Judges as satisficers: a law and economics perspective on judicial liability

Abstract

If judges are guardians of the law, who is to protect the individual member of society from the occasional corrupt, malicious, or reckless judge? The aim of this paper is to provide an answer to the last part of this question, focusing more heavily on cases of negligently inflicted harm. Departing from Simon’s bounded rationality and influenced by other constructs in behavioral law and economics, we view judges as satisficers who make decisions within real-world constraints, such as imperfect information and uncertainty, cognitive limitations and erroneous information. Judges are limited by the commonly observed barriers to the decision making process. Because their goal is not to optimize but to render opinions that are merely satisfactory, they often act as poor agents of their principals’ interests. In this light, it becomes clearer why judges tend to engage in behavior that is “improper”, especially under the circumstances of the currently overloaded judicial caseloads. We first address the differences in judges’ roles in Anglo-American and Continental legal systems. We then present our simple model for judicial misbehavior based on an understanding of judges as “satisficers”. Next we discuss the particularities of judicial errors and introduce a realistic and viable construct of “inexcusable judicial error”. On this basis we evaluate the impact of various incentive schemes on judicial behavior, focusing on the civil liability of judges. We conclude that civil liability for grave judicial errors is the most adequate remedy.

1 Introduction

Judges enjoy privileged status regardless of legal regime. In both civil and common law countries, their profession is the most revered among legal professions.¹ They enjoy high prestige, high incomes and high power. The reasons are many—and they have been well-documented. For the purposes of this paper, we will isolate one: the level of sophistication of “what judges do”. The average layperson is quite ignorant of judicial interpretation and all that it entails. The distance that separates “judge” and “common man” is too great—and even if the instances of judicial error are statistically few, when they occur, they are likely to cause great and sometimes irreparable damage.

Writing on judicial misconduct, Michael Robert King had asked the following question: “Judges are guardians of the law, but who is to protect the individual member of society from the occasional corrupt, malicious, or reckless judge?” (King 1978). This paper aims to provide an answer to this question, focusing on cases of negligently inflicted harm, which have provoked more controversy in both the literature and the case law. For reasons of simplicity, we look to the “average” judge and the factors that will ordinarily guide his or her interpretation.

We proceed with our analysis as follows: First we address the particularities of the role of the judge and the different implications of the judge’s image in Anglo-American and Continental legal systems. We then propose a simple model for judicial misbehavior that views judges as “satisficers” (part 2). We apply this model predominantly in cases of civil litigation because looking into disciplinary and criminal proceedings poses different sets of problems that would overextend the scope of the present analysis. We explore the limitations of judicial decision-making and then map judicial misbehavior as a special application of the principal-agent problem. Next we discuss the particularities of the liability for judicial errors (part 3). We then come up with a working notion of “inexcusable judicial error” (part 4). On this basis we analyze and evaluate the impact of various incentive schemes on judicial behavior, focusing on the civil liability of judges (part 5). Finally, we sup up the results of the analysis and we conclude that civil liability for grave judicial errors is the most adequate remedy (part 6).

2 Why do judges make mistakes?

For the purposes of this paper, we will take a closer look at judges’ civil liability. We are interested in shedding more light on those cases when judges misbehave out of gross negligence or malice. To do this, we need a better understanding of why judges misbehave. But what constitutes “proper” or rather “improper” judicial behavior? And what factors drive judges to conform to the ethical norms of their profession? If we assume that judges indeed “maximize what everybody else does” (as Posner first asserted in 1994), then judges fit nicely within the existing “rational utility-maximizer” model that has found so many applications within the Law & Econ camp. But is “Homo Economicus” an accurate and fitting description of the average judge today? What if judges actually have utility functions that are different from those previously assumed?

As we begin to answer these questions, we need to make some important distinctions: judges in Continental law are quite different from judges in Anglo-American law. The legal systems are different and the legal traditions diverge. Ultimately the legal cultures that have grown out of the different legal evolution processes make it quite difficult, if not impossible to propose a unifying model that will apply for both sides of the Atlantic.

To give an illustrative example, we discuss the different implications of the judge’s image in Anglo-American and Continental law. The civil law judge is not a culture hero or father figure, as he is in the common law. His image is rather that of a civil servant who performs important functions that are not creative. On the contrary, the common law judge is perceived as one who “discovers” and “creates” (Merryman 1969: 38). American judges are perceived by the general public as policymakers, whose politics matter more than their integrity (if one draws conclusions from the public debate surrounding decisions like the US Supreme Court’s Roe v. Wade).² European judges, even if they are given the opportunity, rarely take the initiative to strike down legislation enacted by parliament.

It is obvious and empirically observable that different institutions and different ways of market exchange have translated into different norms and different judicial perceptions in the Anglo-American legal culture as compared to the “situation in Europe”. This might mean “different preferences”, which leads us to question the standard assumptions of the Homo Economicus model (that judicial preferences are largely “exogenous”, fixed, given) and to look for a more appropriate hypothesis.

2.1 Models of judicial behavior

Over the past three decades, judicial behavior has been examined in the light of different models: some give emphasis to socialization variables, attitudes and roles, while others focus on institutional factors, precedent and strategic thinking. In 1983, Gibson had proposed a unifying “integrated” model of judicial activity, which shifted attention to the judge as individual decision maker (Gibson 1983).

