کیفیت پویای قانون : نقش مشوق های قضایی و سرمایه انسانی حقوقی در سازگاری قانون

In this paper I show that the capacity for a legal regime to generate value-enhancing legal adaptation to local and changing conditions through adjudication depends on its capacity to generate and implement adequate expertise about the environment in which law is applied (shared legal human capital). I present a model in which welfare-promoting adaptation of legal rules depends on the incentives of both judges (to risk rule change) and litigants (to bear the cost associated with informing the court about novel features of the environment and alternative rules.) I demonstrate that a legal system will not adapt through adjudication if legal costs relative to damages are either too high or too low, if judicial incentives for change are inadequate or if judicial error rates are initially too high. Comparatively, I show that a legal regime is likely to do better with improved mechanisms for processing information system-wide to reduce judicial error and lower absolute legal costs, but that other apparently beneficial institutional features such as lower relative legal costs, judges who are more willing to adapt law or who make fewer initial errors may or may not lead to lower welfare overall.

Positive economic analysis of the evolution of law has, since Posner (1977), been organized around the claim that the value of the common law is its ability to work out, over time, efficient legal rules. Some authors have rested this claim on the premise that common law judges inherently seek efficiency; often this literature has framed the analysis as an investigation of the different incentives influencing parties interacting with courts and legislatures and as a debate about whether judges or legislators are more susceptible to rent-seeking (Posner, 1977, Rubin, 1982 and Tullock, 1980). Gennaioli and Shleifer (2007) focus on the impact of judicial bias on the capacity of common law to evolve to efficiency. Other authors have rested predictions about the likelihood that the common law will evolve to efficiency on the incentives of litigants to continue challenging inefficient rules (Rubin, 1977, Priest, 1977, Goodman, 1978 and Cooter et al., 1979). Despite differences, these models of the evolution of the common law all share a common recognition that courts do not start out with the right rules. Rather, they move towards them, if at all, only as a product of their interaction with litigants. With few exceptions (Hadfield, 1992, Talley, 1999 and Hylton, 2006), however, the literature on the evolution of common law has not analyzed the process of information acquisition and learning in courts. Similarly, the recent literature comparing the economic effects of common law and civil code regimes has emphasized differences in agency but not learning relationships (Glaeser and Shleifer, 2002 and Mahoney, 2001). Where information acquisition has been emphasized, it has been in a static rather than dynamic setting. Early work comparing rules and standards (Diver, 1983, Rose, 1988 and Kaplow, 1992) and more recent work comparing judicial discretion to rulemaking in legislatures (Shavell, 2007 and Anderlini et al., 2008) compares efficiency when a rule is determined in light of ex ante versus ex post information. The strategic revelation literature (Milgrom and Roberts, 1986, Shin, 1994, Shin, 1998, Dewatripont and Tirole, 1999 and Daughety and Reinganum, 2000) analyzes information acquisition by courts in the context of the efficient application of fixed rules in a static environment. In this paper I develop a model to explore the dynamic relationship between the institutional attributes of a legal regime and the regime’s capacity to learn about and adapt to welfare-relevant changes in the environment. I focus in particular on adaptation through adjudication on the premise that in any legal regime there is welfare-relevant information that is known initially only to those involved in a regulated activity and that rule enforcers (judges) need to know in order to adapt the interpretation and application of legal rules to a changed set of circumstances. The model thus does not evaluate the broader tradeoff between adaptation through a legislature versus adaptation through enforcement institutions such as courts, the question raised by the legal origins literature (La Porta et al., 1997, La Porta et al., 1998, La Porta et al., 2004, Mahoney, 2001, Djankov et al., 2002, Djankov et al., 2003 and Botero et al., 2004), which identifies an empirical relationship between economic variables and legal families broadly identified as those rooted in civil law (French, German, Scandinavian) and common law (English). Rather it presumes that some adaptation through courts is both necessary and unavoidable (in any legal regime, including civil code regimes, adaptation comes about not only through overt rule change but also through rule interpretation and judgments about how to apply rules in concrete circumstances) and evaluates the impact of institutional features on the likelihood that welfare-promoting adaptation occurs through this process. The model focuses on the accumulation of shared legal human capital across judges in a legal regime. Shared legal human capital determines the likelihood that judges make mistakes in adapting an existing rule to novel conditions. I demonstrate that the capacity for a legal regime to adapt its legal rules in a welfare-improving direction over time is a function of the interaction of the incentives of judges and litigants. In particular, I show that for any legal adaptation to occur there must be some judges who face sufficient rewards for rule adaptation to warrant the risk of making errors and litigants must be willing to invest in developing and presenting costly evidence and legal argument in light of the fact that judges make errors in the use of this information. If the risk of judicial error is too high initially, one or both of these conditions may not hold and the legal regime may remain mired at a sub-optimal legal rule with no