Views
More on the Validity of the PDVSA 2020 Bonds
Written by Mark Weidemaier, the Ralph M. Stockton, Jr. Distinguished Professor at the University of North Carolina School of Law, and Mitu Gulati, the Perre Bowen Professor of Law at the University of Virginia School of Law.
Governments with no realistic prospect of paying their debts often gamble for redemption, trying desperately to avoid default. Political leaders, with good reason, fear that a debt default will get them thrown out of office. But in trying to hold power, sometimes by borrowing even more, they often make matters worse for the country and its people. A prime example involves the collateralized bonds issued by Venezuelan state oil company, PDVSA.
Venezuela’s Gamble
In 2016, PDVSA was about to default on its debt, as was the Venezuelan state itself. At that stage, it was already well beyond the point where the debt should have been restructured, given worsening domestic conditions. Instead, the Maduro government gambled. It conducted a debt swap in which investors exchanged unsecured PDVSA bonds for new ones due in 2020. To sweeten the deal, the PDVSA 2020s were backed by collateral in the form of a 50.1% interest in CITGO Holding, the parent company of U.S. oil refiner CITGO Petroleum. The deal bought a few extra years but put at risk the country’s primary asset in the United States.
Even at the time, it was uncertain whether Venezuelan law authorized the transaction. The Venezuelan Constitution requires legislative approval for contracts in the national public interest. Maduro did not seek approval because opposition lawmakers controlled the National Assembly and had made clear they would not grant it. The deal went ahead anyway.
Times have changed. The United States recognizes Juan Guaidó as Venezuela’s interim president (for now). The PDVSA 2020 bonds are in default. The bondholders want their collateral. PDVSA has challenged the validity of the bonds. But the bonds include a choice-of-law clause designating the law of New York. Does this mean that validity is to be determined under New York law? John Coyle recently wrote a terrific post about the case and its significance on this blog. We write to provide some broader context, drawing from our article, Unlawfully Issued Sovereign Debt.
Sovereign Debt and Choice-of-Law Clauses
The story of the PDVSA 2020 bonds is a common one in government debt markets. A government borrows money in dodgy ways or at a time of financial distress. Arguably, the debt contravenes domestic law, although the government may obtain legal opinions affirming its validity. The debt also includes a choice of law clause providing for the application of foreign law, typically that of New York or England. Later, a new government comes to power and disputes the validity of the debt. We have seen this pattern in Venezuela, Mozambique, Ukraine, Zambia, Liberia, Puerto Rico, and in other sovereign and sub-sovereign borrowers. (The pattern goes back even further – for a delightful treatment of the hundreds of such cases from the 1800s involving municipal debt, see here).
These cases raise what seems like a simple question: Does an international bond—i.e., one expressly made subject to foreign law—protect investors against the risk that the bond will later be deemed in violation of the issuer’s domestic law? Despite seeming simple, and how frequently the question arises, there is little clarity about the answer. New York law governs a big part of the sovereign debt markets, and the choice-of-law question in the PDVSA 2020 case has been certified to the New York Court of Appeals. Will that court’s decision offer clarity?
Variations in Clause Language
Count us skeptical. The problem is not just the unpredictability of choice of law rules. It is that many choice-of-law clauses are drafted in perplexing ways, which leave unclear the extent of protection they offer to investors. Consider three examples. The first is from the PDVSA 2020 bond itself where the relevant language is capitalized (as if capitalization has some magic effect):
THIS INDENTURE AND THE NOTES SHALL BE CONSTRUED IN ACCORDANCE WITH, AND THIS INDENTURE AND THE NOTES AND ALL MATTERS ARISING OUT OF OR RELATING IN ANY WAY WHATSOEVER TO THIS INDENTURE AND THE NOTES (WHETHER IN CONTRACT, TORT OR OTHERWISE) SHALL BE GOVERNED BY, THE LAWS OF THE STATE OF NEW YORK WITHOUT REGARD TO THE CONFLICTS OF LAW PROVISIONS THEREOF (OTHER THAN SECTION 5-1401 OF THE NEW YORK GENERAL OBLIGATIONS LAW)
This clause apparently seeks to extend New York law to the widest possible range of questions. Whether that includes the question of whether the bonds were validly issued is, as John’s post puts it, the “billion-dollar question.” And the answer is not clear. The decision by the New York Court of Appeals might provide some clarity on it . . . maybe.
