Article Critique

I. INTRODUCTION

In the wake of the defeat of three Iowa Supreme Court justices in judicial-retention elections in 2010, the President of the New York City Bar Association, Samuel W. Seymour, wrote a letter to the New York Times stating that "When a judge suffers an electoral defeat because he or she exercised judicial independence, we all suffer." n1 Seymour was, of course, mourning the defeat of three judges who had ruled in 2009 that the state constitution of Iowa protected same-sex marriage. n2 The judges' defeats, though ironic in one way, were a very visible illustration of a fundamental dispute between contemporary political scientists and much of the legal community. The defeat of the Iowa judges was ironic in that legal reformers have advocated the selection of judges by a merit-selection system. n3 Merit selection involves a blue-ribbon commission that screens possible judicial appointees; a short list of possible appointees is provided the governor; the governor appoints a judge from the short list; and then the judge runs at a later date in a retention election. n4 In retention elections, there is no opponent, and voters simply vote "yes" or "no" on whether the judge should be retained in office. This merit-selection system is supposed to promote judicial independence from the electorate because judges are rarely defeated in retention elections. But clearly, the Iowa judicial defeats show that retention elections retain an [*222] element of judicial accountability where voters have at least some opportunity to express their electoral preferences regarding judicial decisionmakers. Regardless of one's views about the defeated Iowa judges or the decision that led to their rejection by the voters, Seymour's rejection of the legitimacy of accountability as a goal for a judicial-selection system and his embrace of independence as the only acceptable goal reflects the core dispute between contemporary political scientists and much of the legal community over judicial selection.

Contemporary political scientists (and some law professors) tend to view judicial elections very differently from the mainstream legal community. n5 The result is that many political scientists support judicial elections, especially partisan judicial elections, as a way of achieving electoral accountability by judges. n6 The mainstream legal community, on the other hand, tends to favor systems of judicial selection that support judicial independence from the electorate. n7 Charles Geyh has written:

If, as a significant segment of the political science community believes, independent judges are essentially unconstrained policymakers who decide cases by acting on their personal preferences or attitudes, then ... elections will produce public policies that better represent the citizenry by creating incentives for judges to pay attention to citizen preferences when deciding highly visible and publicly salient issues. If, on the other hand, as the mainstream legal community believes, independent judges do their best to follow the law ... then ... elections create incentives for judges to set the law to one side and pay attention to citizen preferences in deciding cases. n8

Ultimately it is this dispute over how judges reach decisions that is at the core of the debate over the appropriate methods of judicial selection, a topic that has occupied lawyers, judges, and political scientists for decades.

There has been hot debate between the reformers in this group and those doubters who question the reformers' [*223] proposals. n9 In general, lawyers and judges have relied on legal arguments, personal experiences, anecdotes, or opinions rather than on empirical analysis of the arguments for or against particular systems of judicial selection. n10 Much of legal writing reflects a normative or impressionistic approach to judicial selection. Political scientists (and some quantitatively oriented law professors) contribute to that debate by analyzing data that tests the validity of assertions about the various systems of selection. n11

Prior to exploring the political science findings that are of direct relevance to the debate over judicial selection, it is essential to provide an overview of judicial-selection systems to provide a background to this controversy. As one study pointed out, there are many ways to select judges, so many ways that hardly any states have identical selection systems. n12 Most states have hybrid systems where judges will be chosen one way at one level in the court system and another way at another level. n13 [*224] Additionally, some states have different methods for filling vacancies that might occur as a result of the creation of new courts or because of the deaths, resignations, or retirements of judges. n14 Even Texas, which is a partisan-election state to an extreme degree, n15 does not select all of its judges in that manner - vacancies on the bench are filled by appointment, and municipal judges are frequently appointed. n16

However, a general classification scheme for selecting judges would identify at least four major systems of selection: (1) appointment; (2) commission selection or merit selection of judges; (3) nonpartisan election of judges; and (4) partisan election of judges. n17 In recent years, much of the debate over judicial selection at the state level has ignored appointment and, to a considerable degree, nonpartisan election of judges and has focused on merit selection and partisan election of judges, typically with judicial reformers criticizing partisan election of judges and offering merit selection as an alternative system. n18 In this debate between the two systems of selection, it is important to recognize that the two systems reflect two very different theoretical perspectives on the appropriate role of judges.

