The United States Department of Justice appreciates this opportunity to comment on the Tentative Draft Report of the Commission on Structural Alternatives for the Federal Courts of Appeals. The Department commends the Commission on its effort to address a broad range of important issues concerning the structure of the federal courts of appeals. We endorse the Commission's statement that all recommendations on this subject should be free from political considerations and that "[i]t is wrong to realign circuits (or not realign them), and to restructure courts (or leave them alone) because of particular judicial decisions or particular judges." Draft Report at 6. In a relatively short period of time, the Commission has accomplished a great deal and brought new perspectives to bear on difficult issues.
The draft report's recommendations have serious, far-reaching implications for the structure and functioning of the federal courts. We leave to others an assessment from the perspective of the judiciary. We approach these issues from our perspective as a litigant -- a participant in over 40 percent of the cases heard in the federal courts of appeals -- which must comply with procedural rules, brief and argue cases, reconcile tensions in the results and reasoning of decisions in order to give advice to client agencies, and consider whether to seek review of decisions adverse to the government. That difference in perspective may account in part for the fact that we do not share the draft report's assessment of the relevant problems and particular solutions.
To prepare these comments on the draft report's recommendations, the Department of Justice has sought the views of all of its litigating components, including the United States Attorney's Offices. The views expressed in this submission reflect substantial unanimity among the various offices and components of the Department, gathered from the bottom up and reflecting expertise in nearly every area within the subject-matter jurisdiction of the courts of appeals.
We begin with the observation that all available means of non-structural reform should be attempted and assessed before structural changes are imposed on the federal courts. In our June 1, 1998, submission, we expressed the view that structural changes should be undertaken only if a pervasive problem exists, that problem cannot be addressed within the existing structure, and a workable solution can be devised whose advantages outweigh its immediate and potential detriments. Guided by those principles, we agree with the draft report's recommendation that the Ninth Circuit should not be split at this time, and we concur generally in its view that "[t]here is no persuasive evidence that the Ninth Circuit (or any other circuit, for that matter) is not working effectively, or that creating new circuits will improve the administration of justice in any circuit or overall." Draft Report at 27. In our view, the lack of evidence supporting circuit splits also counsels against what we view as the principal recommendation contained in the draft report -- the creation of divisions for the Ninth and other large circuits. That proposal would have potentially adverse repercussions for the administration of justice in the Ninth Circuit, and ultimately across all federal courts of appeals.
We do not mean to suggest that the Ninth Circuit, or the courts of appeals generally, would not benefit from some change or that the need for structural changes will not arise at some future date. Indeed, if current demographic trends continue, structural changes will plainly be necessary -- in the Ninth Circuit and probably elsewhere. In our view, however, current problems may well be resolved through less dramatic, and disruptive, measures. The draft report appears to ground its tentative recommendations on the premise that the Ninth Circuit must now be split, either completely or into semi-autonomous courts of appeals within a single circuit. From our perspective as a litigant and the largest "consumer" of the Ninth Circuit's work product -- its decisions -- that premise is unwarranted. We believe that the Circuit generally operates well, except that it does not employ adequate mechanisms to review and reconcile panel decisions that conflict or are in tension with one another, or that require correction by the court as a whole. Proceeding from that premise, we believe that mechanisms short of a split (divisional or otherwise) should be tried first -- particularly since the steps recommended in the draft report would likely exacerbate, rather than ameliorate, the problem we perceive. Therefore, before recommendations such as those contained in the draft report are implemented by Congress, we urge the adoption of the non-structural reforms suggested in this and our earlier submission.
Because of the approach we have taken, we have not fully assessed the comparative merits of a circuit split versus the adjudicative divisions plan proposed in the draft report. Faced with a binary choice between those alternatives, several components within the Department would likely favor the former; others might support an experiment along the lines of the Commission's proposal, as modified by the suggestions in Section II.A.3 infra. We offer our views at this time on what we regard as the most significant issues presented by the draft report.
II. COMMENTS ON THE DRAFT REPORT'S RECOMMENDATIONS
A. A Divisional Arrangement for the Ninth Circuit Court of Appeals
Draft Report Recommendation:
The draft report recommends that "Congress and the President by statute restructure the
Court of Appeals for the Ninth Circuit into three regionally based adjudicative divisions." Draft
Report at 39. Under this proposal, the Ninth Circuit would be divided into Northern, Middle and
Southern Divisions, with California split between the Middle and Southern Divisions. Between
seven and eleven active judges would serve in each division, with the presiding judge of each
division chosen in the manner that currently exists for the selection of a circuit's Chief Judge.
"A majority of judges serving on each division would be residents of the districts over which that
division has jurisdiction, but each division would also include some judges not residing within
the division, assigned randomly or by lot for specified terms of at least a year." Id. at 41. The
divisions would operate under the Federal Rules of Appellate Procedure and under local rules
adopted by the Circuit as a whole. Judges from each division would hear cases from within the
division's geographic boundaries, and the court would provide by rule for appeals from federal
administrative agencies. Each division would use an en banc procedure to rehear cases from
within the division. One division's decisions, whether panel or en banc, would not bind any
other division, but "should be accorded substantial weight." Ibid. Finally, a non-regional
"Circuit Division" consisting of seven judges would be formed. The Circuit Division panel
would include the Chief Judge of the Circuit, the presiding judge of each regional division, and
one judge from each of the regional divisions. The Circuit Division would have discretionary
jurisdiction to review "square inter-divisional conflicts" of law, but only after a panel decision
had been reviewed by the division en banc or had been denied en banc review. Id. at 42. The
Circuit Division would not have the jurisdiction to review decisions for error, conflict with
another circuit's decision, or exceptional importance.
Department of Justice Comments:
The draft report concludes that "[t]here is no persuasive evidence that the Ninth Circuit
(or any other circuit, for that matter) is not working effectively, or that creating new circuits will
improve the administration of justice in any circuit or overall." Draft Report at 27. We agree
with that finding (with the important exception of the court's en banc processes), and with the
report's conclusion that a split of the Ninth Circuit is not warranted at this time. In our view, that
finding also underscores the lack of justification for creating semi-autonomous adjudicative
divisions within the Ninth and other circuits. From our perspective, the proposal is likely to
exacerbate the problem we have identified, and it may well create others. For that reason, the
Department does not support the draft report's recommendation, at least until alternative non-structural changes are tried that would, we believe, provide meaningful improvement in the
quality of justice for litigants.
