• A Introduction

  A traditional principle of tort law is that victims of a tort have to be compensated, as much as possible, by assessing damages in concreto, implying that their specific individual situation needs to be taken into account when compensation is assessed. Restitutio in integrum aims at bringing the victim into the position they would have been in, had the tort not occurred. This assessment can be a protracted, time-consuming and complicated procedure, especially in cases of personal injury in many legal systems. As a result, victims are often left unsatisfied, receiving compensation only after several trials and procedures that sometimes take years, especially in complicated cases. Victims often indicate that it is not compensation that they are primarily interested in; rather, what they really seek is recognition and expeditious decision-making. The lengthy and complicated procedures often result in additional suffering for victims. Moreover, the lengthy procedures also imply the involvement of (many) lawyers and, therefore, high costs.
  In other words, the way victims of personal injury are traditionally compensated via the judicial system has been increasingly subject to criticism, as the system is not satisfactory from the perspectives of neither the victim nor the society (given the high social costs). Several legal systems have introduced different ways of dealing with personal injury, with some moving away from dispute settlement via the courts and introducing alternative dispute resolution (ADR) via (administrative) boards, and others introducing and following standardized compensation schemes. The latter use a model whereby for a specific type of injury (e.g. the loss of an eye) a bandwidth of compensation is defined, which provides an indication to the decision-maker. The idea is that this system of standardization increases the predictability of the compensation and, thus, increases the likelihood of settlements and reduces the duration of litigations. In cases where no agreement can be reached between the victim and the (insurer of the) injurer, the dispute would then be settled by a personal injury compensation board through ADR. The latter is often cheaper and more rapid than dispute resolution via the court.¹x There are, of course, also countries which make use of standardization but where, if no agreement is reached, litigation before a court ensues. The countries that follow such a model of standardization (inter alia Belgium, Ireland and Sweden) report high degrees of victim satisfaction, especially when it is accompanied with ADR via boards (as in Ireland and Sweden). Overall, standardization efforts mainly relate to non-pecuniary losses.
  These initiatives provide the starting point for this article. The research question that we examine, therefore, is how a system of standardizing compensation for personal injury, typically in combination with ADR, can be assessed from an economic perspective. To this end, first we establish a theoretical framework that sets out the value of speed and accuracy based on a deterrence and compensation perspective. Second, we elaborate using three country-specific case studies how a stronger focus on averaging can work in practice and tentatively evaluate the systems (including user satisfaction) in light of the incentives they set for injurers and victims. Notwithstanding the potential advantages of standardization, there is also some criticism. There is inter alia the fear that a system of standardization would no longer do justice to the individual needs of victims as there would no longer be a damage assessment in concreto. Moreover, from an economic perspective, the goal of compensating personal injury is also to provide incentives for prevention to potential tortfeasors. The question then arises as to whether this goal of prevention can still be served when compensation is provided on the basis of standardized amounts.
  In the sections to follow, in Section II, we discuss a general law and economics framework to analyse personal injury cases; in Section III, sketch three examples of personal injury compensation from a comparative perspective (namely, Belgium, Ireland and Sweden); in Section IV, present an analysis of our findings; and, in Section V, the final section, provide concluding remarks and a summary of our findings.

• B Law and Economics Framework

  The goals of tort law from a law and economics perspective are two-fold: (1) (potential) tortfeasors need to be induced to take optimal care by exposing them to adequate sanctions – that is the essence of the deterrence perspective as originally introduced by Gary Becker.²x Becker, G.S., ‘Irrational Behavior and Economic Theory’, Journal of Political Economy, 1962, Vol. 70, 1-13; Becker, G.S., ‘Crime and Punishment: An Economic Approach’, Journal of Political Economy, 1968, Vol. 76, 169-217. Tort law also provides incentives regarding activity levels and risk spreading, but in our article we focus on the care incentives. See, for example, Shavell, S., Foundations of Economic Analysis of Law, Cambridge, MA, The Belknap Press of Harvard University Press, 2004, 177-193. This way, they internalize the social costs of their behaviour. From this ex ante angle there is the safeguard that incentives are set so as to avoid the commission of a tort in the first place (or, better formulated, to take cost-effective preventive measures). (2) There is the goal of victim compensation.³x Koziol, H., ‘Introductory Remarks’, in Koziol, H. (ed.), The Aims of Tort Law. Chinese and European Perspectives, Vienna, Jan Sramek Verlag, 2017, 3-12, 4-6. Whereas the former perspective takes an ex ante angle, the latter considers the ex post situation. Note that there is an important interrelation between the two goals. The deterrent value of a liability regime is determined – in part if there are other sanctions such as fines on top and in full if there is only civil liability – by the prospect of paying compensation, the likelihood that this will happen, and the magnitude of the tort damages to be paid. Note that from an economic perspective, compensation is in essence not a goal of tort law, but merely a means, an instrument to induce the potential injurer towards efficient prevention.⁴x Posner, R.A., Economic Analysis of Law, 9th ed., New York, Aspen, 2014, 223; Faure, M., ‘Economic Optimisation of Tort Law’, in Koziol, H. (ed.), The Aims of Tort Law. Chinese and European Perspectives, Vienna, Jan Sramek Verlag, 2017, 79-113, 80 and 87. For both deterrence and compensation, it holds that perfection is too costly. It is, therefore, not advisable that legal systems strive for perfect deterrence, with no importance given to the administrative costs involved. Nor is it desirable that compensation is compulsorily rendered perfect no matter how costly it would be to administer it.⁵x Calabresi pointed to the fact that the tort system needs to minimize the total social costs of accidents consisting of primary, secondary and tertiary accident costs. Whereas the primary costs consist of the costs of accident avoidance and the potential damage and the secondary accident costs relate to the costs of loss spreading, the tertiary costs are the administrative costs needed to minimize the primary and secondary accident costs (inter alia the costs of the legal system to provide compensation to victims (Calabresi, G., The Costs of Accidents. A Legal and Economic Analysis, New Haven, Yale University Press, 1970, 26-28).
  As we will elaborate further, both speed and accuracy might have some relevance in determining the adequacy of the deterrence via the tort system. Whereas speed may at first appear to be primarily important for victim satisfaction, we argue that delays may also be undesirable from a deterrence perspective, as a delayed duty to pay compensation might dilute the ex ante incentives of the injurer. The tort system in theory requires accuracy of the compensation in each individual case, but this comes with high administrative costs. We will argue, therefore, that the importance of accuracy and a concrete assessment in each and every individual case is more limited than one might consider at first glance. We develop those arguments, both from the deterrence (B.I) and from the compensation (B.II) perspectives.

  I Deterrence

  1 Deterrence and Speed

  The question being addressed is, how does the speed of dispute resolution affect the ex ante deterrent incentives of potential tortfeasors? This question is important as some of the reforms mentioned in the ‘Introduction’ section were introduced precisely because dispute resolution via the judicial system can often take many years (to some extent due to debates surrounding the uncertainties concerning the appropriate amount of damages). However, the duty to compensate is not only relevant for the position of the victim but is also supposed to provide incentives for prevention to the potential tortfeasor. The question arises as to whether delayed decision-making concerning compensation could equally affect this deterrence. Assuming rationality, it can be argued that this may certainly be the case.⁶x Additional arguments are imaginable from a behavioural perspective: think of over-optimism or over-confidence bias which leads to an over-activity of a certain behaviour. There may further be instances of hyperbolic discounting. A coherent analysis in the frame of behavioural law and economics is beyond the scope of this article. It could, for instance, be argued that a potential injurer may discount the fact of having to pay compensation in 10 years (after a final court order condemns the injurer to pay compensation to the victim). The tortfeasor could, therefore, discount that compensation order to take into account its current net present value. In other words, the probability of, for example, having to pay €50,000 in 10 years to a victim does not have the same value as if the tortfeasor had to pay that amount today. Discounting future damages payments leads to a reduction which, in turn, could reduce the incentives for prevention. In theory, the instrument of legal interest may solve or at least ameliorate this problem, but the interest rate then should be high enough to (partially) offset this discounting. Discounting could also take place since the tortfeasor might consider that, with the passing of time, the probability of the injurer having to pay compensation to the victim might diminish. This could, for example, be related to the possibility that the injurer would become insolvent (and could, thus, not be affected by a judgment any longer) or that with the passing of time the victim might give up their claim given the high costs of the legal procedure. There may be less of an impact on the probability of detection and more on the probability of conviction, which is also a crucial parameter in Becker’s deterrence framework. In addition, if the quality of the evidence deteriorates with time, the injurer may benefit from delay, because the chances of being found liable decrease, and this reduction in expected liability (due to the lower probability) may affect the ex ante care incentives provided by tort law. So far, we assumed that the tortfeasor would themselves be exposed to liability. The situation may be different when the liable injurer has covered the liability through liability insurance. In that case, the incentives for prevention are no longer provided through the tort system to the injurer directly; rather, it becomes a matter of moral hazard control by the insurer. That could take place quicker (e.g. via a premium increase the year after the accident), irrespective of the duration of the procedure. However, (mandatory) liability insurance is not available in all accidents involving personal injury.⁷x This is of course the case with traffic accidents where mandatory insurance for motor vehicles applies.

