Rules and Standards in a Judicial System

Political scientists interested in the structure of legal doctrine have been especially attuned to how it is affected by the judicial hierarchy. The issue is generally framed as whether a higher court will issue a rule or a standard (Tiller and Cross Reference Tiller and Frank2006; Jacobi and Tiller Reference Jacobi and Emerson2007; Heytens Reference Heytens2008; Cross, Jacobi, and Tiller Reference Cross, Tonja and Emerson2012; Lax Reference Lax2012).Footnote ¹ As first defined by legal scholars (Radin Reference Radin1991; Kaplow Reference Kaplow1992), both rules and standards explicate the law, providing information and guidance to other legal and political actors, but they differ in the rigidity of that guidance.

Rules are precise and specific, leaving little room for subsequent interpretation, while standards are more amorphous and place greater discretion in the hands of the subsequent adjudicator. Rules contain explicit categories or factors against which particular facts are measured; standards require nuanced balancing of various elements by the jurist charged with implementation. To borrow the classic example, a rule for speeding would be that no driver may drive over 55 miles per hour. A standard might simply say that the speed should be “reasonable and prudent” (Sullivan Reference Sullivan1992; Lax Reference Lax2012).

When considering how the judicial hierarchy might affect the structure of legal doctrine, most scholars take a principal-agent approach: the higher court is the principal that must rely on its agent—the lower court—to implement its rulings (Carrubba and Clark Reference Carrubba and Tom2012; Hansford, Spriggs, and Stenger Reference Hansford and Anthony2013; Kastellec Reference Kastellec2017).Footnote ² Given that its preferences may not align with the higher court, the lower court can attempt to resist the wishes of the higher court and follow its own predilections.

Legal doctrine is one tool the higher court uses to ensure compliance (Cross et al. Reference Cross, Tonja and Emerson2012; Lax Reference Lax2012; Smith and Todd Reference Smith and James2015). If the lower court is likely to shirk, the higher court can issue restrictive rules; as shirking becomes less likely, the higher court may lean toward the vaguer standard. The propensity for shirking depends primarily on the ideological alignment between higher and lower courts. If the higher court faces an ideologically distant lower court, it will promulgate a rule. If the lower court is more ideologically aligned with the higher court, the guideline can be the more ambiguous standard (Spiller and Spitzer Reference Spiller and Matthew1992; Jacobi and Tiller Reference Jacobi and Emerson2007; Baker and Kim Reference Baker and Pauline2012; Smith and Todd Reference Smith and James2015).Footnote ³

Legal scholars have engaged more normatively in the rules-versus-standards debate, arguing about which type of legal guideline a court should promulgate and why. Sometimes, the claim has been that one structure is inherently inferior to the other: rules are faulted for being overly restrictive and difficult to apply to new factual situations; standards are said to place too much discretion in those charged with implementation of the rule and are criticized for failing to provide sufficient guidance for citizens to resolve their own conflicts (Schauer Reference Schauer1991; Korobkin Reference Korobkin2000; Nance Reference Nance2006; see also Schlag Reference Schlag1985). For others, the trade-off is more complex, with each form generating its own costs and benefits that vary with the legal and political context (Schlag Reference Schlag1985; Korobkin Reference Korobkin2000). Most frequently, scholars’ normative preference for rules or standards depends on the legal area under study; works analyzing the dichotomy across a vast range of doctrinal arenas continue to appear in the legal literature (Diver Reference Diver1983; Lee Reference Lee2002; Overton Reference Overton2002; Crane Reference Crane2007; Phillips Reference Phillips2010).

Beyond Rules versus Standards: Rule Selection and Rule Flexibility

All of these works rest on presumptions about the sources and shape of legal doctrine. They assume that justices ultimately control doctrinal content and that doctrine actually is structured with varying degrees of fluidity. I first question the notion that Supreme Court justices develop whatever legal guideline they wish (cf. Hammond, Bonneau, and Sheehan Reference Hammond, Chris and Reginald2005; Baker and Kim Reference Baker and Pauline2012; Lax Reference Lax2012; Smith and Todd Reference Smith and James2015). The justices certainly can control opinion content, but we do not know for sure that they write legal rules in particular. It may be that to save time and resources, or to fulfill any professional obligation they feel to resolve the dispute at hand, they draw instead from the rules in the briefs submitted by case participants (Corley Reference Corley2008; Collins, Corley, and Hamner Reference Collins, Pamela and Jesse2014; Wofford Reference Wofford2015). Because these proffered guidelines constitute the set of available choices, any study of rules and standards should examine what is offered to, not just formulated by, the justices.

Second, even among the guidelines presented to them, the justices may not confront two dichotomous options. The relevant literature has often labeled doctrinal development as a choice between a (judicially created) precise rule or a vague standard. However, I posit that the Supreme Court almost never faces this type of decision but instead chooses between versions of the same type—rules that are more or less rule-like and standards that are more or less standard-like (Kaplow Reference Kaplow1992).

