1. Background

     In early 1985, the Honorable Pierce Lively, then chief judge of the Sixth Circuit, asked the Federal Judicial Center to assist in evaluating the circuit’s pre-argument conference program, which had been in operation for approximately three years.[1]

     Sixth Circuit Local Rule 18 establishes the following pre-argument conference procedure. Attorneys on the staff of the court conduct conferences, generally by telephone, with the attorneys in most civil appeals to (1) explore settlement possibilities, (2) resolve procedural issues, and (3) clarify issues in the appeal. Settlement negotiations often continue after the conference through follow-up efforts of the court’s conference attorneys. In particular cases, a pre-argument conference order may be entered specially tailoring briefing schedules. All conference-related discussions are kept confidential from the court.[2]

     After two years of the conference program, the court conducted an internal study to evaluate its performance, but the results proved inconclusive.[3] The court thought that a larger, more systematic approach was needed and decided to enlist the resources of the Federal Judicial Center.

     The Center decided to design the evaluation of the Sixth Circuit’s pre-argument conference program as a controlled experiment.[4] Opportunities for conducting formal experiments in an actual court setting are relatively rare and pose a number of challenges to the researchers and court.[5] Despite the challenges, the court opted for the experiment proposed by the Center.

     It should be noted that this study focused exclusively on the conferencing program in the Sixth Circuit, which differs considerably from programs in other courts. The Second Circuit’s Civil Appeals Management Plan (camp), for example, conducts the majority of its conferences face to face with counsel, whereas the Sixth Circuit holds more than 93% of its conferences over the telephone. There are other differences as well in the way in which conferences are conducted, the degree of follow-up, and how the programs are staffed.[6]


2. Objectives and Methodology

     The purpose of the study was to determine whether the Sixth Circuit conference program was meeting its stated objectives. Those were (1) to save judge time by facilitating settlement and early termination of cases without judicial involvement, (2) to lessen case management burdens by clarifying procedural matters, and (3) to simplify and clarify issues on appeal.

     The court also believed that the program was reducing procedural and substantive motions, and it asked the Center to determine whether this was the case. Finally, the court asked the Center whether any particular types of cases are more amenable to settlement than others, so as to learn whether program resources might be most productively focused on particular kinds of cases.

     After extensive discussions with staff of the Sixth Circuit, an evaluation strategy was implemented for studying the pre-argument program.The study was conducted using the control group method. Two groups were randomly drawn from the cases typically eligible for pre-argument conferences. These groups were a “treatment group” (the cases to be conferenced) and a “control group” (cases not to be conferenced). Using the fully randomized control group method enabled us to draw much stronger conclusions about the impact of the program than we could from a less methodologically rigorous approach. The case selection phase lasted for about seventeen months, beginning in March 1985 and ending when there were approximately 1,500 cases in the study. Conference-eligible cases included all civil appeals except prisoner, pro se, Social Security, agency (except for FDIC and Small Business Administration cases), original action, and tax court cases. The case types that were excluded were thought to be not generally amenable to settlement or were otherwise inappropriate for involvement in the conference procedure.


3. Summary of the Major Findings

     Like most innovative efforts, the Sixth Circuit’s pre-argument conference program operates in an environment of high expectations. This study was able to confirm that a number of the settlement and case management objectives of the conferencing program are being met. We were unable, however, to substantiate or confirm all of the expected program effects. This summary of the major findings of the study is organized around the major questions and issues that the study was expected to address.

     The objective of this report is to present the findings and conclusions of the study to the court in a clear and concise manner. Tables containing statistical information have been placed in Appendix A and are referred to in the appropriate discussions within the report.

Does the program increase the numbers of appeals that are settled, voluntarily dismissed, or dismissed for want of prosecution?

     Analysis of the study data points to a very clear and substantial effect on the workload of the court. We were able to establish that about 69% of the conference-eligible appeals in the control group reached argument or submission.[7] For appeals that were subjected to the pre-argument conference procedures, the number of cases argued orally or submitted on the briefs was reduced to 57%. (See Table 1.) This means that the program resulted in a reduction of 12% in the number of conference-eligible cases that would otherwise have been submitted.[8]

What is the impact of the 12% reduction on the workload of the court?

     For calendar years 1985 and 1986, approximately the period covered by the study, data from the pre-argument conference program indicate that a total of 734 cases in 1985 and 668 cases in 1986 received attention from the program. During this time the flow of cases into the program was reduced considerably by the experimental procedures, because one third of the appeals eligible for the conference program were assigned to the control group. Assuming that normally the three conference attorneys are able to process about 750 appeals a year, the estimates of program effect from the study period indicate that about ninety appeals a year are diverted from the court’s argument calendar.

     The implications of this finding to the court can be viewed in terms of savings in judge time. When recommending the creation of new appellate judgeships, the Judicial Conference Committee on Judicial Resources uses a standard of one active circuit court judge for every eighty-five appeals decided on the merits.[9] Using that standard, the data indicate that the conferencing program is doing the work of 1.06 appellate judges.

What degree of confidence can be attached to the 12% finding?

     We are very confident that our estimates of the program effect on the argument calendar are accurate. As with every study of this sort there is a range of uncertainty. In this instance, the range of uncertainty is quite narrow—the chances are better than nine out of ten that the pre-argument conference program diverted between 16.7% and 6.9% of the eligible appeals from the court’s argument calendar. (See Table 1.)

Are certain types of appeals more likely to be settled than others?

     The court asked whether any particular types of cases were more amenable to settlement than others. Because of the relatively small number of certain types of appeals, the analysis of the study data could not discern statistically significant program effects in settling any particular group of cases. The only category for which a statistically significant difference was observed involved diversity claims—the data show that these cases may be slightly less amenable to settlement efforts. On the strength of the study results alone, we cannot make any recommendations to the court concerning any particular case types that should be targeted over others for inclusion in the conferencing program. (See Tables 2 and 3.)

Are unsettled cases likely to be delayed as a result of the
pre-argument procedure?

     When a treatment case did not settle but went on for a judicial decision, the data indicate that briefing was delayed in 50% of these cases compared with 5% of the control group cases that required a judicial decision. (See Table 4.) Some of the delays in the treatment group cases were a result of the conferencing procedures and are consistent with the objectives of the pre-argument program. Briefing is extended when all parties consent and when the conference attorney and the parties perceive that ongoing negotiations offer a realistic possibility for settlement.

     The data also show that treatment cases took an average of twelve days longer to move from docketing to submission. (See Table 5.) We found no statistically significant difference between the treatment and control group in the time it took cases to reach disposition after being submitted to the court for decision. (See Tables 6–8.)

Does the pre-argument program terminate cases at an earlier stage in the appellate process?

     The data indicate that the program does terminate more cases at an earlier stage of the appellate process. About 23% more treatment cases than control group appeals terminate before the filing of the appellant’s brief or the joint appendix.

Does the pre-argument conference clarify procedural matters?

     The study was able to document that there were 14.5% fewer procedural motions filed in treatment cases than in control group cases. (See Table 10.) Our analysis found no statistically significant difference between treatment cases and control group cases in the number of dun notices the clerk’s office was required to send for typical procedural defects such as late briefs or want of prosecution. (See Table 11.)

     When asked whether the program provides assistance to counsel in complying with procedures of the court, 85% of the attorneys surveyed in the study responded in the affirmative. Additionally, 67% expressed the view that the program eliminates procedural motions and 57% reported net savings in the amount of time they had to spend on their appeal as a result of the conference procedure. (See Table 15.)

Does the program appear to simplify or clarify issues?

     The conference attorneys perform a number of functions that are expected to simplify and clarify issues in the appeal. One measure of this effect may be the length of the appellate briefs. Our data showed no statistically significant difference between the length of appellants’ and appellees’ briefs in the treatment and control groups. (See Table 14.) We could not determine any effect from the program on the length of reply and supplemental briefs.