The benefits of complying with the law include maintaining a good reputation and conformity with tradition. We assume that judges’ concern for reputation has positive spillover effects, leading them to foster the creation of pro-social norms. For purposes of understanding the law, people may sacrifice their economic self-interest in order to be, or to appear, fair (Sunstein 2000: 8). Behavioral economists like Bowles and Gintis (2002) have proposed the alternative model of Homo Reciprocans. They have suggested that the institutions of the market can only work if many people (e.g., police, judges, parents, soldiers) do not, in the line of duty, act like Homo Economicus at all, but instead act more like Homo Reciprocans (see also Fehr and Gachter 1998: 337). As noted by Hirschman (1985), a principal purpose of publicly proclaimed laws and regulations is to stigmatize antisocial behavior and thereby to influence citizens’ values and behavior codes. This expressive, value-molding function of the law is just as important as its deterrent and repressive functions.³

Tradition seems to play a pivotal role in all judicial cultures. Judges have a taste for adherence to tradition. Precedent following, also known as stare decisis, is a cornerstone of common law systems. Similarly, jurisprudence constante (the legal doctrine according to which a long series of previous decisions applying a particular rule of law may be determinative in subsequent cases) plays a prominent role in civil law jurisdictions. References to past decisions show how the court is following the course it has set, and maintain stability and tradition. This practice is especially important for higher courts, where stability in justices’ attitudes about the law goes hand in hand with a normative desire on the part of judges to maintain positions that they have taken in the past (Zorn 1998: 4).

Posner (2007: 585) argues that a precedent projects a judge’s influence more effectively than a decision that will have no effect in guiding future behavior. He adds: “If the current generation of judges doesn’t follow precedent, the next generation is less likely to follow the precedents of the current generation because the next generation’s judges are less likely to be criticized for not following their predecessors’ precedents” (id.). Criticism, he explains, which in most walks of life is a weak force, is likely to influence judicial behavior, principally because of the rules of judicial tenure and compensation.

Reasoning from past cases is a heuristic device with wide applicability in judicial decision-making. Because it is so difficult to calculate the expected costs and benefits of alternatives, judges simplify their burdens by reasoning from past cases and thinking analogically (Sunstein 2000: 5). This form of “case-based decision making” is predominant in courts (Gilboa and Schmeidler 1995).

Another key motive for judges is conformity. In game-theoretic terms we could say that judges “play the game” and form their expectations by observing how other “players” (i.e. judges) behave. By following precedent, judges assume a critical role in sustaining a good equilibrium and building an important form of social capital (Coleman 1988). The motive to obey precedent is coupled with and reinforced by judges’ taste for gaining prestige.⁴ This is especially true of younger judges issuing decisions for lower courts.⁵ This builds on an earlier insight by Cooter (1983) arguing that private judges tend to make efficient decisions because they compete for business from litigants and had suggested that public judges tend to do the same to acquire prestige.

In the same vein we find a model proposed by Rasmusen (1994) in which judicial legitimacy is seen as a repeated game. This influence model assumes a game of perfect information with symmetric players.⁶ Rasmusen discusses the possibility of moving to a better equilibrium rather than abandoning hope of a responsible judiciary. He draws on game theory to identify the ways in which judicial expectations are formed: an equilibrium that is Pareto optimal, that uses simple strategies, that has been played out in the past, and that is publicly announced to be the equilibrium becomes a “focal point”, i.e. an equilibrium attractive for psychological reasons (Rasmusen 1994: 78).

One could take into consideration other motives of judges—and their desire to exert a future influence. The standard assumption is that judges want to influence policy. In the US, several empirical and quantitative studies have documented the impact of interjustice influence on Supreme Court decision making. A series of studies (Maltzman and Wahlbeck 1996a, b; Wahlbeck et al. 1998; Caldeira et al. 1999) have focused on strategic influences on such activity as opinion assignment and opinion coalitions.

Judges also want a number of other things: to vote, to register an opinion and have leisure time (Posner 1994)—and to maintain a good reputation with the public. In Posner’s view, judges’ preferences materialize within a utility-maximization framework.⁷ Of course brilliant judges (like Holmes and Cardozo)⁸ are arguably driven by a different set of motives. But looking into these motives is beyond the scope of this paper, which aims to scrutinize the factors that will ordinarily guide the interpretation of the average judge.

Yet another useful perspective is the risk aversion of judges. Incorporating risk attitude into existing models is very important for individual decision making. In the literature judges are portrayed as conservative people who want to avoid controversy at all costs. But the area in which risk aversion finds its primary application is judges’ aversion to having their rulings reversed by appellate courts, which leads them to fit the facts of the current dispute into available precedents.⁹

2.2 Judges as satisficers

Over 40 years ago, Herbert Simon coined the term of “bounded rationality”¹⁰ and distinguished between two types of decision-makers: maximizers and satisficers. The maximizers seek the “best” outcome and exhaustively search all possibilities; the satisficers look for an outcome that is “good enough.” In the literature, managers have been identified as satisficers¹¹—but less attention has been given to judges as satisficers. The theory of satisficing has had an impact in both organizational theory¹² and in traditional public administration theory (Fry 1989), fulfilling the need for some firm foundation that can link human behavior to macropolitics (Jones 2003: 395).