But now consider this clause, from a Brazilian bond (emphasis ours):
The indenture and the debt securities will be governed by, and interpreted in accordance with, the laws of the State of New York without regard to those principles of conflicts of laws that would require the application of the laws of a jurisdiction other than the State of New York . . .; provided, further, that the laws of Brazil will govern all matters governing authorization and execution of the indenture and the debt securities by Brazil.
Does the bold text mean that investors cannot enforce a loan issued in violation of Brazilian law? We aren’t sure. As we discuss in the paper, it can be hard to identify questions of “authorization” and “execution,” especially in the context of sovereign borrowing. Consider the question whether a loan violates a constitutional or statutory debt limit. Does the debt limit negate the sovereign’s capacity to borrow, limit the authority of government officials to bind the sovereign, or make the loan illegal or contrary to policy? How one categorizes the issue will affect the answer to the choice-of-law question. Carve outs like this—which reserve questions of authorization and execution for resolution under local law—appear in around half the New York-law sovereign bonds we examined.
Finally, consider this clause from a Turkish bond (again, emphasis ours):
[The] securities will be governed by and interpreted in accordance with the laws of the State of New York, except with respect to the authorization and execution of the debt securities on behalf of Turkey and any other matters required to be governed by the laws of Turkey, which will be governed by the laws of Turkey
What now? This “other matters” carve out is even odder than the one for questions of authorization and execution. It hints that additional, unspecified matters might be governed by the sovereign’s local law. Indeed, it implies that the sovereign’s own law might determine which issues fall within the “other matters” exception. If so, the clause potentially allows the government to create new exceptions to the governing law clause.
Conclusion
Our discussions with senior sovereign debt lawyers have done little to dispel our uncertainty about the meaning of these clauses. They seem just as confused as we are. All we can say with confidence is that many choice of law clauses include traps for unwary investors. Until drafting practices converge on a consistent and coherent model, the choice-of-law question is likely to remain fodder for litigation.
[This post is cross-posted at Transnational Litigation Blog.]
The Billion-Dollar Choice-of-Law Question
Choice-of-law rules can be complex, confusing, and difficult to apply. Nevertheless, they are vitally important. The application of choice-of-law rules can turn a winning case into a losing case (and vice versa). A recent decision in the U.S. Court of Appeals for the Second Circuit, Petróleos de Venezuela S.A. v. MUFG Union Bank, N.A., is a case in point. The Second Circuit was called upon to decide whether to apply the law of New York or the law of Venezuela to determine the validity of certain notes issued by a state-owned oil company in Venezuela. Billions of dollars were riding on the answer.
In this post, I first review the facts of the case. I then provide an overview of the relevant New York choice-of-law rules. Finally, I discuss the choice-of-law question that lies at the heart of the case.
The Bonds
In 2016, Venezuela’s state-owned oil company, Petróleos de Venezuela, S.A. (“PDVSA”) approved a bond exchange whereby holders of notes with principal due in 2017 (the “2017 Notes”) could exchange them for notes with principal due in 2020 (the “2020 Notes”). Unlike the 2017 Notes, the 2020 Notes were secured by a pledge of a 50.1% equity interest in CITGO Holding, Inc. (“CITGO”). CITGO is owned by PDVSA through a series of subsidiaries and is considered by many to be the “crown jewel” of Venezuela’s strategic assets abroad.
The PDVSA board formally approved the exchange of notes in 2016. The exchange was also approved by the company’s sole shareholder and by the boards of the PDVSA’s subsidiaries with oversight and control of CITGO.
The National Assembly of Venezuela refused to support the exchange. It passed two resolutions – one in May 2016 and one in September 2016 – challenging the power of the executive branch to proceed with the transaction and expressly rejecting the pledge of CITGO assets in the 2020 Notes. The National Assembly took the position that these notes were “contracts of public interest” which required legislative approval pursuant to Article 150 of the Venezuelan Constitution. These legislative objections notwithstanding, PDVSA followed through with the exchange. Creditors holding roughly $2.8 billion in 2017 Notes decided to participate and exchanged their notes for 2020 Notes.