One can arrange the four systems of selection along a continuum where on the one side is judicial independence and on the other side is judicial accountability. n19 An appointment [*225] system, especially one such as the federal judicial appointment system, stresses judicial independence. That is, there is no substantial check on the judge after the confirmation process. Any accountability to the public in such an appointive system occurs only indirectly through the electoral responsiveness of the judicial appointment and confirming officers. Next to an appointive system in stressing judicial independence is a merit-selection system, the most common of which is the Missouri Plan. Under this plan, a commission - composed of lawyers, judges, and lay people - prepares a list of possible judicial nominees for the governor. The governor's appointee must come from the list. Then, after a period of time (frequently one year) the judge runs for office in a retention election. In that election, the judge has no opponent and the voters are asked to vote "yes" or "no" on whether the judge should be retained in office. If the judge is retained, the judge will serve a specified term before facing another retention election. Although the governor is restricted by the commission's recommendation as to whom he or she can appoint, with the exception of the retention election, there is no real check on the judge other than the indirect check that might occur given the electoral responsiveness of the commission or the governor. The retention election might be seen as providing public accountability for the judge, but because they lack an opponent, the defeat rates for judges in retention elections are exceedingly rare. n20 The highly visible defeats of three Iowa Supreme Court justices n21 are a huge exception to the general pattern of incumbent success in retention elections. One study of retention elections from 1990-2004 found that only 1.3% or 3 out of 231 incumbents were defeated in retention elections. n22 This finding is consistent with an earlier study that found that in 1972-78, only 1.6% of judges in retention elections were defeated. n23 [*226] Probably the most comprehensive survey of retention elections looked at both trial and appellate retention elections in ten states from 1964-2006 and found that less than 1% of judges running in retention elections were defeated (56 out of 6306 judges). n24 And, half of the judges who were defeated were in Illinois, a state which requires a super-majority, a 60% favorable vote, for a judge to be retained. n25 These data suggest that any electoral accountability provided by a retention election is minimal at best.

Nonpartisan elections are judicial elections where judges run on the ballot without party labels. Some states, most notably Michigan and Ohio, have such highly partisan judicial elections, even in the absence of party labels on the ballot, that some classify them as partisan election states. n26 Other nonpartisan states are truly nonpartisan with minimal party involvement. n27 Because judges in nonpartisan election states are subject to electoral control, this is a judicial-selection system that leans toward promoting judicial accountability as a goal rather than judicial independence.

Some states have partisan elections that are generally defined (at least with the exception of Michigan and Ohio) as states that elect their judges where the party affiliations of the judges are on the ballot. This system of selection promotes the greatest degree of electoral accountability of judges.

Kermit Hall found that judicial elections were initially seen as a way of promoting judicial independence. n28 Hall claimed that judicial elections were meant to remove judges from the influence of governors and legislators by providing judges with [*227] an independent power base. n29 In an exhaustive study of the rise of judicial elections, Jed Shugerman found that the nineteenth century supporters of judicial elections wanted judges to be separate from legislatures and governors, whom they wanted judges to check. n30 Shugerman concluded that these early supporters also wanted to embolden judges and to legitimize judicial review by connecting judges to the people through popular elections as well as allowing the electorate to choose judges who would defend the rights of the people. n31 Today, most political scientists would argue that whatever the origins of judicial elections, elections by their very nature promote electoral accountability. n32 Charles Geyh well described what has happened with judicial elections over time, stating, "Judicial elections were originally introduced primarily to promote judicial independence by liberating judges from the control of governors and legislators, but they have since morphed into tools that serve primarily to promote judicial accountability." n33

Modern judicial elections are seen by many in the legal community as having nothing but negative effects. Shugerman, for example, describes contemporary judicial elections in this way:

Today, judicial elections weaken state courts and reduce their willingness to defend the rule of law against public opposition or special interests. Recent studies demonstrate that elected judges face more political pressure and reach legal results more in keeping with local public opinion than appointed judges do. Other studies have found that elected judges disproportionately rule in favor of their campaign contributors. It has been a long-established practice for parties and lawyers to donate to the judges who will later hear their cases, but recently the size of such donations has increased dramatically. Spending on judicial campaigns has doubled in the past decade, exceeding $ 200 million in total direct donations from 1999 to 2008. In June 2009, the [*228] U.S. Supreme Court ruled for the first time that an elected judge must recuse himself from a case involving a major campaign contributor. In that case, a coal company CEO who was appealing a $ 50 million verdict spent $ 3 million on the campaign of a challenger for a seat on the West Virginia Supreme Court, financing political attack ads alleging that the incumbent was soft on child molesters. n34

