Nummer 2020-131
Titel Comment on the Hague Convention of 2 July 2019 on the Recognition and Enforcement of Foreign Judgments in Civil or Commercial Matters. Is the Hague Convention of 2 July 2019 a useful tool for companies who are conducting international activities?
Bron NIPR 2020, afl. 1, p. 19-33
PDF /pdf/2020-131.pdf

Comment on the Hague Convention of 2 July 2019 on the Recognition and Enforcement of Foreign Judgments in Civil or Commercial Matters

Is the Hague Convention of 2 July 2019 a useful tool for companies who are conducting international activities?

Catherine Kessedjian*


The Hague Convention on the Recognition and Enforcement of Foreign Judgments in Civil or Commercial Matters, adopted on 2 July 2019, gives some certainty to worldwide trade relations outside regional systems such as the EU, when disputes are submitted to national courts instead of arbitration or mediation. The Convention avoids the difficult issue of ‘direct’ jurisdictional bases and limits itself to ‘indirect’ jurisdictional bases. This choice of policy was one of the keys to its adoption. Another one was the exclusion of many problematic areas of the law where differences in legal systems are too deep to allow consensus. A third one was to allow States becoming Parties to the Convention to make a number of declarations including some to protect their own acts, which may have been considered as acta jure gestionis under international law. Consequently, the Convention has a fairly narrow scope of application. This may induce more States to become a Party, without which the Convention would not have any more success than the old Hague Convention of 1971 which is still on the books, particularly because it still includes a bilateralisation system, albeit an easier one than that included in the 1971 Convention.

1. Introduction

On the 2nd of July 2019, the States gathered under the auspices of the Hague Conference on Private International Law approved the text of the Convention on the Recognition and Enforcement of Foreign Judgments in Civil or Commercial Matters. This is the achievement of a long saga that dates back to 1992, when the US Government addressed a request to the Secretariat of the organization to start negotiations with the effect of agreeing on a set of rules regulating the same topic. At that time, it seemed plausible and feasible to negotiate what was then called ‘a double convention’ i.e. a convention that would both deal with direct jurisdictional bases1 and the recognition/enforcement of judgments, modeled on what was achieved in 1968 by the European Economic Community (as it was then known) with the Brussels Convention.2

This proved to be an impossible scenario. States were too far apart in their conception of direct jurisdiction. In addition, the changing practices of the market towards a digital economy and the divergences on human rights litigation and intellectual property protection, to name only a few issues, meant that no common ground could be found and negotiations had to be halted.3

The only surviving result of that work was, in a way, the 30 June 2005 Choice of Court Convention which is not the most successful document negotiated under the auspices of the Hague Conference.4 Thereafter, it took another 14 years to adopt a convention, this time limited to the recognition and enforcement of judgments, with no attempt being made to regulate direct jurisdiction. This is the text under review here.5

This comment will focus on three issues: the scope of the Convention (section 2); the principle of recognition and enforcement (section 3); and the obstacles to recognition and enforcement (section 4). The comment will generally be inspired by situations faced by a company conducting international activities and will focus on how the Convention may be useful for such a company.6

2. The scope of the Convention

2.1 Scope ratione materiae

One must realize that the 2019 Hague Convention, when in force, will apply to a very small number of situations, even though the Convention, at face value, applies to the classic category of ‘civil or commercial matters’.7

First, any contract that a company will enter into that contains an arbitration clause falls outside the scope of the Convention (Art. 2(3)). The exclusion of arbitration is drafted in broad terms.8 Therefore, even if the arbitration clause is challenged because it may not be valid or applicable, this matter will have to be treated by the court seized of the case under rules that are outside the Convention9 and the decision rendered by that court will not circulate under the Convention, even if it is a decision on the merits, irrespective of the arbitration clause which would have been considered null and void or inapplicable or inoperative, in order for the court of origin to rule on the merits of the dispute. The situation is different if the court of origin did decide on the merits because the defendant appeared before it and defended the case on the merits without pleading the existence of the arbitration agreement. In such a case, the 1958 New York Convention (Art. II.3) does allow the court to proceed without looking at the arbitration agreement. Therefore, it would be quite detrimental to the parties’ will not to allow that judgment to circulate under the 2019 Convention,10 and nothing in the Convention prevents such an outcome.

Second, if the contract contains an ‘exclusive’ choice of court clause, it falls outside the scope of the 2019 Convention. Indeed, any judgment rendered by a court designated in an ‘exclusive’11 choice of court clause will have to be dealt with under the 2005 Hague Choice of Court Convention. Now, one may ask what happens to such a judgment if the court of origin and the requested court are not parties to the 2005 Convention. In such a case, the 2019 Convention does not exclude the judgment from its scope. However, the only judgments covered by the 2019 text are those rendered by a court which has jurisdiction according to an agreement ‘other than an exclusive choice of court’ agreement.12 Therefore, the 2019 Convention cannot be used in the interim period of time when the 2005 Convention is not applicable. The 2019 Convention should not be considered as a gap filler until the 2005 Convention has become widely applicable. In contrast, a judgment rendered by a court when the defendant agrees to the jurisdiction during the proceedings before the court of origin is covered by the 2019 Convention.13 This situation will be rare in commercial dealings.