     Additional support for the observation that the program simplifies or clarifies issues comes from attorneys who were surveyed in the study. Over 77% of those responding indicated that the program clarifies issues in the appeal, with 61% noting that conferencing helps to eliminate issues. (See Table 16.)

Does the pre-argument program reduce the
number of motions that are filed?

     The court asked us to measure whether the pre-argument conference program is making appeals less complex by reducing the number of motions that are filed. As reported above, study results indicate that there were 14.5% fewer procedural motions filed per case in the treatment group. Similarly, the study found a reduction in substantive motions, with the treatment group showing 21.6% fewer substantive motions filed per case than in the control group. (See Table 10.)

     Sixty-seven percent of the attorneys responding to our survey indicated that the program reduced procedural motions. Fifty-six percent of the respondents, on the other hand, believed that the program did not reduce substantive motions. (See Table 15.)

Does the program find support among the bar?

     The study confirmed that there is very strong support among the bar for the program. (See Tables 15–17.) Responses were received from about 88% of the attorneys surveyed. Nearly 80% of these attorneys indicated that in the absence of the pre-argument program, they would not have taken the initiative to approach the opposing side about settlement. Over 50% of the respondents expressed the view that the program results in net savings in time spent on the appeal; only 9% thought the conference procedure increased the time they spent on the appeal. Of the lawyers who have prosecuted appeals with and without the benefit of the conference process, 84% preferred conference program involvement.[10]

     As previously reported, over 67% of the attorneys felt that the pre-argument conference reduced or eliminated procedural motions, though only 15% saw a similar benefit with respect to substantive motions.

     The overwhelming majority of the attorneys who responded to our survey expressed their strong preference for telephone conferences rather than in-person conferences; most pointed to significant savings in time and effort. Similarly, the majority of attorneys did not favor changing the program to require client participation in the conferences.


4. Description of the Program

     The decision to establish a pre-argument conference program in the Sixth Circuit grew largely out of a determination to increase productivity without having to increase substantially the size of the circuit’s bench.[11] Experiencing significant annual growth in the number of appeals filed, the circuit faced a major challenge: how to increase productivity without increasing substantially the number of appeals for which oral argument was needed. The circuit had already undertaken efforts aimed at reducing the number of frivolous, insubstantial, or jurisdictionally defective appeals.

     A number of other case management innovations were explored. Through the efforts of then Chief Judge George Edwards and Circuit Executive James Higgins, the pre-argument conference program came to be viewed as a viable option for enhancing the circuit’s ability to respond to its growing docket. In an effort to better assess the merits of a pre-argument conference program for their circuit, site visits were made by Judge Edwards and Mr. Higgins to the Second and Seventh Circuits.

     The results of their investigation convinced them that a conferencing program would work in the Sixth Circuit. Both programs were, at varying degrees, aimed at settlement and case management. The Second Circuit’s camp program emphasizes settlement, whenever possible, and focuses to a lesser extent on case management. When a settlement is not possible, camp provides for tailored briefing schedules to be issued by the conference attorneys setting deadlines for filing briefs and appendices. The program in the Seventh Circuit, on the other hand, focused primarily on refining issues and addressing procedural matters in the appeal.

     Borrowing features from the programs in the Second and the Seventh, the Sixth Circuit began its program on a trial basis in 1981. The goal was to settle 150 appeals that would have otherwise been submitted.

     The program in the Sixth Circuit is based on two important assumptions about how to settle cases in an appellate setting. One is that lawyers are frequently reticent about initiating settlement negotiations for fear of appearing weak. The other is that the appellate process, unlike that at the trial level, presents few opportunities for the parties to pursue settlement. For example, there are no opportunities at the appellate level for depositions or pretrial conferences during which attorneys for the parties might systematically pursue settlement discussions. The pre-argument conference, therefore, was designed as a vehicle for prompting the attorneys to explore settlement and other related case-management objectives. It is clear that the program does function to get the parties talking about settlement.

     In late 1981, Robert Rack, Jr., was appointed as the senior conference attorney. The staff position was borrowed from the staff attorney’s office along with one secretarial slot. A slot for another conference attorney was transferred from the clerk’s office to the conference program. A third conference attorney position was later authorized by the Administrative Office. When Judge Edwards assumed senior status, one of his two authorized secretarial positions was transferred, at his suggestion, to the pre-argument conference program.

Program objectives

     The main objective of the program is to reduce the judicial workload by increasing the number of appeals that are settled, voluntarily dismissed, or dismissed for want of prosecution. For cases that would have been dismissed anyway, it was hoped that the conference procedure would prompt dismissals earlier in the appellate process. The program also focuses on appeals that do not settle. The objective with this group of appeals centers around clarifying procedural matters for counsel, and simplifying or clarifying issues on appeal.

During the initial months of the program, judges
conducted the conferences

     During the initial months of the program, pre-argument conferences were conducted by active and senior judges of the circuit.[12] After about six months, the conduct of the pre-argument conferences was assumed by the conference attorneys, and the program has since been run almost exclusively by the attorneys.

     It is felt that the initial involvement of the bench resulted in a number of direct benefits to the program. Given that each of the judges tended to differ in the approach and manner in which the pre-argument conferences were conducted, the conference attorneys were able to observe varying styles for conducting the conferences. In addition, judicial involvement demonstrated to the bar that there was significant interest in the program among the judges.

     In 1983, the court voted to conclude the pilot status and to implement the pre-argument conference program on a more permanent basis. Subsequently, efforts were made to fund the program as an independent entity within the court. Following the submission of a funding proposal, Judge Edwards and Senior Conference Attorney Rack appeared before the then Subcommittee on Supporting Personnel of the Committee on Court Administration of the Judicial Conference of the United States to request permanent funding for the program. The request was approved, and program funds were included in the Judiciary’s 1986 appropriation for the following fiscal year.[13]

Nearly all conferences are conducted by telephone

     While in-person conferences are permitted in the Sixth Circuit, the overwhelming majority of the conferences are conducted by telephone. This becomes a major distinguishing feature between the Sixth Circuit’s program and that of camp in the Second Circuit. In the Sixth Circuit, case filings are widely dispersed throughout a four-state area. Consequently, in-person conferences would involve substantial travel costs to the parties or the court. Experimental telephone conferences were found to be effective and were adopted as the routine method of conferencing.

     In the early days of the program, pre-argument conferences were conducted by the conference attorney with only one side of the appeal at a time. It quickly became apparent that this approach resulted in a larger number of calls and callbacks. The approach was changed to allow for pre-scheduled telephone conferences to be conducted with multiple parties in a single call.[14] The conference attorneys are able to conduct discussions with up to seven parties at a time. When circumstances dictate, discussions are held with attorney(s) for one side, while the other parties to the call are placed on hold.

     Approximately 6% of the appeals in the program are settled by the conference attorneys at the first conference or immediately following it. A second telephone conference, again involving attorneys for all parties to the appeal, is conducted by the conference attorneys in approximately 11% of the cases in the program. The great bulk of the conference attorneys’ time and efforts involve follow-up contact. For 80% of the cases that do not settle at the first conference, follow-up contact of some sort is made by the conference attorneys. With rare exceptions, follow-ups almost always involve contact with only one party to the appeal at a time.

Pre-argument conference matters are kept confidential

     Care has been taken to insulate and isolate the functioning of the pre-argument conferencing program from other parts of the court, especially the judges. By design, no information about specific pre-argument conferences is transmitted to the judges. This confidentiality helps to ensure that the parties are not constrained in their discussions with the conference attorneys. In addition, the conference attorneys are able to explore settlement prospects more fully than otherwise might be the case.

     In documents sent to the parties and at the beginning of every pre-argument conference, the parties are informed and reminded that no aspect of the discussion of the issues, the merits of the appeal, or the prospects of settlement is to be communicated to the court. From our observation of the conduct of a number of conferences, it is clear that the assurances of confidentiality set the stage for open and candid discussions during the conferences.

