Edited by Gerrit De Geest
Chapter 7: Pure economic loss
[In: Volume 1, Michael Faure (ed) Tort Law and Economics]
Pure economic loss usually refers to foregone profits or earnings without antecedent harm to plaintiff’s person or property. In contrast to consequential (or parasitic) economic loss, that is, the financial loss that is connected with (even the slightest) damage to the person or property of the plaintiff, there is great variance across legal systems in the recoverability of pure economic loss in tort. In the United States, the economic loss rule states that a plaintiff cannot recover for negligence that causes pure economic loss (the ‘exclusionary rule’). In some cases, however, courts ignore the rule and make their decisions based on an assessment of the particular policy considerations of the case (Schwartz, 2003). For Europe, a distinction can be made between liberal regimes (such as Belgium, France, Greece, Italy and Spain), conservative regimes (such as Austria, Finland, Germany, Portugal and Sweden) and pragmatic regimes (such as England, the Netherlands and Scotland) (Bussani and Palmer, 2003b). Liberal regimes are characterized by the absence of an inprinciple objection to allowing compensation for stand-alone economic harm. On the contrary, in conservative regimes, pure economic loss does not figure among the ‘absolute rights’ which receive protection under tort law. If a remedy is available, it is based on a specific tort provision or on an expansive application of contract principles. In pragmatic regimes, results are not driven by the dictates of a wide tort principle, nor by a checklist of absolute rights. These systems are characterized by a case-bycase approach which studies the socio-economic implications of allowing compensation for pure economic loss (principally through the ‘duty of care’ concept). Two conclusions follow. First, the contrasting approaches do not follow the familiar common law/civil law divide since civil law is divided itself. Second, the radically opposed starting points of the different legal systems often conceal a more complex theoretical substructure. When taken into account, this may lead to more uniformity (Bussani and Palmer, 2003e).
7.2 p. 202A taxonomy of pure economic loss cases
There is a wide range of problems that come under the heading of pure economic loss. Such loss may arise from a varied collection of contexts, ranging from cable cases to oil spill cases to wrong audit cases. However, the following four categories have been proven to be functionally and relationally distinct (see Bussani and Palmer, 2003a; Bussani, Palmer and Parisi, 2003): ricochet loss cases, transferred loss cases, closure of public service and infrastructures cases and flawed professional advice cases. In ricochet loss cases, a physical damage is done to the property or person of one party, which impairs the contract rights of the plaintiff. In these cases, there is a direct victim who incurs physical damage to his property or person, while the plaintiff is a secondary victim who suffers only economic harm. One example is the case in which someone negligently cuts a cable belonging to a public utility, which delivers telephone services to a company which suffers damage caused by the loss of production (‘cable cases’). Another example is the case in which a football star is injured in a car accident, causing the team to drop in the rankings (‘loss of a star cases’).
Transferred loss cases are those cases in which a tortfeasor causes damage to a victim’s property or person, but a contract or the law transfers the loss to a third party. In such cases, a loss that would normally be sustained by the primary victim alone is passed on to a secondary victim, who has a contractual obligation to insure the loss or who only has a contractual interest in the property. Such a transfer may occur when the damaged property is subject to a lease, an insurance agreement, a pending sales contract or when a pay continuation statute requires an employer to pay the salary of an injured employee who can no longer perform his contractual obligations.
In closure of public service and infrastructures cases, an economic loss arises without a previous injury to anyone’s property or person. Any physical damage is to ‘unowned resources’ that lie in the public domain (Goldberg, 1994). Textbook examples of this type of case include the situation in which a negligent act causes the closure of a public highway, resulting in an economic loss to individuals whose production depends on using the highway.
Finally, cases of flawed professional advice deal with the liability of those who provide professional advice, prepare data or render services concerning financial matters. When third parties rely on advice, data or services that are carelessly compiled or executed, they may suffer a pure economic loss. One example is the case of an investor who buys shares in a company at more than its current value because he relies on the audit of an accountant who has carelessly overstated the net financial worth of p. 203the company. Another example is the case of a will that is deemed invalid due to the careless preparation of a lawyer, causing an economic loss to a non-client.
7.3 The insufficiency of explanations not based on efficiency
Traditional explanations seeking to justify the frequent denial of compensation for pure economic loss that are not based on notions of efficiency all lead to practical inconsistencies (see Bussani, Palmer and Parisi, 2003; Dari-Mattiacci and Schäfer, 2007; Parisi, Palmer and Bussani, 2007). A first explanation is grounded in the belief that pure economic losses are more difficult to foresee than physical damage. Applying the traditional foreseeability test to cases of pure economic loss would lead to disastrous levels of liability (Feldthusen, 2000). However, this argument proves too little because not all pure economic losses are difficult to foresee and proves too much because some instances of physical losses seem to cause foreseeability problems that are at least as great, and yet their compensation is not questioned.
