It Takes Three to Tango: A Behavioral Analysis of the Benefits of Having a Mediator in International Disputes

Abstract

This article explores the debiasing role of the mediator in the setting of international disputes. Starting from the rational choice theory, this article examines the choices that international disputants must weigh when deciding how to proceed with their conflict. International actors are assumed to be rational and negotiate settlements that are beneficial to them. Their alternatives to a negotiated settlement include both adjudication and continuation of the conflict. Puzzlingly, despite the advantages of negotiated settlements, many international disputes are not resolved in the form of a settlement. Instead, States seem to prolong conflicts or follow costlier routes of formal adjudication. Behavioral Law and Economics insights on the biases of disputants partly explain this phenomenon. The article contributes to the behavioral discussion by examining two separate categories of biases, i.e., biases when deciding to enter negotiation and biases during negotiation. Following that, it suggests that the specific characteristics of the process of mediation and of the mediator, in particular, can act as debiasing instruments. The article concludes with normative suggestions for wider incorporation of mediation within the international dispute settlement setting.

A. Introduction

Negotiation is considered the first step that two parties take following the emergence of a dispute between them. Ranging from a simple discussion to an offer-counteroffer exchange, negotiations help parties communicate their views, establish the limits of the dispute and search for viable solutions.¹ Nevertheless, it has been observed that negotiation impasses happen frequently, even when the proposed settlement agreements leave both parties better off, compared to an adjudicated outcome.² The reasons for this phenomenon have been widely discussed in the literature, and scholars have identified a range of behaviors leading to destructive results. Specifically, those behaviors are driven by distinct heuristics that humans employ when faced with non-intuitive, difficult decisions. It is these heuristics that lead to biased judgments and erred decisions. Many of these pitfalls associated with negotiation can, however, be avoided through the use of another tool for peaceful settlement of disputes: Mediation.

Mediation is not a new concept that emerged the twentieth century in the United States during the twentieth century.³ Instead, mediation has been around in similar forms in ancient civilizations ranging from China to Africa.⁴ Mediation also has contemporary forms in other cultures, such as gacaca ⁵ in Africa or shuras ⁶ in Afghanistan. Besides its use in private disputes of a civil and commercial nature, mediation has been prevalent on the international legal level⁷ alongside diplomatic approaches⁸ and good offices––diplomatic means for the dispute of settlements.⁹ More recently, there is a tendency in the international sphere towards the judicialization of international legal disputes.¹⁰ However, this does not make mediation obsolete: Several authors¹¹ claim that mediation continues to evolve and can contribute to the resolution of international disputes in terms of increasing peaceful settlements and employing alternative methods.¹²

International Law scholarship has been introduced to the field of rational choice¹³ as well as the Behavioral Law and Economics.¹⁴ In this article, it is assumed that a disputant is a state or any other participant to an international dispute, be it an investor, Internally Displaced People (IDP), or armed group. While acknowledging that many factors generate states’ preferences,¹⁵ this article uses the black-box assumption¹⁶ and presents them as unitary actors.¹⁷ In order to apply rational choice theory, it is assumed for the purposes of this article that the preferences of the disputants are constant. However, after reviewing some shortcomings of rational choice theory, the analysis will borrow insights from the behavioral Law and Economics literature to explain why amicable settlements are not that frequent and negotiation impasses may occur.

This article aims to explain how and why international mediation can contribute to the international dispute resolution toolbox from behavioral law and economics perspective. Section B sets the stage for this endeavor by defining international mediation, different categories of international disputes and conflicts, which the international mediator is, as well as the current stage of incorporation of mediation into practice. Section C then presents the general behavioral law and economics approach to dispute resolution, as well as the standard economic theory predictions of why negotiations fail. Behavioral biases that can occur before and during negotiation are presented in section D. Section E assesses various techniques of the mediator to remove biases from the parties suffering from such negotiation and decision-making biases. The article concludes with Section F in a discussion addressing the normative outlook on the current and future state of international mediation.

B. Defining International Mediation

This section delimitates international mediation from other similar international dispute resolution processes that involve a third party. It also presents some basic divergences from domestic mediation. Lastly, informative limitations of international mediation are discussed which are relevant to the subsequent behavioral analysis of biases.