Third, in addition, a long list of exclusions has been inserted in Article 2. Apart from the classic exclusions of family law and other equivalent topics, the list also contains the following exclusions, which considerably limit the number of hypothetical situations when a company may resort to the Convention: insolvency, composition, resolution of financial institutions and analogous matters (Art. 2(1)(e)); all matters related to the carriage of passengers and goods regardless of how the carriage takes place (Art. 2(1)(f)); intellectual property (Art. 2(1)(m));14 anti-trust and competition matters, except where the judgment is based on conduct that constitutes an anti-competitive agreement or concerted practice (Art. 2(1)(p)).15

Contracts concluded between a private actor and a State or State entity are not excluded, as such, as long as the State has acted jure gestionis. Having said that, at the request of some States, the 2019 Convention includes a number of provisions to carve out specific transactions which could be considered to fall within the category of acta jure gestionis, but these States did not want them to be included. Such is the case for transactions concerning sovereign debt restructuring by unilateral State measures (Art. 2(1)(q)).16

It has to be noted that consumer and employee disputes are not excluded. So the Convention is somewhat broader than if it only applied to business-to-business (B2B) relations.

A final, ‘hidden’ exclusion concerns interim measures that do not fall within the Convention under Article 3(1)(b). Interim measures are the ‘hot potato’ of PIL. It was a contentious matter in the negotiations leading to the ill-fated draft of 2000. The EU has also had difficulties therewith, even between the tight network of courts in the region and the reform of the Brussels I regime has shown a retreat from the original attempt to give effect to interim measures outside the State of origin. This is another major problem for companies around the world. Very often, a market will be lost if no interim measure can be obtained and enforced; litigating on the merits often being too lengthy and expensive to provide them with true protection. But it seems that the distrust among States around the world prevents giving transborder effects to interim measures. We are at a loss to find a plausible explanation for this distrust and particularly for the lack of a serious attempt to try to narrow the difficulties. The International Law Association (ILA) had tried to do exactly that in its report and resolution adopted in 1996 at Helsinki.17 But that work did not prove to be really influential in the end.

2.2 Reciprocity

The Convention only applies if both the State of origin and the requested State are parties to the Convention (Art. 1(2)). In theory, this is a provision that is understandable because the old concept of reciprocity is still a strong component of international private litigation, as conceived by States. However, in practice, it shows that States do not trust each other and their courts. If they did, it would not matter whether the State of origin is also a party to the Convention because the Convention itself contains the safeguards that are necessary not to enforce a judgment rendered in a way that is not satisfactory to the requested State’s standards (see section 4 below). This lack of trust is confirmed by the bilateralisation system inserted in Article 29. The procedure chosen here is different from the classic bilateralisation, whereby a convention has effect among two States only if they have ‘accepted’ each other as partners by a separate ‘positive’ agreement (see the 1971 Hague Judgments Convention, Art. 21).18 The provision of Article 29 of the 2019 Convention is somewhat better because, instead of obliging a State to take an additional step for the Convention to be applicable, it is only ‘negative’ in the sense that a State only has to make a declaration if it does not want to be linked with a State which has ratified the Convention. By requesting a negative declaration, which is always more difficult to make, for diplomatic reasons, than a positive one, the negotiators were probably hoping that fewer declarations to that effect will be made in practice. If their hopes are fulfilled, the Convention will link more countries because most States will remain silent.

The question of reciprocity has recently taken a different twist with the establishment, in some regions of the world, of courts that have powers in civil and commercial matters but are organized jointly by several States (so-called ‘common courts’) to either replace their national courts in certain matters or to complement them.19 This is the case in all parts of the world, even though the most active courts in that sense are those established within the EU system. There are no theoretical reasons why a judgment rendered by one of these courts, if it fulfils the definition of Article 3(1)(b), could not be recognized or enforced under the Convention.20 Clavel and Jault explain that this matter has been somewhat debated, notably because the EU wanted these courts to be considered as part of the Contracting States’ judicial court networks. However, the Convention remains silent and the Report suggests that they should be given that status.

2.3 Additional declarations

States have other possibilities to limit the application of the Convention.

First, Article 17 allows them to declare that they will not apply the Convention when the situation which gave rise to the dispute and hence the judgment was considered ‘internal’ by the requested State. An ‘internal’ situation means that all parties were domiciled in the requested State and that all pertinent circumstances of the dispute were also located in that State. The underlying reason for such a provision is that if the situation was indeed internal to the requested State, there are no justifications for the parties to seek a judgment in another State. By such a provision the concept of the ‘neutral court’ (i.e. a court that has no link with the dispute – whether ratione materiae, rationae personae or ratione loci) is inoperative. This is regrettable because often, for commercial partners, when they do not choose arbitration, they at least want to choose a ‘neutral court’, i.e. one with no link to either party, in the hope that any potential bias when the court is located in the same State where one of the parties has its domicile or seat or was incorporated would be avoided up front. The judgment that will be rendered by such a court does not fall under the 2005 Choice of Court Convention (Arts. 1 and 19) and neither does it fall under the 2019 Convention. This is unfortunate because, at a time when arbitration is widely recognized as being too expensive for many smaller business claims, the parties’ clever and prudent practice in finding a court with no link to the dispute and the parties is, in a way, disapproved of or, at least, not encouraged since the judgment rendered will not benefit from the favourable treatment under the Convention.

