Marc J. Goldstein Arbitrator & Mediator NYC
November 17, 2009

The Fifth Circuit’s Reverse Preemption Decision: An “Upset Victory” for International Arbitration

An arbitration lawyer immersed in the U.S. jurisprudence of the New York Convention, but not necessarily versed in all of the nuances of the Supremacy Clause of the U.S. Constitution, might be tempted to conclude that international arbitration scored a remarkable upset victory, against long odds, in the U.S. Fifth Circuit Court of Appeals last week. (Safety Nat’l Cas. Corp. v. Certain Underwriters at Lloyd’s, 2009 U.S. App. LEXIS 24585 (5th Cir. Nov. 9, 2009)).

By a vote of 15 to 3, the judges of that Court sitting en banc ruled — sustaining the original decision of a three-judge Fifth Circuit panel — that the Convention preempts and nullifies a provision in the Louisiana state insurance law requiring that insurance disputes be resolved only by judges and juries.

To reach that outcome, the majority needed to overcome what is perhaps the most notorious “reverse preemption” statute in U.S. law, the McCarran-Ferguson Act. That Congressional enactment, from 1945 — 25 years before the United States implemented the New York Convention — provides that “No Act of Congress shall be construed” to invalidate any state law enacted “for the purpose of regulating the business of insurance” unless the impairment is by an Act of Congress that “specifically relates to the business of insurance.” In simplest terms, McCarran-Ferguson states that only federal statutes dealing specifically with insurance can preempt state insurance regulation.

Clearly the Convention in its inception was not an “Act of Congress,” it was an international treaty. Its implementing legislation, Chapter Two of the Federal Arbitration Act (sometimes called the “Convention Act”), surely is an “Act of Congress.” But is the Convention, once implemented, an Act of Congress for domestic law purposes, or does it have from that point forward the same force of law domestically as would a treaty that is “self-executing,” i.e. that did not require implementing legislation to have domestic legal effect?

The question before the Fifth Circuit was, in essence, this: To reach the conclusion that the Convention’s command to enforce arbitration contracts preempts Louisiana’s law denying enforcement of such contracts in the insurance industry, must the federal courts “construe” an Act of Congress, FAA Chapter Two, and thus collide with McCarran-Ferguson — or do they merely “construe” the Convention itself, such that McCarran-Ferguson does not come into play?

The dissenting opinion in Safety National methodically demonstrates that the New York Convention has the force of law in the United States only by virtue of its implementation by the Convention Act, and that courts, when they preempt state laws based on treaties that required implementation by Acts of Congress, have consistently stated that they were assigning the preemptive effect to the implementing statutes, not the treaties themselves. The dissent reasons further that the phrase “shall not be construed to invalidate,” in McCarran-Ferguson, does not refer to a process of interpreting the words of an Act of Congress, but rather means simply that Acts of Congress shall not be given legal effect to preempt state law (unless the Act regulates insurance explicitly).

This makes a great deal of sense. And so those of us inclined the handicap whether the Supreme Court of the United States might seize an opportunity here for an important New York Convention decision review carefully the majority opinion to see if its answers to the dissent are convincing.

The majority’s lead-off position is that implementation of the Convention by the Convention Act does not transform the Convention from a treaty into an “Act of Congress.” But this position would seem to beg the fundamental question: whether for McCarran-Ferguson purposes what is being “construed,” in case of a treaty implemented by Act of Congress, is the treaty as a stand-alone legal unit, or the treaty as an incorporated element of the implementing statute.

On this point, the majority declares that “it is the treaty (the Convention), not an act of Congress (the Convention Act), that we construe to supersede Lousiana law.” But one reads on in the majority opinion yearning to be convinced that “construed” in McCarran-Ferguson does not mean “to be assigned preemptive effect,” as the dissent convincingly posits. The majority relies heavily on the notion that the Convention Act merely declares that the Convention shall be enforced, and hence any “constru[ing]” is done in relation to the treaty not the statute. But this reasoning fails to address whether, when a treaty requires implementing legislation to have effect as domestic law, it retains or acquires a domestic legal status after implementation that is separate and distinct from the implementing statute.

The majority defends its position with extensive and rather speculative discussion about what the Congress might have intended, concerning treaties implemented by statute, when McCarran-Ferguson was enacted. But it seems most probable that Congress thought the term “Act of Congress” was not ambiguous and that, treaty implementation having long been a familiar species of legislation, such statutes were among those to which McCarran-Ferguson makes reference.

The majority then states that “[o]ur conclusion that referral to arbitration is proper in this case is bolstered by the congressionally sanctioned national policy favoring arbitration of international commercial agreements.” Indeed. “Bolstered” puts the case mildly; one might well say the majority’s analysis is driven by that policy. But as there is no jurisprudentially-sound basis to say that the pro-arbitration policy of the Convention has priority over the pro-state regulation of insurance policy of McCarran Ferguson, this statement on the majority opinion adds little to its persuasive force.

The persuasive force of the dissent in Safety National, coupled with an apparent split in the Circuits based on a Second Circuit decision in 1995, makes this case a ripe prospect for the grant of certiorari by the Supreme Court. The case will surely be closely-watched for the filing of a certiorari petition and the disposition of that petition.

It will be interesting to see whether the Lloyd’s underwriting syndicate that prevailed in the Fifth Circuit would elect to rely mainly on the majority’s rationale if the case does reach the Supreme Court. The majority opinion did include an elaborate footnote indicating that a strong case could be made that the forum-selection provision of the Louisiana statute is simply not a regulation of insurance within the meaning of McCarran-Ferguson. This could be the path of less resistance to a pro-arbitration solution in the Supreme Court.

Leave a Reply