I. Definition of International Mediation

International mediation can be defined as a voluntary form of conflict management embedded in the international relations domain and dominated by the principles of preservation of actors’ independence and autonomy.¹⁸ There are two adjacent concepts within international law, good offices and conciliation, that share some commonalities with mediation. Good offices involve a third party that attempts to build a communication channel between the two sides and prompts them to resume negotiations.¹⁹ Conciliation mostly takes place when a conciliator leads an independent search into the arguments and information gathered by each disputing party.²⁰

II. Differences from Domestic Mediation

While international mediation shares many commonalities with domestic or civil mediation, it is important to distinguish these two types of mediation, as they are not always governed by the same principles due to the negotiation dynamics being different.²¹ One characteristic divergence from the traditional, domestic mediation process is the motivation of the international mediator. Due to the high-risk position of mediating between states or international players with various power dynamics, a mediator in the international setting must be incentivized with elements of self-interest to be involved in the case.²² At the same time, in order to be effective in conflict management, an international mediator must obtain some type of leverage in order to move a party in the desirable direction.²³ Studies have also found that a mediator within an international mediation can be biased without detrimental effects on the results.²⁴ Experimental studies²⁵ support that a biased mediator does not affect the efficiency of the settlement outcome.²⁶ With that being said, impartiality and neutrality are values that a mediator is expected to demonstrate in civil law mediation.²⁷

The mediator in an international mediation can be any individual,²⁸ or multiple such persons, multiparty mediation.²⁹ It can also be an individual or a group of individuals holding office in another state, be it a bordering state or a state with close connections to the conflict, or representatives of Non-Governmental Organizations.³⁰

On the contrary, the profession of a domestic mediator is explicitly regulated.³¹ A mediator of domestic disputes is usually a trained individual with a professional background in law, psychology, sociology, or also medicine or engineering. In order to become an accredited or certified mediator, the individuals have to attend training seminars or sit exams that aim to educate them on the foundations of negotiation, human behavior and psychology as well as crisis management.³² Depending on the type of mediator, various success rates are documented.³³ For instance, representatives of small governments fare the best with a 56.8 percent success rate. Representatives of regional organizations rank as the runner-up with a 50 percent success rate. The lowest success rate is observed for representatives of international organizations faring only 23.8 percent.

As for the types of disputes in international mediation, the three most prominent categories are international economic disputes, human rights’ disputes and border disputes.³⁴ One can incorporate into the analysis both interstate and intrastate conflicts, the latter usually occurring between the government and a group of people within the country.³⁵ The subsequent behavioral analysis conducted in this article theoretically encompasses every type of inter- and intrastate dispute, with the disclaimer that detailed behavioral analyses for each type of dispute must form part of separate papers.

III. Limitations of Mediation

Some caveats are necessary before proceeding to the main analysis. On the one hand, the scope of the analysis is strictly constrained to behavioral biases before and during negotiations. Therefore, other negotiation failures that may result from strategic behavior and game-theoretical insights will not be addressed.

On the other hand, it is important to note that mediation is not an infallible and flawless procedure; on the contrary, it has certain limitations when it comes to international conflicts. First, mediation’s outcome is unable to create case law and opinio iuris as the mediation sessions and the agreement are bound by confidentiality principles.³⁶ This means that rights and duties that form part of previously mediated agreements are not accessible by future disputants.

A second limitation revolves around the role of legal culture within a state or a community. There are different conflict cultures in various parts of the world and this characteristic drastically affects the perception of and approach toward mediation.³⁷ If the approach of the conflicting parties and the approach employed by the mediator of choice vary greatly among each other, this might limit the capacity and effectiveness of international mediation. Furthermore, some fear that mediation as a process focuses on problem-solving notions that, if generalized, could lead to a dispute resolution culture of colonization.³⁸ The cultural aspect is of high relevance to the possibility of entering or staying in international mediation.

An indispensable prerequisite of mediation is the consent to enter or stay in mediation.³⁹ The prospects of obtaining that consent can differ between economic disputes, where a contractual clause is often the basis for mediation, and human rights or border disputes. The latter types of disputes pose more difficulties because the parties to the dispute have to justify politically their openness to mediation while trying to save face.⁴⁰

C. Behavioral Law and Economics Approach

This section introduces the theoretical framework of the article. The behavioral law and economics approach depart from the standard economic theory of behavior. The decision to negotiate or to go to court has been consistently predicted by classic economics to follow a rational cost and benefit assessment whereas behavioral law and economics introduce the notion of biased disputants whose assessment may be affected by cognitive shortcomings. Subsequently, this assessment is applied to the framework of international disputes accordingly.

I. Background of Behavioral Law and Economics

The behavioral law and economics approach signifies a departure from the standard economic theory of the homo economicus who operates in line with cost and benefit functions and under strict monetary constraints.⁴¹ According to the traditional law and economics literature, an individual will comply with the law as long as the benefits of compliance outweigh the costs of non-compliance.⁴² The literature stream of behavioral law and economics⁴³ introduces the notion of cognitive shortcuts⁴⁴ or rules of thumb⁴⁵ that affect the strict cost and benefit calculations of an individual.

