Is the New York State Court of Appeals still "friendless?" An empirical study of amicus curiae participation.
Amici--because they have a wider perspective, or simply a different perspective--can be of inestimable value to courts in discharging their responsibility that extends beyond the litigants. An amicus can alert the court, as the parties perhaps cannot or would not wish to, that a large issue lurks in an appeal; that a case of seeming insignificance has potentially wide ramifications and will likely have major impact when a ruling is applied in other factual settings; or that a case of obvious major significance could conceivably have wholly unanticipated effects. Amici can sensitize the court--when it may be irrelevant to the litigants, whose objective is to win--to the appropriateness of narrowing or limiting a holding, and with other factual situations in mind they can suggest alternate rationales for achieving results urged by the parties.
Judge Judith S. Kaye (1989) (1)
The New York State Court of Appeals decides a wide range of significant statewide issues which have long-lasting effects on the citizens of the State of New York. Similar to the United States Supreme Court in the federal arena, the primary function of the Court of Appeals is "declaring and developing an authoritative body of decisional law for the guidance of the lower courts, the bar and the public." (2) During any given year, the court decides appeals involving state and federal constitutional law, administrative law, products liability law, environmental issues, zoning matters, rent cases, criminal cases, and tort issues, among many others. Based on this amount of authority, one would expect that non-parties would choose to advocate before the court if there was any mechanism to influence the decision-making process. In fact, there is such a mechanism available: the amicus curiae brief. Through this form of participation, non-parties to appeals are able to engage the court by submitting briefs as amicus curiae, or as a "friend of the court." (3) As of 1989, however, the court has been notably "friendless." (4)
In 1987, the Court of Appeals decided 369 appeals of which only thirty involved amicus briefs. (5) The U.S. Supreme Court, by contrast, consistently decides individual cases in which dozens of amicus briefs are filed. (6) In fact, in 1989, the Court heard one case, Webster v. Reproductive Health Services, (7) in which more amicus briefs were filed than during the entire term of the New York State Court of Appeals. (8) Over the past fifty years, the federal court system has seen a clear and rapid influx of amicus curiae submissions. (9) Whereas between 1946-1955 the percentage of cases decided with at least one amicus brief was only 23% in the U.S. Supreme Court, that number jumped to over 85% of cases between 1986-1995. (10)
The disparity in the number of filings becomes even more surprising considering the Court of Appeals' policy towards amicus briefs. Not only have individual judges encouraged the practice, (11) but the court has expressed interest in amicus filings by amending its rules to enable it, sua sponte, to invite submissions. (12) In December of 1988, the court even added a preamble to its weekly list of new filings which encourages the submission of amicus briefs. (13) Moreover, the court overwhelmingly grants motions for amicus curiae relief. (14)
Not everyone would identify the lack of amicus curiae participation in the court as a negative. Perhaps unsurprisingly, judges have not hesitated at expressing their opinions concerning the value of amicus briefs. Judge Posner has lamented that the "vast majority" of amicus curiae briefs "have not assisted the judges," and are "an abuse" of non-party participation. (15) Justice Scalia, in reference to several amicus curiae submissions in one case, noted that "[t]here is no self-interested organization out there devoted to pursuit of the truth in the federal courts." (16) Justice Alito, however, has argued that "an amicus who makes a strong but responsible presentation in support of a party can truly serve as the court's friend." (17)
The more important inquiry may be what effect these briefs have on a court's decision-making process. Although most literature concerning the manner in which courts engage in decision-making is rooted in a normative approach, or by arguing, on a case-by case basis, that the court simply got the law wrong or right, (18) amicus curiae briefs are peculiarly suitable for study from a social science perspective. That is, by using empirical data to assess their impact on the outcome of litigation. (19) In fact, several studies have done just that by addressing the impact of amicus curiae briefs on the Supreme Court. (20) Notably, these studies have focused almost exclusively on the federal courts.
As a byproduct of these studies, several theories have emerged concerning the manner in which amicus briefs affect the decisionmaking process of courts. The first, and simplest, theory is that amicus briefs have no effect on the outcome of cases. (21) This theory, which statisticians would call the null-hypothesis, is consistent with the attitudinal model of judicial decision-making which suggests that judges have preconceived ideologies and will decide cases in a manner consistent with these views. (22) Thus, even if amicus curiae briefs provide supplemental information to the court, this information would be irrelevant since judges need very little information to decide issues beyond the basic facts and questions presented. (23)
In contrast, two other theories have emerged arguing that amicus curiae briefs do have a quantifiable effect on judicial decisionmaking: the information theory and the affected groups theory. (24) The information theory suggests that certain amicus briefs are more effective as a litigation tool because they provide useful information to the court that is not already present in the record. (25) This theory is consistent with the legal model of judicial decision-making, which holds that judges decide cases based on their interpretation of the relevant law and facts irrespective of outside influences. (26) The statements of Judge Kaye above suggest this theory implicitly. (27)
The affected groups theory suggests that amicus briefs have utility, not from their legal arguments, but simply from being filed. (28) Assuming judges try to resolve cases consistent with public sentiment, the number of amicus briefs filed for a particular party can act as a barometer of public opinion. (29) In turn, judges are then able to construct decisions consistent with public opinion measured by group participation as amicus curiae. This theory is consistent with the interest group model of judicial decision-making, which assumes that judges attempt to satisfy the needs of the most organized interest group before them in court. (30)
Still, there is little doubt that amicus filings have flooded the federal courts as an attempt to control, or at least influence, judicial decision-making. (31) At least as of 1989, the same cannot be said for the New York State Court of Appeals. The question remains, why does the Court of Appeals, which decides issues of statewide significance, have minimal participation from non-parties when amicus curiae briefs are encouraged by the court? Moreover, has it changed? Have the words of Judge Kaye gone unheeded, or have they inspired more amicus filings over the last twenty years? And if it has changed, what impact have amicus briefs had on the court's decision-making process?
In an attempt to answer these questions, I conducted an empirical study of amicus curiae participation in the Court of Appeals over the last twenty years. To accomplish this task, I assembled a large database consisting of all Court of Appeals cases which yielded a published opinion from 1988-2007, as well as selected years in the past for comparison. For each decision, I recorded, among other things, the outcome of the case, whether an amicus brief was filed, the number of amicus briefs supporting appellant, the number of briefs supporting respondent, and the author of each amicus brief. In addition, I also conducted a citation study to identify how often amicus briefs are cited in the court's opinions. After constructing the database, I used statistical techniques to assess trends in amicus filings, as well as the influence of amicus briefs on the outcome of cases in the Court of Appeals.
I. THE RISE OF AMICUS CURIAE BRIEFS IN STATE COURTS
The study of amicus curiae briefs, unsurprisingly, has focused almost exclusively on the federal courts. Amicus curiae participation is booming in the federal court system and plays an important role in judicial decision-making. This section explores the origin of the amicus curiae brief and tracks its evolution in state courts.
A. The Origin and Transformation of the Amicus Curiae Brief
Amicus curiae, or "friend of the court," refers to the tradition of courts permitting non-parties to litigation to submit their opinion on issues involved in the proceedings. (32) Although not always having a personal interest in the outcome of the case, amicus curiae are often particularly informed on the issues, or at least interested in the underlying ruling and its policy implications. In order to participate in a case, an amicus almost always needs permission of the court to file a brief as a non-party. (33)
Amicus curiae have a long history in judicial proceedings. (34) Some scholars assert that amicus curiae are rooted in the Roman Law, and were court appointed to offer non-binding opinions to the court. (35) The more common conception of amicus curiae, however, traces back to the English Common Law in the seventeenth century. (36) First used at common law primarily for oral "Shepardizing," amicus filers also called attention to "manifest error, to the death of a party to the proceeding, and to existing appropriate statutes." (37) For example, in Prince's Case, (38) two amici alerted the court to a provision in the Act of Parliament which they thought helpful to the decision of the case. (39) Common law courts, however, consistently viewed amicus curiae as neutral outsiders or as "detached servant[s] of the court." (40)
Certain characteristics of the federal system in the United States, namely a strict interpretation of federal subject matter jurisdiction and a general hostility towards third-party intervention, (41) led to a transformation of the traditional, common law use of the amicus curiae brief. (42) These institutional restraints encouraged amicus curiae submissions not as a means to provide scholarly advice to the court, but to protect the interests of parties precluded from litigating. (43) And though the restraints on intervention began to subside in the late 1800s, amicus curiae briefs were still tiled with relative ease in the federal courts. (44)
During this period, the identity of amicus filers also began to transform in the United States. By the late 1800s, the Supreme Court began recognizing influential groups, instead of only lawyers as originally required, as amicus curiae. (45) In fact, the Court began identifying amicus curiae briefs by the group that sponsored it, rather than the individual author. (46) This shift in practice permitted interest groups to openly advocate for a litigant's positions--a role as amicus curiae which was unavailable at common law. (47)
The result of these changes was to transform the role of the amicus curiae brief "from neutral friendship to positive advocacy and partisanship." (48) And by the 1940s, amicus curiae briefs had become a powerful tool in litigation for non-parties. (49) Some even argued that these briefs had become a "genuine problem" due to their "time-wasting character." (50) Similarly, there was a perception that amicus briefs treated the U.S. Supreme Court as a "political-legislative body, amenable and responsive to mass pressures from any source." (51) The Court, in responding to this suggestion, amended its procedural rules in 1949 regarding non-governmental amicus participation and required either consent of all parties or, if consent was unavailable, a motion requesting amicus curiae relief. (52) Although the rule changes in 1949 seemed to have some effect on diminishing the number of briefs tiled with the Court, the onset of "public law litigation" in the 1960s and early 1970s saw the participation of amici surge. (53) For instance, the number of amicus curiae briefs has increased by over 800% in the Supreme Court since 1945, and the percent of cases which included amicus filings grew from 35% to 95% during the same period. (54)
B. Amicus Curiae Practice in State Courts
Historically, state courts were less open to amicus curiae participation than the Supreme Court. Most state courts denied motions for amicus curiae relief unless the prospective party was truly a friend of the court. (55) Today, states vary widely concerning the rules which govern filing of amici curiae. Some, consistent with traditional amicus curiae practice, continue to limit the filing of amicus briefs to neutral outsiders without an interest in the litigation. Other states may limit, not by party, but by function, such as cases involving important public issues. (56)
Similar to the federal courts, the filing of amicus briefs in state courts has increased substantially since 1960. (57) One of the few studies to review amicus curiae participation in state courts was conducted on sixteen state courts of last resort (58) between 1965 and 1990. (59) Lee Epstein found that between 1975 and the early 1980s, amicus filings had more than tripled from a base of 4.3 filings per state per year to 12.7 per state per year. (60) Others have confirmed Epstein's findings and found that the increase in amicus participation has continued through 2000. (61)
Notably, even though amicus curiae participation has been increasing, there is significant variation in filings among particular state courts of last resort. Paul Brace and Kellie Butler recently conducted a study of all state courts of last resort from 1995-1998, and identified three specific categories of states, each having substantially different levels of amicus curiae activity. (62) The first category are those states in which between 5 to 20 percent of cases have amici participation; this category includes about 50 percent of states. (63) The second category includes states which have relatively no amici participation--a group that encompasses about 40 percent of all states. (64) The remaining 10 percent of states include those with high levels of amici participation. (65) Beginning with the state with the highest level of amici participation, they were: California, New Jersey, Michigan, Oregon, Oklahoma, Wisconsin, Louisiana, New York, Arizona, and Ohio. (66) Similarly, in a study conducted by Sarah Corbally, Donald Bross, and Victor Flango, which analyzed a random sample of cases from all state courts of last resort to assess amicus curiae participation over a period of three years (19982000), ten states had a high level of amicus participation. (67) Only five states, however, were in this category for both studies: California, Michigan, New Jersey, New York, and Ohio. (68)
II. THE PARTICIPATION OF AMICUS CURIAE IN THE NEW YORK STATE COURT OF APPEALS
Previous studies have suggested that the New York State Court of Appeals has a significant amount of amicus curiae practice as compared to other state courts. But not only has the court received substantial numbers of amicus curiae submissions, it has had a "long tradition of amicus participation." (69) In fact, the court has consistently encouraged amicus filings in court opinions, (70) official documents distributed by the court, (71) and publications by the judges themselves. (72) Still, as of 1988, the court was concerned with the level of amicus participation. This section begins by reviewing the filing requirements for amicus curiae in the Court of Appeals and then explores the trends in amicus curiae briefs from 1988-2007.
