Erie Doctrine Summary
I. Federal Rules (REA)
REA (28 USC 2072) is based on the Supremacy Clause of the Constitution. It gives the
federal courts the power to use the FRCP as long as they do not violate the Constitution
or the Act itself, or as long as the Rules do not “abridge, enlarge, or modify any
substantive right.”
The major issue is whether the limiting condition in the REA has any independent bite.
Warren concludes no – incidental effects on substance are okay so long as rule-makers
intend to regulate procedure. The REA functions, in this view, as a simple delegation
from Congress.
After Hanna, however, many academics – and judges – began to question whether
Warren is right. For example, the First Circuit in Marshall v. Mulrennin refused to apply
then Rule 15 in a situation where it would have deprived a litigant of the right to reach
the merits where the state would have [Link]. The rule-makers amended Rule 15 in
response to Marshall v. Mulrennin.
Marshall v Mulrennin basically adopted Harlan's viewpoint in Hannah that a Rule must
be assessed in context. Still, the US Supreme Court has never invalidated a Rule. Instead,
most work is done at the level of determining whether a Rule (and a statute, see Stewart
v. Ricoh) is on point and really clashes with state law. Although the Court’s cases say to
use the plain meaning approach, in practice, courts try to interpret a Rule to avoid
problems. See Gasperini re Rule 59: "Federal courts have interpreted the Federal Rules
…with sensitivity to important state interests and regulatory policies." Also Semtek, a
case we do at the end of the semester, states explicitly that it is interpreting Rule 41 to
avoid Erie problems. Similarly, Stewart v. Ricoh: "A broad reading that would create
significant disuniformity between state and federal courts should be avoided."
II. When There Is No Rule or Statute on Point (RDA)
The first, crucial thing, to remember is that the state rule is not automatically the default
in these situations because a litigant could always argue for a contrary federal judicial
rule, either a pre-existing one or one advocated for this very litigation. Here is where the
real problems begin.
Note that York, Byrd, Hanna dictum, Harlan concurrence, and Gasperini all co-exist and
no case has explicitly overruled the prior, even though they reflect different approaches.
A. These cases can be divided into two branches:
Branch One. The first is the litigant-oriented approach of York and the Hannah dictum, a
refinement of the outcome determination test, which purports to interpret the phrase
“rules of decision” in the RDA in light of two of the three policies of Erie: forum
shopping and litigant inequality, when substantial variation in applicable rules affecting
the result occurs as a result of the fortuity of diversity (of who is involved in the accident,
contract, etc.)
Branch Two. Harlan's concurrence and Byrd, by contrast, emphasize the third policy of
Erie, not thwarting local state policies – i.e., federalism concerns. Harlan's concurrence
sticks closest to the constitutional concerns in Erie (reserved powers of the states). Byrd
draws not just on primary conduct but also on state policies and analyzes state procedural
rules for substantive interests they might reflect. It asks whether the state rule, whether
denominated by the state as a procedural rule or not, reflects substantive state interests
and whether the federal procedural rule reflects strong federal interests with respect to the
shape and character of the federal judiciary. This branch says the evil of forum shopping
is grossly overstated and too far afield from Erie's original purpose.
B. Gasperini: Effort to Reconcile the Two Branches
Gasperini is a hybrid and close to a reconciliation. NY has ‘deviates materially standard’
for assessing jury damage awards and remitting them as part of its tort reform policy,
which it wants trial courts and appellate courts, exercising independent judgment, to
enforce. This clashes with federal procedural practice of ‘shocks the conscience’
standard, with 7th Amendment reexamination clause, and potentially with Rule 59.
Ginsburg’s opinion is a master of understatement: "Classification of a law as substantive
or procedural for Erie purposes is sometimes a challenging endeavor."
She holds that Rule 59 is inapplicable on the ground that the Rule simply governs the
reasons a new trial should be granted (whatever custom determined) but doesn't specify
the standard for determining excessiveness. Scalia objects. (This was his very argument
in his dissent in Stewart v. Ricoh, which the majority rejected: that Section 1404 spelled
out factors but not the weight given them and for that one must turn to state law.) Also,
he argues that the 7th amendment allows trial courts and even appellate courts to
reexamine damage awards.
Therefore, Ginsburg is in RDA land. She looks at the twin aims test, and sees forum
shopping, but doesn't stop there. She then turns to a Byrd competing governmental
interests analysis and splits the baby: NY has a substantive regulatory stake in the
heightened scrutiny of jury awards, even if the regulatory interests are framed within a
procedural rule. On the other hand, federal courts have a structural stake in not
reassessing closely jury awards because of their history with the 7th amendment, even
though, strictly speaking, trial and appellate courts can constitutionally do so. Note, she
claims there is a crucial difference between the federal trial and the federal appellate
courts. The constitutional permission to scrutinize jury awards is a relatively late
development for the appellate courts and not secure. (What on earth does this mean? It's
only a little constitutional? Presumably she means there is a generalized interest at the
trial level and a heightened interest at the appellate level because of constitutional
overtones and long practice.) So, federal trial courts should use the NY standard of
review in assessing new trial motions, but not appellate courts because of the shadow of
7th amendment concerns, reflecting the federal interest in the character of federal
litigation. Note also that both Ginsburg and Scalia are arguing within the framework of
Byrd, not the Hannah dictum. Ginsburg doesn't care much about the forum shopping
aspect clearly. Scalia too reminds us that forum shopping and outcome determination
weren't supposed to become talismans and thus defeat the procedural autonomy of the
federal courts.
III. Determining the content of state law.
Should a federal court pretend it is the appellate court in the state or a state trial court,
bound by the state’s supreme court rulings? The latter affects the nature of what we think
judging involves. It is a tricky line between an outmoded state precedent drained of
authority by time and developments elsewhere and a federal court with no authority to
change state law who must take the state law as it finds it. Judge Wyzanski had this
advice to give: a federal judge engages in a psychological process of understanding the
jurisprudence of the state judiciary and predicting what it will do. But the federal court
can't take good policy reasons independently into account. Policy is a matter of political
preference left to the will of the legislator.
IV. Federal Common law. Remember to distinguish procedural versus substantive
federal common law. The former are rules or doctrines that federal courts fashion for
procedural purposes that are judge-made (like forum non conveniens) but may need to be
displaced by state substantive law or state procedural law reflecting substantive interests.
The latter are rules or doctrines that judges make to fill gaps in federal statutes or in the
Constitution. Examples we have encountered are: making up a federal rule of decision to
govern a dispute between two states over water rights, implying a damage remedy in the
fourth amendment, or implying a private right of action under a federal statute. Also
remember that the basis of jurisdiction is not really the relevant issue but, rather, whether
the federal court is adjudicating a state based claim or a federal one.