These mechanisms, also known as heuristics, are employed by individuals when facing demanding and complex decisions. Heuristics are not per se negative since they serve to relieve the human cognitive system from unnecessary functions,⁴⁶ though, their application may lead to systematic errors in one’s behavior. To err is human, yet the standard economic theory assumes that people will not err in the pursuit of maximizing their goals while weighing the costs and benefits of each action. Individuals must be biased in their reasoning, thinking or decision-making for an error to result.⁴⁷ This means that heuristics cultivate certain biases that may or may not affect the judgment and decision-making of humans in departing from optimal behavior.

II. Standard Economic Theory in the Context of Disputes

Before discussing the decision-making biases in the international dispute resolution context, it is necessary to examine the law and economics predictions of the disputants’ behaviors.

It is hypothesized⁴⁸ that a party to a conflict will only settle for an amount that it values equal or higher than the expected litigated outcome. To evaluate the expected outcome, the claiming party will have to calculate the probability of winning the claimed amount, add the probability of losing in court and subtract the total costs of going to court. Inversely, the rational defending party will calculate the reversed function, i.e. the probability of paying the full claimed amount in addition to the probability of winning the case and not paying minus the total litigation costs. After these calculations, a bargained settlement between the two disputants is possible if there is a price range between the lowest amount that the potential plaintiff can accept and the highest amount that the potential defendant may offer, also known as the zone of possible agreement (“ZOPA”). The lowest and highest price points are dictated by the expected outcome that the disputants may alternatively receive in court.

Standard economic theory has predicted two economics related reasons for negotiation failure,⁴⁹ namely imperfect information⁵⁰ and risk of opportunistic behavior and strategic bargaining.⁵¹ The behavioral theory predicts that any heuristic employed by the disputing parties in order to calculate the expected outcome in court could potentially lead to a biased assessment.⁵² In turn, this assessment might minimize the ZOPA so that parties cannot successfully negotiate, and go to court instead.⁵³ Lastly, a separate factor that may influence the disputants’ expectations is their counsel, who, according to studies,⁵⁴ can also suffer from biased judgment regarding the potential litigated outcome.

III. Applying Standard Economic Theory to the Context of International Disputes

As discussed above, the context of international disputes differs to a certain extent from that of civil and domestic disputes. It is not always possible to make good on the threat of a day in court,⁵⁵ especially in territorial or human rights disputes. On the other hand, the situation is different for international economic disputes, where subsequent arbitration or forum-adjudication is often foreseen as part of the pertinent treaty or investment/trade agreement.

This section of the article expands the standard economic theory of the behavior of litigants when facing a dispute. For international disputants, the alternative to a negotiated settlement is either a mutually accepted formal mode of dispute resolution, be it arbitration, adjudication, or the continuation of the conflict. In essence, the disputing party will only settle for an agreement that is equal or higher valued than either the expected outcome of a formal dispute resolution process, including the relevant costs of the process, or the benefits minus the costs of the continuation of the conflict.

As for the adjudication alternative, the costs and benefits are similar to the theorized model of litigation. When it comes to the conflict continuation alternative, the relevant benefits may be linked with reputation,⁵⁶ exertion of power and monetary gains. The costs could include costs of armed conflict, costs of reparations for losing the conflict, or even economic sanctions⁵⁷ by the international community. Again, the perception of the costs and benefits as well as the probabilities of the expected outcomes can be influenced, according to the behavioral approach, by heuristics and biases so that negotiations fail, and settlement is not pursued. These specific biases embedded in the negotiation context are analyzed in the next part.

D. Behavioral Biases in a Negotiation Context

The negotiation setting is critical as all participants present their views, preferences and demands referring to the emerged dispute. In legal disputes specifically, i.e., in disputes where parts of the dispute concern legal provisions in the law or in a contract, negotiation is considered the pre-trial stage. This is not necessarily the same for international disputes, as for instance border disputes or intrastate conflicts may lead to violent means of resolution or simply make it impossible for the parties to sit at the bargaining table altogether.

This article distinguishes for the first time in the literature⁵⁸ between two different categories of biases in the negotiation context, first, biases when deciding to enter negotiation, and second, biases during negotiations. This distinction is justified because it considers the extended options a disputant has, namely, to avoid the resolution of the conflict, negotiate with the other side, or pursue a formal adjudicatory process.

