Leaving the bench, 1970-2009: the choices federal judges make, what influences those choices, and their consequences.

• Author: Burbank, Stephen B.
• Position: IV. Retirement through VI. Policy Implications, with footnotes, p. 56-102

4. RETIREMENT

   A. Introduction

   In an effort to present a comprehensive picture of judicial retirement behavior over the years covered by our study, we pursued two parallel efforts. First, in order better to understand the recent history of retirement, we essentially replicated for retirement in the years between 1970 and 2009 what Van Tassel did for resignation (and what we supplemented in Part II), using the same types of public sources. Second, informed by that research, we thought that our understanding of retirement would benefit from a multi-method approach that did not rely exclusively on historical research and secondary sources. Accordingly, we devised certain statistical analyses of the relevant data in the FJC's database and initiated a questionnaire survey of all judges now living that we could identify as having retired from the federal judiciary.

   B. The Demographic Data: Retirements 1970-2009 (233)

   From 1970 through 2009, 2143 judges served on the lower federal courts. Overall, 101 (4.7%) of these judges retired during this period. Of the 1673 district judges eligible, 79 (4.7%) retired; of the 47[degrees] circuit judges eligible, 22 (4.7%) retired.

   1. Frequency of Retirement

      Table 10 presents retirement frequency data for all judges by year within decades. Tables 11 and 12 present the data for district and circuit judges.

      Table 10 reflects a substantial increase in the number of judges eligible to retire beginning in 1984. This increase was due to the statutory amendment effective in July of that year that changed eligibility for judicial retirement from age seventy (with ten years of service) to any combination of age and service totaling eighty starting at age sixty-five. Actual retirements increased as well. This general upward trend appears particularly dramatic when presented visually, as in Figure 4.

      [FIGURE 4 OMITTED]

      As this graph suggests, although retirements peaked in 2000, the longterm trend continues toward increasing numbers of retirements, even if the pace of acceleration has slowed in recent years. Explaining this apparent increase in judicial retirements based on the historical data is more difficult. The shift to the Rule of 80 in 1984 seems to have had an immediate-though slight--effect on retirements, but it alone cannot explain any longterm increase in the retirement rate. The question why pension-eligible judges chose to retire rather than assuming, or after assuming, senior status (or remaining in regular active service) cannot be answered by citing the rules for pension eligibility.

      How the substantial increase in judicial salary in 1991 may have affected the retirement rate is open to several interpretations. For example, judges who previously were considering resigning before becoming pensioneligible had an increased incentive not to leave until they were eligible to retire with an improved retirement benefit. Perhaps resignation and retirement rates are inversely proportional, and substantial salary increases have the effect of delaying departures and encouraging retirements rather than resignations.

      Based on Tables 10 through 12, it is possible to calculate a precise rate of retirement and thus to test whether the apparently dramatic visual trend in Figure 4 accurately portrays judicial behavior over time. As reflected in Tables 13 through 15, we find that, although the rate of retirement increased over the study period, the increase was modest.

   2. Retirements Delayed

      Table 16 presents data on the number of years (or days if less than 365) that all judges, district judges, and circuit judges continued to serve (either in regular active service or in senior status) after becoming eligible to retire before actually retiring.

      These data indicate that judges do not retire at the first opportunity. This observation is borne out by the fact that, as we develop below, most judges who become eligible to retire choose to take senior status first.

   3. Age at Retirement

      Looking next at the 101 judges who retired during the years under study, Table 17 presents retirement age data separately for all judges, district judges, and circuit judges. Figure 5 is a histogram of age at retirement.