Such views about judicial elections have led to a reform movement primarily oriented toward moving states away from judicial elections, especially partisan judicial elections, and toward merit selection. n35

II. THE TWO REFORMER ARGUMENTS

With this framework in mind, judicial reformers focus on two arguments: (1) elected judges are not as independent as they should be; n36 and (2) even if judges should be accountable, judicial elections are inadequate in promoting the accountability of judges. n37

With one major exception, political scientists have little to offer those who make the normative argument that judges should be independent. For example, when the leading judicial reformer, Justice Sandra Day O'Connor, argues, "I do not think that I can make a stronger argument than history has already made on behalf of an independent judiciary," n38 she is making a normative argument in favor of judicial independence. Given that an appointment system for state judges or a merit-selection system would eliminate or vastly reduce accountability to the voters, judicial elections would not be the preferred system of selection.

[*229] However, in one area, the normative argument of judicial reformers provokes dispute by political scientists and many lawyers. That argument is that the judiciary should be removed from "politics" because judges are not "politicians in robes," they are neutral decisionmakers, and judicial elections harm that role. After all, wrote one judicial reformer, "The judiciary is different, providing a unique institutional role with unique institutional norms in unique institutional settings." n39 Justice O'Connor claims, "No amount of reform will remove the politics inherent in partisan judicial elections because they specifically aim to infuse politics into the law." n40

Political scientists (and many lawyers) see the judge's role as inherently political. As one legal scholar wrote,

The Legal Realists taught us nearly 100 years ago, language is often ambiguous. The Legal Realists famously demonstrated this in the early twentieth century by showing that precedents, statutes, and constitutional provisions can often be read in more than one way. Consequently, judges often cannot render decisions based solely on legal texts; they must incorporate - whether consciously or subconsciously - other considerations in order to resolve ambiguities. This is especially true of state court judges. Not only do state court judges have the power to shape the vague commands of statutes and constitutions, but, much more so than their federal counterparts, they also have the power to make common law. n41

Political scientists overwhelmingly share that legal realist view, and they support that view with a substantial body of research done in the years since World War II that shows judges make policy choices. n42 True, the bulk of this research has been on the United States Supreme Court, but there has been a growing body of research on state supreme courts that shows [*230] that state judges, at least state high court judges, have flexibility in their decisionmaking and make policy choices. n43 As Bonneau and Hall have pointed out,

These empirical studies confirm what practicing attorneys and appointing presidents already know intuitively. Skilled attorneys understand quite well that which judge sits will affect the likelihood of winning ... . Although it is the modern equivalent of declaring that the emperor has no clothes to say so, politically astute observers fully recognize that the basic political preferences of judges influence their votes. n44

One may reasonably choose judicial independence as a value over judicial accountability. But political scientists would overwhelmingly reject the argument that judicial-selection systems should emphasize judicial independence if the justification for this choice is that judges are neutral decisionmakers who do not make policy choices and therefore should be removed from the political fray.

Nevertheless, there is research that at least suggests that the political science/legal realist perspective on judicial decisionmaking is overplayed. While no research denies that ideology explains some judicial decisionmaking, two major criticisms have been made about political science claims that "the basic political preferences of judges influence their votes." n45 One criticism is that a majority of the research on the influence of judges' political preferences on judicial decisions is based solely on United States Supreme Court decisions. n46 The United States Supreme Court hears relatively few cases. It is also a high-discretion Court in the sense that it is less constrained by precedent than other courts and is likely to deal with many cases where there are conflicts in precedent and interpretations of the law and where there is great room for judge-made law. n47 Essentially, the argument is that by assuming [*231] research on the United States Supreme Court applies to other courts, political scientists are weighting their results in favor of their assumption that all courts make policy like the United States Supreme Court. n48 The other major criticism is that political scientists ask the wrong question in their research. n49 The question, it is argued, should not be whether judges' political preferences affect their decisions; it should be how much do judges' political preferences affect their decisions. n50 If courts other than the United States Supreme Court are examined, there is evidence that the answer to the how much question is not much at all. n51 Brian Tamanaha, in summarizing the extensive research on federal appellate court decisions by Frank Cross, made the following point,

