PANEL COMPOSITION AT THE COURT OF APPEAL FOR ONTARIO.

• Author: Drucker, Jeremy

I INTRODUCTION 12 II BACKGROUND OF THE COURT 13 I. The Court of Appeal for Ontario 13 III RELEVANT LITERATURE 15 I. Yahya and Stribopoulos: an Empirical Study of the Court 15 II. A Check for Randomness: Regressions 16 III. A Check for Randomness: Panel Composition and Simulations 18 IV ANALYSIS OF THE COURT 18 I. Scope of Analysis 18 II. Statistical Analysis: Panel Presence 20 III. Statistical Analysis: Panel Presence and Regressions 22 IV. Statistical Analysis: Panel Composition 24 V DISCUSSION AND CONCLUSION 28 I. Discussion of Findings 28 II. Study Limitations 29 III. Conclusion 30 I INTRODUCTION

Canada is a nation governed by the rule of law; this is a "fundamental postulate of our constitutional structure." (1) An independent and impartial judiciary is critical to maintaining the rule of law. Indeed, in criminal matters the right to "a fair and public hearing by an independent and impartial tribunal" is guaranteed in Section 11(d) of the Canadian Charter of Rights and Freedoms. (2) Canadians therefore have a right to receive a fair hearing. Further, as the Supreme Court of Canada has held, "the courts should be held to the highest standards of impartiality... Fairness and impartiality must be both subjectively present and objectively demonstrated to the informed and reasonable observer." (3)

The personal attributes of the judge--whether age, sex, political views or otherwise--should therefore have no bearing on the outcome of a given case. And yet, judges are human and carry biases and beliefs that may influence their decisions. Indeed, empirical studies have shown that judges' political affiliations may affect their judicial decisions. (4) When courts hear cases in panels, as opposed to en banc, the possibility exists that the composition of the panel will influence the outcome of the case. The process of appointing judges to panels therefore becomes critical. Are chief justices selectively assigning judges to panels to craft the law in their preferred direction? Or, are panels selected randomly, ensuring that human bias does not affect panel composition? Should panel selection necessarily be random, to avoid any potential for manipulation of panels?

From a normative perspective, random panel composition is not without downsides. Ensuring a random process can be virtuous, as it helps to avoid panels being gamed to sway case outcomes. Yet, it may also fail to take advantage of the beneficial aspects of a non-random process. For example, a judge with criminal law expertise may be uniquely qualified to hear a complex criminal matter, while a judge with corporate law expertise may bring a valued perspective to a corporate matter. Sitting junior judges alongside senior judges may enhance their training. Ensuring diversity of age, gender and culture among judges on a panel may also have advantages. A strictly random process would fail to take advantage of these possible benefits.

As the highest court for the province of Ontario, the Court of Appeal for Ontario (the "Court") is one of the most important appellate courts in the nation. Given the large number of judges on the Court and the small panel sizes used, there is the chance that case outcomes are affected by the composition of the panels. That is, the identity and personal attributes of the specific judges on a panel could be impacting the outcome of a given case. However, research to date has not examined panel composition at the Court in a detailed manner.

This paper will examine panel composition at the Court, seeking to determine whether panel composition is random or non-random. The sample analyzed in this study comprises all judgments of the Court in the 2013 calendar year. Overall, this study has led me to conclude that: (i) panel composition was consistent with randomness for full judgments of the Court, but (ii) panel composition was not consistent with randomness for endorsements of the Court. However, caution must be taken so as not to over-emphasize these findings. As will be discussed, they are subject to significant qualifications and limitations.

In Part II I will provide a brief overview of the Court and its panel selection process. In Part III I will discuss literature pertinent to the Court specifically and panel selection more generally. In Part IV I will review my statistical analysis of the Court and in Part V I will conclude with a discussion of my findings and potential implications.

II BACKGROUND OF THE COURT

1. THE COURT OF APPEAL FOR ONTARIO

   The Court serves as the highest court for the province of Ontario, and hears appeals primarily from decisions of the Ontario Court of Justice and the Superior Court of Justice; the Court also hears a small number of appeals from the Ontario Review Board. The Court receives and disposes of a significant number of appeals. Over the five-year period from 2009 to 2013, the Court disposed of an average of 1,615 appeals per year. (5) Yet, fewer than 3% of the Court's decisions are appealed to the Supreme Court of Canada. (6) For the vast majority of cases, therefore, the Court is the final court of appeal.