1. If enacted, the proposal would likely create significant problems in the
administration of justice in the Ninth Circuit
In this section, we highlight our concerns about possible effects of the adjudicative
divisions plan so that the Commission may evaluate whether the potential costs of going forward
with the draft report's proposal now would outweigh the benefits sought to be achieved.
A basic tenet of American jurisprudence is that federal law should be applied as uniformly as possible within and across circuits. National uniformity and predictability are particularly important to the Department of Justice, which must enforce federal law, and advise federal agencies about the meaning of that law, throughout the country. The Department also plays a special role in the process of unifying the meaning of federal law: as the most frequent litigant in the federal courts, the Department, through the Solicitor General, exercises considerable restraint in choosing which cases the United States brings to the courts of appeals.
It is of paramount importance that federal law be interpreted consistently regardless of the location of the court or the composition of the judicial panel. Rather than reduce the amount of intra- and inter-circuit conflicts created by the Ninth Circuit, we believe the draft report's suggested divisional structure would effectively legitimize, and even encourage, the development of such conflicts. Indeed, the draft report is explicit that "[e]ach division would function as a semi-autonomous decisional unit" and that "[d]ecisions made in one division would not bind any other division." Draft Report at 41. That directive, however, is not absolute. Panels would be required to afford the decisions of other divisions "substantial weight as the judges of the circuit endeavor to keep circuit law consistent." Ibid.
Those instructions appear to point in opposite directions, with the ultimate result that clarity and consistency in the law would suffer. If the Commission were to propose that the first decision of a panel of any division be given the usual precedential effect of a court of appeals' decision, consistency in decisional law would be enhanced, because subsequent panels in other divisions would be obliged to follow that precedent (as is now required) unless the earlier decision were overruled by the en banc court. Conversely, if the Commission were to propose giving no weight to other-division panel decisions, the result would undoubtedly be more inconsistency in the law. By straddling those two regimes, however, the draft report's approach would cause potential additional problems for litigants. They would be required either to accept treatment inconsistent from that afforded other litigants in the same circuit or proceed through a time-consuming multi-stage process to resolve the conflict, by first seeking division en banc review and, if that fails, Circuit Division rehearing. Moreover, the imposition of a "substantial weight" standard would invite litigants to seek panel or divisional en banc rehearing on the ground that the panel gave insufficient weight to a decision of another division that the panel decided not to follow. Thus, we believe that the regime suggested by the tentative draft report would not only reduce clarity and uniformity in circuit law, but it also would generate unproductive litigation on an issue that is at least one step removed from the merits of the case.
The draft report suggests a way of resolving conflicts among divisions through the
mechanism of a "Circuit Division." The Circuit Division's only role, however, would be to
resolve "square inter-divisional conflicts" between the regional divisions. It is unclear from the
draft report what a "square conflict" is and how such a conflict is different from the existence of
other decisions that are difficult to reconcile but which nonetheless point the law in different
directions.(1) Often, the creation of a conflict is not clear, much less immediately clear. And
because the decisions of other divisions are not binding precedents, judges would be less likely
to distinguish, discuss, or even cite decisions from outside their division. Overall, the Circuit
Division mechanism, as proposed, does not seem at all likely to resolve the many intra-circuit
inconsistencies that the semi-autonomous division system would produce.(2)
b. The proposal would unnecessarily delay the administration of justice
The interest in achieving an expeditious appellate process is important for all kinds of cases, but it is particularly acute in two areas in which the Ninth Circuit has large caseloads: criminal cases, in which the defendant's liberty is at stake; and immigration cases, in which the Ninth Circuit currently reviews as much as 50 percent of the nationwide caseload and in which delay defers the alien's expulsion and can encourage new case filings. A swifter and less cumbersome process in such matters is in the interest of both the government, which must enforce the law, and the individual, whose resources typically cannot sustain vigorous multi-tier litigation.
By adding another layer of review, the suggested Ninth Circuit restructuring would delay
the completion of the judicial process for litigants. Following an adverse panel decision, an
aggrieved litigant could seek en banc review by the Division en banc court, as would now be true
of the Circuit as a whole. A denial of such a petition would, in many cases, precipitate a further
request for rehearing at the Circuit Division level.(3) The evaluation of a case for alleged conflicts
with a decision of another panel would only add to what is already a protracted period for finally
resolving cases.
The effect of the proposal on the ability of litigants to obtain Supreme Court review is a
matter of significant concern. The narrow jurisdiction of the Circuit Division to hear only
"square inter-divisional conflicts" means that many cases decided by the divisions would not be
subject to circuit-wide review. Categories of cases excluded from review would include matters
of exceptional importance, cases that conflict with decisions of other circuits, and cases in which
the intra-circuit conflict is significant but not "square." Such cases would be decided solely at
the divisional level, and those decisions would not be binding circuit-wide. That structure would
inevitably multiply the number of decisions within the Ninth Circuit that conflict with decisions
of other circuits, while simultaneously creating a possible impediment to Supreme Court review.