The dominant finding that runs through the book, Cross concluded, "is the importance of law in determining judicial outcomes." "It [was] also noteworthy how very limited the explanatory power of the non-legal variables was." The influence of ideology shows up across the range of areas studied, but "the effect was small." Ideology, gender, wealth, the identity of the litigants, the composition of appellate panels, and other factors were found to have statistically significant correlations with decisions in certain categories of cases, but the impact was not substantial in any particular instance or collectively, and all were negligible in comparison to law. Cross's core findings about the relative insignificance of ideology and the substantial significance of legal factors in determining judicial decisions is consistent with a growing body of quantitative studies of lower federal courts. n52

Of course most of the findings that question the significance of ideology in explaining judicial decisionmaking have been studies of federal courts, especially federal appellate courts. n53 The assumption is that these findings are more applicable to state judiciaries involved in the debate over the appropriate system of judicial selection than are studies of the United States [*232] Supreme Court where ideology appears to be considerably more significant in explaining judicial decisionmaking. n54

While this research showing a limited effect of ideology on judicial decisions supports those who argue against the value of accountability in selecting judges, the research does not completely undermine the argument for judicial accountability. As Gregory Sisk pointed out in a review of Cross's research,

In addition, a small effect does not necessarily mean an inconsequential effect. That ideology plays any role in judicial decision making is an important and substantive finding, even with the qualifying understanding that the effect is constrained. Furthermore, certain variables, including ideology, are more likely to emerge and have greater substantive effect in certain types of cases ... . n55

The second reformer argument is also an empirically testable one - that judicial elections do not achieve the goal of judicial accountability. It is to this argument that political scientists can make a major contribution.

III. DO JUDICIAL ELECTIONS PROMOTE ACCOUNTABILITY?

One of the major criticisms of judicial selection is that voters do not know who their judicial candidates are and, therefore, voters cannot hold the candidates accountable. n56 There is some empirical support for at least part of this argument. One such study was published in 1991 about name recognition of judicial candidates in Texas. n57 It involved telephone interviews with one thousand Dallas County voters who were asked if they recognized the names of a number of public officials including judges. n58 A United States Senator from Texas had very high name recognition, as did the Mayor of Dallas. n59 A Texas Supreme Court justice had moderately high [*233] name recognition, as did a Dallas County district judge who had been involved in a very visible and controversial case that had been the subject of a movie. n60 Another district judge from Dallas County had moderately high name recognition because of a nationwide controversy over remarks he had made about gay murder victims in a trial he had conducted. n61 No district judge in the survey had high name recognition except for one judge who had remarkably high name recognition - almost as high as the United States Senator and the Mayor of Dallas. n62

The voters were then asked to recall the "public office" held by these officials. n63 The "public office" held by the exceptionally visible district judge was "disc jockey." n64 The reason was that the judge shared the name of the most popular radio personality in Dallas at the time - a disc jockey. n65 It appeared that the judge was benefiting from the high name visibility created by the disc jockey and probably explained why, at a time when Dallas County was heavily Republican, the judge was the last-surviving Democrat. n66

Studies such as this one show that voters tend to be unaware of judges unless the judge has gained unusual media attention. But there is a seminal body of research done by Phillip Dubois that argues that voters can use the party affiliations of judges as a "cue" to their values and can make reasonable choices as to the values they would like to see reflected on the bench by voting on the basis of their preferred party affiliation. As Dubois pointed out:

In most kinds of elections, party affiliation has served as the mechanism by which voters make these assessments. Because they are aware of the differences between the parties on those issues that are most salient to them and tend to identify with the party that most closely reflects their own attitudes on public issues, voters' reliance on the partisan label choices is, in a very real sense, a rational act. This is no less true in judicial elections. "Knowing [*234] relatively little about the candidates, voters can be expected to rely heavily on what they do know, and studies suggest that the most readily available cues are important in structuring the vote." Thus, research has repeatedly demonstrated that where the partisan cue is available, judicial voters will rely upon it. The availability of the party label both prompts voters to exercise a choice, thereby increasing the percentage of the eligible electorate participating in the election, and results in the expression in the aggregate of the voters' preferences for the direction of judicial policy. n67

From this perspective, of course, partisan elections are ideal as a method for voters to express their preferences in judicial elections since the party cue is readily available unlike nonpartisan elections where information on the values of judicial candidates is harder to obtain.