A second set of arguments maintains that intangible wealth should not be treated on the same level as bodily integrity or physical property. Not all interests are equally good. Greater protection should be given to tangible property than to intangible wealth (Weir, 2000). However, it’s easy to see that barring recovery for pure economic loss cannot simply be the consequence of an ordering of interests. How could this be reconciled with the fact that intentionally inflicted pure economic losses and consequential economic losses are generally recoverable?
A third frequently invoked explanation concerns the problems of open-ended liability and derivative litigation. In a world of economic networking, pure economic losses are likely to be serially linked to one another. Permitting recovery of pure economic loss would (in some cases) overwhelm the courts, place an excessive burden upon the defendant (producing overdeterrence) and reinforce the general trend toward expansion of tort liability (Spier, 1998). The empirical foundation for this argument however is weak. In the liberal regimes in Europe, where this argument has not been a restraint, no dire consequences have resulted. Moreover, the pattern of adjudication in Europe in ricochet loss cases is not in accordance with an explanation based on concerns for open-ended litigation. The risk of open-ended litigation is far greater in ‘cable cases’ than in ‘loss of a star cases’, and yet many more jurisdictions grant compensation in the former type of cases.
A final argument refers to the need to reduce the costs of litigation by allowing recovery in tort only to those p. 204plaintiffs who suffered physical loss (for example, the telephone company in ‘cable cases’) and thus to induce plaintiffs who suffered pure economic loss (for example, customers of the telephone company who suffered a blackout) to seek protection through contractual arrangements with the first type of plaintiffs, who then recover from the tortfeasor (Rizzo, 1982b). However, this solution does not work when no plaintiff suffers physical harm (Gomez and Ruiz, 2004). Also, this solution carries the risk that plaintiffs will be induced to exaggerate the magnitude of the loss channeled through the contract. It reduces the involvement of the courts in tort liability, but may increase their role in contract cases (as guarantor of the reasonableness of contractual arrangements) (Rizzo, 1982b). Whether litigation costs would rise or fall with this solution is questionable (Bishop, 1986).
7.4 Private versus social loss
Bishop (1982a) presented the first economically oriented analysis of economic losses caused by accidents. He noticed that in many cases in which the law is reluctant to award compensation to the victim for economic loss, there is a divergence between the private loss of the victim and the social loss (the loss that the accident inflicts upon society at large). There is an important difference between cases of physical harm and cases of pure economic loss. Physical harm always implies the destruction of a societal resource. There is a one-to-one relationship between the private loss of the victim and the resulting social loss. The case of pure economic harm is different. Such harm may or may not amount to a social loss because the tortfeasor may cause by a single act an economic damage to a victim and a gain to a third party. Pure economic harm may, to a certain extent, simply mean a redistribution of wealth from one subject to another. For example, if a firm loses profits due to an accident, a competing firm may benefit from an increase in profits as a consequence of that accident (see also Shavell, 1987). Compensation should only be based on the existence of a social loss. If injurers are made liable for more than the harm they cause to society, they will take levels of precaution that are excessive from society’s point of view. Many subsequent analyses of the pure economic loss problem have been based on these theoretical foundations (see for example Shavell, 1987; Landes and Posner, 1987; Goldberg, 1994).
Dari-Mattiacci and Schäfer (2007) have stressed the limitations of Bishop’s transfer argument by arguing that even when the victim’s private loss and the third-party earnings are the same, there may be a social loss deriving from the impairment of socially valuable resources. Impairment and destruction are social losses of the same nature; the difference between them is only quantitative. The value of a resource corresponds to the utility that individuals derive from its use. While destruction means the impossibility of deriving utility from a resource ad infinitum, impairment p. 205is the impossibility of deriving utility from a resource for a limited period of time. The objection that another resource may be used as a substitute is not a convincing reason for accepting that impairment does not amount to a social loss. This objection would still hold when resources are destroyed. The only difference is that a destroyed resource ought to be substituted permanently, while an impaired resource need only be substituted for a limited period of time.