It is uncertain whether Supreme Court Justices would vote to grant certiorari in cases that present
conflicts between only one division of the Ninth Circuit (rather than the Circuit as whole) and
another circuit. The discretionary nature of certiorari jurisdiction suggests that parties opposing
review will argue that the Supreme Court should give the Ninth Circuit as a whole an
opportunity to overturn a divisional decision so as to bring the division into harmony with the
other circuit's decision. The proposed divisional structure therefore might serve to insulate
decisions of the Ninth Circuit from further review, effectively isolating it from the rest of the
federal court system.(4)
d. Dividing California is undesirable
Under the draft report's proposal, the State of California would be split between the
Middle and Southern Divisions of the Ninth Circuit, neither of which would be required to
follow the precedent of the other.(5) We do not support dividing any State in this manner, because,
as much as possible, federal rights and responsibilities should be the same for all citizens within
a State. Splitting California between two divisions that are not bound by each other's precedent
would yield different interpretations of federal and state law, and could result in inconsistent
federal court rulings regarding the constitutionality of the same California law. Nor would there
be any guarantee of quick or certain Supreme Court resolution of those issues. Instead, the
review and conflict resolution process could be protracted, particularly with the requisite added
layer of Circuit Division (following divisional en banc) review. In addition, the existence of
different divisions within one State (or even within the Circuit as a whole) could encourage
forum shopping among those seeking to assure a more favorable audience to adjudicate questions
of federal and state law, as well as delays in the reconciliation of conflicting decisions.(6)
We are concerned that enactment of the tentative draft proposal might unintentionally set the stage for an eventual split of the Ninth Circuit, even if objective circumstances would not otherwise warrant that step, simply because the alternative course of fully reunifying the Ninth Circuit would be seen as too difficult. Although we favor the recommendation that a "sunset provision" end the divisional "experiment" in seven years, which would enable the experience to be assessed, our concern is that in practice the effects of the "experiment" might be difficult to reverse. At the conclusion of that period, each division within the Ninth Circuit would have developed different bodies of case law that reflected judicial disagreements short of "square conflicts." That case law would require cumbersome and time-consuming procedures of reconciliation in the event Congress concluded that the divisional experiment should not be made permanent. At the very least, the reunified court would have to decide what precedential effect to give decisions rendered only within a single division. It could well decide not to give any effect to such decisions, which would invite potentially substantial relitigation of issues. The Commission thus should consider the consequences of the suggested experiment in the event it proves unsuccessful. After the seven-year period, it may be that putting the Ninth Circuit back together would seem too difficult, and creating three circuits based on the geographic boundaries of the three divisions would be viewed as the only practical alternative. We find such a potential outcome troubling, as it would create a problem where none currently exists and promote as a solution the very circuit split the Commission has found to be unwarranted.
2. The proposal is not justified by the reasons given for the change in Ninth
Circuit structure
In this section, we analyze the justifications put forward in the draft report for its adjudicative divisions plan. The draft recognizes that "[t]here is no persuasive evidence that the Ninth Circuit * * * is not working effectively." Draft Report at 27. It nonetheless seeks to justify
its proposed reform by highlighting three objectives it hopes will be achieved. Draft Report at
42-45. We are doubtful that the reforms would achieve the stated objectives.
a. The creation of smaller decisional units would not promote consistency and
predictability in federal law
The draft report suggests that the creation of smaller adjudicative divisions would promote consistency and predictability, because circuit judges would be in a better position to monitor the law as it develops within their division, and because district judges would know who would be reviewing their decisions. It also suggests that smaller, more stable groups of reviewing judges would lead to more predictable outcomes and that each of the court's divisions would become much more of a "known bench" to those who practice before it, "fostering judicial accountability and public confidence." Draft Report at 43. For the reasons stated above, however, we envision the divisional alignment creating less uniformity and less consistency in the Circuit's law.
Although it is true that the Ninth Circuit currently generates a large body of published decisions, the creation of smaller decisional units would not relieve judges of the obligation to monitor the opinions of divisions other than their own, especially if each division must give substantial weight to the decisions of other division panels. In addition, because the proposal contemplates that a number of judges would be assigned outside their division of residence for substantial periods of time, it is unlikely that judges would benefit substantially over the long run by ostensibly being relieved of the burden of monitoring other divisions' opinions. While serving outside their division of residence, they would presumably be expected to keep abreast of the decisions of at least two divisions -- their division of permanent residence and their division of temporary assignment. And if, over a three-to-five year period, they might be assigned to all three divisions, that monitoring responsibility would be hindered by a failure to have kept up with the output of all three divisions. Whatever benefit might accrue to individual judges with respect to the burden of monitoring opinions, therefore, is likely to be only modest and incomplete.
In any event, any gains in consistency within a particular division would likely be negated by tensions and unresolved conflicts among divisions that do not rise to the level of "square conflicts" that are subject to resolution by the Circuit Division. Creating three different decisional units instead of one could only exacerbate inconsistencies in the law rather than ameliorate them, particularly when the concept of a "conflict" is highly elastic and subjective and the means for resolving disagreement is left to the discretion of the judges.
We also question whether the divisional arrangement would provide litigants with a greater sense of predictability by creating a smaller pool of judges from which a panel would be selected. Because the composition of judges in the divisions would change on a yearly basis and very few appeals reach fruition in under a year, litigants would face only somewhat less uncertainty over the likely decisionmakers in their case than they presently encounter. And to the extent discernible judicial approaches lead to "predictability" within particular divisions, that development itself is not without disadvantages. It is precisely to discourage litigants from attempting to tailor their arguments for particular judges that many circuits do not publicly announce the judges on the panel until shortly before argument. And under the proposed divisional plan, predictability may encourage forum shopping (especially within California) or tactics to delay decisions on appeal while pending cases raising the same issue might be more favorably resolved in a different division (and therefore accorded "substantial weight" in the division where the appeal is pending). A unified circuit avoids those anomalies.
b. The proposed en banc and Circuit Division procedures are not likely to promote clearer, more consistent law
The draft report anticipates that its proposal would provide more effective en banc procedures than those currently operating in the Ninth Circuit, with an attendant increase in consistency and clarity in the development of circuit law. The proposed structure, however, would legitimize and encourage the development of three different bodies of law within the Ninth Circuit and would limit the ability of the Circuit to create consistent and coherent case law. Indeed, the Circuit Division's narrow mandate -- to hear only "square inter-divisional conflicts"-- would appear to result in fewer, not more, cases being resolved by the Circuit as a whole.
The draft report also cites dissatisfaction with the small size and composition of the current en banc process as a reason for its proposal to create the two-tier divisional en banc/Circuit Division structure. Yet the draft report tentatively proposes that only seven judges serve on the Circuit Division panel. A majority of four judges could thus bind the entire Circuit of 28 active judges, a number that is even less than under the current system. And the composition of the small Circuit Division means that judges chosen for it would wield a disproportionate amount of judicial power on the most divisive legal issues. The seven judges on the Circuit Division would include the Chief Judge of the Circuit and the three presiding judges all of whom would be among the most senior active judges in the Circuit at a given time. Thus, it is quite possible that a majority of the Circuit Division could represent a distinct minority of the composition of the Ninth Circuit as a whole. If an aim of en banc review is to ensure that the Circuit Division speaks for the court as a whole and makes a decision that all judges will recognize as binding, the draft report's proposal might well be counterproductive, because it fails
to provide a sufficiently representative and substantial composition to be perceived as speaking
for the full court and to make all of the Circuit's judges feel invested in the en banc process.
c. The proposed divisional structure would not effectively promote any appropriate legitimate goal of regionalism
The tentative draft posits that the proposed structure would heighten the regional character of the Circuit, "restoring a sense of connection between the court and the regions within the Circuit." Draft Report at 44. We question whether circuit judges should be animated by regionalist concerns in deciding cases involving the interpretation of federal law (especially for subregions within a single circuit), since typically the laws enacted by Congress are of general applicability. Even assuming that such concerns should be given weight, however, the proposal would not necessarily promote that goal.