It is the case that voters do not vote for all offices on the ballot. One would not, for example, expect a candidate for district judge, sheriff, or constable to get the same number of votes as the candidate at the top of the ticket. These candidates will get fewer votes, of course, and this is known as ballot roll-off. Ballot roll-off in state supreme court races can be defined as the percentage of voters casting votes for the top of the ticket that did not vote in each supreme court race. n68 Overall, in state supreme court elections from 1990-2004, ballot roll-off averaged 22.9%, which means that a little less than one of every four voters for top-of-the-ticket candidates did not vote in state supreme court elections. n69 However, there is a vast difference in ballot roll-off in various kinds of state supreme court elections. Average roll-off in partisan elections was 14% compared to 27% for nonpartisan elections and 26% for retention elections. n70 When only contested elections are examined, roll-off was only 11% for partisan elections and 22% for nonpartisan elections. n71 [*235] For the advocates of judicial accountability, these data are important because partisan elections produce the highest degree of citizen control of the judiciary, and the number of voters who cast ballots in contested partisan judicial elections compared to the number of voters who cast ballots at the top of the ticket is reasonably high - suggesting considerable voter involvement in judicial races.

One study of judicial elections (partisan, nonpartisan, and retention elections) found that between 1980-2000, 64 of 735 state supreme court justices were removed from office. n72 This is a defeat rate of 8.7%. However, the defeat rates show great variability between partisan, nonpartisan, and retention elections. n73 Defeats in retention elections are very rare. Of the 327 incumbent justices in retention elections during this time period, only 6 (1.8%) of the justices were defeated. n74 In contrast, with nonpartisan elections, seventeen justices were defeated for a defeat rate of 7.4%. n75 By far the most competitive judicial elections are partisan elections. During this time period, there were 179 incumbents running in partisan elections and 41 (22.9%) of the justices were defeated. n76 Interestingly, the defeat rates for justices in partisan and nonpartisan elections during this time period are greater than the defeat rates for incumbent members of the United States House of Representatives. House members had only a 5.6% defeat rate. n77 That study also found that there was significant electoral competition in state supreme court races (nonpartisan and partisan) during the period from 1980-2000. n78 Of the 229 incumbents running for reelection in nonpartisan elections during that time period, 116 (50.7%) were challenged by an opponent. n79 In partisan judicial elections, there [*236] were 179 incumbents running for reelection during that time period and 121 (67.6%) were challenged by an election opponent. n80 Another measure of electoral competitiveness in judicial elections is the percentage of the vote for incumbents in state supreme court elections in cases where the elections are contested. In contested nonpartisan elections, from 1980-2000, incumbents received an average of 58% of the vote and in contested partisan elections, incumbents received an average of 61% of the vote. n81

Clearly, accountability works in the sense that, in the words of political scientist Melinda Gann Hall, "accountability is manifested simply by virtue of the fact that citizens have formal control over who holds office, including the ability to oust sitting judges ... ." n82 Additionally, "accountability can be seen as a product of electoral competition, produced by the willingness of challengers to enter the electoral arena and the propensity of the electorate not to give their full support to incumbents." n83

Judicial elections - especially partisan judicial elections - produce significant electoral competition for judges. The question remains, of course, as to whether judges are actually responsive to the electorate in their judicial decisionmaking. One study that attempts to explore this issue is by Paul Brace and Brent Boyea. n84 They explored reversals of capital-punishment decisions by appellate judges and found that in states where public support of capital punishment was the highest, capital punishment reversals were the lowest in the states that elected their judges. n85 In those states, the capital punishment reversal rate was 20%. n86 In contrast, in the states that appointed their judges where the support for capital punishment was greatest, reversals occurred 29% of the time. n87 [*237] That is, of course, almost a 50% difference in reversal rates. n88 Brace and Boyea concluded that such a finding suggested, "elected judges were sensitive to public opinion and electoral pressure." n89 Of course, that notion reflects the whole concept of judicial accountability.

In another article, Paul Brace and Melinda Gann Hall studied the votes of state supreme court justices in death penalty cases. They found that votes to uphold the death penalty are more likely in states with elected judges that are politically competitive states. n90 That finding, of course, supports the Brace and Boyea study in that it suggests that judges in politically competitive states might be inclined to be sensitive to an electorate's tough-on-crime views.

One study of damage awards suggests that elected judges award higher damages in cases involving out-of-state businesses than appointed judges do. n91 Strangely ignoring the role of juries in assessing damages, Alexander Tabarrok and Eric Helland argued that most plaintiffs were likely to be in-state voters and many defendants would be out-of-state. n92 Elective judges, they argued, had an incentive to aid plaintiffs by transferring wealth from out-of-state businesses to in-state plaintiffs. n93 They found that damage awards were much higher in states with an elective judiciary when the defendant was an out-of-state business. n94 Their statistical analysis showed that damage awards were about $ 240,000 higher in states with partisan elected judges where the defendant was from out of state compared to states with other judicial-selection systems. n95 If the authors are correct in their interpretation, this study shows a remarkable sensitivity to a judges' constituency in elective states.