Rizzo (1982a) has objected to Bishop’s transfer argument by remarking that it is based on the assumption that substitutes are readily available on the market at no cost. In reality, third parties will only make gains when they have excess capacity. According to Rizzo, such excess capacity is inefficient because it diverts resources from alternative valuable uses. If economic loss is not compensated, incentives to increase capacity will increase since there will be more accidents. Due to excess capacity, any private loss should be considered a social loss in the real world (Rizzo, 1982a). However, it can be argued that increased capacity by the victim and the gainer is a form of precaution (Schäfer and Ott, 2004, Dari-Mattiacci, 2004; Dari-Mattiacci and Schäfer, 2007). Such precaution leads to a reduction in the social loss. The optimal solution can thus be a combination of some levels of precaution by the injurer, the victim and the gainer. Overcapacity is thus not inefficient per se. The real question is what the optimal level of capacity is. In many cases, the optimal level will be higher than in a world without accidents.
Even when the third-party gain comes at no impairment or increasedcapacity cost, the loss is purely economic, the social loss is smaller than the private loss and there is no explicit contract between the injurer and the victim, some cases should lead to compensation (see for example Dari-Mattiacci and Schäfer, 2007). The reason is that someone else besides the victim may have paid for protection against an involuntary wealth transfer. An example is the case in which the seller of a house hires an expert to value the house with the explicit purpose of facilitating the sales transaction. Another example concerns the liability of an auditor who approves a financial statement for the initial public offering of a corporation. The purpose of hiring the expertise of the auditor (at least in this case) is to reduce the asymmetry of information between the management, the inside investors and the public.
According to Dari-Mattiacci (2004), the private versus social loss antinomy does not furnish a general economic explanation for the legal doctrines concerning pure economic loss. He argues that the pure economic loss problem originates from the use of the tort law system for a hybrid task: the simultaneous internalization of negative and positive externalities. However, tort law is a mechanism primarily conceived to deal with p. 206the internalization of accidental harm and not the internalization of accidental benefits. In an ideal world in which both negative liability and positive liability could be used, legal remedies could provide all parties with optimal incentives. However, no traditional tort rule is compatible with such a framework of decoupled (positive and negative) liability. A liability system that merely links liability to the measure of the net social loss provides optimal incentives for the injurer but not for the victim (see further at Section 7.6). One of the conclusions of Dari-Mattiacci is that the economic explanation of the seemingly disparate treatment that the pure economic loss problem receives in different legal systems may be improved by taking into account additional factors that influence the incentive effects of liability rules (the determination of negligence, the presence of incentives provided by taxes, subsidies or regulation, the way the socially relevant loss depends on the precautions taken by the injurer, the victim and the gainer). From a different perspective, Kalss (2007) argues that the issue of liability for misleading information in capital markets cannot be fully understood without taking into account the global picture of enforcement of informational duties.
7.5 Private loss, social loss and market structure
Some authors have pointed out that the discrepancy between the primary victim’s private loss and the social loss depends on the structure of the affected market. The most extensive analysis comes from Schweizer (2007). He examines the case in which an injurer causes a temporary blackout to a firm (the primary victim) which also affects customers (secondary victims) and competitors of the firm (gainers). When the primary victim serves the market as a monopolist, by definition there are no competitors that could benefit from the monopolist’s blackout. The customers, however, lose their surplus. The social loss thus exceeds the private loss of the primary victim by the amount of the customer’s surplus. In the case of a market governed by perfect (short-run) competition, prices equal marginal costs due to competitive pressure. If an accident happens, the competitors will make up for part of the reduction in supply but the price will rise, given increasing marginal costs. Once again, the social loss exceeds the private loss of the primary victim. In the case of perfect competition with constant marginal costs, the accident would cause neither a private nor a social loss. In the case of imperfect competition with free entry, the primary victim’s private loss exceeds the social loss. Schweizer relies on an earlier finding on free entry under imperfect competition to find this result: when competition is less than perfect, free entry leads to a range where social welfare is decreasing (von Weizsäcker, 1980, Mankiw and Whinston, 1986). As a result, social welfare in the case of an accident, net of the victim’s fixed costs, exceeds p. 207social welfare without accident. In addition, without the accident, the primary victim would earn revenues covering both fixed and variable costs. It follows that the social loss is smaller than the primary victim’s private loss. Note, however, that the social loss is not zero but positive. The intuition goes as follows. The social loss does not depend on the level of fixed costs, since fixed costs arise with and without accidents. In addition, fixed costs are not relevant for quantity choice (although they may affect the entry decision). As a result, the social loss from an accident in a setting with fixed costs would be the same as in the absence of fixed costs. And we know that in the absence of fixed costs, an accident will lead to higher prices since competition is lessened. It follows that the social loss will be positive since higher prices imply a lower sum of producer’s and consumer’s surplus.