As proposed, each of the divisions would consist not only of judges from that region, but also of a share of judges who are assigned to that division on a yearly basis. The draft report is careful to provide that "a majority of the judges in each division come from the geographic area each division serves." Ibid. Yet no assurance is given that, in any given case, a majority (or indeed any) of the judges selected for the panel would come from that particular geographic division. Thus, because three-judge panels would bind a division, the rule of law for a particular region of the Circuit could be made by judges with no geographic tie to that area. Moreover, since only "square" inter-divisional conflicts would make their way to the Circuit Division, judges from a different division who disagree with a panel's decision on an issue with consequences across divisional lines may never have an opportunity to participate in the case. The Ninth Circuit has long adjudicated cases in important areas of the law involving issues that arise throughout the Circuit, including those concerning public lands, Indian Tribes, environmental law, and immigration, to name just a few. The ability of the Circuit as a whole to address cross-cutting issues of Circuit-wide concern would be hindered under the Commission's proposal.
Finally, the recitation of statistics in one part of the draft report suggests that the "regional" character of decisionmaking under the proposal would, as a practical matter, be a one-way street. See Draft Report at 46-50. In its analysis of why the Ninth Circuit should not be split into more than one circuit, the draft report notes that the per-judge caseload is significantly lower in the districts composing the proposed Northern Division than in those composing the proposed Southern and Middle Divisions. Even assuming that all vacancies have been filled, if equalization of caseload is a significant consideration, any assignment of circuit judges under the proposed scheme would likely result in Northern Division judges routinely sitting on Southern (and to a lesser extent Middle) Division cases. If "regional character" (Draft Report at 44) in the decision of individual cases is a desirable interest in the federal courts of appeals, it is appropriate to acknowledge that the approach suggested by the draft report would appear to serve that purpose only as to one of the three divisions it advocates that Congress recognize -- the Northern Division. The other two regions would, to varying degrees, continue to experience more fully the effect of "out-of-region" judges deciding cases in that division. From our perspective, that differentiation in treatment is unwarranted. An advantage of the existing structure of the Ninth Circuit is that judges throughout the Circuit may be called upon to decide cases and all litigants are treated equally in that random assignment process. Any change in the present system should not produce the practical effect of unequal consequences for the citizens and litigants of different parts of the Circuit.
3. Conclusion and Alternative Recommendations
For the foregoing reasons, at this time we cannot endorse the draft report's recommendations for restructuring the Ninth Circuit in the manner suggested. From our perspective as litigants, the Ninth Circuit's shortcoming is traceable not principally to its large number of judges or geographical size, but rather to its failure effectively to address erroneous panel decisions in important cases and to review cases in which a meritorious claim of conflict is presented. Those problems, while admittedly difficult to quantify, nonetheless appear susceptible to amelioration by nonstructural means, as suggested in our original submission to the Commission. If that course is followed, structural changes might ultimately prove to be unnecessary and their attendant difficulties and dislocations avoided. After a period of experience with non-structural alternatives and an assessment of legal and demographic trends, the need for implementation of the proposals identified in the draft report and elsewhere might become clearer. a. In particular, we suggest that the Commission consider ways to improve the en banc process in the Ninth Circuit. The Commission should evaluate whether to recommend that Congress grant the Ninth Circuit a dispensation to lower the statutory requirement that a majority of the Circuit's active-service judges must vote affirmatively to rehear a case en banc. The success of the Supreme Court in exercising its discretionary review based on the votes of less than a majority is a model that should be studied for application in the court of appeals' en banc process. The Commission also might consider a requirement that the Ninth Circuit set a regular en banc calendar that could be filled out of the available pool of petitions for en banc review. Regularly scheduled en banc sittings would promote collegiality by bringing judges together and serve as a focal point for routinized resolution of disagreements in the development of case law.
Although a system of increased availability of rehearing en banc would require some investment of judicial resources, it seems likely that time expended en banc in clarifying the law of the circuit and resolving issues of exceptional importance would in the long run be repaid by a corresponding reduction in litigation and an enhanced ability of the Ninth Circuit as a whole to speak through the en banc procedure. Litigation within the Ninth Circuit is certainly encouraged by intra-circuit inconsistencies in the law. The short-term costs of increased en banc review may well pay substantial long-term dividends.
In addition to more effective use of en banc procedures, the Ninth Circuit should employ other mechanisms to reduce intra-circuit conflict and promote uniform development of the law. For example, opinions to be published that distinguish or disagree with existing precedent should be circulated among the judges of the Circuit before publication. Staff personnel should be deployed to review panel decisions for potential conflict with other circuit decisions. Staff attorneys could also be used to alert panels that the same or similar issues are pending in other cases in the Circuit, thereby facilitating communication between panels that might forestall conflicting decisions. These alternatives should be attempted before structural reforms are imposed.
b. In the long term, we recognize, demographic changes in the Nation's population may well necessitate structural changes in the court of appeals system. If and when that occurs, suggestions of the kind advanced in the draft report will provide valuable insight on the respective merits of dividing circuits into small courts of appeals or establishing adjudicative divisions within circuits. We express no view about which approach would best serve the needs of litigants and
the judiciary at that time. However, we offer the following suggestions with respect to the Commission's tentative proposal.