The first category refers to the decision a disputant faces between entering negotiation or following a different route, be it litigation or continuation of the conflict. An erred assessment of the expected outcome may drive a disputant to ignore negotiation initiatives and continue with litigation or prolongation of the conflict. Consequently, the first category incorporates all behavioral biases that operate on the prediction of outcomes. The second category describes what happens during the decision-making process at the negotiation table. Because opposing disputants have to come to a mutual decision, the complexity⁵⁹ and thus the susceptibility to cognitive biases⁶⁰ increases. In this category, the analysis deals with behavioral biases which occur when one is faced with a hostile opponent or with a less than optimal offer. Disputants apply differently heuristics in these two different decision-making settings and thus, different cognitive biases emerge.⁶¹ The following analysis focuses on the biases found in these two different settings⁶² and is supported by experimental findings.⁶³

I. Biases When Deciding Whether to Enter Negotiation

1. Conjunction Fallacy

The conjunction fallacy is attributed to two different heuristics, the adjustment and anchoring heuristic⁶⁴ and confirmation heuristic.⁶⁵ It describes the wrong belief that the probability of the occurrence of an event composed of two independent events, conjunctive is higher than the probability of the occurrence of one of the independent events. Consequently, individuals tend to overestimate the probabilities of conjunctive events such as winning in court and underestimate the probabilities of disjunctive events such as the chain of actions⁶⁶ leading to the positive judgment of the lawsuit.

2. Insensitivity to Predictability

The insensitivity to predictability is linked with the representativeness heuristic⁶⁷ and refers to the situation where the probabilities of an event happening are only assessed by considering the favorable information and evidence.⁶⁸ In other words, a disputant assesses the stakes of winning in lawsuit by only seeing the favorable evidence and ignoring the other side’s arguments and respective evidence. In sum, this bias increases the optimism of disputants and will likely induce them to make a risk-prone choice based on a false probability assessment. The absence of arguments or information from the other side is common in international disputes, caused by hostile feelings and avoidance of channeling information to the other side.

3. Insensitivity to Base Rates

Another bias that stems from the representativeness heuristics is the so-called insensitivity to the base rate. Due to this bias, one ignores the past frequency or probability of an event happening and instead, bases the future frequency on irrelevant information.⁶⁹ For instance, a disputant may be influenced by the fact that a similar conflict has recently been resolved in a way favoring its side and does not consider the base rate of winning to losing lawsuits in that specific court or regarding the type of dispute. This bias in particular can lead to worse outcomes in the international community, as the examples of dispute resolution are scarcer and therefore, do not offer a sufficient sample size to deduce predictability.

4. Illusion of Validity

The illusion of validity bias looms whenever there are enough correlated input facts that seem to lead to the desired outcome. The accumulation of more input information relevant to each other increases the confidence that the outcome will be achieved.⁷⁰ The illusionary validity bias occurs whenever a party to a conflict which has acquired many pieces of evidence, yet of low predictive validity becomes more confident that they will win in trial and does not consider other evidence that may decrease the probabilities of the desired win.⁷¹

5. Confirmation Trap

Falling into a confirmation trap describes the bias present in individuals who are primed to search for and believe only in favoring information. Unlike the insensitivity to predictability, the confirmation bias does not incorporate a probabilistic assessment and solely refers to the preparation before an uncertain event. The relevant scenario of this bias could be that a party to a dispute actively seeks and believes only the evidence that confirms their legal standing. Additionally, they might believe the legal interpretation and facts that their legal advisor presents only if they align with their demands and position.⁷² The non-confirming facts are not part of the investigation and fact-collection of the disputant. In this way, they are trapped in a positively reinforcing environment that wrongly seems to promise a sure win in upcoming litigation or a successful continuation of the conflict.

6. Framing

When faced with a risk, framing it as a potential gain or a potential loss can shift the risk preferences of the individual accordingly. Prospect theory relates that if a value is framed as an expected gain, people will choose the more risk-averse option, i.e., the option that secures lower stakes and lower gains.⁷³ Reversely, if the same value is framed as an expected loss, people will opt for the more risk-loving option, one of higher stakes and higher loss. In the context of legal disputes, there is a natural framing depending on the position of each disputant as plaintiff or defendant. Plaintiffs usually expect potential gains while defendants expect potential losses, making the former seek a safe settlement and the latter object to negotiations and seek the highest risk option.⁷⁴

7. Focalism

The tendency of a person to focus exclusively on the occurrence of one of equally likely alternative scenarios is called focalism.⁷⁵ If affected by focalism, disputants might focus solely on the event of winning the trial or the dispute so that they are not interested in joining a negotiation with the perspective of settling on a mutually beneficial agreement.

II. Biases During Negotiation

The biases that are recurring in negotiation environments have been systematized in a more consistent way. ⁷⁶ Nevertheless, there is still room to apply this systematization of biases in negotiation on the international disputes’ framework.

