Thinking Like (Law-And-) Economists – Legal Rules, Economic Prescriptions and Public Perceptions of Fairness

4.1 Coasian bargaining – background

Property law concerns the assignment of property rights and the determination of remedies for property disputes. Law and economics provides guidance regarding the achievement of efficient assignment of rights and settlement of disputes. The seminal work of Coase (1960) brought transactions costs to the foreground of the analysis of externalities. In the absence of transaction costs, the legal assignment of rights will have no bearing on the ultimate allocation of rights; whether by legal assignment or post-trial bargaining, the party who values the property right more highly will obtain it.

In his analysis of externalities, Coase (1960) explicitly cited the case of Sturges v. Bridgman, which involved a property dispute between a dentist (who recently built an extension to his examination room) and a candy maker (whose noise now imposed an external cost on the dentist due to the fact that the new examination room was closer to the noise). Vignette P1, presented word-for-word in Supplementary Table S1 of the online appendix and summarized in the first row of Table 1, is based directly on the details of this case. Smith Recording Studio is a successful business located next to Jones Dentistry. Jones has recently built an extension that makes his waiting room closer to the studio. After the extension, noise now drives away a few of Jones’s customers and takes away a small portion of business. Respondents are told that the dispute goes to a trial, and that a judge rules in the dentist’s favor; the ruling implies that the recording studio will have to shut down unless the two parties can reach a deal whereby the dentist allows the studio to remain open and the studio compensates the dentist for the lost business from the noise. The two versions differ, however, in the level of transaction costs that affect the post-trial bargaining process. In the version that does not follow the economic prescription, respondents are told, at the very end of the story, that the owners of the dentistry and the studio are unlikely to reach a deal because they are enemies. In the version that follows the economic prescription, respondents are told that the two parties are friendly and a deal that allows the studio to stay open would be extremely likely. Given the logic of the Coase theorem, the judge’s ruling in the low-transaction-cost version is efficient in that it will not require the successful business to shut down. The ruling in the high-transaction-cost version, in contrast, is inefficient because it causes a successful business to shut down rather than allow it to impose small external costs on the dentist.

4.2 Coasian bargaining – results

Across both versions of P1, deontological attitudes toward external costs, property rights, victimization and the merits of noise complaints could shape respondents’ fairness perceptions of the ruling against the noisy business. Any difference in responses across the two versions will point to the varying conditions for Coasian bargaining exerting an additional effect on perceptions of fairness.

The first row of Table 2 shows the descriptive statistics and Mann-Whitney test that shed light on whether the economic prescription for Coasian bargaining aligns with intuitions of fairness. Figure 1 shows frequency distributions of responses across the two versions.

Figure 1:

Distributions of fairness assessments – Coasian bargaining following an injunction.

When high transaction costs prohibit post-trial bargaining, the median and modal response to a ruling that requires the studio to shut down is “very unfair.” In contrast, when low transaction costs lead to likely post-trial bargaining, the median and modal response is “somewhat unfair.” Results from a Mann-Whitney test (U = 5.31, p < 0.0001) that compares the distribution of responses across the two versions identify a highly significant difference. Comparing the two frequency distributions from Figure 1, there is a rightward shift in fairness ratings, indicating stronger levels of perceived fairness, as the outcome shifts from inefficient to efficient. These results demonstrate that, in this particular context, the economic prescription for allocating property rights in the shadow of post-trial bargaining aligns with intuitions of fairness. A judge who employs Coasian logic – being concerned with transaction costs and the likelihood that bargaining will facilitate efficient outcomes – acts in accordance with lay intuitions of fairness.