error-reducing accumulation of legal human capital. More generally, the capacity for legal adaptation through adjudication is shown to depend on five parameters: exogenous legal human capital, system-wide information-processing (which translates individual case information into shared expertise for judges), the distribution of judicial rewards for rule adaptation, the costs of legal processes to judges and litigants and the level of damages. These parameters are determined by the institutional features of a legal regime such as the use of a formally trained career judiciary as opposed to a capstone judiciary, court specialization, publication practices such as the used of signed decisions and the extent and distribution of judicial opinions and the reliance on sequenced hearings as opposed to a single event trial (Hadfield, 2008). I also compare legal regimes in terms of their capacity for legal adaptation, based on differences in these parameters. I show that lower absolute legal costs and improved information processing unambiguously improve the capacity for a legal regime to adapt its legal rules to changes in the environment through adjudication. Other differences that we might expect to improve legal adaptation – such as more widespread judicial incentives for change or higher initial legal human capital – however, turn out to have ambiguous effects. These apparently salutary attributes can reduce welfare if they cause excessive levels of experimentation with novel rules before enough legal human capital has accumulated to enable judges to implement novel rules (or existing rules in novel ways) with sufficient accuracy. A further surprising result is that a regime does not necessarily do better if it is better at discouraging opportunistic litigants who seek to induce judges to change rules in welfare-reducing rather than welfare-enhancing directions. This depends on assumptions about the nature of information processing: if increased information across a diversity of cases contributes to the systemic capacity to distinguish when rules should adapt and when they should not, then legal regimes that initially allow in information from opportunistic litigants may reduce judicial error more effectively and benefit more from rule change over time. The key results thus emphasize that long-run improvements in the capacity for a legal regime to respond to a changing environment require some but not too much experimentation with legal rules to promote judicial learning. Section 2 lays out the model and then assesses litigants’ incentives to invest in the costly effort to educate a court about the new or changed environment and judicial incentives to adapt the rule, given the risk that departing from an established rule will be costly for the judge. This section establishes conditions under which a system will adapt and conditions under which it will remain stuck at an established rule, despite the existence of a welfare-improving adaptation. Section 3 then evaluates the welfare implications of rule adaptation. I show that even if a welfare-improving rule exists, it may not be optimal for a regime to adapt in light of the costs of the proceedings necessary for judges to acquire information and the cost of errors incurred in the process. If sufficient learning benefits accrue, however, it can be optimal for a regime to adapt even when the initial impact of rule adaptation is welfare-reducing. Section 3 compares the level of welfare achieved in regimes that differ in initial legal human capital, legal costs, damages, potential learning benefits and judicial incentives. Section 4 summarizes the results, discusses limitations and extensions and relates this paper to the literatures on legal origins, the evolution of law, the choice between rules and standards, and the strategic revelation of information in courts.

The model in this paper is a first step to connect the analysis of information processes in different types of legal institutions to the dynamic process of legal adaptation. The model identifies five parameters that influence the capacity of a legal regime to adapt law to local and changing conditions through adjudication so as to promote social welfare: (1) the distribution of judicial rewards for rule-adaptation; (2) the cost of producing evidence and legal argument for presentation to a court; (3) the level of damages; (4) the initial or exogenous level of judicial error and (5) the extent to which the legal regime transforms individual case information into informative – error-reducing – shared legal human capital. I demonstrate that in order for a legal regime to adapt law through adjudication it is not enough for judges to face incentives to adapt law. They must also anticipate that the errors they will make in adapting law to novel circumstances will not be too high. Moreover, they depend on the incentives of litigants to bear the cost of educating them about the environment. These incentives also depend on judicial errors: if judges make too many errors, good litigants for whom legal change is welfare-promoting will be discouraged from investing in these costs. A regime may thus be stuck at an existing rule despite the existence of a welfare-improving adapted rule because the mistakes judges are likely to make in a novel environment discourage judges, litigants or both from investing in the development of the shared legal human capital necessary to reduce error. Assuming a legal regime does not remain stuck at an existing sub-optimal rule, the model presents some surprising results about how variation in the parameters that affect the process of legal human capital accumulation through adjudication can impact the dynamic quality of the law. Legal regimes will achieve higher welfare over time through rule adaptation if they have lower absolute legal costs or a more effective process for translating case-specific information into system-wide error reducing legal human capital. Because the process of legal human capital accumulation requires a balancing of costs and benefits over time, however, attributes that might