Viewing judges as satisficers helps explain both “what it is that judges do” and “why it is that judges misbehave or make serious mistakes”. In any given legal dispute that comes before the courts, the players involved (judges, litigants, attorneys) are not the perfectly informed, risk-neutral rational actors assumed in traditional economic accounts. Behavioral economics has shown the impact of biases on the decision-making process of rational actors. It is now widely accepted that we all suffer from biases which affect our judgment and may lead us to inaccurate perceptions. However, some types of bias (such as hindsight bias, optimistic bias, status quo bias and extremeness aversion) have serious implications for legal judgment.

Under Simon’s bounded rationality, the decision maker endeavors to make a rational choice within the real world constraints—e.g., imperfect information and uncertainty, cognitive limitations, erroneous information, etc. Judges as decision-makers are limited by the commonly observed barriers to the decision making process (Hagle 1990): the first is that they normally cannot control all the elements of a problem. The second is that of unintended results: even if the various actors react as predicted, additional unforeseen results may also occur. Third, it is often not possible to make an accurate cost-benefit analysis of the possible solutions to a problem; data are commonly missing or unavailable and the information is private. Fourth, the goals of judges as decision-makers may not be known or adequately identified. Finally, their decisions may be based on inaccurate forecasts of the consequences of actions.¹³

Faced with such cognitive limitations and biases, the legal actor cannot identify alternative choices that lie outside the boundaries of institutional procedures. The consideration of alternatives is limited and the actor will make decisions by using various simplifying heuristic devices¹⁴ (what Simon called “functional heuristics”) for recognizing and solving problems (Brisbin 2004: 12). Procedural bounded rationality (or instrumentalism) assumes that the individual choice of the decision maker transpires through a practice of “appropriate deliberation” to choose a means to gain a set of ends, to complete a specific task or address a specific problem (Simon 1982: 426; Jones 1999: 301–302).

More recent models of bounded rationality do not provide a general theory of decision making (see esp. Gigerenzer 2002), but instead consist of simple rules that function well under the particular state of available information and cognitive ability available to the user of those rules. Hence, every agent carries a box of simple tools (rules), each of them adapted to a particular problem, rather than an all-purpose sophisticated computer, i.e. a general approach to decision making (Gigerenzer and Selten 2002).

In most legal systems worldwide, judges have been increasingly burdened with the costs of an ever-expanding case load. Courts, especially in civil law jurisdictions, do not have mechanisms for dividing labor¹⁵—so each judge must deal with all the information from each case and read as much of the material as is necessary to come to a decision. The information judges acquire is often incomplete. To acquire more information for each case on their docket would impose tremendous time costs.¹⁶ Thus, the litigation explosion has imposed significant costs on judicial decision-making, distorting the judges’ incentives to comply with the law and also increasing the likelihood of legal errors.¹⁷

The bounded rationality framework can also be used to explain judicial conformity defined as judges’ commitment to existing and well-embedded patterns of choice. Institutional rules and requirements can induce a reluctance to depart from existing policy or doctrine. They also reduce the psychological costs of decision making under conditions of ambiguity. Most legal choices are satisficing for the reasons described above and allow for an evolutionary adaptation of legal policy. Over time, these satisficing practices become a source of path dependence—that is, sequential political or legal choice within a relatively stable or narrowing range of alternatives (see Thelen 2003: 217–222).¹⁸

We could better understand the behavior of judges-as-satisficers if we frame our analysis as a particular application of the “principal-agent” problem as it was employed by behavioral economists studying the firm.¹⁹ In this sense, much like the firm’s managers, judges do not strive to maximize a value (e.g. to serve the higher ideal of justice or pursue the social goals of efficiency and wealth maximization). Instead, they try to reach a minimum but satisfactory level of value, especially through adherence to precedent in order to secure the support of other judges, but mainly through their efforts to come up with acceptable solutions to the problems they are called upon to adjudicate. Consequently, the main concern of judges is to produce results that are merely “good enough” so that they may be free to pursue their other goals. These other goals could for example include rent-seeking, the pursuit of political ambitions, more leisure time, and so on, depending on the legal system, on the idiosyncrasy of the particular judge, etc.

For each individual judge it then becomes crucial to weigh the costs and benefits of misbehaving. In order to minimize the instances of judicial liability, we propose reducing these benefits or increasing these costs. We must assume that judges will respond to normative incentives.²⁰ We propose the introduction and enactment of “inexcusable judicial error” to handle cases of judicial behavior that is grossly negligent.²¹ By increasing the costs of misconduct, judicial liability can become an effective vehicle for protecting litigants from judges who abuse their power.

Of course there are other ways to deal with judicial liability: the state can use external incentive-inducing mechanisms such as harsher penalties for judges who misbehave (e.g. impeachment), cut-backs in the number of judges, improvements in the judicial selection process, promotion of ADR methods to alleviate the judicial caseload,²² a more efficient system of allocating cases to judges,²³ better means of monitoring judges for possible misconduct or changes in procedural rules, etc. Then it may be able to form preferences in such a way that judges will behave more responsibly even after the external threat is removed.