Second, Article 18 makes it possible for States to exclude an area of the law from the Convention in addition to the exclusions already present in Article 2. It is true that the provision endeavours to limit States’ freedom to do so by requesting that ‘the declaration is no broader than necessary and that the specific matter excluded is clearly and precisely defined’.21 The exclusion will have effect in the requested State that has made such a declaration, but also in all other States parties to the Convention, when they become the requested State and the judgment was rendered in the State that made that declaration. This kind of provision, if clearly included to ‘facilitate ratification’ by States, decreases legal certainty for economic actors and may likewise decrease the number of cases falling under the Convention. Foreseeability for private parties may also be diminished, even though the declaration does not have retroactive effect. Indeed, retroactivity is considered as of the time when the litigation commenced. However, in contractual matters, parties do negotiate their agreements considering the law as it is at the time of the negotiations. Any declaration made by a State after the agreement has been concluded but before the start of the litigation will apply, even though it may upset the legitimate expectations of the parties.

Third, even though the Convention is not supposed to affect the rules on State immunity (Art. 2(5)), Article 19 adds that States, which do not consider themselves to be sufficiently protected by Article 2(5), may also make a declaration stating that they will not apply the Convention when they or a governmental agency are party to the dispute, whether as a claimant or as a defendant.22 As for the exact scope of Article 19 and contrary to Article 18, requested States other than the State that made the declaration may still apply the Convention and recognize or enforce the judgment despite the existence of such a declaration. Indeed Article 19(2) is drafted as an option given to the State and not in mandatory terms. Because of the combination of Articles 2(5) and 19, a company which contracts out with a State or a governmental agency will be well inspired to include an arbitration clause in the contract. When the State in question does not accept arbitration, a choice of court clause may be the second-best solution so that it falls under the 2005 Convention, although it does protect rules on immunity as well (Art. 2(6)). If none of the options above is possible, the company has very little hope of enforcing a judgment obtained against the State unless it falls squarely under the exceptions of immunity accepted by customary international law or the UN 2004 Convention on immunity and the requested court belongs to a State that has developed case law that is in accordance with either one of these two sources of law.

2.4 National law more favourable

The drafters of the 2019 Hague Convention have been inspired by Article VII of the 1958 New York Convention which allows Contracting States to apply more favourable rules to an award than the ones included in the New York Convention itself. For a country such as France, whose non-conventional rules are deemed to be more favourable than the rules of the New York Convention, the case law stemming from the New York Convention has almost entirely dried up. Article 15 of the 2019 Hague Convention is built upon the same philosophy although it is somewhat less generous than Article VII of the New York Convention. First, it does not apply to rights in rem. This result stems from the classic characterization of jurisdiction for rights in rem as ‘exclusive’ (Art. 6). This confirms that the negotiators did not have in mind only B2B situations where enterprises may very well want to deal with rights in rem together with other disputes or before another forum than the one located at the place where the property is situated. Why would States be worried about such an occurrence when no weaker party is concerned? What kind of public policy did the negotiators try to protect by such a rule, inserted without exceptions? What kind of public policy inherent in rights in rem may be at stake which would not be covered by the public policy exception? No answers to these questions may be found in the Convention or the Report.23 These are the kinds of provisions that will induce enterprises to include arbitration clauses in their deals involving commercial property. Second, the more generous rules must exist under ‘national law’, when Article VII of the New York Convention also provides for the application of treaties entered into by the requested States. Why the negotiators of the 2019 Convention did not go all the way as did the negotiators of the New York Convention remains a mystery. The Report offers no explanation. In any case, for a country such as France, it is plausible that the Convention will not be applied, or only marginally, since French non-conventional law is definitely more favourable to foreign judgments than the Convention. In addition, for judgments rendered within the European Union, if the law of the EU covers them, these rules pre-empt in any case those of the Convention (Art. 23(4)). If they are not covered, French non-conventional law will apply.

3. The principle of recognition/enforcement

The Convention is drafted with the objective to ‘facilitate rule-based multilateral trade and investment’ and to enhance predictability and certainty so that foreign judgments are globally recognized and enforced particularly in the commercial sphere. The core of the Convention is based on Articles 4 to 6.

The definition of a ‘judgment’ under the Convention does not warrant a lengthy comment. As already mentioned above, interim measures of protection are not considered to be ‘judgments’. On the contrary, judgments that include a settlement (‘transaction judiciaire’) reached before a court and which are enforceable as a judgment in the State of origin, are considered as judgments for the purpose of the 2019 Convention (Art. 11). These settlements have been excluded from the 2019 Singapore Convention (not yet ratified) on mediated settlements with a view of preserving the scope of the Hague Convention. No overlap should occur if judges carefully analyze the way in which a settlement has been reached. There is, however, a difficulty with court-ordered mediations, when the mediation takes place during the proceedings before the court, and the judge incorporates the mediated solution in the judgment. If all goes well, this judgment should be recognized and enforced under the 2019 Hague Convention. However, a State authority could be influenced by the fact that the settlement has been obtained through mediation and therefore be tempted to apply the Singapore Convention. The training of judges will be essential to ensure that they carefully delineate the scope of application of both Conventions.

