Concept
Content of the Federal Rule
Federal common law
Content of the Federal Rule
Clearfield Trust Co. v. United States held that federal common law may govern an issue implicating uniquely federal interests even though no federal statute directly controls, so that the negotiability of checks payable by the United States is decided by a uniform federal judge-made standard rather than by varying state laws. That holding rests on two decisions, and students tend to run them together. The Court first decided that federal law, and no state's law, would govern the government's rights on its own paper. It then decided what that federal law would say, and it chose one national standard, because the government's checks pass through several states and need the same answer wherever they are cashed. The earlier sections of this chapter explained the first decision. This section explains the second.
The two decisions answer different questions. The first is one of competence: do federal lawmaking power and a uniquely federal interest exist, so that a federal court may displace state law as the source of the rule of decision? The second is one of discretion: with federal law governing, should the court fashion a nationally uniform rule from federal sources, or should it adopt state law as the federal rule? The questions come apart because a federal interest can call for federal control of a subject without calling for a rule different from the state's. If the interest is only that federal law, not state law, settle the question, the state's existing answer may serve as well as any new one. If the interest is uniformity across all the states, deferring to each state's own law defeats it, because the states differ.
The cases that follow show both outcomes. Clearfield chose a uniform standard. The preclusion rule for diversity judgments, discussed earlier, went the other way: a federal rule was needed there, but not a uniform one, so the federal court took its content from state law.1 A rule borrowed in that way is still federal law, which is why the second step cannot be skipped by assuming that state content means state authority. The first subsection below follows a court that builds a uniform rule. The second follows a court that borrows. The third explains why a borrowed rule differs from the state law that governs of its own force under Erie.
Fashioning a Uniform Rule
Once Clearfield settled that federal law governs the government's rights on the checks it issues, the Court still had to say what that law provides, and it did not write the rule from nothing. It turned to the federal law merchant that federal courts had developed for about a century under Swift v. Tyson and called it “a convenient source of reference for fashioning federal rules applicable to these federal questions,” while acknowledging that those rules were general commercial law rather than a choice of a federal rule designed to protect a federal right.2 Deciding which lawmaker controls and deciding what the controlling rule says are therefore separate steps.
When no federal statute supplies the rule, the court must make it, and the making is disciplined. The court considers the policies relevant to the field, basic principles of equity, and rules of law established in analogous fields.3 It also weighs scholarly writing and other considerations, so the rule that results is a creative enterprise but not an unguided or arbitrary one.4 The discipline matters because a federal rule displaces state law and binds state courts as well as federal ones, and a rule drawn from a developed body of law arrives with its consequences already known.
The need for uniformity justified a federal rule in Clearfield, and it also directed the search for the rule's content. The Treasury issues commercial paper on a vast scale, and transactions in that paper commonly occur in several states, so applying state law would leave the government's rights and duties in exceptional uncertainty and would make identical transactions come out differently under the laws of the several states.5 A national body of commercial law answers that problem, which is why the Court looked to the law merchant and passed over Pennsylvania's rule. The Court even suggested that the pre-Erie law merchant could serve as the starting point for federal common law in this field.6
How far the search extends depends on whether state law has any claim to apply. Where states are categorically barred from regulating the subject, as with foreign relations, the court has no state rule to weigh and simply finds an appropriate substantive rule.7 Where state authority exists, the first question is whether the federal rule should incorporate the state rule, the possibility taken up in the next subsection.8
The national body of commercial law that courts consult most often today is the Uniform Commercial Code, and they use it as Clearfield used the law merchant. Congress has enacted the Code for the District of Columbia but not for federal transactions, so absent a controlling federal statute or regulation the federal presence makes federal common law the governing law, and state law is not a choice.9 The Code therefore furnishes content only if a court chooses to adopt it, and many courts have drawn heavily on it because it controls most commercial transactions in the United States.10 In United States v. Wegematic Corp. the court reasoned that a Code enacted for the District and in over forty states was well on its way to becoming a truly national law of commerce, so insisting on a different law for transactions with the United States would be a “distinct disservice.”11 That reasoning follows the line Clearfield drew. A drawee that is the United States stands in no different light than any other drawee and does business on business terms, and a rule drawn from ordinary commercial law keeps those dealings in step with the rules everyone else follows.12
A source informs the court's choice without commanding it, and courts have set the Code aside where a different federal policy applies. Where two navy enlisted men furnished the payee's name on a Treasury check and then indorsed and cashed it, the court refused to apply the Code policy that would have placed the loss on the United States rather than on the cashing bank. In United States v. Philadelphia National Bank, the court added that the United States arguably has more difficulty insuring against loss and controlling its employees.13 Courts have likewise departed from the Code where a federal deposit insurance agency enforces an instrument. The rule of D'Oench, Duhme & Co. v. Federal Deposit Insurance Corporation was derived from statutes reflecting a policy of protecting the agency and the public funds it administers against misleading appearances created by undisclosed arrangements, and it estops a person who executes an instrument in the form of a binding obligation from asserting, as a defense, a simultaneous secret agreement that the instrument would not be enforced.14 A court deciding a dispute over federal paper therefore neither must apply the Code nor is barred from consulting it. The Code supplies persuasive content, and that content yields to a contrary federal policy.