     The senior conference attorney reports to the circuit executive on all administrative matters and to the chief judge on matters of policy. All documents and records relating to the pre-argument conferences are maintained in separate, secure files in the offices of the conference attorneys.

     At no time during the course of the study did we find evidence of any activity that would suggest that confidentiality requirements had been compromised in any of the study cases.

Most civil appeals are eligible for pre-argument conferences

     Over the years, the experience of the conference attorneys suggested that the objectives of the pre-argument program are more likely to be reached in certain types of cases than in others. Three categories of appeals are excluded routinely from the program: (1) prisoner civil rights and habeas corpus petitions, (2) pro se cases, and (3) most agency cases (except when requested).

     Prisoner civil rights and habeas corpus petitions are viewed as especially difficult to settle or even as inappropriate for conferencing. The pro se cases were excluded largely to avoid problems that tend to arise when conducting settlement negotiations with parties who are not trained in the law.[15]

     Agency cases are excluded because the conference attorneys had considerable difficulty conducting settlement negotiations with attorneys for federal agencies. Settlement negotiations with the federal government are often difficult because of the number of layers of review required before authority can be obtained to settle. Also, the government’s position is frequently based more on principle than on expediency, making compromise difficult. In benefit eligibility cases, for example, the government generally takes the position that the claimant either is entitled to the benefit or is not and will not consider splitting the benefit to settle. However, agency appeals are conferenced if a specific request is made.

The pre-argument statement

     Local Rule 18(b) requires the appellant to file an original and two copies of a pre-argument statement with the clerk of the Sixth Circuit within fourteen days after the notice of appeal is filed.[16] All other parties to the case are then served with a copy of the pre-argument statement. The pre-argument statement is routed directly by the clerk’s office to the conference attorneys.

     In addition to the names and addresses of the attorneys representing the parties, the pre-argument statement contains information necessary for an understanding of the nature of the appeal, including (1) the nature of the trial court’s decision; (2) the relief denied or obtained below; (3) a statement of the jurisdictional basis of the appeal; (4) issues to be raised in the appeal; (5) any pending related appeals or cross-appeals; (6) an indication of any court cases or statutes that may be crucial to a resolution of the appeal, including whether the case presents a potential conflict of law within the circuit or between the Sixth and some other circuit; and (7) any other information that would help in understanding the appeal.

     Side one of the pre-argument statement provides information about the action in the trial court. In addition to the caption, district court docket numbers, date of filing, and the names, addresses, and phone numbers of counsel for all parties, it also identifies the district judge, the basis of federal court jurisdiction, the nature of the case, and the district court’s disposition.

     Side two of the pre-argument statement requires the appellant to provide more subjective information about the issues in the appeal, along with the citation of any cases or statutes the interpretation or application of which the appellant feels will determine the outcome of the case. The appellant is also asked to indicate whether the case presents a possible conflict of law within the Sixth Circuit or between the Sixth and other circuits. Information is also requested as to whether there are any other pending cases in the court of appeals or district court that relate directly to the instant action.

The decision to schedule a pre-argument conference

     For the majority of civil appeals in the Sixth Circuit, the senior conference attorney decides which cases will enter the pre-argument program and whether such cases should be conferenced by one or two attorneys. During the case selection phase of the study, all civil appeals, except those excluded from the program, were randomly assigned to conferencing. This represented a major departure from the usual case selection and assignment procedure.

     The second and most frequent method is by request of either or both parties. Currently, the clerk sends out a notice about the program to all parties in civil appeals. This notice advises of the existence of the program and authorizes requests for conferences. The fact that a conference has been requested is not disclosed to the non-requesting party. The scheduling and handling of these cases is otherwise identical to the others.

     Another, and somewhat rare, method through which cases enter the program is by referral by the court before, or even after, oral argument.

     During the first two years of the program, the conference attorneys attempted to identify appeals with great potential for settlement. Cases thought to have a strong chance of settling were then scheduled for pre-argument conferences. After reviewing the results of these efforts, no increased settlement or enhanced case management patterns could be discerned for any particular case type or group of cases. Consequently, the court decided to eliminate the screening for great potential for settlement.

     Normally, cases to be conferenced are assigned to one of the three conference attorneys by a secretary, following verification of the names and addresses of the attorneys for all of the parties in the case. Before making the assignment, the secretary reviews the files in the clerk’s office to determine whether transcripts have been received and whether there have been any briefs or motions in the case.

     If an appellate brief has been filed, a pre-argument conference is not normally scheduled, on the theory that the prospects of settlement diminish as the parties’ investment in the appeal increases. Preparation and filing of a brief represents a significant financial and psychological investment in an appeal.

     If no briefs or motions have been filed, the case is scheduled for a conference to take place in two to three weeks. The secretary assigns the case to the conference attorney having the next available opening on the conference program calendar.

     Following assignment to a conference attorney, notices indicating the date and time of the conference are sent to the attorneys listed on the pre-argument statement. The notices to the attorneys also explain the purposes of the pre-argument conference as being (1) to identify and attempt to resolve any matters that may interfere with the smooth handling or disposition of the proceeding, (2) to clarify issues presented in the appeal, and (3) to explore possibilities of settlement.[17] Finally, the notice advises the attorneys to be prepared to discuss the merits of the case for settlement purposes and that each attorney should be certain to have the authority to terminate the litigation consistent with the interests of his or her client.

     The pool of conference-eligible civil cases tends to vary somewhat. Occasionally, appeals are assigned for conferences that would be conducted on a date that comes after the briefing date. This occurred primarily during the study period, because of delays caused by the randomization procedure.[18] In all such instances, the conference attorneys usually extend the briefing date until after the date on which the conference is scheduled. Any changes in the scheduling of a conference because of a conflict in the schedule of an attorney representing one of the parties is done by the conference attorneys’ secretaries.

     Most appeals are assigned to a single conference attorney. It is his or her responsibility to review the pre-argument statement as well as the initial docket sheet and, if available, the record and opinion below. Until April 1989, two conference attorneys were involved in the conduct of the initial conferences in about 30% of the cases. This was done with the expectation that the involvement of two conference attorneys would manage multiple parties and issues more effectively. Generally, a single conference attorney conducts all follow-up discussions to the initial conference. The study did not discern any increase in settlements in co-mediated cases. This was largely because less than 6% of all appeals settled at the initial conference.

Post-conference activity[19]

     If there is any reasonable likelihood of settlement and the parties do not object, the filing date for the briefs may be extended to enable the parties to reach a settlement. This practice is based on the assumption that chances of settlement diminish once briefs are written. The conference attorneys believe the power to extend briefing serves as a valuable tool in facilitating settlements.

     The extensions are usually short—one to three weeks—but in some instances they may be as long as a few months if settlement is likely and some action is required that is not within the control of the parties, such as a decision from another court. If a lawyer seeks an extension of time for briefs in a case where the conference attorney does not see serious good-faith settlement negotiations or where a party objects, the lawyer will be required to file a motion for enlargement of time with the court.

     The conference attorney continues to be involved in settlement negotiations in a case until it settles or until he or she concludes that the case is not going to be settled or that his or her participation is not likely to have any effect. Sometimes this is immediately clear as a result of the conference; sometimes it is not clear until the lawyers have consulted further with their clients, with the conference attorney privately, or with each other.

     Usually the lawyers communicate their clients’ responses directly back to the conference attorney, who ensures that all offers are answered and that all responses include some explanation or rationale.

     When settlement agreements are reached, the conference attorneys send form stipulations to dismiss to counsel for their signature and return by a given date. Again, if an extension on a brief or stipulation due date is necessary to enable the parties to execute settlement, that time is given.