1. Anchoring

Negotiations deal with the reservation prices of the opposing sides.⁷⁷ In general, anchors act as primers in that they are easily retrievable numerical information that the mind of the negotiator resorts to when faced with a complicated question.⁷⁸ Consequently, the initial asking price of one of the disputants may greatly influence the final value of the settlement.⁷⁹ Research confirms the large effect of anchors during negotiation in many experimental settings.⁸⁰ The anchoring bias can affect either the final value of the settlement or the reservation prices⁸¹ of each of the parties. As a result, sophisticated parties that are aware of the anchoring bias may opportunistically boost their asking prices as a means to achieve more gains in a settlement.

2. Framing

During bargaining, offers and counteroffers may include trade-offs, distributional and integrative effects⁸² between the negotiators. Framing a choice as yielding gains or losses can influence the attitude of the decision-maker and their respective risk preferences. It has been found⁸³ that positively framed negotiating parties exhibit more cooperative behavior than negatively framed ones. The parties tend to accept more offers, settle with a higher propensity, and agree with a higher likelihood to mutually beneficial deals. However, a bargaining setting usually has both negatively and positively framed disputants who expect either gains or losses. Together with exacerbated risk preferences, such a negotiation may lead to break-off or a less than optimal mutual agreement.⁸⁴ One example is that risk-loving, yet negatively framed parties do not seek a high loss agreement. When they are faced with risk-averse counterparties, they take advantage of the latter’s wish for a secure low gain agreement.⁸⁵

3. Information Selection

The information selection bias occurs when an individual is influenced by their informational set which they have previously isolated as relevant.⁸⁶ The process of selecting specific information can be driven by accessibility, the evaluation of this information as important, or ignorance. As a result, disputants may select and rely on a narrower set of important information when negotiating, rather than being able to understand and consider a broader range of alternatives or opportunities. Coupled with an opportunistic withholding of valuable information, the information selection bias compromises a mutually beneficial agreement.

4. Status Quo

The status quo bias is the general tendency of individuals to maintain the current state of affairs with their decisions. This bias is mainly responsible for individuals conforming to default options rather than actively changing an option. Apart from the transaction costs foregone, it seems that people value preserving the status quo by accepting a widely accepted option.⁸⁷ This preference could also become relevant in a bargaining situation. A person suffering from status quo bias might accept a less than optimal agreement if convinced that it constitutes a widely accepted default option.

5. Emotion and Cognition Collision

Every individual is sometimes affected by their emotional state, state of fear, shock or excitement. Overloading oneself with the additional task of decision-making during an emotional state is said⁸⁸ to inhibit System 2 from taking over the more difficult decisions and thus allowing System 1⁸⁹ to automatically approve or reject certain options without proper consideration. Such a situation is not foreign to negotiation where parties are usually antagonistic and emotionally affected by facing their opponent and bargaining with them. If a decision in negotiation is made under emotion and cognition collision i.e., when the emotional side and the cognitive side of the brain collide, it is likely that the decision-maker would revoke the decision thus taken after more careful reasoning.⁹⁰

6. Reactive Devaluation

One specific emotional bias, part of the affect heuristic, is the reactive devaluation bias.⁹¹ This bias leads to rejecting an initially desirable offer or concession⁹² and is caused by four alleged different reasons,⁹³ two of which belong to more strategic thinking and two belong to strictly behavioral biased thinking. As for the first category, reactively devaluating an offer could happen due to a lack of private information that may make an offer appear desirable or due to perceiving the offer as a signal of further bargaining potential. With regard to the strict behavioral reasons, reactive devaluation can occur due to a malevolent utility function that makes the offeree have antagonistic and spiteful feeling against the offeror. A second behavioral reason for reactive devaluation can be the impression that “available states of the world are simply less alluring than the unachievable or the uncertain”.⁹⁴

The individual suffering from this bias devaluates the offer made by the opposite side as a reaction to them being opponents in a subjective manner. By lowering the value of every offer or agreement term, the reservation value of the parties increases drastically so that no common ground for settlement exists. Experimental findings on reactive devaluation during bargaining tend to prove the existence of this bias.⁹⁵ Though, these studies are limited due to the difficulty in understanding the channel of behavior that is activated.

7. Self-serving bias

The vast majority of experiments with biases in negotiating settings deal with the self-serving bias, as it is endemic in “morally ambiguous settings in which there are competing focal points.”⁹⁶ Parties maintain their roles as plaintiff-defendant, victim-tortfeasor, employer-employee, or government-group with sovereign demands during a negotiation process. Consequently, the parties are influenced by the nature of their roles as they tend to assess choices and decisions according to their role’s threshold.⁹⁷ The same holds true for negotiating parties when exchanging offers.