  2 Deterrence and Accuracy

  Deterrence is based on the idea of making tortfeasors internalize externalities by way of an anticipated damages payment. Such a prospect will retroact to the ex ante decision-making to engage in wrongful behaviour or at least on the decision of how much care to take and how often to engage in the activity. This is determined by the details of the liability regime in place. The seminal paper that sets out how accurate compensation payments have to be for deterrence to be effective was written by Kaplow and Shavell in 1996.⁸x Kaplow, L. and Shavell, S., ‘Accuracy in the Assessment of Damages’, The Journal of Law & Economics, 1996, Vol. 39(1), 191-210; Baumann, F. and Friehe, T., ‘On the Superiority of Damage Averaging in the Case of Strict Liability’, International Review of Law and Economics, 2009, Vol. 29, 138-142. Also see Kaplow, L., ‘The Value of Accuracy in Adjudication: An Economic Analysis’, Journal of Legal Studies, 1994, Vol. 23, 307-401. The main conclusion is: for deterrence purposes, it is enough if judges are correct on average. This has a sufficient effect on the rational potential tortfeasor, exactly because of the ex ante effect. The potential tortfeasor estimates the expected liability, and, for this expected value, it makes no difference if tortfeasors compensate the actual or the average losses.⁹x For example, if there is a 50% chance that the losses are €10,000 and a 50% chance that they are €30,000, the expected loss is 0.5*10,000 + 0.5*30,000 = €20,000. For the expected liability it then makes no difference whether the tortfeasor faces a 50% chance of having to pay €10,000 and a 50% chance of having to pay €30,000, on one hand, or a 100% chance of having to pay the average losses of €20,000 on the other. In both cases, the expected liability is €20,000 (this example of course assumes risk neutrality of the parties involved). Hence, the compensation payment does not have to correspond to each victim’s individual loss. As long as courts assess the losses correctly on average and base compensation payments on this average, adequate behavioural incentives are set. There is no need for accuracy in each and every case, but systematic mistakes, such as a constant omission of certain damage components or systematic overestimations of certain types of losses, have to be avoided. In conclusion, the goal of optimal deterrence is achieved by payments that are correct on average. The expected liability remains the same from the tortfeasors’ point of view, irrespective of whether tort damages are based on the actual losses or on the average losses. More accuracy may even be undesirable, because it does not result in better incentives but only increases the costs of damage assessment.
  In the above line of reasoning, the tortfeasor has no information about the size of the actual losses they may cause. In such situations, they can only base their behaviour on an estimation of the average losses. This changes if the tortfeasor does have information about the actual losses they may cause. If damages are still based on the average losses, tortfeasors who know that they will cause above-average losses receive inadequate care incentives. Tortfeasors who know that they will cause below-average losses on the other hand, receive excessive care incentives. Accurately assessed damages then provide better incentives, but this advantage must be weighed against the increase in costs of damage assessment.¹⁰x If the tortfeasor does not yet have this information but can acquire it by additional research, the costs of this research must be included in the weighing of costs and benefits as well. See Kaplow and Shavell (1996, 195); Kaplow (1994, 319).
  The smaller the possible gap between the average loss and the actual loss, and the higher the costs of an accurate assessment, the better it is to base damages on the average loss.¹¹x Kaplow and Shavell (1996, 194); Kaplow (1994, 315). With personal injuries, the losses can differ extensively between cases, and this in principle pleads for actual individual assessment. However, in the current section, which deals with the deterrence incentives, the essential question is whether the tortfeasor ex ante has information about the actual loss they may cause. If they do not have such information, more accuracy has no added value.
  In our view, this results in the following: there is a value in having subgroups. If it is relatively easy to distinguish between types of cases on the basis of the loss they may cause, tort damages should be based on the average loss for such types of cases. If it is difficult to predict the concrete loss within such types of cases, further accuracy is not desirable. To give a simple example, the types of losses caused by car drivers differ from those caused by cyclists. Therefore, tort damages for car drivers and for cyclists should not be based on the average loss caused by both groups together. After all, it is predictable that car drivers on average will cause higher losses than cyclists; hence, car drivers should receive more incentives to take care than cyclists. If it is not possible for car drivers to predict the actual loss more accurately than ‘the average loss caused by car drivers’, further accuracy within the group of car drivers has no added benefit from the deterrent perspective and would only cause higher assessment costs.
  A further complication is connected to the possible risk aversion of tortfeasors. Risk-averse tortfeasors prefer average damages over actual damages because they then know that they will not be confronted with high outliers where the losses are much higher than average. They will also not benefit in cases of low outliers, but due to risk aversion the possible disadvantage of high outliers is perceived to be larger than the possible advantage of low outliers. In as far as risk-averse tortfeasors take excessive care measures to reduce the risk of being held liable, basing damages on average losses has the added benefit of avoiding such excessive care.¹²x In the standard law and economics literature it is, furthermore, suggested that strict liability requires full compensation whereas negligence could also provide correct incentives with incomplete compensation, as long as taking due care is cheaper than running the risk of being held liable. However, this view overlooks the impact of the legal requirement that the negligence must be the cause of the losses, as is convincingly shown by Kahan, M., ‘Causation and Incentives to Take Care Under the Negligence Rule’, Journal of Legal Studies, 1989, Vol. 18, 427-447; see Visscher, L., ‘Wrongfulness as a Necessary Cause of the Losses – Removing an Alleged Difference between Strict Liability and Negligence’, Economic Analysis of Law Review, 2011, Vol. 2, 188-203.
  So far, we have focused on the tortfeasor. In settings where also the potential victim can affect the expected accident losses, the so-called bilateral accidents, accuracy may affect the behavioural incentives of the victim as well. After all, if damages are based on the average loss, victims who expect to suffer above-average losses regard these tort damages as under-compensatory. This may induce them to take excessive care and/or to engage in the activity less often, because this reduces the likelihood of being involved in an accident where part of their losses remains uncompensated. This issue is not problematic if there is a negligence rule in place. Under negligence, injurers (at least in theory) receive incentives to behave carefully, which implies they will not be liable in the first place. Victims, hence, expect to have to bear their own losses anyway. But under a rule of strict liability, where victims do expect their losses to be compensated, the under-compensatory level of average damages indeed may result in excessive incentives for the victim.
  Victims who expect to suffer below-average losses may receive inadequate behavioural incentives if damages are based on the average loss, because they would in sum benefit from an accident: the damages they receive are higher than their losses. Again, under negligence this is no problem, because careful injurers are not liable, and victims then do not receive compensation at all. Under strict liability, the problem indeed can occur, but here the defence of comparative or contributory negligence counters a too low care level of the victim. It is, however, not possible to address the possibly excessive activity level of the victim.
  So, average damages may result in incorrect incentives for victims under a rule of strict liability: victims with above-average losses may take too much care and engage in the activity too little, while victims with below-average losses may engage in the activity too often. If risk aversion of the victims is also included in the analysis, the first problem is exacerbated while the second is mitigated.

  II Compensation

  1 Compensation and Speed

  Victim satisfaction can be affected by a variety of different elements.¹³x On the aspects that constitute victim satisfaction, see Akkermans, A., ‘Reforming Personal Injury Claims Settlement: Paying More Attention to Emotional Dimension Promotes Victim Recovery’, 8 (26 February 2009). Available at SSRN: https://ssrn.com/abstract=1333214 or http://dx.doi.org/10.2139/ssrn.1333214. One of those is the speed at which a solution can be reached. Generally speaking, being involved in a compensation process for a long time can negatively influence the victim’s well-being.¹⁴x Shuman, D.W., ‘When Time Does Not Heal: Understanding the Importance of Avoiding Unnecessary Delay in the Resolution of Tort Cases’, Psychology, Public Policy, and Law, 2000, 6, 880-897 with further references on the effect of delays; Elbers, N.A., Akkermans, A., Cuijpers, P. and Bruinvels, D.J., ‘What Do We Know about the Well-being of Claimants in Compensation Processes?’ Recht der Werkelijkheid, 2012, Vol. 33(2), 65-78. This is also called ‘secondary victimization’ and refers among other aspects to the lack of control over procedures and lack of information.¹⁵x See Akkermans (2009, 3). The extent to which a speedy solution is possible at all will depend on the type of case. One important way to increase the speed of the procedure is to use a system of standardization that would no longer be based on a detailed damage assessment in concreto but would rather work with an averaging based on categories of harm. Victims might be willing to accept less accurate compensation if speedy recovery, rather than a precise accurate outcome, matters to them.

  2 Compensation and Accuracy

  The principal idea for the compensation perspective is that each victim is fully compensated, that is, to be put back into the situation as it would have been had the infringement not occurred. Calculation and assessment challenges taken aside, the victim is supposed to be indifferent then between the previous and later situation. So to start with, this calls for an accurate assessment of individual victim loss.
  There is limited but important research on the prospect of averaging from the victims’ point of view. Importantly, in situations where victims ex ante do not know whether they will have a weak or a strong claim, they assess the expected value of their claim. Again, in expected terms, there is no difference between assessing each individual case (differently) on one hand and the average amount on the other. Hay and Rosenberg argue that when a victim has ex ante uncertainty on whether it will have a strong or a weak claim, the victim might favour averaging above a damage assessment in concreto.¹⁶x Hay, B.L. and Rosenberg, D., ‘The Individual Justice of Averaging’, The Harvard John M. Olin Discussion Paper Series, Discussion Paper No. 285, 2000, 34ff. Ex ante, the expected value of either averaging or assessment in concreto is the same, but (1) averaging leads to lower administrative costs (as a result of which there is more left for victim compensation) and (2) it provides more certainty (the victim knows the average amount that will be allocated irrespective of whether ex post it appeared that the victim had either a weak or a strong claim). Ex post, victims with a strong claim obviously prefer an in concreto assessment. This insight is the reason why – practical challenges aside – it is argued that the system would need to be mandatory, namely, due to the selection effect on victims with strong claims.¹⁷x Ibid., 15ff.
  In practice, as the second part of the article will show, systems which resemble mandatory ones do provide for the possibility of deviation (concretization). One could see this (deviation within a mandatory system) as the price to be paid for introducing an overall cheaper system of handling personal injury cases. Another explanation might be that perhaps this theoretical risk of a selection effect (victims with strong claims opting for individual assessment) might not materialize in practice. It seems, furthermore, possible to limit the occurrences of an ex post opt-out of averaging, precisely so as not to entirely undermine the ex ante incentives the system set.
  The fact that averaging is sometimes regarded as problematic because it may leave some victims undercompensated can be explained by the ‘natural’ focus of many lawyers on the ex post perspective: the loss has already occurred and the actual victim should be fully compensated. The emphasis is placed on victims with above-average losses, who are not likely to agree to averaging.
  This view however overlooks various issues. First, there will also be victims with below-average losses, who will likely agree on averaging ex post. The ex post opinion of victims about averaging will, therefore, depend on their individual circumstances. Second, as mentioned earlier, from the ex ante perspective (so before a victim knows whether they will have a strong or a weak claim), victims overall prefer averaging because the lower system costs leave more money available for compensation, and it avoids the risk of victims ending up with a weak claim ex post.
  Here we need to mention an important specification: this latter point holds for all differences between cases, except the magnitude of pecuniary losses. Victims ex ante prefer averaging when it comes to, for example, the strength of the proof, the victim-friendliness of the legal rules, the burden of proof, and the size of the pain and suffering damages, because this avoids the risk of ending up in a situation where they do not get any compensation at all because they cannot meet the high burden of proof or because the legal system does not recognize their type of losses. However, regarding the size of the pecuniary losses, they prefer individualization, because averaging creates the risk that they ex post turn out to have suffered above-average pecuniary losses for which they only receive average compensation.¹⁸x Ibid., 37ff.
  Importantly, this preference for accuracy regarding pecuniary losses has to be balanced against the preference for averaging regarding all other differences and against the lower system costs of averaging. This balancing will likely not result in a preference for a fully individualized system. For example, a victim with high pecuniary losses but weak proof regarding causation may, in an individualized system, remain empty-handed because the causal link between the tort and the losses is not established. This victim may prefer an averaging system which applies proportional liability so that they at least receive average compensation instead of nothing. In general terms, the more important and significant differences there are in pecuniary losses between victims, the stronger the arguments for individual assessment become. However, if insurance against pecuniary losses for victims is available, this whole issue becomes less important, because the insurance then covers the risk of suffering above-average pecuniary losses. It is interesting to note that such first-party insurances often already depart from fully covering individualized losses, because they apply maximum coverage and/or fixed percentages (of, for example, functional disability). In any case, in our view, it is important not to solely focus on the ex post preference of victims with above-average losses, because this neglects important benefits of averaging for the group of victims as a whole.
  This implies that victims ex ante have a preference for averaging when it comes to pain and suffering damages, because these target non-pecuniary losses. This aspect is especially relevant for the topic of personal injury. For the pecuniary losses, there can still be reasons to assess them on an individual basis, especially if the assessment costs are not so high. The costs of medical treatment and possible costs of adapting the house to lasting disabilities due to injuries are relatively easy to assess and can differ significantly between cases, so averaging of such costs makes little sense.¹⁹x In fact, in many European legal systems, the pecuniary losses are often covered by social security carriers (e.g. the medical costs are paid by health insurers) as a result of which most debates in tort cases anyway take place concerning non-pecuniary losses. But pain and suffering damages connected to injuries are difficult, if not impossible, to assess on an individual level. Therefore, it is desirable to assess them in a more standardized manner, for example, with the use of tables which connect a certain range of compensation amounts to certain types of injuries. Within this range, some fine-tuning is still possible through using the circumstances of the case, but the general magnitude of compensation amounts is standardized. Hence, victims with comparable injuries then receive the same amount in pain and suffering damages, which is not assessed in detail individually.