The text does not differentiate between recognition and enforcement; both effects of the judgment must be granted after the conditions set forth in the text have been fulfilled. Therefore, there is no ‘reconnaissance de plein droit’ or, in other words, it is not possible for a company to use a judgment in a subsequent proceeding as a shield only, so that an issue fully discussed in the first instance and upon which the foreign court has decided, would not be reopened, even though no ‘enforcement’, strictly speaking, is necessary.24 There is no ‘automaticity’ in the workings of the Convention for recognition purposes. Therefore, in order to be recognized, a judgment must undergo the same scrutiny organized by the Convention for enforcement. However, because the Convention does not regulate the procedure under which the requested court must work (Art. 13), it is plausible that a State party to the Convention would grant the authority to any of its courts, seized of a matter for which a judgment has been rendered in a foreign State, to scrutinize that judgment as an incidental question before it renders its decision on the merits. It will be entirely up to the parties in dispute to bring to the attention of the court seized the existence of the foreign judgment and to the State to organize the procedural aspects of such scrutiny, a process which falls outside the scope of the Convention.

Article 4(2) provides for the classic rule according to which the requested court shall not review the merits of the foreign judgment. The enforcement procedure must not be used to retry the substance of the dispute. This rule aims at ensuring certainty in international private dispute resolution. Once the matter has been decided, deference should be given to the decision. This is sometimes expressed by the doctrine according to which parties should enjoy their day in court, but only one day in court. The only exception to the rule is of very limited effect. Indeed, the requested judge has the right to consider what has been decided in order to fulfil the requirements of the Convention (see section 4 below). Any such reconsideration will be limited to the facts ascertained during the foreign proceedings. But sometimes, other elements of the dispute will have to be considered so that the enforcement is effective.

The main verification that will have to be made by the requested judge is to check whether the judge, in the country of origin, had jurisdiction according to the Convention. In the entire history of the recognition and enforcement of foreign judgments around the world, the jurisdiction of the court of origin has always been the main factor of concern by the requested court and, as such, is always the target of that court’s verification.25 This is usually designated under the expression ‘compétence indirecte’, but the Convention does not use that expression and speaks only of ‘bases for recognition and enforcement’ in a fairly cryptic expression (Arts. 5 and 6). It is true that some of the bases enumerated in Article 5 are not, strictly speaking, jurisdictional bases. For example, Article 5(1)(c) is not worded in terms of jurisdiction. According to that provision, it suffices that the plaintiff (i.e. the person who seized the court) in the country of origin lost its case and the defendant in that action (presumably) presents the judgment for recognition and enforcement. In that case, one may presume that the plaintiff chose the court of origin (whatever the jurisdictional basis was) and that the defendant accepted it (probably tacitly by an appearance). Because it is the same defendant that wants to avail itself of the judgment, there is no reason to upset what may have been a ‘tacit’ agreement in the court of origin for its jurisdiction. Why this provision of Article 5(1)(c) was drafted and placed there in the Convention is unclear.

Before studying the multiple bases provided in Article 5, we need to clarify that the Convention, although it appears to provide for a single exclusive jurisdictional basis concerning rights in rem in immovable property (Art. 6), also includes another exclusive basis of jurisdiction in Article 5(3), for a residential lease or the registration of immovable property. The judgment deciding on such rights will be recognized or enforced ‘only if’ the immovable property at stake is situated in the country of origin. Why were the rules in Articles 5(3) and 6 separated into two different provisions when the policy behind both is identical? The only reason seems to be the applicability of Article 15. As explained above, States have the right to apply their own national law to judgments decided on matters covered in Article 5(3), but do not have that right when the judgment dealt with matters covered by Article 6. The difference in treatment is not immediately understandable. Practice will show whether it is a viable distinction.

The Convention provides for two other rules dealing with real estate and rights in rem. Article 5(1)(h) provides for the same rule as in Article 5(3), but applies it to a lease (tenancy) other than a residential lease. This is the provision that will apply to commercial leases. The difference with Article 5(3) is that the jurisdiction of the place where the property is located is not an exclusive jurisdiction. This is a good provision because, in commercial relations, there is no reason to limit the will of the parties to choose another court. There is also no reason to limit litigation strategies when a plaintiff wishes to amalgamate several claims against the same defendant and chooses to sue in another court having jurisdiction according to any other of the bases provided in Article 5 (for example, the defendant’s principle place of business or habitual residence). The same is true when the contractual obligation is secured by a right in rem in immovable property when both claims (contractual and the security) are brought together. Although, in that case, the Convention provides for the jurisdiction of the State where the property is situated (Art. 5(1)(i)), it is not necessarily the case that both issues would be brought at the same time, before the same court. The 2019 Convention will influence litigation practice for these disputes. The 2019 Convention closes a gap in the 2005 Convention from which all leases and tenancies were excluded (Art. 2(2)(l)).