Judicial reformers see campaign contributions in judicial elections as an evil that is endemic to judicial elections. As [*238] Justice Sandra Day O'Connor said recently in a lecture at Columbia University Law School, the system of electing judges in New York and other states was "a form of corruption." n96 She said, "There's too much cash in the courtrooms and we need to get it out." n97 That cash amounts to huge sums. In 1990-99, judicial-campaign fundraising amounted to $ 83.3 million; in 2000-09, fundraising was $ 206.9 million. n98 Candidates in partisan state supreme court elections raised $ 153.8 million in 2000-09 compared to $ 50.9 million for nonpartisan state supreme court elections and $ 2.2 million for retention elections. Some nonpartisan elections, however, such as Wisconsin in 2007-08 and Georgia in 2006 have proven as costly as partisan elections. n99 The money, to an enormous degree, comes from interests concerned with tort law where business and conservative groups back candidates opposed by plaintiffs' lawyers and unions. n100

Some research does show a strong correlation between campaign contributions and judicial decisions. n101 The problem, however, is one of causation. While the research can show that contributions from a specific interest are related positively to judicial decisions favoring that interest, the direction of the causation is unclear. That is, do the contributions cause the judicial decisions or is the money going to a judicial candidate who is sympathetic to the perspective of the contributing interest? James Gibson concluded, "No clear conclusions emerge from this literature [on campaign contributions and judicial decisions] on whether contributions actually affect decisions." n102

[*239] Reformers such as O'Connor claim these sums at least lead to the perception that "campaign contributions affect courtroom decisions." n103 There is poll evidence that shows that the public and even a substantial proportion of judges believe that campaign contributions affect judicial decisions. n104 Additionally, there is experimental research that shows that campaign contributions from those with direct interests before a court (or other governmental institution) do lead citizens to perceive that decisionmaking will be biased. n105 Indeed, James Gibson's experimental research explores the effect of campaign contributions on the legitimacy of courts as well as the recent use of policy pronouncements by judicial candidates and the use of attack ads on the legitimacy of courts. n106 The experimental study, done in Kentucky, uses a series of vignettes to show a sample to voters to examine how contributions, policy pronouncements, and attack ads affect legitimacy. n107 He concludes:

Given the research design employed in this analysis, I am entitled to conclude with considerable certitude that when candidates for public office receive campaign contributions from those with direct business interests before the institution, many (if not most) citizens perceive policy making as biased and partial and the policy-making institution as illegitimate. Similarly, the use of attack ads causes many to question institutional legitimacy. Just as certain are my conclusions that policy pronouncements by judicial candidates cause little harm to courts ... . n108

However, if judicial accountability is a goal, campaign contributions from parties with interests before the courts create considerable problems. These campaign contributions do lead to a perception of impropriety and challenge the legitimacy of courts. n109 At the same time, as Bonneau and Hall point out,

[*240]

competitive, well-financed campaigns substantially improve the willingness of citizens to participate in elections. Given this fact, we are not entirely sure why we should be concerned about amounts of money spent in supreme court campaigns, absent a showing of adverse effects on candidate entry, quality recruitment, or other important aspects of the political process like citizens' perceptions of courts and their observable consequences. If judicial elections are to be efficacious mechanisms of democratic control, then it is imperative that voters participate in them after being sufficiently educated about the candidates. In a healthy democracy where citizens garner knowledge about candidates and are mobilized to vote through election campaigns, we should be open to the fact that lively well-financed campaigns actually might be preferable to lackluster and poorly financed campaigns that fail to serve voters. n110

In a quantitative analysis of state supreme court races from 1990-2004, Bonneau and Hall found that contested races for state supreme courts are expensive if the race is a closely contested one; if only one seat on the court is up for election; if the race is single-member; if the race is in a nonpartisan election; if the judicial term of office is longer; and if the court hears a relatively high proportion of tort cases. n111 It should of course, come as no surprise that tort law is a driving force in making state supreme court elections high-dollar elections. n112 [*241] Thus, state supreme courts with a large tort-law docket should reasonably be expected to generate significant campaign funding from interests concerned with tort law.