In 2019, the United States recognized Venezuela’s Interim President Juan Guaidó as the lawful head of state. Guaidó appointed a new PDVSA board of directors, which was recognized as the legitimate board by the United States even though it does not control the company’s operations inside Venezuela. The new board of directors filed a lawsuit in the Southern District of New York against the trustee and the collateral agent for the 2020 Notes. It sought a declaration that the entire bond transaction is void and unenforceable because it was never approved by the National Assembly. It also sought a declaration that the creditors were prohibited from executing on the CITGO collateral.
Choice of Law
If the 2020 Notes were validly issued, they are binding on PDVSA, and the CITGO assets may be seized by the noteholders in the event of default. If the notes were not validly issued, they are not binding on PDVSA, and the CITGO assets may not be seized by the noteholders in the event of default. Whether the Notes were validly issued depends, in turn, on whether the court applies New York law or Venezuelan law. This is the billion-dollar choice-of-law question. If New York law applies, then the notes will almost certainly be deemed valid and the noteholders can seize the pledged collateral. If Venezuelan law is applied, then the notes may well be deemed invalid and the noteholders will be stymied. With the stakes in mind, let us now turn to the applicable choice-of-law rules.
A federal court sitting in diversity must look to the choice-of-law rules of the state in which it sits—here, New York—to decide which jurisdiction’s law to apply. N.Y. General Obligations Law 5-1401 states that a New York choice-of-law clause should be enforced whenever it appears in a business contract worth more than $250,000 in the aggregate. The 2020 Notes contain New York choice-of-law clauses. Since the aggregate value of the 2020 Notes is far greater than $250,000, and since the 2020 Notes have no relation to personal, family or household services, it may seem that the court should simply apply New York law and call it a day.
There is, however, another New York choice-of-law rule that may trump Section 5-1401. Section 5-1401 states that it shall not apply to any contract “to the extent provided to the contrary in . . . section 1-301 of the Uniform Commercial Code.” Section 1-301(c) states that if N.Y Commercial Code Section 8-110 “specifies the applicable law, that provision governs and a contrary agreement is effective only to the extent permitted by the law so specified.” Section 8-110(a), in turn, states that “[t]he local law of the issuer’s jurisdiction . . . governs . . . the validity of a security.”
All of this suggests that the applicable choice-of-law rule may not be the one laid down in Section 5-1401. Section 8-110 directs courts to apply the local law of the issuer’s jurisdiction—here, Venezuela—to resolve issues relating to the “validity” of the security. The billion-dollar question is what exactly the word “validity” means in this context.
On the one hand, the term may be interpreted broadly to refer to both the corporate law of Venezuela and to Venezuelan law more broadly. Under this interpretation, the 2020 Notes may not be validly issued because they were never approved by the National Assembly as required under Article 150. On the other hand, the term “validity” may be interpreted to refer only to the corporate law of Venezuela. Under this narrower interpretation, it is irrelevant whether the National Assembly approved the 2020 Bonds because all of the corporate formalities needed to validly issue a security—approval by the board of directors, approval by the shareholders, etc.—appear to have been followed.
Interpretation in the District Court
In a lengthy decision decided on October 16, 2020, the U.S. District Court for the Southern District of New York (Judge Katherine Polk Failla) concluded that the term “validity” should be given a narrow interpretation and that New York contract law governed the issue of validity.
The court began its analysis by observing that the strongest argument in support of a broad interpretation is based on plain language. This term “validity” is not generally understood to refer solely to corporate formalities. It is understood to encompass the many reasons why a contract may not be enforceable as a matter of contract law. While this plain language reading is compelling at first glance, the court ultimately concluded that it did not mandate the application of general rules of Venezuelan law given the broader context of Article 8.