First, in our view, the Circuit Division en banc panels should be larger and more
representative of the Circuit as a whole; a Circuit Division panel majority of four judges should
not have the authority to bind a Circuit of 28 active-service judges. Second, instead of
automatically including division presiding judges, the process should more randomly select
judges for the panel, with appropriate consideration given to regional representation;
representativeness should be emphasized over seniority in the selection of the Circuit Division en
banc panel. Third, the limitation of Circuit Division en banc jurisdiction to "square conflicts"
should be eliminated and the standards made to comport with those established under the Federal
Rules of Appellate Procedure for normal en banc review.(7) And fourth, the Commission should
clarify which judges will be available for service on division en banc courts; with the
requirement that judges be re-assigned for substantial periods of time to divisions other than their
division of residence, it is unclear whether that assignment also applies to the division en banc
process.
B. A Divisional Organization of the Courts of Appeals
Draft Report Recommendation:
The draft report recommends that Congress give all appellate courts with more than 13
judgeships discretion to adopt a divisional arrangement such as the one set out for the Ninth
Circuit and require appellate courts with more than 17 judgeships to do so. Draft Report at 52-53. The draft report states that this recommendation gives "considerable leeway" to the courts of
appeals in designing a divisional structure, even though it would require the creation of a "Circuit
Division" modeled on the one set out for the Ninth Circuit and the use of en banc procedures
within each division to resolve intra-division conflict. Id. at 53. This recommendation also
contemplates that each division would have an odd number of judges (between 7 and 11), that
each judge would be assigned to a specific division for a substantial period of time, and that each
division would exercise exclusive jurisdiction over the appeals assigned to it.
Department of Justice Comments:
The Department of Justice does not support the recommendation that the remaining circuits be required (or, as an initial matter, allowed) to split themselves into semi-autonomous adjudicative divisions when they reach a certain number of judgeships. We do not believe such a significant change in the federal appellate structure is justified, particularly before any experience with the Ninth Circuit divisional plan has been obtained and especially before non-structural alternatives of the type we have suggested are implemented and their effects evaluated.(8)
The implementation of a nationwide adjudicatory divisions plan would create for each circuit the types of problems we have identified in Section II.A of this submission. Moreover, widespread enactment ultimately would result in a completely restructured system overall, adding a fourth layer of review throughout much of the federal judicial system, creating differing paths of access to the Supreme Court depending on geography, and allowing varying bodies of law to be developed by numerous mini-courts of appeals in relative isolation from one another. We are
uncertain how to evaluate the need for that change in view of the Commission's statement that it is proposing "no general realignment of the circuit structure." Draft Report at 51.(9)
As an alternative to the proposal in the tentative draft report, we recommend implementing non-structural changes of the type described in our prior submission, including the en banc and other suggestions outlined in Section II.A.3 above. At a minimum, we suggest that the proposal be deferred for a period of time sufficient to ascertain the existence of systemic problems in other circuits sufficient to warrant such a change and to allow litigants and judges an opportunity to assess whether the proposed structural changes would improve the quality of justice. If the Commission goes forward with the draft report's recommendation that Congress create divisions for the Ninth Circuit on an experimental basis, that experiment should be studied before being imposed elsewhere. In the event the Commission rejects the foregoing suggestion,
we submit that the proposal for a divisional arrangement nationwide should also contain a sunset
provision like that provided for the Ninth Circuit.
C. Two-Judge Panels
Draft Report Recommendation:
The draft report recommends that Congress authorize federal appellate courts to use two-judge panels, and to allow the courts to designate by rule those case types suitable for such
disposition. Draft Report at 53-55. The draft report suggests that two-judge panels should be
used to decide primarily cases "in which the outcome is clearly controlled by well-settled
precedent." Id. at 54. Finally, the draft report states that "[i]f the two judges disagree or are in
doubt, they can enlist a third judge to participate in the decision, or refer the case to a regular
three-judge panel for hearing." Ibid.
Department of Justice Comments:
The Department's experience with various screening procedures employed by the courts
of appeals, including summary affirmance, leads us to question whether it is necessary for
Congress to mandate two-judge panels and whether such panels would actually conserve judicial
resources. We have further questions regarding whether this proposal ensures both adequate
procedures for assessing how cases are selected for decision by such panels and necessary
safeguards for determining how a third judge is to be brought into the process when the two-judge panel reaches an impasse. An evaluation of whether "the outcome [of a case] is clearly
controlled by well-settled precedent" that is made prior to full consideration of the merits will
inevitably produce some erroneous designations, the correction of which would further delay the
process. Draft Report at 54. We also are concerned about how this proposal would affect the
public's perception of the administration of justice by the courts. If the two-judge panel
recommendation is to be adopted at all, we believe it would best be implemented as an
experiment for a limited duration in a few courts to allow Congress, courts, and litigants an
opportunity to assess the change.
D. District Court Appellate Panels (DCAP)
Draft Report Recommendation:
The tentative draft report recommends that Congress "authorize judicial councils to create
district court appellate panels [DCAPs] within the circuit and provide by rule for the assignment
of certain categories of cases to those panels." Draft Report at 55. Each circuit's judicial council
would create a "district court appellate panel service" with district and circuit judges from the
circuit. Ibid. The judicial council would specify categories of cases appropriate for DCAP
jurisdiction and the panel would have exclusive jurisdiction over those cases. The draft report
states that diversity cases would be likely prospects for DCAP jurisdiction, as well as sentencing
appeals and cases that "generally require the reviewer to apply well settled legal rules to varying
fact patterns." Ibid. Panels created from the DCAP service would consist of two district judges
and one circuit judge designated by the chief judge of the circuit. District judges would not
review judgments from the courts on which they serve. Further review of decisions by a DCAP
would be discretionary in the court of appeals. In addition, the panel itself could transfer a case
to the court of appeals if disposition involved a determination of a question of law it deemed
appropriate for the court of appeals.
Department of Justice Comments:
The draft report notes that it does "not express an ultimate judgment as to the value of district court appellate panels." Draft Report at 56. In our view, this proposal would likely result in a net cost to litigants and to the judicial system as a whole, even if it produced an incidental reduction in the burdens on the courts of appeals. Accordingly, we are not persuaded that the creation of DCAPs is warranted or desirable.
