John C. Coffee Jr. John C. Coffee Jr.

Everyone knows that Morrison v. National Australia Bank, 561 U.S. 247 (2010), ended the ability of those who purchased or sold securities outside the United States to participate in U.S. class actions. Everyone knows this—and therefore, unsurprisingly, it turns out to be not quite true. As usual, people miss the exceptions to generalizations that they think are universal rules. In fact, there are at least two routes that permit a foreign purchaser to participate in a U.S. class action. Neither has been much exploited to date, but both may be pushed aggressively in the near future.

First, although a litigation class cannot be certified unless it can be shown (without an individual inquiry into each class member) that only domestic purchasers or sellers are covered, a settlement class can be broader and can contain non-domestic purchasers or sellers. This was shown last year in In re Petrobras Securities Litigation, 317 F. Supp. 3d 858 (S.D.N.Y. 2018), when Judge Rakoff certified the class and approved a $3 billion settlement against Petrobras, the Brazilian oil company, even though the Second Circuit had earlier reversed his certification of the class because the “domesticity” of various bond purchasers was in doubt. See In re Petrobras Secs. Litig., 862 F.3d 250 (2d Cir. 2017). Because determination of the domesticity of the bond purchasers seemed to require an individualized inquiry, the Second Circuit had overturned Judge Rakoff’s certification of the class, finding that it flunked the “predominance” requirement of Rule 23(b)(3) of the Federal Rules of Civil Procedure (which requires common questions of law and fact to “predominate” over individual questions).

But then how could Judge Rakoff subsequently certify the class for settlement purposes? The answer lies in an aspect of Morrison that not everyone notices. Morrison said that Rule 10b-5 applies only in the cases of domestic transactions, but that this issue of domesticity was a “merits” issue and not an issue of subject matter jurisdiction. That means that, at settlement, the defendant can waive this issue (and Petrobras did). As Judge Rakoff explained, even a hopelessly non-meritorious case can be settled if the defendant is willing to waive “merits” issues.

This idea that settlement classes can be broader than litigation classes opens up a world of possibilities. Suppose for example that a Canadian issuer trades 80% in New York on the NYSE and 20% in Canada on the Toronto Stock Exchange. Next, assume that a settlement is reached of the U.S. portion of this dispute. The defendant might wish to settle the Canadian purchasers’ claims as well in one global settlement, for any of several reasons, including: (1) it wants global peace and an end to costly litigation (where its own attorneys may cost as much as the settlement); (2) if the dispute were to settle instead in overlapping, parallel class actions in the United States and Canada, the defendant might have to overpay because it would be uncertain where claimants will file and many claimants would wait to see which settlement is higher (and some claimants will try to obtain recoveries in both); or (3) the defendant may feel it can overreach the Canadian plaintiffs through collusion by using a friendly plaintiff’s attorney to include them within the global settlement at a price below what they would receive in Canada.

Can a U.S. judge certify and approve such a settlement when a significant percentage of those covered are foreign purchasers? Although Judge Rakoff said that defendants can waive the merits and settle as they want, the Petrobras case he approved did not really test this principle, as it was only unclear where the Petrobras bonds had traded. Going even a step further, consider next an extreme case where a foreign defendant’s stock trades 90% in London, while its American Depository Receipts (ADRs) trade only in the United States and account for the other 10% of the trading. Can a settlement of the claims held by ADRs holders in the United States be used as a vehicle for a global settlement of all claims based on the defendant’s waiver of the domesticity issue? Even if subclassing and separate counsel are used, this outcome seemingly renders Morrison a nullity in the settlement class context. International comity may also be threatened, as this procedure arguably is an end run around U.K. law, which does not recognize a securities class action.

Another legal route to this same end is also possible, but no less controversial. Under 28 U.S.C. §1367, federal courts can hear a case over which they would lack subject matter jurisdiction, but only if the case shares the same nucleus of operative facts as a case before the court over which the court does have jurisdiction. The purpose of conferring such “supplemental jurisdiction” is to encourage integrated resolution of sprawling cases and avoid more costly piecemeal resolution. Its core requirement that there be a common set of shared facts will often (and probably generally) be true with respect to securities litigation, where the key factual issues in both actions are likely to be: (1) the presence or absence of a material misstatement or omission, or (2) the presence or absence of scienter by the defendant. Assume now that a litigation class is brought in the same U.S./Canadian dispute discussed above, and a class of Canadian purchasers seeks to intervene in the U.S. action, based on supplemental jurisdiction, but asserting Canadian law claims. Of course, under §1367(c), the court need not accept a case that presents “novel or complex issues,” but most Canadian class actions will not present “novel or complex” issues. In truth, a Canadian class action (in which scienter and reliance do not have to be shown under Canadian law) is really a simpler version of the U.S. litigation.

