The Yale Law Journal

June 2016

Reimagining Finality in Parallel Patent Proceedings

Civil ProcedureIntellectual PropertyProperty

Parties may challenge the validity of issued patents in federal courts and before the Patent and Trademark Office (PTO) and its administrative tribunal, the Patent Trial and Appeal Board (PTAB). Recently, the Court of Appeals for the Federal Circuit, which has exclusive appellate jurisdiction over cases arising under the patent laws, has struggled to manage cases contested in parallel judicial and administrative proceedings.

In Fresenius USA, Inc. v. Baxter International, Inc., the Federal Circuit held that a district court’s judgment may be modified by an “intervening” judgment arising out of a parallel administrative proceeding unless all issues have been fully adjudicated in the district-court action.1 This requirement remains controversial. At least five of the court’s eleven judges oppose the court’s finality rule,2 and this debate has led to vigorous dissents from panel decisions and denials of petitions for rehearing en banc.3Nevertheless, the rule has attracted little academic attention, and commentators that have addressed the rule have analyzed it under formalistic constitutional and preclusion doctrines, ignoring other approaches.4

This Comment argues that none of the arguments for or against the Federal Circuit’s finality rule are legally determinative. Given that the current debate need not define the scope of possibilities, I propose a more flexible rule that looks outside the parameters of this debate for a solution. Under the proposed rule, district courts would have the discretion to adhere to their prior remedy decisions despite intervening administrative judgments of invalidity. Such a rule would allow courts both to prevent litigants from abusing administrative challenges and to preserve the viability of injunctions. In addition, the proposed rule would encourage courts to more carefully communicate their decisions. It might also secure approval from those who disagree with the Federal Circuit’s constitutional analysis.5

Furthermore, courts could implement the proposed rule without waiting for legislative action. Congress could solve the problems identified in this Comment by limiting administrative reexamination6 or eliminating it entirely.7 However, such a development is unlikely and would run counter to current congressional trends. The Leahy-Smith America Invents Act expanded, rather than limited, reexamination,8 and subsequent legislative proposals have favored patent challengers.9 Consequently, courts remain the actors best positioned to address the problems caused by parallel proceedings.

This Comment proceeds in three Parts. Part I describes patent law’s system of parallel judicial and administrative adjudication. Part II analyzes the Federal Circuit’s finality jurisprudence and arguments for and against the court’s adopted rule. Part III proposes an alternative rule and explains its advantages.

I. parallel judicial and administrative proceedings

Administrative patent validity proceedings have existed for more than thirty years. In 1980, Congress authorized the PTO to conduct ex parte reexaminations of issued patents and to cancel claims found invalid.10 Supporters believed reexamination would “permit efficient resolution of questions about the validity of issued patents without recourse to expensive and lengthy infringement litigation.”11 Accordingly, the 1980 statute makes reexamination broadly available. “Any person at any time” may request reexamination, and the PTO will grant any request that raises a “substantial new question of patentability” regarding the patent’s novelty or obviousness.12 The “substantial new question” standard is easily met, and a request may raise a “substantial new question” even if a court has previously considered the prior art references asserted by the request.13

Ex parte reexamination procedures are procedurally similar to those used in an initial examination.14 If the examiner finds the patent invalid, the patentee may appeal the examiner’s determination to the PTAB and then directly to the Federal Circuit.15 Requests may be anonymous, and requestors are not precluded from challenging the patent’s validity in subsequent proceedings.16 Rather, the reexamination statute’s drafters believed that courts would exercise their stay power to “prevent costly pretrial maneuvering which attempts to circumvent the reexamination procedure.”17

Between July 1981, when the ex parte reexamination statute entered into force, and September 2014, the PTO received 13,217 requests.18 The PTO granted ninety-two percent of these requests and changed or cancelled claims in seventy-eight percent of reexaminations.19 Contrary to the drafters’ expectations, requestors commonly use ex parte reexamination as a supplement, not an alternative, to litigation. Parallel judicial and administrative proceedings are common, and thirty-three percent of reexaminations involve patents known to be in litigation.20 Courts have been reluctant to grant stays in these cases in part because they recognize that the outcome of the reexamination will be unlikely to resolve all litigation issues.21 Furthermore, appeals may require four or more years to resolve, and resolution of these appeals often does not prevent further legal action between the parties.22