One interesting aspect of these findings is that long judicial terms are related to expensive judicial races. One proposed reform in judicial elections to reduce the political battles that have occurred in state supreme court races in recent years has been to lengthen judicial terms. n113 The idea has been, essentially, that longer terms mean fewer elections that mean fewer costly and noisy supreme court election battles. n114 However, this finding suggests that shorter judicial terms will lead to less expensive judicial races than longer ones. Most importantly, this study identifies an expected and key finding. That is, nonpartisan elections tend to be costlier than partisan elections. n115 Reformers have opposed partisan elections because of the role of political parties in these elections and have frequently offered nonpartisan elections as a preferred alternative to partisan elections. n116 However, Bonneau and Hall stress that,

Partisan elections reduce the costs of campaigning and provide critical cues to voters selecting between competing candidates for judicial office. Alternatively, nonpartisan elections increase information costs to voters, reduce the likelihood that voters will cast ballots in supreme court elections even when voters already are at the polls to vote for other important offices, and raise the costs for candidates seeking office. n117

In another statistical analysis, Bonneau and Hall establish that increased spending in state supreme court elections substantially increased voter participation in these races. n118 As they state,

[*242]

Whether measured as the overall spending in each election or in per capita terms, the fact remains that expensive campaigns serve to encourage participation in elections for the state high court bench once those voters already have gone to the polls to cast ballots for other important elections. In short, money means voters in supreme court elections. n119

Finally, a mainstay argument of reformers is that accountability does not work because voters are so uninterested and have such limited information that they make irrational choices in electing judges. n120 A more sophisticated version of this argument is that elections, unlike merit selection, are not designed to focus on judges' qualifications because elections do not have formal screening or evaluation systems as do many merit-selection systems. n121 However, if elections are so poor a mechanism for judicial selection, one would expect that the qualifications of judges chosen in elections would be dramatically lower than the qualifications of judges chosen in merit-selection systems where, at least in theory, there is such attention to choosing judges on the basis of their qualifications. Henry Glick observed that when commissioners in fifteen commission-selection jurisdictions were asked to identify characteristics that were important in judicial selection, they gave fifty-two responses that often involved vague, subjective criteria such as temperament, neutrality, and ability. n122 When Glick compared merit-selected judges with elected judges on measurable criteria, he found that the merit-selected judges did "not consistently produce obviously superior judges in terms of quality education, cosmopolitan backgrounds, previous judicial experience or nonpartisan careers." n123 And when Glick and Craig Emmert examined the characteristics of 324 state supreme court justices who were serving from 1980-81 in terms of [*243] possible quality variables such as localism, education, legal experience, and political experience, they found that state judges have more similarities than differences. n124

Stephen Choi, Mitu Gulati, and Eric Posner have done the most sophisticated comparison of appointed judges to elected judges. n125 They compared these judges on three dimensions of judicial performance: effort, skill, and independence. n126 Effort was measured by comparing the productivity of judges in terms of the number of opinions written. n127 Skill examined the quality of judicial opinions measured by the frequency of out-of-state citations of judicial opinions. n128 Independence is measured by whether judges write opinions against judges of the same or the opposite party. n129 The assumption is that judges who write dissents against majority opinions authored by judges of the same party (or majority opinions against dissents of same party judges) are more independent than judges who rarely dissent or who do not dissent against judges of the same party. n130 In terms of productivity, Choi, Gulati, and Posner found that elected judges write more opinions than do appointed judges. n131 However, they also found that while appointed judges write fewer opinions, those opinions garner more citations than do the opinions of elected judges. n132 In reference to the independence dimension, they found that "the independence levels of elected and appointed judges are not clearly different." n133 Choi, Gulati, and Posner also found that there are some differences in the characteristics of elected and appointed judges, some of which were not found in earlier research. They found, for example:

Compared with Appointed judges, Elected judges make more campaign contributions; are paid less; are on less [*244] stable benches; and have shorter tenures. We also examined the law school that our sample judges attended. A little under 70% of our Election Partisan judges attended an in-state law school compared with only 33% of the Appointed judges... . In addition, the US News ranking... of the law school for Election Partisan judges was on average equal to 57.9, whereas the mean ranking for Appointed judges was equal to 32.3... . Election Partisan judges are more likely to have gone to a law school in the state in which they sit and are more likely to have gone to a lower-rank law school. They are, in short, more politically involved, more locally connected, more temporary, and less well-educated than appointed judges. They are more like politicians and less like professionals. Politicians are likely to see their role as judges as predominantly one of resolving disputes (as many as possible), whereas professionals are more likely to see their role as advancing the law. This might explain the productivity and quality differences. n134

