When Courts Make Federal Law

When Courts Make Federal Law

In Erie, Justice Brandeis wrote that “[t]here is no federal general common law.” On the same day, in another case, he wrote for the Court that whether the water of an interstate stream must be apportioned between two states is a question of “federal common law” on which neither state’s statutes nor decisions can be conclusive.¹ Erie therefore ended one kind of federal judge-made law and left another in place, and federal common law continues to exist despite the common misperception of what Erie said.²

Clearfield Trust shows what remains. In 1943 the Court considered whether state or federal law controlled the effect of delay in notifying a prior endorser that an endorsement on a federal check was forged.³ Pennsylvania law barred the United States from recovering against Clearfield Trust because of the government’s “unreasonable delay,” and the district court, applying Erie, followed Pennsylvania law.⁴ The Supreme Court took a different view. Reciting the great federal interest in the uniformity of law governing federally issued paper, it held that federal law would govern, and it applied a federal common-law rule it had announced decades before Erie.⁵

That holding raises a worry. No federal statute answered the question, so if silence in the statute books were enough to let a federal court write its own rule, federal judges could decide any question that came before them, and the general federal common law that Erie rejected would return.⁶ Two requirements prevent that result. Federal courts create federal common law when power to do so exists and a federal interest makes federal lawmaking appropriate.⁷

The requirements ask different questions. Power asks whether federal legislative authority reaches the subject at all. Interest asks why a federal rule, rather than state law, should decide the case. The second question arises because state law is the default. The Rules of Decision Act directs that state laws “shall be regarded as rules of decision” in federal trials at common law except where the Constitution, treaties, or federal statutes “otherwise require or provide,”⁸ and the Act is correctly read to make state law, statutory or common, govern on substantive matters.⁹ A court that departs from the default needs a dominant federal interest that justifies displacing the state law that would otherwise govern.¹⁰

The sections below take the requirements in order. A corollary follows them, because a grant of jurisdiction over a kind of case does not by itself authorize federal courts to make substantive law for it. Later sections turn to the content of the federal rule and to displacement by Congress.

Federal Lawmaking Power

Federal courts have power to declare independent federal common law, on substance and procedure alike, in areas in which there is federal legislative authority.¹¹ That is the power requirement, and it works as a threshold. A subject that federal legislative authority does not reach is closed to federal judicial lawmaking, however useful a uniform rule might be.

Erie explains the threshold. Federal courts have no power to create a body of general federal common law governing substantive rights, because that power is reserved to the states under the Tenth Amendment.¹² In Erie the federal government lacked the power to create rules of tort and property law.¹³ Clearfield Trust distinguished Erie on exactly that ground. “[W]hen the United States disburses its funds or pays its debts, it is exercising a constitutional function or power,” and the Court emphasized that “[t]he authority to issue the check had its origin in the Constitution and statutes of the United States.”¹⁴ That authority was “in no way dependent on the laws of Pennsylvania.”¹⁵ Federal legislative authority reached the government’s own checks, so a federal court that filled the gap acted within federal power.

Federal lawmaking authority has identifiable sources. Congressional legislation can federalize an area, and the Constitution can allocate power to the federal courts or to another branch of the federal government.¹⁶ The Sherman Act illustrates the first. Congress’s command there is so broad and indefinite that it virtually invites the federal courts to supply much of the statute’s substantive content, and after 130 years the courts, acting in a manner not strikingly different from a common-law court, have provided much of the Act’s meaning.¹⁷ Admiralty illustrates the second. The constitutional grant of admiralty jurisdiction implies authority in the federal courts, exclusive of state authority, to evolve and apply a national substantive law, and the courts are expected to do so where Congress has not acted.¹⁸ Suits between states are the other most prominent area. It would have been impracticable to use state law there, and Congress furnished no substantive law to speak of.¹⁹ More recently the Court has gone beyond those two areas, and Clearfield Trust is the leading example.²⁰

The same test sorts the rules that federal courts announced while Swift v. Tyson was in force. Some of those rules involved areas where the Constitution granted no federal power, and Erie threw them out. The remainder involved areas with a constitutional “hook,” and after Erie their characterization changed from general law to federal law.²³ Some of today’s federal common law therefore comes from cases that the federal courts decided during Swift’s reign.²⁴