Today, ex parte reexamination is not the only, or even the most popular, administrative option for challenging the validity of issued patents. In 2011, the America Invents Act expanded parties’ administrative options by creating inter partes review and other adversarial proceedings.23 Despite the availability of these paths of review, reexamination continues to provide a meaningful avenue for challenging patent validity. While the number of reexamination requests decreased significantly in 2013, after the America Invents Act’s review provisions entered into force, this decline stabilized in 2014.24 Furthermore, inter partes review proceedings might themselves create opportunities for conflicting judicial and administrative judgments. Despite estoppel provisions25 and strict statutory deadlines,26 inter partes review proceedings themselves may require nearly two years to resolve, enough time for a district court to enter judgment.27

II. the federal circuit’s finality rule

A. The Rule’s History

The Federal Circuit’s finality rule emerged out of this parallel system of district court and administrative proceedings. In 2003, Fresenius filed a declaratory judgment action against Baxter, alleging that three of Baxter’s patents, including U.S. Patent No. 5,247,434 (the ’434 patent), were invalid and noninfringed.28 Baxter counterclaimed for infringement.29 In 2007, the jury awarded Baxter $14,266,000 in preverdict damages, and the district court further awarded Baxter a permanent injunction and postverdict damages, calculated at a reasonable royalty rate.30

In Fresenius I, the Federal Circuit affirmed the district court’s findings that the ’434 patent was valid and infringed, but found the other patents invalid.31 Accordingly, the court vacated the injunction and postverdict damages award and remanded for “further proceedings consistent with [its] opinion.”32 Significantly, Fresenius did not appeal the district court’s preverdict damages award, and the Federal Circuit did not expressly vacate this portion of the district court’s judgment.33 The Federal Circuit panel disagreed over the remand order’s legal effect. Judges Dyk and Newman concurred separately, attempting to shape the order in opposite directions. Judge Dyk argued that on remand, the district court could stay the litigation pending the resolution of a parallel ex parte reexamination proceeding before the PTO.34 Judge Newman, by contrast, argued that stay would be inappropriate and that “[o]nly an issue of adjustment of damages and a modification of the permanent injunction . . . remain[ed] on remand.”35

Subsequently, the district court declined to stay the litigation36 and proceeded to adjudicate the remaining issues. Simultaneously, the PTO found the claims at issue invalid.37 While the appeal from the reexamination proceeding was pending, the district court entered judgment for Baxter.38 The court’s judgment preserved the $14,266,000 preverdict damages award but, in compliance with the Federal Circuit’s remand order, reduced postverdict damages.39

Baxter’s victory was short-lived. The Federal Circuit affirmed the PTO’s finding of invalidity.40 The district court’s contrary decision did not bind the PTO “because the two proceedings necessarily applied different burdens of proof and relied on different records.”41 Subsequently, Fresenius appealed the district court’s judgment, arguing that although Fresenius would ordinarily be precluded from relitigating infringement and preverdict damages, the Federal Circuit was obligated to apply its own intervening judgment holding the claims invalid.42

In Fresenius II, the Federal Circuit agreed and ordered the district court to dismiss the action entirely.43 The court explained that although the district court’s judgment was “final for purposes of appeal” and “might have been given preclusive effect in another infringement case between these parties, it was not sufficiently final to preclude application of the intervening final judgment” affirming the PTO’s finding of invalidity.44 “To rise to that level, the litigation must be entirely concluded so that [the] cause of action [against the infringer] was merged into a final judgment . . . one that ‘ends the litigation on the merits and leaves nothing for the court to do but execute the judgment.’”45 However, Fresenius I “left several aspects of the district court’s original judgment unresolved,” including postverdict damages and injunctive relief.46 Thus, the court reasoned, under “general res judicata principles governing the preclusive effect of a judgment,” there was “no final judgment binding the parties.”47