The court first quoted the following language from the Prefatory Note to Article 8:
[Article 8] deals with the mechanisms by which interests in securities are transferred, and the rights and duties of those who are involved in the transfer process. It does not deal with the process of entering into contracts for the transfer of securities or regulate the rights and duties of those involved in the contracting process (emphasis added).
The court observed that if the term “validity” were given a broad scope, it would “swallow whole any choice of law analysis involving the formation of a contract for securities.” The court cited state legislative history indicating that the term “validity” in Article 8 referred merely to whether a security “ha[d] been issued pursuant to appropriate corporate or similar action.” The court also quoted the authors of a leading treatise on Article 8 as saying that:
Obviously, the concept of “invalidity” as used in this section must have a narrower scope than one might encounter in other legal contexts, e.g., in a dispute about whether the obligation represented by the security is “enforceable” or “legal, valid, and binding.”
Finally, the district court noted the virtual absence of any New York case law supporting the broad interpretation of the validity favored by the plaintiffs. If the term was as sweeping as the plaintiff claimed, the court reasoned, there would be more cases where the courts had applied Section 8-110. The lack of any such cases cut against giving the term a broad interpretation. The district court’s analysis of this issue has attracted support from some commentators and criticism from others.
After concluding that the term “validity” in Section 8-110 should be interpreted narrowly to select only Venezuelan corporate law, the district court applied New York contract law. It held that the 2020 Notes were valid and enforceable and that the defendant trustee was entitled to judgment in the amount of $1.68 billion. The plaintiffs appealed.
Interpretation in the Second Circuit
On October 13, 2022, the U.S. Court of Appeals for the Second Circuit declined to provide a definitive answer as to the interpretive question discussed above. After reviewing the various arguments for and against a broad interpretation of “validity,” the court certified the question to the New York Court of Appeals. In so doing, the court commented on the issue’s importance to “the State’s choice-of-law regime and status as a commercial center.” It also noted the importance of the choice-of-law issue to the ultimate outcome in the case:
If the court concludes New York choice-of-law principles require the application of New York law on the issue of the validity of the 2020 Notes, and that Article 150 and the resolutions have no effect on the validity of the contract under New York law, then we would affirm the district court’s decision to apply New York law and uphold the validity of the bonds. On the other hand, if the court concludes Venezuelan law applies to the particular issue of PDVSA’s legal authority to execute the Exchange Offer, then we would likely remand for an assessment of Venezuelan law on that question and, if necessary, for consideration of the Creditors’ equitable and warranty claims.
The fate of the 2020 Notes—and the billions of dollars those notes represent—is now in the hands of the New York Court of Appeals.
Conclusion
There will be additional updates and commentary on Petróleos de Venezuela S.A. v. MUFG Union Bank, N.A. at Transnational Litigation Blog in the weeks and months ahead. In the meantime, please feel free to mention this case the next time a student or a colleague questions the importance of choice-of-law rules. These rules matter. A lot.
[This post is cross-posted at Transnational Litigation Blog.]
What is an international contract within the meaning of Article 3(3) Rome I? – Dexia Crediop SpA v Provincia di Pesaro e Urbino [2022] EWHC 2410 (Comm)
The following comment has been kindly provided by Sarah Ott, a doctoral student and research assistant at the University of Freiburg (Germany), Institute for Comparative and Private International Law, Dept. III.
On 27 September 2022, the English High Court granted summary judgment and declaratory relief in favour of the Italian bank Dexia Crediop SpA (“Dexia“) in its lawsuit against the Province of Pesaro and Urbino (“Pesaro”), a municipal authority in the Marche region of Italy. This judgement marks the latest development in a long-running dispute involving derivative transactions used by Italian municipalities to hedge their interest rate risk. Reportedly, hundreds of Italian communities entered into interest rate swaps between 2001 and 2008 having billions of Euros in aggregate notional amount. It is also a continuation of the English courts’ case law on contractual choice of law clauses. Although the judgments discussed in this article were, for intertemporal reasons, founded still on Art. 3(3) of the Rome Convention, their central statements remain noteworthy. The Rome Convention was replaced in almost all EU member states, which at the time included the United Kingdom, by Regulation (EC) No 593/2008 (“Rome I”), which came into effect on 17 December 2009. Article 3 Rome I Regulation contains only editorial changes compared to Article 3 of the Rome Convention. As a matter of fact, Recital 15 of the Rome 1 Regulation explicitly states that despite the difference in wording, no substantive change was intended compared to Article 3(3) of the Rome Convention.