A. An Open-Door Policy Towards Amicus Curiae Briefs
The requirements for filing amicus curiae briefs in the Court of Appeals immediately suggest that the court is receptive to amicus curiae practice. Any non-party, other than the New York State Attorney General, seeking to file an amicus curiae brief must obtain permission by motion from the court. (73) The criteria for admission are broad and require the filer to demonstrate that: "(i) the parties are not capable of a full and adequate presentation ...; (ii) the amicus could identify law or arguments that might otherwise escape the Court's consideration; or (iii) the proposed amicus curiae brief otherwise would be of assistance to the Court." (74) Notably, amicus curiae relief for the Attorney General is governed by separate provisions which permit the Attorney General to file an amicus brief without leave of the court. (75)
Compared to other states, the Court of Appeals has more restrictive rules concerning amicus curiae submissions. (76) Still, the court rarely denies motions to file. For instance, in 1999, the court granted almost 80% of motions for amicus curiae relief. (77) Similarly, in 2004, almost 95% of motions for amicus curiae relief were granted. (78) When the court does deny motions for amicus curiae relief it is typically a result of a failure by the parties to follow procedural rules rather than apathy towards amicus curiae participation. Thus, in 1999, the reasons for denial of amicus curiae briefs included: failing to file motions in sufficient time to permit proper review by the court, and a failing to present law or arguments not already presented by the parties. (79)
Beyond a mere tendency to grant motions for amicus curiae relief, the Court of Appeals explicitly encourages them, (80) and even provides notice to the New York State Bar Association when issues are particularly appropriate for amicus curiae participation. (81) In fact, the court noted "with concern the substantial decline in the number of motions ... for amicus curiae relief in 2000." (82) During that year, the court received only 59 motions for amicus curiae relief, as compared to 87 in 1999. (83) Specifically, the court was surprised that only four motions were filed in the first appeal regarding the 1995 capital punishment statute in People v. Harris. (84) By 2001, however, the court reported that the downward trend from the year 2000 had been "soundly reversed." (85) In 2001, 110 motions for amicus curiae relief were filed of which 94 were granted. (86)
B. Trends Over Time
1. Patterns in Amicus Curiae Filings: 1988-2007
To assess the level of amicus curiae activity in the Court of Appeals, my study provides a complete assessment of amicus activity over a twenty-year period (1988-2007). (87) In addition, I collected data on amicus curiae participation for the years 1959, 1969, and 1979, to assess trends over time. My study suggests that amicus activity in the Court of Appeals, although not progressing at the rapid level of the federal courts, has steadily increased over the last twenty years. An examination of Figure 1 shows a steady increase in amicus briefs over the last twenty years and a rather sharp increase over the past five years.
Notably, during the first five years of the study (1988-1992), 257 briefs were filed with the court. Over the last five years (2003-2007), 399 briefs were filed, an increase of over 55%. When comparing the number of amicus briefs filed in 1959, 1969, and 1979 in Figure 2, the increase in amicus filings becomes clearer. Whereas the average number of briefs filed per year from 19882007 was 62, there were only 34 briefs filed in 1959, 12 in 1969, and 25 in 1979.
The steady increase in amicus briefs is Similar when expressed in terms of the percentage of total argued cases. (88) Figure 3 displays the percentage of cases with an amicus brief over five-year periods. Although not displaying a particularly compelling increase in the percentage of amicus briefs per year, the statistics become more meaningful when compared to 1959, 1969, and 1979. As shown in Figure 4, amicus briefs were filed in only 12.96%, 6.18%, and 9.04% of cases in each of those years respectively. In comparison, over the period of my study, the only years which had less than 20 percent of cases having an amicus brief were 1988 (17.16%) and 1993 (18.18%).
In addition to reviewing the number of briefs per year and the percentage of cases with an amicus brief, the average number of amicus briefs per argued case provides another important measure of amicus participation. (89) Notably, the mean number of briefs per case has increased over each five-year period in the study. While the mean was only 0.4 amicus briefs per case in the first five-year period (1988-1992), over the last five-year period (2003-2007) almost 0.7 briefs were filed per case. Furthermore, when expressed in terms of the cases in which one or more amicus briefs were filed, the mean number of amicus briefs per case increased every five-year period except from 1997-2002, which can be explained in part by the low number of briefs received in 2000. (90)
Not surprisingly, and most likely due to the duration of the study, the median and modal number of amicus briefs per case remained the same over each five-year period. This may suggest, however, that the mean number of briefs per case provides a misleading measure of amicus participation since previous studies in the federal courts have found that a small number of cases attract unusually high numbers of amicus briefs. (91) For instance, between 1986 and 1995 the Supreme Court heard 119 cases which had 10 or more briefs filed. (92)
Since the Court of Appeals receives far fewer amicus briefs than the Supreme Court, cases with unusually high number of briefs were identified as those with five or more briefs filed. (93) From 1988-2007, I identified twenty-seven cases (94) with six or more briefs filed. Of those cases, most involved constitutional law or education. (95) The case with the most number of amicus briefs was Hernandez v. Robles with twenty-four briefs filed, (96) which involved whether the New York Constitution compelled recognition of same-sex marriages. (97) Another case with an unusually high number of amicus briefs was Campaign for Fiscal Equity, Inc. v. State of New York, yielding thirteen amicus briefs. (98) In this case the court was asked to invalidate the state education financing system as violative of the New York Constitution as well as title VI of the Civil Rights Act of 1964. (99) Other notable cases with large numbers of amicus briefs include: Nicholson v. Scoppetta (100) and Zanghi v. Niagara Frontier Transportation Commission, (101) both involving negligence issues with ten briefs filed per case.
2. Identity of Amicus Filers
My database also permitted me to analyze the identity of amicus filers before the court, another measure of amicus activity over my study period. In addition and as will be discussed below, those who regularly file amicus briefs may produce briefs that are more useful to the court, as they have far more experience and interaction with the process than other amicus filers. (102) And identifying those who participate regularly as amicus curiae will be useful as a comparison to other studies which have found that certain parties enjoy higher rates of success in the federal courts as amicus curiae. (103)
Table 1 displays the top ten amicus fliers over the period of my study. Notably, the identity of top amicus filers is consistent with previous studies analyzing amicus curiae participation. Of particular note, the New York State Attorney General was the top amicus curiae filer from 1988-2007. Based on the wide variety of duties conducted by the Attorney General's Office, it is largely unsurprising that the Attorney General chose to participate as amicus curiae significantly more than any other particular group. (104) This level of participation is entirely consistent with that of the U.S. Solicitor General as amicus curiae before the Supreme Court. (105)
Figure 5 displays the number of briefs filed per year by each individual Attorney General over the period of the study. Although not a flawless measure of Attorney General participation as amicus curiae due to the varying length of terms, it does suggest that recent administrations have chosen to participate more readily in this capacity. This is largely unsurprising considering the success the Attorney General achieves when appearing as amicus curiae in support of one of the parties. (106)
3. Subject Matter of Cases Involving Amicus Curiae Briefs
Table 2 displays the top ten subject matters for cases in which at least one amicus brief was filed. Out of all cases with at least one amicus brief, 14.4% were criminal cases. Notably, this was also the subject matter for which the Attorney General was most likely to file an amicus brief (19.3%).
Similarly, when expressed in terms of total number of amicus briefs for each specific subject matter, the results change very little. Table 3 displays the number of amicus briefs filed for each specific subject matter, as well as the percent of total amicus briefs filed.
C. Citation Rates of Amicus Curiae Briefs in Court Decisions
The mere influx of amicus curiae briefs may not necessarily represent whether it affects the court's final product--the case opinion. The rate of citation of amicus curiae briefs, however, does provide a limited assessment of the utility of these briefs for the court. (107) In order to assess the level of quotation or citation of amicus briefs, I conducted a computer assisted search (108) and recorded in each case where there was at least one amicus brief filed whether an amicus brief was cited or quoted in any portion of the opinion. (109)
Figure 6 displays the percentage of cases with an amicus brief cited over each five-year period of the study. (110) Over the past twenty years, (111) an amicus brief was cited in 15.7% of cases with at least one amicus brief filed. (112) There has, however, been a notable increase in the rate of citation over the past ten years. The rate at which amicus briefs are cited has almost doubled from 10.8% between 1988-1997 to 20.6% between 1998-2008. (113)
The increase in the percentage of cases with an amicus brief cited could be explained by the steady increase in the total amount of briefs filed over the past ten years. (114) Unfortunately, it is difficult to measure how many briefs are cited in each case since the court typically refers to amicus curiae by stating "the amici argue" making it difficult to assess who and how many they are actually citing. Nevertheless, the rate of citation has almost doubled in the last ten year period, whereas the total number of briefs has increased by less than 50%. (115) As a result, it seems that the court has been more likely to cite amicus briefs over this period even when considering the increase in the total number of briefs filed.
In addition, when citing amicus curiae briefs, the court has been particularly receptive to the value of the information provided. For instance, in Niesig v. Team I, the court noted the contribution by many of the amici in that case. (116) Specifically, the court stated that amicus curiae submissions have "enlarged our comprehension of the broad potential impact of the issue presented." (117) And the court has even relied on amicus curiae submitted in previous cases in framing issues. (118)
III. STUDIES CONCERNING THE INFLUENCE OF AMICUS CURIAE PARTICIPATION ON CASE OUTCOMES
Even as amicus curiae participation has increased in the Court of Appeals over the past twenty years, the more important inquiry is whether these briefs have any utility for the court. If amicus filings do have an effect on the outcome of litigation, it would explain, at least in part, why non-parties have increasingly utilized amicus curiae briefs. In contrast, if these briefs have no impact on judicial decision-making, it would help to explain why non-parties have been hesitant to file amicus briefs in the past. (119) Since there are few studies focusing on the effect of amicus curiae briefs on the decision-making of state courts of last resort, this section reviews studies conducted on the Supreme Court.
Nathan Hakman conducted one of the earliest reviews of amicus curiae participation in the Supreme Court and argued that their influence was an example of "Political Science Folklore." (120) Hakman concluded that organized interest groups did not play much of a role in Supreme Court decision-making. (121) But Hakman did not engage in a statistical review of amicus participation and failed to assess the effect of specific amicus filers, e.g. governmental entities, states, etc. (122) The shortcomings of Hakman's approach seem to be consistent with several other studies reviewing amicus curiae participation. Although a number of legal scholars explore the effect of amicus curiae briefs, many have focused on the role of amicus briefs in particular issues, notable cases, or specific substantive areas. (123) These studies have focused on assessing the impact of amicus briefs on issues such as federalism, (124) affirmative action, (125) the constitutional rights of women, (126) and in several other areas. (127) For instance, a recent article reviewed the influence of amicus curiae submissions on the Supreme Court's decision in Gonzales v. Raich, (128) which upheld Congress's authority to prohibit the possession and distribution of marijuana under the Controlled Substances Act. (129) No doubt, these studies are valuable in assessing particular issues. But they are narrow in scope, and do not provide the depth or breadth of analysis required to assess the impact of amicus filings on a more strategic level.