Judge Newman dissented. In her view, the court’s requirement that all aspects of a judgment be fully adjudicated “violate[d] the constitutional plan” that court judgments “may not lawfully be revised, overturned or refused faith and credit” by other branches.48 Further, she argued, the court’s decision was “contrary to the precedent of every circuit,” which “impose[d] finality and preclusion” on issues “finally decided in full and fair litigation” even if other issues remained unresolved.49

Since then, the Federal Circuit’s decisions have split along similar lines. In Fresenius III, the court denied Baxter’s motion for rehearing and rehearing en banc.50 Judge Dyk concurred,51 and Judges O’Malley and Newman dissented,52 reiterating their established positions. In ePlus, Inc. v. Lawson Software, Inc., the Federal Circuit extended Fresenius II’s rule to civil contempt orders.53 The court explained that a contempt finding for noncompliance with an injunction, like a damages award, could not survive an intervening judgment finding the patent invalid.54 As before, the ruling divided the court. Judges Moore and Newman, joined by Judges O’Malley, Reyna, and Wallach, dissented from the court’s decision to deny rehearing and rehearing en banc.55

B. Arguments for the Current Rule

Despite continuing controversy, none of the arguments for or against the Federal Circuit’s finality rule are legally determinative. Precedent neither requires nor forbids the court’s approach.

Proponents of the rule rely heavily on the Supreme Court’s decision in John Simmons Co. v. Grier Bros. Co.56 There, the district court ruled for the patentee on patent infringement and unfair competition claims and issued a decree granting a permanent injunction and accounting.57 The Third Circuit affirmed the unfair competition holding but ruled the reissued patent invalid and ordered the district court to modify its decree.58

Subsequently, in a separate case, the Supreme Court found the reissued patent not invalid, and the patentee moved to vacate the district court’s modified decree, entered pursuant to the Third Circuit’s order.59 On appeal, the Simmons Court held that the decree should be vacated because it was not final.60 Instead, the Court required the parties to hold an “accounting” to determine the patentee’s award for the unfair competition claim.61 Thus, “there was nothing to take the case out of the ordinary rule that there can be but one final decree in a suit in equity.”62

Superficially, Simmons appears to support the Fresenius II court’s conclusion that an order is not immune from the application of an intervening judgment if any issues remain unadjudicated. However, its precedential value is suspect. First, the Federal Rules of Civil Procedure, adopted after Simmons, abrogated the traditional rule that “there can be but one final decree in a suit in equity” by allowing courts to “direct entry of a final judgment as to one or more, but fewer than all, claims or parties.”63 Second, the Simmons decree was much less definitive, in terms of concluding the action, than Fresenius’s preverdict damages award. Simmons involved an accounting, a multistep procedure involving a court-appointed special master, whose recommendations were subject to objection by the parties and review by the court.64 Third, federal appellate review has changed significantly between Simmons and Fresenius. Caseloads have dramatically expanded, and appellate courts have increasingly entrusted matters of trial management to their district-court colleagues.65 Simmons embodied an earlier attitude, now abandoned, in which appellate courts strictly reviewed district courts’ managerial decisions.66

C. Arguments Against the Current Rule

Similarly, the legal arguments against the Federal Circuit’s rule do not compel its abandonment. Opponents argue that the rule “violates the constitutional plan,” enshrined in Article III, that judgments “may not lawfully be revised, overturned or refused faith and credit by another Department of Government.”67 In other words, they argue, the Federal Circuit’s decision to affirm an administrative finding of invalidity, notwithstanding an apparently contrary court judgment, effectively allows the agency to overturn the lower court.