United States v. Standard Oil Co. of California shows the second step ending with no rule in the government's favor. In February 1944 a soldier, John Etzel, was hit by a Standard Oil truck at a street intersection in Los Angeles. The government paid $123.45 for his hospital care and continued his pay of $69.31 during his disability, and Etzel released the company from his own claims for $300.15 The United States then sought to recover those outlays, and the value of the soldier's lost services, from the company as a tortfeasor. The Court concluded that state law should not be selected as the federal rule for that question.16 The earlier section on the rights of the United States explains why federal law governed. The government nonetheless lost, because the Court declined to create the right it asserted.
The Court took that step with open reluctance. Its opinion begins by observing that the Court is not often asked, since Erie, to create a new substantive legal liability without legislative aid and as at the common law.17 The Court did not doubt its authority to create a right of indemnification. It noted, however, that the issue arises frequently, that Congress had never created such a right by statute, and that adopting one by judicial decision would upset settled expectations, so on balance it seemed wiser to defer to Congress.18 The restraint has a structural basis. A court that borrows a developed commercial rule adopts content whose effects are already known, whereas a court that creates a new liability reassigns losses among parties who arranged their affairs without it. The question here was chiefly one of federal fiscal policy, a field in which the legislature is better placed than a court to decide who should bear the cost.19
Clearfield and Standard Oil pair as two claims by the United States that federal law governed and that came out differently at the second step. In Clearfield the government's right of recovery already existed in the law merchant, which treated the drawee's right against a presenter of a forged check as accruing when payment was made, and the Court's task was to decide what effect delay in giving notice should have on that right.20 In Standard Oil nothing in federal law gave the government a claim to indemnity, and recovery would have required the court to invent the right itself. A holding that federal law governs the government's claim is not a holding that the government recovers.
Borrowing State Law
A court that decides federal law should govern an issue may still take the content of its rule from state law. The resulting rule is federal in authority and state in content, which differs from leaving the issue to state law altogether. The court may choose that course where state law would serve the relevant purpose without frustrating the objectives of federal law, and the choice is appropriate where a federal rule is needed but need not be uniform in all federal cases.24
Clearfield shows the opposite choice, and the reason lies in a feature of the government's paper that many other federal questions lack. The same check can be cashed in any state, so a rule taken from one state would have let the government's rights change with the place of cashing. Where that feature is absent, borrowing becomes the sensible option.
United States v. Kimbell Foods shows the choice coming out the other way. The government sued to enforce contract-based security interests in certain property and claimed that federal common law gave its liens priority over those of private creditors, whereas state law would have ranked the government with other creditors according to the nature of its interests and the date on which each lien was perfected.25 The Court accepted that federal common law applied because the dispute involved the government's proprietary interests, yet it used state law to decide priority. The government sought an advantage that could well upset the expectations of private parties who had presumably looked to state law to determine priority and made their lending decisions on that basis.26 A lender prices a loan by where its lien will rank, so a federal priority rule announced after the loan would change the value of security the lender had already accepted. Adopting the state ranking protected that reliance and left the government in line with every other lienholder.
