the removal of a specific person for specific misconduct, handled quietly. In the middle sits the signalling case: an expulsion used to express displeasure with another government's policy. At the far end sits the mass, coordinated case: dozens or hundreds of removals used as a headline instrument of pressure. All three rest on the same short article, and all three enjoy the same freedom from any duty to explain. The flexibility of the rule is exactly what allows it to stretch across this whole range. Limits, Controversies, and Critiques For all its usefulness, #persona_non_grata attracts serious criticism. A fair study must set out the main objections and weigh them honestly. The first criticism concerns arbitrariness. Because the host need give no reasons, nothing on the face of the law prevents an expulsion that is unfair, mistaken, or purely political. A diplomat may be removed on a thin suspicion, on a case of mistaken identity, or as a convenient scapegoat during a crisis. There is no hearing at which the person can answer, and no court that can test the host's judgment. Critics argue that this offends basic ideas of fairness and due process, especially when the host publicly accuses the diplomat of grave conduct such as spying. A public accusation can damage a person's career and reputation, and under the "no reasons" rule that harm can be inflicted without any obligation to prove the claim. Defenders respond that the diplomat is not a defendant in a criminal case and that the sending state, not the individual, is the proper party to any dispute. They add that requiring reasons would be worse, because it would force hosts to reveal intelligence and would turn every expulsion into a public trial with no neutral judge. On this view, the lack of individual #due_process is a deliberate price paid for a workable system, and the sending state's own diplomatic channels are the correct place to raise any grievance. The debate here is genuine, and reasonable people land on different sides of it. The second criticism concerns scale and escalation. When the tool moves from single removals to mass, coordinated expulsions, it starts to resemble a form of collective pressure that can spiral. Each side feels bound to match the other, and rounds of retaliation can gut the staffing of embassies on both sides. Reduced missions mean fewer channels for communication at exactly the moments when communication matters most, during crises and confrontations. Critics warn that treating expulsion as a routine weapon can therefore undermine the very purpose of diplomacy, which is to keep states talking even when they are angry. There is a paradox in using a diplomatic mechanism in ways that weaken diplomacy itself. The third criticism links persona non grata to the wider problem of abuse of #diplomatic_immunity. Some scholars argue that the immunity system as a whole has grown too generous, shielding serious wrongdoing from any accountability, and that expulsion is a weak substitute for justice. When a diplomat commits a grave offence and the sending state refuses to waive immunity, the victim may be left with no remedy at all while the offender simply goes home. Writers in this vein call for reforms that would strike a better balance between privilege and accountability, for example by narrowing immunity for the gravest crimes or by making waiver harder to refuse (Butt, 2024; Akani, 2024; Abusamra, 2024). Persona non grata, on this account, is part of a system that too often ends in impunity rather than responsibility. A fourth strand of debate asks whether human rights norms should ever override diplomatic protections. Some commentators suggest that when a diplomat is implicated in torture, trafficking, or slavery, the host should be free to act more forcefully than mere expulsion. Others reply that opening such exceptions would destabilise the whole regime and expose diplomats everywhere to politically motivated prosecutions. This tension between immunity and higher norms remains unresolved, and it forms part of the living argument around the field. It is fair to note the limits of these criticisms as well. The "no reasons" rule does not authorise mistreatment of the diplomat, who remains inviolable. It does not create a criminal record or a legal finding of guilt. And it operates within a culture of #reciprocity that gives every state an incentive not to abuse the tool too freely, since today's aggressor may be tomorrow's target. These built in restraints do not answer every objection, but they show that the rule is not as lawless as its harshest critics imply. Balancing Sovereignty and Stability: An Analysis Having set out the rule, its history, its logic, its practice, and its critics, we can now weigh the whole picture. The central question is whether the freedom to expel without giving reasons is, on balance, a wise design. The answer offered here is a qualified yes. The strongest argument in favour of the rule is that it makes the diplomatic system stable and self correcting. Diplomacy depends on strong immunity, and strong immunity creates the risk of abuse. A system that granted immunity but gave the host no remedy would be unbalanced and would eventually collapse, because hosts would not tolerate protected wrongdoers with no way to remove them. #Persona_non_grata supplies the missing remedy in a clean, predictable form. It lets the host protect itself without breaking immunity, without seizing embassies, and without resorting to force. In this sense the rule is what keeps the immunity bargain acceptable to both sides. The "no reasons" element, though troubling on fairness grounds, serves real and defensible ends. It protects the secret sources on which national security often depends. It spares relationships the poison of endless public argument over contested facts. It respects the deep principle that a state need not justify to a foreign power its choice of whom to host as a representative. These are not trivial values, and any reform that stripped the rule of its confidentiality would carry heavy costs of its own. At the same time, the analysis cannot ignore the dangers. The very qualities that make the tool useful, its silence, its speed, and its freedom from review, are the qualities that make it easy to abuse. When expulsions are used not against genuine wrongdoers but as blunt instruments of political anger, and especially when they are used in escalating mass waves, they can hollow out the channels of communication that prevent conflicts from spinning out of control. The instrument designed to protect diplomacy can, if overused, weaken it. The way to hold these truths together is to see the rule as sound in design but demanding in use. The law rightly gives hosts a strong, unexplained power of removal. The responsibility for restraint therefore falls not on the text of Article 9 but on the judgment of states. Good practice means reserving the tool for real cases of misconduct and interference, using the soft formula of "activities incompatible with diplomatic status" to allow both sides to save face, keeping deadlines reasonable, and resisting the temptation to let every dispute become a numbers game of matched expulsions. The health of the system depends less on changing the rule than on using it wisely. For students of #international_law, the deeper lesson is about the relationship between law and power. Article 9 is a small piece of text, yet it channels one of the sharpest edges of state power into an orderly form. It does not abolish the underlying reality that states will remove representatives they distrust; it shapes and civilises that reality. It replaces the risk of violent quarrels with a predictable procedure, and it substitutes a quiet notice for an open rupture. That is what good international law often does. It does not erase power, but it gives power a form that other states can live with. Conclusion Persona non grata is one of the oldest and most durable ideas in the law of nations, and one of the most quietly powerful. It rests on a simple truth: no state can be forced to keep a foreign representative it no longer trusts. The #Vienna_Convention on Diplomatic Relations of 1961 took that ancient custom and gave it a modern, universal form in Article 9, a short provision that lets a #receiving_state remove a foreign #diplomatic_agent at any time and without any duty to explain. This article has argued that the "no reasons" rule is a deliberate and largely sound design. It protects #sovereignty by keeping the choice of whom to host with the host. It protects national security by allowing states to act on secret intelligence without exposing it. It protects stability by keeping expulsions out of courtrooms and out of endless public disputes. And it supplies the essential remedy that makes the whole system of #diplomatic_immunity workable, since a host that cannot arrest or try a diplomat can at least send the person home. The same article has also shown the darker side of the tool. Freed from any duty to justify itself, the power to expel can be turned into a #diplomatic_weapon, used for political signalling, for reputational attack, and for escalating rounds of retaliation that shrink the very channels diplomacy needs. The mass removals of recent years, from the coordinated response to the Salisbury poisoning to the sweeping expulsions that followed the invasion of Ukraine, show both the reach of the instrument and the risk that its overuse will corrode the practice it was meant to serve. The right conclusion is not to weaken the rule but to understand it, and to insist on its responsible use. The text of Article 9 is unlikely to change, and there are good reasons for that. What can change is the wisdom with which states wield it. For the reader who encounters the next expulsion headline, the task is to look past the drama and ask the legal questions: Was this a genuine case of misconduct or a political signal? Was the deadline reasonable? Will it trigger a matching response? Behind every short, cold sentence about an unwelcome person lies this rich and demanding body of #international_law, and learning to read it is the first step toward using it well. References Abusamra, L. (2024). Mechanisms that could be used to prevent diplomats from abusing the immunities and privileges of their authority. Z Problematyki Prawa Pracy i Polityki Socjalnej, 1 to 17. Ahmad, N. (2021). Diplomatic immunities from the perspective of the receiving state and the balance with local law. In studies on the Vienna Convention regime. (Discussed alongside contemporary practice.) Ahmad, N., Lilienthal, G., and Asmad, A. B. H. (2024). Abuse of diplomatic immunities and its consequences under the Vienna Convention: A critical study. Transnational Law and Contemporary Problems, 33(1), 147 to 178. Akani, N. (2024). A critical analysis of diplomatic immunity in international relations: Myth or reality? Journal of International Trade Law and Policy. Butt, J. S. (2024). The abuse of diplomatic immunity: Examining cases and implications for international relations, a research. Acta Universitatis Danubius. Relationes Internationales, 17(2), 42 to 79. Denza, E. (2025). Diplomatic law: Commentary on the Vienna Convention on Diplomatic Relations (5th ed.). Oxford University Press. https://doi.org/10.1093/law/9780198974857.001.0001 Moritani, Y., and Akiyama, H. (2023). Securitisation behind persona non grata: Implications to the theory and the cases regarding the Russian invasion of Ukraine in 2022. F1000Research, 12, 269. https://doi.org/10.12688/f1000research.129876.1 Nagieva, A. A., and Samadova, A. E. (2023). The history of the formation and development of the institution of diplomatic immunities and privileges. Law Enforcement Review. Nurhartanto, G. S. (2021). Diplomatic immunities from the perspective of criminal, civil, and administrative jurisdictions of the receiving state. Tanjungpura Law Journal. Primary legal instruments Vienna Convention on Diplomatic Relations, adopted 18 April 1961, entered into force 24 April 1964, United Nations Treaty Series, vol. 500, p. 95. Vienna Convention on Consular Relations, adopted 24 April 1963, entered into force 19 March 1967, United Nations Treaty Series, vol. 596, p. 261. Selected cases referred to in the text United States Diplomatic and Consular Staff in Tehran (United States of America v. Iran), International Court of Justice, Judgment of 24 May 1980. Basfar v. Wong, United Kingdom Supreme Court, 2022. hashtags #Persona_Non_Grata #The_Diplomatic_Weapon #Legal_Contours_Of_Persona_Non_Grata #Vienna_Convention_On_Diplomatic_Relations #Article_9_VCDR #Diplomatic_Immunity #Diplomatic_Expulsion #Sovereign_Right #Host_State_Powers #Expelling_Foreign_Diplomats #Without_Legal_Explanation #Coercive_Diplomacy #International_Law_For_Students #Diplomatic_Law_Explained #STULIB_Research

This article examines one of the clearer host-state victories in modern investment arbitration: Adel A Hamadi Al Tamimi v. Sultanate of Oman, decided by an ICSID tribunal in 2015. The dispute arose from a limestone quarrying project in the Jebel Wasa mountains, run by a United States investor through two companies that leased land from an Omani state-owned enterprise. When the investor repeatedly broke local environmental law and continued operating without valid permits, Oman terminated the leases, imposed fines, and eventually had the investor arrested. The investor then brought a treaty claim under the United States and Oman Free Trade Agreement, alleging expropriation, a breach of the minimum standard of treatment, and a breach of national treatment, and asking for several hundred million dollars in compensation. The tribunal dismissed every claim and ordered the investor to pay most of Oman's legal costs. The decision matters because the tribunal treated the good-faith enforcement of environmental protection rules as a legitimate exercise of regulatory sovereignty rather than as a wrong against the investor. This article sets out the facts, the legal framework, and the tribunal's reasoning in simple terms, then explains why the award became a reference point for the right to regulate, the police powers doctrine, and the wider debate over how far a host state can go in protecting its environment without owing money to foreign investors. It closes with practical lessons for students, states, and investors. Keywords Investment treaty arbitration; ICSID; environmental protection; right to regulate; minimum standard of treatment; expropriation; Oman; police powers doctrine; foreign investment; sustainable development. 1. Introduction International investment law is often described as a system that protects foreign investors against unfair treatment by the states that host them. That description is accurate, but it is only half of the story. States that sign investment treaties do not sign away the power to govern their own territory. They keep the power to make laws, to protect public health, and to protect the natural environment. The hard question in many disputes is where the line sits. When does a genuine act of regulation cross over and become a wrong that the state must pay for, and when does an investor simply have to accept the consequences of breaking the rules that everyone else must follow? Adel A Hamadi Al Tamimi v. Sultanate of Oman is a useful case for students because it answers that question in a way that is easy to follow. The facts are not complicated. An investor was given a licence to quarry limestone within a defined area, on defined conditions, in a mountain region of Oman. He went beyond what the licence allowed, took material he was not permitted to take, put up buildings without approval, cleared trees, and kept working after his lease had ended. The Omani ministries and the state-owned company that had leased him the land responded with warnings, fines, termination, and finally the involvement of the police. The investor treated all of this as harassment and as an attack on his foreign investment. The tribunal did not agree. It found that Oman had done nothing more than apply and enforce its own environmental protection laws in good faith, and that doing so did not breach the treaty. The value of the case is not only in its result. It is in the language the tribunal used. The tribunal said plainly that it would be slow to second-guess a government that is enforcing environmental rules, and that only serious and shocking conduct by the state would count as a breach of the minimum standard of treatment. That statement has been quoted many times since, and it feeds directly into the modern argument that states have a protected policy space to regulate for public welfare. This article walks through the whole dispute, then places it inside the larger conversation about environmental protection and regulatory sovereignty in investment arbitration. 2. The parties and the investment The investor, Adel A Hamadi Al Tamimi, was a citizen of the United States. He had been born in the region that later became the United Arab Emirates, but he had become a naturalised American, and the tribunal treated him as a national of the United States for the purposes of the treaty. His American nationality mattered, because the treaty he relied on was the Free Trade Agreement between the United States and Oman, which only protects investors of one party investing in the territory of the other. The #host_state was the Sultanate of Oman. The Omani side of the arrangement was led by the Oman Mining Company LLC, a state-owned enterprise usually shortened to OMCO. OMCO was looking for operators who could supply large quantities of limestone, partly to meet demand from the nearby construction market in the Gulf. The investor set up two companies to carry out the project. The first was Emrock Aggregate and Mining LLC, which was properly registered in Oman. The second was a company usually referred to as SFOH, which was not properly registered in Oman and did not hold the business licences it needed there. Each of these companies signed a separate lease agreement with OMCO in 2006 for the quarrying of limestone in the Jebel Wasa area, in the municipality of Mahda, close to the border with the United Arab Emirates. This structure matters for the rest of the story. Only one of the two companies, Emrock, was a properly registered Omani entity. That single fact later helped decide which parts of the dispute the tribunal could even hear. 3. How the relationship broke down The project began with permits. After the leases were signed, the Omani Ministry of Finance and the Ministry responsible for the environment issued the initial permissions that the investor needed. Both ministries told the investor, in clear terms, to stay inside the boundaries of the licensed quarrying area and to limit his activity to extracting limestone. The investor started operating in the Jebel Wasa mountains in September 2007. The relationship soured quickly. The environmental authority issued a series of complaints, warnings, and fines against the investor's companies and against OMCO. The reasons given form a pattern of #environmental_law breaches. The investor was accused of taking material, including sand and stone, from outside the area he was licensed to work. He was accused of running heavy machinery without the necessary permits. He was accused of building housing on the site without approval. He was accused of uprooting trees. On the record before the tribunal, the investor did not correct course in response to these complaints. The situation moved from warnings to termination. OMCO decided to end the lease with Emrock, and it treated the lease with the unregistered company, SFOH, as null and void because that company had never registered or obtained licences in Oman. A letter in July 2008 told the investor about these decisions. A second letter in February 2009 confirmed that the Emrock lease was terminated. Even after all of this, the investor kept operating at the quarry. The ministries and OMCO issued further warnings. In the end, at the request of the environmental authority, the Royal Oman Police arrested the investor in connection with his continued and now unauthorised presence at the site. He was released after giving an undertaking that quarry operations would stop. The dispute also had a criminal side inside Oman. In his role as chairman of Emrock, the investor was tried and convicted by an Omani court in November 2009 on two misdemeanour counts. The first was taking sand and stone without a permit. The second was violating Omani #environmental_law by carrying out quarrying and crushing without the required permissions. He was fined and given a short prison sentence that was suspended. These convictions were not the subject of the investment claim, but they show that Oman's complaints were not invented for the arbitration. They had already been tested and upheld in Oman's own courts. 