7.6 Strict liability versus negligence
Several authors have examined the efficiency properties of the rules of strict liability and negligence when the victim is entitled to full compensation of his economic loss and when he is only entitled to compensation of the (smaller) socially relevant loss (Dari-Mattiacci, 2004; Dari-Mattiacci and Schäfer, 2007; Schweizer, 2007). The most elaborate analysis comes from Dari-Mattiacci (2004), who considers three general cases: unilateral precaution accidents (only the injurer can take precautions), bilateral precaution accidents (either the injurer and the victim or the injurer and the gainer can take precautions) and trilateral precaution accidents (injurer, victim and gainer can take precautions). When the victim and the gainer are not passive, they are only able to mitigate the magnitude of the loss by increasing their capacity. The injurer is able to reduce the probability of the accident. Further, the gainer is never asked to pay compensation for the gain received. Under the ‘recovery rule’, the victim is entitled to full compensation for his loss. Under the ‘exclusionary rule’, he is only entitled to compensation for the socially relevant loss (note the difference with the standard definition of the ‘exclusionary rule’).
7.6.1 Unilateral precaution accidents
When only the injurer can take precautions, the exclusionary rule is optimal under strict liability, while both the exclusionary rule and the recovery rule are optimal under simple negligence. In general, if we want to induce efficient precaution under a rule of strict liability, the injurer should face damages equal to the social loss. This is the case only under the exclusionary rule. If damages are lower than the social loss, a suboptimal amount of precaution will be taken. If damages are larger than the social loss (as under the recovery rule), too many precautions will be taken. A negligence rule provides incentives for efficient precaution if the standard of conduct p. 208is equal to efficient precaution and a negligent injurer owes damages not below the social loss. This last condition is fulfilled under the recovery rule as well as under the exclusionary rule.
7.6.2 Bilateral and trilateral precaution accidents
When the injurer and the victim can take precautions (and the gainer cannot), neither the recovery rule nor the exclusionary rule is generally optimal under strict liability or under simple negligence. Under strict liability, the injurer will choose the optimal amount of precaution if compensation is equal to the pure social loss (exclusionary rule) and if the victim takes optimal precautions. Consequently, the victim bears the remaining private loss. This is not efficient because the victim should also bear the social loss in order to have incentives to take the optimal level of precaution. In general, it is impossible to set a level of compensation so that both the victim and the injurer take optimal precautions. Under simple negligence, both the exclusionary rule and the recovery rule optimally incentivize the injurer when due care is set at the optimal level and the victim also takes optimal precautions. Consequently, the non-negligent injurer pays no compensation to the victim, who bears his full loss. The victim will take too many precautions as he bears a higher cost than the social cost.
When the injurer and the gainer can take precautions (and the victim cannot), the exclusionary rule is optimal under strict liability, while both rules are optimal under simple negligence. Under strict liability, the gainer will act so as to maximize his expected gain minus his gain-enhancing costs. Since he bears the full gain, he will take the optimal level of gain-enhancing precaution if the injurer takes an optimal level of precaution. The injurer will take an optimal level of precaution if he faces damages equal to the social loss (exclusionary rule) and if the gainer takes optimal care. Consequently, only the exclusionary rule gives both the injurer and the gainer incentives to take optimal levels of precaution. Under simple negligence, the injurer’s minimization problem is the same as in the unilateral precaution case and the gainer’s maximization problem is p. 209the same as under strict liability. Both rules are efficient if the required level of precaution is set at the optimal level.
When the injurer, the victim and the gainer can take precautions, neither the recovery rule nor the exclusionary rule is generally optimal under strict liability or under simple negligence. Under strict liability, the gainer bears the full gain and will take the optimal level of precaution if the other parties take their optimal levels. This is not the case. As in the case in which only the injurer and the victim can take precautions, it is impossible to set a level of compensation so that both the victim and the injurer take optimal precautions. Under simple negligence, the gainer’s maximization problem is the same as under strict liability. Thus the gainer will take optimal precautions if both the injurer and the victim take optimal precautions. However, this is generally impossible (once again we can refer to the case in which only the injurer and the victim can take precautions).
7.7 Intentional versus negligent torts
The US exclusionary rule does not apply to economic loss caused by intentional wrongdoing. Intentionally inflicted pure economic loss is recoverable in all European legal systems as well. The economic rationale for recovery in such cases is straightforward. A rule of no recovery would enable an intentional tortfeasor to impose a pure economic loss on a victim, creating an economic benefit for a third party, without having to face any tortious liability. These zero-sum transfers would generate the potential for a spiral of reciprocal takings with extensive rent dissipation for society as a whole (Bussani, Palmer and Parisi, 2003; Parisi, Palmer and Bussani, 2007).