The list of the bases of jurisdiction admitted by the Convention is long (13 grounds). They are all alternatives, i.e. only one of these jurisdictional bases is sufficient to trigger the ‘validity’26 of the court of origin’s jurisdiction and, therefore, allow the recognition and enforcement of the judgment (unless one of the obstacles studied in section 4 below is present in the case before the requested court). However, if the judgment is entered between a plurality of parties, the jurisdiction of the court of origin must be assessed for each party separately. The Convention is silent on this issue but the Draft Report makes that issue clear (§ 136).27 In order to facilitate an understanding of the Convention, we have grouped the bases according to the main philosophy behind them.

1) A series of provisions are based on the defendant’s consent:

– Simple consent, expressly made, at any point during the proceedings in the country of origin (Art. 5(1)(e)). This provision does not apply to litigation against a consumer or an employee, unless the consent was expressly worded (orally or in writing), before the court, once the proceedings have been commenced (Art. 5(2)(a)).28

– The defendant’s silence as to the jurisdiction of the court of origin, while defending on the merits (Art. 5(1)(f)). This procedural behaviour must be evaluated according to the country of origin’s law as to the time limit and the conditions under which the contest on jurisdiction must have been made. If it appears to the requested judge that the contest on jurisdiction was doomed under the law of the country of origin, then it may consider that there could not have been implicit consent by the defendant. This jurisdictional basis does not apply when the defendant is a consumer or an employee (Art. 5(2)(b)).

2) The parties’ agreement may also provide a proper basis for jurisdiction (other than Art. 5(1)(c) mentioned above):

– If the judgment ruled on a contractual obligation, the court of origin’s jurisdiction was based on the place of performance of that obligation as provided for in the contract (Art. 5(1)(g)(i)). This provision is similar to the one found in the Brussels 1 Regulation. It may be unfortunate to have included it here when one knows the difficulties the Court of Justice of the European Union (CJEU) has had when trying to interpret such a provision. One special difficulty is whether the court may also take into consideration the place of performance as defined in the INCOTERM inserted in the contract by the parties. This provision does not apply when the defendant is a consumer or an employee (Art. 5(2)(b)).

– In the case of a judgment concerning internal matters of a trust, the State of origin was designated in the trust instrument as the State whose courts had jurisdiction for the matters decided29 or as the State in which the principal place of administration was situated. This jurisdictional basis works only among the persons who are or were within the trust relationship and for issues concerning the validity, construction, effects, administration or variation of the trust (Art. 5(1)(k)).

– In all cases when the court of origin’s jurisdiction was based on a choice of court agreement other than an exclusive one (Art. 5(1)(m)). This provision is obviously made to preserve the 2005 Convention and to complement it for all choice of court agreements that fall outside the scope of that Convention. The definition of an exclusive choice of court agreement is found in Article 3(a) of the 2005 Convention. This provision does not apply when the defendant is a consumer or an employee (Art. 5(2)(b)).

3) A series of provisions take into consideration the defendant’s ‘presence’ in the country of origin:

– the habitual residence of the defendant in the State of origin at the time the proceedings were commenced in that State (Art. 5(1)(a)). This basis must be understood in the classic way. For a company, Article 3(2) does not provide for any new developments than the classic definition of its ‘residence (statutory seat or incorporation, or central administration or principal place of business)’. This is the only ‘general’ basis of jurisdiction provided for by the Convention, i.e. it applies whatever the claim may be and even if the claim arises out of a transaction or an activity outside the defendant’s place of residence.

– The defendant’s principal place of business in the State of origin, if the action was based on activities arising out of that business (Art. 5(1)(b)). This is akin to ‘transacting business’.

– The defendant had a branch, agency or establishment in the country of origin (without a separate legal personality) and the action stemmed out of the activities of that branch, agency or establishment (Art. 5(1)(d)).

4) A series of provisions are based on the defendant’s activities in the country of origin:

– In contractual obligations, the place of performance according to the law applicable to the contract, unless it clearly did not constitute ‘a purposeful and substantial connection’ to the State of origin (Art. 5(1)(g)(ii)). This is a combination of the old US requirement that the activities of the defendant in the State equate to a ‘purposeful availment of the laws of that State’ and the requirement by Canada that the link between the court of origin and the case is ‘substantial’. This is also the classic basis of indirect jurisdiction that French law has known for decades under the seminal Simitch case: ‘si le litige se rattache d’une manière caractérisée à l’Etat d’origine’.30

– In case the action was based on a non-contractual obligation, the act or omission directly causing the harm occurred in the State of origin (Art. 5(1)(j)). The Convention is very restrictive on this issue because it only takes into consideration the causal act, without any consideration being given to the place where the damage occurs (hence where the victim is situated). It is clearly favourable to tortfeasors. In most cases, it will not be different to the habitual residence of the defendant if it is an individual or, if it is a company, from the place of establishment of that company.