Does this attempt offend the principles underlying Morrison? At its core, Morrison seemed most concerned about U.S. law seeking to dominate the world, thereby riding roughshod over principles of international comity. But the Canadian action would be applying Canadian law so there would not be any extraterritorial application of U.S. law.

To date, securities classes of foreign purchasers have met a mixed response from U.S. courts. (I cover these cases in a recent article. See John C. Coffee, Global Settlements: Promise and Peril (April 9, 2019). Those courts that have refused to permit a foreign class to be appended to a domestic U.S. class action have chiefly raised the issue of comity. In addition, they have doubted that the United States has any interest in providing a forum for foreign shareholders to resolve their claims in a U.S. court. For example, in 2018 in In re Mylan N.V. Secs. Litig., 2018 U.S. Dist. LEXIS 52084 (S.D.N.Y. March 28, 2018), the court observed that the United States has only a “minimal interest, if any, in providing a forum to litigate the claims of foreign stockholders under foreign securities laws.”

That seems true enough in the case of a foreign corporation that is merely traded on a U.S. exchange. But what if the corporation is a U.S. company that trades 80% in the United States and 20% on some foreign exchange (such as Canada). In the case of a settlement class, this U.S. company has a clear interest in resolving all the litigation against it expeditiously (both to reduce costs and to place the matter behind it). Thus, if a global settlement can be negotiated, it seems entirely consistent with the purpose of supplemental jurisdiction to allow that foreign purchaser class to join in the settlement before the U.S. court. Principles of comity are also seldom implicated by a settlement class.

Suppose instead that the corporation is defending a litigation class and a class of shareholders who purchased outside the United States wishes to join the fray. Now, there is both an issue as to comity and as to whether the United States has any “interest” in permitting foreign purchasers to use its courts. But such an “interest” may be discoverable. One possible such interest is that asserted by Judge Henry Friendly when he first announced the “conduct” test that long prevailed in U.S. courts prior to Morrison. He posited that the United States had an interest in not allowing itself to be used as a “base for fraud.” See Bersch v. Drexel Firestone, Inc., 519 F.2d 974 (2d Cir. 1975); I.T.T. v. Vencap, Ltd., 519 F.2d 1001 (2d Cir. 1975). Even though that consideration has been rejected by Morrison as a basis for the extraterritorial extension of U.S. law, we are here considering the very different question of what factors should motivate a U.S. court to exercise supplemental jurisdiction. Morrison’s concern about U.S. law being read to rule the world is simply not applicable here because the foreign purchasers will be suing based on foreign law. Thus, a claim that the alleged fraud was planned or implemented in the United States arguably could justify permitting a class of foreign purchasers to join a class of domestic purchasers. In contrast, however, if U.S. domestic purchasers brought suit in the United States against a German corporation listed on the NYSE, a class of foreign purchasers should not be permitted to join this litigation, at least absent some showing that the fraud was planned in the United States.

Supplemental jurisdiction is likely to receive increasing judicial attention in the near future. One basis for this prediction is the 2018 decision of the Israeli Supreme Court in Cohen v. Tower Semiconductor that, in the case of a dual listed stock, the law of the foreign jurisdiction will govern. Thus, even in a suit in Israel by purchasers in Israel against an Israel company that is jointly listed on U.S. and Tel Aviv stock exchanges, the U.S. securities laws will govern. Israel has a large number of companies cross-listed in both the United States and Israel. Because most of the trading in these dual listed stocks occurs in the United States, it would be simpler and more efficient to resolve this litigation in the U.S. courts, rather than to have parallel suits be litigated in the United States and Israel. Certainly, there is no comity problem here because Israel has elected to apply U.S. law, and thus supplemental jurisdiction would not impose foreign law on them. Indeed, if the U.S. litigation eventually settles, it seems predictable that the parties might want to add the claims of Israeli foreign purchasers to the settlement. This could be done either under the Petrobras theory that the defendant can waive a “merits” issue or pursuant to supplemental jurisdiction.

That settlement classes covering a subclass of foreign purchasers may be coming to U.S. courts is not necessarily a good thing that should be accepted routinely. Settlement classes have a long history of being designed to shortchange class members. See Ortiz v. Fibreboard Corp., 527 U.S. 815 (1999). A class of foreign purchasers seems particularly vulnerable (as notice may not reach many of the foreign class members). At a minimum, subclassing and separate counsel should be mandatory.

All in all, for nearly a decade, Morrison has sealed off foreign purchasers in securities cases from access to U.S. courts. But nature abhors a vacuum, and those foreign purchasers seem about to appear on the doorstep of U.S. courts.

John C. Coffee Jr. is the Adolf A. Berle Professor of Law at Columbia University Law School and Director of its Center on Corporate Governance.