First, the use of DCAPs would not reduce the overall judicial workload -- instead, it would simply divert much of the workload for some appeals from busy appellate judges to busy district court judges. Although the factual justification underlying this recommendation is unstated, it may be a response to the statistical trends recorded in Table 2-3, which suggest that in the past century the per-judge caseload for circuit judges has increased five-fold while that for district judges has only doubled. Draft Report at 14. Without a more careful analysis of the workload of district judges, however, it would be premature to base conclusions on those numbers alone. The statistics do not capture the increasing complexity of time-consuming pre-trial practice, trials, and sentencing proceedings, as well as district judge assignments to court of appeals cases. Absent more definitive data, it seems unwarranted to conclude that district judges are sufficiently underutilized that they may absorb the extra work contemplated by this recommendation. Indeed, overall the proposal may require even more judicial resources than are now required at both the district court and court of appeals level, because in at least some instances the court of appeals would grant permission to take a further appeal after a DCAP decision and would in any event have to consider requests for the exercise of discretionary review. Thus, the courts (as well as the parties) could incur the expense of conducting two appeals instead of just one before seeking Supreme Court review.
Second, the draft report's proposal calls for judicial councils, rather than Congress, to determine the class of cases to be adjudicated by DCAPs. That assessment, however, involves policy decisions about the nature of the underlying legal disputes, including a substantive evaluation of the applicable law. Such significant policy decisions, such as whether diversity cases should be handled in a distinctive manner, should be made by Congress, rather than by the judicial councils.(10)
Moreover, we question whether the administration of justice would be served by creating a class of appellate courts inferior to circuit courts of appeals and assigning cases deemed to be less significant to them. Certainly service on such courts is not made to seem attractive as described in the draft report, since it seems unlikely that a circuit judicial council would assign the most interesting classes of cases to any court other than its own court of appeals. Finally, the draft report states that "[f]urther review of decisions by a district court appellate panel would be in the courts of appeals, but only by leave of that court." Draft Report at 55. Such discretionary review raises the possibility that a litigant might be foreclosed from having the right to seek Supreme Court review of a decision that the court of appeals declined to review.
For the foregoing reasons, the Department opposes the proposal to create District Court
Appellate Panels. At a minimum, the recommendation should be adopted only as a temporary
pilot project that would operate in a single court, in carefully and explicitly designated categories
of cases selected by Congress, and only for a limited period of time.
E. Bankruptcy Appeals
Draft Report Recommendation:
The Commission recommends that Congress not implement the recommendation of the
National Bankruptcy Review Commission (NBRC) that would provide for appeal of certain
decisions of bankruptcy judges directly to the regional court of appeals. Draft Report at 57. The
Commission does, however, suggest three alternatives to the NBRC recommendation for
consideration by Congress, each of which would substantially reduce the district court's
oversight of bankruptcy judges. Id. at 59-61. Under current law, bankruptcy judges may issue
final orders or judgments only in "core" matters arising directly under the bankruptcy laws or
arising in a bankruptcy case and in "non-core" matters in which the parties consented to the entry
of the final order or judgment by a bankruptcy judge. 28 U.S.C. § 157(b)(1), (c)(2). In "non-core" matters, absent the consent of the parties, a bankruptcy judge may enter only a
recommended decision. The final order or judgment is entered by a district judge following de
novo review of the bankruptcy judge's proposed findings of fact and conclusions of law after
considering those matters to which any party has timely and specifically objected. Id. §
157(c)(1). In those proceedings in which a bankruptcy judge enters a final order or judgment --
that is, core proceedings and non-core proceedings in which the parties consented to the entry of
the final determination by a bankruptcy judge -- the parties may appeal that order or judgment to
the district court. Id. § 158(a). In lieu of district court review, the parties may consent to review
by a bankruptcy appellate panel (BAP) if the judicial council for the circuit has created a BAP
service and the district judges for the district in which the appeal arises have authorized the
service to hear and determine appeals. Id. § 158(b). In all circuits that have created a BAP
service, bankruptcy appeals automatically are referred to the BAP unless one of the parties "opts
out" of the BAP to appeal to the district court.
As we understand it, each of the Commission's proposals would reduce or eliminate the
district court's role in reviewing bankruptcy judge decisions in core proceedings and in non-core
proceedings in which the parties consented to the entry of the final order or judgment by a
bankruptcy judge. Under the first proposal, an appeal from a final order or judgment of a
bankruptcy judge would be taken directly to the court of appeals, but the court could employ
BAPs to hear the case in the first instance and to prepare a Report and Recommendation for its
review. Unless a party objected or the court disagreed with the BAP's recommendation, the
Report and Recommendation would be adopted as the court of appeal's decision. Under the
second proposal, litigants could opt for review of a bankruptcy judge's final order or judgment
(and certain interlocutory orders) in either the BAP or the court of appeals. Parties would
automatically appeal to the BAP unless they "opted out" of the BAP to go to the court of appeals.
If the litigants chose BAP review, the court of appeals would have discretion to review the
BAP's decision; if the litigants chose court of appeals review, the court could still employ a BAP
to prepare a Report and Recommendation for its review. Under the third proposal, litigants
would be required to seek review of a bankruptcy judge's final order or judgment (and certain
interlocutory orders) with the BAP before proceeding to the court of appeals. Under all three
proposals, district courts would still provide de novo review of bankruptcy court findings of fact
and conclusions of law in non-core matters in which the parties had not consented to final
determination by the bankruptcy judge. The court of appeals would have jurisdiction of appeals
from district court rulings.
Department of Justice Comments:
The Department concurs in the recommendation that Congress not adopt the NBRC proposal, but opposes the alternative suggestions for changes in bankruptcy appellate procedure.
We have concerns about the implications of the proposals. All three alternatives provide for expanded use of BAPs and eliminate or curtail an appellate function for the district court. In core proceedings and non-core proceedings in which the parties have consented to a final determination by a bankruptcy judge, all of the proposals would permit courts of appeals to employ BAPs to determine appeals in the first instance, even without the parties' consent. In the Department's experience, while BAPs are well suited to reviewing cases that involve the bankruptcy courts' discretion, their specialized knowledge is of far less value in appeals that involve pure questions of law, including the construction of state laws, non-bankruptcy federal statutes, and constitutional issues. District courts and courts of appeals have greater experience, as generalists, in dealing with those types of issues, and therefore are better suited to adjudicating cases involving them.