Power is only the threshold, and a court that finds it has answered half the question. The existence of power does not mean that it will be exercised in every case, and a court that does exercise it may adopt state law as the federal standard instead of announcing a new federal rule.²⁵ Restraint also protects the policies of Erie, which aim at results that do not differ materially between state and federal courts.²⁶ The Court made the point in Wallis. It is “by no means enough that, as we may assume, Congress could under the Constitution readily enact a complete code of law governing transactions in federal mineral leases among private parties.”²⁷

A Uniquely Federal Interest

Power tells a court that it may supply a rule. Interest tells it whether it should. The Supreme Court has framed the test in several ways, and each asks whether a federal concern is strong enough to displace the state law that would otherwise govern. On any vertical choice-of-law question, the inquiry is whether some dominant federal interest justifies that displacement.³³ In Wallis the Court required that “a significant conflict between some federal policy or interest and the use of state law in the premises must first be specifically shown.”³⁴ The Clearfield approach adds a second element to federal competence. Competence alone would not preempt equally valid state law, so preemption in favor of federal decisional law requires a finding that a strong need exists for national uniformity.³⁵

The Wallis Court applied its own test strictly. Because it found no significant threat to any identifiable federal policy or interest, it did not press on to the other questions that bear on federal common law, such as the strength of the state interest in having its own rules govern and the feasibility of creating a judicial substitute.³⁶ The conflict showing therefore comes first, and a federal connection that creates no conflict ends the inquiry.

The government’s checks satisfy both showings. Federal competence is clear, and the Court recited the great federal interest in the uniformity of law governing federally issued paper.³⁷ The rights and duties of the United States on commercial paper it issues are governed by federal rather than local law.³⁸ The interest is distinctively federal because the dispute concerned the government’s own rights on its own commercial paper, and the holding did not turn on the government’s status as a party.³⁹

Interstate disputes show the same logic from another direction. When the action of one state reaches into the territory of another, the extent and limits of the two states’ rights become a justiciable dispute, and the Court settles it in a way that recognizes the equal rights of both and establishes justice between them.⁴⁰ Using the law of either state would have been impracticable, and Congress furnished no substantive law to speak of.⁴¹

Disputes between private parties can supply the interest too. Despite concurrent state lawmaking power, the Court has sometimes decided that the federal interest in private litigation is great enough to warrant federal common law, and it typically finds the interest sufficient only when application of state law would interfere with some obvious federal purpose.⁴² Boyle v. United Technologies Corp. is an instance.⁴³ Bank of America v. Parnell and the other private-party cases come in a later section, and Wallis already marks the contrast. Leases among private parties, with no identifiable federal policy at risk, left state law in place.

In a case that reaches federal court only on diversity, the test takes a more structured form. State law will usually govern unless the matter is in an area of federal lawmaking authority and two further conditions are met.⁴⁴ First, a significant conflict must be specifically shown between some legitimate federal policy or interest and the use of state law on the specific point in issue. Second, the court must determine that an overriding necessity for uniformity requires a uniform federal rule. The second condition balances the two sides. It asks about the degree of state interest, the degree of federal interest, and the feasibility of a judicial substitute, including the ease of formulating a rule and the added burden on the court in administering it.⁴⁵

Beyond those showings, the decisions turn on a cluster of considerations that resist a formula. Among other factors, the Court has focused on these:⁴⁹

• the nature of the federal interest;
• constitutional implications;
• relevant congressional actions or silences;
• the likely impact of a proposed federal rule on the interests of the states; and
• the workability of relying on state law as an alternative.