   As of March 1, 2015, there are 28 justices of the Court. The majority of cases are heard in panels of three, though occasionally panels of five are used or an appeal is disposed of by a single judge. Given the large number of judges and small panel sizes, the potential for panel effects looms large. That is, given the possibility that one panel of three judges may not always reach the same decision as another panel of three judges, the composition of the panels can impact case outcomes. Indeed, as Ian Greene and others put it, there is a "luck of the draw dimension" to the process of appellate decision making. (7)

   The Ontario Courts of Justice Act (the "Act") is largely silent on panel selection and appears to leave the process entirely within the discretion of the Chief Justice. The Act states that "the Chief Justice of Ontario has general supervision and direction over the sittings of the Court of Appeal and the assignment of the judicial duties of the court." (8) The Act further states that panels must be composed of an uneven number of judges. (9) Another empirical study of the Court, discussed later in this paper, notes that the Registrar is responsible for determining panel composition. (10)

   In practice, the Court strives to assign judges to cases using a random process, while making efforts to have at least one senior judge and one subject-area specialist per panel. (11) For example, a criminal panel would have at least one senior judge and at least one judge with expertise in criminal law. While, as a generalist court, the judges hear appeals across different areas of law, in practice, judges with expertise in a particular area may hear more appeals in that area.

   Every week, the Court holds at least one criminal panel and one civil panel. Before judges are assigned to panels, cases are scheduled by a coordinator. The coordinator is unaware of the legal substance of the cases. Once the cases have been scheduled, the Senior Legal Officer, taking into account scheduling conflicts, manually assigns judges to panels. The number of panels depends on the number of cases scheduled to be heard. Often there are three or four panels in any given week. Judges will typically hear cases for two weeks, followed by two weeks of crafting judgments. While the Chief Justice and Associate Chief Justice may review the draft schedules, they very rarely intervene.

   Overall, this process attempts to strike a balance between taking advantage of judicial resources--specifically, practice area expertise and seniority--while attempting to maintain panel composition consistent with a random process. Objectively, however, a process designed to take advantage of seniority and expertise is not random.

   III RELEVANT LITERATURE

1. YAHYA AND STRIBOPOULOS: AN EMPIRICAL STUDY OF THE COURT

   Yahya and Stribopoulos' 2007 article analyzed all decisions of the Court and all votes from 1990 to 2003, tracking the type of litigant, political party that appointed the judge (hereinafter referred to as a judge's political party) (12) and gender of the judge. While the focus of this study was on case outcomes, as a rare empirical study of the Court it nonetheless holds valuable lessons for students of the Court.

   Yahya and Stribopoulos note that across their sample of over 4,000 cases, 95% of the Court's decisions were unanimous. (13) Further, there was little variation in this rate across case categories. Yahya and Stribopoulos did not, however, include endorsements; these shorter decisions represent a significant amount of the Court's work but due to their brevity and, in the pre-internet age, general lack of publication, they are of little precedential value.

   While the greater part of Yahya and Stribopoulos' study was on case outcomes, they did conduct a brief analysis of panel composition. (14) For example, they found that 59% of panels were of mixed gender, 40.5% were all male and just 0.5% were all female. (15) Further, they noted that 73% of panels were mixed by political party, while 18% were composed of all Conservative judges and 9% were composed of all Liberal judges. (16) They also described the percentages of votes cast by female judges, male judges, Conservative judges and Liberal judges. However, none of these figures were analyzed for statistical significance, and the discussion of panel composition was brief.

   The bulk of the paper was devoted to an analysis and discussion of case outcomes as a function of the area of law, the disposition of the trial court, the nature of the complainant, and the gender and political party of the judges. while noting a "remarkable amount of cohesiveness" (17) on the Court, the authors found some influence of political party of appointment on judicial decisions. For example, in certain contexts, Conservative judges preferred more conservative outcomes in Charter litigation...