5) One additional provision is based on what is known as a ‘derived jurisdictional basis’:

– The judgment rules on a counterclaim (understandable because the jurisdiction on the counterclaim follows the jurisdiction on the main claim). But the provision does not state this. It adds two hypothetical situations. The first one is that of a judgment in favour of the counterclaimant. In that case, the counterclaim must have arisen out of ‘the same transaction or occurrence’ as the main claim (Art. 5(1)(l)(i)). The second one is that of a judgment deciding against the counterclaimant. In such a case, it will be enforced ‘unless the law of the State of origin required the counterclaim to be filed in order to avoid preclusion’ (Art. 5(1)(l)(ii)). The idea behind such a provision is to protect the defendant in case it was obliged to file the counterclaim, instead of being a voluntary decision on its part.

– No other derived jurisdictional basis is allowed under the 2019 Convention. Particularly, there is no provision on jurisdiction against ‘co-defendants’. This issue had given rise to a great deal of discussion in earlier efforts to harmonize the matter, since some countries ignore such a jurisdictional basis which is, nonetheless, very useful in litigation strategies for both plaintiffs and defendants. Therefore, it is not surprising that the Convention is silent on the matter. When co-defendants were joined in the court of origin, the only way to be able to enforce the judgment is to look at the jurisdictional basis against each of them.31

To conclude this section of our comment, one more missing basis of jurisdiction must be mentioned: denial of justice. Again, it is not surprising that such a basis of jurisdiction was not included. Although it has been recognized in many legal systems for a very long time, it is certainly not widely developed. It is useful essentially for victims located in countries where the judicial system is deficient. The fight against a denial of justice may grant them a forum in another country, when no other forum would be available. The fact that the 2019 Convention does not take that kind of jurisdictional basis into consideration is not surprising, as it has been negotiated essentially with a commercial model in mind. However, in the fight against violations of fundamental rights by corporations around the world, civil actions have become the only remedy that the victims of these violations may have to obtain redress. Some countries may use the fight against a denial of justice as the only jurisdictional basis available to victims. These judgments do not benefit from the 2019 Convention and remain within the ambit of national law.

4. The obstacles to recognition/enforcement

The obstacles to recognition/enforcement are necessarily limited because of the general philosophy behind the Convention favouring recognition and enforcement.32 However, they are sufficient (seven in total) to filter judgments rendered after proceedings that do not conform with the minimum of due process. This part of our comment will be necessarily limited, the provisions of the 2019 Convention (Arts. 7-10) being everything but novel.

All the grounds for refusal included in Article 7 are optional for the requested Court. Therefore, there is no obligation, under the Convention, to look at any of the obstacles sua sponte, even if it is a question of public policy. The question is, however, whether a State becoming a Party to the Convention could, under cover of this being a procedural issue,33 impose an obligation on its courts to look at these issues even if the person against which the judgment is sought to be recognized or enforced does not raise any of the obstacles included in Article 7. We are of the opinion that a State should not be free to do so, because of the liberal philosophy behind the Convention and because national laws can only be ‘more favourable’ than the Convention itself.34

The first obstacle (Art. 7(1)(a)) deals with the manner in which the notification of the commencement of the proceedings was performed in the country of origin. This is a classic requirement. It covers both the timing and the substantive requirements of the notification. Concerning the timing, the defendant must have had sufficient time to defend its cause. The only exception that is admitted is if the defendant appeared in the court of origin and did not challenge the notification, unless the law of the State of origin does not allow such a challenge. Concerning the substantive requirement, the requested State is only allowed to impose its ‘fundamental principles’. One can assume that the language of the notification would be part of these principles. If the documents were notified in a language that the defendant did not understand, the requested court could consider that the notification was not properly done and that the judgment must not be recognized or enforced.

The second obstacle concerns fraud (Art. 7(1)(b)). This is again a classic obstacle. The Convention does not provide any detail about the kind of fraud at stake here apart from the fact that the fraud must have been instrumental in ‘obtaining the judgment’. Consequently, one may assume that the fraudulent acts may both relate to the procedural aspects of the case (the claimant has lied about the whereabouts of the defendant) and to the substance of what was submitted to the court of origin (false documents, for example). Although other provisions of Article 7 may also cover procedural fraud, the negotiators felt that it was preferable to provide for a stand-alone provision covering both.

The third obstacle deals with the public policy exception (Art. 7(1)(c)). There is one novel aspect in this provision as it includes in the concept of public policy an infringement of security or sovereignty of the requested State. One may wonder why this was necessary at all. It has to be understood as a pendant to all the provisions elsewhere in the Convention about State activities and immunities. The States which insisted on having all these provisions included also wanted to make sure that, at the recognition and enforcement level, they could use their sovereignty as part of their public policy.

The fourth obstacle relates to jurisdiction (Art. 7(1)(d)) and is more surprising since one could have assumed that all jurisdictional issues were dealt with in Article 5. Article 7(4) is specifically geared towards protecting a choice of court agreement, so that if the court of origin ‘wrongly’ (in the eyes of the requested court) considered that the agreement was null and void or otherwise inapplicable, the requested court, being of a different opinion, may refuse recognition or enforcement. The philosophy behind such a provision is commendable in a liberal society when party autonomy is a very high standard. However, it is certainly not proper to end litigation. Instead, it would have been much better to provide some kind of cooperation between the two courts so that they could discuss their different approaches in interpreting the choice of court agreement. The same policy also underlies Article 8(1) which is meant to protect the exclusive jurisdictional basis in Article 6.