     Copies of the stipulation transmittal letters, which set forth the due dates, are sent to the deputy clerks responsible for monitoring the briefs in the case, and a docket entry is made of the due date and monitored by the clerk’s office. Generally, once these stipulations are mailed the conference program considers the case terminated, taking no further action or interest in the case unless contacted by one of the parties or the clerk’s office. If the stipulations are not returned by the due date, the clerk issues a show cause order, eventually dismissing the case for want of prosecution.

How a typical pre-argument conference is conducted

     The pre-argument conference is the focal point of all program activity. The attorney conducting the conference begins by explaining clearly the rule and expectations for confidentiality. He then proceeds to call counsel’s attention to the Sixth Circuit’s sometimes troublesome joint appendix procedure, Local Rule 11, which provides for the preparation and submission of a deferred joint appendix. Counsel are urged to call the clerk’s office with any problems or questions they encounter in the preparation of the appendix.

     Counsel are then invited to raise any procedural problems they have or anticipate in the appeal that might require motions work or delays and that might be susceptible to resolution by agreement of the parties. Typical problems raised at this time include, but are not limited to, difficulties with obtaining transcripts; questions regarding the transmittal of exhibits; the necessity for particular portions of transcript or documents in the record; modifications of the routine briefing schedule, particularly in cross-appeals and cases involving multiple parties and issues; the consolidation or grouping of cases; and questions of jurisdiction, often related to interlocutory appeals and appeals from non-final orders.

     Questions pertaining to records and transcripts are either answered or referred to the clerk’s office. Briefing schedules are established by agreement of the parties under the authority of Local Rule 18 and are transmitted to the clerk’s office for implementation. Jurisdictional questions are explored and may result in a joint stipulation to dismiss or the filing of a motion to dismiss or a motion in the district court for Local Rule 54(b) or § 1292(b) certification to take an interlocutory appeal.

     In cases where the lawyers express a preference to waive oral argument and submit their case to the court on briefs, a form joint motion to waive oral argument is provided with the understanding that the hearing panel will decide whether to waive or require oral argument.

     The conference attorney then explains that the primary purpose of the conference is to examine issues being brought to the Sixth Circuit and possible grounds for settlement. Counsel for appellant may be asked to explain specifically where he or she believes the district court erred. Opposing counsel is invited to respond, and discussion usually ensues concerning the law and facts of the case as necessary to assess the possible outcomes of the appeal.

     Counsel are asked about their settlement efforts since the trial court decision and about their clients’ settlement needs. The conference attorney may then engage in a series of private discussions with the lawyers for each of the parties until he or she has a reasonably accurate perception of each side’s settlement needs and flexibility. If settlement appears impossible, the parties are advised of this and the conference is terminated.

     Few cases settle during a first conference. Usually, settlement proposals are developed later in joint and separate discussions with counsel. Often the lawyers will be asked to consult again with their clients in light of the discussions in the conference. In some instances, the conference attorney may propose a specific settlement, asking each lawyer to seek authority from his client to accept it. Second conferences are sometimes scheduled. Data from the conference attorneys indicate that an average conference lasts about an hour, with some lasting several hours.

In-person conferences

     Nearly all of the conferences held in the treatment group cases were conducted by telephone. Before the study was begun, some conferences were held in person, either in the conference program’s offices or in federal courthouses in the cities where counsel reside. In-person conferences held outside of Cincinnati were scheduled in batches of five or more over two days in districts that generated enough appeals to allow such scheduling.

     There has been discussion about increasing the number of face-to-face conferences in order to assess whether the settlement rate is sufficiently higher in those than in telephone conferences to justify the additional time and expense of travel. Procedures for the conduct of personal conferences are the same as for telephone conferences except counsel are required to have clients present or immediately available by phone to respond to settlement proposals as they arise.


5. Evaluation Methodology

     A computer-generated procedure was developed to randomly assign appeals to two groups, a treatment and a control. The random assignments were made from the universe of conference-eligible cases for which pre-argument statements were filed in the clerk’s office during the case selection phase of the study. Preliminary information on case filings in the circuit indicated that approximately one year of filings would yield samples of 1,000 treatments and 500 controls. In actuality, it took seventeen months of filings to generate the targeted sample of eligible appeals.

     All cases for which pre-argument statements were filed during the case selection period were eligible for inclusion in the study, except for the following types of appeals: (1) criminal; (2) prisoner; (3) pro se; (4) Social Security; (5) agency appeals, except FDIC and Small Business Administration appeals; (6) original actions; and (7) tax court cases. These were the only cases excluded from the study.

Random assignment of cases

     Our initial analysis indicated that approximately 1,500 cases would be needed in order to have samples that would generate findings within reasonably acceptable statistical boundaries. It took a total of seventeen months to generate the desired number of cases. Cases in the study were all filed over a period beginning on March 12, 1985, and ending on August 20, 1986.[20] This allowed a more than ample case selection period, thereby ensuring that any seasonal fluctuations in the circuit’s filings are reflected in the sample.

     We did not divide the universe of conference-eligible cases equally into two groups because we wanted to avoid the possibility of ending up with an insufficient number of cases necessary to fill the conference program’s calendar. In order to avoid that possibility, a decision was made to assign one third of the cases to the control group and two thirds of the cases to the treatment group. This allowed for a large enough pool of cases for the conference attorneys to work with during the case selection period. Statistical computations indicated that the projected sample could be expected to generate results within reasonably acceptable bounds of statistical confidence.

     A relatively straightforward random assignment process was employed to generate the two samples. At the time the pre-argument statement was filed, the case processing section of the clerk’s office made a copy of the form and indicated the case type, as set out on the docket sheet. In addition, a notation was made on the form to indicate whether the case was related, a cross appeal, or consolidated with another case. A notation was also made to indicate if the fee was paid and whether appearance and transcript order forms were filed. If the necessary documents had not been filed, the case was not placed in the pool for randomization until the filing requirements were satisfied. This helped to ensure that cases in the study were not in danger of being procedurally dismissed after randomization.

     The form was then reviewed to determine whether it met the case type criteria for inclusion in the study, as well as to ascertain if any related case, cross appeal, or consolidated case had already been randomized. If the case was the proper case type to be included in the study, the case title and number were entered into the computer for group assignment.

     Pre-argument statements for the treatment group then were given to the conference attorney. Summary listings of all control group cases and cases excluded from the study were provided to the conference attorney on a regular basis to ensure that these cases were not conferenced.

     A rule was developed for assignment of related, cross, and consolidated appeals.[21] The assignment of the lead case determined to which group any subsequent related, cross, or consolidated appeal would be assigned. Some conferenceable cases were deemed ineligible for inclusion in the study because they were related to a case in which a pre-argument statement was filed before the beginning of the study. In a few instances, it was learned sometime after group assignment that a particular case was a cross, consolidated, or related appeal. A decision was made to retain the original assignment for these cases, even though this may have meant the cases ended up in a different group than their lead cases.

     Cases with associated appeals having the pre-argument statement filed before the beginning of the study were automatically excluded if the associated case was active within three months of the beginning of the study. (This did not prevent the case from being conferenced.)

     For purposes of the study, we coded as conference activity any written or oral contact by the conference attorney’s office, including telephone contact, with any party in the case. Our principal source of information regarding what happened to cases assigned to the treatment group were internal assignment and termination forms (see Appendix C) maintained by the conference attorneys. The forms summarize case information at the time the conference attorneys complete their involvement with the case.

About the sample

     Using the procedure outlined above, a total of 1,082 cases were randomly assigned to treatment and 525 cases to the control group. As explained below, some adjustments to the sample were made during the randomization phase of the study, and we ultimately ended up with sample sizes of 1016 treatment cases and 509 controls.

