Measuring the effects of specialization with circuit split resolutions.

• Author: Hansford, Eric

INTRODUCTION I. SPECIALIZATION AND THE FEDERAL COURTS OF APPEALS A. Previous Work on the Effects of Judicial Specialization B. Specialization in the Circuits II. MEASURING JUDICIAL PERFORMANCE WITH CIRCUIT SPLITS A. Judicial Performance and the Federal Courts of Appeals B. Measuring Judicial Performance with Circuit Split Resolutions, Not Reversal Rates III. DATA COLLECTION AND SUMMARY STATISTICS A. Data Collection and Adjustments B. Summary of Data IV. RESULTS A. Effects of Partial Specialization by Subject B. Effects of Partial Specialization, Aggregated. C. Quantifying the Effect of Partial Specialization CONCLUSION APPENDIX: DATA SOURCES AND COLLECTION A. Identifying Circuit Splits B. Constructing Yearly Cases per Judgeship INTRODUCTION

Suggestions for a specialized judiciary are a lightning rod in legal academia, apt to attract passionate defenses of the idealized "generalist judge." But the dichotomy of specialist versus generalist judges is a false one. In fact, all judges specialize to some extent thanks to variety in the subject matters of their cases. Within the regional federal courts of appeals ("circuits"), the Second Circuit is known for securities law, the Fifth for immigration law, and the D.C. for administrative law. By recognizing that judicial specialization by subject matter is not binary but instead lies along a continuum, this Note aims to test empirically the predicted benefits of specialization. In particular, this Note examines whether partial specialization improves judicial performance. A court partially specializes when it decides more cases than other courts in a particular subject matter and gains a relative advantage in deciding that kind of case. A court partially specializes relative to other courts: the Second Circuit partially specializes in securities law because it handles more securities cases than the other federal courts of appeals, but it has not fully specialized because it also hears nonsecurities cases. Partial specialization is about load relative to other courts; full specialization, on the other hand, is about narrowed subject matter jurisdiction of a specific court. So partial specialization, unlike full specialization, requires a comparison group. Economists have long concluded that full specialization leads to better performance. This conclusion should apply to partial specialization as well. But empirically testing these benefits for courts runs into an immediate problem: how does one assess when a court performs "better"?

This Note addresses the difficulty of measuring judicial performance by exploiting a quirk in the federal appellate system that pits courts of appeals directly against each other: the Supreme Court typically decides cases in order to resolve conflicts ("splits") among the circuits. Whether the Supreme Court agrees with a circuit's position in a given conflict provides one meaningful measure of the circuit's judicial performance. The regional circuits (1) are considered generalist courts, but partially specialize relative to each other. In this Note, I analyze which circuit "prevails" in the Supreme Court. I sort the cases by subject matter, and then track whether the Supreme Court agrees more often with a court that is partially specialized in that subject matter. A finding that the Court agrees with partially specialized courts more often would indicate that partial specialization improves judicial performance.

Part I gives the theory behind measuring partial specialization in the regional circuits and defines the variables. Part II introduces this Note's key innovation: measuring circuit performance with circuit split resolutions instead of reversal rates. Part III describes how I collected and adjusted the data and gives summary statistics, including a comparison of circuit split resolutions and reversal rates for the 2005 through 2008 Terms. Part IV presents the results, which are limited by the small data set. No single test produces a statistically significant result, but the results are consistent with partial specialization improving judicial performance. The Conclusion considers how to make circuit split resolutions more useful in future empirical research.

1. SPECIALIZATION AND THE FEDERAL COURTS OF APPEALS

   All judges specialize in one sense: their only job is judging. (2) Courts also often specialize geographically. (3) But the literature on specialized courts focuses on one particular kind of specialization: specialization in subject matter. Some quibble over when to call a court "specialized" as opposed to "generalist," (4) but I follow conventional use and define a court as "generalist" if it has jurisdiction to hear nearly all cases in nearly all subject matters (like the D.C. Circuit) and "fully specialized" if it hears cases in only a few (like the Federal Circuit). (5)

   Specialization should increase both judicial efficiency and expertise. (6) Increased judicial efficiency means that opinions can be written more quickly: an experienced securities judge can write a securities opinion more easily than could an inexperienced judge because the experienced judge already knows the relevant statutes, regulations, and case law. Increased judicial expertise leads to better opinions: a judge with a deep understanding of securities law has an advantage over a judge who knows little about the subject. Efficiency goes to the quantity of opinions, while expertise goes to the quality of opinions.