4. The legal framework: the treaty and the ICSID system To understand the claim, students need to know two instruments. The first is the Free Trade Agreement between the United States and Oman, usually called the US and Oman FTA, which entered into force on 1 January 2009. Chapter 10 of that agreement contains the investment protections. The provisions that mattered in this case were the guarantee of #national_treatment, the #minimum_standard_of_treatment tied to #customary_international_law, and the rule against #expropriation without compensation. The chapter also contained an environmental clause, modelled on the older North American Free Trade Agreement, which made clear that nothing in the investment chapter should stop a party from adopting or enforcing measures to make sure that investment activity is carried out in a way that is sensitive to environmental concerns. The second instrument is the Convention that created the International Centre for Settlement of Investment Disputes, known as #ICSID, which has been in force since 1966. ICSID is part of the World Bank group and provides a forum where an investor of one country can bring a claim directly against another country, without needing that investor's own government to take up the case. The investor filed his request for arbitration at the end of 2011, and it was registered by ICSID in December of that year. The tribunal that heard the case had three members. The president was David A. R. Williams of New Zealand, appointed by agreement of the parties. The investor appointed Charles N. Brower of the United States. Oman appointed J. Christopher Thomas of Canada. All three are well known figures in #investment_arbitration, which added weight to the outcome. 5. The claims and the amount sought The investor advanced three main claims under Chapter 10 of the treaty. The first was that Oman had expropriated his investment, in the indirect and creeping sense, by taking a series of steps that together destroyed the value of what he had built. The centre of this #expropriation argument was the termination of the two leases, which had given him the right to work in the Jebel Wasa area in the first place. The second claim was that Oman had failed to treat his investment in line with the #minimum_standard_of_treatment required by the treaty, which the parties agreed was tied to the standard owed under #customary_international_law. This is the standard that many people know as #fair_and_equitable_treatment. The third claim was that Oman had breached #national_treatment by treating him worse than it treated comparable domestic operators. The money at stake was large. The investor put the value of his claims in the hundreds of millions of United States dollars, and the compensation he sought has been reported at around 560 million dollars. Oman asked the tribunal to dismiss everything and to order the investor to pay Oman's costs. The size of the claim is worth remembering, because the gap between what was demanded and what was awarded, which was nothing, is part of why the case drew so much attention. 6. Jurisdiction ratione temporis: the timing problem Before the tribunal could look at whether Oman had done anything wrong, it had to decide whether it had the power to hear the claims at all. This is the question of #jurisdiction #ratione_temporis, which is a Latin phrase that simply means jurisdiction with respect to time. The problem was that the US and Oman FTA only came into force on 1 January 2009, but the two leases had been signed back in 2006. A treaty generally does not apply to events that happened before it existed. So the tribunal had to work out which of Oman's actions fell inside the treaty's time window and which fell outside it. The tribunal handled the two leases differently. For the Emrock lease, it looked closely at the two termination letters. It decided that the first letter, in July 2008, had not actually ended the lease, because OMCO and the Omani authorities kept dealing with Emrock about the lease after that date, as if it were still alive. The tribunal treated the later letter, in February 2009, as the one that truly ended the lease. Because February 2009 was after the treaty came into force, the events surrounding the Emrock lease fell within the treaty's reach, and the tribunal could examine them. For the SFOH lease, the tribunal reached the opposite result. It found that OMCO's decision to treat that lease as null took effect in July 2008, before the treaty came into force. It also found that OMCO was entitled to treat the SFOH lease as void, because SFOH had never registered or obtained the licences it needed in Oman. So the tribunal declined #jurisdiction over the SFOH lease. This is a small lesson on its own. An investor who does not follow the basic requirements of local company law can find that the parts of its investment built on that failure fall outside treaty protection entirely. 7. The expropriation claim and why it failed The heart of the #expropriation claim was the loss of the right to work in the Jebel Wasa area. The investor argued that the termination of the leases was unlawful and that it stripped him of his investment. The tribunal did not accept this framing, and its reasoning is instructive. The tribunal drew a line between two very different things. One is a sovereign act, where the state uses its public power to take or destroy an investment. The other is a private contract dispute, where one commercial party ends an agreement with another. The tribunal viewed the termination of the leases as belonging to the second category. In its view, the investor lost his investment because his contract with OMCO came to an end, and that was a matter of ordinary contract law, not a matter of #public_international_law. A state does not commit #expropriation every time a company it owns ends a commercial contract on ordinary commercial grounds. The tribunal added a further point that closed the door on the rest of the claim. Once the leases had ended in February 2009, the investor no longer held any property right to work in the area. Actions taken by Oman after that point, including the arrest and the removal of the investor from the site, could not have taken away a property right that had already ceased to exist. In simple terms, you cannot expropriate something that the claimant no longer owns. For these reasons the tribunal dismissed the #expropriation claim in full. 8. The minimum standard of treatment: the core of the case The most important part of the award, and the part that later scholars return to, deals with the #minimum_standard_of_treatment. The treaty tied this standard to #customary_international_law, so the tribunal had to explain what that standard requires. It looked at earlier decisions, many of them from the North American context, such as the well-known cases involving Canada and Mexico, which had already tried to describe the content of the standard. The tribunal set the bar high. It said that to prove a breach, the investor had to show that Oman had acted with a gross or flagrant disregard for the basic principles of fairness, consistency, even-handedness, #due_process, or natural justice that all states are expected to respect. This is a demanding test. It is not enough for an investor to point to some inconsistency, some delay, or some imperfect handling of paperwork. The state's conduct has to be seriously and obviously wrong before it crosses the line. The tribunal then said something that has become the signature line of the case. It explained that not every minor misapplication of a state's laws or regulations will meet that high standard. And it went further, saying that this is especially so where the conduct in question concerns the good-faith application or enforcement of a state's laws or regulations relating to the protection of its environment. In other words, when a state is enforcing #environmental_protection rules in good faith, a tribunal should be even more reluctant to find a breach. The tribunal reinforced the point by borrowing a phrase from an earlier decision, saying that it had no open-ended mandate to second-guess government decision-making, particularly in the face of a strong defence of environmental values. Applied to the facts, this reasoning made the outcome almost inevitable. Even if the Omani ministries had been somewhat inconsistent at times in how they handled permits, that fell far short of #manifest_arbitrariness or a complete lack of transparency and candour, which is the kind of conduct that would be needed for a breach. The measures used to make the investor stop, including his arrest, were treated as ordinary consequences of his unlawful presence at the site after the lease had ended, and as steps taken in line with Omani law. The tribunal dismissed the #minimum_standard_of_treatment claim. The tribunal also rejected a separate argument built on #proportionality. The investor had argued that terminating the leases and using the police against him were disproportionate responses. The tribunal held that the #minimum_standard_of_treatment under #customary_international_law does not contain a free-standing requirement of #proportionality that would let a tribunal review every state action for whether it was the least intrusive option available. This is a significant statement, because #proportionality is a favourite tool of investors who want tribunals to weigh the state's aims against the harm to the investment. The tribunal declined to turn that tool into a general obligation. 9. The national treatment claim The third claim, based on #national_treatment, did not take up much space in the award. The investor argued that Oman had treated him worse than comparable domestic quarry operators. The tribunal gave two reasons for rejecting this claim. The first followed from its earlier findings on timing and on the end of the leases. Any action taken after the leases had been terminated could not breach the investor's treaty rights, because there was no longer a protected investment for those rights to attach to. The second reason was evidential. The investor's proof of worse treatment rested mainly on an informal discussion he had with the operator of a neighbouring quarry, and the tribunal did not consider that reliable enough to establish a breach. A #national_treatment claim needs solid evidence that a genuinely comparable domestic operator was treated better in genuinely comparable circumstances, and that evidence was missing here. 10. Environmental protection as a sovereign defence The reason this case is remembered is the way it treated #environmental_protection. It is worth slowing down to see exactly what the tribunal did and did not decide, because the case is sometimes described too loosely. The tribunal did not create a rule that any measure a state labels as environmental is automatically lawful. It did not say that #environmental_protection is a magic word that defeats every claim. What it did was narrower and, for that reason, more durable. It read the treaty as a whole, including the environmental clause in Chapter 10, and concluded that the agreement placed a high value on #environmental_protection and that this value coloured how the investment obligations should be read. It then applied the general #minimum_standard_of_treatment with extra care in a setting where the state was enforcing environmental rules in good faith. The key phrase, good faith, is doing a lot of work. The tribunal was protecting genuine enforcement of real rules, not a pretext dressed up in green language. This approach fits a wider movement in #investment_arbitration and in treaty drafting. Newer investment treaties increasingly contain express clauses that preserve the #right_to_regulate for public welfare, including the environment. Older treaties were often silent on the point, which left tribunals to work out the balance for themselves. In the Al Tamimi case, the tribunal did not need a bespoke exception to reach a sensible result. It used the environmental language that already existed in the treaty, combined with the high threshold of the #minimum_standard_of_treatment, to give Oman room to govern. Seen this way, the case is an example of what commentators call the #police_powers doctrine at work through the back door. The #police_powers doctrine holds that a state's normal, non-discriminatory regulation in pursuit of a genuine public purpose does not require compensation, even if it hurts an investment. The doctrine is usually discussed in the context of #expropriation, where the question is whether a regulation counts as a compensable taking. In Al Tamimi, the same underlying idea appeared through the #minimum_standard_of_treatment and through the tribunal's refusal to second-guess good-faith environmental enforcement. The label was different, but the instinct was the same: a state that is honestly protecting its environment and enforcing its own laws is exercising #regulatory_sovereignty, not committing a wrong. 11. The role of the investor's own conduct One theme runs through every part of the award, and students should not miss it. The investor lost partly because of what Oman did, but mostly because of what the investor himself did. He operated outside his licensed area. He ran machinery without permits. He built without approval. He cleared vegetation. He kept working after being told to stop, and after his lease had ended. He was warned repeatedly and did not change. He was even convicted in Oman's own courts for two of these breaches. This matters for the legal analysis because many treaty protections carry an unwritten assumption that the investor has behaved lawfully and reasonably. The idea of #legitimate_expectations, for example, protects an investor who relied in good faith on the state's conduct. It offers much less to an investor who ignored clear conditions and continued in breach. When an investor comes to a tribunal with a record of repeated law-breaking, the state's response looks less like harassment and more like ordinary enforcement. The Al Tamimi tribunal did not have to strain to find that Oman acted in good faith, because the investor's own conduct made Oman's position easy to accept. This is sometimes described in the wider literature under the heading of investor conduct or clean hands, and it is one of the quiet but powerful lessons of the case. 12. Costs: a real financial consequence The tribunal did not stop at dismissing the claims. It also dealt with #costs, and it did so in a way that had real financial bite. The investor was ordered to bear his own costs and to reimburse Oman for 75 percent of Oman's arbitration costs, legal fees, and related expenses. The tribunal did not order full reimbursement, partly in recognition that Oman had not won every preliminary point, including some of its objections to #jurisdiction. Still, a 75 percent costs award against a losing claimant is a serious outcome. For students, the costs decision is a reminder that #investment_arbitration is expensive and that losing can be very costly. The general trend in these cases has moved somewhat toward the principle that costs follow the event, meaning the losing side pays a large share of the winner's costs. A claimant who brings an inflated claim and loses on every point exposes itself to exactly this risk. The costs order in Al Tamimi sent a message that weak claims against genuine regulation can carry a heavy price. 13. Why the case became a reference point Several features combined to give this award lasting importance in the field of #investment_law. The first is the clarity of the language on #environmental_protection. Tribunals often hedge. This tribunal said, in plain words, that it would be especially slow to condemn a state that is enforcing environmental rules in good faith. That sentence is easy to quote and easy to apply, which is why later scholarship keeps returning to it. The second is the outcome. A complete defence win for a #host_state, with every claim dismissed and most costs shifted to the investor, is less common than a partial result. A clean win makes a stronger precedent for the proposition that #regulatory_sovereignty and #environmental_protection can defeat a treaty claim. The third is the region. Cases involving Middle Eastern governments in #investment_arbitration were relatively rare at the time, and the case gave a concrete example of a Gulf state successfully defending its regulatory choices under a modern free trade agreement. The fourth is timing within the larger debate. The award came as the whole system was moving toward a stronger recognition of the #right_to_regulate. Treaty drafters were beginning to write explicit public welfare exceptions into new agreements, and academics were arguing for a presumption that reasonable, non-discriminatory regulation should not trigger compensation. Al Tamimi fit that mood and gave it support from an actual decided case rather than from theory alone. 