More recently, quantitative data analysis has been used to assess the impact and utility of amicus curiae briefs on the outcome of litigation. Still, these studies have reached divergent conclusions concerning the effect of amicus briefs on the Court's decisionmaking process. For instance, using data from ten terms of the Supreme Court over a twenty-year period (1967-1988), Donald Songer and Reginald Sheehan attempted to measure the impact of amicus briefs by identifying "matched pairs" of cases in which, in one case, only one party was supported by an amicus curiae brief submitted by an interest group and, in the other case, neither party had amicus support. (130) Songer and Sheehan found no evidence of amicus group influence and noted that the success rate of parties supported by interest group amicus briefs was virtually identical to parties without any amicus support. (131) But others have concluded that amicus briefs do have a quantifiable impact on case outcomes. For example, using multivariate regression analysis Kevin McGuire found that the likelihood of success before the Supreme Court was significantly related to the level of amicus support for a party. (132)
Several studies have focused on the influence of particular entities or authors participating as amicus curiae. For instance, Gregg Ivers and Karen O'Connor found that the American Civil Liberties Union and Americans for Effective Law Enforcement achieved minimal success as amicus curiae and only when the Supreme Court was ideologically predisposed to reach the outcomes they favored. (133) Still, the most often studied amicus curiae filer has been the U.S. Solicitor General. Jeffrey Segal, analyzing all cases decided between the 1952-1982 terms, found that the Solicitor General achieved unprecedented success as amicus curiae. (134) The Solicitor General's success rate applied for all issues, all changes in Court membership, and regardless of whether the Solicitor General was supporting a liberal or conservative result. (135)
Taking an alternative approach, Linda Simard recently conducted a nationwide survey of federal judges to assess the influence and utility of amicus curiae briefs from the perspective of the judiciary. (136) Based on her survey results, Simard concluded that most judges find amicus curiae briefs helpful to: (1) find new legal arguments not presented by the parties, (2) alert the court to the implications of their decision beyond the immediate litigants, and (3) assist a party who is not adequately represented. (137) Consistent with the studies above, Simard also found that briefs submitted by the Solicitor General are favorably viewed by the federal courts. (138)
To date, Professor Joseph Kearney and Thomas Merrill have conducted the most thorough review of amicus curiae participation in the Supreme Court through an empirical assessment of the influence of amicus curiae from 1945-1995. (139) By assembling a database consisting of fifty years of Supreme Court decisions, Kearney and Merrill, through the use of a variety of statistical techniques, assessed the utility of amicus briefs against three models of judicial decision-making: the legal model, the attitudinal model, and the interest group model. (140) They were able to confirm that the Solicitor General enjoys great success as an amicus filer. (141) In addition, they found no evidence that large disparities of amicus support for one side relative to the other results in a higher probability of success for the supported party. (142) They cautiously interpreted their results as showing support for the legal model of judicial decision-making in which Justices rely on quality information provided by amicus filers to decide the outcome of cases. (143)
Although the study conducted by Kearney and Merrill provides the most robust statistical analysis of the effect of amicus curiae participation on the Supreme Court, literature concerning the affect of amicus curiae briefs on state courts is lacking. (144) Moreover, present studies of amicus curiae participation tell us little about the effect of amicus curiae participation on the New York State Court of Appeals. They are, however, helpful in formulating hypotheses concerning the manner in which amicus curiae briefs affect the court's decision-making process. In essence, they collectively provide a solid foundation to undertake an empirical study of amicus briefs.
IV. THREE MODELS OF JUDICIAL DECISION-MAKING: THE UTILITY OF AMICUS CURIAE BRIEFS AND THEIR INFLUENCE ON OUTCOMES
Emerging from these studies are several competing hypotheses concerning the utility of amicus curiae briefs. Interestingly, each hypothesis fits within one of the several models of judicial decision-making that have been formulated by legal scholars to explain how courts, and typically the U.S. Supreme Court, decide cases. But nothing prevents their application to the New York State Court of Appeals. (145) In fact, since many of the same institutional characteristics relate to both the Court of Appeals and the Supreme Court, the application of these models is particularly appropriate under these circumstances. This section begins by briefly explaining each of the three models of judicial decision-making. It then explains how each model generates a hypothesis concerning the utility of amicus curiae briefs and their influence on the outcome of litigation.
A. The Attitudinal Model: The Null Hypothesis
The preeminent model adopted by political scientists studying the Supreme Court is the attitudinal model. (146) Considered a subset of "rational choice theory," the attitudinal model assumes that judges can "maximize their utility" by deciding cases based on results which will please them ideologically, without regard to other considerations. (147) In other words, "Rehnquist vote[d] the way he [did] because he [was] extremely conservative; Marshall voted the way he did because he was extremely liberal." (148)
Under the attitudinal model, the values of the judiciary are fixed and can be ideologically scaled. (149) Justices may be characterized along a liberal to conservative scale, or even more specifically, such as along a pro-life to pro-choice scale. When facts or case stimuli arise that trigger these values, the decision of the Court will depend on how many Justices share a common value set. (150) For instance, if an issue involves an unreasonable search and seizure and a majority of Justices are in favor of individual liberty interests, then the party representing the individual challenging the governmental action would likely win.
Several factors make the attitudinal model particularly suitable to the Supreme Court. First, the members of the Court "lack electoral or political accountability." (151) Justices serve for life and although Congress can impeach them, this has only happened on one occasion. (152) Second, since serving on the Court is the pinnacle of the legal profession, Justices have no ambition for higher office. (153) In fact, over the last 100 years, only three Justices resigned to take other positions. (154) Third, the Court determines and controls its own caseload. (155) As a result, the ability to decide what cases to hear increases the opportunity for Justices "to make policy by deciding only the most important/controversial/difficult cases." (156) Since the New York State Court of Appeals shares many of the same characteristics regarding appointment and institutional control as the Supreme Court, it follows that application of the attitudinal model is appropriate in this study.
For present purposes, the important part is identifying the relationship between the attitudinal model and amicus curiae briefs. Since the attitudinal model assumes that judges need minimal information to decide an issue beyond the underlying issues and bare facts, it follows that amicus curiae briefs will be irrelevant for the decision-making process. As a result, this generates the null hypothesis for my study, or that amicus briefs have no effect on case outcomes.
Notably, political scientists have recently suggested that judges do modify their positions on issues in light of information about how other institutional actors will respond to their decisions. (157) Accordingly, scholars have sought to explain this behavior by application of a "strategic actor" model which incorporates the attitudinal model but also suggests that judges "are rational actors who modify their voting behavior in order to maximize the chances of their ideological preferences actually being adopted as policy." (158) For instance, if a Justice thought that a lower court decision should be reversed, but knew that if it was granted certiorari the case would be affirmed by the Court, then that Justice would vote to deny certiorari in order to prevent affirmance by the Supreme Court. (159)
Articulating a workable hypothesis for the "strategic actor" version of the attitudinal model is more problematic. Still, one could hypothesize that based on the "strategic actor" model, judges will pay particular attention to specific types of litigants who represent the interests of the executive branch--for this study specifically, the Attorney General. (160) In addition, one would expect that amicus briefs submitted by municipalities may have more success in influencing decisions since municipalities often must implement judicial decisions. (161)
B. The Legal Model: The Information Hypothesis
The second, and more traditional, model of judicial decisionmaking is the legal model. Accepted by the majority of legal scholarship, this model assumes that judges decide cases in accordance with their understanding of the law and facts irrespective of outside influences. (162) To resolve issues, the ideological preferences of judges are irrelevant. Rather, judges will look to sources of law--such as statutes, constitutions, and precedents--for guidance. (163) In turn, the most formalistic version of the legal model assumes that issues have one correct answer that judges discover through legal scholarship. (164)
Recently, the legal model has drawn sharp criticism among many legal theorists. (165) Some assert that the model is not falsifiable and, therefore, impossible to test. (166) For instance, if judges decide a case in a certain manner, the legal model explains that decision by assuming that it was the only correct answer, or at least the answer which the deciding judge or judges believed to be the correct legal answer. Thus, the legal model cannot predict future outcomes. Rather, the legal model simply serves as an after-the-fact rationalization for decisions. (167)
In response, proponents of the legal model have argued that even though legal issues often involve a complex interplay between law and facts, judges are still guided primarily by traditional sources of law. (168) Based on this landscape, proponents argue, it is unrealistic to expect the legal model to predict future outcomes. (169) In addition, proponents of the more modern formulation of the legal model have sought to incorporate a judge's ideological preferences while maintaining that judges still decide cases through legal scholarship. Scholars have asserted that the legal model assumes that a judge's decision "reflect[s] his own intellectual and philosophical convictions in making [a] judgment, [which] is ... very different ... from supposing that those convictions have some independent force in his argument just because they are his." (170)
The legal model provides the framework for the information hypothesis which assumes that amicus briefs have value from the quality of information they provide to the court. (171) The information provided in amicus briefs, in turn, provides the framework by which judges can decide the law. Notably, the rules surrounding amicus curiae briefs submissions to the Court of Appeals suggest that amicus briefs provide an informational value to the court. (172) Amicus briefs will be accepted if they "could identify law or arguments that might otherwise escape the court's consideration; or ... otherwise would be of assistance to the court." (173) The informational value of amicus briefs is also suggested by Judge Kaye's comments in the New York State Bar Journal. (174)
Testing the information theory, however, is difficult since it depends on identifying briefs which provide useful information to the court. In other words, the information hypothesis is that high quality briefs can influence the outcome of a case. (175) And repetitive briefs or those which do not provide valuable information to the court should not influence outcomes. (176) As a result, testing the legal model depends, in part, on how one defines high quality briefs. (177)
C. The Interest Group Model: The Affected Groups Hypothesis
The third model of judicial decision-making, the interest group model, assumes that judges attempt to satisfy the needs of the most organized interest group before them in court. (178) In contrast to the attitudinal model which assumes that judges have fixed ideologies and seek to decide cases based on them, the interest group model holds that judges have no fixed beliefs but rather use their position to satisfy the most private good. (179)
This model runs afoul of many of the traditional notions of judicial decision-making, e.g. judges are immune from majoritarian pressures. (180) Two explanations have been formulated to explain why even Supreme Court Justices may be influenced by public opinion. First, since Justices share policymaking authority with the other branches of government, they have an incentive not to circumvent public opinion. (181) In other words, if Justices were to "stray too far from public opinion ... it is likely that the legislature may attempt to alter or override their decision or the executive may indifferently enforce the decision." (182) Second, in order to "ensure the institutional legitimacy of the Court," Justices are responsive to public opinion. (183) Since the Court cannot enforce their decisions but must rely on the goodwill of the public to maintain their legitimacy, Justices may feel a need recognize popular sentiment. (184)
The implications from the interest group model are relatively clear. Courts operating under such a theory would need information about public opinion on certain issues. Although data may be available on controversial topics such as abortion or the death penalty, for the vast majority of issues the court will need an alternative measure of public opinion. In turn, the Court can rely on interest group participation as a measure of public opinion. (185)
The interest group model, therefore, generates the affected groups hypothesis. These briefs are, arguably, a barometer for public opinion. Therefore, under the affected groups hypothesis the court will decide cases based on the level of amicus participation for one party versus another, irrespective of the quality of information provided. (186) In fact, the court would need very little information to decide the case beyond how many groups have filed for one side, the information contained in the briefs would be irrelevant. (187)
V. AN EMPIRICAL STUDY OF AMICUS PARTICIPATION IN THE NEW YORK STATE COURT OF APPEALS
To test the competing hypotheses concerning the influence of amicus curiae briefs on the New York State Court of Appeals, I conducted an empirical study. Since there is no database available which includes quantitative data relating to amicus participation in the court, I decided to create my own. In doing so, I constructed a database of Court of Appeals decisions over the last twenty years, as well as selected years in the past for comparison.
My database consists of all decisions which yielded a published opinion from 1999-2008, as well the years 1959, 1969, and 1979. (188) For each of the 2,503 cases, I recorded a number of descriptive variables. The outcome of each case was classified into one of three categories: A-wins (short for appellant wins), meaning that the judgment below was reversed or set aside; A-loses (short for appellant loses), meaning that the judgment below was sustained; and mixed result, meaning that the judgment below was partially sustained and partially set aside, (189) In addition, the subject matter of each case was noted. (190) I also recorded whether an amicus brief was filed, the number of total amicus briefs filed, the number of amicus briefs filed for appellant, the number of amicus briefs filed for respondent, and the number of briefs which could not be characterized as associated with either party. (191) Furthermore, for each amicus brief filed, I recorded the entity who submitted the brief and the author for the entity, if applicable. (192)
Before conducting an analysis of the separate hypotheses raised above, I used my database to compute a benchmark rate of success. That is, the rate of success for appellants and respondents when no amicus briefs are filed. This approach is consistent with other studies which have sought to measure the effect of amicus participation. (193) These benchmark rates of success permit a useful comparison to cases involving amicus briefs. For instance, according to the interest group model, one would expect that filing more briefs for one side should raise their overall rate of success relative to the benchmark rate of success. If, however, cases involving disproportionate amicus support for one side have the same rate of success as the benchmark rate it would tend to cast doubt on the interest group model.
In order to analyze benchmark rates of success, I first filtered out all cases which included an amicus brief and then computed the average rate of success for appellant and respondent. For my twenty-year study period, I found that appellants were successful 36.5% of the time, respondents were successful 56.1% of the time, and mixed results occurred 7.5% of the time. Figure 7 summarizes the benchmark rates of success for each five-year period and overall. (194)
After computing the benchmark rates of success, I was able to present a bivariate analysis showing each independent variable's individual effect on the benchmark rate of success. Since these preliminary indicators suggest that certain amicus variables are quite successful, I decided to further explore this influence by constructing multivariate regression models. In particular, these models help to confirm the effects of different variables on the probability of producing a certain outcome--in this case an appellant win.