7.8 Economic analysis of the various types of cases
Bussani, Palmer and Parisi (2003) have argued that the substantive applications of the economic loss rule in European jurisdictions are consistent with the predicates of economic analysis (see also Parisi, Palmer and Bussani, 2007).
7.8.1 Ricochet loss cases
A very large majority of European jurisdictions refuses recovery of pure economic loss in ‘loss of a star cases’, while about half grant compensation in ‘cable cases’. This difference can be explained in economic terms. In ‘loss of a star cases’, any liability of the tortfeasor toward the team would probably amount to duplicate compensation for the same loss. Given the position of star players as monopolistic sellers of non-fungible services, star players already capture most of the surplus that the team expects to earn from the player. If compensation received by the victim already includes his lost wages, the duplicate compensation could cause overdeterrence. The situation is different for ‘cable cases’. In these cases, the asset’s market price cannot be assumed to capture the full surplus that third parties obtain from its use. For example, it’s unlikely that the price of telephone services captures the full consumer benefit of using these services. If liability would be limited to the losses of the telephone company, compensation would be lower than the social losses of the accident.
7.8.2 Transferred loss cases
A split among European legal systems can be observed in the treatment of transferred loss cases. This is not unexplainable from an efficiency p. 210perspective. As long as the full social cost of the accident is borne by the tortfeasor and there is no duplication of the tortfeasor’s liability for the same loss, both solutions (recovery or not by the third party) lead to efficient precautions. Whether compensation is paid to the primary victim, to the secondary victim or split among the two, is irrelevant from a deterrence point of view. Criteria of efficiency are thus neutral on the question whether compensation for pure economic loss should be granted in transferred loss cases.
7.8.3 Closure of public service and infrastructures cases
Almost all European courts deny compensation for pure economic loss in closure of public service and infrastructure cases. Once again, this seems consistent with efficiency considerations. First, in many of these cases, the pure economic loss may be much larger than the social loss. The social loss is maximally equal to the difference between the foregone profit and the profit of the second-best opportunity available to the plaintiff. Also, some of the lost profit for the plaintiff may result in a windfall gain for other suppliers. With perfect market elasticity, there will be no net social loss. Second, in those instances in which the private loss might generate a social loss due to inelastic market conditions, the efficiency gains may not justify the large administrative costs necessary to implement a full liability system. Indeed, it is this category (perhaps together with the next one) that raises the greatest concerns for open-ended liability and litigation.
7.8.4 Flawed professional advice cases
While virtually all European jurisdictions allow recovery of pure economic loss in case of flawed services provided by lawyers and notaries, they are less enthusiastic about doing so in the case of auditors and accountants. This dichotomous approach is consistent with the economic model of optimal liability. There is an important qualitative difference between services provided by lawyers and notaries and services provided by auditors and accountants. The services of lawyers and notaries usually benefit exclusively the client or a limited group of third parties. These third parties are likely to be intended beneficiaries of the client who is paying for the services. The price of services incorporates the expected cost of professional liability. The client is thus paying for an implicit warranty of quality. In the case of accountants and auditors, due to the intrinsic nature of the services provided, many more individuals may rely upon the information provided by these professionals (for example, third-party investors and other financial institutions). If third parties other than intended beneficiaries can rely on the information and claim compensation in case of flawed information without contributing to the cost of the service (directly or p. 211indirectly), those who acquire the professional services would pay for the larger potential cost of liability in the form of higher fees. In turn, this would lead to a suboptimal demand for professional services.
Traditional legal theories are unable to explain the observed boundaries between compensable and non-compensable pure economic losses. The economic approach has proven to be rather successful in explaining some large trends in the treatment of pure economic loss cases, but is less advanced when it comes to explaining the variety of legal solutions to the problem of pure economic loss. One possibility is that we need to look at a broader picture that includes other legal elements which differ among various legal systems and which also provide incentives for the parties involved. Another possibility is that attitudes toward pure economic loss are still evolving and changing (and this at a different pace in different countries). In the past 40 years for example, England and Italy have become much more resilient to the recoverability of pure economic loss (see Bussani and Palmer, 2003e). Still another possibility is that the different patterns reflect an element of randomness in rulemaking.
Dari-Mattiacci, Giuseppe (2004), ‘The economics of pure economic loss and the internalization of multiple externalities’, in Willem van Boom, Helmut Koziol and Christian Witting (eds), Pure Economic Loss: Vol. 9. Tort and Insurance Law, New York/Vienna: Springer, pp. 167–90.