3 What makes liability for judicial errors special?

Due to the abovementioned cognitive limitations, judges as decision makers are prone to making mistakes. Holding the judges liable for each and every mistake they make would be rather unreasonable; the cost of issuing a flawless decision is immense in terms of time and would mean considerable delays in deciding other cases, especially taking into account the work load of most judges. In this respect, the case of judicial liability does not seem to differ from other cases of professional liability, at first glance. Hence, from a law and economics perspective the “Learned Hand Formula” could apply, according to which actors should be held liable if the costs in trying to avoid the occurrence of damage are lower than the expected losses.²⁴ However when it comes to judicial liability there exist additional factors, apart from the burden of the judge and the loss of the victim, which ought to be taken into account:

First of all, a system of judicial liability has to be compatible with the need to preserve the independence of the judiciary. Judicial independence is protected in all legal orders, often at a constitutional level.²⁵ In this context judges are granted guarantees which are meant to enable them to decide each case impartially, free from interferences and in accordance with their conscience.²⁶ Moreover, it is of vital importance to safeguard judges’ discretion when performing their duties, in order for pioneering solutions to emerge. Judges’ liability for judicial errors renders them vulnerable and may be thus seen as a threat to judicial independence.²⁷

Moreover, it is generally accepted that judicial proceedings should not be lengthy and protracted. Judicial liability is confronted with skepticism as it may lead to endless proceedings, thus endangering the concept of “res judicata”, which is essential for the preservation of legal certainly and social peace.²⁸ Closely connected with this last point is the argument that liability for judicial errors would undermine the authority of Justice as such. Courts and judges must be respected in order to able to fulfill their task effectively. Thus, concealing judicial errors is thought to have an inherent expressive function from a societal viewpoint.²⁹

A regime of liability for judicial errors should strike a balance between all the above variables. Without doubt, the discretion of the judge in deciding a case is to be preserved, but, in view of its rationale, its exercise should remain within reason; it may not be exercised in a way which would override its purpose. As regards the concerns about the violation of the res judicata and the authority of justice, we believe that they are of less significance. Even if a judge is held liable for his behavior, this does not necessarily need to lead to the review of the judgment; the remedy of damages does not impair the effect of the initial decision (Grunsky 1974: 150–151; van Bogaert 2006: 198). Finally, the social value of tolerance towards judicial errors should not be overestimated. As long as people are individually dissatisfied by the administration of justice, its authority can hardly be retrieved. As Alexander Hamilton put it over two centuries ago, “[o]ffences which proceed from the misconduct of public men, or, in other words, from the abuse or violation of some public trust … relate chiefly to injuries done immediately to the society itself”.³⁰

4 The concept of “Inexcusable Judicial Error”

In view of the above considerations, the description of a single desirable “socially efficient” behavior of judges is not only elusive, but it is also of questionable legitimacy. If every deviation from this standard would signify judicial liability, judicial independence would be at stake. Hence, in order to draw a line between judicial errors which may be excused and those which give rise to issues of judicial responsibility, we opt for the opposite goal: we aim at identifying the cases where the act (or omission) of the judge could be admittedly qualified as unacceptable.

Indisputably, intentional behavior on the part of a judge (e.g. bribery) which leads to the infliction of illegitimate harm to a litigant falls under the category of “inexcusable judicial error”. When the judicial error is due to negligent behavior the question is more complicated.³¹ Negligence is commonly defined as a deviation from the behavior of a reasonable person (Black’s Law Dictionary 2004). Thus, a negligent judge is one who acted in a way that an average hypothetical judge would not. In view of the considerations of judicial independence, a slightly negligent behavior may not be characterized as inexcusable. This does not hold in the case of gross negligence, i.e. when deviation of the behavior of the judge in question is beyond the ordinary. In this context, an error would be qualified as grossly negligent if a hypothetical average conscientious judge would never have committed it.³² Nevertheless, the assessment of the behavior of a non-existent hypothetical average judge is necessarily based on speculation, while the proof of culpability and the distinction among its degrees are thorny in practice.

The abovementioned problem could be set aside if we follow a more objective approach which would focus on the outcome of judicial behavior. Adopting the recent position of the French Cour de Cassation (in plenum),³³ we consider a judicial error to be inexcusable when it leads to a “defective judicial service”.³⁴ Judges render “defective judicial services” when they fail to accomplish their assigned task, namely to resolve a legal dispute in a manner that ensures social peace.

In this sense, an “inexcusable judicial error” would fall within (at least) one of the following categories:

• Denial of justice or inordinate delays in pending cases.³⁵ Although judges cannot as a general rule be “held accountable” for their excessive caseload and several other factors are at play for this particular problem, we think that individual judges can be better or worse managers of their own case-loads. Judges do not possess the same level of multi-tasking skills or time-management skills. Thus, they may not have much choice in the number of cases assigned to them, but may be more (or less) skilled and efficient in their tasks.

• Serious violations of the procedural rules which influence the outcome of the trial (e.g. violation of the right to be heard, illegal rejection of the submission of important evidence, failure of the judge to disqualify himself because of conflict of interests etc). The assurance of a fair trial constitutes the absolute minimum possible professional standard for judges. Excess of jurisdiction would also fall within this category.