In addition, the draft report's proposals would produce inconsistent treatment of bankruptcy litigants. District courts would still provide de novo review of bankruptcy court findings of fact and conclusions of law in non-core matters in which parties had not consented to a final determination by the bankruptcy judge, and such matters could be appealed directly to the court of appeals. All other parties would ordinarily have to proceed from the bankruptcy court to a BAP. That inconsistent treatment seems anomalous. Moreover, a district court may withdraw the reference of an entire bankruptcy case or a core bankruptcy proceeding. Presumably, parties appealing core matters determined by a district court judge would proceed directly to the court of appeals; parties appealing core matters determined by a bankruptcy judge would not have that option. Accordingly, there is potential for even core matters to be treated differently on appeal, depending on whether they had been initially adjudicated by a district judge or a bankruptcy judge. Litigants may be tempted to seek withdrawal of the reference of bankruptcy proceedings to the district court to avoid BAP review.
Finally, we urge the Commission carefully to consider the constitutional implications of
the draft report's proposals in light of Northern Pipeline Construction Co. v. Marathon Pipe Line
Co., 458 U.S. 50 (1982); Commodity Futures Trading Comm'n v. Schor, 478 U.S. 833, 851-53
(1986); Granfinanciera, S.A. v. Nordberg, 492 U.S. 33, 56 n.11 (1989), and other relevant
precedents. Such a study is particularly warranted to avoid triggering the kind of disruption to
the bankruptcy system occasioned by constitutional challenges brought after enactment of the
1978 Bankruptcy Code. At a minimum, any change in this area will likely generate substantial
litigation.
F. The Court of Appeals for the Federal Circuit
Draft Report Recommendation:
The draft report addresses three types of cases "that have been frequently discussed as
potential candidates for the Federal Circuit's jurisdiction." Draft Report at 63-65. Those include
tax appeals, copyright cases, and Social Security appeals. Although the draft report makes no
explicit recommendation, it seems to suggest that those types of cases should be transferred to
the exclusive jurisdiction of the Federal Circuit.
Department of Justice Comments:
The Department believes that any proposal to create specialized subject-matter courts or
to add to the jurisdiction of the United States Court of Appeals for the Federal Circuit should not
be viewed as a panacea for easing appellate dockets. Each such proposal should be analyzed on
its own terms, and should not be implemented without careful study and deliberation. On
balance, we believe all three categories of appeals discussed in the draft report would best remain
with the regional courts of appeals.
1. Tax Appeals
The Department does not favor placing tax appeals under the jurisdiction of the Federal Circuit or any other specialized court of appeals.(11) First, the removal of tax cases from the jurisdiction of the general courts of appeals would not significantly reduce the caseload of those courts. Tax cases do not constitute a significant portion of the courts of appeals' dockets, and there is no reason to believe that the number of appeals in tax cases will increase substantially in the future. Nor has the Department perceived any significant problem with unresolved circuit conflicts in tax cases.Moreover, the administration of justice in tax cases would not benefit from assigning them to the Federal Circuit. The Federal Circuit has not had the opportunity to develop any more specialized expertise with respect to tax issues than other circuits and in fact has considerably less than most of them. While the Court of Federal Claims has concurrent jurisdiction with the district courts over tax refund litigation, the number of appeals from the Court of Federal Claims to the Federal Circuit in tax cases has been quite small compared to those going from the district courts and the Tax Court to the regional courts of appeals.(12) The Federal Circuit has virtually no experience with broad categories of tax litigation, such as criminal cases, summons enforcement cases, bankruptcy cases, and lien and levy disputes.
In addition, a strength of our appellate system is that it allows generalist judges to draw
upon their expertise in a full range of legal matters and apply the lessons they have learned to the
development of the law in a specialized area. Tax disputes touch on virtually every aspect of the
law and economic life of the Nation. Because of the specialization of its jurisdiction, the Federal
Circuit is largely isolated from legal developments in the areas of general law that bear on the
resolution of tax issues. The Department believes that, in the long run, the development of tax
law would be better served by leaving appellate review of tax cases primarily to the regional
courts of appeals. Finally, a requirement that tax appellate parties incur the costs of travel to
Washington, D.C., instead of to a regional court closer to their homes is likely to create a
perception of unfairness that may affect public confidence in the system for resolving tax
disputes.
2. Social Security Appeals
In the Department's view, the Federal Circuit also is not the appropriate forum for Social
Security appeals. As with tax cases, Social Security cases do not make up a significant portion
of the caseload of the general courts of appeals. Even in the boom period of the mid-1980s,
Social Security and benefit repayment cases accounted for only five percent of filed appeals,(13)
and the courts of appeals have developed methods for disposing of most such cases under
summary procedures. Moreover, Social Security appeals normally involve fact-intensive review
of benefits eligibility, a task that does not require the services of a specialized court, especially
when the appellants are not in a position to pay the expenses of travel to Washington for a
hearing. Finally, Social Security cases sometimes present important and difficult issues of
statutory, administrative, and constitutional law. Appeals presenting issues of those types should
be decided by the generalist judges of the regional courts of appeals.
3. Copyright Cases
The Commission notes that "some have suggested that it would be sensible to have patent
and copyright claims ultimately adjudicated in the same court." Draft Report at 65. Adjudication
of copyright claims, however, almost never requires the same level of technical background that
patent cases require. At the same time, as the Commission recognizes, copyright law involves
some of the most important and complex legal issues raised in today's world, and must do so
within a framework of dynamic change. Although we have not yet had the opportunity to study
this issue as carefully as we intend to, we believe that copyright adjudication may well best be
served by leaving appellate review of copyright cases to the generalist judges of the regional
courts of appeals, where the cutting-edge issues of this increasingly important area of the law can
be variously presented, decided and analyzed, creating a broad and rich record for Supreme Court
review when necessary.
III. CONCLUSION
The Commission has performed a valuable service in studying the United States court of
appeals system and proposing ideas for its future organization. In our view, however, the draft
report does not substantiate a present need that mandates structural, as opposed to non-structural,
reforms. The modifications suggested in the draft report themselves risk creating greater
inconsistencies in the law and delay in the system. Accordingly, while we believe that
circumstances may eventually warrant the adoption of structural changes, other measures should
be tried first, and the need for proposals of the type suggested in the draft report evaluated in
light of that experience.