     Of the initial group of 1,082 cases, a total of 66 appeals were deleted from the sample, largely because they had not been randomized and processed consistent with the study’s design. Most of the cases were removed from the sample after it was determined that the case was in one of four categories: (1) inadvertently randomized twice; (2) procedurally dismissed and should not have entered the study; (3) of a type that was excluded from the conferencing program, e.g., pro se or miscellaneous docket cases; or (4) never received by the pre-argument conference program.

     For cases in the last category, there were no records to indicate that the conference attorneys ever received notice of the cases being assigned to the program. Pre-argument statements were apparently never received by the conference attorneys, no program files on the cases were ever created, and no conferences were scheduled or held. Most of these cases were filed relatively early in the randomization process, when the study procedures were still new to all of the actors involved. There is no reason to believe that the inconsistent processing of the cases was in any way deliberate or that the problem affected the sample adversely.

Data collection

     The study’s evaluation design called for most of the case file data to be collected by the clerk’s office. Specific data collection procedures were discussed and agreed upon before the implementation of the study.

     The study used the following types of data:

(1)   conference activities logged and maintained by the conference attorneys, which included case-specific data on the frequency and nature of all program-related contacts;

(2)   case file information routinely maintained by the clerk’s office;

(3)   questionnaire data from surveys of the judges in the Sixth Circuit and attorneys who represented parties in cases that were conferenced; and

(4)   participant-observer data from a sample of conferences conducted during the course of the study.

     The clerk’s office distributed and collected the judges’ questionnaires for study cases that reached submission. Judges did not know whether a case was in the treatment or control group.

     The clerk’s office also mailed the attorney questionnaires, along with a cover letter from then Chief Judge Lively explaining the study. The attorneys were asked to send their completed questionnaires directly to the Federal Judicial Center and a franked self-addressed return envelope was included for that purpose.

     The conference attorneys maintained information in their files indicating cases in the treatment group actively conferenced; cases in the control group conferenced because of a referral or request;[22] and any treatment cases in which an extension to file a brief or appendix was granted by the program under Local Rule 18.

     As might be expected, there were important differences in the manner in which certain events are characterized by the clerk’s office and the conference attorneys. For example, cases with settlement stipulations as well as cases dismissed for failure to submit briefs were reported as conference program settlements by the conference attorneys. The clerk’s office reports all such cases as dismissals for want of prosecution under their Local Rule 4(f). Similarly, the clerk reports as judicial dispositions all joint motions to remand to the district court as a condition of settlement of a conferenced case under the Sixth Circuit’s First National Bank of Salem v. Hirsch, 535 F.2d 343 (1976). The conference attorneys, on the other hand, count these as program-induced settlements. For purposes of our analysis, a case was counted as a settlement if it did not proceed to a judicial decision by being submitted on the briefs or argued.


6. Findings

Impact of conferencing on appeals that would be submitted

     Among the expected or anticipated effects of the conferencing program, the impact on the court’s argument calendar is of primary interest. Given that cases in the study were randomly assigned, any difference between the proportion of treatment and control group appeals argued or submitted can be taken as the strongest evidence of program effect.

     The first part of the analysis attempted to determine whether the pre-argument conference program reduced the number of cases that were submitted to the court. We found important differences between the treatment and control groups in the number of cases settled, voluntarily dismissed, or dismissed for want of prosecution.

     Study results indicate that the pre-argument conference program has a substantial impact on the number of appeals diverted from the argument calendar. Additional details on this aspect of the study’s findings are presented in Table 1.

     With 68.9% of the control cases argued or submitted compared with only 57.1% of the treatment appeals, the observed difference between the two groups was 11.8%. As shown in Table 1, confidence intervals were calculated at both the 95% and 68% levels, although the magnitude of the difference in the upper and lower confidence bounds at the 95% level was relatively narrow. As discussed below, the impact of the program on the number of appeals reaching submission is even greater than the observed difference between the two groups.

     The extent of the actual effect of the pre-argument conference program can perhaps be best understood in terms of its impact on the workload of the court. As noted above, the program reduced the number of appeals reaching the argument calendar from approximately 69% of the filings to approximately 57%. Without the pre-argument conference program, approximately 69% of the appeals in the eligible group would reach argument or submission, with the remaining 31% being settled, dismissed, or withdrawn. The conference program results in the withdrawal or settlement of about 12% of the appeals that go through the program (and see footnote 8 regarding reported increases in settlement rates since the study period).

     A certain number of the appeals that settled would have done so in the absence of the conferencing program. In order to arrive at a more specific estimate of the impact of the program on the number of conference-eligible appeals, we employed a relatively straightforward calculation used earlier by Partridge and Lind in their reevaluation of the Second Circuit’s camp.[23] Using this calculation, our best estimate is that the pre-argument conference program reduces the total number of conference-eligible cases that would have been argued by about 17%. This, no doubt, represents significant savings of judicial, administrative, and litigant resources. This estimate takes into account those appeals that would have settled even in the absence of the pre-argument conference program.

     During the seventeen months when appeals were being randomized for study, the flow of appeals into the conferencing program was constrained significantly. Approximately one third of the conference-eligible appeals were assigned away from the program so that they might be studied as part of the control group.

     For calendar years 1985 and 1986, the period roughly covered by the study, data from the conference program indicate that 734 cases were processed by the program in 1985 and 668 cases in 1986.[24] This averages out to about 700 cases a year and comes close to the 750 cases projected by the conference attorneys to be processed by the program during a one-year period. Assuming that the three conference attorneys are able to process approximately 750 cases a year, the estimates of program effect from the study period indicate that about 90 appeals a year are diverted from the court’s argument calendar.

     The Judicial Conference Committee on Judicial Resources, when recommending the creation of new appellate judgeships, uses a standard of one active circuit court judge for every eighty-five appeals decided on the merits.[25] Using that standard, the conferencing program is doing the work of 1.06 judges. Each active appellate judge in the Sixth Circuit is authorized to have three law clerks and two secretaries,[26] and savings in judicial support staff should be factored into the equation when the court assesses the full implications of this finding.

Is the program’s impact on the submission rate more pronounced for certain types of appeals?

     We also examined the study appeals to determine whether the program was more or less effective in settling certain types of cases. For two reasons, some caution should be exercised in interpreting the results. First, for certain types of appeals the number of cases in our study was relatively small. Second, the observed differences were not statistically significant.

     Two types of treatment cases, those involving civil rights and federal question issues, showed a somewhat lower submission rate, suggesting that the conference program may be more effective in diverting those types of appeals. Four types showed an increase in arguments. Table 2 contains more details of these analyses. Again, caution must be exercised in interpreting these findings. We would not recommend that the court initiate changes in the program, such as the targeting of certain types of appeals, based on these findings alone.

     Similar analyses were undertaken of the cases that settled to determine whether any significant differences existed between the two groups that might further suggest a program effect. (See Table 3.) Certain types of appeals in the study groups were quite small. This was especially true for cases in the other civil and civil U.S. categories. The observed differences between the two groups in terms of the percentage that settled were highest for three of the five types of appeals: diversity, U.S. civil, and other civil. The results suggest that the conferencing program may be somewhat more effective in reaching settlements in the three case types noted in Table 3. Except for the diversity appeals, the observed differences between the two groups in the table are not statistically significant and may not represent actual program effects.

Impact on time to and stage of termination

     Having determined that the pre-argument conference program does divert a significant number of cases away from the argument calendar, we now turn to the question of the program’s impact on disposition time. Generally, the active period of a case is the time from the filing of the notice of appeal to either (1) the termination short of judicial disposition (e.g., dismissal) or (2) judicial disposition of the case following submission.

     It should be noted, however, that the conference attorneys do not begin their settlement or case management efforts until the pre-argument statement, which has been filed in the clerk’s office, arrives in the conference attorneys office. Under Local Rule 18, the pre-argument statement must be filed within fourteen days of the filing of the notice of appeal. The study did not collect systematic data on the average time from filing of the notice of appeal to receipt of the pre-argument statement by the conference attorney program. Our monitoring of the process makes us certain, however, that the randomization of the conference-eligible study cases did not delay the receipt of the pre-argument statements by the conference attorneys.