  III Interim Conclusion

  Summarizing: the tendency that can be noticed in many legal systems to speed up adjudication of personal injury claims can be understood from both deterrence and compensation perspectives. Delayed compensation may dilute the incentives for prevention by the tortfeasor, and it could equally lead to secondary victimization simply due to the lengthy procedure and reduced victim satisfaction. Standardization of compensation can be one of the tools (in addition to introducing ADR) to speed up this compensation. Also, this standardization can to a large extent be supported from an economic perspective. From a deterrence perspective, the accuracy of damage assessment is not needed if the tortfeasor does not have information about the size of the losses they may cause and also cannot acquire this information against acceptable costs. It is then sufficient that damages are on average accurate, but that does not imply that damage assessment needs to be accurate in concreto in every specific case. In other words, the social value of accuracy is low, at least lower than the private value. From the society’s perspective, the major advantage of standardizing the compensation is that administrative costs can be lower. From the victim’s perspective averaging can also be attractive, as long as there is a margin with sufficient room for individualization. The individualization can be expected to take place anyway as far as the pecuniary losses are concerned simply because (e.g. concerning income losses) individualization is possible at relatively low administrative costs; for the non-pecuniary components of the damages, standardization is particularly advantageous. But in any case, the individual situations of victims could still be taken into account in a model of standardization. This way, a balance is struck between the advantages of standardization and the advantages of individualization. In this balancing act, it would only be important to avoid gross and systematic over- or under-estimations.
  If the victim can influence the probability of incurring personal injury, in light of the legal concepts of damage mitigation, an entitlement to full compensation might again not be in order. From a law and economics point of view that share of the harm needs to be internalized by the victims. However, a defence of comparative or contributory negligence is a better instrument than providing systematically incomplete compensation, which would under-incentivize the tortfeasor.
  From both deterrence and compensation perspectives, a look at administrative costs signals immediately that optimization in light of costs incurred should be aimed for, rather than seeking perfection irrespective of costs. When seeking to flesh out this angle, it becomes clear that a wrongdoer receives the right behavioural incentives if courts impose compensation payments that are on average correct. Victims are supportive of averaging as well under certain conditions, and their satisfaction is determined by striking a balance between speed and accuracy. Hence, from both perspectives total accuracy may not be the superior guideline.
  As an aside, satisfying two goals simultaneously with one instrument is obviously more challenging than the satisfaction of one goal only.²⁰x Jan Tinbergen mentioned that N-problems require N-solutions (referring to the fact that national banks should employ an equal number of instruments as the number of political targets (Tinbergen, J., The Theory of Economic Policy, North-Holland Publishing Company, 1952). In this particular case, we examined whether it is possible to reach the twin goals of speedy decision-making and accuracy. This section mainly served the purpose to motivate the view that from both core perspectives it is worth looking into legal systems where averaging compensation is applied with a view to speeding up processes. This averaging does not have to be at the expense of deterrence, and it even has several advantages for victims themselves in many situations. Moreover, one could (probably incorrectly) assume that a lengthier procedure would necessarily lead to more accuracy in damage assessment, but obviously that should not always be the case. The procedure could lengthen with the involvement of representatives of victims and injurers (the lawyers); however, such involvement may increase only the length of the procedure but not improvethe accuracy of damage assessment in any event. Also from that perspective, lengthier adjudication does not always increase accuracy (if that were already desirable). The fact that some victims ex post will receive less compensation than they would have under individual assessment of losses does not detract from the advantages of averaging we have discussed.

• C Personal Injury Compensation in Belgium, Ireland and Sweden

  Several legal systems have changed the resolution of disputes concerning personal injury in remarkable ways, compared to a traditional in concreto damage assessment by the courts. One important step consists of introducing standardized amounts of compensation; in other words, opting for averaging instead of accurate individual damage assessment. The other reform consists in increasing the speed of compensation by introducing ADR mechanisms. Those systems have been introduced in a variety of legal systems (e.g., in France and Spain). In this contribution as case studies we describe the examples of Belgium, Ireland and Sweden which, as we will show, have some similarities as well as differences, in how they have reformed dispute resolution in personal injury cases. Although standardization has ‘most use’ for non-pecuniary losses, to some extent it could be used for pecuniary losses as well. In the three examples we will provide, standardization also relates to pecuniary losses, but given our focus on non-pecuniary losses we will mostly focus on those.

  I Belgium

  In personal injury cases under Belgian law, many cases are settled between the victim and the insurer (of the responsible offender). The model is nevertheless one where importance is given to the intervention of the judge and the legal profession. If a victim in Belgium does not receive satisfaction from the insurer, the first logical step will be to go to a lawyer. Belgium does not have a system where non-lawyers (personal injury councillors) also intervene to settle personal injury disputes. The victim will, therefore, in principle call on a lawyer, which also means that there are relatively many legal proceedings in Belgium, though extrajudicial costs (except for the system of limited legal compensation) are not reimbursed.²¹x Van Dort, R., Dikke Pech In ’t Buitenland, Maastricht, 2020 (ISBN 978-94-92741-34-9), 37. Note also that the burden of proof in essence fully comes to rest with the claimant side.²²x Cousy, H. and Droshout, D., ‘Compensation for Personal Injury in Belgium’, in Koch, B.A. and Koziol, H. (eds.), Compensation for Personal Injury in a Comparative Perspective, Vienna, Springer, 2003, 37-75, 65. Strict liability regimes and legal presumptions come into play (especially as far as traffic accidents are concerned). However, the victim still needs to prove the existence and the actual amount of damage. Lawyers in Belgium are paid on an hourly basis. In the neighbouring country, the Netherlands, success fees are being experimented with, by allowing lawyers to multiply their normal hourly tariff in case of success with a factor 2-2.5 (depending on some details), with a maximum of 25%-35% of the outcome in cases of personal injury and fatal cases.²³x Faure, M.G., Fernhout, F.J. and Philipsen, N.J., Resultaatgerelateerde Beloningssystemen voor Advocaten. Een Vergelijkende Beschrijving van Beloningssystemen voor Advocaten in een Aantal Landen van de Europese Unie en Hong Kong, The Hague, Boom Juridische Uitgevers, 2009, 14. Such success fees are, however, unknown in Belgium.
  The Belgian model is particularly interesting because the judges who are primarily involved in claims settlement (judges in the courts of first instance, justices of peace, and police judges)²⁴x The Union Nationale des Magistrats de Premiere Instance/Nationaal Verbond van de Magistraten van Eerste Aanleg and the Union Royale des Juges de Paix et des Juges de Police/Koninklijk Verbond van Vrede- en Politierechters. took the initiative for the first time in 1995 to draw up a so-called indicative table. This could to some extent be considered as a form of judicial activism.²⁵x Cornelis, L., ‘De toegevoegde waarde van rechterlijk activisme: uitdagend (3)’, in Vansweevelt, T. and Weyts, B. (red.), De indicatieve tabel 2016: kansen en kritiek, Antwerp, Intersentia, 2018, 129-139. The original objective was solely to help judges with the damage assessment by proposing a standardization of the various damage components as well as providing a detailed assessment model. The drafters, therefore, wanted to provide more legal certainty and create more uniform case law regarding compensation for personal injury.²⁶x Boyen, A., ‘Forfaitaire vergoedingen en de (privé) indicatieve lijst’, De Verzekering, 2002, nr. 338, 55-76, 57. The table’s importance in legal practice has now become extraordinarily high.²⁷x Durant, I.C., ‘Belgium’, in Koziol, H. and Steininger, B.C. (red.), European Tort Law 2010, Berlin, Walter de Gruyter, 2011, 30-60, 53. Although the table is only indicative and, therefore, not mandatory,²⁸x Boyen (2002, 60); Court of Cass. (Belgium) 11 September 2009, Nieuw Juridisch Weekblad 2010, 25 with case note by G. Jocqué. See also Durant, I.C., ‘Belgium’, in Oliphant, K. and Steininger, B.C. (red.), European Tort Law 2012, Berlin, Walter de Gruyter, 2013, 57-93, 85. it is in principle applied by the judges of the facts when handling personal injury cases. If sufficient proof is available, deviations from the table are justified.²⁹x So Vansweevelt, T. and Weyts, B., ‘De indicatieve tabel 2012: van (te) normerend naar betwist?’ R.W., 2014/2015, Vol. 78, 243-253, 245. Since 1995, the table has been amended six times. The current version (from 2020) is the eighth.³⁰x https://www.schadeweb.be/sites/default/files/indicatieve-tabel-2020-tableau-indicatif-2020.pdf Ideally, the table is revised every four years. The table distinguishes three types of damage³¹x Vansweevelt and Weyts (2014, 247-248).: (1) Bodily harm; (2) damage to objects and costs (this mainly concerns vehicles); and, finally (3), interests and advance payments. By making this distinction, the indicative table is in line with the summa divisio in Belgian damages law between personal injury and property damage, which also has its roots in case law. Within personal injury, Belgian law (and, therefore, also in the indicative table) makes a distinction between temporary damage and permanent damage.³²x Van Dort (2020, 19 en 23-29). An important role in medical liability cases is played by the medical expert.³³x Ibid., 15: on the importance of the medical file. Whereas the table serves as the basis for damage assessment, court-appointed medical experts are a standard element of litigation.³⁴x Vansweevelt and Weyts (2014, 248); Ulrichts, H., Schaderegeling in België, Mechelen, Wolters Kluwer, 2018, 65. Their duties are established in the text accompanying the indicative table. Ideally, both the victim and the injurer bring in experts in the procedure. A third expert may then be needed to take the final decision.
  Although in Belgium there is general satisfaction with the indicative table (advantages include, among other things, better predictability of compensation, legal certainty³⁵x Weyts, B., ‘De indicatieve tabel 2016: krachtlijnen en enkele kritische bedenkingen’, in Vansweevelt, T. and Weyts, B. (red.), De indicatieve tabel 2016: kansen en kritiek, Antwerp, Intersentia, 2018, 1-16, 5. and reduction of legal proceedings³⁶x There is more out-of-courts settlement: Vandenwijngaert, G., ‘De indicatieve tabel, vloek of zegen voor de verzekeraar?’ In Vansweevelt, T. and Weyts, B. (red.), De indicatieve tabel 2016: kansen en kritiek, Antwerp, Intersentia, 2018, 121-128, 128.), there is also some criticism. This criticism focuses mainly on (1) the fact that certain normative choices made in the table have not been further explained,³⁷x Jocqué, G., ‘Enkele bedenkingen bij de nieuwe indicatieve tabel’, R.W., 2012-13, afl. 38, Vol. 76, 1519-1520, 1520. (2) there is a risk of a mismatch between payable compensation amounts and actual damage suffered³⁸x Among others, Van Steenbergen, J., ‘Kritische bedenkingen bij de indicatieve tabel’, in Van den Bossche, M. (red.), De indicatieve tabel. Een praktisch werkinstrument voor de evaluatie van menselijke schade, Brussels, Larcier, 2001, 16. and (3) the table is drawn up exclusively by judges and it is not clear to what extent the views of other stakeholders (consumers, victims’ associations, lawyers, insurers and academics) are taken into account.³⁹x Ulrichts, H., ‘De nieuwe indicatieve tabel: de stem van de verzekeraar’, De Verzekering, 2002, nr. 338, 84-104, 84.

  II Ireland

  Ireland opted for a radical change in 2003 by, in principle, having personal injury claims settled through the Personal Injuries Assessment Board (PIAB) as set out in the PIAB Act 2003.⁴⁰x www.irishstatutebook.ie/eli/2003/act/46/enacted/en/index.html. In 2007, an important reform concerning the fee system took place, see Quill, E., ‘Ireland’, in Koziol, H. and Steininger, B.C. (red.), European Tort Law 2007, Vienna, Springer, 2008, 352-372, 352-353: it sought in particular to reduce the incentives to reject the assessment without justification. This board since then exists as a necessary pre-step or, ideally, as an alternative to the court structure. The PIAB offers a cheap and low-threshold system on which a victim who is unable to obtain relief from the tortfeasor (or their insurer) can rely. The applicant must pay an amount of €45 for an online request and €90 for a paper request. The respondent has to pay a fee of €600, and the system is largely financed through those fees. The procedure is entirely written and can, in principle, be followed by a victim without a solicitor,⁴¹x Quill, E., ‘The Functioning of the Personal Injuries Assessment Board in Ireland’, Journal of Personal Injury Law, 2019, 74-81, 76. Originally it was even forbidden, which was later argued to be unlawful. although solicitors are increasingly intervening in the procedure in Ireland, too.⁴²x In essence 90%, see Aldus Cost of Insurance Working Group, Report on the Cost of Motor Insurance, January 2017, 92. https://assets.ie/6254/060219172049-067f7ed921f44343a7f41144ac3d4940.pdf. If involved, as a first step, the victim’s solicitor will usually contact the responsible offender or their insurer. The aim is to reach a settlement through negotiations. A PIAB procedure is only a next step. The PIAB is an administrative body. Anyone who has suffered any form of personal injury to which the law applies (which in fact applies to all forms of personal injury) is required to submit an application to the PIAB in accordance with Section 11 of the PIAB Act 2003. Section 10(b) even forbids the victims to go directly to the court if an application has not been made to the PIAB first. Hence, the authorization to litigate in court can only be obtained by the PIAB. Parties have the option to accept or reject an assessment by the PIAB. However, if it is accepted by both sides, that assessment has the same legal effect as a court judgment and can, therefore, be enforced.⁴³x For a summary, see Quill, E., ‘Ireland’, in Koziol, H. and Steininger, B.C. (red.), European Tort Law 2003, Vienna, Springer, 2004, 245-265, 246-247. If a claimant rejects an assessment by the PIAB but the assessment was accepted by the defendant, the claimant cannot claim costs from the defendant when the amount eventually awarded by the court (or accepted by the claimant in a settlement) is not higher than the assessment made by the PIAB. This is clearly meant as a mechanism to incentivize claimants to accept the PIAB assessment. Whether it is effective is doubtful, because courts often award higher damages than the PIAB.
  One of the formal tasks of the PIAB according to the law is to establish a Book of Quantum, that is, standardization. After one year in action, in 2004, the PIAB created the first Book of Quantum; a second edition was published in 2016, and, in 2021, the Book of Quantum was replaced by the Personal Injury Guidelines.⁴⁴x See https://claimsauthority.ie/injury-guidelines/. An important difference between the Book of Quantum and the Guidelines is that application of the latter is mandatory, so the trial judge must refer to these Guidelines. Any departure from the Guidelines must be justified.⁴⁵x See, for example, https://kennedyslaw.com/thought-leadership/article/the-personal-injuries-guidelines-2021/ and www.addleshawgoddard.com/en/insights/insights-briefings/2021/litigation/all-change-new-personal-injury-guidelines/. Under the Book of Quantum, courts could deviate from it,⁴⁶x Section 22(2) of the Civil Liability and Courts Act 2004: ‘Subsection (1) shall not operate to prohibit a court from having regard to matters other than the Book of Quantum when assessing damages in a personal injuries action’. without giving a reason.⁴⁷x Quill, E., ‘Ireland’, in Karner, E. and Steininger, B.C. (eds.), European Tort Law 2019, Berlin, De Gruyter, 2020, 289-313, No. 32, 18. In some cases, the courts ignored Section 22 of the Civil Liability and Courts Act and, thus, determined compensation without even mentioning the Book of Quantum.⁴⁸x Quill, E., ‘Ireland’, in Karner, E. and Steininger, B.C. (red.), European Tort Law 2015, Berlin, Walter de Gruyter, 2016, 282-303, 298. References to the Book of Quantum in court judgments are even ‘relatively sparse’ according to the literature.⁴⁹x McKeown, A., ‘Court of Appeal: More Regard Should Be Had to Book of Quantum’, Irish Legal News, 13 August 2020, www.irishlegal.com/article/court-of-appeal-more-regard-should-be-had-to-book-of-quantum. This will change under the Guidelines because, as said, referencing is mandatory.
  The Book of Quantum was based on a statistical analysis (by consultants) of amounts awarded on average in case law for certain types of damage items. It contained guidelines which were drafted by Verisk Analytics, an international consultancy firm, specialized in statistical analyses. They examined 51,000 terminated personal injury cases, dealt with in 2013 and 2014, by either the courts, the PIAB, or in settlements with insurers. The Personal Injury Guidelines seem to result in significantly lower awards than the Book of Quantum.⁵⁰x Idem.
  Unlike in Belgium, the Book of Quantum in Ireland was not so much normatively intended to promote the further development of case law, but rather a restatement of what is usually attributed in case law with regard to the individual damage items.⁵¹x It has been criticized that also settlement amounts were included, see Quill, The Functioning…, 79. The Guidelines, however,

  seek to promote a better understanding of the principles governing the assessment and award of damages for personal injuries with a view to achieving greater consistency in awards, even though cases will invariably have their own unique features.⁵²x Personal Injury Guidelines (2021, 5).

  Irish law makes a first broad distinction between two types of damages. In the first place, there are general damages to cover harm resulting from pain and suffering. The Guidelines give an indication of the pain and suffering damages that can be awarded for each type of injury, including injuries resulting in foreshortened life expectancy. In addition, there are special damages for loss of earnings, medical bills, loss of future income, and costs of future medical care. In addition, Irish law also provides for so-called aggravated damages that can be awarded in exceptional circumstances when the damage to the victim as a result of the conduct of the offender has increased. Exemplary or punitive damages may be awarded in exceptional cases for punishment and deterrence.⁵³x Quill, E., ‘Ireland’, in Wininger, B., Koziol, H., Koch, B.A. and Zimmermann, R. (red.), Digest of European Tort Law. Vol. 2: Essential cases on damages, Berlin, Walter de Gruyter, 2011, 39-41, 41.
  There is only limited experience with the Guidelines yet, so we cannot comment on this. Regarding the Book of Quantum, there is much more experience: the PIAB’s award was accepted in 60% of the cases, but victims (also under pressure from their solicitors) increasingly went to court because they usually award higher amounts than the Book of Quantum. Note, therefore, that compensation awarded by the board and by courts does differ. Compared to court judgments, the procedure used by the PIAB was on average much shorter (9 versus 36 months⁵⁴x PIAB Annual Report 2004. According to Quill, the real term is closer to 1.5 years and there is also the risk that a significant backlog would develop, see Quill, E., ‘Ireland’, in Koziol, H. and Steininger, B.C. (red.), European Tort Law 2006, Vienna, Springer, 2007, 281-298, 281-282.) and cheaper (costs are only 10% of the total compensation versus 46%⁵⁵x Quill, E., ‘Ireland’, in Koziol, H. and Steininger, B.C. (eds.), European Tort Law 2005, New York: Springer, 2006, 348-367, 349-350.) than in the case of settlement through the courts. In reality, the PIAB procedure may only be the first step, followed by litigation in the courts. Recent data confirmed the superiority of resolution via the PIAB compared to the court: the 2018 PIAB Annual Report mentions an administrative cost of the PIAB of 6.1% of the value of accepted awards compared to average legal costs of 63% of the amount of compensation in the court.⁵⁶x Quill (2020, 306-307). The 2020 PIAB Annual Report mentions a cost of PIAB resolutions of 11% and litigation 66% (Quill, E., ‘Ireland’, in Karner, E. and Steininger, B.C. (eds.), European Tort Law 2021, Berlin, De Gruyter, 2022, 283-305, 298-299). Recent data also confirm speedier decision-making via the PIAB (on average 2.3 years) than via litigation (an average of 4.1 years).⁵⁷x See Regulatory Impact Analysis. Personal Injuries Resolution Board Bill 2022, available at: https://enterprise.gov.ie/en/legislation/legislation-files/ria-personal-injuries-resolution-board-bill-2022.pdf, p. 5 (last accessed 29 November 2022).
  It is important to emphasize that standardization via the PIAB was apparently not undisputed, with the result that the Book of Quantum was often ignored by the courts in practice. There was also criticism that the courts themselves were in no way involved in the establishment of the standard.⁵⁸x Cost of Insurance Working Group (2017, 116), Recommendation 18; see also Quill, E., ‘Ireland’, in Karner, E. and Steininger, B.C. (red.), European Tort Law 2017, Berlin, Walter de Gruyter, 2018, 283-309, 303. Notice that in Ireland the criticism was exactly the opposite as in Belgium, where the criticism related to the fact that the indicative table is drawn up exclusively by judges. This is fundamentally different in the new situation, because the Personal Injuries Guidelines Committee consists of judges only.⁵⁹x Quill, The Functioning…, 79. Because this change is still recent, it is difficult to estimate its consequences. But it is in any case clear that the magistracy will be more closely involved in the drawing up of the standard, more in line with the Belgian model (with the important difference that in Ireland, unlike in Belgium, a legal basis for the standard exists). On 2 August 2022, the Personal Injuries Resolution Board Bill 2022 was published. The goal of the Bill is to give the PIAB a new function (to offer mediation as a means of resolving a claim), to promote public awareness and to provide PIAB with additional time to assess claims where an injury is to settle rather than releasing to litigation.⁶⁰x www.gov.ie/en/publication/55ae3-personal-injuries-resolution-board-bill-2022/ (last accessed 29 November 2022).

  III Sweden

  The Swedish system differs significantly from the Belgian and Irish models, since in Sweden liability law plays no role whatsoever in compensation for personal injury. For specific types of accidents (traffic accidents, medical accidents, pharmaceuticals and work-related accidents), direct insurance policies taken out by involved parties (motor vehicle owners, medical professionals, pharmaceutical manufacturers and employers) have been created whereby the insurer directly, basically on a no-fault basis, compensates the victim’s damage.⁶¹x Nyquist, S. and Persson, E., ‘Sweden’, in Faure, M. and Hartlief, T. (red.), Financial compensation for victims of catastrophes. A comparative legal approach, Vienna, Springer, 2006, 227-260, 227; Dufwa., B.W., ‘Compensation for Personal Injury in Sweden’, in Koch, B.A. and Koziol, H. (red.), Compensation for Personal Injury in a Comparative Perspective, Vienna, Springer, 2003, 293-324, 297. It is rather ambiguous that literature still refers to a strict liability mechanism, although it in fact constitutes a duty to compensate upon the liability insurer which exists irrespective of any liability of the insured. An examination of the liability of the insured is in other words no prerequisite for the duty to compensate by the liability insurer. The victim, therefore, no longer has to prove a fault on the part of a possible responsible offender but can rely directly on insurance protection. In principle, a victim, therefore, reports to the (direct) insurer and the victim is compensated by that insurer. The victim will receive full compensation according to the rules of tort law. About 60%-70% of all personal injury cases are handled by insurers in less than two months.⁶²x Interview with Mr Stefan Andersson, from the insurance company Folksam, on 31 August 2020. This is possible due to the comprehensive compulsory insurance system.⁶³x Strömbäck, E., ‘Personal Injury Compensation in Sweden Today’, Scandinavian Insurance Quarterly, 2000, 89-106, 89: only 10% would not have a liability insurance. Only if the victim is not satisfied with the compensation provided by the insurer will the case get to one of the many Boards, such as the Traffic Injury Board (TIB), that have been set up with a view to compensation. This Board then formulates advice that is in principle always followed by the insurer.
  Liability law no longer plays any role. Compensation takes place primarily through social security (which would pay the medical costs and an important part of the loss of income⁶⁴x Macleod, S., Urho, M. and Hodges, Chr., ‘Sweden’, in Macleod, S. and Hodges, Chr. (red.), Redress Schemes for Personal Injuries, Oxford, Hart, 2017, 167-189, 186.), secondarily through insurance, and only for those who would not be covered by any insurance, liability law would still play a role.⁶⁵x Interview with Henrik Jerkrot, from the insurance company, on 28 August 2020. This de facto means that the most important part of the damage for which the special insurance models are responsible is the compensation for pain and suffering damages.
  It is the aforementioned Boards that have established a standard: tables that give indications about appropriate compensation for certain loss items. These tables are also used in case law. The members of such boards involved in this standard-setting are judges, former judges, insurers (with law training and extensive experience in claims handling) and laymen. It is precisely because of the broad composition of the boards that their advice (and, therefore, also the tables produced by them) can count on broad support.⁶⁶x Nyquist and Persson (2006, 251). The TIB plays an important role in this, which has also provided detailed rules regarding its operations on its website. There are also other boards (for injuries other than traffic), but the tables developed by the TIB are apparently also followed for injuries other than traffic. Even the Swedish Supreme Court often uses the tables de facto.⁶⁷x Ibid. The tables generally indicate a range of possible compensations for various loss items. The decision on the exact amount of damage is then determined, taking into account the medical report and, therefore, the individual situation of a victim. The TIB website clearly states that each damage assessment is done on an individual basis.⁶⁸x TIB, www.trafikskadenamnden.se/vid-skada/, 19.
  In general, personal injury is determined via the tables and standards as developed by the TIB.⁶⁹x Sandell, H., ‘Sweden’, in Koziol, H. and Steininger, B.C. (red.), European Tort Law 2002, Vienna, Springer, 2003, 393-406, 396. For example, in the year 2000, the Swedish Supreme Court determined that relatives of a deceased victim could receive 25,000 SEK (about €2,700) in compensation. This is based on the assumption that surviving relatives experience grief at least one year after death and one year of distress corresponded to SEK 25,000 according to the TIB standard at the time. A higher compensation is possible, but proof must be provided that in the specific case the next of kin experienced more-than-average suffering as a result of the death.⁷⁰x Ibid., 394. Compensation for damages takes place according to the tables of the TIB, which provide standardized amounts based on objectively calculable criteria. The amount depends, for example, on the length of hospitalization, the pain suffered during the treatment, the period during which the victim was bedridden, and other factors that can be objectively determined.⁷¹x Strömbäck (2000, 99). The TIB itself indicates that they adjust their compensation tables every year.⁷²x www.trafikskadenamnden.se/vid-skada/, 1: ‘Each year, the Traffic Injury Board determines its compensation tables. They are also used by courts and when calculating compensation for other insurances, for example accident insurance.’
  The Boards are financed by insurers, but there is also government intervention as the composition of the Boards is approved by the Swedish government. Decision-making by the Boards is cheap (in principle free of charge for the victim) and fast. Replacing the courts in essence by the Boards has led to a major reduction in costs.⁷³x Dufwa (2003, 296). Interview with Mr Henrik Jerkrot, Personal Injury Lawyer in Sweden, on 31 August 2020.

• D Comparison and Analysis

  When the cursory overview of the three legal systems that has just been presented is examined more closely, it is striking that the models in which the standardization of personal injury has been established differ considerably between the relevant legal systems. This standardization cannot be viewed separately from the various models that have been developed for the settlement of personal injury claims.
  In the first place, there are important differences between the three systems with regard to the settlement of personal injury claims. In Belgium, in principle, this is done consensually between the victim and the insurer of the responsible offender. However, when no agreement can be reached between them, the case is taken to court. In Ireland, the model is different, in the sense that any personal injury case in which the victim is dissatisfied with the insurer’s offer must pass through the PIAB. A victim cannot directly bring the case before the court, but must have an ‘authorization’ from the PIAB for this purpose. About 60% of personal injury cases end with a recommendation from the PIAB; in other cases, victims still go to court in the hope of getting a higher compensation, which they almost always succeed in. In Sweden, the role of the judge is even smaller than in Ireland. In principle, the case is once again settled between the victim and the insurer. But in the absence of an agreement, the matter can be submitted to one of the many Boards. In addition, in Sweden there are also legally prescribed cases where the case has to be submitted to the Board (e.g. if a traffic accident leads to injuries of more than 10% disability). The advice of a Board in Sweden is always followed by the insurer.
  In the countries in which an ADR model exists (the PIAB in Ireland and the Boards in Sweden), it appears that the costs are lower and the duration shorter compared to personal injury cases dealt by the courts. The duration of procedure before the PIAB in Ireland is on average a lot shorter than the duration of litigation before the courts (9 versus 36 months). The administrative costs of dispute resolution are equally substantially lower: with the PIAB, the tertiary costs are only 10% of the total compensation, whereas this is 46% in case of compensation via the courts.⁷⁴x Quill, E., ‘Ireland’, in Karner, E. and Steininger, B.C. (eds.), European Tort Law 2018, Berlin: De Gruyter, 2019, 299-325, 319-320. A report by the Irish Central Bank of 2018 assessed that on average €24,208 was awarded to victims by the courts with an average legal cost of €15,139, which is 63% of the total compensation. The report also indicated that the average duration of a procedure with PIAB in the period 2015 to 2018 was 2.5 years, whereas this was 4.2 years before the courts.⁷⁵x Quill (2020, No. 31, 18). The PIAB could count on an average consumer satisfaction of 90%.⁷⁶x Quill (2006, 349-350). This high degree of consumer satisfaction is striking, as in fact only 60% of the PIAB assessments are accepted by victims. The reason victims still do not accept the PIAB assessment in 40% of cases is that they expect that the courts will award higher damages, but it apparently does not lead to a lower customer satisfaction. In practice, when cases go to court, insurers often make an offer to the claimant for an amount just above the PIAB assessment in order to avoid the high costs of a legal procedure.⁷⁷x Quill, The Functioning…, 79.
  A similar picture concerning customer satisfaction also appears from Sweden. The procedure before the Boards is considered faster and cheaper than the court system. The most important Board, the TIB, publishes monthly statistics indicating the average treatment time. Recently that varied from 58 days in March 2021 to 50 days in January 2022.⁷⁸x TIB, www.trafikskadenamnden.se/om-oss/#Link12167. There is a formal commitment that a case is dealt with by the Boards within a period of maximum 90 days. The procedure before the TIB is, moreover, free of charge for the victim.
  In Belgium, the legal profession has an important role to play in the settlement of personal injury, partly because in the event of a conflict the case must in principle be submitted to the court. The Irish PIAB was originally hesitant about the involvement of solicitors, but has had to drop that resistance (due to constitutional concerns), and solicitors are now increasingly involved in the procedure with the PIAB. Although the treatment for the Swedish Boards is easily accessible, lawyers can assist the victim in certain cases.
  In two of the three cases we examined (Ireland and Sweden), the standardization of personal injury claims was related to the specific model of personal injury adjudication. In Belgium, no specific ADR model was developed. The adjudication process also determines the authority that creates the standardization. In Belgium, it is the judges who have taken the initiative to establish a standard in the indicative table. It is also the judges who deal with personal injury claims in the first de facto instance (police and peace judges, as well as judges in the court of first instance) who have jointly worked out the indicative table. In Ireland, there is a legal basis for the Book of Quantum provided in the PIAB Act 2003 which expressly states that one of the functions of the PIAB was to create a Book of Quantum. This task was taken away from the PIAB; the Book of Quantum was withdrawn in 2021, and, instead, Personal Injuries Guidelines were published by the Judicial Council. However, the PIAB remains important as the primary body for personal injury settlement through which all cases must pass. In Sweden, it is the Boards involved in claims settlement that have created the tables. There is, thus, a difference between the three systems concerning the authority that created the standardization: in Belgium, the tables are developed by the judges; in Ireland, until recently by the (administrative) PIAB, and now by a committee of judges; and in Sweden, by the Boards consisting of judges, consumers and insurers. The heads of damages to which the tables apply exactly also differ per country, but in general it concerns both material (mainly income) damage and immaterial damage. The form of the tables differs, but there are basic principles that apply to each damage component. For example, the age of the victim is usually taken into account and a range is usually indicated at which a choice can be made between the highest and lowest amount, depending on the severity of the injury. The intention in the three systems is always that use is made of a medical report that has been drawn up on the basis of the criteria that can be found in the tables. In principle, the body responsible for determining the actual compensation can, therefore, award a specific amount for each loss item on the basis of the medical report and the table. In addition, the bandwidth is precisely intended to allow individualization within a standardization model, that is, to take into account the individual situation of the victim.
  The tables are extensively used in legal practice in Belgium. In Ireland, the Book of Quantum was used by the PIAB, but to a lesser extent by the judges. Even the legal requirement to state the amount that should be awarded according to the Book of Quantum was often ignored by judges. With the legal change due to the 2019 Act, Section 22 of the Civil Liability and Courts Act 2004 was amended, and courts must now have regard to the Personal Injuries Guidelines, and, importantly, where they depart from the guidelines, the courts must state the reasons for such departure in their decision. In Sweden, in principle, only the tables developed by the Boards are used, including by judges. A common denominator is that the tables in all three systems are regarded as indicative and not as binding. Even in Ireland, where judges were required to quote the amount from the Book of Quantum (even though they often ignore this), they could still deviate from the compensation amounts recommended by the Book of Quantum. Also according to the new rules, a departure from the Personal Injuries Guidelines is possible if the reasons for this departure are stated in the judicial decision.
  In view of the different formulations of the standards, there are also differences with regard to the precise intention of the amounts specified in the tables. In Belgium, the judges indicate what they consider to be reasonable compensation amounts for specific damage items. They naturally look at the amounts awarded in the past, but they also make a normative assessment of what they believe would be reasonable compensation for certain loss items. The indicative table can, therefore, also lead to a development in case-law. In Ireland, the Book of Quantum was much more a restatement of court decisions. The Book of Quantum was even created by consultants on the basis of a statistical analysis (which is also one of the criticisms that has recently led to a change in the law, so that standardization is no longer established via the PIAB). A mixed system seems to exist in Sweden, in the sense that on one hand previous rulings are taken into account, but the Boards also determine normatively what would be appropriate compensation for a particular loss item.
  Related to this is the question of whether the tables are dynamic or not. The current version of the indicative table in Belgium is the eighth. This table has, therefore, been regularly updated since 1995. In Ireland, only a second version of the Book of Quantum has seen the light since 2004 (in 2016) and the adaptations are, therefore, less dynamic (which also explains the criticism in case law). The Personal Injuries Guidelines Committee is to review the guidelines more frequently, at least every three years. In Sweden, the tables are adapted in the most dynamic way, as they are redefined every year.
  The question obviously arises of how the experience from the three countries relates to the economic insights on averaging. The information provided via the country studies only allows for some tentative and indicative conclusions. First of all, not all the case studies provide detailed information, for example, on whether standardization actually allowed an increase in the speed of the adjudication; in addition, we only discussed three examples (and not even in full detail) so one obviously has to be cautious, as it is not known to what extent these three examples are really representative for other systems where standardization has been introduced. Nevertheless, looking at the case studies in light of the law and economics insights from Section II, the following can be said: first of all, it was held that speed is desirable from both compensation and deterrence perspectives. The cases of Ireland and Sweden do indicate that the speed of adjudication has increased to a large extent compared to adjudication via the courts. For Belgium, there is no information on that point. One has to be careful, however, to deduce from this finding that it would be the standardization alone that would have been the cause of speedier decision-making. After all, in Ireland and Sweden, the standardization fits into an ADR model of decision-making. It is, therefore, rather the combination of standardization and ADR that might have caused the reduction of delays in adjudication. The effects are not disentangled. Although Belgium does not have an ADR model, it is held in legal doctrine (however, without providing precise data) that the standardization would have reduced the number of legal proceedings. More particularly, the increased legal certainty (by providing amounts in the indicative tables) would have increased the number of out-of-courts settlements. And in Belgium, this was realized without a specific ADR procedure. Although the evidence is obviously very limited, this may be an indication that also standardization as such can speed up decision-making (also when it is not combined with an ADR model of adjudication). The logic is obviously that standardization may provide information to all stakeholders in the dispute, as a result of which they know ex ante the margin within which a court would most likely decide the case. That could, therefore, reduce the incentives to take the case to court and could increase out-of-courts settlements.
  A second economic criterion is whether standardization would reduce the administrative costs of adjudication (Calabresi’s tertiary costs). Again, in Belgium, there is speculation that standardization would have reduced the costs of adjudication, and the same is true in the case of Sweden. Only for Ireland there is a clear indication that the administrative costs compared to total compensation are substantially lower under the PIAB adjudication than when the case goes to court. But again, standardization may have facilitated this. The most important reason is probably decision-making via ADR, rather than via a more adversarial adjudication via the court. Still, the legal doctrine in Belgium indicates that the standardization (via the indicative table) would have created more legal certainty by making the amounts of compensation more predictable and, thus, reducing legal proceedings and increasing settlements (and, therefore, presumably also reducing administrative costs).
  A third, and from an economic perspective, probably the most relevant question is whether the standardization leads to a situation where tortfeasors are on average exposed to the correct amount of the losses. Obviously, this is an almost impossible question to answer, as it would require assuming that it would be possible to indicate what for a particular type of injury a ‘correct’ amount of damages would be. Here, we wish to make our analysis a bit more loss-specific. In the theoretical part, it was outlined how averaging does not disadvantage victims in relation to non-pecuniary losses (unlike for pecuniary losses). For those types of losses, the argument in favour of standardization is stronger. Given the high uncertainty involved in assessing non-pecuniary losses, assessment costs may be high and standardization can be justified from an economic perspective. In the three countries examined, damages regarding pain and esthetical losses were assessed in a standardized manner which fits into the economic reasons for standardization mentioned earlier. We observed – in line with Section II – less standardization for pecuniary damages in the countries. For particular costs, such as medical costs in a case of temporary disability, the indicative table (e.g. in Belgium) allows the victim to provide a detailed overview of the costs, supported by evidence. But the table indicates a maximum amount which provides a safeguard against unacceptable high costs. Still, the table allows the victim to show its individual medical costs, supported with invoices. For loss of income on the other hand, which is obviously highly personal, an individualized approach is followed. It can also be understood that for these types of pecuniary losses an individual assessment is followed because (1) risk-averse victims might prefer this; (2) on this point of pecuniary losses, there can be substantial differences between victims and, in other words, also large divergences from the average and the actual values; and (3) for these types of pecuniary losses, the assessment costs are not prohibitively high.⁷⁹x It also makes economic sense to assess loss of income in an individualized manner, since there can be a wide variety in income losses, and this is obviously pecuniary. Averaging income would cause too much risk and potential losses for victims. There are, therefore, good reasons to keep the assessment of income losses largely individualized. However, there is de facto often some averaging in income losses, for example, as far as predictions are concerned about career development, retirement, aid and mortality risks. Those are often based on general statistics. But again, those statistical analyses will be applied to the individual wage of the victim. Even though we can obviously not argue that the cases would support that on average injurers would be exposed to the correct amount of damages, we do see that the heads of damages for which standardization is chosen correspond with the economic model.
  Fourth, a noneconomic, but still highly important argument in order to introduce standardization was the argument of victim satisfaction. Recall that the desire to speed up adjudication (avoiding lengthy trials) was inspired by the idea that this might increase victim satisfaction. On that point, the three cases unanimously argue that after standardization victim satisfaction has become higher. Again, for Ireland and Sweden, this may also be due to the ADR model of adjudication (which is of course facilitated by standardization), but in Belgium the argument is made that the standardization would have increased victim satisfaction, even though there is no ADR-system. But obviously also in Belgium, standardization increased the number of out-of-courts settlements by providing more certainty on the amounts of damages.
  Finally, the question could also be asked as to how the change to averaging personal injury claims could be done from a law reform perspective and which methods are preferred over others. Theoretically, the choice is between, on one hand, the legislator (who would fix the standards in legislation) and, on the other, the judiciary (to develop this in its case law or otherwise agree to it). One could also imagine more hybrid systems where, for example, a legislative mandate towards harmonization would be provided, but the fixation of specific standards would be delegated. And in that respect, still distinctions could be made between delegation to a technical board or to practitioners (mostly judges). From a law reform perspective, one could argue that the legislative route is always more complex, as it requires a majority within the legislative system to agree with the reform. The advantage of legislation might be that it provides a more solid basis for the reform. It could then be the authoritative document for judicial and ADR solutions. The disadvantage would be (certainly if legislation would amount to a statute whereby the legislator itself would also determine the standards) that it might not be sufficiently flexible and would be out-dated fast. The advantage of the alternative (determination by a board of judges) would of course be the increased flexibility. But the disadvantage would be that it would lack democratic legitimacy.⁸⁰x Also if an ADR body set standards for itself, this issue might emerge. But from a law reform perspective, an introduction of standardization, for example, through a common agreement between judges on which type of amounts to allocate for which type of injuries, could in theory be easier to accomplish than to do so via a legislative change.
  At the end of the day, there may not be one ideal ‘one size fits all’ solution, as the preferred method to realize this law reform may also depend on specific characteristics of the legal system where the reform would have to be implemented. That is also what can be observed in the countries that we examined. In Ireland, there was a legislative basis for the Book of Quantum. That should in theory have led to a greater acceptance, but in fact it did not, as the judges did not agree to the amounts that were determined in the Book of Quantum from a technocratic perspective. This shows the importance of a support base for the reform among those who have to work with it. In Sweden, the reform could be realized through a common agreement to create the TIB consisting of judges, consumers and insurers. In that country, it was apparently possible to create the reform based on a consensual agreement by involving all relevant stakeholders. This consensus safeguards the just-mentioned support base. In Belgium, the guidelines lack a formal legal basis, but it was simply the judges who agreed to create the tables and who equally adapt them as well. The Belgian example shows that it is apparently possible to create the reform also without a formal legal basis. The way it functions in Belgium has, moreover, the advantage of being flexible as regular adaptations of the tables are equally done. However, the tables are also criticized for the fact that it is only the judges who determine the amounts and that there is no involvement of other stakeholders (like academics or personal injury lawyers). In sum, there are many ways in which this standardization could be realized and the three examples we discussed show that there is not necessarily one optimal way of doing this, as it may largely depend, on one hand, on the trade-off between providing a solid statutory basis and flexibility and, on the other, on the specific characteristics of the jurisdiction where the law reform would have to be executed.
  To summarize, the three practical cases under investigation to a large extent are consistent with the theoretical insights that call the criterion of total accuracy into question. Importantly, even though they present differences in the institutional details of the systems, the three countries share the combination of ‘speedy adjudication’ with ‘standardization’, and scholarship reports positive effects in terms of speed and adjudication costs. Deviations from the standardized amounts are possible under certain conditions, in particular when assessing pecuniary loss. And, perhaps most importantly, they all report that this combination has led to higher victim satisfaction. Legal resistance against standardization may be caused by a too narrow focus on the ex post position of victims with potentially above-average losses. Moreover, an important reason for this resistance may also come from interest group politics, as personal injury lawyers may fear that they are less needed than in an adversarial model of adjudication via the courts. The clarity provided by standardization makes it obviously easier for victims and their representatives (not necessarily being lawyers) to reach an agreement with the (insurer of the) liable injurer, without the necessity of having a professional lawyer involved.

• E Conclusion

  We started by outlining how in various countries initiatives have been taken to speed up adjudication as lengthy trials are generally regarded as a problem. One of the methods introduced in particular jurisdictions to reach that speeding up has been the standardization of compensation. There have, however, been concerns that speeding up may come at the expense of accuracy. In this contribution we have analyzed from an economic perspective under which conditions accuracy is more important, and when it is less important. From the perspective of social costs, damages only need to be on average correct in order to provide optimal deterrence. Accuracy in the determination of damages in an individual case is, therefore, less important (at least from the economic deterrence perspective) than is sometimes presumed. Only systematic errors should be avoided. The value of the deterrence perspective, of course, depends on the extent to which the individual tortfeasor themselves are really directly exposed to compensation payments or whether this takes place via a liability insurer. We equally explained that in many countries that rely on the court system to engage in an individual assessment of damages, the ‘price’ the victim has to pay is, especially in complicated cases, often that they have to wait for a long time for compensation whereby the various procedures and the time spent waiting can lead to additional suffering and secondary victimization. Also from an economic perspective, a long time-lapse before the injurer is bound to pay compensation is undesirable, as it may endanger the incentives for prevention. In other words, both from prevention and compensation perspectives, it could be argued that the speed with which a victim receives compensation should be increased. In other words, the endeavour to speed up litigation by ‘sacrificing’ some accuracy can be supported from an economic perspective. And that is exactly what several countries have done. Standardizing (averaging) damage amounts is one way of increasing speed; another one is introducing ADR.
  The three countries we examined give rise to optimism that averaging actually brought about the positive effects that theory would predict. New forms of standardization seem to have led to increased predictability and, thus, stimulated settlements. Victims could better understand why particular damages are awarded. The effect on tertiary costs seems to be positive. In all countries where averaging is applied, there is, moreover, room for individualization. The standards provide a margin of compensation for a particular head of damages, within which the decision-maker (ADR board or judge) can fine-tune the amount of compensation to be awarded, taking into account the specific circumstances of a victim. In line with the theoretical section, more room for individualization was left with a view to pecuniary losses. Moreover, in all the three systems, the standards are only indicative prescriptive for the decision-maker and not mandatory, even though they are (certainly in Belgium and Sweden) almost always applied. It is, however, difficult to argue that it is merely averaging the damage compensation that led to increased speed and higher victim satisfaction thanks to a more adequate compensation system. In Ireland and Sweden, the averaging was combined with the introduction of an ADR mechanism which can equally have increased speed and victim satisfaction. The effects of both measures (averaging and ADR) can hardly be separated. However, also in Belgium where no specific ADR structure was introduced, the system is perceived well. It would be very interesting if future research would take this analysis as the starting point for an empirical analysis to concretely measure the effects of the different regimes on the behaviour of the stakeholders involved. More countries where either only standardization or standardization in combination with ADR is pursued could be investigated. In that exercise, a more loss-type-specific analysis may be warranted.
  An important note of criticisms concerns the way in which the standards were created. That equally provides interesting lessons for the way in which a law reform towards averaging these kinds of personal injury claims could be realized. For example, there is in Belgium criticism of the fact that the judges determine the standards without justifying particular normative choices and without involving the opinion of other stakeholders. In Ireland, to the contrary, the judges were critical of the Book of Quantum as they had not been involved in its drafting, as a result of which the system has recently changed in Ireland. This shows that if a country were to introduce a standardization system, it is important to design a carefully thought-through procedure taking into account stakeholder involvement. It is also important to keep the standards sufficiently dynamic and flexible. In that respect, there were substantial differences between Ireland (which has also seen two versions of the Book of Quantum so far (however, the Personal Injuries Guidelines need to be updated more frequently)), and Sweden (where the tables are amended every year) and Belgium being in-between the two. But notwithstanding these specific design issues where the countries examined (as well as others) could mutually learn from each other’s experiences in order to create best practices, the countries we examined seem to indicate that models of damage averaging can be supported both from an economic and a legal perspective. They can lead to a more rapid victim compensation and victim satisfaction, while still respecting the economic requirement of deterring potential injurers by exposing them to the costs of their activities and the legal demand of doing justice to victims in individual cases.

Noten

• 1 There are, of course, also countries which make use of standardization but where, if no agreement is reached, litigation before a court ensues.

• 2 Becker, G.S., ‘Irrational Behavior and Economic Theory’, Journal of Political Economy, 1962, Vol. 70, 1-13; Becker, G.S., ‘Crime and Punishment: An Economic Approach’, Journal of Political Economy, 1968, Vol. 76, 169-217. Tort law also provides incentives regarding activity levels and risk spreading, but in our article we focus on the care incentives. See, for example, Shavell, S., Foundations of Economic Analysis of Law, Cambridge, MA, The Belknap Press of Harvard University Press, 2004, 177-193.

• 3 Koziol, H., ‘Introductory Remarks’, in Koziol, H. (ed.), The Aims of Tort Law. Chinese and European Perspectives, Vienna, Jan Sramek Verlag, 2017, 3-12, 4-6.

• 4 Posner, R.A., Economic Analysis of Law, 9th ed., New York, Aspen, 2014, 223; Faure, M., ‘Economic Optimisation of Tort Law’, in Koziol, H. (ed.), The Aims of Tort Law. Chinese and European Perspectives, Vienna, Jan Sramek Verlag, 2017, 79-113, 80 and 87.

• 5 Calabresi pointed to the fact that the tort system needs to minimize the total social costs of accidents consisting of primary, secondary and tertiary accident costs. Whereas the primary costs consist of the costs of accident avoidance and the potential damage and the secondary accident costs relate to the costs of loss spreading, the tertiary costs are the administrative costs needed to minimize the primary and secondary accident costs (inter alia the costs of the legal system to provide compensation to victims (Calabresi, G., The Costs of Accidents. A Legal and Economic Analysis, New Haven, Yale University Press, 1970, 26-28).

• 6 Additional arguments are imaginable from a behavioural perspective: think of over-optimism or over-confidence bias which leads to an over-activity of a certain behaviour. There may further be instances of hyperbolic discounting. A coherent analysis in the frame of behavioural law and economics is beyond the scope of this article.

• 7 This is of course the case with traffic accidents where mandatory insurance for motor vehicles applies.

• 8 Kaplow, L. and Shavell, S., ‘Accuracy in the Assessment of Damages’, The Journal of Law & Economics, 1996, Vol. 39(1), 191-210; Baumann, F. and Friehe, T., ‘On the Superiority of Damage Averaging in the Case of Strict Liability’, International Review of Law and Economics, 2009, Vol. 29, 138-142. Also see Kaplow, L., ‘The Value of Accuracy in Adjudication: An Economic Analysis’, Journal of Legal Studies, 1994, Vol. 23, 307-401.

• 9 For example, if there is a 50% chance that the losses are €10,000 and a 50% chance that they are €30,000, the expected loss is 0.5*10,000 + 0.5*30,000 = €20,000. For the expected liability it then makes no difference whether the tortfeasor faces a 50% chance of having to pay €10,000 and a 50% chance of having to pay €30,000, on one hand, or a 100% chance of having to pay the average losses of €20,000 on the other. In both cases, the expected liability is €20,000 (this example of course assumes risk neutrality of the parties involved).

• 10 If the tortfeasor does not yet have this information but can acquire it by additional research, the costs of this research must be included in the weighing of costs and benefits as well. See Kaplow and Shavell (1996, 195); Kaplow (1994, 319).

• 11 Kaplow and Shavell (1996, 194); Kaplow (1994, 315).

• 12 In the standard law and economics literature it is, furthermore, suggested that strict liability requires full compensation whereas negligence could also provide correct incentives with incomplete compensation, as long as taking due care is cheaper than running the risk of being held liable. However, this view overlooks the impact of the legal requirement that the negligence must be the cause of the losses, as is convincingly shown by Kahan, M., ‘Causation and Incentives to Take Care Under the Negligence Rule’, Journal of Legal Studies, 1989, Vol. 18, 427-447; see Visscher, L., ‘Wrongfulness as a Necessary Cause of the Losses – Removing an Alleged Difference between Strict Liability and Negligence’, Economic Analysis of Law Review, 2011, Vol. 2, 188-203.

• 13 On the aspects that constitute victim satisfaction, see Akkermans, A., ‘Reforming Personal Injury Claims Settlement: Paying More Attention to Emotional Dimension Promotes Victim Recovery’, 8 (26 February 2009). Available at SSRN: https://ssrn.com/abstract=1333214 or http://dx.doi.org/10.2139/ssrn.1333214.

• 14 Shuman, D.W., ‘When Time Does Not Heal: Understanding the Importance of Avoiding Unnecessary Delay in the Resolution of Tort Cases’, Psychology, Public Policy, and Law, 2000, 6, 880-897 with further references on the effect of delays; Elbers, N.A., Akkermans, A., Cuijpers, P. and Bruinvels, D.J., ‘What Do We Know about the Well-being of Claimants in Compensation Processes?’ Recht der Werkelijkheid, 2012, Vol. 33(2), 65-78.

• 15 See Akkermans (2009, 3).

• 16 Hay, B.L. and Rosenberg, D., ‘The Individual Justice of Averaging’, The Harvard John M. Olin Discussion Paper Series, Discussion Paper No. 285, 2000, 34ff.

• 17 Ibid., 15ff.

• 18 Ibid., 37ff.

• 19 In fact, in many European legal systems, the pecuniary losses are often covered by social security carriers (e.g. the medical costs are paid by health insurers) as a result of which most debates in tort cases anyway take place concerning non-pecuniary losses.

• 20 Jan Tinbergen mentioned that N-problems require N-solutions (referring to the fact that national banks should employ an equal number of instruments as the number of political targets (Tinbergen, J., The Theory of Economic Policy, North-Holland Publishing Company, 1952).

• 21 Van Dort, R., Dikke Pech In ’t Buitenland, Maastricht, 2020 (ISBN 978-94-92741-34-9), 37.

• 22 Cousy, H. and Droshout, D., ‘Compensation for Personal Injury in Belgium’, in Koch, B.A. and Koziol, H. (eds.), Compensation for Personal Injury in a Comparative Perspective, Vienna, Springer, 2003, 37-75, 65.

• 23 Faure, M.G., Fernhout, F.J. and Philipsen, N.J., Resultaatgerelateerde Beloningssystemen voor Advocaten. Een Vergelijkende Beschrijving van Beloningssystemen voor Advocaten in een Aantal Landen van de Europese Unie en Hong Kong, The Hague, Boom Juridische Uitgevers, 2009, 14.

• 24 The Union Nationale des Magistrats de Premiere Instance/Nationaal Verbond van de Magistraten van Eerste Aanleg and the Union Royale des Juges de Paix et des Juges de Police/Koninklijk Verbond van Vrede- en Politierechters.

• 25 Cornelis, L., ‘De toegevoegde waarde van rechterlijk activisme: uitdagend (3)’, in Vansweevelt, T. and Weyts, B. (red.), De indicatieve tabel 2016: kansen en kritiek, Antwerp, Intersentia, 2018, 129-139.

• 26 Boyen, A., ‘Forfaitaire vergoedingen en de (privé) indicatieve lijst’, De Verzekering, 2002, nr. 338, 55-76, 57.

• 27 Durant, I.C., ‘Belgium’, in Koziol, H. and Steininger, B.C. (red.), European Tort Law 2010, Berlin, Walter de Gruyter, 2011, 30-60, 53.

• 28 Boyen (2002, 60); Court of Cass. (Belgium) 11 September 2009, Nieuw Juridisch Weekblad 2010, 25 with case note by G. Jocqué. See also Durant, I.C., ‘Belgium’, in Oliphant, K. and Steininger, B.C. (red.), European Tort Law 2012, Berlin, Walter de Gruyter, 2013, 57-93, 85.

• 29 So Vansweevelt, T. and Weyts, B., ‘De indicatieve tabel 2012: van (te) normerend naar betwist?’ R.W., 2014/2015, Vol. 78, 243-253, 245.

• 30 https://www.schadeweb.be/sites/default/files/indicatieve-tabel-2020-tableau-indicatif-2020.pdf

• 31 Vansweevelt and Weyts (2014, 247-248).

• 32 Van Dort (2020, 19 en 23-29).

• 33 Ibid., 15: on the importance of the medical file.

• 34 Vansweevelt and Weyts (2014, 248); Ulrichts, H., Schaderegeling in België, Mechelen, Wolters Kluwer, 2018, 65.

• 35 Weyts, B., ‘De indicatieve tabel 2016: krachtlijnen en enkele kritische bedenkingen’, in Vansweevelt, T. and Weyts, B. (red.), De indicatieve tabel 2016: kansen en kritiek, Antwerp, Intersentia, 2018, 1-16, 5.

• 36 There is more out-of-courts settlement: Vandenwijngaert, G., ‘De indicatieve tabel, vloek of zegen voor de verzekeraar?’ In Vansweevelt, T. and Weyts, B. (red.), De indicatieve tabel 2016: kansen en kritiek, Antwerp, Intersentia, 2018, 121-128, 128.

• 37 Jocqué, G., ‘Enkele bedenkingen bij de nieuwe indicatieve tabel’, R.W., 2012-13, afl. 38, Vol. 76, 1519-1520, 1520.

• 38 Among others, Van Steenbergen, J., ‘Kritische bedenkingen bij de indicatieve tabel’, in Van den Bossche, M. (red.), De indicatieve tabel. Een praktisch werkinstrument voor de evaluatie van menselijke schade, Brussels, Larcier, 2001, 16.

• 39 Ulrichts, H., ‘De nieuwe indicatieve tabel: de stem van de verzekeraar’, De Verzekering, 2002, nr. 338, 84-104, 84.

• 40 www.irishstatutebook.ie/eli/2003/act/46/enacted/en/index.html. In 2007, an important reform concerning the fee system took place, see Quill, E., ‘Ireland’, in Koziol, H. and Steininger, B.C. (red.), European Tort Law 2007, Vienna, Springer, 2008, 352-372, 352-353: it sought in particular to reduce the incentives to reject the assessment without justification.

• 41 Quill, E., ‘The Functioning of the Personal Injuries Assessment Board in Ireland’, Journal of Personal Injury Law, 2019, 74-81, 76. Originally it was even forbidden, which was later argued to be unlawful.

• 42 In essence 90%, see Aldus Cost of Insurance Working Group, Report on the Cost of Motor Insurance, January 2017, 92. https://assets.ie/6254/060219172049-067f7ed921f44343a7f41144ac3d4940.pdf.

• 43 For a summary, see Quill, E., ‘Ireland’, in Koziol, H. and Steininger, B.C. (red.), European Tort Law 2003, Vienna, Springer, 2004, 245-265, 246-247.

• 44 See https://claimsauthority.ie/injury-guidelines/.

• 45 See, for example, https://kennedyslaw.com/thought-leadership/article/the-personal-injuries-guidelines-2021/ and www.addleshawgoddard.com/en/insights/insights-briefings/2021/litigation/all-change-new-personal-injury-guidelines/.

• 46 Section 22(2) of the Civil Liability and Courts Act 2004: ‘Subsection (1) shall not operate to prohibit a court from having regard to matters other than the Book of Quantum when assessing damages in a personal injuries action’.

• 47 Quill, E., ‘Ireland’, in Karner, E. and Steininger, B.C. (eds.), European Tort Law 2019, Berlin, De Gruyter, 2020, 289-313, No. 32, 18.

• 48 Quill, E., ‘Ireland’, in Karner, E. and Steininger, B.C. (red.), European Tort Law 2015, Berlin, Walter de Gruyter, 2016, 282-303, 298.

• 49 McKeown, A., ‘Court of Appeal: More Regard Should Be Had to Book of Quantum’, Irish Legal News, 13 August 2020, www.irishlegal.com/article/court-of-appeal-more-regard-should-be-had-to-book-of-quantum.

• 50 Idem.

• 51 It has been criticized that also settlement amounts were included, see Quill, The Functioning…, 79.

• 52 Personal Injury Guidelines (2021, 5).

• 53 Quill, E., ‘Ireland’, in Wininger, B., Koziol, H., Koch, B.A. and Zimmermann, R. (red.), Digest of European Tort Law. Vol. 2: Essential cases on damages, Berlin, Walter de Gruyter, 2011, 39-41, 41.

• 54 PIAB Annual Report 2004. According to Quill, the real term is closer to 1.5 years and there is also the risk that a significant backlog would develop, see Quill, E., ‘Ireland’, in Koziol, H. and Steininger, B.C. (red.), European Tort Law 2006, Vienna, Springer, 2007, 281-298, 281-282.

• 55 Quill, E., ‘Ireland’, in Koziol, H. and Steininger, B.C. (eds.), European Tort Law 2005, New York: Springer, 2006, 348-367, 349-350.

• 56 Quill (2020, 306-307). The 2020 PIAB Annual Report mentions a cost of PIAB resolutions of 11% and litigation 66% (Quill, E., ‘Ireland’, in Karner, E. and Steininger, B.C. (eds.), European Tort Law 2021, Berlin, De Gruyter, 2022, 283-305, 298-299).

• 57 See Regulatory Impact Analysis. Personal Injuries Resolution Board Bill 2022, available at: https://enterprise.gov.ie/en/legislation/legislation-files/ria-personal-injuries-resolution-board-bill-2022.pdf, p. 5 (last accessed 29 November 2022).

• 58 Cost of Insurance Working Group (2017, 116), Recommendation 18; see also Quill, E., ‘Ireland’, in Karner, E. and Steininger, B.C. (red.), European Tort Law 2017, Berlin, Walter de Gruyter, 2018, 283-309, 303. Notice that in Ireland the criticism was exactly the opposite as in Belgium, where the criticism related to the fact that the indicative table is drawn up exclusively by judges.

• 59 Quill, The Functioning…, 79.

• 60 www.gov.ie/en/publication/55ae3-personal-injuries-resolution-board-bill-2022/ (last accessed 29 November 2022).

• 61 Nyquist, S. and Persson, E., ‘Sweden’, in Faure, M. and Hartlief, T. (red.), Financial compensation for victims of catastrophes. A comparative legal approach, Vienna, Springer, 2006, 227-260, 227; Dufwa., B.W., ‘Compensation for Personal Injury in Sweden’, in Koch, B.A. and Koziol, H. (red.), Compensation for Personal Injury in a Comparative Perspective, Vienna, Springer, 2003, 293-324, 297. It is rather ambiguous that literature still refers to a strict liability mechanism, although it in fact constitutes a duty to compensate upon the liability insurer which exists irrespective of any liability of the insured. An examination of the liability of the insured is in other words no prerequisite for the duty to compensate by the liability insurer.

• 62 Interview with Mr Stefan Andersson, from the insurance company Folksam, on 31 August 2020.

• 63 Strömbäck, E., ‘Personal Injury Compensation in Sweden Today’, Scandinavian Insurance Quarterly, 2000, 89-106, 89: only 10% would not have a liability insurance.

• 64 Macleod, S., Urho, M. and Hodges, Chr., ‘Sweden’, in Macleod, S. and Hodges, Chr. (red.), Redress Schemes for Personal Injuries, Oxford, Hart, 2017, 167-189, 186.

• 65 Interview with Henrik Jerkrot, from the insurance company, on 28 August 2020.

• 66 Nyquist and Persson (2006, 251).

• 67 Ibid.

• 68 TIB, www.trafikskadenamnden.se/vid-skada/, 19.

• 69 Sandell, H., ‘Sweden’, in Koziol, H. and Steininger, B.C. (red.), European Tort Law 2002, Vienna, Springer, 2003, 393-406, 396.

• 70 Ibid., 394.

• 71 Strömbäck (2000, 99).

• 72 www.trafikskadenamnden.se/vid-skada/, 1: ‘Each year, the Traffic Injury Board determines its compensation tables. They are also used by courts and when calculating compensation for other insurances, for example accident insurance.’

• 73 Dufwa (2003, 296). Interview with Mr Henrik Jerkrot, Personal Injury Lawyer in Sweden, on 31 August 2020.

• 74 Quill, E., ‘Ireland’, in Karner, E. and Steininger, B.C. (eds.), European Tort Law 2018, Berlin: De Gruyter, 2019, 299-325, 319-320.

• 75 Quill (2020, No. 31, 18).

• 76 Quill (2006, 349-350).

• 77 Quill, The Functioning…, 79.

• 78 TIB, www.trafikskadenamnden.se/om-oss/#Link12167.

• 79 It also makes economic sense to assess loss of income in an individualized manner, since there can be a wide variety in income losses, and this is obviously pecuniary. Averaging income would cause too much risk and potential losses for victims. There are, therefore, good reasons to keep the assessment of income losses largely individualized. However, there is de facto often some averaging in income losses, for example, as far as predictions are concerned about career development, retirement, aid and mortality risks. Those are often based on general statistics. But again, those statistical analyses will be applied to the individual wage of the victim.

• 80 Also if an ADR body set standards for itself, this issue might emerge.