This conclusion is mistaken. First, from a practical perspective, the constitutional objections are inconsistent with settled understandings regarding the differences between administrative and judicial proceedings. In Ethicon, Inc. v. Quigg, the Federal Circuit held that civil and administrative proceedings involve different inquiries because of their different standards of proof.68 “Courts do not find patents ‘valid,’ only that the patent challenger did not carry the ‘burden of establishing invalidity in the particular case before the court’ . . . .”69 Subsequently, in In re Swanson, the court applied this logic to uphold the constitutionality of parallel reexaminations.70 Second, from a normative perspective, the outcome of the constitutional analysis depends on whether the court’s finality rule is correct. Article III protects final judicial decisions from modification.71 Assuming arguendo that a district court’s judgment is nonfinal, constitutional objections fail.

Additionally, opponents of the rule argue that under modern claim and issue preclusion principles, a court’s judgment may be final with respect to issues that are finally decided in full and fair litigation, even if other issues remain unadjudicated.72 However, this argument conflates preclusion and immunity from intervening judgment.73 These concepts are analytically and legally distinct. Preclusion prevents parties from bringing the same claims in subsequent proceedings; it does not forbid the application of intervening judgments in the original proceeding.74 Thus, a court may be bound to apply an intervening judgment even if the parties would have been precluded from raising that issue.75

Thus, just as proponents of the rule err in concluding that the rule is legally necessary, opponents of the rule err in concluding that it is constitutionally prohibited.

III. towards a discretionary alternative

As Part II has argued, neither the arguments presented for nor against the Federal Circuit’s finality rule are convincing. In the absence of a legally compelled result this Part recommends that the court replace the rule with a flexible discretionary alternative.

Thus far, the Federal Circuit has analyzed finality in mandatory terms.76 However, a better approach would be to allow district courts, at their discretion, to adhere to their prior remedy decisions despite intervening judgments of invalidity. Under the proposed rule, the Federal Circuit would allow district courts to declare individual remedy decisions final for the purposes of immunity from intervening judgments, but only if those courts expressly justified their decisions to do so. On appeal, the Federal Circuit could expressly vacate or modify those decisions based on its assessment of the district court’s justification. If not expressly vacated or modified by the Federal Circuit’s remand order, those decisions would become final. Thus, cases like Fresenius II would turn on both the clarity and thoroughness of the district court’s remedy order and the clarity of the Federal Circuit’s remand order. And unlike in Fresenius II, an ambiguous remand order would be construed narrowly, not broadly.77

The Supreme Court’s retroactivity jurisprudence provides some support for the proposed rule. Generally, discretionary doctrines aimed at promoting finality and efficiency allow courts to depart from their prior decisions if intervening judgments undermine the basis for those decisions.78 However, in analogous contexts, federal courts have declined to apply intervening changes in law based on fairness concerns. For example, the Court has “refused to apply” intervening changes in law where doing so “would infringe upon or deprive a person of a right that had matured or become unconditional.”79 While the Court has discussed this principle as an element of statutory retroactivity analysis, it has applied the principle to agency actions, where a party’s rights “matured” under a prior judicial decision before they were extinguished by the agency.80

Unfortunately, the Court’s jurisprudence provides little guidance in determining when a right is sufficiently “mature” to justify ignoring an intervening change in law.81 However, the Court’s recognition of fairness interests suggests that, in at least some cases, district courts would be justified in adhering to their prior remedy decisions, notwithstanding intervening judgments of invalidity. Adherence to a prior remedy decision may be appropriate where it is clear that the accused infringer or third party has abused the reexamination system or where the only outstanding issues involve injunctions.

The proposed rule would have four principal benefits. First, the proposed rule would prevent accused infringers or third parties from abusing the reexamination system. The drafters of the reexamination statute believed that creating an administrative alternative would “greatly reduce, if not end, the threat of legal costs being used to ‘blackmail’ [patentees] into allowing patent infringements or being forced to license their patents for nominal fees.”82 However, the Federal Circuit’s finality rule encourages challengers to pursue their claims in multiple forums and delay litigation while reexamination progresses. Even more troublingly, the rule creates opportunities for third parties to interfere with patentees’ licensing activities. Such parties may threaten to file low-probability, high-consequence reexamination requests to induce patentees to grant transferrable licenses to high-value patents.83 The low threshold for granting an examination request, combined with the high threshold for litigation misconduct, make this strategy easy to execute.84