In the case at hand, Pesaro and Dexia entered into two interest rate swap transactions in 2003 and 2005. Each of the transactions was subject to the 1992 International Swap Dealers Association (“ISDA”) Master Agreement, Multicurrency – Cross Border and a Schedule therto. During the 2008 financial crisis, the swaps led to significant financial burdens for Pesaro. In June 2021, Pesaro commenced legal proceedings in Italy seeking to unwind or set aside these transactions. Dexia then brought an action in England to establish the transactions were valid, lawful and binding on the parties.
A central question of the dispute was the law applicable to the contract. Pesaro claimed breaches of Italian civil law in its proceedings, while Dexia argued that only English law applies. As correctly stated by the court, the applicable law is determined by the Rome Convention, as the transactions between the parties took place in 2003 and 2005. According to Article 3(1) Rome Convention, a contract is governed by the law chosen by the parties. The ISDA Master Agreement in conjunction with the Schedule contained an express choice of law clause stating that the contract is to be governed by and construed in accordance with English law. Of particular importance therefore was whether mandatory provisions of Italian law could nevertheless be applied via Article 3(3) Rome Convention. This is the case if “all the [other] elements relevant to the situation at the time of the choice are connected with one country only […]”. In order to establish weather Article 3(3) applied, the court referred to two decisions of the English Court of Appeal. Both cases also concerned similar interest rate swap transactions made pursuant to an ISDA Master Agreement with an expressed choice of English law.
In Banco Santander Totta SA v Companhia de Carris de Ferro de Lisboa SA [2016] EWCA Civ 1267, the Court of Appeal extensively discussed the scope of this provision in connection with the principle of free choice of law, more precisely, which factors are to be considered as “elements relevant to the situation”. This was a legal dispute between the Portuguese Santander Bank and various public transport companies in Portugal. First, the Court of Appeal emphasised that Article 3(3) Rome Convention is an exception to the fundamental principle of party autonomy and therefore is to be construed narrowly. Therefore, “elements relevant to the situation” should not be confined to factors of a kind which connect the contract to a particular country in a conflict of laws sense. Instead, the Court stated that it is sufficient if a matter is not purely domestic but rather contains international elements. Subsequently the court assessed the individual factors of the specific case. In so far, the Court of Appeal confirmed all factors the previous instance had taken into account. Relevant in the case was the use of the “Multi-Cross Border” form of the 1992 ISDA Master Agreement instead of the “Local Currency-Single Jurisdiction” form, that the contract included the right to assign to a foreign bank and the practical necessity for a foreign credit institution to be involved, as well as the foreseeability of the conclusion of hedging arrangements with foreign counterparties and the international nature of the swap market. These factors were found sufficient to establish an international situation.
In Dexia Crediop S.P.A. v. Comune di Prato [2017] EWCA Civ 428, the Court of Appeal addressed the issue again and concluded that already the fact that the parties had used the “Multi-Cross Border” form of the 1992 ISDA Master Agreement in English, although this was not the native language of either party, and the conclusion of back-to-back hedging contracts in connection with the international nature of the derivatives market was sufficient.
In the present case, Dexia again relied on the use of the ISDA Master Agreement, Multicurrency – Cross Border and on the fact that Dexia hedged its risk from the transactions through back-to-back swaps with market participants outside Italy. But as the relevant documents were not available, the second circumstance could not be taken into account by the court. Nevertheless, the court considered that the international element was sufficient and Article 3(3) of the Rome Convention was not engaged.