     The study focused on describing differences between the treatment and control groups in terms of the percentage of cases terminating during each stage of the case and the median time cases remained in a particular stage.

     We inferred that the goal of terminating cases earlier in the appellate process had been served if the treatment group had a significantly lower median days active than the control group for the periods from docketing to (1) disposition for all study cases; (2) disposition for cases submitted; (3) disposition for cases that were settled or dismissed; and (4) closing for all cases.[27]

     When considered as a group, cases assigned to the conferencing program took an average of twenty-five fewer days from filing to disposition by settlement, dismissal, or judicial decision than did cases that were not conferenced. Table 8 presents additional details. The twenty-five-day difference in overall disposition times between the two groups was statistically significant. In general, the data presented in the table is consistent with statistics for 1986 as contained in the circuit’s annual report for 1986[28] as well as with Administrative Office data for fiscal year 1987.[29]

     For appeals that were submitted, the cases in the treatment group took an average of about eleven days longer to reach disposition than did appeals that were not conferenced. Table 6 provides a more detailed picture of this statistically significant finding.

     For appeals that were settled or dismissed, those in the conference group took an average of five days fewer to settle than did appeals that were not conferenced. This finding is not statistically significant. Additional findings for both groups of settled or dismissed appeals are set out in Table 7.

     The submitted appeals in the conferenced group took an average of about twelve days longer to reach submission from docketing than did appeals in the control group. The additional results are contained in Table 5. This finding is statistically significant.

     The time from docketing to closing was found to be considerably shorter for conferenced appeals. Appeals in the treatment group took an average of 446 days from docketing to closing, compared with 484 days for cases in the control group. This represents a difference of 38 days. The date of closing is the date on which the clerk’s office determines that all activity regarding an appeal in the court has been completed. While focused largely on administrative considerations, i.e., removing the case from the active to the inactive category for reporting purposes, the elapsed time to closing does provide a rough measure for comparing the total amount of time that a case is active. Table 9 provides more details, and the findings here are statistically significant.

     Finally, data from the clerk’s office were analyzed to determine whether there were any differences between the treatment and control groups in terms of the number of times that briefing was held up or delayed. For cases that are subjected to conferencing, the attorneys in the program must often make important decisions about the briefing dates based on their assessments of the progress of discussions with the attorneys for the parties. This is especially true, for example, when appeals are assigned for conferences that would not be conducted until after the briefing date. When this happens, the conference attorneys will usually suspend the briefing date until after the scheduled conference.[30]

     Table 4 contains a breakout of the study data on this point. It should be emphasized that the numbers in the table reflect the number of times the briefs were held up, rather than counts of specific cases in which briefing was delayed.

Impact on issue simplification and clarification

     We incorporated into the study a number of indicators of issue simplification and clarification. We inferred that those goals had been served if the observed differences between the two case groups favored the treatment cases in terms of

(a)   the average number of pages in appellants’ briefs;

(b)   the average number of pages in the appellees’ briefs;

(c)   the number of motions disposed by the clerk’s office;[31]

(d)   the average number of motions filed that were brought to a single judge for disposition;

(e)   the average number of motions disposed by a three-judge panel;

(f)    the number of dun notices regarding procedural defects sent by the clerk’s office to the parties; and

(g)   the number of cases disposed of with written opinions.

Our findings on each of these measures are discussed below.

Average number of pages in appellants’ briefs

     The average number of pages in appellants briefs was lower in the group of treatment cases compared with the control group cases, but the difference was not statistically significant. Data were collected on the lengths of briefs, including reply and amicus briefs. A reduction in brief length is not seen as a primary objective of the program. It is, however, generally expected that conference attorney efforts should lessen the complexity of issues in those appeals that are presented to the court, thereby reducing brief length and increasing the number of briefs that are filed on time.

     As might be expected, there was considerable variability in the page length of briefs in the study cases, with a range of 17 to 289 pages.

     On a related measure, a slightly larger percentage (2.3%) of the control group appeals had briefs filed on time. (See Table 14.) This difference between the two groups did not reach statistical significance.

Average number of pages in appellees’ briefs

     There were no statistically significant differences between the two groups in the length of appellees’ briefs.

Number of motions disposed of by clerk’s office

     Overall, the program clearly reduces the number of procedural and substantive motions that are filed. A reduction in the number of motions filed translates into direct savings in judicial time and resources. In order to get a better sense of where the specific savings are, we analyzed the data on motions to determine whether there were fewer motions filed that would have required disposition by the clerk’s office, a single judge, or a panel of judges. There were differences, but none of the breakouts were determined to be statistically significant.

     There were differences between the groups in the average number of motions filed that were disposed of by the clerk’s office. The control group had more motions disposed of by the clerk’s office than did the treatment group. Again, the difference was not statistically significant. This measure is a very indirect indicator of the conferencing program’s effect in the sense that the conference attorneys efforts may result in the parties’ filing fewer motions that require the attention of the clerk’s office.

Average number of motions filed brought to a single judge

     There were no statistically significant differences between the groups in the average number of motions filed that were brought to a single judge for disposition.

     Of the treatment group appeals dismissed on motions, 50% were for lack of jurisdiction under Local Rule 8(a),[32] compared with 41% of the controls. This finding was statistically significant. This suggests that the conference attorneys may be especially adept at processing appeals with issues involving jurisdiction.

Average number of motions disposed of by three-judge panel

     We found that the treatment cases had an average of about 1% more motions that were brought to a three-judge panel for disposition. Again, the difference between the two groups was very small and did not reach statistical significance.

     Table 10 presents additional findings with respect to the number and manner of disposition of motions filed in the two study groups.

Number of dun notices

     When we examined the total number of dun notices sent to counsel in study cases,[33] the treatment group had a slightly larger percentage of notices for late briefs. We found only slight differences between the two groups in the percentage of notices for procedural defects involving return of briefs, filing a late appendix, or return of an appendix. None of the differences was statistically significant. Table 11 presents the results of this analysis in terms of the number of dun notices per case. It should be noted that the conference attorneys enter the picture at a point well after issues relating to the fee and appearance have been resolved. As such, they have little or nothing to do with any of the procedural defects involving fee or appearance aspects of the case.

Number of cases disposed of without written opinion

     One measure of the extent to which the pre-argument conference program clarifies or reduces issues in appeals is the rate at which conferenced appeals are disposed of without written opinion. Fewer treatment group cases than control group cases were disposed of with written opinions, suggesting that the conferencing program may reduce the complexity of issues that are ultimately presented to the court. The difference between the two groups on this measure was about 6% and was found to be statistically significant. This translates into a difference of about 10% in rate of opinion dispositions. More details are contained in Tables 12 and 13.

     As shown in Table 13, a significantly larger proportion of the treatment cases, nearly 59%, were disposed of by an order. In most instances, dispositions by orders present considerably less burden on the court than do those that terminate in a written opinion.

Differences in panel ratings and attorney assessments

     Through questionnaires, we attempted to determine whether the judges perceived any differences in the presentation of issues in the treatment and control group cases assigned for judicial decision.[34]

     Because questionnaires could not be administered for every treatment and control group case that reached submission, we can draw only very limited conclusions from the judge questionnaires, and the findings are not statistically significant.[35]

     Data from 579 completed panel questionnaires indicate that the judges found virtually no differences between cases in the two study groups in terms of the clarity, relevancy, and completeness of the presentations. We think that this result may be an artifact that was prompted by poor construction of the survey questionnaire. A review of the data from the completed questionnaires strongly suggests a response set bias brought on by an odd number of response options (five) for each item on the panel questionnaire. With numerical choices ranging from 1 to 5, the judges were asked to rate each of the presentations, with the number 1 equaling “far below average” and 5 representing “far above average.” The number 3, indicating an average rating, was selected in just about every response. In short we believe that the placement of the response items on the questionnaire may have overly influenced the responses.