appear to promote or defeat welfare can have surprising effects. If more judges face higher payoffs for rule adaptation there may be excessive adaptation in early stages when judicial errors are relatively high. Lower initial levels of judicial error (higher exogenous legal human capital) may also lead to an initial overinvestment in rule adaptation. Even lower legal costs, relative to the damages at stake, do not unambiguously improve the dynamic quality of the legal regime: if relative legal costs are low, “bad” litigants who seek to exploit the likelihood that judges will mistakenly adapt rules in welfare-reducing ways may flood the system and, anticipating this, judges will be less willing to allow any litigants to proffer the evidence and legal argument necessary for them to learn and reduce the likelihood of error. As I explore in more detail in a companion paper (Hadfield, 2008), each of the parameters in the model is linked in important ways to the institutional environment that defines a legal regime. Judicial rewards for rule-adaptation, for example, are a function of the organization of the judiciary and the nature of the information available to those who make up a judge’s audience. Information processing is a function of how much factual detail and legal argument is included in judicial decisions and how widely those decisions are distributed. These and other features that affect the parameters I analyze differ systematically across legal systems. The model in this paper thus contributes to but certainly does not resolve the issue posed by the legal origins literature and implicit in the literature on the efficiency of the common law about the relative economic value of different legal institutional settings. An important limitation of the paper is the focus exclusively on legal adaptation through litigation. Any legal system will experience some adaptation through this mechanism: legal rules do not only change when the text changes (which is what legislatures can do) but also when the interpretation and implementation of text changes (which is what only those who enforce rules can do). But learning through adjudication is clearly not the only mechanism for infusing legal rules with information. The contribution of this model is to emphasize the need to focus on the process by which information available to those with presumably the lowest costs of learning the information (those involved in the regulated activity) is transmitted into the system of regulation. I have shown how this process in courts depends in subtle ways on the interaction of the incentives of judges and litigants. Further work would more explicitly address the comparable issue of how different institutional settings affect the capacity of legislatures to learn and adapt law, particularly in light of the inevitable process of adjudication, interpretation and implementation in courts. Further extensions of the model could also integrate this work with the strategic revelation literature. I have modeled the information process in courts in reduced form and suppressed any decisionmaking by plaintiffs. Provided the equilibrium is not fully revealing (Milgrom and Roberts, 1986), this reduced form approach is without loss of generality: whatever type 1 and type 2 errors result following the defendant’s presentation of its case can be interpreted as the outcome of a strategic game between defendant and plaintiff. It remains to be seen, however, whether further insight into the dynamic accumulation of legal human capital can be generated by embedding the error process more explicitly in a game theoretic model. Game theoretic modeling of the information process in courts could be particularly important to address another important limitation of the model presented here. I have assumed that presentations of evidence and argument from both good and bad defendants can be equally informative when processed by the system over time. Indeed, information from bad defendants, even if it distorts outcomes in particular cases, may be valuable in the long run because those who sift and synthesize information across cases may be better able to extract error-reducing legal human capital from a diversity of cases rather than those of a single type. (By the same token, variation in a sample improves the capacity to identify significant relationships in empirical work.) Nonetheless, it is likely the case that if a system attracts too much information from bad defendants that the system as a whole experiences increases rather than decreases in error rates. The strategic interaction between defendants and plaintiffs is likely to play a role in determining the balance of information that accumulates. Finally, a fuller analysis must also take into account the impact on information acquisition and legal human capital accumulation of several features suppressed here such as activity choices, settlement, norms of deference to the decisions of other judges (precedent and jurisprudence constante) and appeal. Hadfield (1992), for example, points to ways in which information acquisition by courts can be biased by the activity choices of agents along the evolutionary path under particular institutional assumptions regarding the impact of precedent and case or controversy requirements.6Talley (1999) considers the potential for information cascades in courts in a framework that can be easily interpreted to have relevance to the comparison between different institutional settings. Hylton (2006) examines the impact of settlement on the mix of cases that go to trial and thus form the basis for rule evolution. Ultimately, the impact of different legal institutional structures will also require integration of the analysis of the information processes in courts and legislatures and the agency considerations (corruption, judicial preferences, etc.) that have dominated the legal origins literature to date. With further theoretical work we can then hope to unpack more of the empirical results and policy implications surrounding the design of legal systems to support economic activity.