To accomplish this task I used logistic regression analysis. Logistic regression permits me to predict a binary outcome, e.g. success/failure, from a set of independent variables. For purposes of this study, I am attempting to predict what amicus variables help predict whether appellant wins. Thus, where my dependent variable is A-win (coded 1 or 0), either the case resulted in an appellant win or it did not. Similarly, each of my independent variables--which are discussed in more detail below--are coded as dummy variables (coded 1 or 0), either the independent variable is present or it is not. This allowed me to assess the influence of specific case characteristics (e.g., Attorney General files amicus curiae brief for appellant) on the outcome of the case.
A. Testing the Null Hypothesis
The attitudinal model predicts that judges decide cases based on their personal, political, or social ideology. (195) Thus, certain amici should have a high rate of success since these groups represent the political viewpoints of the judges. If amici reflecting various political viewpoints had a higher rate of success as compared to the benchmark rates above, this would seem to cast doubt on the attitudinal model. As a result, using the benchmark rates from above, I assessed whether the filing of an amicus brief for appellant or respondent increased the likelihood of an appellant win. In other words, I focused on whether filing a brief for appellant increased appellant's win rate and, similarly, whether filing a brief for respondent would decrease appellant's win rate. (196) I decided not to present results using different benchmark rates of success for appellant and respondents since doing so would create results which would likely be confusing. (197)
Figures 8 summarizes the extent to which amicus filers supporting appellant have achieved results that diverge from the benchmark appellant win rate across each five-year period of the study and overall. Over the past twenty years, amicus briefs supporting appellant have achieved a rate of success which is 12.34% higher than the benchmark rate of success. And for every five-year period there is a minimum 13% increase in appellant's rate of success. In addition, for the five-year period of 1998-2002, appellant's rate of success jumped by 30.20% compared to the benchmark rate of success. (198)
In contrast, amicus briefs supporting respondent have not had as strong an impact as compared to appellant's benchmark win rate. Figure 9 shows that amicus briefs supporting respondent achieved a rate of success for appellant which is 6.81% less than the benchmark rate of success over the twenty-year period of the study. Furthermore, during the same time period that amicus briefs supporting appellant saw a high level of success, amicus briefs supporting respondent had no effect on the benchmark rate of success. In fact, from 1998-2002, appellant's rate of success actually increased for amicus briefs supporting respondent. (199)
Table 4 sets forth the results from the logistic regression model used in conjunction with Figures 8 and 9. Although not yet controlling for other variables, the model confirms what was suggested by the bivariate analysis--the filing of an amicus brief has a significant impact on case outcomes. For instance, if an amicus brief is filed for appellant, the odds of an appellant win are 2.2 times greater than otherwise. Similarly, if an amicus brief is filed for respondent, the odds of an appellant win are decreased by almost 50%. Of course, since other variables are not included in this model, its explanatory power is limited. It does, however, suggest that amicus curiae have at least some influence on case outcomes.
B. Testing the Information Hypothesis
The legal model assumes that judges decide cases based upon their understanding of the law irrespective of outside influences. (200) Thus, as discussed above, the legal model generates the information hypothesis, or that quality amicus briefs should have a high rate of success since they assist the court in deciding cases in a manner consonant with the law. The problem becomes deciding how to identify high quality briefs. Clearly, reading each amicus brief would have been unduly burdensome and impractical. And even if I had read each amicus brief individually, any assessment of its quality would have been largely subjective. As a result, based on the composition of the dataset, I chose specific, 'institutional' litigants as a measure of amicus brief quality. I defined institutional litigants as those appearing most often as amicus curiae before the court. This is, arguably, a strong indicator of amicus brief quality since these litigants likely have more experience before the court and in other judicial proceedings. In addition, institutional litigants are more likely to have resources to devote to non-party participation.
Based upon my analysis of amicus author identity, (201) I identified five institutional litigants over the study period: (1) the New York State Attorney General, (2) the City of New York, (3) the New York State District Attorneys Association, (4) the New York State Trial Lawyers Association, and (5) the Legal Aid Society. For each, I computed the change in appellant's win rate when each litigant filed for appellant or respondent as compared to the benchmark rates of success.
Figure 10 displays the changes in the benchmark rates of success when the Attorney General files an amicus brief in support of appellant or respondent. When the Attorney General files a brief for appellant or respondent the benchmark rate of success increases or decreases respectively by almost 50%. This deviation in the benchmark rate of success was the most pronounced among the institutional litigants. (202)
As shown in Figure 11, the change in appellant's win rate is not as large when the City of New York files as amicus curiae. When the City of New York files an amicus brief in support of appellant, appellant's win rate increased by 21% as compared to the benchmark rate of success. When filing for respondent, however, the benchmark rate of success also increased even though the information hypothesis suggests it should decrease. This may be due, in part, to the total number of cases in which the City of New York filed as amicus for respondent (n=13). (203)
The New York State District Attorneys Association, when filing as amicus curiae for appellants or respondents, saw the expected changes in appellant's win rate over the twenty-year period of the study. Figure 12 shows that the District Attorneys Association achieved a positive influence on appellant's win rate when filing for appellant. As expected, there was a corresponding decrease in appellant's win rate when the District Attorneys Association filed for respondent. Again, the underlying sample size of cases is rather small over the twenty-year period. (204)
An entirely different pattern arises when viewing the change in appellant's win rate when the New York State Trial Lawyers Association files as amicus for appellant or respondent. In fact, the exact opposite of the expected result occurs. Appellant's win rate decreases when the Trial Lawyers Association files for appellant and increases when it files for respondent. Notably, this was the only institutional litigant which saw a decrease in appellant's win rate when filing for appellant as amicus curiae. Unfortunately, when splitting cases between those in which the Trial Lawyers Association files for appellant or respondent as amicus, the underlying pool of cases becomes quite small making it difficult to draw statistically sound conclusions. But even with the amount of cases included, the Trial Lawyers Association was the only institutional litigant to see unexpected changes to appellant's win rate when it was present as amicus curiae. (205)
An equally surprising result occurs when analyzing cases in which the Legal Aid Society appears as amicus for appellant or respondent. As shown in Figure 14, whereas appellant's benchmark rate of success increases by over 11% when the Legal Aid Society files an amicus brief in support of appellant, it also increases by 19.1% when the Legal Aid Society files an amicus brief in support of respondent. (206)
The results from the regression model underlying figures 10 to 14 is displayed below in Table 5.
One variable immediately stands out--the Attorney General as an amicus filer. The Attorney General has a statistically significant effect on case outcomes when filing for appellant or respondent. In fact, when the Attorney General files for appellant, appellant is about six times more likely to win the case than when the Attorney General does not file for appellant. Similarly, if the Attorney General files for respondent, the chances of an appellant win decrease by over 80%. Notably, the only other variable to show a statistically significant effect on case outcomes was the District Attorneys Association filing a brief for appellant. The model tells us that when the District Attorneys Association files an amicus curiae brief in support of appellant, appellant is 3.6 times more likely to win than when the District Attorneys Association does not file an amicus brief.
C. Testing the Affected Groups Hypothesis
In order to assess the affected groups hypothesis, I analyzed cases in which there was a disparity of amicus support for one side of the litigation. Although there are several possible measures for analyzing this disparity, (207) I chose to look at cases in which an amicus brief was filed for one side and no amicus briefs filed for the other side. In other words, I studied cases in which one, two, or more than two briefs were filed for appellant, none filed for respondent, and none classified as other briefs (x-0-0 cases), and then studied cases in which there were one, two, or two or more briefs filed for respondent, none for appellant, and none classified as other briefs (0-x-0 cases). If cases with a large disparity of amicus support have higher rates of success as compared to cases with a smaller disparity in amicus support, then it would tend to show evidence of the affected groups hypothesis.
Figure 15 displays the change in the benchmark rate of success for cases involving amicus support for appellants only. Based on the interest group model, one would expect that cases with a larger disparity of
amicus support for one side, in this case appellant, should have a higher rate of success as compared to the benchmark rate, as well as across the categories of disparity analyzed. The findings from Figure 15 are entirely consistent with the affected groups hypothesis. As shown in the figure below, cases involving one appellant brief for appellant, none for respondent, and no other briefs (1-0-0 cases) have an appellant win rate of almost 18% higher than the benchmark rate of success. When two appellant briefs are filed under similar circumstances (2-0-0 cases), appellant's win rate rises to almost 24% above the benchmark rate of success. Moreover, cases involving three amicus briefs present an even higher rate of success as compared to the previous categories. In cases involving two or more amicus briefs filed for appellant, the success rate for appellant rises by almost 30%. (208)
When analyzing cases in which amicus support was in favor of respondent only, I found a somewhat similar pattern. Again, the interest group model predicts that in cases involving larger amicus disparities in support of respondent there should be a corresponding decline in appellant's win rate. As shown in Figure 16, when one amicus brief is filed for respondent, none for appellant, and no other briefs filed (0-1-0 cases), appellant's win rate decreases by approximately 10%. Similarly and as predicted by the interest group model, when the amicus disparity for respondent increases to two filings for respondent, none for appellant, and no other briefs filed (0-2-0 cases), appellant's rate of success decreases by almost another 8%. Notably, for cases involving three briefs for respondent, none for appellant, and no other briefs filed (0-3-0 cases), appellant's win rate falls back closer to the benchmark rate of success. (209)
The above analysis seems to show some support for the affected groups hypothesis, especially when amicus support favors appellant. That is, cases involving amicus support for one side and none for the other have a higher rate of success. As shown above, however, the Attorney General has a large effect on appellant's rate of success when appearing as amicus in support of appellant or respondent. (210) Notably, a corollary to the rate of success of the Attorney General in the Court of Appeals can be found in the federal courts with the Solicitor General. (211) Many commentators have found that the Solicitor General has had unprecedented success both when appearing as amicus curiae or as a party before the U.S. Supreme Court. (212) Therefore, I decided to run the same amicus disparity analysis while filtering out cases involving the Attorney General as amicus curiae or as a party.
As shown below in Figure 17, when controlling for the Attorney General in cases involving amicus disparity in favor of appellant, the rates of success decrease across each category of cases as compared to Figure 15. Still, cases involving a disparity of amicus briefs in support of appellant show a higher rate of success than the benchmark rate. For instance, in cases involving one brief filed in favor of appellant while controlling for the Attorney General, the success rate of appellant rises by 16%. Moreover, when a second brief is filed in favor of appellant, the win rate increases by 20%. (213)
Figure 18 displays a similar result when the Attorney General is filtered out in cases involving amicus disparity in favor of respondent. Although each of the three categories of disparity presented shows the expected decrease in the success rate of appellant when there is a disparity of support in favor of respondent, controlling for the Attorney General has mitigated that effect. For cases involving one amicus brief in favor of respondent, the percentage change in the success rate dropped to 8% when controlling for the Attorney General, down from above 10% in Figure 16 above. Similarly, the success rate dropped to below 13% when filing two amicus briefs in favor of respondent while controlling for the Attorney General as compared to over 18% in the previous analysis. Notably, the success rate of appellant actually decreased more when controlling for the Attorney General in cases involving two or more briefs filed for respondent. Still, it is important to note that based on the sample size of 0-2-0 and 0->2-0 cases it is difficult to draw definitive conclusions from the data. (214)
The results of the regression model underlying Figures 15 to 18 are displayed in Table 6. Based on the effect of the Attorney General in the previous analysis, I also included those variables in this model. Although the Attorney General maintains an effect on the outcome of cases, only one of the disparity variables included in the analysis is significant--respondent files one amicus brief and no other briefs are filed. This could be due to the small sample size provided by several of these variables. It also seems to suggest that the affected groups hypothesis does not necessarily explain how amicus curiae briefs affect the Court of Appeals.
My analysis did not indicate that one theory alone explains the affect of amicus curiae on the court's decision-making process. The results, however, do suggest at least three conclusions: (1) filing an amicus brief influences the decision-making process of the Court of Appeals; (2) the Attorney General as amicus curiae achieves an unprecedented rate of success when controlling for other variables; and (3) as a whole, the results show the most support for the information hypothesis.