November 6, 1998
1. In that connection, it might be useful for the Commission to elaborate on some key questions if it were to adopt the proposal in the draft report: What does the phrase "square conflict" mean (or have to offer) given the subjectivity of the inquiry and historical differences in view (even among Supreme Court Justices) about when a conflict is present? Would it be the same standard used by the Supreme Court in the certiorari process? Would the Circuit Division be required to resolve a conflict expressly recognized by a Division panel? Should reasons for denying a Circuit Division rehearing petition be required of the judges to provide notice to litigants? Should those reasons, if provided, include an indication of what asserted conflicts are totally nonmeritorious (and therefore should not be asserted in the future), simply distinguishable, or tolerable until a suitable case emerges to resolve the conflict?
2. Other comments on the proposed Circuit Division arrangement, which would permit four judges to set the law for the entire Circuit, are set forth, infra, in Sections II.A.2.b and II.A.3.
3. Although it is difficult to demonstrate a "square conflict" between two or more judicial decisions, our experience opposing petitions for a writ of certiorari in the Supreme Court suggests that a large number of litigants nonetheless will try.
4. That concern is not theoretical. In the area of criminal law, the Supreme Court in recent Terms has reversed decisions of the Ninth Circuit in which that Circuit alone has held the particular view of the issue presented and been in conflict with every other circuit to have considered that issue. See United States v. Ramirez, 118 S. Ct. 992 (1998), rev'g, 91 F.3d 1297 (9th Cir. 1996); United States v. Hyde, 520 U.S. 670 (1997), rev'g, 92 F.3d 779 (9th Cir. 1996); United States v. Watts, 519 U.S. 148 (1997), rev'g, 78 F.3d 1386 (9th Cir. 1996) and 67 F.3d 790 (9th Cir. 1995); United States v. Armstrong, 517 U.S. 456 (1996), rev'g, 48 F.3d 1508 (9th Cir. 1995) (en banc); United States v. Mezzanato, 513 U.S. 196 (1995), rev'g, 998 F.2d 1452 (9th Cir. 1993); United States v. Shabani, 513 U.S. 10 (1994), rev'g, 993 F.2d 1419 (9th Cir. 1993); United States v. X-Citement Video, 513 U.S. 64 (1994), rev'g, 982 F.2d 1285 (9th Cir. 1992); United States v. Padilla, 508 U.S. 77 (1993), rev'g, 960 F.2d 854 (9th Cir. 1992); see also Almendarez-Torres v. United States, 118 S. Ct. 1219 (1998) (overruling United States v. Gonzalez-Medina, 976 F.2d 570 (9th Cir. 1992)); Neal v. United States, 516 U.S. 284 (1996) (overruling United States v. Muschik, 49 F.3d 512 (9th Cir. 1995)). A process that insulated from Supreme Court review those types of erroneous division panel decisions that conflicted with other circuit decisions would be unfortunate. In our view, rather than creating a structure that might insulate such decisions from Supreme Court review, the Ninth Circuit should employ a more vigorous en banc procedure to address those types of conflicts and erroneous decisions.
5. The draft report states that this type of situation exists in California at the state court level. But the fact that California may allow for such divergence among its own appellate courts in a unitary judicial system does not mean that federal law, in its establishment of the overlay of a federal court system, should impose a regime of divergent precedent on a single State.
6. We have not had an opportunity to assess to what extent the geographical division of California would create the possibility of conflicting jurisprudence on a range of substantive areas of law of particular interest to the United States, such as public land and ecosystem management, wildlife and marine resource issues, and Native American rights and interests, but consideration of such effects should be undertaken before any division plan is implemented.
7. We note that while the draft report proposes a Circuit Division whose "only authority would be to resolve square inter-divisional conflicts" (Draft Report at 42), the draft bill is slightly different, providing that "the Circuit Division may review the decision of a panel within a division only if en banc review has been sought and denied by the division." Id. at 83, Appendix C, § 3(a).
8. The Commission's view is that "the maximum number of judges for an effective appellate court functioning as a unitary decisional unit is somewhere between eleven and seventeen." Draft Report at 27. That conclusion appears to be based on the perceptions of a majority of those circuit judges who expressed an opinion. Ibid. & n.66. While our views admittedly reflect the perspective of litigants, not judges, the Department's experience is that size -- whether considered in terms of numbers of judges or geography -- does not necessarily determine how well a court can function.
9. In addition, the proposal to create divisions
in the courts of appeals may result in the development, over time, of even more
complex and varied local rules of procedure. The Department has worked extensively
with the Advisory Committee on Appellate Rules to develop simplified, centralized
rules of appellate procedure and to reduce the number and range of local appellate
rules. The draft report proposes that all federal courts of appeals, however
organized, use the same procedural rules. We fully endorse that position. Nonetheless,
the draft report's recommendation gives considerable flexibility to the courts
of appeals in creating independent divisional systems that enable circuits to
"implement the divisional arrangement in different ways according to their needs."
Draft Report at 53. Thus, we remain concerned that the proposed structural rearrangement
could derail efforts to develop nationally uniform procedural rules.
Moreover, the considerable leeway afforded to circuits other
than the Ninth to develop divisions does not foreclose the possibility that
circuits might create special subject-matter divisions. For the reasons stated
in our original submission and in this submission, we would be concerned about
the creation of subject-matter divisions. Such a possibility would add an element
of potentially great variability in practice and procedure among different areas
of practice.
10. The draft report does not contain any formal recommendation concerning how diversity cases should be treated, so we have not included an analysis of that issue in this submission.
11. The draft report states that the Senate Judiciary Committee received substantial support in 1979 for the creation of a court of tax appeals, and mentions a similar recommendation made by the Federal Courts Study Committee in 1990. Draft Report at 64. The statement that the 1979 recommendation garnered "substantial support" is inaccurate. The Department of Justice and the organized tax bar as a whole vigorously criticized the court of tax appeals concept and have consistently maintained that the tax system would suffer in the event that tax appeals were centralized in one court.
12. Over the last ten years, according to Department records, an average of only about 25 tax appeals per year have gone to the Federal Circuit, out of approximately 800 new tax cases each year in the appellate courts as a whole.
13. Federal Judicial Center, Stalking the Increase In the Rate of Federal Civil Appeals 10 (1995).