     Questionnaires were sent to each attorney who represented a party in one of the treatment appeals.[36] The focus of the attorney questionnaires was on the conference program in general and not on the specifics of any given appeal. Attorneys in the control group cases were not surveyed.

     Table 15 presents some of the results from the attorney questionnaires that focused on their assessments of selected aspects of the pre-argument conference program. The majority of the respondents indicated that they believed that the program assisted counsel in complying with court procedures.

     A small percentage of the respondents, about 6%, expressed the view that the program did not assist counsel in complying with the procedures of the court.

     Over 67% of the attorneys felt that the program reduced or eliminated procedural motions. However, the majority of the attorneys did not feel that the program reduced or eliminated substantive motions.

     It is clear from the survey responses that a large number of attorneys believe the program to be helpful and effective in reducing procedural type motions. The questionnaire sought to determine whether the attorneys believed that conferencing produced other benefits for their clients. As shown in Table 16, the majority of the attorneys expressed the view that the program both clarifies and eliminates issues in the appeal. The responses to this question are consistent with the results of attorney surveys reported in the two studies of civil appeals management programs done earlier by the Federal Judicial Center.[37]

     As a follow-up to questions about the impact of the program on the issues in the appeals, the attorneys were asked about the net effect, if any, of the program on the amount of time spent on a case that reaches argument or submission. It is reasonable to expect that argued or submitted appeals entail a greater amount of attorney effort than cases that are dismissed or settled before submission. As Table 17 indicates, a little over half of respondents expressed the view that the program has a positive effect in that it results in a net savings in the amount of time expended on submitted appeals. The significance of this finding should be considered in light of the finding discussed earlier that treatment cases that were submitted took somewhat more time to reach submission and disposition than did similar cases in the control group. It would appear that while it took more time for treatment cases to reach submission or disposition, the attorneys did not spend more of their time working on the case. We did not attempt to determine whether any reported savings in attorney time were passed on to the attorneys’ clients; presumably, at least some of the savings were.

     Finally, it should be noted that nearly 80% of the attorneys responding to our questionnaire indicated that in the absence of the pre-argument conference program, they would not have taken the initiative to approach the opposing side about settlement. It is clear that the pre-argument conference program facilitates the process by which parties begin to consider and discuss the prospects of settlement.


7. Conclusion

     Like most studies, this evaluation of the Sixth Circuit’s pre-argument conference program was guided by a number of questions: (1) does conferencing result in cases being settled that would otherwise proceed to a judicial decision; (2) does the program make cases more manageable by simplifying and clarifying the issues that are ultimately presented to the court; (3) are any savings realized by the court as a result of the program; (4) does the practicing bar favor the program; and (5) is telephone conferencing, rather than in-person face-to-face meetings, a viable approach?

     Our analysis indicates that the answer to all of these questions is affirmative. The program in the Sixth Circuit clearly settles cases that would otherwise proceed to a full judicial disposition on the merits. We now know that 12% of the cases docketed by the court will be diverted from the argument calendar by the program. The immediate implications of that finding is that the program does the work of 1.06 appellate judges. As such, it is fair to characterize conferencing in the Sixth Circuit as essentially a settlement program. Moreover, the program settles cases at an earlier stage in the appellate process than would otherwise be the case. While we did not attempt any calculations to determine how much in dollars and cents is saved by getting appeals out of the process sooner, it is clear that significant savings in litigant and court resources may result. Limitations in our data did not permit us to identify any particular types of cases for which the program was more or less effective.

     Not only does the pre-argument conference program achieve its major objective of settling cases, it enjoys very strong support among the bar as well. The majority of the attorneys who responded to our survey indicated that, but for the efforts of the conference attorneys, they would not have initiated any discussions about settlement with opposing counsel. Concerns about not wanting to appear to be weak to opposing counsel were often cited by attorneys in our survey as the basis for not initiating discussions aimed at settling an appeal. Like similar appellate settlement programs in the Second, Eighth, Ninth, and District of Columbia Circuits, the Sixth’s initial efforts urging settlement come before any briefs have been written. It is not surprising, then, that half of the attorneys who responded to our survey felt that the program resulted in a net savings in the time they spend on an appeal. Of those who have handled appeals in the Sixth with and without the benefit of conferencing, most expressed a clear preference for the pre-argument program.


     [1]. Rule 33 of the Federal Rules of Appellate Procedure authorizes courts to conduct pre-argument conferences for purposes of simplifying issues and exploring settlement. The Sixth Circuit adopted Local Rule 18 establishing the pre-argument conference procedure in April 1981. The procedure was made applicable to all civil appeals from the district courts in which the notice of appeal is filed on or after August 1, 1981. The program was implemented with the hiring of the first conference attorney in November 1981 and became fully operational in January 1982. Rule 18 is contained in Appendix C of this report.

     [2]. Sixth Circuit Local Rule 18(c)(3) & (4).

     [3]. The internal study focused on a group of appeals filed in the first six months of 1983. Cases in this group were compared with a sample of appeals filed during the first six months of 1981. The 1981 appeals would have been subjected to the conferencing procedure had Local Rule 18 been in effect. (A summary of the results of the study is on file with the Federal Judicial Center’s Information Services Office.)

     [4]. Two earlier Center evaluations of the Civil Appeals Management Plan in the Court of Appeals for the Second Circuit had been conducted as controlled experiments. See J. Goldman, An Evaluation of the Civil Appeals Management Plan: An Experiment in Judicial Administration (Federal Judicial Center 1977), and A. Partridge & A. Lind, A Reevaluation of the Civil Appeals Management Plan (Federal Judicial Center 1983). See also J. Goldman, Ineffective Justice (1980); Goldman, The Civil Appeals Management Plan: An Experiment in Appellate Procedural Reform, 78 Colum. L. Rev. 1209 (1978).

     [5]. See Experimentation in the Law: Report of the Federal Judicial Center Advisory Committee on Experimentation in the Law (Federal Judicial Center 1981).

     [6]. For an overview of camp-type programs in the U.S. courts of appeals, see Note, “camp”ing Is on the Rise: A Survey of Judicially-Implemented Pre-Argument Programs in the United States Circuit Courts of Appeal, 1987 J. Dispute Resolution 89. As of the completion of this study, conferencing programs are in place in the following circuits: District of Columbia, Second, Eighth, Ninth, and Tenth.

     Conference programs have been implemented in a number of state court systems. See D. Steelman & J. Goldman, The Settlement Conference: Experimenting with Appellate Justice: Final Report (National Center for State Courts 1986).

     [7]. The term submission is used throughout this report to include both appeals argued orally and those submitted on the briefs. Cases that are submitted require specific attention of the judges on the panel to which a case is assigned. The pre-argument conference program aims to reduce the number of appeals that ultimately reach submission.

     [8]. According to annual reports submitted by the court, there has been a steady increase in settlement rates over the course of the program. Based on data compiled by the conference attorneys during the period of the study, reported rates increased by 5% per year. The trend has continued, with the court reporting a 49% settlement rate for 1988. If the 31% settlement rate found by the Center’s study in control group cases has held constant, the program now may be removing 18% of all conference-eligible cases from the court’s calendar. It should be noted the rates cited by the Conference Attorney’s office are raw calculations of the percentage of conferenced appeals settled or withdrawn before oral argument or submission on the briefs. As such, any observed trends may be largely the result of factors external to the program. (See 1988 Annual Report of the Pre-Argument Conference Program; 1989 Annual Report of the Sixth Circuit. Both reports are on file at the Federal Judicial Center’s Information Services Office.)