14. Placing the case in the wider debate To see the case clearly, it helps to place it beside the broader argument about how #investment_law should treat the environment. For a long time, critics worried that #investment_arbitration tilted too far toward investors, and that it could produce a chilling effect, where states become afraid to pass strong environmental rules because they fear being sued. Some early decisions in the field fed that worry, especially those that focused only on the effect of a measure on the investment and ignored the public purpose behind it. Under that approach, sometimes called the sole effects approach, even a measure taken to protect the environment could look like a compensable taking if it hurt the investment enough. The counter-current, which has grown stronger over the last decade, insists that a state's #right_to_regulate for genuine public welfare, including #environmental_protection, is part of the background against which all treaty obligations must be read. On this view, honest and even-handed regulation is not a wrong at all, and the burden sits with the investor to show something more than lost profits. The Al Tamimi award belongs to this counter-current. It did not use the phrase #police_powers in the way an expropriation case might, but it reached a result that respects the same principle. It gave a state room to enforce its environmental laws without paying an investor who broke them. This does not mean every environmental defence succeeds. Later cases in other regions, including mining disputes in Latin America, show that tribunals still divide on how far a state can go, especially where the state has made specific promises to an investor and then reversed course, or where the treaty contains particular language on exceptions. The lesson from the run of cases, taken together, is that #environmental_protection is a strong shield when the state acts in good faith, applies its rules even-handedly, and does not single out the foreign investor for worse treatment. It is a weaker shield when the state has induced reliance and then changed the rules in a way that looks arbitrary. Al Tamimi is a strong-shield case, because the state was enforcing existing rules against an investor who was plainly in breach. 15. Lessons for host states Several practical lessons flow from the award for governments that want to defend their #regulatory_sovereignty while still attracting #foreign_investment. The first lesson is to write the rules down and apply them consistently. Oman's position was strong because the permits set clear conditions, the ministries issued documented warnings, and the breaches were recorded and even tested in the domestic courts. A paper trail that shows good-faith, even-handed enforcement is one of the most valuable assets a state can have in an arbitration. The second lesson is to keep environmental enforcement separate from any appearance of targeting the investor. The tribunal accepted that Oman was enforcing general rules, not persecuting a foreigner. Where a state can show that domestic and foreign operators face the same rules, the #national_treatment claim weakens and the good-faith finding becomes easier. The third lesson is to use the treaty's own environmental language. The tribunal drew on the environmental clause in the treaty to read the investment obligations in a way that respected #environmental_protection. States that negotiate treaties should include clear #right_to_regulate and environmental provisions, and states that defend claims should point to those provisions rather than relying only on general principles. 16. Lessons for investors The case is just as instructive for investors and their advisers. The first lesson is that treaty protection is not a substitute for following local law. An investor who breaks the #environmental_law of the host country, ignores permit conditions, and continues operating after being told to stop is in a very weak position, no matter how large the claim. The treaty protects lawful, good-faith investment, not conduct that the host state has already found unlawful. The second lesson concerns corporate structure and basic compliance. One of the investor's companies was never properly registered in Oman, and that failure caused an entire branch of the claim to fall outside the tribunal's #jurisdiction. Simple steps, such as registering companies and obtaining the right licences, can determine whether an asset is protected at all. The third lesson is about the cost of overreach. The investor asked for a very large sum, lost on every point, and was ordered to pay most of the state's costs. Bringing an inflated claim against genuine regulation is not free. The risk of a large adverse #costs award is real and should shape how and whether a claim is brought. 17. Some limits and open questions It would be wrong to read Al Tamimi as settling every question about #environmental_protection in #investment_law. A few limits deserve mention so that students form a balanced view. First, the environmental point was tied tightly to good faith. The tribunal protected honest enforcement of real rules. It did not decide what happens when a state uses an environmental label as a cover for something else, such as favouring a domestic competitor. That harder case was not before the tribunal. Second, the case turned in large part on the investor's own conduct and on the timing problem created by the treaty's late entry into force. Those features made Oman's defence unusually strong. A different case, with a compliant investor and a treaty already in force, might test the environmental principle more severely. Third, the tribunal's refusal to recognise a general #proportionality requirement within the #minimum_standard_of_treatment is not accepted by everyone. Other tribunals and many scholars favour some form of #proportionality or reasonableness review, especially where the state has interfered with #legitimate_expectations. The debate over the correct standard of review is still live, and Al Tamimi is one voice in it, not the final word. Fourth, the wider system continues to change. Newer treaties, reform discussions at the international level, and a growing focus on climate policy are all reshaping how tribunals think about the #right_to_regulate. Al Tamimi remains relevant, but it should be read alongside the newer treaty texts and the newer awards rather than on its own. 18. Why students should study this case For a student of international #investment_law, Al Tamimi is close to an ideal teaching case. The facts are concrete and easy to picture: a quarry in the mountains, a licence with clear conditions, an investor who kept breaking them, and a state that responded with warnings, fines, termination, and finally the police. The legal issues cover most of the core topics in the field in a single decision: #jurisdiction and the problem of timing, #expropriation and the difference between sovereign and commercial acts, the #minimum_standard_of_treatment and the meaning of #fair_and_equitable_treatment, #national_treatment, #proportionality, #environmental_protection, and #costs. The case also teaches a habit of mind that is useful well beyond this one dispute. It shows that the result of an arbitration often depends less on grand principles and more on the specific facts: who did what, when, and whether they acted in good faith. The investor's repeated breaches and the state's careful, documented enforcement drove the outcome. Students who learn to read cases this way, by asking what each side actually did rather than only what each side argued, will understand #investment_arbitration far better than those who memorise doctrines in the abstract. 19. Conclusion Adel A Hamadi Al Tamimi v. Sultanate of Oman is a landmark for the simple reason that it shows a #host_state defending its #environmental_protection choices and winning completely. The tribunal found that Oman had done nothing more than apply and enforce its own #environmental_law in good faith against an investor who had broken that law again and again. It set a high bar for the #minimum_standard_of_treatment, it refused to treat the ending of a commercial lease as #expropriation, it declined to read a general #proportionality requirement into #customary_international_law, and it made clear that it would be especially slow to condemn good-faith environmental enforcement. It then dismissed every claim and made the investor pay most of Oman's #costs. The award does not give states unlimited power, and it does not turn the word environment into a shield against all accountability. Its protection is reserved for honest, even-handed regulation. But within that limit, the case stands as a clear statement that #regulatory_sovereignty and #environmental_protection are not obstacles to be overcome by foreign investors. They are legitimate public interests that #investment_law is built to respect. For students trying to understand how the balance between investor protection and the #right_to_regulate is struck in practice, few cases make the point as plainly as this one. References Alschner, W. (2022). Investment Arbitration and State-Driven Reform: New Treaties, Old Outcomes. Oxford University Press. Dolzer, R., Kriebaum, U., and Schreuer, C. (2022). Principles of International Investment Law (3rd ed.). Oxford University Press. Hailes, O. (2025). Environmental Clauses in Investment Arbitration: Deep Roots, Green Shoots and Dead Wood. ICSID Review - Foreign Investment Law Journal, 40(2), 399-440. https://doi.org/10.1093/icsidreview/siaf003 Marisi, F. (2020). Environmental Interests in Investment Arbitration: Challenges and Directions. Kluwer Law International. Titi, C. (2022). The Right to Regulate in International Investment Law (Revisited). International and Comparative Law Research Center. SSRN Working Paper No. 4058447. Vinuales, J. E. (2020). Defence Arguments in Investment Arbitration. In J. E. Vinuales and M. Waibel (Eds.), ICSID Reports, Vol. 18 (pp. 9-108). Cambridge University Press. Vinuales, J. E. (2025). Investment Treaties and Climate Change Policy: A Possible Legal Pathway. ICSID Review - Foreign Investment Law Journal, 40(2), 216-238. https://doi.org/10.1093/icsidreview/siaf010 Waibel, M. (2021). Subject-matter Jurisdiction: The Notion of Investment. In J. E. Vinuales and M. Waibel (Eds.), ICSID Reports, Vol. 19 (pp. 25-82). Cambridge University Press. Legal authorities cited Adel A Hamadi Al Tamimi v. Sultanate of Oman, ICSID Case No. ARB/11/33, Award, 3 November 2015. Agreement between the Government of the United States of America and the Government of the Sultanate of Oman on the Establishment of a Free Trade Area (US and Oman FTA), in force 1 January 2009, Chapter 10. Convention on the Settlement of Investment Disputes between States and Nationals of Other States (ICSID Convention), in force 14 October 1966.

This paper reviews one of the defining #investor_state_dispute cases of the early twenty-first century: Azurix Corp. v. The Argentine Republic, decided by an International Centre for Settlement of Investment Disputes tribunal on 14 July 2006. The dispute grew out of the privatization of regional water and sewerage services in the Province of Buenos Aires during the years surrounding Argentina's severe #economic_crisis. A United States water company, controlled at the time by the energy group Enron, won a thirty-year concession in 1999, ran into conflict with provincial authorities over water quality and tariffs, abandoned the concession, and then sued Argentina under the United States-Argentina #bilateral_investment_treaty. The tribunal found that Argentina had breached its treaty duties of #fair_and_equitable_treatment and #full_protection_and_security, rejected the claim that its conduct amounted to #expropriation, and awarded the investor about US$165 million. An annulment committee later upheld the award in full. The paper explains the facts and the legal reasoning in plain terms, situates the case within the broader wave of arbitrations that Argentina faced after its crisis, and uses it as a window onto three enduring debates: the clash between the #human_right_to_water and treaty protection for foreign investors, the fear of #regulatory_chill, and the pressure for #ISDS_reform. It argues that Azurix remains valuable less for any single legal holding than for exposing the structural strain between private investment protection and public control of essential services. Keywords: international investment law; water privatization; bilateral investment treaties; ICSID arbitration; fair and equitable treatment; human right to water; Argentina; regulatory chill 1. Introduction Few sectors bring the tension between private profit and public need into sharper focus than the supply of drinking water. Water is a commodity that can be sold, metered, and traded, yet it is also a basic condition of life that no household can do without. When governments hand the running of a water system to a private company, and especially to a foreign one, they mix two logics that do not always sit comfortably together. One logic asks whether the investor can earn a fair return. The other asks whether ordinary people can afford to drink, wash, and stay healthy. The case of #Azurix Corp. v. The Argentine Republic sits precisely at that meeting point. This paper examines the case as a study in #water_politics and in the workings of #bilateral_investment_treaty law. The dispute arose when a foreign company took over the water and sewerage service for a large part of the #Province_of_Buenos_Aires in Argentina, encountered serious operational and political problems, walked away from the contract, and then turned to an international #arbitral_tribunal for compensation. The award that followed, issued in 2006, ordered Argentina to pay roughly US$165 million, and a later committee refused to overturn it. For students of international law, economics, and public policy, the case offers a compact but rich example of how treaty commitments made to attract capital can later collide with a government's handling of an essential service during a national emergency. The argument developed here is that Azurix should be read on three levels at once. On the first level, it is a technical dispute about the meaning of treaty standards such as #fair_and_equitable_treatment and the correct way to measure #damages. On the second level, it is a political story about the failure of a specific #water_privatization project and about who should bear the cost when such a project collapses. On the third level, it is a piece of evidence in the wider argument over whether the #investor_state_dispute system strikes the right balance between protecting #foreign_direct_investment and preserving a state's freedom to govern in the #public_interest. A word on scope and method is in order. This paper is a doctrinal and contextual case study rather than an empirical survey. It relies on the published record of the award and the annulment decision, on secondary commentary, and on recent scholarship in international investment law to reconstruct what the tribunal decided and why, and then to place those decisions in their political and intellectual setting. The aim is explanatory rather than advocative. Where the field is genuinely divided, as it is on the meaning of the treaty standards and on the wisdom of the whole arbitration model, the paper sets out the competing positions and the evidence behind them instead of declaring a winner. Readers should treat the case not as a source of settled rules but as a lens through which the fault lines of the system become visible. The paper proceeds as follows. Section 2 sets out the background of Argentina's crisis and its embrace of water privatization. Section 3 explains the legal architecture of investment treaties and the #ICSID system. Section 4 narrates the facts of the dispute. Section 5 analyzes the tribunal's reasoning. Section 6 turns to the #annulment stage. Sections 7 through 10 draw out the broader themes of human rights, regulatory freedom, comparative case law, and reform. Section 11 concludes. Throughout, the aim is to keep the language clear enough for students new to the field while remaining faithful to the record of the case and to recent scholarship. 2. Background: Argentina's Crisis and the Turn to Water Privatization To understand why a company from Houston ended up running water pipes in a Latin American province, one has to start with the economic ideas that swept through the region in the 1990s. After a decade of high inflation and heavy public debt, Argentina adopted a package of market reforms. The centerpiece was the convertibility plan, which pegged the Argentine peso to the United States dollar at a rate of one to one. This peg was meant to kill inflation by removing the government's power to print money freely, and for several years it worked. Prices stabilized and foreign capital flowed in. Alongside the currency peg came a broad program of #privatization. State-owned companies in energy, telecommunications, transport, and water were sold or leased to private operators, many of them foreign. The reasoning was straightforward. Public utilities were seen as inefficient, underfunded, and politically manipulated. Private firms, it was argued, would bring capital, technology, and discipline. Water and sewerage services, long neglected in much of the country, were an obvious target. Large portions of the population lacked reliable connections, and the systems that existed were often old and leaking. Selling concessions to run these systems promised both investment and improved coverage without the government having to raise the money itself. The water sector was handed out in pieces. The best known concession covered the City of Buenos Aires and its surrounding districts, and it went to a consortium led by the French group that operated under the name Aguas Argentinas. Other provinces ran their own tenders. In 1999 the Province of Buenos Aires, a separate jurisdiction from the federal capital, awarded a #concession for two of its service regions to a group led by Azurix Corp. The winning company created a local vehicle, Azurix Buenos Aires S.A., to act as the #concession holder, and paid a very large up-front fee, known in Spanish as the canon, of around 438 million pesos, which at the fixed exchange rate equaled the same figure in dollars. This #canon_payment was, in effect, the price of the right to run the service for thirty years. The design of the tender helps explain what went wrong later. Concessions of this kind were commonly awarded to whichever bidder offered the largest up-front payment, a structure that pushed companies to bid aggressively. A very high canon looked attractive to a cash-strapped provincial government, because it delivered a large sum immediately, but it also loaded the winning operator with a heavy financial burden that could only be recovered over decades through tariffs and efficiency gains. If anything then disrupted those tariffs or those gains, the operator would find itself unable to earn back what it had paid. In hindsight, critics argued that the tender rewarded optimism over prudence, and that a bidder confident of rapid growth in a booming, dollar-pegged economy had every incentive to overpay for a system whose real condition and real revenue potential were uncertain. When the boom turned to bust, the size of the original bet became part of the problem rather than a sign of strength. Azurix itself is central to the story. The company had been carved out of Enron, the American energy giant, and floated on the stock market in 1999 with the ambition of becoming a global water utility. It carried the confidence of a firm that believed water services around the world were ripe for consolidation by well-capitalized operators. That confidence would prove costly. Enron collapsed spectacularly in late 2001 amid an accounting scandal, and Azurix, always dependent on its parent, unwound as a going concern. By the time the arbitration reached its later stages, the corporate group behind the claim had already fallen apart, yet the legal claim survived because the treaty right belonged to the company, not to the fortunes of its founders. The macroeconomic backdrop turned from stable to catastrophic almost exactly as the concession troubles peaked. By 2001 the currency peg had become a straitjacket. The overvalued peso hurt exports, unemployment climbed, capital fled, and the government could no longer service its debts. In late 2001 and early 2002 Argentina defaulted on much of its sovereign debt, abandoned the peg, and converted many dollar obligations into pesos at forced rates, a process known as pesification. Bank accounts were frozen. Street protests brought down the government. The peso lost most of its value against the dollar within months. This #economic_crisis is the shadow that falls across every Argentine investment case of the period, because so many contracts had been written in or linked to dollars, and the collapse of the peg changed the economics of every one of them. Water privatization, then, was launched in a moment of optimism and tested in a moment of disaster. The gap between the promises made when the concessions were signed and the reality of the crisis years created the conditions for dozens of legal claims. Azurix was one of the earliest and, because it concerned water rather than gas or finance, one of the most politically charged. The dispute cannot be separated from this larger picture of a country that opened itself to global capital, tied its currency to the dollar, and then watched the whole arrangement come apart. 