As a result, rather than study whether the Court issues rules or standards, this study employs a more general concept of rule flexibility. Drawing from the legal literature (Radin Reference Radin1991; Sullivan Reference Sullivan1992; Sunstein Reference Sunstein1995; Korobkin Reference Korobkin2000), I suggest that all legal guidelines—which I term “rules”—exist on a continuum of flexibility. Rules, therefore, can be extremely rigid, extremely ambiguous, or somewhere in between. Rule flexibility is the extent of discretion the rule imparts to lower courts—that is, the amount of leeway that lower court judges have to use and implement the rule. The more flexible a rule, the more discretion it offers, and the greater the freedom lower court judges have to apply the rule; the less flexible a rule, the less discretion it offers, and the more constrained lower court judges are in implementation.

Rule flexibility is shaped by several factors. To begin, there is the precise language that constitutes the rule (Cross et al. Reference Cross, Tonja and Emerson2012). Certain words such as “reasonable,” “equitable,” or “substantial” produce rules that offer lower court judges more discretion. Rules with absolute words such as “never” or “always” or that set specific guidelines for time or amount (i.e., a reply must be filed within 60 days or an individual under the age of 16 cannot be executed) do not allow a lower court judge much leeway.

The grammatical complexity of a rule also can affect its flexibility. Certain rules have a very simple structure, with straightforward explanations for implementation written as declarative sentences (e.g., the lower court must “determine if the amount of work leave requested falls within the statutory time period of thirty days”).Footnote ⁴ Other rules, however, are more complex, involving multiple phrases, subclauses, or factors for a judge to consider (e.g., the judge must “balance the reasonableness of the work leave request with the needs of the employer, attending to the facts and circumstances of each case”).Footnote ⁵

Lastly, the content of a rule can affect the discretion given to lower courts. For instance, rules that grant (or deny) jurisdiction to a lower court, grant (or deny) the lower court the capacity to issue a legal remedy, or determine which statutory or constitutional provision a lower court should apply shape what the lower court can or cannot do in the immediate case.

With justices potentially selecting from a set of rules presented to them, rules that vary in their flexibility, the key question then becomes not whether a justice will formulate a rule or standard but how, given the Court’s position at the apex of the judicial hierarchy, the flexibility of a proffered rule might influence whether it is favored by the justices. It is this query to which this preliminary study is directed.

Data: Legal Rules and the Justices’ Votes

To explore this question, I gathered all the legal rules suggested to and adopted by the justices in 500 randomly selected Supreme Court cases from 1954 to 2000, drawn from the Spaeth et al. (Reference Spaeth, Lee, Andrew, Jeffrey, Theodore and Sara2018) data set. To ensure that the justices were faced with a choice of competing legal rules, I used only cases Spaeth et al. identify as involving intercircuit conflict among the lower courts.Footnote ⁶

To gather all the legal rules suggested by all case participants, I consulted every brief submitted to the justices by litigants and amici, using both Lexis and Westlaw. From these briefs, I then identified and extracted every legal rule suggested to the justices.Footnote ⁷ Any case without briefs was dropped and replaced with the next case in the list.Footnote ⁸

Who suggested the rule to the justices was then recorded, including all petitioners, respondents, and amici.Footnote ⁹ Amici had to express explicit support for a particular legal rule to be counted as supporting that rule; general claims about what outcome the court should reach or why were not sufficient.

I next read all the court opinions (majority, concurring, and dissenting), coding whether each justice supported a particular rule. If a justice wrote or joined the majority opinion, that justice was presumed to support the rule favored by the majority opinion; if a justice wrote or joined a dissenting opinion that supported another rule, that justice was presumed to support that other rule. For concurring justices, the justice was coded as supporting the majority rule unless the justice explicitly rejected the rule, refused to join that portion of the majority opinion that contained the rule, or endorsed a competing rule.

Occasionally, I had to code a justice’s vote as missing. In general, if a justice did not express or join an opinion that explicitly expressed support for a particular rule, that justice was coded as missing. Dissenting or concurring justices, for example, were coded as missing if they rejected the majority’s rule but did not clearly adopt a rule of their own; they were also coded as missing if their decision was based on a different legal question.Footnote ¹⁰ Justices who did not participate in the case also were coded as missing. When complete, this process produced a data set of 4,449 individual judicial votes, with 91 coded as missing.Footnote ¹¹

Measuring Rule Flexibility

Evaluating rule flexibility requires the creation of a novel measure. I developed a four-point scale that ranges from 1 (very flexible; gives a lot of leeway), 2 (flexible; gives a fair amount of leeway), 3 (somewhat flexible; gives a bit of leeway), to 4 (not at all flexible; gives no leeway). Two third-year law students were hired for their legal expertise as coders. They were provided with the legal rules offered by the petitioner and the respondent in 500 randomly selected Supreme Court cases (1954–2000) drawn from the Spaeth et al. (Reference Spaeth, Lee, Andrew, Jeffrey, Theodore and Sara2018) data set, and they were instructed on how to code the flexibility variable.Footnote ¹²

To highlight the facial validity of this measure, consider the following two cases from the data set. In Business Electronics Corp. v. Sharp Electronics Corp. (485 U.S. 717 [1988]), the Court had to decide whether a certain type of agreement between a manufacturer and distributor violated antitrust law. One party suggested that the Court adopt the “rule of reason,” in which “the factfinder weighs all of the circumstances of a case in deciding whether a restrictive practice should be prohibited as imposing an unreasonable restraint on competition.” The other proffered rule simply stated that the type of agreement at issue “constitute[d] a per se violation” of the applicable statute. Both coders gave the first rule a score of 1 (highly flexible) and the second rule a score of 4 (highly inflexible).