     [9]. See A. Partridge, The Budgetary Impact of Possible Changes in Diversity Jurisdiction 53–54 (Federal Judicial Center 1988), citing a benchmark of 255 case participations per appellate judgeship in appeals decided on the merits—or 85 filings per judgeship decided on the merits as the best objective estimator available.

     [10]. The court reports that in 1988, 30% of all the program’s conferences were scheduled at the request of one or more of the parties. See 1988 Annual Report of the Pre-Argument Conference Program (on file at the Federal Judicial Center’s Information Services Office).

     [11]. For a more detailed discussion of the genesis of the pre-argument conference program in the Sixth Circuit, see Rack, Pre-Argument Conferences in the Sixth Circuit Court of Appeals, 15 Tol. L. Rev. 921 (1984). Much of the substantive background information on the pre-argument program is based on the discussion contained in a background paper prepared by Sixth Circuit Executive James Higgins and Senior Conference Attorney Robert Rack, Jr., author of the article cited above.

     [12]. Judge Edwards had conducted pre-argument conferences in a small number of cases before the formal implementation of the program. During the six months following the implementation of the program, over half of the judges in the circuit were involved in conducting pre-argument conferences.

     [13]. Appropriations Act of October 18, 1986, P.L. 99-300. It is important to note that the appropriations provide authorization for separate funding of the pre-argument conference program positions for settlement functions only. The positions are not interchangeable with other legal positions within the court.

     [14]. The court’s pre-argument conference notice is set out in Appendix C.

     [15]. The conference attorneys indicate that it is especially difficult to avoid pro se litigants’ perception that attorneys speak for the court. Such litigants also tend to look to the conference attorney for legal advice.

     [16]. In all civil cases, Local Rule 18 also requires the clerk of district court, upon filing of a notice of appeal, to transmit to the clerk of the Sixth Circuit a number of documents, including (1) the full docket sheet of the court or agency from which appeal is taken; (2) a copy of the court or agency’s order, judgment, or decision that is being appealed; (3) any opinion or findings issued below; and (4) any report or recommendation made by the U.S. magistrate. The conference attorneys are provided copies of each of the above documents.

     [17]. Appendix C contains a copy of the pre-argument conference notice.

     [18]. The randomization procedure is described in Chapter 5.

     [19]. The discussion in this section and the one that follows is taken from a program description prepared by Senior Conference Attorney Rack and submitted to the Center at the beginning of the study.

     [20]. A pre-test of the random assignment procedures began earlier, on February 6, 1985. None of the cases from the pre-test phase were included in the final study sample. The randomization process was suspended for four weeks during June 1986 to accommodate the planned absence of the senior conference attorney.

     [21]. For the purpose of this study, cross and consolidated appeals were defined as those cases so designated on the court of appeals docket sheet. Related cases were defined as cases that were not cross or consolidated appeals but which had the same case number in the lower court.

     [22]. Conferences were conducted in a small number of the control group cases, generally in response to a request from one of the parties. There were fewer than a dozen such instances.

     [23]. A. Partridge & A. Lind, supra note 4, at 34–35.

     [24]. See B. Rack, Annual Report of the Conference Attorney Program for 1985, internal memorandum to the Judges of the Sixth Circuit (Jan. 13, 1986); Annual Report of the Conference Attorney Program for 1986 (Feb. 24, 1987). Totals for both years were verified from internal conference program logs maintained by Teresa Mack of the conference attorney’s office.

     [25]. See A. Partridge, supra note 9, at 53–54, citing a benchmark of 255 case participations per appellate judgeship in appeals decided on the merits—or 85 filings per judgeship decided on the merits as the best objective estimator available.

     [26]. United States Court of Appeals for the Sixth Circuit, Internal Operating Procedures 5 (1988).

     [27]. For purposes of our analysis, a case was considered as closed when the clerk’s office reported it as having been administratively and statistically removed from its docket of active cases. In other words, all activity in the case has been completed.

     [28]. United States Courts—Sixth Circuit, 1986 Annual Report Presented to the Sixth Circuit Judicial Conference 18, figure 5.

     [29]. Administrative Office of the U.S. Courts, 1987 Annual Report of the Director 152, Table B4.

     [30]. It should be noted that we did not count as a delay every instance in which the conference attorney extended the briefing date. For example, an extension by the conference attorney pursuant to Local Rule 18 was considered as on time, in contrast to a motion for extension filed by the parties.

     [31]. We examined all procedural and substantive motions filed in the treatment and control group cases during the period of the study. Rule 27 of the Federal Rules of Appellate Procedure and Sixth Circuit Local Rule 19 govern motions practice in the Sixth Circuit. Our analysis focused on identifying differences between the treatment and control group cases in terms of the number of motions that were disposed by the clerk’s office, a single judge, or a panel of judges. The assumption is that motions disposed by the clerk’s office normally require no judge time and therefore consume fewer judicial resources than do motions that require judicial attention. Most procedural motions, such those involving withdrawal or appointment of counsel, corrections of briefs or records, extensions of briefs, joint appendices, voluntary dismissal, etc., are ruled on by the clerk’s office. Other types of procedural motions such as requests to waive oral argument must be ruled on by a judge. All substantive motions such as those to dismiss the case on the merits or for an injunction or interlocutory appeal require judge action. Motions for a rehearing or reconsideration require a ruling by a three-judge panel.

     [32]. Sixth Circuit Local Rule 8(a) Motions and Motions Practice states in part: “(1) For Lack of Jurisdiction. At any time after a notice of appeal is filed a party may file a motion to dismiss on the ground that the appeal is not within the jurisdiction of the court. Motions to dismiss ordinarily may not be filed on grounds other than lack of jurisdiction.”

     [33]. Dun notices covering the following matters are sent by the clerk’s office when there are problems: filing fees; appearance; transcript; late brief; brief returned; appendix late; appendix returned; jurisdiction is premature; pre-argument statement; want of prosecution; brief correction; no-show at a pre-argument conference; status report late; errata sheets; and failure to file a pre-argument.

     [34]. The questionnaire, which focused on the briefs in each case, was administered to each panel judge about four weeks before argument. The judges were asked to return the questionnaires at the end of the two-week panel session.

     The questionnaire used a Likert-type scale on which judges were asked to rate the presentation of the issues in both the appellant’s and appellee’s briefs in terms of clarity, relevancy, and completeness. To rate clarity, the judges were asked to assess whether the issues, as presented, appeared to be correctly understood and accurately articulated.

     To rate relevancy, the judges were asked to assess whether the issues presented were appropriate to the case, with few, if any, extraneous issues presented. Judges were also asked to assess whether all of the relevant issues and arguments were presented, with few, if any, important issues being omitted. Questionnaires were administered in treatment and control group cases.

     [35]. Distribution and collection of the questionnaire presented some unanticipated burdens on the judges and staff of the court. At the request of the court, the questionnaire was discontinued before most of the panels with argued or submitted study cases could be surveyed.

     [36]. A questionnaire was administered to the attorneys for both the appellant and the appellee. A total of 2,560 questionnaires was sent to attorneys representing parties in the treatment group only. We attempted to send a question to each attorney of record in a case. We received 2,260 responses, giving us a response rate of 88%. Three of the questionnaires had to be eliminated because we could not decipher the responses.

     The questionnaire focused on, among other things, issue simplification and clarification effects of the program and not on the specifics of the conferences held in the appeals selected for study. The questionnaire was directed at obtaining an assessment of the pre-argument conference program in general. The questionnaire was mailed to the attorneys just after briefing, or, in situations where the case terminated before briefing, at the time of termination.

     [37]. See J. Goldman, supra note 4, at 82, reporting that 62% of attorneys representing treatment group cases that were argued or submitted noted that camp clarified issues in the appeal; A. Partridge & A. Lind, supra note 4, at 69, noting that about 25% of the attorneys responding believed that the conference resulted in improvements in the quality of brief or arguments.