Thus, this new decision not only continues the very broad interpretation of the Court of Appeal as to which elements are relevant to the situation, but also lowers the requirements even further. This British approach appears to be unique. By contrast, according to the hitherto prevailing opinion in other Member States, using a foreign model contract form and English as the contract language alone was not sufficient to establish an international element (see, e.g., Ostendorf IPRax 2018, p. 630; Thorn/Thon in Festschrift Kronke, 2020, p. 569; von Hein in Festschrift Hopt, 2020, p. 1405). Relying solely on the Master Agreement in order to affirm an international element seems unconvincing, especially when taking Recital 15 of the Rome I Regulation into account. Recital 15 Rome I states that, even if a choice of law clause is accompanied by a choice of court or tribunal, Article 3(3) of the Rome I Regulation is still engaged. This shows that it is the purpose of this provision to remove the applicability of mandatory law in domestic matters from the party’s disposition. The international element must rather be determined according to objective criteria. With this interpretation, Article 3(3) of the Rome I Regulation also loses its effet utile to a large extent.
Unfortunately, the Court of Appeal considered its interpretation to be an acte clair and therefore refrained from referring the case to the CJEU. Since Brexit became effective, the Rome I Regulation continues to apply in the United Kingdom in an “anglicised” form as part of national law, but the English courts are no longer bound by CJEU rulings. As a result, a divergence between the English and the Continental European assessment of a choice of law in domestic situations is exacerbated.
This also becomes relevant in the context of jurisdiction agreements. In the United Kingdom, these are now governed by the HCCH 2005 Choice of Court Convention which is also not applicable according to article 1(2) if, “the parties are resident in the same Contracting State and the relationship of the parties and all other elements relevant to the dispute, regardless of the location of the chosen court, are connected only with that State”. As there is a great interest in maintaining the attractiveness of London as a the “jurisdiction of choice”, it is very likely that the Court of Appeal will also apply the standards that it has developed for Article 3(3) Rome I to the interpretation of the Choice of Court Convention as well.
One can only hope that in order to achieve legal certainty, at least within the European Union, the opportunity for a request for referral to the CJEU will present itself to a Member State court as soon as possible. This would allow the Court of Justice to establish more differentiated standards for determining under which circumstances a relevant foreign connection applies.
News
Second Issue of the Journal of Private International Law for 2024
The second issue of the Journal of Private International Law for 2024 has just been published. It contains the following articles:
Reid Mortensen & Kathy Reeves, The common law marriage in Australian private international law
The common law marriage is a curiosity in the private international law of marriage in the Commonwealth and Ireland. In some cases, a marriage that is invalid under the law of the place where it was solemnised (lex loci celebrationis) may nevertheless be recognised as valid if it meets the requirements of a common law marriage. These originate in the English canon law as it stood in the eighteenth century and include the central requirement of the parties’ present declaration that they are married. The parties also had to meet the essentials of a Christian marriage as described in Hyde v Hyde (1866): “a voluntary union for life of one man and one woman to the exclusion of all others”. Read more
Review of Afifah Kusumadara, Indonesian Private International Law, Oxford: Hart Publishing, 2021, 288 pp, hb $140
After reading and reviewing a thought-provoking book on the choice of law in international commercial contracts in Indonesia last year, I decided to delve further into the subject by picking up a book on Indonesian private international law. The book, titled Indonesian Private International Law, is part of the prestigious Hart series on Private International Law in Asia. Authored by Dr. Afifah Kusumadara, with contributions from a team of Indonesian scholars (hereafter referred to as “the authors”), this work was published during the COVID-19 pandemic. Spanning 226 pages across six chapters, the book aims to be the leading English-language text on private international law in Indonesia. This review provides an overview of its content.
Review of Kazuaki Nishioka, Treatment of Foreign Law in Asia, Oxford: Hart Publishing, 2023, 327 pp, hb £117
It is a great pleasure to review the book titled Treatment of Foreign Law in Asia, edited by Kazuaki Nishioka. This volume contains 17 chapters, including an introduction and conclusion, spanning 298 pages (excluding the series editor’s preface, table of contents, bibliography and index). The book examines 15 Asian jurisdictions, representing a variety of legal systems, including civil law (China, Taiwan, Japan, South Korea, Vietnam, Cambodia, Indonesia, and Thailand), common law (Hong Kong, Singapore, Malaysia, Myanmar, and India), and mixed jurisdictions (Philippines and Sri Lanka). Read more