Self-serving bias leads⁹⁸ to bargaining impasse for various reasons. The first instance of self-serving bias is when a party to a dispute interprets one’s position in a favoring or self-serving way so that its reservation values are conflated with this belief⁹⁹ and no common ground for a ZOPA is possible. The self-serving bias can also affect one’s fairness perception in the sense that believing one’s position to be the only objectively fair one will create suspicion of the other side’s offer,¹⁰⁰ viewed as less than fair, or even aggressive. As a result, negotiating parties will never accept an offer deemed less than their own self-served belief of fair.¹⁰¹ The literature confirms that self-serving bias exists in an experimental setting with negotiation¹⁰² and that it creates bargaining impasses.¹⁰³

8. Overconfidence

This term is used in everyday life and signals the attitude of someone that believes strongly in oneself, in one’s judgment and abilities. Overconfident individuals are convinced that they are rarely mistaken, which can lead to inefficient outcomes, when present in negotiation settings.¹⁰⁴ In fact, overconfident negotiators¹⁰⁵ tend to insist on their initial positions and avoid concessionary moves. Alternatively, overconfident parties may fear the scenario of losing face during a negotiation, which makes them reject even small concessions.

9. Fixed-pie Error

The fixed-pie error¹⁰⁶ concerns one party’s perceptions of the opposite party and their respective demands, beliefs and interests in the negotiated prize. It is activated when one party expects the opposition to share similar priorities, demands and perceptions on the disputed matter. This shortsightedness can lead to a less than optimal agreement if the fixed-pie erring party ignores potential opportunistic behaviors by the other party. This bias can be accentuated if the parties belong to different cultures, as cultural background can deepen the communication gap while increasing the probability of having various negotiation expectations and attitudes. Even without tactics and opportunism, a party with a fixed-pie bias may agree with terms that seem fair to them and satisfy their priorities without further exploring room for potential gain.¹⁰⁷

E. Mediation as a Debiasing Tool

The previous section concluded that cognitive infringements that prevent parties from realizing the common ground of agreement produce weaker settlements. Selected biases were analyzed in connection to the international dispute setting and the corresponding choice of the disputant. This section explores whether these biases can be reduced with the help of the mediation process.

I. Theory of Debiasing Tools

The existence of systematic and consistent biases during decision-making has not left legislators and policymakers uninterested. There is an ever-growing attempt to tackle biased behaviors with various techniques and tools. Law, being one of the major drivers of human behavior, is considered a debiasing tool¹⁰⁸ according to two different approaches. The first approach is to construct legal rules that induce behaviors immune to biases.¹⁰⁹ The second approach is departing from a paternalistic tendency of imposing the optimal legal rule and behavior. Instead, it relies on smartly designed legal rules that operate directly on the errors allowing for error minimization without the corresponding choice deprivation.

Another relevant distinction¹¹⁰ that offers a comprehensive guide with tools for all types of judgment and cognitive biases is between debiasing/modifying the decision-maker¹¹¹ and debiasing/modifying the environment where the decision process takes place. As for the techniques used for debiasing the decision-maker, they include education, training, use of decision analysis and models as well as checklists. Conversely, the modification of the decision-making environment requires either economic levers or nudges. Economic levers consist of economic incentives to induce smart choices as well as accountability rules and information provisions to inform on social norms. Nudges, apart from switching the default option, can be separated into categories based on their function.¹¹²

These techniques and tools are generally applicable to all decision-making biases and are not context- and bias-specific.¹¹³ Bazerman and Moore¹¹⁴ documented a series of biases found in negotiation settings and proposed strategies to overcome the decision-making pitfalls during negotiation.¹¹⁵ Some of these strategies are decision analysis tools in order to correct departures from the expected utility, acquiring expertise to aid with the decision, following a stricto sensu debiasing training,¹¹⁶ analogical thinking, taking an outsider’s view, and understanding biases in others.

II. Can the Process of Mediation Effectively De-bias?

Some of the proposed debiasing techniques could be successfully employed in combatting the two different types of biases present in negotiation. It also becomes apparent that most of these techniques require an unbiased individual that can help the biased negotiators to apply them and, subsequently, monitor the debiasing process as well as provide relevant feedback. A potential limitation of some debiasing techniques is that the domain of international law does not welcome mandatory legal modifications or the introduction of nudges due to a lack of enforcement and consent.¹¹⁷ Against this background, I introduce the notion that a mediation environment with the presence of a third-party neutral is a situation that can accommodate many debiasing techniques for biased international disputants.¹¹⁸

Findings of experimental literature allude to mediation being an effective mechanism to counter biases and negotiation failure. Framed persons seemed to minimize their framing behavior when asked to provide a rationale for behaving in a specific way.¹¹⁹ The same holds true for self-served persons when they were asked to write down the weaknesses of their bargaining position.¹²⁰ Informing subjects about the existence and function of the bias did not have a major effect in practice.¹²¹ Similarly, one can tackle overconfidence¹²² by questioning one’s judgment,¹²³ giving contradicting arguments or considering the opposite scenario.¹²⁴ More recent research¹²⁵ shows that micro-interventions in the form of simpler questions such as “Why” and “How certain am I” do not yield any significant outcome. In contrast, some voices in the literature¹²⁶ propose the use of Bayesian networks as decision aids.¹²⁷