Kimbell is also the decision that spells out a two-step process. The Court first determines whether federal interests justify the creation of federal law, and only then considers whether to fashion a nationally uniform rule or to look to state law instead.27 The steps are often described as questions of federal competence and federal discretion: at the first stage the court asks whether federal interests, from whatever source, make it competent to fashion federal common law, and at the second stage it exercises discretion over the content of the rule.28
The division answers a criticism of Clearfield, which was said to have considered only competence and to have slighted the interests of the states, and Kimbell conducts the analysis explicitly in two steps.29 Hold the structure loosely, though, because it organizes the decisions and is not a test the Court recites in every case. In Boyle v. United Technologies Corp. the Court framed its inquiry as the preemption of state law and cast doubt on the need for a separately stated second step.30 Whatever the labels, the underlying question remains: once a federal interest is found, does that interest require its own rule, or can the state's rule serve?
The cases do not reduce that question to a formula, but decisions to adopt state law rest on a recurring set of considerations, and any of them can support borrowing.31 They are the following:
- Congressional intent, as expressed in a statute or its legislative history.
- A substantial state interest that incorporating local law would serve.
- Federal and state law that are closely interwoven on the subject.
- The furtherance of equity.
- Convenience, when federal law is silent on the particular issue.
Each consideration works through a mechanism that the decisions make visible. Congress sometimes dictates the answer itself, as in the Federal Tort Claims Act, which makes the government liable for the torts of its agents “in accordance with the law of the place where the act or omission occurred.”32 In that setting the state rule controls regardless of the federal interest in the matter.33 A substantial state interest and equity drove Kimbell, since lenders had ordered their affairs by state priority rules. Convenience matters most when a federal statute is silent: statutes commonly create a right of action without stating when it must be brought, and the usual rule is to apply the limitations period the state provides for analogous suits.34
De Sylva v. Ballentine combines the state-interest and interwoven-law considerations. The question was whether the illegitimate son of a copyright owner is one of the owner's “children” within the meaning of the copyright statute.35 The Court recognized that the meaning of the word in the statute is a federal question, but because there is no federal law of domestic relations and the matter is primarily one of state concern, it drew on “the ready-made body of state law” to define what Congress meant.36 The next subsection returns to the control the Court kept over that borrowed definition.
The same approach appears wherever a federal rule is needed but the rule need not be uniform in all federal cases. In Kamen v. Kemper Financial Services, Inc. the Court incorporated state law as the federal rule of decision in a federal shareholder derivative action, filling a gap about the demand requirement under the Investment Company Act.37 In Semtek International Inc. v. Lockheed Martin Corp. the Court held that federal common law governs the claim-preclusive effect of a federal diversity dismissal and then declined to displace the law of the forum state in determining that effect, adding that this federal reference to state law will not obtain where the state law is incompatible with federal interests.38 Incorporation has two benefits that explain its appeal: it preserves uniformity between state and federal law, and it permits departure when necessary to protect federal interests.39 The second benefit keeps the federal court in charge of the result, because borrowing is a choice made on the strength of state law's fit, and a court that finds incompatibility need not follow the state rule.
United States v. Yazell shows borrowing in a suit the government itself brought. The United States sued Mrs. Yazell to recover under a loan contract negotiated between her husband and the Small Business Administration, and the Court barred recovery by applying the state's law of coverture, which still prevented married women from binding their property without a court order.40 The Court emphasized that the contract was negotiated and entered into individually and drafted with state law clearly in mind, and it did not decide whether state law applied because the Rules of Decision Act required it or because the Court adopted it as a federal principle.41 That setting of a single negotiated contract is the opposite of Clearfield, where thousands of government checks were issued, and of Bank of America v. Parnell, where thousands of government bonds were involved.42
The hinge in these decisions is whether the federal interest depends on one answer across a mass of transactions. In Clearfield it did, because a state's rule adopted as the federal rule would have given the government's checks different consequences in different states. A single contract written with local law in view creates no comparable need. A holding that federal common law governs therefore does not require a new uniform federal rule, as Kimbell shows by ending that holding in the state's own rule.