3. The Legal Architecture: BITs, the US-Argentina Treaty, and ICSID Investors do not sue states in a vacuum. The right to bring a claim like Azurix's rests on a specific legal structure that spread across the world from the late 1950s onward. The building blocks are #bilateral_investment_treaty agreements, usually shortened to #BIT, and the arbitration institutions that give them teeth. A bilateral investment treaty is an agreement between two countries in which each promises to treat investors from the other country according to certain standards. These treaties typically guarantee protection against uncompensated #expropriation, promise #fair_and_equitable_treatment, promise #full_protection_and_security, forbid arbitrary or discriminatory measures, and allow the free transfer of funds. Crucially, most of them contain a clause that lets a foreign investor bypass the host country's own courts and take a dispute directly to international arbitration. This last feature is what makes the system distinctive. It removes the investor's most sensitive disputes from the national legal system and places them before a private international tribunal, a shift that supporters describe as depoliticization and critics describe as a loss of #sovereignty (Beysulen Angin, 2025). The treaty at the heart of Azurix was the agreement between the United States and Argentina, signed in 1991 and in force from the mid-1990s. Like many such treaties of its generation, it was written in broad language. Its promise of fair and equitable treatment did not spell out exactly what conduct crossed the line, and its protection against measures short of outright seizure left much to interpretation. This vagueness is not an accident. General standards were meant to be flexible enough to cover situations the drafters could not foresee. The price of that flexibility, however, is that the real meaning of the treaty gets worked out case by case, by arbitrators reading words such as fair, equitable, and arbitrary and deciding what they require in a given set of facts (Dolzer, Kriebaum, and Schreuer, 2022). The forum for Azurix was the International Centre for Settlement of Investment Disputes, or #ICSID, an arbitration body linked to the World Bank and created by a 1965 convention. ICSID does not decide cases itself. It administers arbitrations conducted by tribunals of, usually, three arbitrators, one chosen by each side and a president chosen by agreement or by the institution. An ICSID award is meant to be final and binding, enforceable in the courts of member states almost as if it were a local judgment, and it can only be challenged through a narrow internal process called #annulment rather than through ordinary appeal. This design was intended to give investors confidence that a favorable award would actually be paid, and to reassure states that awards could not be endlessly relitigated (Collins, 2023). The system was built on a promise that ran in both directions. States hoped that by binding themselves to protect foreign investors, they would attract more capital and development. Investors hoped that neutral arbitration would shield them from hostile local courts and political interference. Whether the promise has been kept is one of the great debates of the field. Empirical work on the operation of the regime has questioned how far treaties actually increase investment, how consistent the arbitration outcomes are, and whether the process treats states and investors even-handedly (Behn, Fauchald, and Langford, 2022). Argentina became the most-sued country in the world after its crisis, and the sheer volume of claims turned it into a natural laboratory for these questions. It matters, finally, that in Argentina's federal structure the water concession was granted not by the national government but by a province. Provincial officials made the decisions that Azurix later attacked. Yet the treaty bound Argentina as a state, and under international law a country is responsible for the conduct of its provinces and local organs. This is a recurring feature of investment disputes. The level of government that causes the harm and the level of government that pays the bill are often not the same, which can create a sense of unfairness on both sides and complicate the politics of compliance. 4. The Facts of the Dispute The concession began with high hopes and a very large check. In 1999 the Azurix-led group won the tender for two service regions of the Province of Buenos Aires, paid the canon of around 438 million pesos, and set up its local company to operate the system. The service area covered a substantial population across a mix of towns and rural districts, with water and sewerage infrastructure that had suffered years of underinvestment. The concessionaire was expected to improve coverage and quality over time, funded partly by tariffs charged to users and partly by its own capital. Azurix later poured well over a hundred million dollars of additional money into the operation on top of the up-front fee. Trouble arrived quickly, and it arrived through the tap. In early 2000, less than a year into operations, an outbreak of #algae in a reservoir serving the Bahia Blanca area turned the water cloudy, foul-smelling, and alarming to residents. The company and the province blamed each other. Azurix argued that the underlying cause was infrastructure that the province had promised to fix and hand over in good condition but had not, so that the raw water reaching its plants was already compromised. The province and its water regulator treated the episode as a service failure by the operator. The regulator ordered the company not to bill customers for water during the affected period and to bear the costs, a highly visible penalty that framed Azurix in the public mind as a foreign firm that could not deliver clean water. From that point the relationship deteriorated on several fronts at once. The #tariff_regime became a battleground. The concession contract set out a framework for what the company could charge and how those charges could be adjusted, but Azurix complained that the province blocked or delayed adjustments it was entitled to, applied the rules inconsistently, and treated tariff decisions as political questions rather than contractual ones. Public officials made statements that, in the company's telling, encouraged customers not to pay and undermined the concession. The promised transfer of certain completed public works, which were supposed to remedy long-standing problems in the system, did not happen as expected. The company argued that without a stable and honored framework it could not raise the financing needed to carry out its investment plan, and that the province had effectively made the concession unworkable. The financing squeeze deserves emphasis, because it explains how a quality incident in one area could threaten the whole enterprise. A water concession is a capital-hungry business. The operator must borrow heavily against the promise of future tariff revenue to pay for pipes, treatment plants, and connections. Lenders will only advance that money if they believe the revenue stream is secure and the rules will be honored. Once the relationship with the province soured, once tariffs were frozen or contested and officials were seen to side against the operator, the company's ability to raise capital dried up. Without new capital it could not carry out the investment program that was supposed to justify the concession in the first place. In this way a dispute that began over discolored water in one region spiraled into a general crisis of confidence that made the entire project unfinanceable. Each side could point to the other as the cause of that spiral, and both accounts contained some truth: the operator's performance had genuine shortcomings, and the province's conduct genuinely undermined the commercial basis of the deal. The province, for its part, portrayed the company as a poor operator that had overpaid for the concession, failed to meet its service obligations, and looked for someone else to blame. It stressed that the users of an essential public service were entitled to safe water at reasonable prices, and that a regulator has both the right and the duty to intervene when a private operator falls short. In this framing the state was not sabotaging a foreign investor but protecting citizens from the consequences of a bad bargain and a struggling firm. By 2001 the situation had become untenable, and the wider national collapse made everything worse. In the second half of 2001 Azurix gave notice that it was terminating the concession, blaming the province for breaking its commitments. The province rejected that account, treated the company as the party at fault, and moved to take the service back into public hands, eventually placing it with a new provincial entity. The formal end of the concession came in early 2002, at almost exactly the moment the national economy was imploding, the peg was abandoned, and dollar values were being converted into a rapidly depreciating peso. The local operating company later went into insolvency in Argentina. Having lost the concession and its investment, Azurix turned to the treaty. It had already begun ICSID proceedings under the United States-Argentina #BIT, and it pursued a large claim, seeking in the region of half a billion dollars or more in compensation for what it characterized as Argentina's destruction of its investment. The company's core argument was that the province's conduct, taken as a whole, had denied it the treatment the treaty promised and had deprived it of the value of its investment. Argentina's core defense was that this was, at bottom, a commercial dispute about a contract gone wrong, that the company had failed as an operator, that the regulator had acted properly in defense of consumers, and that a breach of contract by a province was not the same thing as a violation of an international treaty. 5. The Tribunal's Reasoning The tribunal that decided the case was chaired by Andres Rigo Sureda, with Marc Lalonde and Daniel Hugo Martins as the two other arbitrators. Before the merits could be reached, Argentina tried to remove the president, arguing that his connections to a law firm active in other investor cases created a conflict of interest and cast doubt on his impartiality. The other two arbitrators rejected the challenge, and the point resurfaced later at the annulment stage. This episode is worth remembering because arbitrator independence and the small, repeat-player world of investment #arbitral_tribunal appointments are among the most cited concerns in the modern critique of the system (Behn, Fauchald, and Langford, 2022). 5.1 Jurisdiction and the shareholder question Argentina argued that Azurix had no standing to bring the claim. The contractual rights under the concession belonged to the local operating company, it said, and Azurix was merely a shareholder, so it could not sue in its own name over harm done to the company's contract. The tribunal disagreed. Under the treaty's broad definition of investment, an indirect shareholding counted as a protected investment, and the harm to the company's value was harm to the shareholder's investment. This reasoning, common in the case law, means that the corporate form does not shield states from claims by the ultimate foreign owners, but it also raises hard questions about who really stands behind a claim and about the risk that the same underlying loss might be pursued through more than one route. 5.2 Fair and equitable treatment The heart of the award was #fair_and_equitable_treatment. Here the tribunal adopted a demanding version of the standard, drawing on the influential formulation from the earlier #Tecmed case. In that reading, fair and equitable treatment requires the host state to act consistently and transparently, to respect the investor's #legitimate_expectations, and to maintain a stable and predictable framework for the investment. Measured against this yardstick, the tribunal concluded that the province's conduct fell short. The politicized handling of tariffs, the public discouragement of payment, the failure to deliver on commitments about infrastructure, and the overall treatment of the concession as a political football rather than a legal arrangement combined to deny the investor the treatment the treaty promised. This holding is important and, in later scholarship, controversial. The Tecmed formulation has been criticized for setting the bar so high that almost any change in a state's regulatory posture could be recast as a breach of the investor's expectations. Critics argue that a government is entitled to adjust its policies, especially in a crisis, and that treating stability as a near-absolute entitlement tilts the balance too far toward the investor and squeezes the state's #right_to_regulate (Beysulen Angin, 2025). Defenders reply that fair and equitable treatment is meant to catch exactly the kind of arbitrary, opaque, and bad-faith conduct that the tribunal found here, and that honoring reasonable expectations is a condition of attracting long-term capital. 5.3 Expropriation The tribunal declined to find #expropriation. Azurix had argued that the province's measures, taken together, were equivalent to a taking of its investment, entitling it to full compensation on that basis. The tribunal applied the usual test, which asks whether the state's interference deprived the investor of the use and value of the investment to such a degree, and with such permanence, that it amounted in substance to a seizure, even without a formal transfer of title. It concluded that the threshold was not met. The company had, in the tribunal's view, retained a degree of control and the interference did not reach the level of a substantial and lasting deprivation of the whole investment in the way the expropriation standard requires. The loss, serious as it was, flowed from breaches of other standards rather than from a taking. This distinction between expropriation and lesser breaches is more than a technicality. Expropriation carries a strong association with compensation at full market value, while breaches of standards such as fair and equitable treatment leave more room for argument about the proper measure of loss. As will be seen, the tribunal's decision to award market-value-style #compensation despite finding no expropriation became one of the sharpest points of contention. 5.4 Full protection and security, and arbitrary measures The tribunal also found that Argentina had failed to provide #full_protection_and_security. Traditionally this standard was understood to concern physical protection, for example a duty to use police power to protect an investor's premises and staff from violence. The Azurix tribunal read it more broadly, treating it as extending to the stability of the legal and commercial environment in which the investment operated. On top of this, the tribunal held that certain provincial measures were arbitrary and impaired the investor's use and enjoyment of its investment, in breach of the treaty's ban on arbitrary or discriminatory conduct. The expansive reading of full protection and security has drawn particular criticism. Recent scholarship argues that stretching a clause originally aimed at physical safety into a general guarantee of a favorable business climate blurs the boundaries between distinct treaty standards, feeds inconsistency across cases, and deepens the #legitimacy_crisis around investment arbitration (Yacoub, 2023). The same author's earlier work on the due diligence content of the clause, and related work on state responsibility for failing to prevent harm, underline how uncertain the true scope of the obligation remains and how much depends on the interpretive choices of individual tribunals (Howse and Yacoub, 2022). 5.5 Umbrella clause and transparency Not every argument went the investor's way on the merits. Azurix pressed a so-called #umbrella_clause theory, which would have converted the province's alleged breaches of the concession contract into breaches of the treaty itself, and it argued that the state's failure to publish certain regulations breached a duty of transparency. The tribunal was more cautious here. Because the concession contract was between the local operating company and the province, rather than between Azurix and the state directly, the tribunal did not accept that the umbrella clause turned those contract disputes into treaty violations in the investor's favor. The transparency concern was folded into the broader analysis of fair and equitable treatment rather than standing as an independent, decisive ground. The result was a mixed picture on the legal theories, even though the investor prevailed overall. 