4.3 Competing resource claims – background

Vignette P2, summarized in the second row of Table 1 and presented word-for-word in Supplementary Table S2, describes a judge settling a dispute over competing claims to a resource; it is based on two cases that involve disputes over the blockage of sunlight. The two cases illustrate how precedents evolved in response to evolving economic conditions. In the earlier case, Fontainebleau Hotel Corp. v. Forty-Five Twenty-Five, Inc., one hotel’s planned vertical extension would restrict the sunlight that another hotel’s pool received. The court allowed the extension, denying the claim that any one party could possess a right to sunlight. In the later case, Prah v. Maretti, the opposite conclusion was reached when a disputed extension would have reduced the amount of sunlight that an owner of a solar-powered house would receive. Using economic logic in the decision, the court in Prah sought a ruling that would not inhibit the development of new industries that relied upon sunlight.

Motivated by these cases, the vignette’s two versions parallel the core difference between Fontainebleau and Prah. In each version, Evergreen Hotel and Jackson Lab are neighbors; Evergreen Hotel is planning on building an extension that would partially block the sunlight that Jackson Labs would receive. In both versions, the court rules against Jackson Labs, allowing the extension to be built. In the version that follows the economic prescription, modeled after Fontainebleau, sunlight is important to Jackson Labs because its employees use a pool on the premises for exercise. In the version that does not follow the economic prescription, modeled after Prah, sunlight is important to Jackson Labs because they conduct research on solar panels. Comparatively, a difference exists to the extent that inhibiting promising research and development is more likely to negatively impact the size of the objective economic surplus compared to inhibiting recreational activity. In the written opinion in Prah, Chief Justice Shirley Abrahamson of the Supreme Court of Wisconsin relied on this logic explicitly, noting that, “Access to sunlight as an energy source is of significance both to the landowner who invests in solar collectors and to a society which has an interest in developing alternative sources of energy.”

4.4 Competing resource claims – results

Across both versions of P2, deontological attitudes toward property rights, specifically the importance of first possession and the willingness to tolerate external costs imposed on others, could shape respondents’ fairness perceptions of the ruling that allows the obstruction of sunlight. Any difference in responses across the two versions will point to the varying benefit of sunlight exerting an additional effect on perceptions of fairness.

The descriptive statistics and Mann-Whitney test that shed light on whether the economic prescription for assigning property rights amidst competing land uses are found in the second row of Table 2. Figure 2 shows frequency distributions of responses across the two versions. The judge’s decision to allow the extension in the version that does not follow the economic prescription (when the extension impedes scientific research) is considered to be less fair than the same decision made in the version that follows the prescription (when the extension impedes employee leisure). The median and modal response when the ruling does not follow the prescription is “somewhat unfair.” Although the version that follows the economic prescription has the same modal response, the median response when the ruling follows the prescription is 3.5, which is between “somewhat unfair” and “somewhat fair” due to the greater frequency of “somewhat fair” responses A Mann-Whitney test (U = 5.31, p < 0.0001) indicates that the difference in perceived fairness between the two versions is highly significant. The rightward shift in fairness ratings across the two frequency distributions in Figure 2 shows how perceptions of fairness change across the two versions. Overall, the difference in fairness ratings across the two versions demonstrates that, in this particular context, following the economic prescription for allocating property rights in the face of competing resource claims aligns with perceptions of fairness. An assignment of property rights that allows a nuisance is perceived as being increasingly unfair when the nuisance inhibits beneficial economic activity.

Figure 2:

Distributions of fairness assessments – competing claims to sunlight.

4.5 Fugitive property – background

Vignette P3, described in the last row of Table 1 and presented word-for-word in Supplementary Table S3, explores whether perceptions of fairness align with economic prescriptions for allocating fugitive property. The vignette is based on Pierson v. Post, which involved a dispute between a foxhunter and an opportunist who tracked the actions of the hunter and took possession of the fox after the hunter had cornered it (and just before the hunter could complete the final stage in the hunt). The court held that the property right belonged to the free-riding party, in part because of the difficulty associated with identifying possession at an earlier stage. A dissenting judge, however, using economic logic, explicitly described how, since foxhunting was time-consuming and costly, the court’s ruling would adversely affect the incentives to hunt.