   This Note focuses exclusively on whether specialization improves the quality of opinions--a predicted result of expertise. It ignores the efficiency benefits of specialization, which may allow judges to decide more cases or shift their time to other tasks. Because it ignores efficiency, this analysis likely understates the benefits of specialization.

   1. Previous Work on the Effects of Judicial Specialization

      The literature on judicial specialization usually discusses specialization in the abstract. (7) Most commentators agree that specialization will improve judicial performance. (8) But they debate whether the improvements in performance will outweigh the disadvantages that may come with full specialization. The disadvantages range from judicial boredom (9) to judges overstepping their bounds and trying to legislate from the bench. (10)

      Determining the actual effects of specialization requires empirical evidence. (11) Some studies have tried to find the evidence by focusing on a single fully specialized court (often the Federal Circuit), (12) or a group of fully specialized courts. (13) But for most of the fully specialized courts, "a good point of reference for comparison does not exist"; (14) typically no other court decides cases involving the subject matter. (15) Even aside from the dearth of comparison groups, measuring judicial performance is tricky, and no study convincingly resolves whether specialization improves judicial performance. As Lawrence Baum explains: "[S]pecialization of judges might have significant effects on their work. Debates over specific proposals for specialized courts and the general movement toward greater specialization in the courts reflect a belief that such effects exist. The empirical evidence on the impact of specialization, however, is limited." (16)

      This Note looks to the federal courts of appeals to provide some of this missing evidence. By examining partially specialized courts rather than fully specialized ones, this Note can control for many of the factors that might otherwise hinder identification of a relationship. More importantly, partially specialized courts have a clear comparison group, which permits the identification of differences in judicial performance.

   2. Specialization in the Circuits

      The United States has thirteen courts of appeals: the eleven numbered regional circuits (the First Circuit through the Eleventh Circuit), the D.C. Circuit, and the Federal Circuit. The numbered circuits and the D.C. Circuit have jurisdiction over appeals from almost all of the decisions of the federal district courts they contain, (17) so they are generalist courts. (18) The Federal Circuit, which hears cases involving only certain subject matters, is the only specialist federal court of appeals. (19) The Federal Circuit rarely opposes other circuits in circuit splits before the Supreme Court, so this study excludes it.

      While the regional courts of appeals are generalists, the subject matters of their cases vary based on the kinds of cases that arise in their districts. The D.C. Circuit presents the most extreme example, its caseload crowded with administrative law cases. (20) Similarly the Second Circuit (containing New York) has more than one-third of the federal appeals on securities law, (21) and the Fifth Circuit (containing Texas) has about half of the federal immigration law appeals, (22) although neither has a docket dominated by these cases. These levels of specialization are the most extreme. (23) But in any given subject, some circuits partially specialize relative to the other circuits.

      Although at least one commentator has declared that studies of judicial specialization should focus on the judge, not the court, (24) here the most logical unit of analysis is the court. Courts of appeals make decisions in panels of (4th ed. 2003) ("State judiciaries have general jurisdiction and may therefore hear all causes of action unless there is a statute denying them subject matter jurisdiction. But federal courts have limited subject matter jurisdiction; that is, they are restricted in what cases they may adjudicate and may exercise jurisdiction only if it is specifically authorized."). But colloquially, most consider the federal courts of appeals (other than the Federal Circuit) generalists, given that they have authority over diverse subject matters. See, e.g., Ruth Bader Ginsburg, An Overview of Court Review for Constitutionality in the United States, 57 LA. L. REV. 1019, 1021 (1997) ("With...