The fifth to seventh obstacles (Art. 7(1)(e) and (f), and 7(2)) deal with situations where there was concurrent jurisdiction and proceedings, whether two judgments were rendered concurrently or whether the situation is still one of lis pendens. As is classic in those situations, the requested State’s decision has priority over the one rendered by the court of origin. If the case is still pending in the requested court, it will have priority if it was seized before the court of origin and the requested State has a close connection with the dispute.

Additional obstacles provide for the situation where the judgment dealt with preliminary questions (Art. 8) or with damages (Art. 9). Article 8 has been drafted so as to protect the material scope of the Convention, so that a decision that deals with a preliminary question falling outside that scope cannot circulate. On Article 9, suffice it to say that this provision has become customary in the field so as to prevent the effect of decisions containing exemplary or punitive damages that do not compensate a party for the actual loss or harm suffered. This shows that the concept of ‘private enforcement’ or the deterrent effect of private litigation is still looked upon as being evil in many legal systems.

5. Conclusion

For companies involved in litigation around the world that could not include an arbitration clause in their contract, the 2019 Hague Convention gives them some reasons to be satisfied: (a) If they were the defendant in the court of origin, they are protected not only in the way the jurisdictional bases are formulated but also in the provisions that the requested court has at hand to refuse recognition and enforcement; (b) If they were the claimant in the court of origin and the proceedings were in conformity with the most current due process requirements, they should be able to see the decision rendered being given effect in all other Contracting States.


* Professor Emerita – University Panthéon-Assas Paris II, Arbitrator and Mediator.

1 ‘Direct jurisdiction’ designates the jurisdiction of the court which was asked to decide on the merits of the dispute, and thus renders the judgment which is the object of the recognition or enforcement proceedings in another State. It contrasts with ‘indirect jurisdiction’, an expression which designates the criteria used by the requested court to verify the jurisdiction of the rendering court. Despite the recurring difficulties, the Conference decided to embark upon an exercise dealing with direct jurisdiction. A first report is expected in 2020.

2 That Convention has been transformed into a Regulation now known as Regulation (EU) 1215/2012 of the European Parliament and of the Council on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, OJ 2012, L 351/1.

3 A diplomatic conference was convened in 2001, but, in a rare move, the delegates decided to end the conference without reaching an agreement on a text.

4 The Convention has entered into force thanks to ratification by the European Union for its Member States. The UK and Denmark have ratified separately. Outside the EU, only three States have ratified the Convention: Mexico, Montenegro and Singapore. China, Ukraine and the USA have signed the text but have not ratified it. During the first Obama administration, there was a strong political desire, at the federal level, to ratify and implement the Convention. However, the National Conference of Commissioners on Uniform State Laws insisted that the matter fell within the ambit of the states’ authority. This internal obstacle could not be overcome and therefore the project was abandoned and will most certainly not be revived at any time soon. The text of the Convention is available at

5 As far as the EU is concerned, the situation is somewhat complex. One could say that because the EU has regulated the matter internally, only the EU has the competence to negotiate the Convention. In addition, one must not forget that the EU is a full member of the Hague Conference, having been admitted in 2007 (at the time it was still the European Community), after an amendment to the Organization’s bylaws. However, because the Member States are still full members of the Conference, they are present in the room but, under the general principle of bona fide cooperation, they are not allowed to take another position than the one adopted during the coordination meetings. This is only partially true for Denmark, which is not a party to the PIL rules, even though the principle of cooperation applies to Denmark. The UK, being on the eve of leaving the EU, has more freedom in the negotiations.

6 In order to draft this comment we benefited from the presentation made by Andrea Stein before the Groupe européen de droit international privé (GEDIP) in September 2019, and the communication given by Sandrine Clavel and Fabienne Jault-Seseke before the Comité français de droit international privé in October 2019 (preliminary text available on the website of the CFDIP). We also consulted the preliminary Report by P. Garcimartin and G. Saumier (hereinafter ‘the Report’), the final Report not being publicly available at the time this comment was written.

7 This expression has become the ‘buzz word’ in private international law (whether in the EU or for rules adopted under the auspices of the Hague Conference). However, it means very little in reality. What is really meaningful concerns the multiple exclusions listed in the text. In practice, the difficult issues will be: (a) to interpret the exclusions themselves and (b) to have a judgment enforced under the Convention even though it may be based on another decision that is excluded from the Convention (Art. 8), or it includes a decision on a preliminary question itself excluded (Art. 9), or the decision was rendered when an issue pertaining to an exclusion was discussed but was not the ‘object’ of the decision (Art. 2(2)).

8 Compare with Art. 2(4) of the 2005 Choice of Court Convention. The difference in drafting may not have any impact on the interpretation of the exclusion which must be understood to preserve the 1958 New York Convention on the recognition and enforcement of foreign arbitral awards and all other international texts and national laws dealing with arbitration.