III. Four Debiasing Features of Mediation

In the next subsections, four mediation features are examined with regard to their potential to remove biased decision-making in international disputes: the party autonomy, the private caucus, the enforceability of the agreement and the role of the mediator i.e., the tools and methods they can use to remove bias.

a. Party Autonomy

One of the main features of mediation as an alternative dispute resolution process is the party autonomy and the facilitation of face-to-face interaction between the disputing parties.¹²⁸ Before and during traditional negotiation talks, international state actors may be accompanied by experts and legal advisors. The latter may enhance biases of optimism, overconfidence, self-serving and confirmation trap by fueling the disputants with confirmatory information that increases their perceived likelihood of succeeding in an upcoming trial.¹²⁹

On the contrary, mediation encourages the parties to sit together and face each other, communicate directly, narrate their side of the dispute, and share their proposals, offers and alternatives. Direct communication may be successful at allowing parties to gain a wider perspective, be exposed to the other side’s arguments and viewpoints so that the optimism, confirmation and self-serving bias may be decreased.

b. Private Caucus

The next feature of mediation that is relevant for debiasing is the private caucus option¹³⁰ or shuttle diplomacy. It works as an intermittent with the mediator upon the request of either side. The mediator enters a private caucus with one party where they discuss private information and potential offers and proposals to be delivered to the other side. The private caucus is covered by strict confidentiality and the mediator is not allowed to disclose any information shared within, unless specifically allowed.¹³¹ Private caucuses provide the perfect setting for mediators to adjust anchors and reframe the disputants’ views. Having a private discussion with the disputant, mediators are free to re-express the information exchanged during the joint sessions in a more tailor-made, context-specific way. Framing is one of the leading biases during negotiations, based on the plaintiff/defendant role of the party. Thus, private caucuses allow mediators to frame the options on the table as gains instead of perceived losses.

c. Enforceability of the Mediation Agreement

Another helpful debiasing feature of mediation is the enforceability of the agreement. Parties may agree on terms or prices while under the influence of emotional biases or reactive devaluation. The weak enforceability of the mediated agreement allows for a cooling-off period,¹³² which alleviates the emotional distress and the heated reactions of the disputing parties. As a consequence, parties have the opportunity to resume mediation or negotiation, or in the case that their biases led to undesirable outcomes, can decide not to conform to the agreement without any further legal implication.¹³³ The same option is not available to the same degree after an obtained court ruling.

d. The Role of the Mediator

Due to their qualification and training, mediators are equipped with additional knowledge and expertise when dealing with disputants in conflict. A mediator can be particularly helpful in cross-border and culture-divergent negotiations.¹³⁴ Cultural differences are often accountable for accentuating various biases in negotiation, as noted above. Hence, a mediator that can observe and, following that, neutralize potential cultural differences may be successful at reducing the influence of some of those biases. Yet it is important to acknowledge that mediators can also suffer from biases while attempting to remove biases from disputants, such as the blind-spot bias¹³⁵ that allows them to understand biases in everyone else except for themselves.

A method that mediators use is the so-called reality testing or checking tool.¹³⁶ A party is reality tested when the mediator poses certain questions regarding the alternatives to the negotiation/mediation or their views on their private interests concerning the dispute. It is considered a very broad-reaching technique because it allows the mediator to understand from which biases both parties suffer. Having established that, the mediators can tailor the subsequent approach so that they successfully address most biases.

At the outset, one has to distinguish between the facilitative and evaluative mediation style at this point.¹³⁷ A facilitative mediator enables the parties to exchange views and opinions and can reformulate subjective and emotional opinions to be more objective and clearer. In this way, disputants experience the narration of the other party and cannot ignore the other side’s viewpoint.¹³⁸ These techniques could reduce phenomena of self-serving bias, information selection and overconfidence. On a similar note, the exchange of views can drastically help the problem of fixed pie error, i.e., the divergence of reality from private belief regarding the other party’s interests in the negotiation. To this end, the mediator can help establish the points of convergence¹³⁹ and underline them for both parties to acknowledge.

An evaluative mediator has the ability to develop some further methods while keeping the methods of the facilitative style in the toolbox.¹⁴⁰ Parties are allowed, upon mutual agreement, to ask for an evaluation of their case, i.e., their probabilities in adjudication. Hearing an outsider’s neutral perspective may change the perceptions of the parties regarding their winning or losing probabilities as well as the reservation values and ZOPA. ¹⁴¹ Biases such as insensitivity to predictability, validity illusion and focalism can be greatly reduced once the mediator expresses an objective estimate. ¹⁴² Self-serving bias may also decrease since it has been found¹⁴³ that listening to reasons why the opposite side might win has a diminishing effect on this particular bias.