The vignette in the survey replaces the foxhunting context with the search for a treasure. Like the actual case, a free-rider acquired possession of a resource and the court allowed him to keep it. In the version that does not follow the economic prescription, the treasure hunter spent years tracking the treasure and developing a strategy to find it; in the version that follows the economic prescription, the tracking and strategy development took place over a few hours. Across both versions, the ruling that allows the opportunist to maintain possession would reduce the incentives for treasure hunters to engage in costly search. But, when search takes place across a few years, the ruling would have a more perverse effect on the decision to search. In contrast, when the treasure hunter only stands to lose a few hours of work if an opportunist free rides off of the initial investment, the perverse effect on incentives would not be as severe. Since the treasure is a valuable resource, someone who prioritizes the effect of incentives would be less likely to support the ruling when years, as opposed to hours, of search are preempted.

4.6 Fugitive property – results

Across both versions of P3, deontological attitudes toward rewarding effort and penalizing opportunism could shape respondents’ fairness perceptions of the ruling that rewards the free-riding party. Any difference in responses across the two versions will point to the variation in the initial time of investment exerting an additional effect on perceptions of fairness.

The third row of Table 2 shows the results from P3. Figure 3 shows frequency distributions of responses across the two versions. Median fairness ratings in both versions are both “somewhat unfair,” and the mean ratings across the two conditions are statistically indistinguishable (3.44 compared to 3.48). A rank-sum test that compares the two distributions shows no difference (U = 0.47, p = 0.6395); the comparison of the two frequency distributions likewise shows no shift in perceived fairness. The ruling in favor of the freeriding party, therefore, is considered to be equally unfair regardless of whether the investment that lead to the discovery of the fugitive property took place over a few years or over a few hours. Unlike the results with respect to Coasian bargaining and disputes over competing claims, the perceived fairness in a scenario where first possession overrides initial investment does not align with economic prescriptions for assigning rights to fugitive property.

Figure 3:

Distributions of fairness assessments – fugitive property.

5.1 Hand rule – background

Justice Learned Hand, in U.S. v. Carroll Towing, Co. (169 F2d 169 [1947]), explicitly relied on economic logic when deciding whether a barge owner should be liable for the collision and ultimate sinking of the ship. Hand argued that a standard of reasonable care would have been met if the burden of precaution (B) was less than the expected accident cost without precaution, found by multiplying the probability of an accident without precaution (P) and the losses from an accident (L). Brown (1973) translated Hand’s logic into marginal terms; B can be interpreted as the marginal cost of more precaution, while PL can be interpreted as the marginal benefit that comes from decreased expected accident costs.

Two vignettes, based on two separate cases, explore the perceived fairness of the Hand rule’s logic for determining whether standards of care were met. One vignette explores how fairness assessments change as the marginal benefit of an act of precaution changes (P*L given the terms listed above). A second vignette addresses how intuitions of fairness change given different marginal costs of precaution (the B term listed above).

5.2 Hand rule – marginal benefit of precaution

Davis v. Consolidated Rail serves as an example in which a judicial opinion considers the marginal benefit of an act of precaution that was not taken. Writing the court’s decision for this case, Posner employed the Hand-rule calculus to determine if a railroad company was liable for damages caused when a railroad car ran over the leg of an employee who was inspecting underneath a train. Hearing the case on appeal, Posner’s decision explains that the court could find no legitimate reason to overturn an initial ruling that found the railroad liable. Using the terms described above, Posner reasoned that the jury could have reasonably concluded that the marginal cost of precaution was less than the marginal benefit of precaution, as it would have been easy to blow a warning horn before moving the train and doing so would have substantially decreased the probability of an accident occurring. The jury was therefore reasonable to conclude that the railroad should be liable for not taking sufficient care.