5.6 Damages Having found several breaches but no expropriation, the tribunal turned to #damages, and its approach here is among the most studied parts of the award. It decided to value the investment at its #fair_market_value as of the date the concession effectively came to an end in early 2002, and it applied a modified version of an actual investment method. Under this method the tribunal began from the real amounts the investor had put into the project, the canon plus the later capital contributions, and then adjusted downward where it thought reductions were justified. Although the amounts actually invested exceeded half a billion dollars, the tribunal took only about US$165 million into account, splitting the figure between a valuation of the canon and a portion of the additional investment, and it added interest. Two features of this reasoning provoked debate. First, the tribunal used a fair market value measure even though it had rejected the expropriation claim, effectively borrowing the compensation logic usually associated with takings and applying it to breaches of other standards. Second, the modified actual investment method produced a number well below the total invested, which pleased neither side. Argentina argued that the choice of method and the reasoning behind the figure were flawed and inconsistent. Azurix, having sought far more, recovered a fraction of its claim. The addition of interest, accruing from the valuation date until payment, further increased the state's exposure and reminded both parties that delay in these disputes is never neutral. Time favors whichever side is owed money, and arbitrations of this magnitude can run for years. The award thus illustrates a persistent problem in the field: the measurement of #compensation in non-expropriation cases lacks a settled, predictable formula, and reasonable tribunals can reach very different numbers from similar facts. 6. The Annulment Proceeding An ICSID award cannot be appealed in the ordinary sense. A losing party can only ask an #annulment committee to set it aside on a short list of grounds, such as that the tribunal was improperly constituted, that it manifestly exceeded its powers, that there was a serious departure from a fundamental rule of procedure, or that the award failed to state the reasons on which it was based. Annulment is not a second chance to argue the case on the merits. It is a limited check on the integrity of the process, and committees have generally been reluctant to use it to second-guess a tribunal's substantive conclusions. Argentina applied for annulment in late 2006, and a committee composed of Gavan Griffith, Bola Ajibola, and Michael Hwang was constituted to consider the request. Argentina raised a range of grounds. It revived the complaint about the president's independence, arguing that the tribunal had not been properly constituted. It challenged the tribunal's handling of the applicable law, its jurisdictional findings, and, most substantially, its calculation of #damages. On the damages point Argentina argued, among other things, that it made no sense to apply a fair market value standard after finding no expropriation, and that the tribunal had not adequately explained the link between its findings of liability and the figure it produced. The committee rejected every ground in its 2009 decision. On the damages question it held that the tribunal had given sufficient reasons and had not contradicted itself. Fair market value, the committee reasoned, was not reserved exclusively for expropriation cases and could properly serve as the measure for breaches of other treaty standards. It also accepted that the modified actual investment method, which started from the amounts invested and adjusted them, was a coherent way to reach a market value figure. On the independence challenge and the other procedural and jurisdictional points, the committee found no defect serious enough to justify annulment. Argentina was left responsible for the institution's costs of the proceeding, and the award stood. The annulment outcome carries a lesson beyond the parties. It confirms how narrow the grounds for undoing an award really are. A state that loses at the merits stage and believes the tribunal simply got the law or the economics wrong has almost no avenue to correct that within the system. Supporters see this finality as the price of an enforceable regime that investors can trust. Critics see it as a serious weakness, because there is no appellate body to harmonize the law, correct clear errors, or bring consistency to a field where tribunals often disagree, and this absence is a central theme in current arguments for #ISDS_reform (Franck, 2023). 7. Water as a Human Right Meets Investor Protection The most distinctive feature of the Azurix dispute, and the reason it continues to attract attention, is that it was about water. Water is not just another asset. Access to safe drinking water and sanitation has come to be recognized as closely tied to the rights to life, health, and dignity. In 2002 the United Nations Committee on Economic, Social and Cultural Rights issued an authoritative interpretation treating access to water as an element of existing economic and social rights, and in 2010 the United Nations General Assembly adopted a resolution explicitly recognizing the #human_right_to_water and sanitation. These developments postdate the start of the Azurix concession, but they reflect a shift in thinking that was already underway and that gives the case its lasting resonance. Argentina tried to bring this human dimension into the arbitration. Its experts argued that the investment treaty had to be read alongside the state's human rights obligations, and that where the two pointed in different directions, the protection of consumers and the public interest in water should take precedence over the commercial interest of a private operator (Yacoub, 2023). The claim, in essence, was that a regulator acting to protect people's access to safe and affordable water was performing a public duty that the treaty could not simply override. Azurix responded that the interests of users were already protected within the concession framework and that the human rights argument was being used to excuse the province's mistreatment of the investment. The tribunal did not embrace the human rights framing. Like other water-sector tribunals of the period, it treated the investment protection question and the human rights question as largely separate, and it resolved the case within the four corners of the treaty without giving the #human_right_to_water real weight in its analysis. This reluctance is a recurring pattern. Scholars examining the water disputes have noted that tribunals of that era tended to acknowledge such arguments and then set them aside, declining to explain how a state's human rights duties should shape its treaty obligations (Choudhury, 2022). The effect was to leave states in an awkward position: bound by human rights commitments to guarantee access to water, yet exposed to liability under investment treaties when they acted on those commitments in ways that hurt an operator. This tension is not merely academic. Water services involve people who cannot substitute another supplier, prices that ordinary families must be able to pay, and infrastructure that a government cannot allow to fail. When a private operator underperforms or when a crisis makes contracted prices unbearable, the state faces intense pressure to intervene, and intervention is exactly what triggers investor claims. The deeper question that Azurix raises is whether the ordinary tools of investment law, designed for factories and mines and financial holdings, are the right tools for judging conduct in a sector where the underlying good is a condition of human survival (Violi, 2021). The affordability question sits at the core of this tension and is easy to overlook in a purely legal analysis. A tariff that makes commercial sense for an operator carrying a large debt may be unaffordable for households already struggling through a collapse in the value of their wages and savings. When a currency loses much of its value in a matter of months, prices that were reasonable under the old exchange rate can become impossible for ordinary families almost overnight. A government that lets such prices stand risks cutting people off from water they cannot live without; a government that freezes or lowers them protects consumers but wrecks the operator's finances and, under the prevailing reading of the treaties, exposes itself to liability. There is no costless choice. The privatization model assumed that a stable economy would keep this dilemma from ever becoming acute, and when that assumption failed, the whole arrangement was caught between two duties that pointed in opposite directions. Later cases would begin, tentatively, to take the human rights dimension more seriously, but Azurix stands as an example of the earlier approach that kept the two worlds apart. 8. Regulatory Chill and the Right to Regulate A second theme that Azurix illuminates is the fear of #regulatory_chill. The idea is simple to state. If a government knows that regulating a foreign investor might lead to a multimillion-dollar claim and years of costly arbitration, it may hesitate to regulate at all, or may water down a measure it believes is in the public interest, in order to avoid the risk. The chill, in this account, is not only about awards actually paid. It is about the shadow that the mere possibility of a claim casts over everyday governance (Beysulen Angin, 2025). The Azurix facts show how this concern arises. The province's regulator ordered a foreign operator not to bill for contaminated water and pressed it on service quality and tariffs. From a consumer protection standpoint, these are exactly the kinds of actions one wants a regulator to take. Yet the same actions became the raw material for a treaty claim and, ultimately, a nine-figure award. A different provincial official, contemplating similar intervention in a later concession, might reasonably think twice. That hesitation, if it occurs, is the chill in action, and it points to a real cost of the system that does not show up in any single award. The picture is more complicated than the simple story suggests, and recent empirical work cautions against assuming that chill is uniform or automatic. States are not single-minded actors. Different ministries, agencies, and levels of government weigh investment risk very differently, and in many settings the political and public health pressures to regulate outweigh the distant threat of arbitration. Detailed study of dispute data indicates that regulatory chill is real in some contexts but far from universal, and that its intensity depends heavily on the capacity, wealth, and internal structure of the state in question (Garcia Sanchez, 2024). A wealthy country with strong institutions may shrug off the risk, while a poorer or crisis-stricken government may feel it acutely. This connects to the broader debate about the #right_to_regulate and #policy_space. Investment treaties were written to constrain state conduct on purpose, to reassure investors that the rules would not shift arbitrarily against them. The hard question is where legitimate protection ends and undue interference with democratic governance begins. Critics argue that the standards developed in cases like Azurix, especially the expansive readings of fair and equitable treatment and full protection and security, have pushed the balance too far toward investors and left too little room for states to respond to changing circumstances, public health needs, and social demands (Sornarajah, 2021). Reformers have therefore focused much of their attention on redrafting these standards, adding explicit language that preserves the right to regulate, and building safeguards for essential public services into new treaties. 9. Azurix Within the Wider Wave of Argentine Claims Azurix did not happen in isolation. Argentina's crisis produced a flood of investor claims, and by the mid-2000s it was the most frequently sued state in the ICSID system. Placing Azurix alongside its neighbors clarifies both what was typical about it and what was distinctive. Several of the best-known Argentine cases arose in the energy sector, where investors held gas and electricity assets whose economics were upended by the collapse of the currency peg and the emergency measures that followed. In cases such as those involving major gas transmission and generation companies, tribunals wrestled with Argentina's defense that a genuine national emergency justified its conduct, a defense grounded in the treaty and in the customary law doctrine of necessity. The results were strikingly inconsistent. Some tribunals accepted that the emergency excused or reduced liability for a period; others rejected the defense on similar facts. This inconsistency, more than any single outcome, damaged confidence in the predictability of the system and became a standard exhibit in the case for reform (Franck, 2023). The necessity debate is worth pausing on because it captures the essential difficulty. A country in the middle of a genuine collapse, with banks failing and unrest in the streets, plainly must take emergency action, and most legal systems recognize some allowance for a state acting to preserve itself in a true emergency. Yet the investors argued, not without force, that treaties exist precisely to protect capital against the risk of political and economic turmoil, and that if a state could escape all liability simply by pointing to a crisis, the protection would be worth little at the moment it was most needed. Different tribunals struck the balance between these two truths in different places, sometimes reading the treaty's own emergency clause narrowly, sometimes broadly, sometimes conflating it with the stricter customary law defense and sometimes keeping them apart. The lack of a shared method for handling emergencies left states unable to predict, even roughly, how their crisis measures would be judged. Within the water sector specifically, Azurix belongs to a cluster of disputes. The concession for the City of Buenos Aires and surrounding districts generated its own arbitration, in which non-governmental organizations were, notably, allowed to file submissions as friends of the tribunal precisely because the case touched the #human_right_to_water and the interests of millions of users. A separate dispute concerned a water concession in the province of Tucuman, litigated through a long saga that produced important rulings on the relationship between contract claims and treaty claims and on the standard for annulment. These cases share a common shape with Azurix: an ambitious privatization, a clash over tariffs and service quality, a breakdown of the relationship, and a turn to international arbitration. The comparison with #Urbaser is especially instructive because it shows the direction the law began to move. In a later Argentine water case, a tribunal took the unusual step of engaging seriously with the argument that investors themselves might bear obligations connected to the #human_right_to_water, and it entertained a counterclaim by the state framed in those terms, even though it ultimately did not hold the investor liable on that basis. Whatever the precise outcome, the willingness to treat the human right as a live legal consideration, rather than a distraction to be set aside, marked a shift away from the approach seen in Azurix (Choudhury, 2022). Read together, the Argentine water cases trace an evolution from an early period in which investment protection dominated toward a later period in which tribunals began, cautiously, to make room for public interest and human rights considerations. Beyond Argentina, the water sector produced its own iconic controversies elsewhere in the region. The dispute over a water concession in a Bolivian city, which followed steep price increases and mass protests and led to the reversal of the privatization, became a global symbol of resistance to the commodification of water and fed directly into the political backlash against investment arbitration. Azurix, less dramatic on the streets than that episode, nonetheless belongs to the same family of cases in which the privatization of an essential service, protected by an investment treaty, collided with public anger and state intervention. 10. Reform Debates and the Legacy of Azurix By the time the Azurix annulment decision came down in 2009, the sense that the investment arbitration system faced a #legitimacy_crisis was already widespread, and it has only grown since. The complaints are by now familiar. Awards are inconsistent, so that similar facts yield different results. The standards are vague, so that outcomes depend heavily on which arbitrators are appointed. The same small group of individuals serves as arbitrators, counsel, and experts across many cases, raising concerns about independence and bias. There is no appeal, so errors cannot be corrected and the law cannot be harmonized. Proceedings are expensive, and the costs fall hard on states, especially poorer ones. And the system, critics say, gives foreign investors a powerful tool that ordinary citizens and domestic firms do not have (Behn, Fauchald, and Langford, 2022; Burgert, 2026). Azurix illustrates several of these complaints in one package. The disqualification challenge against the tribunal's president speaks to the independence concern. The expansive readings of fair and equitable treatment and full protection and security speak to the vagueness concern. The contested damages methodology, upheld on annulment despite serious objections, speaks to both the inconsistency concern and the absence of meaningful review. And the underlying subject matter, water for a large population during a national emergency, speaks to the deeper worry that the system was not designed for disputes with such heavy public stakes (Yacoub, 2023). The response has been a broad, still unfinished, project of #ISDS_reform. Governments negotiating new treaties have moved away from the open-ended language of the 1990s. Newer agreements often define fair and equitable treatment more narrowly, tie it to a closed list of prohibited conduct, spell out that non-discriminatory regulation for public welfare objectives is not expropriation, and include explicit statements preserving the #right_to_regulate. Some treaties carve out sensitive sectors or add human rights and sustainable development language. At the multilateral level, a long-running intergovernmental process has debated structural options, ranging from a standing investment court with tenured judges and an appeal mechanism, to an appellate body layered on top of existing arbitration, to a greater role for mediation and for the prevention of disputes before they harden into claims (Beysulen Angin, 2025; Franck, 2023). Transparency and public participation form another strand of the reform effort, and here the water cases were an early testing ground. Older arbitrations were often conducted almost entirely behind closed doors, with pleadings, hearings, and sometimes even the existence of the case kept confidential. When the subject matter is a purely commercial dispute between two firms, secrecy may be defensible. When it is the water supply of millions of people, the argument for secrecy weakens considerably, because the outcome affects the public far beyond the two parties at the table. The decision in the Buenos Aires City water case to accept submissions from community and civil society groups marked a step toward opening the process, and later reforms have built on it, adopting rules that make documents public, open hearings to observers, and allow interested non-parties to be heard in appropriate cases. These changes do not alter the substantive standards, but they respond directly to the complaint that a system deciding questions of public importance should not operate in the dark. Each option carries trade-offs. A permanent court might improve consistency and independence but would take years to build and would require states to agree on its design. An appellate mechanism could correct errors and harmonize the law but would add time and cost. Narrowing the standards protects policy space but may reduce the confidence that attracted investment in the first place. And no reform to the forward-looking treaties changes the thousands of older agreements that remain in force and under which claims can still be brought. Meanwhile some states have gone further and withdrawn from the ICSID system or terminated treaties altogether, though exit brings its own complications, since many treaties keep protecting existing investments for years after termination (Sornarajah, 2021). Where does this leave the legacy of Azurix. The case did not create any of these problems, and it did not single-handedly change the law. Its holdings on fair and equitable treatment and full protection and security have been criticized and, in the newer treaties, partly reversed by drafters who wanted to rein in such expansive interpretations. Its damages reasoning remains a cautionary example of how uncertain #compensation can be in non-expropriation cases. Its treatment of the #human_right_to_water represents an approach that later tribunals have begun to move beyond. Yet precisely because it packs so many of the system's strains into one well-documented dispute over such a fundamental service, Azurix endures as a teaching case. It shows students, in concrete detail, why a set of treaties designed to encourage cross-border investment ended up at the center of a fierce argument about #sovereignty, public services, and the limits of private power over shared resources (Dolzer, Kriebaum, and Schreuer, 2022). 