There is simply no convincing evidence that merit-selection judges are better judges than are partisan-elected judges; indeed, as the authors of this study point out, the results suggest "a more complicated story" that suggests that the conventional wisdom that appointed judges are better than elected judges needs to be reexamined. n135

IV. CONCLUSION

At its core, the debate on judicial-selection methods is a debate over two conflicting values - judicial independence and judicial accountability. Indeed, many of the research findings presented in this paper that support judicial accountability could easily be used by an advocate of judicial independence to criticize judicial elections. For example, advocates of judicial accountability argue that substantial campaign funds aid judicial candidates in presenting their views to the electorate and increased voter awareness of and voter participation in judicial elections. Advocates of judicial independence, on the other hand, are concerned that these contributions create an unhealthy [*245] alliance between judicial candidates and campaign contributors and that, at the minimum, an appearance of impropriety is created. It is the choice between these core values of independence and accountability that should determine which system of judicial selection is appropriate. That choice is a normative one and political scientists are unable to contribute much to that decision.

However, the debate over judicial selection has ranged far afield. Given a huge body of research over the past sixty years by political scientists on judicial decisionmaking where appellate judges (as least United States Supreme Court Justices) make policy choices and where the attitudes of judges have proven an important determinant of judicial decisions, it makes no sense to argue that judges are neutral actors who simply apply the law and so should be independent of the electorate. It is more appropriate to argue that ideology is involved in judicial decisionmaking, but to a limited degree - whether that degree of judicial discretion justifies judicial elections is, of course, a debatable point. Reformers also argue that judicial elections do not promote judicial accountability, but political science research does show that elections - most notably partisan elections - lead to relatively high levels of voter participation, high levels of electoral competition, and high levels of defeats of incumbent judges. It is astounding that such a level of electoral involvement in judicial elections can be seen as non-accountability. While it might be argued that voters do not know who judges are and, therefore, voter control over the judiciary is a random process, with partisan elections at least, political scientists have shown that the party label is a crucial cue to the attitudes and values of judicial candidates, and voters can reasonably use the party label as a way to vote for judges who share their values. Reformers have also argued that campaign contributions in judicial elections have a corrupting effect on state judiciaries. There is some evidence to support this claim, but much stronger evidence that campaign contributions from parties with interests in case outcomes do create an appearance of impropriety and illegitimacy of court decisions. On the other hand, if one's primary goal is insuring accountability of the judiciary, campaign money increases voter [*246] interest and voter awareness of judicial candidates and, from that perspective, has value in enhancing accountability.

There is research that shows elected judges can be sensitive to their constituency in such areas as the death penalty and damage awards that raises fundamental questions of fairness in a system where judges must cultivate voter preferences. Yet, it might be argued that judges should have some sensitivity to those values of the community over which they preside.

Elective systems may produce different judicial personalities than appointive systems. For example, elected judges may be more parochial and more "political" personalities than appointed judges. However, it is a stretch to then claim that appointed judges are "better" judges. Elected judges are more productive than appointed judges; though appointed judges may write better quality opinions in the sense that their opinions are more cited out of state. Rather than arguing over which system produces the "best" judges, this research suggests that different systems of selection produce judges who have different strengths.

Although a good argument can be made that some political science research has been unnecessarily polarizing in attacking judicial reforms and reformers, n136 this does not mean that political scientists are necessarily opposed to all judicial reforms - either systemic or incremental. Certainly no political scientist has ever claimed that judicial accountability should be the only goal of a judicial-selection system. Nor has any political scientist taken an absolutist position such that any behavior that enhances judicial accountability is justified. No political scientists have ever claimed that there are no problems with judicial elections. Indeed, it is frightening to contemplate the consequences of a judicial-selection system that only considered judicial accountability as a desirable goal. However, political science does emphasize that many of the perceived problems with judicial selection have simply not been empirically tested. We do not know, for example, the degree of harm that the loss of judicial experience through judicial defeats [*247] actually has on the administration of justice. n137 What empirical research done by political scientists does suggest is that judicial elections are useful mechanisms in promoting judicial accountability, although this research has also identified flaws in judicial-election systems. Finally, this research has been useful in modifying and adjusting some of the more extreme claims of reformers who only see evil in electing judges.

Legal Topics:

For related research and practice materials, see the following legal topics:

Civil ProcedureJudicial OfficersGeneral OverviewGovernmentsCourtsJudgesGovernmentsCourtsJudicial Precedents

FOOTNOTES: