Tilburg University The Sorry Clause Srivastava, Vatsalya Publication date: 2016 Document Version Early version, also known as pre-print Link to publication in Tilburg University Research Portal Citation for published version (APA): Srivastava, V. (2016). The Sorry Clause. (CentER Discussion Paper; Vol. 2016-008). CentER, Center for Economic Research. General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Download date: 14. mrt.. 2023 THE SORRY CLAUSE_ by Vatsalya Srivastava January 21, 2016 TILEC Discussion Paper No. 2016-004 CentER Discussion Paper No. 2016-008 ISSN 2213-9532 ISSN 2213-9419 http://ssrn.com/abstract=2737286 The Sorry Clause∗ † Vatsalya Srivastava Tilburg University, CentER, TILEC January 21, 2016 Abstract When players face uncertainty in choosing actions, undesirable outcomes cannot be avoided. Accidental defections caused by uncertainty, that does not depend on the level of care, require a mechanism to reconcile the players. This paper shows the existence of a perfect sorry equilibrium in a game of imperfect public monitoring. In the sorry equilibrium, costly apology is self-imposed in case of accidental defections, making private information public and allowing cooperation to resume. Cost of the apology required to sustain this equilibrium is calculated, the efficiency characteristics oftheequilibriumevaluatedandoutcomescomparedtothosefromotherbilateralsocial governance mechanisms and formal legal systems. It is argued that with the possibility of accidental defections, other social mechanisms have limitations, while formal legal systems can generate perverse incentives. Therefore, apologies can serve as a useful economic governance institution. Keywords: Apology, Sorry, Imperfect Public Monitoring, Uncertainty, Social Norms, Economic Governance, Legal Institutions, Courts, Incentives. JEL Classification: D08, K04, Z01. ∗IamgratefultoJensPru¨fer,EricvanDamme,MichaelPowell,JohnBarkleyRosser,Jr.,MarkRamseyer, Alex Teytelboym, Avner Greif and all participants of ISNIE (2015, Harvard), ESNIE (2015, Corsica) and CLEEN (2015, Tilburg) conferences for their insightful comments and feedback. All errors are my own. †CentER, TILEC, Tilburg University, P.O. Box 90153, 5000 LE Tilburg, The Netherlands; [email protected] 1 Introduction The word sorry, is an almost universally acknowledged expression of apology. As a regretful acknowledgmentofanoffenceorfailure,itisusedtoexpressregret,seekforgiveness,reconcile differences and enable cooperation in the future in case of breach of a social contract. An apology is tendered in a multitude of different circumstance to mitigate ill-will and resolve conflicts: from individuals who bump into someone in a crowd, to national governments who have kidnapped aboriginal children in the past. 1 The concept of apology is time honoured and finds expression in many different cultures (Allen, 2006; Al-Zumor, 2011; Bar-Siman- Tov, 2003). If the wide variety of Sorry greeting-cards available were not evidence enough, there are firms now who apologise on behalf of their clients for a fee: ”Tianjin and the central city of Xi(cid:2)an now boast new, successful apology companies. And apology call-in shows on state radio have also started to appear ...On behalf of clients, the apologisers write letters, deliver gifts and make explanations.” 2 The ubiquitous usage of apologies, its important role in society and complexities have been duly acknowledged and investigations have spawned a vast literature in a host of differ- ent disciplines. There is evidence for the role that apologies play in business (Folkes, 1984; Hearit, 1994; Abeler et al., 2010). A vast empirical literature in social and child psychology investigates the various aspects of apologies (Schlenker, 1980; Schlenker and Darby, 1981; Darby and Schlenker, 1982; Ohtsubo and Watanabe, 2009; Ohtsubo et al., 2012) and its impact on forgiveness and justice (Witvliet et al., 2008; Wallace et al., 2008). Other fields of study like linguistics (Cohen, A. D., Olshtain, 1985; Reiter, 2000), history (Trouillot, 2000; Gibney, 2008) and politics (Lind, 2011; Nytagodien and Neal, 2004) have also explored the effect of apologies. Economic experiments have also been used to verify the efficacy of apolo- gies in repairing damaged reputation and enabling reconciliation (Fischbacher and Utikal, 2013; Ho, 2012). In a departure from the positive exploration and documentation of apologies in other subjects, the legal literature has taken a normative stand of calling for a greater role for apologies in the judicial system of the U.S.A (Wagatsuma and Rosett, 1986; Shuman, 2000; Petrucci, 2002). The virtues of apologies have been evaluated and extolled in keeping down crime and recidivism rates (Haley, 1982, 1995, 1998; Robbennolt, 2003). Its role in providing cathartic relief to affected parties in wide variety of cases including, medical malpractice (Keeva, 1999), divorce (Schneider, 2000) and civil rights abuse (White, 2005), has also been found to be positive. These studies posit apology as a redressal mechanism outside the 1https://www.aadnc-aandc.gc.ca/eng/1100100015644/1100100015649 2http://www.nytimes.com/2001/01/03/world/tianjin-journal-for-a-fee-this-chinese-firm- will-beg-pardon-for-anyone.html 1 formal judicial process. Their argument being that the justice system stands to benefit from the restorative, conciliatory effects that apologies engender if they are incorporated into the judicial process. Nonetheless, despite its pervasiveness in society, offering an apology or determining when or how much to apologise is no simple matter: On May 7, 1999, during the NATO bombing of Yugoslavia, five US guided bombs hit the People’s Republic of China embassy in the Belgrade district of New Belgrade, killing three Chinese reporters and outraging the public. President Clinton promptly expressed deep regrets and tendered an apology. However, the apology was dismissed as being too casual, delivered while Mr. Clinton was outdoors and wearing a polo shirt. The claim of much of the Chinese media was that ”it was not really an apology at all.” (Stamato, 2008) Theefficacyofanapology, seemstobedeterminedbynormsthatarealsolikelytoprovide a measure of their sincerity. The economic literature on social norms maintains that formal legal institutions are not indispensable to covenants of cooperation (that enable productive economicactivities)(Ostrometal.,1992;Ostrom,2000;Posner,1997). Thenormsthatallow for such cooperation, reflect a social contract, an implicit agreement among the members of a society. There is plethora of evidence supporting the role of apology in governing interaction and maintaining social organisation, making it one such social contract and an therefore, an instrument of economic governance. This paper is an attempt to analyse apologies as a social norm. A game-theoretic frame- work, of imperfect public monitoring, is used to rationalise the existence and usage of apolo- gies. The questions that structure this analysis are: What in the nature of socio-economic interactions necessitates the need for apology? Is apology an equilibrium outcome? What is the cost of such an apology? What are the characteristics of such an equilibrium? How do the welfare outcomes compare to that in other modes of community governance, like os- tracising defaulting members? What are some of the additional constraints that courts face compared to apologies? In answering these questions, this paper contributes to the literature on imperfect public monitoring games by establishing the existence of a sorry equilibrium. In a sorry equilibrium, while the equilibrium strategy is not public 3, a costly apology al- lows private information to be made public. It also contributes to the legal discussion on apologies by providing a more rigorous analytical framework which enables delineation of the trade-offs inherent in the use of apologies, along with inferences about the limitations of other governance mechanisms. The remainder of the paper is organised as follows. In the next section, some of the em- pirical facts that have led to legal scholars emphasising the role of apologies are presented. In section 3 the relevant extant literature on social interaction and economic governance 3It conditions on private information. 2 is reviewed while the characteristics of apologies are identified. The model is outlined in Section 4; laying out all the assumptions. In section 5 the equilibrium strategy is proposed and it is shown that it constitutes an equilibrium; cost of apology in equilibrium and some comparative statics are discussed. In section 6, some characteristics of the sorry equilib- rium are shown. The welfare outcomes of different community enforcement mechanisms are compared to apologies in Section 7. In section 8 a model of formal court governance is out- lined, its limitations identified and some policy suggestions are made. Section 9 concludes by summarising the results and making suggestions for future research. 2 USA, India and Japan The criminal justice system in the United States of America has 6.94 million people or 1 in every 35 American adults under its purview, of which approximately 2 million are incarcerated (Glaze and Herberman, 2012). In fact, the U.S is next only to Seychelles in its prison population rate. 4 These statistics are staggering, but the socio-political impact of high incarceration rates and the huge burden they put on the taxpayer notwithstanding, these number do not reflect the full scale and scope of the impact that the legal system has on the economy at large. 5 The tort system in the US has also been under attack for being costly and inefficient(Shuman, 2000). While some claim that it lets corporations get away with huge excesses, others claim that it imposes too high a cost on businesses (Krauss, 2012). In particular, American courts have gained notoriety in their handling of ”second-order law”: class actions, sexual harassment claims, or medical malpractice (Ramseyer and Ras- musen, 2010). 6 Their penchant for awarding large punitive damages in many high profile cases is well documented (Wagatsuma and Rosett, 1986; Ramseyer and Rasmusen, 2010). The unpredictability that stems from such cases can have major repercussions for the econ- omy: ”They can profoundly affect the social relations and economic structure within a coun- 4Number of people incarcerated per capita: http://www.prisonstudies.org/highest-to-lowest/ prison_population_rate?field_region_taxonomy_tid=All. 5Therearealsomajorconcernsaboutthecompositionofprisonpopulation. In2006, about60percentof jail inmates reported having had symptoms of a mental health disorder in the twelve months prior to their imprisonment (James and Glaze, 2006). There are other long term impacts of imprisonment. Many people arereturnedtojailfornon-paymentoffinesandfees(Evans,2014). Accordingtoanotherstudy, individuals who do manage to find work after release earn less on average than their counterparts who have never been incarcerated. Among formerly incarcerated men in the study, ’two-thirds of whom were employed before being incarcerated’ hourly wages decreased by 11 percent, annual employment by nine weeks, and annual earnings by 40 percent as a result of time spent in jail or prison (Western and Pettit, 2010). 6The authors distinguish between first-order law and second order law. In this classification first order law is pertains to the typical disputes over automobile accidents and contract claims. 3 try, but not because they are common or because they protect property rights. They affect social relations and economic structure because despite their scarcity they discourage invest- ment and cause firms to take precautions of little social value.”(Ramseyer and Rasmusen, 2010, p. 6) The Indian Judicial System, servicing the world’s second most populous country is an- other massive legal institution. There are about 20 million pending cases in lower courts and 3.2 million such cases in the high courts (Hazra and Micevska, 2004). More recent estimates put the total number closer to 30 million.7 70% of the incarcerated population is under-trial and many for those awaiting trials have been in jail longer than a formal sentence would have required them to be (Krishnan and Raj Kumar, 2010). The costs of such a logjam in the courts, both social and economic, are so large that while often spoken of, hardly any comprehensive and reliable estimates are available. In India, the limited capacity of the legal system is a binding constraint on the number of cases that can be handled by the courts. In the US, a relatively more efficient legal system with a propensity to grant multi-million dollar damages and commit people to supervision has probably created perverse incentives whereby, it is individually rational for citizens to access the courts even when it is not socially optimal to do so. 8 The nature of the problems facing the legal systems of the two countries and their sources are disparate, but it is apparent that both the systems require improvements. In contrast to the U.S and India, in the Japanese legal system the attitude of the accused and their willingness to apologise and confess is crucial to the decision of whether to prosecute or not (Haley, 1982; Stephens, 2008). In fact, relying on the willingness of the accused to apologise after the offence, at one time in Japan ”33% of all cases involving non-traffic related offences were suspended by the prosecution” (Haley, 1982, p.271). Further, even when prosecution did proceed and the defendant was found guilty, the court suspended jail sentences in more than two-thirds of such cases. Apology is therefore used in the Japanese legal system as an informal sanction which reduces the likelihood that a dispute will be taken to court (Haley, 1998). The Japanese legal system provides a great example of apologies being effective in a wide-variety of scenarios; resolving disputes and fostering less acrimonious attitudes. It has also been associated with the low crime and recidivism rates in Japan (Haley, 1998). But lest the case of Japan be considered an isolated cultural anomaly, there are other examples from all over the world. In the U.S., evidence from a study on settlement offers at the time 7http://timesofindia.indiatimes.com/india/Courts-will-take-320-years-to-clear-backlog- cases-Justice-Rao/articleshow/5651782.cms. 8Optimal here refers to social optimal where (marginal cost)=(marginal benefit) given the law and insti- tutional capacity. This is not to deny the role of supply side activities in altering the optimal level. 4 of an accident indicates the relevance of apologies. 73% of the respondents would accept the settlement offered if a full apology was tendered compared to 52% when no apology was offered (Robbennolt, 2003). Since 1989, community conferences in New Zealand have dealt with more than half of the juvenile offenders in an alternative system that relies on getting the victim and the offender to meet and utilises apologies for dispute resolution and victim rehabilitation(Scheff,1998). InAustraliathesamesystemisusedforbothadultandjuvenile crimes. In fact, it is claimed that tort plaintiffs often profess that what they really wanted was an apology and brought a suit only when it was not forthcoming (Shuman, 2000). 3 Characterising Apology The existence of pro-social preferences, norms and social enforcement mechanisms (replying onmonitoringandpunishments)canallowbi-lateralormulti-lateralinteractionstoovercome the problems inherent in social dilemmas (Abreu, 1988; Greif, 1993; Sobel, 2006; Dixit, 2009). The mechanism that allows for sustained cooperation in such a setting has usually been modelled in the literature as self-reinforcing equilibria, made possible by repeated interactions between agents 9. In such models of cooperation without external enforcement, the punishment never gets played out in the equilibrium, the threat exists merely to ensure that players cooperate in equilibrium. This is also true for the role of legal sanctions in law and economics literature, where sanctions are usually put in place to deter cheating (Hermalin et al., 2007). The equilibrium outcomes of such models, however, are at odds with observations of the day-to-day usage of apologies. They are used rather frequently, both in long-standing relationships as well as in interactions with comparative strangers. The first 2 unique char- acteristics of an apology that distinguish it from other forms of social enforcement: (1) The frequent usage of apologies suggests that it constitutes an equilibrium outcome and is not to be relegated to the off-equilibrium path to merely act as deterrence. (2) In order for apologies to be an equilibrium outcome there must be scope for fail- ures/offences in equilibrium. Moreover given the usage of apologies in a wide-variety of circumstances, the cause of such failures must be sufficiently general to find manifestation in many different types of events. Studies in social psychology posit that an apology is a plea for the offended to not take a negative event as representative of the intentions and character of the offender (Schlenker and Darby, 1981; Schlenker, 1980; Darby and Schlenker, 1982). (3) An apology is an attempt to establish a lack of intent on part of the offending party, 9Insituationswherereputationsareimportant,casespecificex-postarrangements(Bull,1987)infinitely repeated interactions can also lead to cooperative outcomes. 5 along with being an acknowledgment of harm or offence caused. So, while the harm caused is not disputed, the contention is that any such harm was accidental and not the result of an intentional action. There must therefore be a separation in intent and outcome in the game. The possibility of allowing for mistakes in selection of actions to explain the need for apologies has already been explored in the literature (Fischbacher and Utikal, 2013). It has been shown that in an experimental setting, players use apologies if their intentions cannot be easily inferred. In such situations, the evidence shows that ”An apology is a strong and cheap device to restore social or economic relationships that have been disturbed” (Fischbacher and Utikal, 2013). However, this particular attempt at explaining the usage of apologies relies on players having other regarding preferences. Such an approach requires additional assumptions about preferences and limits the possibility of exploring why such preferences might have been socialised in the first place. 10 The separation of intent and outcome has been also been treated in a more general set- ting in the economics literature, as games of imperfect public monitoring (Rubinstein, 1979; Fudenberg and Maskin, 1986; Fudenberg et al., 1994). This literature does not rely on ad- ditional behavioural assumptions; instead the offended player is responsible for statistically distinguishing between an accidental offence and a deliberate one. There are however limita- tions to the solutions provided to the problem of accidental defection in this literature. The complications of statistical identification aside, the equilibria proposed either rely on mixed strategies (strategies are not specified, instead, payoffs that can be generated by some mixed strategies are the focus) (Fudenberg and Maskin, 1986; Fudenberg et al., 1994) or players who do not discount (Rubinstein, 1979). An apology is fundamentally different from these approaches; it requires the offending player to take on the burden of resolution. Apologies, in most social conventions can only be offered, not taken. This implies that an apology is a self-imposed sanction. As an apology must be offered, the offending player can condition his actions on private information. An apology therefore, may be offered only when the player knows that the offence was not deliberate. 11 This is a departure from public strategies hitherto used in games of imperfect monitoring. (4) A village elder can order it or friends can goad for it, but eventually an apology has to be offered by the offender to the offended. This distinguishes it from more conventional punishment that can be met out by some authority or an outside agent. In fact, an apology has a redeeming quality built into it. This characteristic is most visible in a social dictum 10Therefore it is instructive to identify the usefulness of apologies without pro-social preferences that support its usage. 11In fact, it has been shown that apologising after a deliberate offence can often make the situation worse (Fischbacher and Utikal, 2013; Struthers et al., 2008). 6 regarding apology that emphasises the need to feel sorry, when an apology is offered. This need to feel sorry can be understood as the cost that needs to be incurred for the apology to serve its purpose. (5) Apologising is costly. This observation while not self-evident is a critical one. An apology performs the purpose of placating the affected player. If there was no cost attached to apologising, the party causing the harm could always apologise and consequently there is no reason why the affected party should take any apology seriously. So there must be some psychological, social or even monetary costs attached to apologising. This argument is akin to claiming, that in general, apologising is not cheap talk. This is not to deny that in certain cases, apologies may well be cheap talk. However, cheap talk is useful (and a cheap talk apology likely to be honest), only if the interests of the two players are aligned to a substantial degree (Crawford and Sobel, 1982; Farrell and Rabin, 1996). So, if the objective of the strategic interaction between players is akin to a coordination game, then a cheap talk apology might indeed be useful (Ohtsubo and Watanabe, 2009; Ho, 2012). But the objective of this paper is not to investigate all possible uses of the word sorry in the English language, but to establish the role of apologies as a resolution and reconciliation mechanism. As social dilemmas do not afford the luxury of useful cheap talk, apologies are in general considered to be a costly self-imposed sanction. Furthermore, even seemingly cheap talk apologies might have costs associated with them. If apologies are useful tools to mitigate conflicts, then it is entirely feasible that preferences favouring their use be socialised in members of a society. This might imply that there may be psychological costs attached with apologising. 12 (6) There are different levels of sorry. The social psychology literature differentiates between at least 5 different levels of apologies (Schlenker and Darby, 1981). The easiest way to think of this claim is to imagine the varying levels of sincerity with which sorry can be said. A simple sorry, or a profuse apology or even buying a sorry greeting card to convey the apology. This implies that the cost incurred for an apology can change to reflect some underlying differences in the circumstances. Definition 1 Apology is a self-inflicted cost, undertaken by the offending party to convince the offended party that any mistakes committed were accidental. These accidental infractions or mistakes are caused by the residual stochasticity inherent in actions. While the law and economics literature on torts and liability emphasises the level of due care (Brown, 1973), residual stochasticity refers to that left over component of uncertainty that does not depend on the level of care taken by an individual. 12Apologising might involve shame, disappointment, embarrassment and other negative emotions. So, while the model developed here does not rely on pro-social preferences, they might reflect a part or the entirety of the cost of apology posited here. 7