First, my results permit me to reject the null hypothesis--that amicus briefs have no affect on the court's decision-making--since cases involving amicus briefs have uniformly yielded different rates of success when compared to the benchmark rate. In fact, as discussed above, the mere filing of amicus brief for appellant makes the odds of an appellant win 2.2 times greater than if no amicus brief was filed for appellant. (215) And throughout my results, I find a consistent pattern in which cases involving amicus briefs affect the benchmark rate of success suggesting support for both information and affected groups hypotheses. Almost uniformly, this affect on the benchmark rate is as expected for the specific hypothesis tested. For instance, cases involving amicus disparity for appellant show an expected rise in appellant's win rate. (216) Similarly, cases involving amicus disparity in favor of respondent show an expected decrease in appellant's win rate, though not a statistically significant effect. (217)
Since the Court of Appeals is not inundated with amicus briefs like the Supreme Court, it may make their presence in litigation that much more forceful. Given that only about 20% of argued cases involve amicus curiae participation for the Court of Appeals, the judges may be more attentive to amicus filings. Thus, the Court of Appeals may lack an amicus fatigue that may arise for the Justices of the Supreme Court who decide individual cases with dozens of amicus filings.
Furthermore, the conclusion that amicus briefs affect the court's decision-making process is largely unsurprising given the steps the court has taken to encourage amicus participation. Clearly, the court finds value in the information provided in amicus curiae briefs, even if it may not be entirely clear how it affects the outcome of cases. Undoubtedly, there are numerous factors which affect how the Court of Appeals, or any court, decides a case. My findings only suggest that amicus briefs are one of the many factors which play a role in the court's final decision.
Second, the results of my analysis indicate that the Attorney General achieves unprecedented success as an amicus filer even when controlling for other variables. My regression models for both the information and affected groups hypotheses showed that the Attorney General, when filing as amicus curiae for appellant or respondent, has a significant impact on the odds of appellant prevailing.
The rate of success for the Attorney General may be due to a variety of factors and, under these circumstances, it is helpful to analogize to the success of the Solicitor General as an amicus filer. Previous studies have consistently found that the Solicitor General achieves significant success as an amicus filer before the Supreme Court.21s There are numerous explanations offered to explain the Solicitor General's success as amicus curiae including that the Justices defer to the views of the Solicitor General for political reasons since the views of the Solicitor General reflect the view of the executive. (219) Another, more plausible, explanation is that the Solicitor General is successful because of a reputation for accurately stating the law, e.g., the information hypothesis. (220)
My results suggest support for the latter explanation, or that the Attorney General is successful as an amicus filer before the Court of Appeals because of an ability to objectively and accurately state the law. The Attorney General is the most seasoned amicus filer, at least over the last twenty years, and it is reasonable to assume that this experience helps the Attorney General formulate cogent legal arguments to which the court is receptive. (221) Moreover, the success of the Attorney General as an amicus filer explains why the process for filing an amicus brief for the Attorney General is more liberal as compared to other prospective amici. (222)
Finally, I tentatively interpret my results as showing support for the information hypothesis. The evidence which leads to this conclusion involves two specific litigants: the Attorney General and the New York State District Attorneys Association. As discussed above, the Attorney General had a significant effect on the probability of appellant prevailing when appearing as an amicus filer for appellant or respondent. Similarly, when the District Attorneys Association appeared as an amicus filer supporting appellant it significantly affected the probability of an appellant victory.
Notably, my results were not always as expected when assessing the influence of the institutional litigants on the benchmark rate of success. I attribute that the other institutional litigants did not display a similar affect on appellant's chances of prevailing to two factors: (1) sample size, and (2) my identification of institutional litigants. First, by identifying the cases in which each institutional litigant filed an amicus brief for appellant or respondent, it inevitably led to sample sizes below thirty. Where sample sizes are small, it is appropriate to exercise caution when interpreting the results. Second, it was difficult to properly identify amicus filers which provided high quality briefs consistent with the information hypothesis. Although I believe that my methodology was the most appropriate approach considering that there have been few studies concerning amicus curiae participation in state courts and due to the small sample size of many amicus filers before the court, I do highlight how this process may limit the conclusions which can be drawn from this study.
In 1989, Judge Kaye noted with dismay that the level of amicus curiae participation in the New York State Court of Appeals was minimal. Even as the U.S. Supreme Court was seeing a significant increase in the level of amicus curiae briefs, the Court of Appeals was struggling to encourage additional amicus curiae submissions. For instance, the Court of Appeals has encouraged amicus filings in court opinions, official documents distributed by the court, and publications by the judges themselves. The first focus of this study was to assess whether anyone was listening, or is the Court of Appeals still friendless today?
By constructing a database including every case which yielded a published opinion from 1988-2007, I was able to assess the level of amicus curiae participation in the Court of Appeals. It is now clear that the words of Judge Kaye have not gone unnoticed--amicus curiae participation in the court is on the rise. In fact, I have found that generally the rate of filing, as well as the rate of citation of amicus curiae in court opinions, has seen a notable increase over the last twenty years.
Still, the more important inquiry is whether these briefs actually affect the decision-making process of the court. In turn, I decided to test my database against several competing hypotheses concerning the utility of amicus curiae briefs. Previous studies of amicus curiae participation have identified three competing theories to explain the usefulness, or lack thereof, of amicus curiae briefs. More specifically, these theories are: (1) that amicus curiae briefs play no role in the decision-making process, (2) that amicus curiae briefs provide relevant information that helps the court decide the case, and (3) that amicus curiae briefs act as a barometer of public opinion so that judges can rule in favor the party which holds the most public sentiment.
I tested these hypotheses by conducting a variety of statistical techniques and found that I could flatly reject that amicus curiae briefs play no role in the decision-making process of the court. In addition, my results show that the Attorney General as an amicus filer achieves an unprecedented rate of success in affecting the outcome of decisions. This suggests that although the Court of Appeals is receptive to amicus curiae briefs generally, it pays particular attention to any submission by the Attorney General. Based on this information, I was able to reasonably conclude that the Court of Appeals finds utility from amicus briefs based on the quality of information the briefs provide.
APPENDIX A: CASES WITH SIX OR MORE AMICUS BRIEFS 1988-2007 CASE SUBJECT MATTER BRIEFS 1 Hernandez v. Robles, 855 N.E.2d 1 (N.Y. Same sex marriage 2006 (constitutional aspects) 24 2 Campaign for Fiscal Equity, Inc. v. State of N.Y., 801 N.E.2d Education (constitutional 326 (N.Y. 2003) aspects) 13 3 Zanghi v. Niagara Frontier Transp. Comm'n, 649 N.E.2d 1167 (N.Y. 1995) Negligence 10 4 Nicholson v. Scoppetta, 820 N.E.2d 840 (N.Y. 2004 Family (parent child) 10 5 In re Westchester County Med. Ctr., 531 N.E.2d 607 (N.Y. Health (right to refuse 1988 treatment) 9 6 Hope v. Perales, 634 N.E.2d 183 (N.Y. Social services 1994 (constitutional aspects) 9 7 Cahill v. Pub. Serv. Comm'n, 556 N.E.2d 133 (N.Y. 1990) First Amendment 8 8 Juarez v. Wavecrest Mgmt. Team, Ltd., 672 N.E.2d 135 (N.Y. 1996 Negligence 8 9 Parker v. Mobil Oil Corp., 857 N.E.2d 1114 (N.Y. 2006) Evidence (scientific) 8 10 Campaign for Fiscal Equity, Inc. v. State of N.Y., 861 N.E.2d Education (constitutional 50 (N.Y. 2006) aspects) 8 11 Braschi v. Stahl Assocs. Co., 543 N.E.2d 49 (N.Y. Landlord tenant (rent 1989 regulation) 7 12 Lloyd v. Grella, 634 N.E.2d 171 (N.Y. Education (discrimination 1994 based on sexual orientation) 7 13 Hamilton v. Beretta U.S.A. Corp., 750 N.E.2d 1055 (N.Y. 2001 Negligence 7 14 Sheppard-Mobley v. King, 830 N.E.2d 301 Negligence (medical (N.Y. 2005) malpractice) 7 15 Courtroom Television Network LLC v. State of N.Y., 833 N.E.2d 1197 (N.Y. Courts (camera coverage of 2005 courtroom) 7 16 Balbuena v. IDR Realty LLC, 845 N.E.2d 1246 (N.Y. Damages (claim by 2006 undocumented alien) 7 17 Desiderio v. Ochs, 791 N.E.2d 941 (N.Y. 2003 Damages (jury award) 7 18 Immuno AG v. Moor- Jankowski, 549 N.E.2d 129 (N.Y. 1989 Libel and slander 6 19 In re Allison D., 572 N.E.2d 27 (N.Y. 1991 Family (parent child) 6 20 Grace Plaza of Great Neck, Inc. v. Elbaum, 623 N.E.2d 513 (N.Y. Health (right to refuse 1993 treatment) 6 21 Rent Stabilization Assoc. of N.Y City, Inc. v. Higgins, 630 N.E.2d 626 (N.Y. Rent regulation 1993 (constitutional aspects) 6 22 Daily Gazette Co. v. City of Schenectady, 710 N.E.2d 1042 Freedom of information (N.Y. 1999) (police records) 6 23 Levin v. Yeshiva Univ., 754 N.E.2d Civil rights (discrimination 1099 (N.Y. 2001) based on sexual orientation) 6 24 Wittlinger v. Wing, Damages (attorneys' fees 786 N.E.2d 1270 under the Equal Access to (N.Y. 2003) Justice Act) 6 25 The Argo Corp. v. Greater N.Y. Mut. Ins. Co., 827 N.E.2d Insurance (duty to 762 (N.Y. 2005) indemnify) 6 26 Catholic Charities of the Diocese of Albany v. Serio, 859 N.E.2d Freedom of religion 459 (N.Y. 2006) (constitutional aspects) 6 27 People v. Taylor, 878 N.E.2d 969 (N.Y. Death penalty (constitutional 2007 aspects) 6
(1) Judith S. Kaye, One Judge's View of "Friends of the Court", N.Y. ST. BAR. J., Apr. 1989, at 8, 13.
(2) ARTHUR KARGER, THE POWERS OF THE NEW YORK COURT OF APPEALS 3-4 (3d ed. 2005).
(3) Kaye, supra note 1, at 9.
(4) Id. at 8.
(5) Id. at 12.
(6) See, e.g., Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265 (1978) (54 amicus briefs filed); Pac. Mut. Life Ins. Co. v. Haslip, 499 U.S. 1 (1991) (31 amicus briefs filed); Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833 (1992) (31 amicus briefs filed).
(7) 492 U.S. 490 (1989).
(8) In Webster, 78 amicus briefs were filed, whereas only forty-nine were filed for the entire 1989 term of the New York Court of Appeals. See id.; infra Part II.B.1.
(9) Joseph D. Kearney & Thomas W. Merrill, The Influence of Amicus Curiae Briefs on the Supreme Court, 148 U. PA. L. REV. 743, 744 (2000); see also John Harrington, Amici Curiae in the Federal Courts of Appeals: How Friendly Are They?, 55 CASE W. RES. L. REV. 667, 680 (2005) (finding that the number of amicus curiae briefs at the federal courts of appeals increased by over 14% from 1992 through 2002).
(10) Kearny & Merrill, supra note 9, at 753.
(11) See Kaye, supra note 1, at 13.
(12) N.Y. COMP. CODES R. & REGS. tit. 22, [section][section] 500.12(e), 500.23(a)(1) (2007).
(13) Kaye, supra note 1, at 9 ("The subject matter of the newly filed cases may suggest appropriate motions and participation which the Court welcomes.").
(14) STUART M. COHEN, ANNUAL REPORT OF THE CLERK OF THE COURT TO THE JUDGES OF THE COURT OF APPEALS OF THE STATE OF NEW YORK 11 (2000), http://www.nycourts.gov/ctapps/news/annrpt/annrpt00.pdf.
(15) Ryan v. Commodity Futures Trading Comm'n, 125 F.3d 1062, 1063 (7th Cir. 1997).
(16) Jaffee v. Redmond, 518 U.S. 1, 35-36 (1996) (Scalia, J., dissenting).
(17) Neonatology Assocs., P.A. v. Comm'r of Internal Revenue, 293 F.3d 128, 131 (3d Cir. 2002).
(18) Richard A. Posner, Foreword: A Political Court, 119 HARV. L. REV. 31, 32 (2005).
(19) See Kearney & Merrill, supra note 9, at 748.
(20) See Andrew P. Morriss, Private Amici Curiae and the Supreme Court's 1997-1998 Term Employment Law Jurisprudence, 7 WM. & MARY BILL RTS. J. 823 (1999); Donald R. Songer & Reginald S. Sheehan, Interest Group Success in the Courts: Amicus Participation in the
Supreme Court, 46 POL. RES. Q. 339 (1993); Gregg Ivers & Karen O'Connor, Friends as Foes: The Amicus Curiae Participation and Effectiveness of the American Civil Liberties Union and Americans for Effective Law Enforcement in Criminal Cases, 1969-1982, 9 LAW & POL'Y 161, 162 (1987).