• Grave legal errors in the judicial opinion itself. Such errors may regard the implementation of the wrong law (e.g. implementing an important law that had been abrogated, especially if the attorney had mentioned the new law) or its erroneous interpretation. An interpretation may be considered erroneous if it is incompatible with the letter of the law as well as the unanimous opinion of legal theory and case-law.³⁶ This category would also comprise cases of violation of obligations arising out of European law.³⁷

• Grave error in the evaluation of the facts of the case in the judicial opinion, especially when the facts are so erroneously taken into account that the decision seems to refer to a different case (e.g. “mistaken identity” cases, in which the judge confuses the facts of different cases) or when crucial facts are disregarded without adequate reason and against the teachings of experience³⁸ (e.g. awarding physical custody of the child to the mother, who is extremely aggressive due to a chronic, severe mental illness and ends up killing the child³⁹).

• Abuse of power (e.g. determining maintenance after divorce at an amount which is manifestly well beyond the ability of the defendant to pay⁴⁰). The abuse of judicial discretion often demonstrates the judge’s lack of impartiality. Thus, cases of corruption fall within this category.

In addition, as the French Cour de Cassation has ruled in 2001,⁴¹ the judicial service may also be considered defective if the judgment in question contains so many minor errors that the faith of the litigant in the smooth functioning of justice is shattered.

“Inexcusable judicial errors” will in practice arise out of the judge’s grossly negligent or intentional behavior. In any case, the concept of “inexcusable judicial error” can become particularly useful as a professional standard for judicial behavior, since it provides a more objective set of criteria for determining when a judge has not acted with due diligence. Thus, it provides a low-cost mechanism that allows the legal system to detect the cases which give rise to judicial liability. Finally, it has the added advantage of flexibility. Indeed, “inexcusable judicial error” may be applied in both civil and common law countries, adjusted according to the local legal culture, the roles of judges in each legal system and the overall civic development of each society.

5 Remedies for inexcusable judicial errors

The present section of the paper focuses on the civil liability for damage caused by judges in the exercise of their duties. However, the first step of the analysis is to establish the necessity of civil liability within a broad-based system of judicial accountability for acts of misconduct. We do so by discussing its main alternatives, namely the right to revise the judgment, the political accountability of judges, as well as their criminal and disciplinary (or quasi-disciplinary) liability.

5.1 Review of the judgment

The appearance of the institution of appeal is correlated with the development of the doctrine of judicial immunity in English common law: since litigants could appeal against a decision which they considered erroneous, there was no longer a need for litigation against the judge (Comisky and Paterson 1987: 233).

Indeed the right of appeal is granted in most jurisdictions against most judgments.⁴² This alternative presents the advantage that the initiative for launching the process belongs to the litigants, who are better informed on the occurrence of an error (Shavell 1995: 381).⁴³ Ideally, the review of the decisions of a certain judge would impair his reputation, thus providing him with incentives to perform his duties in a diligent way (Shavell 2006).

Nevertheless, we believe that the deterrent effect of appeal is weak, for the following reasons: First, it is not at first glance apparent how many decisions of each judge have been reversed. Even if it were, this would not necessarily signal the inadequacy of the judge, since judgments may be reversed also in cases of minor defects. In effect, the fact that a judgment has been reversed does not even mean that the initial decision was an erroneous one; its review could be simply due to the difference in legal, or even political, preferences between the judge who issued the initial judgment and the one who revised it (Cf. Spitzer and Talley 2000: 649).⁴⁴ Moreover, even if it has become clear that a specific judge is unscrupulous, it is rather unlikely that any action (e.g. disciplinary proceedings) will be initiated against him or her.⁴⁵

Finally, it is worth noting that not only does the possibility of review itself lack a deterrent effect, but in the absence of an actually functioning incentive scheme (e.g. liability), it may even lead to an increase of flawed decisions: judges of first instance courts, who have a significant case load to sift through, know that even if their decisions are erroneous, their “faults” can be corrected by higher courts. This may relieve them from a possible guilty conscience.

5.2 Political accountability

The political accountability of elected officials can be traced back to classical Athens, where the standard penalties of officials were impeachment (“eisaggelia”) and political trials in the Assembly (Hansen 1999). The standard model of accountability is construed in the principal-agent framework.

Political accountability is a classic remedy for judicial misconduct in the United States, where state judges, unlike federal judges, do not enjoy a life-tenured position. State judges hold their office for an initial period of time, after which they must be re-elected (or re-appointed) in order to keep it (Haley 2006: 281). Nevertheless, political accountability is a remedy of limited application. In order to guarantee judicial independence, a judge has to be appointed for a rather lengthy initial term in office (Schuck 1989: 672).⁴⁶ Moreover, this remedy is not reliable, especially in cases where judges have to be re-elected. Non-lawyers lack the knowledge to understand and evaluate the performance of a judge, whereas pre-election periods bring about a substantial waste of resources.

In the United States political accountability inevitably leads the discussion to impeachment. Impeachment is an extraordinary remedy which aims at the removal of the judge from his office, against his will. However, impeachment proceedings as they apply to the judiciary are not political in nature. “There is no basis to interpret the Constitution to allow the removal of a judge for political reasons. To do so would be the antithesis of creating and sustaining an independent judiciary” [Hastings v. United States of America, US District Court, District of Columbia, 802 F.Supp. 490]. Nevertheless, in the case of impeachment, judicial accountability is in practice confounded with political accountability. More concretely, to be impeached, a life-tenured judge must be brought up on real charges, i.e., “high crimes and misdemeanors”, and receive a real trial before the full Senate as clearly required by the Constitution. A problem that arises in this context is defining “high crimes and misdemeanors”: should judges or other officials be impeached only when they have broken written, statutory law—or do certain “moral indiscretions” qualify as impeachable offenses? Historically, the term “high crimes and misdemeanors” included non-statutory misdemeanors. Alexander Hamilton and Judge Joseph Story defined “misdemeanor” as “political malconduct”.⁴⁷ Therefore, political considerations filter into the process at all levels.

5.3 Liability

The inadequacy of procedural and political tools to provide incentives for the elimination of “inexcusable judicial errors” points to the necessity of a system of judicial liability. Judicial liability may be criminal, disciplinary or civil. In the sections that follow, we briefly discuss criminal and disciplinary proceedings against judges. Then we take a closer look at the possible systems of civil liability of judges and finally we evaluate them focusing on their incentive potential.

5.3.1 Criminal liability

Notwithstanding certain procedural constraints on the initiation of criminal proceedings against a judge in some legal orders, judges are in principle liable for the offences they commit when performing their duties (Goré 2007: 8).⁴⁸

The specific criminal provisions judges may fall within while performing their duties vary considerably among different legal orders. Such crimes may pertain, among others, to bribery (meaning here the acceptance of an illegitimate advantage),⁴⁹ abuse of power,⁵⁰ obstruction of justice,⁵¹ illegal deprivation of freedom,⁵² denial of justice⁵³ or generally violation of duties.⁵⁴ Some of these offences can be committed only by judges (e.g. denial of justice) while others can be committed by any public servant (e.g. bribery, abuse of power, violation of duties) or more generally by any person (e.g. illegal deprivation of freedom, obstruction of justice). In any case the judicial status of the actor may be considered an aggravating circumstance.⁵⁵

Anyhow, all abovementioned offences have one common element, namely that they require intentional behavior on the part of the actor. Without doubt the criminal liability of judges is a powerful incentive tool. However, its scope of application is limited since it focuses on cases of blatant violations and thus fails to cover all instances of “inexcusable judicial errors”.

5.3.2 Disciplinary liability

Disciplinary liability is imposed on judges by means of internal proceedings and aims at the maintenance of a minimum professional standard. The particular violations which may give rise to disciplinary liability are usually not described in detail (Goré 2007: 17). In general they pertain to deviations from judicial “deontology”, which may be codified in a Code of Judicial Conduct. Thus, disciplinary liability can be seen as a mechanism for self-protection that the judiciary uses to maintain its prestige. The range of penalties which may be imposed on a judge is wide; it varies from admonition or imposition of a fine to temporary or even permanent removal from his office (Goré 2007: 17–18).

The scope of disciplinary proceedings is considerably broader than that of criminal proceedings. Nevertheless, the deterrent effect of disciplinary liability is in practice weak. First of all in legal orders where the initiation of the procedure rests on the initiative of the Minister of Justice or the magistracy itself,⁵⁶ the procedure is not likely to be initiated. The disciplinary system may be more successful if individuals can lodge complaints against judges to independent authorities (e.g. Ombudsmen in Israel and in Finland).⁵⁷ Even in these cases however, it could be reasonably expected that the instances of conviction would remain rare due to “professional solidarity”.⁵⁸ More often than not, the body which has the competence to decide is comprised, solely or in majority, of judges.⁵⁹

An additional effect that “dilutes” the practicability of disciplinary liability is that even when the relevant procedure is launched and the judge is held liable, the reputational costs of this decision remain, in principle, low: with the exception of dismissal, any other penalty imposed against the judge does not really have an impact on his reputation, because of the confidentiality or in any case because of the limited publicity and the lack of transparency of the procedure (Schuck 1989: 669). Thus, unless judges face the risk of disqualification, which arises in a few, extreme cases, the penalties imposed on them are usually insufficient to provide them with sufficient incentives to comply with the rules.

The abovementioned considerations on disciplinary liability apply also in the quasi-disciplinary procedure of “impeachment”.⁶⁰ The reluctance to impeach judges is a clear indication of how politics play into judicial activities. Thus, life-tenured judges, as “civil officers of the United States”, have been subjected to the impeachment process since 1804 when Judge John Pickering was impeached (see esp. Rehnquist 1992: 127). Since that time, and until 1960, there were only 52 impeachment proceedings at the state level and only 19 judges removed (Brand 1960: 1315). At the federal level, the numbers are even smaller: some 13 judges have been subjected to impeachment and trial and of those only seven were convicted (Haley 2006: 287).