11. Conclusion The story of #Azurix Corp. v. Argentina can be told in a single sentence: a foreign company took over regional water services during a period of reform, the arrangement collapsed amid a water quality scandal and a national economic disaster, and an international tribunal ordered the state to pay about US$165 million for breaching its treaty duties. But the value of the case lies in everything that sentence leaves out. It leaves out the optimism of the privatization era and the depth of the crisis that followed. It leaves out the technical craft of interpreting standards such as #fair_and_equitable_treatment and #full_protection_and_security, and the unresolved puzzle of how to measure #damages when there has been no #expropriation. It leaves out the human beings on the other end of the pipe, whose access to safe and affordable water is not a line item on a balance sheet. Three lessons stand out. First, the design of investment treaties matters enormously, because vague standards written to attract capital can later be read in ways that constrain a government's ability to protect its citizens, and the meaning of those standards is settled not by legislatures but by tribunals deciding cases like this one. Second, essential public services such as water strain the ordinary categories of investment law, because the interests at stake are not merely commercial, and the failure of early tribunals to grapple with the #human_right_to_water helped fuel the backlash that reform now tries to answer. Third, finality without appeal is a double-edged feature: it gives investors the certainty that awards will be honored, but it also means that contested reasoning, once upheld on the narrow grounds of #annulment, becomes part of the law with no mechanism to correct it. For students entering this field, Azurix is worth studying not because it settled anything but because it exposed how much remained unsettled. The pressures it revealed, between #investor_state_dispute protection and the #right_to_regulate, between global capital and local control, between the price of water and the right to it, have not gone away. If anything they have intensified, as climate stress and scarcity make control over water even more contested. The fallout of Azurix, in that sense, is still settling, and the questions it raised about #water_politics and #bilateral_investment_treaty law will occupy courts, negotiators, and scholars for years to come. References Behn, D., Fauchald, O. K., and Langford, M. (Eds.). (2022). The Legitimacy of Investment Arbitration: Empirical Perspectives. Cambridge University Press. Beysulen Angin, B. (2025). The right to regulate vs investment protection: Unveiling the causes of imbalance and the limits of current reform efforts in international investment law. ICSID Review - Foreign Investment Law Journal, 40(1), 11-41. https://doi.org/10.1093/icsidreview/siaf007 Burgert, G. J. (2026). A system in "crisis"?: An outlook on investor-state dispute settlement. Houston Law Review, 63(3), 723. Choudhury, B. (2022). Human rights provisions in international investment treaties and investor-state contracts. In J. Scheu et al. (Eds.), Investment Protection, Human Rights, and International Arbitration in Extraordinary Times. Nomos. Collins, D. (2023). An Introduction to International Investment Law (2nd ed.). Cambridge University Press. Dolzer, R., Kriebaum, U., and Schreuer, C. (2022). Principles of International Investment Law (3rd ed.). Oxford University Press. https://doi.org/10.1093/law/9780192857804.001.0001 Franck, S. D. (2023). Reforming World Bank dispute resolution: ICSID in context. Michigan Journal of International Law, 44(3), 349. Garcia Sanchez, G. J. (2024). Defrosting regulatory chill. University of Pennsylvania Journal of International Law, 45(3), 597-680. https://doi.org/10.58112/jil.45-3.2 Howse, R., and Yacoub, A. R. (2022). Litigating terror in the Sinai after the Egyptian spring revolution: Should states be liable to foreign investors for failure to prevent terrorist attacks? Michigan Journal of International Law, 43(3), 595. Sornarajah, M. (2021). The International Law on Foreign Investment (5th ed.). Cambridge University Press. Violi, F. (2021). Contracting in land and natural resources: A tale of exclusion. International Journal of Law in Context, 17, 145. Yacoub, A. R. (2023). Bridging the gap between international investment law and human rights. Hofstra Law Review, 52(1), 65. #Azurix_v_Argentina #water_politics #bilateral_investment_treaties #ICSID_arbitration #investor_state_dispute_settlement #fair_and_equitable_treatment #human_right_to_water #water_privatization #regulatory_chill #right_to_regulate #international_investment_law #ISDS_reform #Province_of_Buenos_Aires #expropriation_law #WaterPolitics #BITs #AzurixCase #InvestmentArbitration

This article studies one of the most discussed decisions in the history of international investment law: the dispute between Occidental Petroleum and the Republic of Ecuador before the International Centre for Settlement of Investment Disputes. The case is famous for two reasons. First, in 2012 the tribunal issued what was then the largest known award in the Centre's history, ordering Ecuador to pay Occidental about USD 1.77 billion, a figure that grew to roughly USD 2.3 billion once interest was added. Second, the reasoning of the tribunal placed the principle of proportionality at the centre of the debate on how far a #host_state may go when it punishes a #foreign_investor who has broken a rule. The state had a genuine legal basis to act, because Occidental transferred part of its rights in an oil project without asking for the approval that the law required. Yet the tribunal decided that cancelling the entire contract and seizing the whole project was too harsh a response, and that this excess turned a lawful sanction into an unlawful taking. The article explains the facts, the legal standards, the quantum analysis, and the later partial annulment that cut the award by forty per cent. It then draws practical lessons for students about the meaning of #expropriation, the limits of state power, the discipline of sovereign damages, and the risk that comes from splitting ownership of a protected asset. The study argues that Occidental is best read not as a victory for either side, but as a warning to both governments and companies about the cost of ignoring proportionality and clean legal structure. Keywords: international investment law; expropriation; proportionality; fair and equitable treatment; ICSID; damages; contributory fault; beneficial ownership; oil and gas arbitration; bilateral investment treaty. 1. Introduction International investment law sits at a difficult crossroads. On one side stands the sovereign right of a country to control its own natural resources, to write its own laws, and to enforce those laws against anyone who breaks them, including large multinational companies. On the other side stands the promise that many countries have made to foreign investors, usually through a treaty, that they will treat those investors fairly and will not take their property without proper compensation. When these two forces collide, the result is often a long and expensive arbitration. The dispute between Occidental Petroleum and Ecuador is one of the clearest examples of that collision. It involved a real breach of the law by the investor, a real and lawful power to punish held by the state, and a final outcome in which the state was nevertheless found responsible and ordered to pay an enormous sum. For students trying to understand how #investor_state_dispute_settlement actually works, the case is close to perfect, because nothing about it is simple. Occidental was not an innocent party. Ecuador was not acting without legal justification. And yet the tribunal, after weighing everything, concluded that the punishment did not fit the fault. The central legal idea in the case is #proportionality. In plain language, proportionality asks whether the harm a government inflicts on an investor is reasonably balanced against the goal the government is trying to reach. A speeding driver can be fined, but the state cannot burn the driver's car to the ground. That intuition, familiar from everyday life and from constitutional law in many countries, was applied by the tribunal to the world of oil contracts and treaty protection. The tribunal held that terminating the whole #Participation_Contract, and taking control of the entire project, was the equivalent of burning the car for a parking violation. This article has several goals. It sets out the facts in enough detail that a reader new to the field can follow them. It explains the legal framework, including what a #bilateral_investment_treaty is, what the Centre does, and what standards such as #fair_and_equitable_treatment and expropriation mean. It walks through the tribunal's reasoning on liability and on the calculation of #sovereign_damages. It then examines the 2015 decision that partially cancelled the award, which is just as important as the award itself. Finally, it collects the lessons that flow from the whole saga, both for states that want to keep their right to regulate and for companies that want to keep their protection. The argument advanced here is that Occidental should be studied as a lesson in restraint. States must learn that a real breach by an investor does not give them unlimited freedom to respond. Investors must learn that treaty protection is not automatic and can be lost through careless legal structuring. And both sides must learn that the size of an award depends heavily on technical questions of valuation and ownership that are easy to overlook until they cost, or save, hundreds of millions of dollars. 2. The Background and the Facts 2.1 How Occidental came to Ecuador Occidental Petroleum Corporation and its subsidiary, Occidental Exploration and Production Company, were United States companies active in the oil business. The parent company was incorporated in Delaware and the operating subsidiary in California. Occidental began working in Ecuador in the mid nineteen eighties under a services contract with the national oil company, then known by its Spanish initials and later succeeded by #Petroecuador. Over time, the commercial relationship changed shape, as often happens when oil prices move and governments rethink how they want to share the profits of their resources. The key agreement for this dispute was signed on 21 May 1999. Under this #Participation_Contract, Occidental was given the right to explore and exploit hydrocarbons in an area of the Ecuadorian Amazon known as #Block_15. A participation contract is a common structure in the oil industry. Rather than being paid a flat fee for its services, the company takes a share of the oil that is produced, which means it carries more risk but can also earn more when the project succeeds. Block 15 became a productive and valuable asset, and it is the loss of that asset that lies at the heart of the case. To understand why the relationship soured, it helps to remember what was happening in the wider region at the time. From the year 2000 onward, the price of oil rose steeply. Contracts that had been signed in the nineteen nineties, when prices were low, suddenly looked far too generous to the foreign companies and far too stingy to the host governments. Across Latin America, states began to demand a larger share of the windfall. Countries such as Venezuela, Bolivia, Argentina, and Ecuador renegotiated terms, changed their tax rules, and in some cases took back control of projects outright. This wave is often described as resource nationalism, the reassertion of state control over natural wealth. When renegotiation failed to produce a deal that both sides could accept, disputes followed, and many of them ended up before the Centre. Occidental v. Ecuador belongs to this larger story. It is not simply a quarrel about one missing signature, but part of a structural conflict between old contracts and new market realities. Keeping this backdrop in mind helps explain why a technical breach was met with such an aggressive response. 2.2 The Farmout Agreement In 2000, Occidental entered into a transaction that would later destroy its position in Ecuador. It signed a Farmout Agreement with a company then called Alberta Energy Company, referred to in the proceedings as AEC. Under this agreement, which was governed by the law of New York, AEC acquired the right to forty per cent of Occidental's share of Block 15's production, together with the obligation to contribute to the costs of the project. The parties also signed a joint operating agreement to put the arrangement into practice. At first sight this looked like a normal commercial deal in which one company brings in a partner to share risk and expense. The problem was legal rather than commercial. Under Ecuadorian law, and specifically under the #Hydrocarbons_Law, a contractor could not transfer or assign its rights under a participation contract, or bring in a partner through a private agreement, without the authorization of the Ministry of Energy and Mines. The Participation Contract itself contained a matching clause requiring ministerial approval for such transfers. Occidental did not obtain that approval before entering the Farmout Agreement. It later communicated with the Ministry and, at various points, sought approval, but the approval it needed was never granted in the form required. This omission, small in appearance but serious in law, gave Ecuador the argument it would use years later to justify the most drastic response available under its legal system. 2.3 The earlier tax dispute Before turning to the termination, it helps to note an earlier clash between the same parties, because it shaped the atmosphere in which later events unfolded. Ecuador's tax authority had, for years, refunded value added tax to oil companies. Around 2001 it changed course, refused future refunds, and even sought to claw back refunds already paid. Occidental treated this as a breach of Ecuadorian tax law and of the treaty and started a separate arbitration. In 2004 a tribunal issued a VAT award in Occidental's favour for about USD 75 million, finding Ecuador's conduct unfair and discriminatory. That result angered parts of the Ecuadorian government and, according to many observers, made a hard line on the #Farmout_Agreement more likely. The two disputes are legally distinct, but the tax victory forms part of the human background to the later termination. 2.4 The Caducidad Decree In August 2004, the Attorney General of Ecuador ordered the Ministry of Energy and Mines to terminate the Participation Contract through a declaration known in Spanish as caducidad. Caducidad is a form of forfeiture. It is the harshest sanction available under the #Hydrocarbons_Law, because it does not merely fine the contractor or suspend its rights. It ends the contract, and it requires the contractor to hand over all of the assets of the project to the state, without compensation. For roughly eighteen months Occidental tried to persuade the authorities not to take this step, arguing that the transfer had not truly harmed the state and that lesser measures would be more appropriate. These efforts failed. On 15 May 2006, the Minister of Energy and Mines issued the Caducidad Decree, formally terminating the Participation Contract and the related agreements. As a result, Occidental was forced to turn over all of the Block 15 assets to Ecuador and received nothing in return. Two days later, on 17 May 2006, Occidental and its parent filed a request for arbitration with the International Centre for Settlement of Investment Disputes. The stage was set for one of the most important expropriation cases of its generation. 3. The Legal Framework 3.1 The treaty The claim rested on the treaty between the United States and Ecuador concerning the encouragement and reciprocal protection of investment. This kind of agreement is usually called a #BIT, short for bilateral investment treaty. The United States and Ecuador signed their treaty in 1993, and it entered into force in 1997. Like most treaties of its type, it promised American investors in Ecuador, and Ecuadorian investors in the United States, a set of protections. Two of these protections mattered most in the case. The first was the protection against uncompensated expropriation, set out in the article of the treaty dealing with the taking of investments. Under this provision, a state may take an investor's property only for a public purpose, in a non discriminatory way, in accordance with due process, and on payment of prompt, adequate, and effective compensation reflecting the fair market value of the investment. A taking that fails any of these conditions is an unlawful expropriation. The second was the guarantee of #fair_and_equitable_treatment, often abbreviated as #FET. This is a broad and somewhat open ended standard. Over the years, tribunals have read it to include protection of the investor's legitimate expectations, protection against arbitrary or grossly unfair conduct, and, in some decisions, a requirement that state measures be proportionate. It is through the FET standard, more than through the expropriation clause, that #proportionality entered the reasoning in Occidental. 3.2 The Centre and the Convention The forum for the dispute was the International Centre for Settlement of Investment Disputes, usually shortened to ICSID. The Centre is part of the World Bank group and was created by an international treaty, the Convention on the Settlement of Investment Disputes between States and Nationals of Other States. Its purpose is to provide a neutral place where foreign investors and host states can resolve their disputes through arbitration, rather than in the national courts of the host state, where investors often fear they will not get a fair hearing. Two features of the ICSID system are worth remembering. First, an ICSID award is meant to be final and directly enforceable in the courts of member states, almost as if it were a judgment of those courts. Second, the only way to attack an ICSID award is through a special internal procedure called #annulment, decided by an ad hoc committee, and available only on a short list of narrow grounds. There is no ordinary appeal on the merits. This design tries to balance finality against a minimum guarantee of procedural integrity, and, as we will see, that balance was tested in Occidental. 3.3 The applicable standards on expropriation International law has long accepted that every state has the right to expropriate property, including the property of foreigners, as an attribute of sovereignty. This is not controversial. What is controversial is the price of that right. Under the modern treaty standard, and under customary international law, an #expropriation is lawful only if it meets the conditions described above and is accompanied by compensation. Where those conditions are not met, the state has committed an internationally wrongful act and owes #full_reparation, a concept discussed in more detail in the section on damages. Expropriation can be direct, where the state openly seizes title to the property, or indirect, where the state's measures destroy the value or control of the investment without a formal transfer of title. The line between a lawful regulation and an #indirect_expropriation is one of the hardest questions in the field. Many tribunals have used a proportionality test at exactly this point, asking whether the burden imposed on the investor is reasonable in relation to the public aim, an approach traceable to the well known reasoning in the Tecmed case against Mexico. Recent scholarship continues to debate how much room this test should leave for the state's #right_to_regulate, and whether tribunals should defer more to democratic choices made by host governments. 4. Jurisdiction: Could the Tribunal Hear the Case at All Before the tribunal could rule on whether Ecuador had done anything wrong, it had to decide whether it had the power to hear the dispute at all. Ecuador raised a clever objection. It argued that the Participation Contract, read together with Ecuadorian law, carved caducidad disputes out of arbitration and reserved them for the Ecuadorian administrative courts. In other words, Ecuador said that the very type of decision at issue, a forfeiture decree, could only be challenged at home, not before an international #arbitral_tribunal. The tribunal rejected this argument. It reasoned that any exception to the availability of ICSID arbitration must be written in clear language. Silence is not enough. If the parties had really wanted to remove #Caducidad_Decree disputes from the reach of the Centre and hand them to the local courts, they could have said so plainly, and they had not. The tribunal also relied on a more fundamental principle. A state cannot use its own domestic law as a shield to escape the jurisdiction it agreed to grant under a #BIT. If that were allowed, every host state could rewrite its internal rules to defeat treaty protection whenever it became inconvenient. This part of the decision may look technical, but it carries a broad message. Treaty protection would be worth little if a state could avoid it simply by declaring, in its own legislation, that certain of its actions are beyond review. The finding on jurisdiction preserved the basic bargain of the treaty, namely that disputes about state conduct affecting protected investments can be taken to a neutral forum. 5. The Merits: Expropriation and the Proportionality Analysis 5.1 Occidental did break the law It is important to be honest about this point, because much of the criticism of the award ignores it, and much of the defence of the award depends on it. The tribunal did not find that Occidental was blameless. On the contrary, it agreed with Ecuador that the #Farmout_Agreement effected a transfer of rights that required ministerial approval, that Occidental never obtained the approval in the required form, and that this was a breach both of the Participation Contract and of the #Hydrocarbons_Law. Occidental had a real legal problem, and the state had a real legal grievance. If the case had stopped there, Ecuador would have won. A state is normally free to enforce its own laws against a company that breaks them. The question the tribunal had to answer was not whether Ecuador could sanction Occidental at all, but whether the particular sanction it chose, the complete forfeiture of a valuable project, was consistent with the state's obligations under the treaty and under international law. 5.2 The proportionality principle Here the tribunal turned to #proportionality. Drawing on Ecuadorian administrative law, which itself recognizes a principle of proportionality, and on the international standard of #fair_and_equitable_treatment, the tribunal held that any administrative goal must be balanced against the investor's own interests and against the true nature and effect of the conduct being punished. A sanction cannot be justified merely because the law permits it in the abstract. It must be reasonable in the concrete circumstances. It helps to break proportionality down into the steps that lawyers, especially those trained in constitutional and human rights law, usually apply. The first step asks whether the state's goal is legitimate. The second asks whether the measure chosen is suitable, meaning capable of achieving that goal. The third asks whether the measure is necessary, meaning that no less restrictive option would have achieved the same goal. The fourth and final step, often called balancing in the strict sense, asks whether the benefit to the public outweighs the burden placed on the private party. A measure can pass the first three steps and still fail the fourth if the harm it causes is simply out of proportion to the good it does. The tribunal made several findings that map onto these steps and that pointed toward disproportion. It accepted that the approval requirement had a rational aim, so the goal was legitimate. But it observed that there were lesser measures available to Ecuador, such as fines or other sanctions short of total forfeiture, which means the necessity step was not satisfied. It said that forfeiture should have been a measure of last resort rather than a first response. It questioned whether the state had actually suffered any real harm from the failure to seek approval, noting that even if the approval could lawfully have been refused, the loss of the chance to review the transfer did not, on the evidence, cause the state any genuine injury. That finding weakened the case at the balancing stage, because the public benefit of the sanction looked thin. Finally, it weighed the aim of deterrence, the idea that harsh sanctions discourage other companies from breaking the rules, against the drastic effect on Occidental, which was the loss of an entire investment worth billions of dollars. The tribunal concluded that the deterrent purpose did not justify a penalty of that magnitude. In the language of the balancing test, the weight imposed on the investor was far greater than the weight of the interest the state was protecting. Putting these findings together, the tribunal held that the #Caducidad_Decree was a disproportionate response to Occidental's breach. Because it was disproportionate, it violated the treaty's guarantee of fair and equitable treatment and it breached Ecuadorian law and customary international law. The application of proportionality, the later annulment committee would say, was one of the cornerstones of the award. 5.3 From proportionality to expropriation Having found a violation of the fair and equitable treatment standard, the tribunal went one step further and held that Ecuador's measures were also tantamount to #expropriation. In effect, the disproportionate seizure of the entire project, without compensation, amounted to an unlawful taking of Occidental's investment. This part of the reasoning has drawn heavy criticism, and it is worth understanding why. Critics argue that the tribunal did not carry out a separate and rigorous analysis of the expropriation clause. Instead, they say, it moved almost automatically from its finding on #FET to its finding on expropriation, treating the second as a natural consequence of the first. Some commentators, including the arbitrator who partly dissented, felt that the tribunal stretched both standards and gave the state too little credit for enforcing its own valid law. Others defend the outcome, arguing that when a state destroys the whole value of an investment through a measure that is itself unlawful because it is disproportionate, calling that measure an expropriation is not a leap but a description. For a student, the key point is that liability rested on the combination of two ideas: the treaty forbids grossly unfair treatment, and it forbids uncompensated takings, and a single disproportionate act can offend both at once. 5.4 The dissent The tribunal was not unanimous on everything. One of its three members, a distinguished professor of international law, wrote a strong separate opinion. Her concerns focused less on the general idea of proportionality and more on two specific points: the treatment of the #Farmout_Agreement, and, above all, the calculation of quantum. In her view, the majority had effectively decided the fate of rights belonging to a company that was not a party to the arbitration, and had awarded Occidental compensation for an interest that Occidental no longer beneficially owned. This dissent proved influential. Several of its themes reappeared, with decisive effect, in the later annulment proceeding. 5.5 Occidental in the company of other cases The proportionality reasoning in Occidental did not appear from nowhere. It drew on a line of decisions that had slowly built the idea into investment law. The most cited ancestor is the Tecmed case against Mexico, in which the tribunal said there must be a reasonable relationship of proportionality between the burden placed on the investor and the aim the state is trying to achieve. Later cases against Argentina, arising from that country's financial crisis, developed related ideas about necessity and public interest. Seen in this light, Occidental is less a bolt from the blue than a high profile confirmation of a trend that had been developing for years. It is also useful to compare Occidental with cases where investor misconduct led to a reduction of damages rather than a denial of liability. In the Yukos arbitrations against Russia, the tribunal found that Russia had acted unlawfully, but it also found that the investors had engaged in abusive tax practices, and it cut the damages by twenty five per cent to reflect their share of responsibility. The parallel with Occidental is close, because in both cases the investor was wronged by the state yet had also behaved badly, and in both cases the tribunal split responsibility rather than treating either side as wholly innocent. This pattern, in which liability and fault are separated and then weighed against each other, is one of the most useful things a student can take from these disputes. Wrongdoing by the state and wrongdoing by the investor are not mutually exclusive, and a modern tribunal will often try to account for both. 6. Quantum: Calculating Sovereign Damages 6.1 The full reparation standard Once a tribunal decides that a state has committed an internationally wrongful act, it must decide how much the state has to pay. The governing principle comes from a famous 1928 decision of the Permanent Court of International Justice in the #Chorzow Factory case. The principle of #full_reparation holds that #compensation must, as far as possible, wipe out all the consequences of the wrongful act and put the injured party in the position it would have occupied if the act had never been committed. In investment cases, this usually translates into paying the #fair_market_value of the lost investment, plus interest. Fair market value is the price a willing buyer would pay a willing seller for the asset, with both acting freely and with good information. Determining that value for a producing oil field is a matter for expert economists and can involve enormous sums and equally enormous uncertainty. The most common method for a project that generates a stream of income is the #discounted_cash_flow method, often abbreviated as #DCF. This method estimates the future cash the project will produce and then reduces those future amounts to a single present value using a discount rate that reflects risk and the time value of money. 6.2 Valuing Block 15 The tribunal valued the full rights to Block 15 under the Participation Contract at about USD 2.36 billion. This figure represented one hundred per cent of the value of the project. From that starting point, two adjustments would eventually bring the number down, one made by the tribunal itself and one made later by the annulment committee. The choice of valuation method and valuation date always attracts argument, because small changes in assumptions produce large changes in the result. To see why the #discounted_cash_flow method is so sensitive, imagine estimating how much oil a field will produce every year for the next fifteen or twenty years, then guessing the price at which that oil will sell in each of those years, then subtracting the cost of getting it out of the ground, and finally deciding how heavily to discount those distant and uncertain profits back to today. Change the assumed oil price by a few dollars, or the discount rate by a single percentage point, and the final value can swing by hundreds of millions. This is why opposing experts in these cases can produce valuations that differ by billions, each defended with charts and models that look equally rigorous. Recent academic work has stressed that tribunals should be careful not to dress up deep uncertainty as ordinary risk, because doing so can inflate awards to levels that bear little relation to what the investor truly lost. Where a project is speculative, some tribunals have preferred to award only the money the investor actually spent, its sunk costs, rather than a projection of profits that may never have come. Where a project is already producing, as Block 15 was, a discounted cash flow figure is easier to justify, but the assumptions still deserve close scrutiny. The date of valuation is a further battleground. Should the loss be measured as of the day the state acted, or as of the day of the award, which may be many years later, after oil prices have moved and after more is known about the field? A leading recent study argues that for an unlawful taking the better approach is to value the loss as of the date of the award, using information reliably available at that time, because this is what truly restores the investor to the position it would have held. The debate over the valuation date, and over how far later information should be taken into account, remains unsettled and is one of the liveliest topics in the current literature on #sovereign_damages. For a student, the lesson is that the phrase fair market value hides a set of difficult choices, and that whoever controls those choices often controls the size of the award. 6.3 Contributory fault The most striking feature of the tribunal's quantum reasoning, and the one most relevant to the theme of proportionality, is its treatment of #contributory_fault. Because Occidental had itself broken the law by failing to seek ministerial approval, the tribunal decided that Occidental had contributed to the loss it suffered. It therefore reduced the damages by twenty five per cent. Contributory fault is a general principle recognized in the law of state responsibility. The idea is fair and familiar. If the injured party's own wrongful or careless conduct helped bring about the harm, it should not recover the full amount, because part of the loss is its own doing. Investment tribunals have applied this idea in several well known cases, sometimes using the same figure of twenty five per cent, as in the Yukos arbitrations against Russia. The percentage is rarely explained with mathematical precision, and tribunals openly acknowledge that they enjoy a wide discretion in fixing it. Scholars have criticized this vagueness, but they generally accept that some reduction for contributory fault is appropriate when the investor's misconduct is real and connected to the harm. In Occidental, the twenty five per cent reduction meant that the starting value of about USD 2.36 billion fell to roughly USD 1.77 billion. This adjusted figure, precisely USD 1,769,625,000, was the principal amount the tribunal ordered Ecuador to pay in its 2012 award. 6.4 Interest Compensation without interest is incomplete, because the investor is deprived of the use of its money for the years between the wrongful act and the payment. The tribunal ordered Ecuador to pay pre award interest at a rate of about 4.19 per cent per year, compounded annually, running from the date of the #Caducidad_Decree in May 2006 until the date of the award in October 2012. It also ordered post award interest from the date of the award until payment. Compound interest over several years on a principal of this size adds a great deal. It is because of this interest that the total exposure was often described as approaching USD 2.3 billion, even though the principal was about USD 1.77 billion. 