(21) See Kearney & Merrill, supra note 9, at 781.
(24) See Linda Sandstrom Simard, An Empirical Study of Amici Curiae in Federal Court: A Fine Balance of Access, Efficiency, and Adversarialism, 27 REV. LITIG. 669, 681 (2008).
(25) Id. at 682.
(26) Kearny & Merrill, supra note 9, at 748.
(27) See Kaye, supra note 1, at 12 13.
(28) Simard, supra note 24, at 681.
(30) Kearny & Merrill, supra note 9, at 783.
(31) See id. at 744.
(32) Notably, the definition of amicus curiae has changed little over time. See Samuel Krislov, The Amicus Curiae Brief: From Friendship to Advocacy, 72 YALE L.J. 694, 694-95 (1963) (comparing modern dictionary definitions of amicus curiae which define the term as "friend of the court," to Holthouse's Law Dictionary, a vintage source, with a similar definition).
(33) Id. at 716.
(34) Id. at 694.
(36) Id. at 695.
(38) 77 Eng. Rep. 481, 516, 8 Coke 1a, 29a, 29b (Ch. 1606).
(39) Simard, supra note 24, at 676 n.30.
(40) Krislov, supra note 32, at 697; see also Campbell v. Swasey, 12 Ind. 70, 72 (1859) (noting that amicus curiae "acts for no one, but simply seeks to give information to the Court").
(41) Krislov, supra note 32, at 697.
(42) The first recorded amicus curiae brief in the United States was in Green v. Biddle, 21 U.S. (8 Wheat.) 1 (1823), filed by the state of Kentucky. Id. at 700.
(43) See id. at 697-98.
(44) See Simard, supra note 24, at 676-77.
(45) Krislov, supra note 32, at 703.
(47) Simard, supra note 24, at 677.
(48) Krislov, supra note 32, at 697.
(49) See id. at 705 (noting that Charles J. Bonaparte, as Attorney General, was the innovator of the use of the amicus curiae brief by the government as a form of advocacy to implement, or at least influence, major social changes and broad public policies).
(50) Fowler V. Harper & Edwin D. Etherington, Lobbyists Before the Court, 101 U. PA. L. REV. 1172, 1172 (1953) (noting that amici curiae "are repetitious at best and emotional explosions at worst").
(51) Id. at 1173.
(53) Simard, supra note 24, at 679.
(54) Kearney & Merrill, supra note 9, at 752.
(55) See People v. Gibbs, 38 N.W. 257, 258 (1888) (refusing to grant permission for an amicus filer since they represented the position of one of the parties and were not a true friend of the court). But see People ex rel. Attorney-General v. Sec. Life Ins. Co., 71 N.Y. 222, 225 (1877) (noting that a non-party to litigation who wants to advocate for a position before the court should appear as amicus curiae).
(56) See, e.g., City of Grand Rapids v. Consumers' Power Co., 185 N.W. 852, 854 (Mich. 1921) (noting that amicus curiae relief is almost always granted for cases involving important public issues).
(57) Lee Epstein, Exploring the Participation of Organized Interests in State Court Litigation, 47 POL. RES. Q. 335, 348 (1994).
(58) Id. at 338. "State courts of last resort" refers to the fifty-two highest appellate state courts. Notably, Texas and Oklahoma have two courts of last resort, one with criminal and one with civil jurisdiction.
(59) Id. at 339.
(60) Id. at 341.
(61) Sarah F. Corbally et al., Filing of Amicus Curiae Briefs in State Courts of Last Resort: 1960-2000, 25 JUST. SYS. J. 39, 44 (2004).
(62) Paul Brace & Kellie Sims Butler, New Perspectives for the Comparative Study of the Judiciary: The State Supreme Court Project, 22 JUST. SYS. J. 243, 253 (2001).
(67) Corbally et al., supra note 61, at 46 (defining a high level of amicus curiae participation as over twenty cases, which translates from their twenty percent sample to one hundred plus cases with amicus curiae participation during the three year sample period).
(69) Kaye, supra note 1, at 12.
(70) See, e.g., Niesig v. Team I, 558 N.E.2d 1030, 1036 (N.Y. 1990) (noting the important contribution made by various amici).
(71) STUART M. COHEN, 1999 ANNUAL REPORT OF THE CLERK OF THE COURT TO THE JUDGES OF THE COURT OF APPEALS OF THE STATE OF NEW YORK 11 (1999), http://www.nycourts.gov/ctapps/news/annrpt/annrpt99.pdf.
(72) Kaye, supra note 1, at 13.
(73) N.Y. COMP. CODES R. & REGS. tit. 22, [section] 500.23 (2007).
(74) N.Y. COMP. CODES R. & REGS. tit. 22, [section] 500.23(a)(4) (2007).
(75) N.Y. COMP. CODES R. & REGS. tit. 22, [section][section] 500.23 (b), 500.12(e), 500.11(h) (2007).
(76) Corbally et al., supra note 61, at 48-50.
(77) See COHEN, supra note 71, at 11. In 1999, there were 87 motions for amicus curiae relief of which 69 were granted. For the reasons discussed below, the number of amicus curiae briefs identified by my study was only 65. See infra note 194.
(78) STUART M. COHEN, 2004 ANNUAL REPORT OF THE CLERK OF THE COURT TO THE JUDGES OF THE COURT OF APPEALS OF THE STATE OF NEW YORK 7 (2004), http://www.nycourts.gov/ctapps/news/annrpt/AnnRpt2004.pdf. In 2004, the court granted 88 of 93 motions for amicus curiae relief.
(79) COHEN, supra note 71, at 11.
(80) Id. ("Given that the Court hears the majority of appeals by its own permission, and that the questions presented are generally novel and of statewide importance, the Bar should take special note that the Court encourages appropriate requests for permission to file amicus curiae submissions."). This quote, or similar language, appears in every annual report by the clerk of the court from 1998-2007.
(81) See, e.g., Notice to the Bar, Amicus Curiae Participation, Stuart M. Cohen, Clerk of the Court, New York Court of Appeals (Sept. 11, 2002), http://www.nycourts.gov/ctapps/news/nottobar/noticesep11.pdf (alerting the Bar that the court invites amicus curiae submissions for In re Allen, 794 N.E.2d 18 (N.Y. 2003) from those who are "qualified and interested").
(82) COHEN, supra note 14, at 11.
(84) 740 N.E.2d 227 (N.Y. 2000).
(85) STUART M. COHEN, 2001 ANNUAL REPORT OF THE CLERK OF THE COURT TO THE JUDGES OF THE COURT OF APPEALS OF THE STATE OF NEW YORK 9 (2001), http://www.nycourts.gov/ctapps/news/annrpt/annrpt01.pdf.
(86) Id. at 9-10.
(87) For a full explanation of my research methodology, see infra Part V.
(88) The average number of argued cases which yield a published opinion was approximately 125 from 1988-2007.
(89) Kearney & Merrill, supra note 9, at 754.
(90) The mean, median, and modal number of amicus briefs per case during each five-year period and the mean, median, and modal number of briefs in each case with at least one amicus brief (noted "w/1+") in each five-year period are as follows.
Mean Median Mode Mean w/+1 Median w/l+ Mode w/1+ 1988-1992 0.4 1.74 0 1 0 1 1993-1997 0.44 1.97 0 1 0 1 1998-2002 0.46 1.72 0 1 0 1 2003-2007 0.69 2.39 0 1 0 1 All 0.49 1.97 0 1 0 1
(91) Kearney & Merrill, supra note 9, at 754-55.
(92) Id. at 755 n.29.
(93) Cases with five or more briefs filed represented the 95th percentile of amicus briefs per case. The number of cases with five or more, ten or more, and twenty or more briefs per five-year period are as follows.
[greater than or [greater than or [greater than or equal to 5 equal to 10 equal to 15 1988-1992 10 0 0 1993-1997 12 1 0 1998-2002 3 0 0 2003-2007 19 3 1 All 44 4 1
(94) The twenty-seven cases with six or more briefs and the subject matter of each case is set forth in Appendix A.
(95) There were eight cases with six or more briefs with constitutional law as the subject matter (29.6%).
(96) 855 N.E.2d 1 (N.Y. 2006).
(97) Id. at 5.
(98) 801 N.E.2d 326 (N.Y. 2003).
(99) Id. at 328.
(100) 820 N.E.2d 840 (N.Y. 2004).
(101) 649 N.E.2d 1167 (N.Y. 1995).
(102) See infra Part V.B. I expect that repeat amicus filers have experience that should help them formulate arguments which will be receptive to the court.
(103) See infra notes 136-38 and accompanying text.
(104) See Office of the Attorney General of the State of New York, http://www.oag.state.ny.us/our_office.html (last visited Apr. 17, 2009) ("[T]he Attorney General serves as the guardian of the legal rights of the citizens of New York....").
(105) See infra notes 137-38 and accompanying text.
(106) See infra Part V.B.
(107) See, e.g., Kearney & Merrill, supra note 9, at 757 ("The only publicly visible manifestation of the impact of amici is the frequency with which their briefs are cited or quoted in the opinions of the Justices.").
(108) Since reviewing every case for references to amicus briefs would have been overly time consuming, I chose to perform a Westlaw search. This approach is consistent with previous studies which have explored the rate of citation or quotation of amicus briefs. See, e.g., Kearney & Merrill, supra note 9, at 844-46 (summarizing the data-gathering methodology in Appendix B). Using the New York State Cases Database (NY-CS), I searched for any reference to "amicus," "amici," "friend(s) of the court," & "friend(s) to the court." My search syntax was as follows CO(HIGH) & OP("AMICUS" "AMICI" "FRIEND! OF THE COURT" "FRIEND! TO THE COURT") & da(aft 1958 & bef 1960) % (MOTION /S AMICUS AMICI). Any references to "amicus" or "amici" within the same sentence as "motion(s)" were excluded from the search since the database includes motions to file amicus briefs which made the original result difficult to navigate.
(109) A case was coded as citing an amicus brief if there was any reference to the brief in the opinion--it did not necessarily require a full citation to the brief.
(110) This chart represents the percentage of cases in which at least one amicus brief was filed, and where the court's opinion cites to an amicus brief.
(111) There were only a total of five briefs cited or quoted in 1959, 1969, and 1979 combined. Three were cited in 1959, one in 1969, and one in 1979.
(112) At least one amicus brief was cited in each of ninety-nine cases from 1988-2007.
(113) The data underlying Figure 6 is provided below:
Cases with Cases with No Briefs Cited Brief Cited 1988-1992 19 129 1993-1997 15 151 1998-2002 30 119 2003-2007 35 132 All 99 531
(114) See supra Part II.B.1.
(115) See supra Figure 1.
(116) 558 N.E.2d 1030, 1036 (N.Y. 1990).
(118) See Campaign for Fiscal Equity, Inc. v. State, 655 N.E.2d 661, 672 (N.Y. 1995) (Levine, J., concurring) (relying on amicus curiae from Bd. of Educ. v. Nyquist, 439 N.E.2d 359 (N.Y. 1982), to provide a historical perspective of state funded education in New York).
(119) Kearney & Merrill, supra note 9, at 767-68.
(120) Nathan Hakman, Lobbying the Supreme Court--An Appraisal of "Political Science Folklore," 35 FORDHAM L. REV. 15 (1966-67).
(121) Id. at 50.
(122) Id. at 16, 18.
(123) See, e.g., Morriss, supra note 20, at 909-10 (concluding that the participation of interest groups in three Title VII cases is an example of how amicus curiae briefs can be more burdensome than helpful to the Supreme Court); Stephen Calkins, Supreme Court Antitrust 1991-92: The Revenge of the Amici, 61 ANTITRUST L. J. 269, 269 (1993) (noting that amicus curiae participation significantly influenced the Supreme Court's opinions in three high profile anti-trust cases).
(124) See Paul Chen, The Informational Role of Amici Curiae Briefs in Gonzales v. Raich, 31 S. ILL. U. L.J. 217, 218 (2007) (arguing that information provided in amicus briefs in Gonzales v. Raich had an impact on the outcome of the decision of the Supreme Court).
(125) See TIMOTHY J. O'NEILL, BAKKE & THE POLITICS OF EQUALITY: FRIENDS AND FOES IN THE CLASSROOM OF Litigation 165-67 (1985) (reviewing the influence of amicus briefs on the Supreme Court's ruling in Regents of the University of California v. Bakke, 438 U.S. 265 (1978)).