The resulting decisions do not form a predictable pattern. Federal common law arises here and there for reasons that one can rationalize in retrospect but cannot often predict in advance. Within the enclaves where it is established, lower federal courts and state courts work out answers to new questions, while in fields it has not reached, courts proceed with greater caution and greater deference to state interests.⁵⁰ That caution matches the Court’s description of the role of federal common law. In Rodriguez v. FDIC the Court said that judicial lawmaking in the form of federal common law “plays a necessarily modest role under a Constitution that vests the federal government’s ‘legislative Powers’ in Congress and reserves most other regulatory authority to the States.” Absent a demonstrated need for a federal rule of decision, the Court has taken the prudent course of adopting the ready-made body of state law as the federal rule of decision until Congress strikes a different accommodation.⁵¹

Jurisdiction Alone Is Not Enough

Article III reaches, among others, admiralty cases, diversity cases, and controversies to which the United States is a party.⁵⁴ A court told to hear a kind of case can look as though it has also been told what law decides it, but it has not. The fact that the Constitution confers jurisdiction over a kind of case does not, by itself, imply that it has empowered the federal courts to make substantive law for those cases. The key inquiry is whether the area is one in which the federal government has a distinctively important interest.⁵⁵

Before Erie, jurisdiction alone could justify an exercise of lawmaking authority. After Erie it no longer sufficed, and the Court had to explain how far federal common law remained valid within its established enclaves and to justify any new ones.⁵⁶ If a grant of jurisdiction were enough, every category of cases would bring its own judge-made law, and the general common law that Erie rejected would return through the side door.⁵⁷

Diversity jurisdiction supplies the negative example. Justice Brandeis made clear in Erie that the constitutional grant of diversity jurisdiction was not in itself an adequate basis for judicial lawmaking power.⁵⁸ The purpose of the grant explains why. Diversity jurisdiction was evidently created to provide a neutral forum, not uniform substantive law, and a neutral forum is fully served whichever state’s rule the court applies.⁵⁹ State law therefore continues to govern on substantive matters.⁶⁰

Admiralty reaches the opposite result from a grant that looks parallel on the page. In Southern Pacific Co. v. Jensen the Court ruled that the constitutional grant of admiralty jurisdiction gave the federal courts, and Congress, power to develop a uniform body of substantive federal maritime law, which would constitute part of “national law.”⁶¹ The federal courts are expected to develop and apply that law where Congress has not acted.⁶² The usual explanation is that national uniformity in maritime matters was in the minds of the drafters of Article III and of the Judiciary Act of 1789, while diversity was created for a different purpose.⁶³ The dissent in Jensen argued that Article III speaks only of establishing jurisdiction and does not prescribe the substantive law that governs, and Erie later treated diversity in much that way. The Jensen majority view has nevertheless been reaffirmed since Erie, and federal statutory and common law governs in admiralty cases.⁶⁴ What separates the two grants is the reason each exists, not the form in which each is written.

The third grant is the one that Clearfield Trust puts to the test. Article III authorizes the federal courts to hear controversies to which the United States is a party, and the judicial code has long provided jurisdiction over actions the United States brings as plaintiff.⁶⁵ One might think that any case in which the United States is a party implicates a uniquely federal interest, and that this is why Article III confers party-based jurisdiction.⁶⁶ The Court nevertheless placed little weight on the simple fact of jurisdiction.⁶⁷ The holding rests on the distinctively federal interest in the rules governing the rights and duties of the United States on its commercial paper, and not on the fact that federal jurisdiction was based on the government’s being a party.⁶⁸

The Court may well have been on solid ground in declining to rest on party status. The United States may be formally a party to a case whose substance does not implicate any important or uniquely federal interest.⁶⁹

The agency in that scene and the Treasury in Clearfield Trust stand on the same jurisdictional footing. The difference lies in the substance of the dispute. The Treasury’s own check implicates an interest that justifies a federal rule, and a boundary line between private neighbors does not. Jurisdiction opens the courthouse door in both cases and supplies the rule in neither.

The three grants share a pattern. A jurisdictional grant is evidence of lawmaking authority only to the extent that its purpose reveals a distinctively important federal interest. Admiralty’s uniformity purpose reveals one, diversity’s neutral-forum purpose does not, and party status does so only when the dispute itself touches a federal function such as the government’s commercial paper. When a fact pattern recites that a federal court has jurisdiction, the next question is therefore what the grant was meant to accomplish and whether the dispute implicates an interest of that kind. The next part surveys the fields where the showing has been made, among them the rights of the United States, interstate disputes, admiralty, and foreign relations, where Banco Nacional de Cuba v. Sabbatino arose.