In Pioneer Investment Services Co. v. Brunswick Associates Ltd. Partnership (507 U.S. 380 [1992]), the Court confronted a statute that required courts to accept late filings by litigants if the delay was caused by “excusable neglect.” The petitioner suggested that excusable neglect existed when “failure to file timely [was] due to circumstances beyond [the litigant’s] reasonable control.” The respondent argued that determinations of excusable neglect should be made by evaluating a late filer’s conduct against “several factors, including: (1) whether granting the delay will prejudice the [other party]; (2) the length of the delay and its impact on efficient court administration; (3) whether the delay was beyond the reasonable control of the person whose duty it was to perform; (4) whether the [litigant] acted in good faith; and (5) whether clients should be penalized for their counsel’s mistake or neglect.” Here, the rules had similar levels of flexibility, with scores from both coders of 3 and 2, respectively

In terms of intercoder reliability, there was an extremely high level of intercoder agreement: 90.55% for petitioner’s rule and 87.43% for respondent’s rule (Freelon Reference Freelon2010). To account for expected agreement, I also calculated intercoder reliability statistics. In terms of coding petitioner’s rule, the bivariate weighted Kappa value was .52; for respondent’s rule, it was .47. Both values indicate “moderate” agreement (Landis and Koch Reference Landis and Koch.1977). The intercoder reliability score (Kappa) on which rule was more flexible was .68, indicating “substantial” agreement (Landis and Koch Reference Landis and Koch.1977).

In an ideal world, there would be a more objective measure of rule flexibility. One option would be to measure the length of the legal rule by counting the number of words. Scholars have used the length of texts as a proxy for the complexity of Court opinions and congressional statutes (Huber and Shipan Reference Huber and Charles2002; Randazzo, Waterman, and Fine Reference Randazzo and Jeffrey2006; Black and Spriggs Reference Black2008). Others have used “readability” scores (Coleman Reference Coleman2001; Law and Zaring Reference Law and David2010). There also are content-analysis software programs, which offer more sophisticated measures of linguistic clarity and complexity and are becoming increasingly popular (see, e.g., Law and Zaring 2010; Owens and Wedeking Reference Owens and Justin2011; Owens, Wedeking, and Wohlfarth Reference Owens, Justin and Patrick2013).

The problem with all these approaches, however, is that they measure complexity, not flexibility, concepts that are not interchangeable. The more complex a rule, the more difficult it may be to understand and apply, but this does not mean that it will necessarily give a lower court judge more discretion. A more complex rule might be more flexible: a rule with more words or a greater number of clauses, for example, might give judges more “wiggle room” by providing additional language the lower court judge could interpret at will. At the same time, a longer or more grammatically intricate rule could just as easily restrict a judge’s discretion by providing more language that limits how a judge should apply the rule. Simpler and shorter rules work the same way, potentially providing more or less discretion depending on the nature of the words themselves. Viewed this way, the amount of discretion turns as much on the content of the words as the complexity or clarity of the rule. Extant approaches to complexity unfortunately do not capture this distinction.

Other scholars have used the Linguistic Inquiry and Word Count (LIWC) program to study other aspects of Court opinions (Cross and Pennebaker Reference Cross and James2012; Collins et al. Reference Collins, Pamela and Jesse2014; Corley and Wedeking Reference Corley and Justin2014). Originally developed by psychologists, LIWC is a dictionary-based program that can assess documents for the use of certain types of words. Unfortunately, while LIWC can, for instance, determine how much “certainty” or “negative emotion” exists in a document or how many times a speaker uses a “big word” (more than six letters), it contains no dictionary that maps to flexibility or ambiguity, even under the most generous definitions of those terms.

The software does permit the user to build an original dictionary, but it is unclear what words would need to be included in such a dictionary for this study. Words such as “reasonable” or “equitable” would be obvious candidates, but including every word or phrase that could render a rule more flexible is likely to be vastly overinclusive (including many words that might never appear in legal doctrine) and underinclusive (missing words that only generate flexibility when used in the context of a specific legal issue). At the very least, constructing such a dictionary is beyond the scope of this project. Until technology advances sufficiently, employing human coders seems the best approach to measuring the precise concept I study here.

Two additional factors mitigate the subjectivity of the measure. Others have used similar measures of related concepts with success (Spriggs Reference Spriggs1997; Breitmeier, Young, and Zürn Reference Breitmeier and Michael2006). The process undertaken by coders also replicates what the justices themselves likely do: use their legal training to assess how much discretion the rule provides to lower court judges. Law students are not justices, of course, and would lack the latter’s experience with promulgating rules for lower courts, but the ultimate task of reading, interpreting, and evaluating a legal rule in light of the judicial hierarchy should be quite similar.