(126) See KAREN O'CONNOR, WOMEN'S ORGANIZATIONS' USE OF THE COURTS 100-15 (1980) (reviewing various women's groups that file amicus briefs); Mary-Christine Sungaila, Effective Amicus Practice Before the United States Supreme Court: A Case Study, 8 S. CAL. Rev. L. & WOMEN'S. STUD. 187, 190-96 (1999) (exploring the influence of an amicus curiae brief filed by the National Organization for Women Legal Defense and Education Fund in Davis v. Monroe County Board of Education, 526 U.S. 629 (1999)).
(127) See, e.g., STEPHEN L. WASBY, RACE RELATIONS IN AN AGE OF COMPLEXITY 219-26 (1995) (exploring the possible effects of amicus curiae briefs in several civil rights cases); H.W. PERRY, JR., DECIDING TO DECIDE: AGENDA SETTING IN THE UNITED STATES SUPREME COURT 135-37 (1991) (exploring the effects of amicus curiae at the certiorari stage); John Howard, Retaliation, Reinstatement, and Friends of the Court: Amicus Participation in Brock v. Roadway Express, Inc., 31 HOW. L.J. 241, 270 (1988) (reviewing the influence of amicus curiae in a procedural due process case and noting that amici have become "powerful participants" in Supreme Court litigation).
(128) 545 U.S. 1 (2005).
(129) Chen, supra note 124, at 217, 221.
(130) See Donald R. Songer & Reginald S. Sheehan, Interest Group Success in the Courts: Amicus Participation in the Supreme Court, 46 POL. RES. Q. 339, 345 (1993).
(131) Id. at 345-46.
(132) See Kevin T. McGuire, Repeat Players in the Supreme Court: The Role of Experienced Lawyers in Litigation Success, 57 J. POL. 187, 193 (1995); Kevin T. McGuire, Obscenity, Libertarian Values, and Decision Making in the Supreme Court, 18 AM. POL. Q. 47, 48 (1990).
(133) See Ivers & O'Connor, supra note 20, at 169.
(134) See REBECCA MAE SALOKAR, THE SOLICITOR GENERAL: THE POLITICS OF LAW 145-50 (1992) (finding that the Solicitor General supported the winning side in 72% of cases in which he filed an amicus brief); Jeffrey A. Segal, Amicus Curiae Briefs by the Solicitor General During the Warren and Burger Courts: A Research Note, 41 W. POL. Q. 135, 136 (1988) (finding that Solicitor General amicus filings have supported the winning side approximately 75% of the time).
(135) Segal, supra note 134, at 142; see also Jeffrey A. Segal & Cheryl D. Reedy, The Supreme Court and Sex Discrimination: The Role of the Solicitor General, 41 W. POL. Q. 553, 563 (1988) (finding that when the Solicitor General files an amicus brief in a sex discrimination case it is very influential on the Court's decision while controlling for other variables).
(136) Simard, supra note 24, at 675.
(137) Id. at 691-93.
(138) Id. at 697.
(139) See Kearney & Merrill, supra note 9, at 749.
(140) Id. at 774-75.
(141) Id. at 797-98.
(142) Id. at 817-18; see also Paul M. Collins, Jr., Friends of the Court: Examining the Influence of Amicus Curiae Participation in U.S. Supreme Court Litigation, 38 LAW & SOC'Y REV. 807, 812-13 (2004) (studying amicus participation in the Supreme Court from 1953 to 1985, and concluding that disparities in amicus support did not lead to higher rates of success).
(143) Kearney & Merrill, supra note 9, at 820-23.
(144) Notably, one study did review the effect of amicus curiae briefs on three state courts of last resort--Georgia, South Carolina, and North Carolina--and found that amici support was significantly related to a party's chances of success. Donald R. Songer & Ashlyn Kuersten. The Success of Amici in State Supreme Courts, 48 POL. RES. Q. 31, 39 (1995).
(145) See JEFFREY A. SEGAL ET AL., THE SUPREME COURT IN THE AMERICAN LEGAL SYSTEM 38 (2005) (noting that nothing precludes the application of the attitudinal model to state appellate courts).
(146) Frank B. Cross, Political Science and the New Legal Realism: A Case of Unfortunate Interdisciplinary Ignorance, 92 NW. U. L. REV. 251, 265 (1997).
(148) JEFFREY A. SEGAL & HAROLD J. SPAETH, THE SUPREME COURT AND THE ATTITUDINAL MODEL REVISITED 86 (2002).
(149) Id. at 89.
(150) Id. at 324.
(151) SEGAL ET AL., supra note 145, at 39.
(157) Kearney & Merrill, supra note 9, at 781.
(158) Id. at 782; see also SEGAL & SPAETH, supra note 148, at 98 ("When an actor considers the ramifications of his or her actions in a game-theoretic situation and makes the best response to that situation given available information, that actor may be said to have behaved strategically."); LEE EPSTEIN & JACK KNIGHT, THE CHOICES JUSTICES MAKE 10 (1998) ("[J]ustices are strategic actors who realize that their ability to achieve their goals depends on a consideration of the preferences of other actors, the choices they expect others to make, and the institutional context in which they act.").
(159) SEGAL & SPAETH, supra note 148, at 98-99.
(160) Kearney & Merrill, supra note 9, at 782 (arguing that the strategic theory model would support the hypothesis that justices will respond more favorably to briefs submitted by the Solicitor General).
(161) Id. at 782.
(162) Cross, supra note 146, at 252; see also Osborn v. Bank of the U.S., 22 U.S. (9 Wheat.) 738, 866 (1824) (Marshall, C.J.) ("Judicial power is never exercised for the purpose of giving effect to the will of the Judge; always for the purpose of giving effect to the will of the Legislature; or, in other words, to the will of the law.").
(163) Kearney & Merrill, supra note 9, at 776.
(164) See SEGAL ET AL., supra note 145, at 22.
(168) RONALD DWORKIN, TAKING RIGHTS SERIOUSLY 38 (1977).
(169) Kearney & Merrill, supra note 9, at 777.
(170) DWORKIN, supra note 168, at 118.
(171) Simard, supra note 24, at 682.
(172) N.Y. COMP. CODES R. & REGS. tit. 22, [section] 500.23(a)(4) (2005).
(174) See Kaye, supra note 1, at 13.
(175) See Simard, supra note 24, at 682 (suggesting that the "information theory," which "suggests that amicus briefs are effective ... because they supplement the arguments of the parties by providing information not found in the parties' briefs," helps to explain the effect of amicus curiae on courts).
(176) Kearney & Merrill, supra note 9, at 776, 778-79.
(177) See infra Part V.B.
(178) Kearney & Merrill, supra note 9, at 782-83.
(179) Id. at 783.
(180) See ALEXANDER M. BICKEL, THE LEAST DANGEROUS BRANCH: THE SUPREME COURT AT THE BAR OF POLITICS 17-20 (1986).
(181) Collins, supra note 142, at 812-13.
(183) Id. at 813.
(185) Kearney & Merrill, supra note 9, at 785.
(186) See id. at 787 ("The greater the disparity of amicus support for one side, the more likely the Court will rule for that side.").
(187) See id.
(188) Cases which yielded only a memorandum opinion are excluded from the dataset. These opinions are in a separate portion of the New York Reports Volumes. This does present a shortcoming of the study since some of these cases include amicus filings. I made the decision not to record these cases since the New York Reports does not provide descriptive information for these filings, namely, who the filer supports. Since I did not include amicus filings from Memorandum Opinions, the number of amicus curiae, as well as the number of cases decided by the court, varies slightly from the Annual Reports provided by the New York State Court of Appeals.
(189) Although most cases were adjudicated "affirmed," "reversed," or "modified," classifying the outcomes of decisions was not an exact science since the court's opinions did not always fit neatly into one of the three categories. For example, certified questions presented a particularly difficult situation since it required reading the opinion and assessing which side prevailed. See also Kearney & Merrill, supra note 9, at 788 n.148 (noting that "Supreme Court decisions do not come neatly packaged as victories or defeats for the parties").
(190) Similarly, classifying cases under a particular subject matter is problematic due to the variety of issues which can be included in one particular case. I relied on the headnotes provided by the New York Reports Volumes as a guide and also briefly read the information provided in Points of Counsel. My goal was to classify cases broadly, and I acknowledge that many cases could be classified in multiple categories.
(191) Unfortunately, New York Reports does not classify briefs as being for, or against, a particular party. As a result, I had to briefly review the description of each brief and assign them to appellant or respondent.
(192) New York Reports lists the organization who submits the brief as well as the author of the brief.
(193) See Kearney & Merrill, supra note 9, at 789; Songer & Sheehan, supra note 20, at 345 (comparing success rates of parties supported by amici with success rates of parties not supported by amici).
(194) Notably, these benchmark rates of success directly contrast with similar analyses on the Supreme Court. See Segal, supra note 134, at 140 (noting that the typical success rate for respondents in the Supreme Court is 33%). Studies have found that the Supreme Court is more likely to rule in favor of the petitioner rather than respondents, presumably because part of the reason the Court decided to grant certiorari was due to a belief by at least four of the Justices that the case was incorrectly decided below. See Virginia C. Armstrong & Charles A. Johnson, Certiorari Decisions by the Warren and Burger Courts." Is Cue Theory Time Bound?, 15 POLITY 141, 150 (1982).
(195) See supra Part IV.A.
(196) In other words, in Figure 8, I computed the number of amicus briefs supporting appellant whose appellants win divided by the total number of amicus briefs supporting appellant. In Figure 9, I computed the number of amicus briefs supporting respondent whose appellants win, divided by the total number of amicus briefs supporting respondent.
(197) See Kearney & Merrill, supra note 9, at 791 (noting that "using separate benchmarks for filers supporting petitioners and filers supporting respondents would have made the exposition of results more complicated and potentially confusing").
(198) The data underlying Figure 8 is provided below:
All years 1988-1992 1993-1997 1998-2002 Number of appellant amicus briefs, appellant wins 416 96 113 101 Number of appellant amicus briefs, appellant loses 318 63 80 53 Number of appellant amicus briefs, mixed result 22 1 8 1 Total number of appellant amicus briefs 756 160 201 155
(199) The data underlying Figure 9 is provided below:
All years 1988-1992 1993-1997 1998-2002 Number of respondent amicus briefs, appellant wins 142 29 39 38 Number of respondent amicus briefs, appellant loses 333 64 85 64 Number of respondent amicus briefs, mixed results 4 2 1 1 Total number of respondent amicus briefs 479 95 125 103
(200) See supra Part IV.B.
(201) See supra Part II.B.2
(202) The data underlying Figure 10 is provided below:
ATTORNEY GENERAL AS AMICUS CURIAE Attorney General Filing for Appellant AG AG A-Win A-Loss Benchmark Percent Count Rate Rate A-win Rate Change 40 82.9% 17.1% 36.5% 46.48% Attorney General Filing for Respondent AG AG A-Win A-Loss Benchmark Percent Count Rate Rate A-win Rate Change 47 110.6% 189.4% 36.5% -52.91%
(203) The data underlying Figure 11 is provided below:
CITY OF NEW YORK AS AMICUS CURIAE City of New York Filing for Appellant City City A-Win A-Loss Benchmark Percent Count Rate Rate A-win Rate Change 33 57.6% 36.4% 36.5% 21.13% City of New York Filing for Respondent City City A-Win A-Loss Benchmark Percent Count Rate Rate A-win Rate Change 13 38.5% 61.5% 36.5% 2.01%
(204) The data underlying Figure 12 is provided below:
NYS DISTRICT ATTORNEYS ASSOCIATION AS AMICUS CURIAE District Attorneys Association Filing for Appellant DA DA A-Win A-Loss Benchmark Percent Count Rate Rate A-win Rate Change 16 75.0% 25.0% 36.5% 38.55% District Attorneys Association Filing for Respondent DA DA A-Win A-Loss Benchmark Percent Count Rate Rate A-win Rate Change 19 21.1% 78.9% 36.5% -15.40%
(205) The data underlying Figure 13 is provided below:
NYS TRIAL LAWYERS ASSOCIATION AS AMICUS CURIAE NYS Trial Lawyers Filing for Appellant Lawy. Lawy. A-Win A-Loss Benchmark Percent Count Rate Rate A-win Rate Change 14 35.7% 64.3% 36.5% -0.74% NYS Trial Lawyers Filing for Respondent Lawy. Lawy. A-Win A-Loss Benchmark Percent Count Rate Rate A-win Rate Change 17 52.9% 47.1% 36.5% 16.49%
(206) The data underlying Figure 14 is provided below:
LEGAL AID SOCIETY AS AMICUS CURIAE Legal Aid Society Filing for Appellant LAS LAS A-Win A-Loss Benchmark Percent Count Rate Rate A-win Rate Change 21 47.6% 52.4% 36.5% 11.17% Legal Aid Society Filing for Respondent LSA LAS A-Win A-Loss Benchmark Percent Count Rate Rate A-win Rate Change 9 55.6% 44.4% 36.5% 19.1%
(207) See Kearney & Merrill, supra note 9, at 794 (discussing other methods for analyzing disparities in amicus support such the difference between the total number of amicus briefs supporting one side and the number of amicus briefs supporting the other side, or comparing the ratio of the number of briefs for one side relative to the number of briefs filed for the opposing side).