Similar to the self-serving bias, framing appears in situations with varying reference points.¹⁴⁴ A study on debiasing of framing revealed that asking for one’s rationale behind deciding between a risky and a certain choice mitigated the framing bias in their subsequent decision.¹⁴⁵ A mediator is able to ask the parties for a rationale. Another way to remove bias from framing is to state that a settlement is a certain gain whereas adjudication is an uncertain win or lose situation, especially when one considers all the certain costs of going forward with it, including legal fees and time.

Reactive devaluation is another bias that can be corrected if the parties share their respective offers with the mediator first so that the latter is responsible for communicating them to the other side.¹⁴⁶ As part of the normal responsibilities of a mediator to act as an intermediary, the communication of an offer can reduce reactive devaluation: It has been shown that in this situation,¹⁴⁷ the opposite side does not perceive the offer as coming from the hostile antagonist. Moreover, the potential reactive devaluation attributed to the feeling that the other party can bargain further can be corrected with the help of the mediator’s evaluation of the dispute.

Another role of the evaluative mediator is to propose alternative solutions upon the parties’ mutual request. This allows the mediator to tackle the anchoring and adjustment biases between the negotiators, as they are influenced by an initial asking price or by their own reservation prices. If the mediator proposes a different value or a different asset, parties may break free from the anchors. In this way, the experience of the mediator can help parties enlarge the zone of possible agreement¹⁴⁸ and correct the fixed-pie error.

Debiasing attempts of a mediator may, however, lead to second-order errors and biases.¹⁴⁹ For instance, a debiasing strategy of the mediator may affect the interactional justice perceptions of the participants to a mediation, for example, a correction of an overconfident offer may lead to a reactional devaluation from the other side. Another limitation is that an accumulation of multiple, differently directed biases may inhibit the mediator from debiasing the parties in an effective way.

F. Conclusion with a Normative Outlook

Following the behavioral law and economics analysis and its application to the mediation setting, it is evident that international mediation can contribute to the successful resolution of conflicts that would otherwise lead to negotiation impasses. Biases that may exist when deciding whether to negotiate or to continue the conflict, for example insensitivity to predictability and base rates, focalism and framing, can be subject to mediators’ debiasing. Furthermore, a trained mediator can address biases occurring at the negotiation table such as anchoring, reactive devaluation and fixed pie errors. By way of conclusion, some institutional suggestions and other fora where the advantages of mediation could bear fruit are discussed.

First, the distinction between biases when deciding to negotiate or continue the conflict and biases during subsequent negotiation reveals that the current modus operandi of mediation might be taking place too little, too late. Escalation of emotions and conflict inhibits the parties’ readiness to make concessions and increases the need for every party to save face. If mediation takes place earlier than at a moment of crisis, it may resolve more aspects of a conflict or the conflict itself in a permanent way. For this to be feasible, the respective institutional capacity must be in place. For instance, a permanent UN body could be endowed with the responsibility to monitor conflicts in escalation and propose mediation at an early stage. Alternatively, as a way to reduce information costs and logistics, local monitoring agencies could be tasked with observing the initial stages of a conflict and contributing with mediation approaches and tools wherever asked to.

Comparable to mediation on civil matters, ICSID or WTO could introduce a preparatory stage where parties to an investment or trade dispute consult with a third neutral, non-adjudicating party before going forth with the formal adjudicatory procedure. These steps would require the broadening of international mediation rules, either with the help of treaties, UN resolutions or soft international law. In any case, this suggestion must be coupled with some type of mediation protocol¹⁵⁰ and appropriate training of the mediators that can curb the limitations presented above,¹⁵¹ such as varying legal and disputing cultures, consent to mediate and enforcement weakness. One encouraging step in this direction is the UN Convention on International Settlement Agreements Resulting from Mediation, adopted in December 2018.¹⁵² The so-called Singapore Convention aspires to encourage mediation by facilitating the enforcement of international settlement agreements on matters of trade.¹⁵³

Second, mediation could also contribute to bias reduction during the negotiation phase for international treaties. Environmental treaties are in the spotlight when it comes to hard negotiation impasses and opt-out clauses from certain international actors. Having a mediation option during these sessions could prove the effectiveness of the debiasing role that mediation can have.

Third and finally, establishing mediation rules and mediation fora can have an effect on international preferences. As states’ preferences can be shaped by the setup of normative ideals,¹⁵⁴ international disputants might be able to recognize their biases in negotiation settings and actively seek more self-determining forms of resolving their disputes.