6.5 The disputed forty per cent The final quantum issue would become the decisive one. Ecuador argued that any award to Occidental should be limited to a sixty per cent interest in #Block_15, because Occidental had transferred forty per cent of its economic interest to AEC under the #Farmout_Agreement, and AEC had later passed that interest on to a Chinese company, Andes Petroleum. In Ecuador's view, Occidental could not be compensated for something it had already given away. The 2012 tribunal rejected this argument and awarded Occidental compensation reflecting the full value of the project, subject only to the twenty five per cent contributory fault reduction. The majority reasoned, in part, that the unauthorized transfer to AEC was itself legally ineffective, so that Occidental remained the party entitled to claim. The dissenting arbitrator disagreed sharply, arguing that this reasoning improperly reached into rights held by a non party and effectively compensated a protected investor for value that truly belonged to an unprotected one. This is the exact point on which the award would later be cut down. 7. The Annulment: When Half the Award Disappeared 7.1 The nature of annulment Ecuador applied to annul the award in October 2012, only days after it was issued. As explained earlier, #annulment is not an appeal. An ad hoc committee cannot simply substitute its own view of the merits for that of the tribunal. It can only annul, in whole or in part, on a few narrow grounds set out in the ICSID Convention. The main grounds are that the tribunal was not properly constituted, that it manifestly exceeded its powers, that there was corruption, that there was a serious departure from a fundamental rule of procedure, or that the award failed to state the reasons on which it was based. The most important ground in Occidental was #manifest_excess_of_powers. 7.2 The committee's decision On 2 November 2015, the annulment committee issued its decision. The committee was composed of three experienced arbitrators and was chaired by a well known Spanish jurist. It rejected almost all of Ecuador's arguments. It upheld the tribunal's finding on #jurisdiction, confirming again that a state cannot rely on its domestic law to escape its treaty commitments. It upheld the tribunal's use of #proportionality as a component of fair and equitable treatment, acknowledging that this was one of the cornerstones of the award and refusing to second guess the tribunal's factual judgment that the sanction was excessive. On the central question of liability, Ecuador lost, and the finding of responsibility survived. But the committee accepted one argument, and it was enough to change the outcome by hundreds of millions of dollars. The committee agreed with Ecuador, and with the earlier dissent, that Occidental had genuinely transferred forty per cent of its economic interest in #Block_15 to AEC, which later held through Andes Petroleum, a company that was not protected under the United States Ecuador treaty. By awarding Occidental compensation for one hundred per cent of the project, the tribunal had compensated a protected investor for an investment that was beneficially owned by an unprotected third party. The committee described this as a #manifest_excess_of_powers, because the tribunal had extended its jurisdiction beyond the persons the treaty was designed to protect. 7.3 The beneficial ownership principle The committee's reasoning turned on the difference between legal ownership and beneficial ownership. Even if Occidental remained the legal title holder on paper, the real economic value of forty per cent of the project had passed to another company through the #Farmout_Agreement. A tribunal established under the United States Ecuador treaty has power only over disputes involving investments protected by that treaty. It cannot, the committee held, compensate a claimant for value that in truth belongs to an investor of a different nationality who must look to its own treaties or to other rules of international law for protection. To hold otherwise would open the door to an uncontrolled expansion of jurisdiction, allowing protected investors to sell hidden interests to unprotected parties while still claiming the full value in arbitration. As a result, the committee reduced the compensation by forty per cent. The award fell from USD 1,769,625,000 to USD 1,061,775,000, plus interest. This was, by a wide margin, the largest amount ever annulled in the history of the Centre, and it was the first annulment there in several years. The committee took care to add that beneficial owners are not left without protection in general. They simply have to bring their claims under the treaties that match their own nationality, which for the Chinese interest meant the treaties available to Chinese investors. 7.4 A note on the scope of annulment The Occidental annulment attracted attention not only for its size but for its method. The committee was willing to look closely at questions that blended fact and law, in order to decide whether the tribunal had overstepped its jurisdiction. Some commentators welcomed this as a healthy check on runaway awards. Others worried that it blurred the line between annulment and appeal, and that it made ICSID awards look less final than the system intends. The debate reflects a deeper tension in the design of #ISDS between finality, which investors value, and correctness, which states demand. For students, the lesson is that even a supposedly final award can be reopened on jurisdictional grounds, and that the way ownership is structured can decide whether an award stands or falls. 8. The Settlement and the Aftermath 8.1 The settlement of January 2016 After the annulment, the reduced award of about USD 1.06 billion still stood against Ecuador, and interest continued to accrue. The two sides had been discussing a possible settlement for some time. In January 2016 they reached one. Occidental agreed to accept a further discount on the amount then owed, reported to be in the region of thirty per cent of that amount, and the parties fixed a net #settlement figure of about USD 979.7 million, with an initial cash payment of USD 100 million and arrangements for the balance. Ecuador's finance minister announced that the state would comply. It is useful to line up the three reductions, because together they tell the whole financial story of the case. The full value of the rights to #Block_15 was assessed at about USD 2.36 billion. The tribunal cut this by twenty five per cent for Occidental's #contributory_fault, producing about USD 1.77 billion. The annulment committee then cut the figure by a further forty per cent because Occidental lacked standing over the interest beneficially owned by Andes, producing about USD 1.06 billion. Finally, in the #settlement, Occidental accepted a further reduction on the amount owed to obtain certain and prompt payment, ending at about USD 980 million. A claim that began in the billions ended at well under half of the project's assessed value. Each reduction has a distinct legal cause, and understanding those causes is a compact lesson in the whole of damages law. 8.2 The downstream dispute over the forty per cent The story did not end there. Because the annulment committee held that the forty per cent belonged beneficially to Andes Petroleum, the relationship between Occidental and Andes over that interest became its own battleground, leading to a separate arbitration between the two companies that produced a further award years later. This aftermath shows how a single decision to bring in a partner without proper approval can generate not one dispute but a chain of them, spanning different forums and different bodies of law, and lasting for the better part of two decades. 8.3 Ecuador and the investment treaty system Occidental also has to be seen against Ecuador's broader and often stormy relationship with the investment treaty system. During the same era, Ecuador grew increasingly hostile to #ISDS, questioned the legitimacy of arbitral tribunals, moved away from the ICSID Convention, and later terminated many of its investment treaties, before once again re engaging with parts of the system in later years. Occidental, with its record breaking award, became a symbol in the political debate inside Ecuador about whether the country should remain in a regime that could impose such large liabilities. Whatever one thinks of that debate, it illustrates how a single high value case can influence a country's entire policy toward foreign investment. 9. Analysis and Lessons 9.1 Proportionality has arrived, but its limits are unclear The most durable contribution of Occidental is its endorsement of #proportionality as a working tool in investment arbitration. The case confirmed that a state cannot hide behind the letter of its own law to justify a crushing response to a modest breach. Even where a sanction is technically authorized, and even where the investor really did do wrong, the state's measure can be tested for reasonableness. This is a significant protection for investors and a significant discipline on states. Yet the very flexibility that makes proportionality attractive also makes it uncertain. There is no fixed formula for deciding when a sanction is too harsh. Tribunals weigh the aim, the availability of lesser measures, the harm caused to the state, and the effect on the investor, but they do so through judgment rather than calculation. Some scholars applaud proportionality as a mature, structured way to balance public and private interests, borrowed from constitutional and human rights law. Others warn that it hands arbitrators a large and loosely defined power to override the choices of sovereign governments, and they call for tribunals to show more deference to the #host_state, especially when the measure serves a genuine public purpose. The right balance between investor protection and the #right_to_regulate remains one of the central unresolved questions of the field, and Occidental sits right at its heart. 9.2 Real fault does not license unlimited punishment A related lesson is aimed at governments. When an investor breaks the rules, the temptation to respond with the most severe measure available is understandable, particularly where public opinion is angry and where the resource at stake is a symbol of national wealth. Occidental teaches that this temptation is dangerous. Ecuador had a valid legal basis to act. It had suffered a technical breach. But by reaching straight for total forfeiture, and by ignoring the many lesser options open to it, the state converted a defensible enforcement action into an internationally wrongful #expropriation that cost it close to a billion dollars even after every reduction. The lesson is not that states may never punish investors. It is that the punishment must be measured, must consider alternatives, and must relate sensibly to the actual harm. 9.3 Structure your investment carefully or lose your protection For investors and their lawyers, the sharpest lesson comes from the #annulment rather than the award. Treaty protection follows nationality. It attaches to investments that are genuinely owned by the protected investor. When Occidental transferred forty per cent of its economic interest to AEC, and that interest moved on to a Chinese company, Occidental put that portion of its investment outside the reach of the United States Ecuador treaty, whether or not it realized what it was doing. The result was that Occidental could not recover the value of the forty per cent from Ecuador in that arbitration at all, because it lacked #standing over an interest it beneficially no longer held. This is a costly reminder that the way an investment is structured is not a mere formality. Beneficial ownership, corporate nationality, and the chain of title decide who can claim, against whom, and under which treaty. A company that reorganizes its holdings, brings in partners, or moves interests between subsidiaries must think about how those changes affect its treaty coverage. Getting this wrong can turn a valuable claim into no claim. 9.4 Damages are where cases are truly won and lost Occidental also shows that the calculation of #sovereign_damages is not a technical footnote but often the main event. Liability was decided in Occidental's favour and stayed that way through the annulment. Yet the amount the company actually recovered was cut, step by step, from the full project value of about USD 2.36 billion to a settlement of about USD 980 million. Each step, the #DCF valuation, the #contributory_fault reduction, the #beneficial_ownership adjustment, and the negotiated discount, moved the number by hundreds of millions. Students who focus only on whether a state is liable, and skip the quantum analysis, miss the part of the case that mattered most to the parties' bank accounts. Modern scholarship on valuation, on the correct date of assessment, and on the treatment of uncertainty is not abstract. It is where real money changes hands. 9.5 Finality is valuable but not absolute Finally, Occidental illustrates the fragile balance between finality and correction in the ISDS system. Investors are drawn to ICSID precisely because its awards are hard to attack. But the annulment machinery exists to catch serious errors, above all errors of jurisdiction. When a tribunal compensates the wrong party, or reaches beyond the persons a treaty protects, annulment is the safety valve. The Occidental committee used that valve to remove forty per cent of the largest award of its time. States can take some comfort that the system can correct grave overreach, while investors are reminded that a favourable award is not truly safe until the annulment window has closed. 9.6 Contract claims and treaty claims are not the same Occidental also teaches a subtle point about the different types of claim an investor can bring. A breach of the #Participation_Contract is, on its own, a matter of contract law, usually to be resolved under the rules the contract itself sets out. A breach of the treaty is a matter of international law, resolved before a treaty tribunal. These two tracks can overlap, especially where the state is also the contracting party, but they are not identical. Part of the controversy in Occidental is that the tribunal moved between contract questions and treaty questions in a way that some critics found too loose, for example when it treated the unauthorized transfer as legally ineffective. Investors should understand which hat they are wearing when they make an argument, because a claim that succeeds as a matter of treaty law may fail as a matter of contract law, and the other way around. Careful pleading keeps the two straight and avoids handing the other side an opening. 9.7 The case reshaped how treaties are written Large awards have consequences far beyond the parties. After Occidental and a handful of other big cases, many states became more cautious about the open ended promises in their older treaties. Newer agreements often define #expropriation more tightly, spell out that genuine public interest regulation is not a taking, limit the reach of the #fair_and_equitable_treatment standard, and describe more carefully how #compensation should be calculated. Some governments withdrew from old treaties or renegotiated them. In this sense, the case is not only a decision about one oil field. It is part of the pressure that has pushed the whole system toward reform, including the long running discussions about how to make #investor_state_dispute_settlement more predictable and more balanced. Students who follow those reform debates will find that cases like Occidental are constantly cited as examples of what went wrong, or of what the system got right, depending on who is speaking. 9.8 Reading the case as a whole None of these lessons stands alone. They interact. Proportionality protected Occidental on liability, but careless structuring cost it on quantum. The state lost on the merits, but a jurisdictional error in the award saved it a fortune. The investor won the principle but lost much of the money. This is why Occidental repays careful study. It refuses to fit the simple narrative of a greedy company against a defenceless state, or of a lawless state against an honest investor. Instead it shows a legal system trying, imperfectly, to hold both sides to account at the same time. 10. Conclusion The dispute between Occidental Petroleum and Ecuador began with a small legal failure, the omission of a required approval, and ended with one of the most important bodies of reasoning in modern investment law. Along the way it produced the largest known ICSID award of its era, a landmark endorsement of #proportionality as a limit on state power, and a record breaking #annulment that reshaped the outcome. For students, three ideas deserve to be carried forward. First, a real breach by an investor does not give a state a free hand. The response must be proportionate, and a disproportionate sanction can transform lawful enforcement into unlawful #expropriation that carries a heavy price. Second, treaty protection is not automatic. It depends on genuine ownership and on the nationality of the true owner, so that careless restructuring can silently strip away the very protection an investor relies on. Third, the calculation of #sovereign_damages is central rather than peripheral. Liability and quantum are separate questions, and a party can win one while losing the other. Read together, these lessons point to a single underlying message about restraint and care. States must temper the exercise of sovereign power with a sense of proportion. Investors must temper their commercial ambition with legal discipline. And tribunals must temper their remedial power with respect for the limits of their own jurisdiction. Occidental v. Ecuador endures as a teaching case because it captures all three of these tensions in one long, expensive, and instructive story. References Baetens, F. (2022). Protecting foreign investment and public health through arbitral balancing and treaty design. International and Comparative Law Quarterly, 71(1), 139 to 181. Beharry, C. L. (Ed.). (2018). Contemporary and emerging issues on the law of damages and valuation in international investment arbitration. Brill Nijhoff. De Brabandere, E., and Merrills, J. G. (2022). Merrills' international dispute settlement (7th ed.). Cambridge University Press. Faccio, S. (2020). Indirect expropriation in international investment law: Between state regulatory powers and investor protection. Editoriale Scientifica. Jarrett, M. (2019). Contributory fault and investor misconduct in investment arbitration. Cambridge University Press. Kriebaum, U., Schreuer, C., and Dolzer, R. (2022). Principles of international investment law (3rd ed.). Oxford University Press. https://doi.org/10.1093/law/9780192857804.001.0001 Marboe, I. (2017). Calculation of compensation and damages in international investment law (2nd ed.). Oxford University Press. Riffel, C. (2022). Indirect expropriation and the protection of public interests. International and Comparative Law Quarterly, 71(4), 945 to 974. https://doi.org/10.1017/S0020589322000343 Sabahi, B., Rubins, N., and Wallace, D. P. (2019). Investor-state arbitration (2nd ed.). Oxford University Press. Titi, C. (2021). The function of equity in international law. Oxford University Press. Ziegler, P., and Gallorini, C. (2023). The case for the ex-post valuation of damages under international investment law. The Journal of World Investment and Trade, 24(1), 115 to 165. https://doi.org/10.1163/22119000-12340286 Topical hashtags #Occidental_v_Ecuador #ICSID #expropriation #proportionality #fair_and_equitable_treatment #sovereign_damages #Caducidad_Decree #contributory_fault #beneficial_ownership #manifest_excess_of_powers #bilateral_investment_treaty #investor_state_dispute_settlement #oil_and_gas_arbitration #international_investment_law #Block_15