The trend in the Court's decisions points the same way. In the area of federal liens, lower courts applied federal rules, while the Supreme Court's decisions, culminating in Kimbell, increasingly looked to state law as the applicable standard.43 Convenience in having one rule does not justify a supreme federal rule, even one that borrows from state law. In O'Melveny & Myers v. FDIC a federal agency acting as receiver for a failed savings and loan sued a law firm for malpractice. The Court held that state law must govern, even if applying state principles might produce varying results in similar cases brought in different forums, because the case lacked any “significant conflict with an identifiable federal policy or interest.”44 O'Melveny shows the first step failing for want of a federal interest, while Kimbell shows the second step ending in borrowing, and both reflect a reluctance to build a federal rule on the appeal of uniformity alone.
The Court exercises its power to make federal common law with reticence, and it has put the default in general terms. In American Electric Power Co. v. Connecticut it said that “[a]bsent a demonstrated need for a federal rule of decision, the Court has taken the prudent course of adopting the readymade body of state law as the federal rule of decision until Congress strikes a different accommodation.”45 The statement places the burden on the party who wants a distinct federal rule, since state law is already available and its consequences are already known to the people who have organized their affairs around it. The closing phrase also marks the borrowed rule as provisional, because Congress can strike its own accommodation at any time, a point the final section of this chapter takes up.
Borrowed Law Versus Erie
A federal court applying a state's rule under Erie and a federal court adopting the same rule as federal common law end up reciting the same words, and the same sentence may even decide both cases. The two situations differ in the source of the court's authority, and that source decides how much freedom the court has to interpret, select, and adjust what it borrows.48
When Erie supplies the rule, state law applies of its own force, because the federal court has no competence to apply anything else. The court therefore functions, in effect, as another court of the state, and its task is to reproduce the result a state court would reach, so that the choice of a federal forum does not change the outcome.49 Three consequences follow. The court is bound by the decisions of the state's highest court, it cannot substitute a rule it considers wiser, and it cannot decide for itself which state's law applies, since it is supposed to mirror the state's own view of the question.50
Adoption as a federal rule reverses each of those points. The court has already decided, on federal grounds, that federal law governs, and the decision to incorporate a state's rule is its own judgment about what the federal rule should say. The district court is therefore not just another court of the state, it is not bound by earlier state-court interpretations of the borrowed law, and it has much greater flexibility in working out the substance of that law.51 The reason lies in who made the choice. A court that applies state law because it lacks power to do otherwise leaves the content and limits of the rule to the state. A court that incorporates state law after deciding that federal law controls keeps control over the intent and the methods of the adoption, because the borrowing is a federal judgment about fit.52
De Sylva v. Ballentine, introduced above, shows that freedom in use. The Court borrowed “the ready-made body of state law” to define what Congress meant by “children,”53 yet it did not hand the word over to the state. A state would not be entitled to use “children” in a way entirely strange to its ordinary usage, so the state definition was adopted only to the extent of the permissible variations in the ordinary concept.54 The Court also indicated that a federal court adopting state law is free to decide which state's law applies, though only one state was involved and no choice was needed.55 Semtek International Inc. v. Lockheed Martin Corp. follows the same pattern: the claim-preclusive effect of a federal diversity judgment is a question of federal common law, and its content comes from the law of the state in which the federal court sat.56
The judge in De Sylva and the judge in that scene both consulted a state's law, but the first exercised a choice that federal law allowed, and the second followed a rule that state law imposed. The difference reaches beyond the courtroom. State courts are bound by federal-court decisions on matters of federal common law, even when those decisions deviate from state law or from the state's interpretation of it, whereas a federal court's reading of state law in an Erie case gives state courts no such obligation.57 It may look odd that a court can find a uniquely federal interest and then borrow the very state law that the finding displaces. The two steps fit together because federal common law rests on a law or policy of Congress that invites the federal courts to develop rules, and borrowing from a state is one way of carrying out that invitation.58
Treating borrowed content as if it carried Erie's authority is therefore a mistake with two costs. It would bind a federal court to state-court decisions that the court is free to depart from, and it would leave state courts free to ignore the federal court's decision.