5.3.3 Civil liability

The issue of civil liability for judicial errors is very controversial, since different legal orders deal with it in a very diverse manner. Depending on whether the judge and/or the state are held liable for the damage caused, the following four patterns of rules may be distinguished:

5.3.3.1 No remedy against the judge or the state—the US law

In the United States judges enjoy absolute immunity from civil claims (see, among others, Shaman 1990; Haley 2006: 283). The historical roots of the doctrine of judicial immunity go back to England: since the King can do no wrong, neither could the judges of the King’s Court, who were the delegates of the royal power. Gradually this privilege expanded to judges of other courts as well, not only in England but also in other common law countries.⁶¹

In the United States, the High Court clarified judicial immunity in the landmark decision Bradley v. Fisher, 80 US (13 Wall.) 335, 20 L. Ed. 646 (1871). Some hundred years later, we find an outrageous case of judicial immunity in Stump v. Sparkman, 435 US 349 (1978).⁶² In this case a mother petitioned the court seeking the sterilization of her 15 year old daughter, on the grounds that the girl was “somehow retarded” and had started dating men. The judge ordered the sterilization of the girl without having heard her or having appointed a guardian ad litem to represent her interests. The young woman, who was not retarded after all (as shown by her school records), found out about the sterilization two years after her marriage, when, in her efforts to conceive a child, she visited a doctor. She brought a claim against the judge who gave permission for her sterilization, but her claim was dismissed on the grounds of judicial immunity.

Although it has been heavily criticized, the doctrine of judicial immunity is well established. In the course of the years the Supreme Court has identified few exceptions to judicial immunity, especially when the act of the judge does not constitute a “judicial act”.⁶³ Moreover, judges are not absolutely immune from declaratory and injunctive relief. These forms of relief require parties to do or refrain from doing a certain thing. If a judge loses a suit for declaratory judgment or injunctive relief, the judge may be spared from paying money damages, but may have to pay the court costs and attorneys’ fees of the winning party.⁶⁴

Finally, in cases of judicial misconduct the injured party may not recover damages from the state. This result is mostly derived by the doctrine of sovereign immunity, according to which the state may not be sued without its consent (Schuck 1989: 667).⁶⁵

5.3.3.2 Remedy only against the state—the French law

According to article L. 141-1 of the new French Code of Judicial Organization,⁶⁶ which was introduced by Ordinance 2006-673 of June 8, 2006, the state has to compensate the damages caused by defective judicial services, namely in cases of grave judicial error⁶⁷ and denial of justice.⁶⁸ Furthermore, Art. 11-1 of the ordinance of December 22, 1958 on Judicial Status, as amended by Law 43 of January 18, 1979, provides that judges may not be sued personally for errors they commit while performing their duties. However, in case the state compensates the injured party, it has a right of recourse against the judge who issued the flawed decision. In practice, however, such recourse action is hardly ever filed (Canivet and Joly-Hurard 2004: 25). At any rate, Art. 22 of the Organic Law of March 5, 2007 stipulates that the cases in which the State affords damages to individuals because of defective judicial services become public; every year the Government shall submit to the parliament a relevant report. This system presents the advantage that the victim is compensated without the dispute being personified, and is thus thought to preserve judicial independence.⁶⁹

A somehow similar regime has been recently adopted in the Netherlands. Since January 1, 2001, the personal liability of judges has been abolished. Thus judges enjoy immunity from civil liability (van Bogaert 2006: 195–196).⁷⁰ The injured party may file an action for damages against the state. Nevertheless, this action succeeds only in cases of violation of the principle of fair trial, in the sense of Art. 6 ECHR (id. at 202). The state has no right of recourse against the culpable judge, who may be held liable only on criminal or disciplinary grounds. Interestingly this reform has been introduced without controversies (id. at 196-197).

5.3.3.3 Remedy only against the judge—the Greek law

Under article 99 of the Greek Constitution, a litigant who suffered damage because of an erroneous judgment may sue the culpable judge personally before the Special Court for Mistrial, which is comprised of three judges, two professors of law and two lawyers. According to article 6 of Law 693/1977 judges are liable only for damages caused by grossly negligent or intentional acts, as well as in cases of denial of justice, including inordinate delays. A claim against the judge can be filed only if there are no other means of reversing the flawed judgment, within six months from the time when the litigant became aware of the illegal act or the flawed judgment became irrevocable. In practice, the suit filed by the plaintiff very rarely succeeds. In spite of the fact that judges do not participate in majority in the Court for Mistrial, the other participants, and especially the lawyers (Kassimatis 1973: 103–106), are reluctant to “convict” a judge to pay damages.

As regards state liability for wrongful judicial acts, according to the case-law, judicial errors do not give rise to state liability⁷¹ on the grounds that since judges are independent, the state may not be held liable for their acts. This argument has been criticized by most legal theorists, who claim that the guarantee of judicial independence is meant to ensure the impartial administration of justice and not to protect the state from liability claims (Kassimatis 1973: 108–109; Makridou 2006: 157, with further notes).

5.3.3.4 Joint and several liability of the judge and the state—the German law

According to article 34 of the German Constitution, in combination with § 839 of the German Civil Code, the plaintiff can sue both the state and the judge for the damages suffered because of judicial misconduct. As regards the conditions under which this liability arises, it is important to make the following distinction: If the (flawed) judicial action led to the issuance of a “judgment in a legal matter”⁷² the judge or/and the state may be held liable only if this act constituted a criminal offence (§ 839 [2]). In the rest of the cases, which also include denial of justice or unreasonable delays, the judge is liable in case of gross negligence.⁷³ In any case the lawsuit of the plaintiff is dismissed if there are other legal remedies (e.g. the right of appeal) which could make up for his loss. If the state compensated the injured party it has a right of recourse against the culpable judge.⁷⁴ In practice, recourse actions against the judges are rare (Wagner 2006: 8).