Second, the proposed rule would preserve the viability of injunctions. Orders involving injunctions are subject to interlocutory appeal85 and are commonly vacated or modified.86 Accordingly, injunctions create opportunities for the kinds of delays that, under the Federal Circuit’s finality rule, subject all issues to modification. Faced with this prospect, patentees may be discouraged from pursuing injunctions in cases involving parallel administrative proceedings (or in which parallel proceedings are likely). Yet injunctions are an important remedy. Although injunctions are not always appropriate, “courts have granted injunctive relief upon a finding of infringement in the vast majority of patent cases . . . given the difficulty of protecting a right to exclude through monetary remedies . . . .”87 Allowing district courts to adhere to prior decisions regarding validity and damages, notwithstanding ongoing litigation over injunctions, would maintain the strategic value of seeking injunctive relief.

Third, the proposed rule would encourage the Federal Circuit and district courts to more carefully consider and communicate the implications of their decisions. Fresenius II was controversial, in large part, because the Fresenius I panel did not expressly address the district court’s preverdict damages award in its remand order. Judge Newman believed that the Federal Circuit’s remand order, as issued, was limited to the district court’s injunction and postverdict damages awards,88 while the majority subsequently held that the order was broader in scope.89 In contrast with the Federal Circuit’s finality rule, the proposed rule would encourage clarity. District courts that did not expressly finalize remedies for immunity purposes would lose the protection that finality provides, and the Federal Circuit, if it were unclear about the scope of its remand order, would forgo the opportunity to reopen the district court’s judgment. Thus, the proposed rule recognizes that appellate panels may disagree about how district courts should proceed on remand90 but discourages them from resolving disagreements through ambiguity.

Finally, the proposed rule might appease critics who believe that the Federal Circuit’s finality rule violates Article III. Specifically, the proposed rule avoids the situation in which an agency would “overturn” a court’s decision by allowing a district court to preemptively decide whether its decision will incorporate a future agency finding of invalidity. While this Comment concludes that these Article III concerns are ultimately meritless,91 the proposed rule might have a comparative advantage over the Federal Circuit’s approach in securing approval from the opponents of the current finality rule.

Although the proposed rule would allow for the “peculiar result” in which a patent holder “secure[s] damages for infringement of a patent that has been conclusively found invalid by the PTO,”92 this result is not as strange as it seems. It is well settled that an adjudged infringer may not recover the cost of a final judgment if the patent is subsequently invalidated.93 Thus, if the Federal Circuit had affirmed the PTO’s finding of invalidity after, rather than during, Fresenius’s thirty-day appellate window, the district court’s judgment would have been unassailable. Whatever the case, the proposed rule’s considerable advantages outweigh its potential peculiarities.


The Federal Circuit’s finality rule, which insists that district court decisions remain open to modification by intervening administrative judgments until all issues have been finally adjudicated, remains controversial. This Comment assesses the arguments for and against the rule and argues that neither side’s arguments are legally determinative. Instead, the court should permit district courts to adhere to prior remedy decisions despite intervening administrative judgments of invalidity. Such an approach would have several advantages. It would allow courts to prevent litigants from abusing administrative challenges, preserve the viability of injunctions, and encourage courts to communicate their decisions more carefully.

Moreover, the proposed rule would draw upon decades of appellate experience. Congress established the Federal Circuit to promote uniformity in patent law. Yet, prominent critics contend that the court’s exclusive jurisdiction should be abolished because the court’s specialization breeds insularity.94 Consistent with this critique, this Comment argues that the Federal Circuit would benefit from looking outside of patent law to solve patent law’s problems. The court’s finality rule is a creature of the court’s tendency to search for patent-specific solutions. However, issues concerning parallel litigation are not exclusive to patent law, and over the course of the past century, federal appellate courts have increasingly entrusted these issues—and others involving trial management—to their district-court colleagues. The proposed rule embodies this trust and provides a model for courts to follow in balancing the interests of litigants, courts, and the public.