9 Most probably, the matter will be decided under either the 1958 New York Convention which has already been ratified by 161 countries (as of December 2019), or national arbitration rules.

10 This situation falls under Art. 5(1)(f) of the 2019 Convention.

11 The meaning of that term is to be understood under the 2005 Convention. See our comment in ‘La Convention de La Haye du 30 juin 2005 sur l’élection de for’, J.D.I. 2006, pp. 813-850.

12 Art. 5(1)(m). This is understandable (but regretful) because the 2005 Convention only deals with exclusive choice of court agreements.

13 Art. 5(1)(e).

14 The exclusion of intellectual property (IP) is symptomatic of the difficulties that area of the law creates. First, one must remember that IP is the basis of a major part of international economic wealth. However, differences between State laws in the field render the issues of jurisdiction and the applicable law even more difficult to negotiate. This is one of the areas that was mostly responsible for the demise of the 2000 draft Convention. It is no surprise, therefore, that this has been one of the most contentious issues in 2019. The exclusion is sharp and broad. Therefore, it should allow States to exclude from the Convention all kinds of IP rights (even if not universally considered as such) and all kinds of decisions. However, the Report suggests that a judgment limited to the matter of royalties would not be excluded from the Convention because it would be based mostly on contract law. Clavel and Jault explain that this is an attempt to save some judgments from the broad exclusion. However, it is hard to see any textual support for such a limitation. Indeed, in practice, a judgment affirming the duty of a party to pay royalties or negating such an obligation is necessarily based on a decision to consider the IP right valid or invalid. The final act of the diplomatic session includes an invitation to the General Affairs Council of the Hague Conference to explore the possibility of further negotiations in the field. From an outsider’s point of view, and considering all the work that has already been done by the Hague Conference, in coordination with the World Intellectual Property Organization (WIPO), by the American Law Institute and by many other study groups (for example, under the auspices of the Max Planck Institute), it is difficult to see how the major differences between the States can be reduced to arrive at a commonly acceptable set of rules.

15 The textual limit of the exclusion of anti-trust matters (which is the product of a heavily discussed compromise – see the full text in the Convention, at shows that the negotiators were aware of the fairly large number of business cases where anti-trust issues are indeed present, notably where they constitute a disruption of market functioning, so that concerted practices must be discouraged. The Report reveals that the 2019 Convention wanted to be in line with the policies of the Organisation for Economic Cooperation and Development (OECD) (see § 69). An additional striking feature of the provision must be highlighted: the exception to the exclusion is limited to practices that were conducted and their effect developed in the State of origin. It is a novel drafting technique to include a jurisdictional link in the definition of a rule on scope. Having said that, in current economic trends, this exception to the exclusion will rarely apply since most anti-competitive practices have effects in several States at the same time. The result is, at best, unfortunate.

16 The entire treatment of State activities in the 2019 Convention is complex. Many provisions deal with these issues, showing, if necessary, that the UN Convention of 2004 on State immunities has not been widely accepted and that, even for States which have ratified it, the uniformity of interpretation of key concepts may not have been achieved.

17 Principles on Provisional and Protective Measures in International Litigation, Resolution adopted at the Sixty-Seventh Conference, Helsinki, 1996, Report, p. 202.

18 This system has always been believed to be the main reason why the 1971 Convention was not a success.

19 This is the case, notably, with the OHADA in Africa, the EU and the Mercosur in South America.

20 The Report (§ 101) suggests that this must be the case.

21 Art. 18(1), last sentence.

22 This is true also if an individual represents the State or the governmental agency in the proceedings. On these issues, see also section 4 below.

23 The Report (§ 233) only mentions ‘proximity’ and the fact that public registers and other public documents may be at stake in such proceedings.

24 There is also no room in the 2019 Convention for any other effect than recognition or enforcement. For example, the ‘effet de fait’ or ‘effet de titre’ known under French law are not covered by the Convention.

25 The only exception is known in systems where the judicial jurisdiction is regulated uniformly at the level of the court of origin. This is the case in the United States of America (although this is in very broad terms) and in the European Union (where it is much more detailed and precise).

26 By using the concept of ‘validity’ we do not mean that the court of origin needs to use one of the bases. In fact, because the Convention only provides for ‘indirect’ bases of jurisdiction, it does not matter what basis, in practice, the court of origin actually used. It suffices that the requested court considers that the court of origin did have jurisdiction under one of the Convention bases.

27 The Draft Report is available at

28 The French version is clearer than its English counterpart.

29 This provision must be interpreted with the concepts of the 1985 Hague Convention on Trusts in mind.

30 French Court of cassation, Civ. 1, 6 February 1985, n° 83-11241. A recent case seems to revisit partially the 1985 decision without major implications as to the overall effect of the rule: French Court of cassation, Civ. 1, 19 December 2018, n° 17-28562, Rev. crit. DIP 2019, p. 818, obs. D.B.

31 See above.

32 See the introduction above. This philosophy is clearly reiterated in Art. 9 allowing the recognition or enforcement of only a part of a judgment if that part is severable from the remainder of the decision.

33 The Convention says nothing (or very little) on the procedure for recognition or enforcement.

34 Implicit in Art. 15 (see above).