(208) The data underlying Figure 15 is provided below:
1-0-0 cases 2-0-0 cases >2-0-0 cases Number of cases 221 50 35
(209) The data underlying Figure 16 is provided below:
0-1-0 cases 0-2-0 cases 0->2-0 cases Number of cases 138 22 15
(210) See supra Part IV.B.
(211) See supra notes 134-35 and accompanying text.
(212) Kevin T. McGuire, Explaining Executive Success in the U.S. Supreme Court, 51 POL. RES. Q. 505, 506-07 (1998) [hereinafter McGuire, Executive Success].
(213) The data underlying Figure 17 is provided below:
1-0-0 cases 2-0-0 cases >2-0-0 cases Number of cases 204 46 28
(214) The data underlying Figure 18 is displayed below:
0-1-0 cases 0-2-0 cases 0->2-0 cases Number of cases 116 17 10
(215) See supra Part V.A.
(216) See supra Part V.C. Figure 15.
(217) See supra Part V.C. Figure 16.
(218) See supra notes 134-35 and accompanying text.
(219) Kearney & Merrill, supra note 9, at 818.
(220) See McGuire, Executive Success, supra note 212, at 522.
(221) See supra Part II.B.2 Table 1.
(222) See supra Part II.A.
Matthew Laroche, Without the unwavering support of my family, this paper would not have been possible. Specifically, I would like to thank my mother whose strength and support has guided me to success. I am grateful to my faculty advisor Professor Raymond Brescia for his insights, assistance, and encouragement over the last year. I am also indebted to Jim Gilmer and KiDeuk Kim for their invaluable contributions to the statistical analyses in this paper. Finally, I must thank my mentor, Professor James Acker, who gave me the idea to pursue this topic.
Table 1: Identity of Top Amicus Filers (1988-2007) % Total Amicus Amicus Briefs Briefs Amicus Author Filed Filed 1 New York State Attorney General 87 7.0% 2 City of New York 47 3.8% 3 New York State District Attorneys Association 37 3.0% 4 New York State Trial Lawyers Association 35 2.8% 5 Legal Aid Society 30 2.4% 6 Association of the Bar of the City of NY 28 2.3% 7 New York State School Boards Association 21 1.7% 8 NYS Conference of Mayors Municipal Officials 20 1.6% 9 AFL CIO 15 1.2% 10 New York Civil Liberties Union 15 1.2% Table 2: Subject Matter of Cases with at Least One Amicus Brief Number % of Subject Matter of Cases Total 1 Criminal 91 14.4% 2 Negligence 47 7.5% 3 Insurance 42 6.7% 4 Constitutional Law 31 4.9% 5 Education 31 4.9% 6 Municipal Corporation 28 4.4% 7 Civil Service 26 4.1% 8 Contract 24 3.8% 9 Health 21 3.3% 10 Tax 20 3.2% Table 3: Subject Matter of Cases with the Most Amicus Briefs Number of % of total amicus amicus Subject matter briefs briefs filed 1 Criminal 132 10.6% 2 Negligence 104 8.4% 3 Insurance 83 6.7% 4 Constitutional Law 79 6.4% 5 Education 71 5.7% 6 Civil Service 52 4.2% 7 Municipal Corporation 50 4.0% 8 Health 49 4.0% 9 Contract 38 3.1% 10 Landlord Tenant 38 3.1% Table 4: Coefficients, Standard Errors, and Odds Ratios from Logistic Regression for Null Hypothesis Independent Variables B S.E. Significance Exp(b) Amicus Filed .805 .111 .000 ** 2.237 for Appellant Amicus Filed .567 .136 .000 ** .567 for Respondent Constant .554 .047 .000 .575 Odds Ratio (95% Conf. Int.) Independent Variables Lower Upper Amicus Filed 1.800 2.779 for Appellant Amicus Filed .435 .740 for Respondent Constant -2 Log Likelihood = 3287.342 Nagelkerke R Square = .032 * p < .05, ** p <.01 Table 5: Coefficients, Standard Errors, and Odds Ratios from Logistic Regression for Information Theory Independent Variables B S.E. Significance Exp(b) AG as amicus for 1.821 .426 .000 ** 6.175 appellant AG as amicus for -1.830 .483 .000 ** .160 respondent City of New York as .482 .370 .193 l.620 amicus for appellant City of New York as -.283 .605 .639 .753 amicus for respondent District Attorneys l.282 .594 .031 * 3.603 Association as amicus for appellant District Attorneys -1.019 .575 .076 .361 Association as amicus for appellant Trial Lawyers -.242 .611 .692 .785 Association as amicus for appellant Trial Lawyers .389 .512 .447 1.476 Association as amicus for respondent Legal Aid Society as .237 .454 .602 1.267 amicus for appellant Legal Aid Society as .187 703.000 .790 1.206 amicus for respondent Constant -.275 .097 .005 .760 Odds Ratio (95% Conf. Int.) Independent Variables Lower Upper AG as amicus for 2.679 14.233 appellant AG as amicus for .062 .414 respondent City of New York as .784 3.347 amicus for appellant City of New York as .230 2.464 amicus for respondent District Attorneys 1.125 11.542 Association as amicus for appellant District Attorneys .117 1.114 Association as amicus for appellant Trial Lawyers .237 2.600 Association as amicus for appellant Trial Lawyers .542 4.023 Association as amicus for respondent Legal Aid Society as .520 3.085 amicus for appellant Legal Aid Society as .304 4.784 amicus for respondent Constant -2 Log Likelihood = 801.139 Nagelkerke R Square = .129 * p <.05. ** p <.01 Table 6: Coefficients, Standard Errors, and Odds Ratios from Logistic Regression for Affected Groups Theory Independent Standard Variables B Error Significance Exp(b) AG as amicus 1.863 .454 .000 ** 6.445 supporting appellant AG as amicus -1.845 .507 .000 ** .158 supporting respondent 1-0-0 cases; AG .381 .231 .099 1.464 filtered out as amicus 2-0-0 cases; AG .545 .350 .119 1.725 filtered out as amicus >2-0-0 cases; AG .571 .424 .178 1.769 filtered out as amicus 0-1-0 cases; AG -.639 .276 .020 * .528 filtered out as amicus 0-2-0 cases; AG -.896 .601 .136 .408 filtered out as amicus 0->2-0 cases; AG -1.103 .812 .174 .332 filtered out as amicus Constant -.283 .184 .123 .754 Odds Ratio (95% Conf. Int.) Independent Variables Lower Upper AG as amicus 2.648 15.688 supporting appellant AG as amicus .058 .427 supporting respondent 1-0-0 cases; AG .931 2.302 filtered out as amicus 2-0-0 cases; AG .869 3.422 filtered out as amicus >2-0-0 cases; AG .771 4.060 filtered out as amicus 0-1-0 cases; AG .307 .906 filtered out as amicus 0-2-0 cases; AG .126 1.325 filtered out as amicus 0->2-0 cases; AG .068 1.628 filtered out as amicus Constant -2 Log Likelihood = 785.314 Nagelkerke R Square = .159 * p < .05, ** p <.01 Figure 1: Amicus Briefs Filed per Year (1988-2007) Year Number of Amicus Briefs Filed 1988 40 1989 49 1990 56 1991 63 1992 49 1993 58 1994 74 1995 73 1996 56 1997 66 1998 43 1999 64 2000 35 2001 61 2002 54 2003 82 2004 60 2005 79 2006 99 2007 79 Note: Table made from line graph. Figure 2: Amicus Briefs Filed per Year (1959, 1969, 1979) Year Number of Briefs Filed 1959 34 1969 12 1979 25 Figure 3: Percentage of Cases with at Least One Amicus Brief Filed (1988-2007) % of cases 1988-1992 23.19% 1993-1997 22.58% 1998-2002 26.61% 1998-2003 28.38% Figure 4: Percentage of Cases with at Least One Amicus Brief Filed (1959, 1969, 1979) % of cases 1959 12.96% 1969 6.18% 1979 9.04% Figure 5: Amicus Briefs Filed Per Year Per Each Attorney General (1988-2007) Briefs per year Abrams (1988-1993) 2.50 Koppell (1994) 4.00 Vacco (1995-1998) 3.00 Spitzer (1999-2006) 5.44 Cuomo (2007) 8.00 Figure 6: Percentage of Cases Citing an Amicus Curiae Brief (1988-2007) % citation 1988-1992 12.8% 1993-1997 9.0% 1998-2002 20.1% 2003-2007 21.0% All 15.7% Figure 7: Benchmark Rates of Success for All Cases Without an Amicus Brief Filed (1988-2007) A-win rate A-loss rate 1988-1992 42.7% 51.8% 1993-1997 36.4% 58.5% 1998-2002 35.0% 54.0% 2003-2007 30.5% 59.8% All 36.5% 56.1% Figure 8: Percentage Change in Appellant Win Rate for Amici Supporting Appellant Relative to the Benchmark Rates of Success % of change 1988-1992 17.32% 1993-1997 19.79% 1998-2002 30.20% 2003-2007 13.68% All 12.34% Figure 9: Percentage Change in Appellant Win Rate for Amici Supporting Respondent Relative to the Benchmark Rates of Success % of change 1988-1992 -12.16% 1993-1997 -5.23% 1998-2002 1.93% 2003-2007 -7.41% All -6.81% Figure 10: Percent Change in Appellant Win Rate for Cases Where Attorney General Appears as Amicus % of change AG as amicus supporting appellant 46.48% AG as amicus supporting respondent -52.91% Figure 11: Percent Change in Appellant Win Rate for Cases Where City of New York Appears as Amicus % of change City of New York as amicus supporting appellant 21.13% City of New York as amicus supporting respondent 2.01% Figure 12: Percent Change in Appellant Win Rate for Cases Where NYS District Attorneys Association Appears as Amicus % of change NYS District Attorneys Association as amicus supporting appellant 38.55% NYS District Attorneys Association as amicus supporting respondent -15.40% Figure 13: Percent Change in Appellant Win Rate for Cases Where NYS Trial Lawyers Association Appears as Amicus % of change NYS Trial Lawyers Association as amicus supporting appellant -0.74% NYS Trial Lawyers Association as amicus supporting respondent 16.49% Figure 14: Percent Change in Appellant Win Rate for Cases Where Legal Aid Society Appears as Amicus % of change Legal Aid Society as amicus supporting appellant 11.17% Legal Aid Society as amicus supporting respondent 19.10% Figure 15: Percent Change in Appellant Win Rate for Cases Involving Amicus Support for Appellant Only % of change 1-0-0 cases 17.85% 2-0-0 cases 23.55% >2-0-0 cases 29.26% Figure 16: Percent Change in Appellant Win Rate for Cases Involving Amicus Support for Respondent Only % of change 0-1-0 cases -10.36% -0-2-0 cases -18.27% 0->2-0 cases -9.78% Figure 17: Percent Change in Appellant Win Rate for Cases Involving Amicus Support for Appellant Only (Attorney General Filtered Out as Amicus and Party) % of change 1-0-0 cases 16.00% 2-0-0 cases 20.07% >2-0-0 cases 20.69% Figure 18: Percent Change in Appellant Win Rate for Cases Involving Amicus Support for Respondent Only (Attorney General Filtered Out as Amicus and Party) % of change 0-1-0 cases -8.00% -0-2-0 cases -12.92% 0->2-0 cases -16.45%
|Printer friendly Cite/link Email Feedback|
|Publication:||Albany Law Review|
|Date:||Jun 22, 2009|
|Previous Article:||Shipping up to Boston: the voting of the Massachusetts Supreme Judicial Court in nonunanimous criminal cases from 2001-2008.|
|Next Article:||The New York State Comptroller as sole trustee of the common retirement fund: a constitutional guarantee?|