5.3.3.5 Comparative remarks—De Lege Ferenda thoughts

From the abovementioned analysis it has become obvious that different legal systems deal with judicial misconduct in very diverse ways, especially as far as civil liability is concerned. The provisions of each legal order can only be evaluated as a whole. Thus, it is not possible to conclude categorically which system is preferable than the other. We believe, however, that no absolute immunity should be granted to judges. A system of civil liability should play a dominant role in the structure of the relevant incentive scheme, because it presents significant advantages when compared to other alternatives as regards deterrence:

First of all, the system of civil liability is decentralized, meaning that the proceedings may (and often will) be initiated by private parties. The danger of frivolous law suits may be mitigated by means of provisions of “qualified procedural judicial immunity”, meaning procedural rules which lead to the “screening” of the cases filed (cf. King 1978: 589), thus increasing the costs of the action.

Second, the magnitude of the deterrent effect of a system of civil liability depends on the specific structure of the liability regime. Apart from the conditions under which civil liability should be admitted, which in our view should be the cases of “inexcusable judicial error”,⁷⁵ it is of crucial importance to consider whether the judge will be held personally liable or not. In case the state compensates the victim, the deterrent effects of civil liability are maintained, provided that the state has the right of recourse against the responsible judge and actually exercises this right. Even if the state does not have, or does not exercise, such a right of recourse against the culpable judge, the liability regime may still lead to efficient results in terms of deterrence, if the state or the competent disciplinary body intensifies its control over the judge in question. This is more likely to happen if compensation is paid to the victim not from the general state budget, but from the budget of a smaller administrative agency, e.g. the budget of a judicial authority (Schuck 1989: 668). In other words, under the said condition a regime of civil liability of the state may enhance the effects of judicial disciplinary liability. At this point it is worth noting that according to an opinion, because of the publicity and the transparency of civil litigation, an action for damages caused by judicial error may, to some extent, have a deterrent effect in itself. This argument would explain the number of civil claims filed against judges in the United States, although there can hardly be any doubt that they will be dismissed (Haley 2006: 286–287).

Third, an important issue to be examined is whether the civil action for damages on the basis of civil liability of the judge should be contingent upon the possibility of judicial review, or, more generally, upon other ways to mitigate the damages caused. Indeed, judicial review of a judgment comes at significantly lower costs and may do away with the harm caused. However, we believe that this should not constitute a ground for the extinction of the liability of the judge who gave the initial decision because it would actually render the judges of first instance courts immune to liability and would thus eliminate its deterrent effect. Even if an appeal is filed and it goes through, this comes with additional costs, so damage persists (van Bogaert 2006: 198). In any case, the application of the doctrine of comparative negligence could provide incentives to the victim to mitigate his loss by making use of all procedural tools offered to him.

Last but not least, the civil liability of judges not only provides them with incentives to behave diligently, but it also has a positive side-effect: the victim is compensated. From this point of view, it could be argued that because of the satisfaction of all involved parties, “social peace” is better served. Compensating the victim satisfies the conditions of economic efficiency, but also of social effectiveness, as it introduces an instrument for more effective social control. By improving the mechanisms for the evaluation of judicial behavior, we are contributing to the construction of a more ordered normative universe, with greater stability and predictability.

6 Conclusion

In this paper we have set forth the idea that judges are satisficers who adjudicate under the constraints described by the bounded rationality model. Further constrained by other exogenous factors (like the time pressure brought on by congested court dockets and the increasing complexity of the cases that come before them), judges cannot serve their principals (the public) at an optimal level. The best that they can do thus becomes “the best that they can do under the circumstances”, which is only good enough, evaluated ex post using criteria like the frequency of judicial errors. In an ideal situation, judges would be perfectly impartial and render decisions under perfect information. However, in recent years adjudication has become more complicated and more permeable to extra-legal influences than ever before. The sources that inform judicial decision making today are multiple and often have a cumulative effect on judges. They are required to process and filter large amounts of information in limited periods of time. In this light, it becomes clearer why judges may engage in behavior that is “improper”. Judges are satisficers not because that is their primary goal. We think that few judges perform at suboptimal levels because they choose to pursue private interests, for example political ambitions. Most judges become satisficers along the way, because satisfying is the best survival strategy in an inefficient system which provides them with disincentives to maximize.

Understanding judges’ incentives in the prism of economic analysis allows us to make normative suggestions about how the state can act upon them. Our concept of “inexcusable judicial error” is a viable suggestion that has many advantages: it has wide applicability in civil litigation, it could serve as a guidepost especially for first instance court judges and it provides a unifying treatment of judicial liability for both Anglo-American and Continental law. The differences in the role of judges across legal systems can be dealt with by providing different interpretations of “inexcusable judicial error”. Our proposal could thus play a crucial role in shaping a new set of professional standards.

Once these professional standards are set, there are numerous incentive schemes in order to induce the judges to keep to the standards. The possibility of judicial review, political accountability, as well as judicial liability (criminal, disciplinary or civil) may contribute to the deterrence of judges. As shown in Sect. 5, each legal order has adopted a different combination of the abovementioned measures, associated with its own advantages and disadvantages. In this paper we do not claim to have modeled the optimal incentive scheme for judges. However, we do believe that civil liability should not be absent from this scheme.