      [FIGURE 5 OMITTED]

      C. What the Historical Record Teaches

      In this part, using the methodology followed in the Van Tassel study, we review and comment on what the historical record tells us about judicial retirements over the years we studied. The preceding part presented and analyzed data on retirement from the database maintained by the Federal Judicial Center. By contrast, the descriptive material in this part is based on public documents, newspapers, and the like. (236)

      Analyzing the causal influences on retirements is difficult because judges have rarely offered recorded explanations, and the judiciary has had no exit interview mechanism. As one respondent to our questionnaire observed, "I have been waiting for 12 years for someone in the federal court system to ask me why I left." (237) The handful of occasions when judges have spoken out may or may not be illustrative of larger trends. For instance, several judges noted dissatisfaction with the social isolation of serving as a federal judge, as well as with a perceived diminution of the importance and authority of judges.

      Figure 6 illustrates the apparent reasons for judges' retirements. As the figure suggests, most of the judicial retirements in the last forty years can be loosely grouped into two broad categories. The first consists of a minority of judges who retired at an advanced age after many years of service. This category includes not simply the small number of judges who explicitly cited advanced age or health in explaining their retirement, but also many of the judges in the "unknown" category, who retired at the median age of eighty after a median ten years of pension eligibility. (238) This group of judges often retired in their eighties or nineties after as many as twenty years in senior status. Few of these judges took other employment after retirement, and many of them died shortly after leaving the bench.

      The second and larger category consists of judges who entered private practice or other private sector employment after retirement (69%, when the categories of inadequate salary, other employment, or dissatisfaction are included). By and large, these judges shared several characteristics. First, they retired relatively young: the median age for judges who entered private practice was 67.5 years. Second, they retired soon after becoming pensioneligible: the median judge in this group waited only one year after eligibility before retiring. Third, although about two-thirds of these judges first assumed senior status (47 of 70), they remained in senior status very briefly. Beyond this, however, it is difficult to generalize about the judges' motivations. Most provided little information about their reasons for retirement, although a handful mentioned inadequate salaries, or the isolation of federal judges, in explaining their motivations. (239)

      The other reasons for retirement were usually idiosyncratic. Three judges retired after relatives were appointed to the same court--retirements needed to avoid the strictures of an anti-nepotism statute. (240) One judge stated that his retirement was motivated by political attacks on him during the 1996 presidential campaign. (241) Two judges retired in protest over the

      The other reasons for retirement were usually idiosyncratic. Three judges retired after relatives were appointed to the same court--retirements needed to avoid the strictures of an anti-nepotism statute. (240) One judge stated that his retirement was motivated by political attacks on him during the 1996 presidential campaign. (241)t Two judges retired in protest over the implementation of the Sentencing Guidelines, (242) while another retired to forestall further investigation of his alleged illegal disclosure of a federal wiretap. (243)

      The large majority of judges (73 of 101) assumed senior status prior to retirement. For many of these retiring judges, however, service in senior status was brief: a matter of a year or even a few months prior to retirement, usually followed by work in the private sector. The large number of judges who adopt senior status only briefly before entering the private sector suggests that judges are adopting senior status as a conscious transition to retirement.

      From 1970 through 2009, retirements occurred more frequently from some courts than others. Among the district courts, New Jersey, the Central District of California, and the Northern District of Illinois had the highest raw numbers of retirements. Among the circuit courts, there was less variation, but the Fifth Circuit had the highest number of retiring judges. Figures 7 and 8 compare the number of retirements against the numbers of authorized judgeships (as of 2010) in the district and circuit courts, respectively.

      Although the retirement rates from some courts seem disproportionately large, some of the absences are just as notable. The Eastern District of New York, for instance, had no retirements from 1970 through 2009, even though it has 15 authorized judgeships (as compared with 17 for New Jersey). These disparities raise the question whether the policies or culture of individual courts concerning judges in senior status have an effect on retirement rates, a question that requires additional research.

      One anomaly seems to have at least a partial explanation. The District Court for the District of Columbia and the Court of Appeals for the District of Columbia have comparatively high retirement rates. There may be greater opportunities for these prominent judges to receive appointments to other public offices. Indeed, the three judges appointed to other public office after retirement during the study period all served on one of these courts. (246)

      D. The Questionnaire Survey

   1. Methodology

      We distributed a questionnaire to...