Toshiba ADR Investors in a Catch-22

Image by ChatGPT

A recurring challenge in defining the geographic scope of U.S. securities law is how to characterize non-exchange-based transactions in American Depositary Receipts (ADRs). Under the Supreme Court’s Morrison test, such transactions have to qualify as “domestic” to trigger the application of U.S. law. If they don’t, the assumption is that investors would have to litigate any claims for securities fraud under the law of the issuer’s home jurisdiction. But what if they lack standing under that law?

The Ninth Circuit’s recent memorandum decision in the long-running litigation involving Japan’s Toshiba Corporation illustrates the Catch-22 that can result.

Background

In 2015, holders of Toshiba ADRs initiated class action litigation against the company and two of its former officers, alleging a scheme of accounting fraud. The investors brought two sets of claims. The first were under the U.S. Securities Exchange Act of 1934. The second, based on supplemental jurisdiction, were under Japan’s Financial Instruments and Exchange Act.

Toshiba stock is listed only on Japanese stock exchanges; however, U.S. investors can purchase and trade unsponsored ADRs over the counter in the United States. In 2015, through its investment managers, the lead plaintiff—Automotive Industries Pension Trust Fund (AIPTF)—placed a market order for 36,000 Toshiba ADRs with Barclays Capital. Barclays purchased Toshiba common stock in Japan. Those shares were deposited with Citibank in New York, which then issued ADRs. AIPTF paid for the ADRs and transfer of title to them was recorded.

Application of U.S. Law

In arguing that their claims fell within the scope of U.S. anti-fraud law under the Morrison test, AIPTF focused on the fact that the common shares of a foreign issuer and ADRs representing those shares are distinct securities. ADRs represent a beneficial interest in shares of a foreign company, but not legal title to those shares. Moreover, ADRs are issued not by the relevant foreign company but by a U.S. depository institution. As a consequence, AIPTF argued, the Morrison test should be applied not to Barclays’s acquisition of Toshiba’s common shares, but to its own acquisition of the ADRs.

On this view, the transaction giving rise to the claims qualified as domestic. AIPTF had incurred irrevocable liability to take and pay for the ADRs in the United States, and title to the ADRs had transferred in the United States.

The Ninth Circuit panel (Judges Wardlaw, Mendoza, and Johnstone), however, rejected this characterization. It focused on Barclays’s acquisition of the common shares in Japan as the first and necessary step in creating the ADRs. Holding that this acquisition triggered irrevocable liability on AIPTF’s part to take and pay for the ADRs, it concluded that “the later conversion of the shares to ADR certificates domestically is irrelevant to the analysis.”

The Ninth Circuit did not explicitly address the question where title to the securities had passed, which other courts have recognized as an alternative means of testing domesticity. (The district court had, but considered only the passing of title to the common shares, not the passing of title to the ADRs themselves.) Indeed, the Ninth Circuit’s opinion seemed to suggest that incurring irrevocable liability was the only relevant test, and passing of title was simply one factual element relevant to that analysis:

“To determine whether a transaction was domestic or foreign, we look to the location where a purchaser incurred irrevocable liability to take and pay for a security. This is a factually driven inquiry. We look to ‘contract formation, placement of purchase orders, passing of title, and the exchange of money.’”

To digress briefly on the matter of domesticity: If the Ninth Circuit is saying that the place where irrevocable liability is incurred is the only test for the location of a transaction, then this opinion illustrates another split among the circuits in applying Morrison. The Second Circuit’s 2012 opinion in Absolute Activist v. Ficeto established a two-part test for determining the domesticity of a non-exchange-based transaction: either irrevocable liability must be incurred in the United States, or title to the securities must be transferred in the United States. The court’s subsequent decisions have confirmed that they are alternative tests, such that satisfying either one establishes that the transaction occurred within the United States.

A difference between the Second and Ninth Circuit approaches on this issue would dovetail with another split between them in applying Morrison. Under the standard the Second Circuit adopted in Parkcentral v. Porsche (2014), a domestic transaction is necessary but not sufficient to invoke U.S. anti-fraud law: if a plaintiff’s claims are “predominantly foreign,” the application of U.S. law would be impermissibly extraterritorial. The Ninth Circuit, by contrast, rejected the Parkcentral approach in an earlier decision in the Toshiba litigation, holding that establishing a domestic transaction automatically triggers the application of U.S. law. The latter approach lacks the flexibility to address cases like Toshiba involving unsponsored ADRs, where there is a domestic security (the ADR) but no involvement of the foreign issuer in U.S. securities markets. The only way to avoid the application of U.S. law is to conclude that the transaction in question is not domestic.

Application of Japanese Law

In the second half of the opinion, the Ninth Circuit affirmed the district court’s grant of summary judgment for Toshiba on AIPTF’s claims under Japanese law. That decision was based on a recent opinion by the Tokyo District Court regarding the standing of shareholders to initiate claims under the Japanese Financial Instruments and Exchange Act (JFIEA).

Article 21-2 of the JFIEA states that corporate officers who make false statements in securities registration documents shall pay damages to the “persons that acquire securities.” Earlier this year, the Tokyo District Court issued a decision holding that this phrase refers only to registered shareholders—that is, persons whose ownership interest is registered by book entry. When ADRs are created, it is the custodian of the underlying common shares that is the registered owner of those shares, not the beneficial owners who hold the ADRs. As a result, the Ninth Circuit held, AIPTF lacked standing to assert claims under the JFIEA.

This puts Toshiba’s ADR holders in a Catch-22. Because it was the acquisition of the common shares that counted in establishing the location of their purchase, they can’t litigate their fraud claims under U.S. law. But because the ADRs are in fact a different security than the underlying shares, they can’t litigate them under Japanese law either.

Conclusion

My understanding is that the Tokyo District Court’s decision was motivated by the concern that conferring standing on beneficial as well as registered owners might lead to double recovery for securities violations. If that decision stands on appeal, attention will presumably shift to the circumstances under which the custodians of shares can assert claims on behalf of beneficial owners. Without some pathway to litigating potential fraud claims, U.S. investors might become more cautious about investing in unsponsored ADRs of Japanese issuers.