The storyline in vignette T1, presented word-for-word in Supplementary Table S4 and summarized in the first line of Table 3, closely follows the case of Davis V. Consolidated Rail. Across both versions, the court finds the owner of the railway yard liable because the owner of the yard should have sounded an alarm before moving the train. The vignette tests the importance of the marginal benefit of precaution by varying the probability that an act of precaution would have prevented the accident. In the version that follows the economic prescription, the alarm would have been unlikely to decrease the probability of an accident; in the version that follows the prescription, the alarm would have likely decreased the probability of an accident. Thus, using the logic of the Hand rule, it would be inefficient to hold the yard owner liable for damages when the marginal benefit of precaution is low, which is the case when the alarm is not likely to matter. Likewise, it would be efficient to find liability when the action that could have been taken would have been likely to prevent the accident. A comparison of fairness ratings across these two versions can show whether intuitions of fairness align with the prescriptions for efficiency, in terms of the marginal benefit of taking precaution, set out by the Hand rule.

5.3 Hand rule – marginal cost of precaution

Winn Dixie Stores, Inc. v. Benton, involved a court considering the marginal cost of accident avoidance. The case involved a slip-and-fall on a puddle of milk in a supermarket. The opinion highlights how the length of time that the puddle was on the floor was the driving factor determining whether the store should be liable for the accident. Based on evidence that the floor had not been swept for 30 minutes prior to the accident, the appeals court upheld the original jury’s finding that the store did not exercise due care. Because intermittent sweeping is not an overly burdensome task, and because such sweeping would have drastically decreased the likelihood of an accident, the logic of the Hand rule points to a finding of negligence.

Vignette T2, presented word-for-word in Supplementary Table S5 and summarized in the second line of Table 3, is directly based on this case. A shopper slips and falls on a puddle of milk, and the court finds the store liable for the accident. In the version that does not follow the economic prescription, the marginal cost of precaution was high; the spill happened one minute after that accident, so it would have been difficult for the store to prevent the fall. In the version that follows the economic prescription, the marginal cost of precaution was low, as the accident happened thirty minutes after the spill and it would have been easy for the store to prevent the accident by cleaning the spill. Differences in fairness ratings across these two versions of the vignette can shed light on whether intuitions of fairness align with variation in the cost-justification of accident prevention.

5.4 Hand rule – results

Across both versions of vignettes T1 and T2, deontological attitudes toward responsibility, labor relationships (for T1), consumer protection (for T2) and compensation for harm could shape respondents’ fairness perceptions of liability. Additionally, given that assignment of liability involves payment of monetary damages, attitudes toward income redistribution could also affect fairness perceptions across the two versions of each vignette. Any difference in responses across the two versions for each respective vignette will point to the variation in the marginal benefit (for T1) or the marginal cost (for T2) of precaution exerting an additional effect on perceptions of fairness.

Vignettes T1 and T2 both show strong support that intuitions of fairness align with the prescriptions for efficiency implied by the Hand rule. The first and second rows of Table 4 show the relevant statistical results. Analyzing how perceptions of fairness vary with the marginal benefit of precaution, the first row shows that the median fairness rating of the finding of liability is “somewhat fair” when the alarm in vignette T1 was unlikely to matter. Median perceived fairness increased, however, to “very fair” when the alarm probably would have prevented the accident. The Mann-Whitney test in the right-most cell in the first row shows that the difference between the two versions is highly significant (U = 6.53, p < 0.0001). The distribution functions shown in Figure 4 provide further evidence of the shift in fairness ratings across the two versions, with the modal response shifting from “somewhat fair” to “completely fair.”

Figure 4:

Distributions of fairness assessments – Hand-rule, marginal benefit of precaution.

The same pattern is observed when analyzing the perceived fairness ratings in vignette T2. The second row of Table 4 shows that the median rating for the finding that the grocery store should be liable following the slip-and-fall is “somewhat fair” when the accident could have only been avoided at a substantial cost. But when the accident could have been prevented at a lower cost, the median fairness rating of the liability ruling increases to “very fair.” A Mann-Whitney test shows that this difference is highly significant (U = 12.78, p < 0.0001), and Figure 5 further demonstrates the shift in fairness ratings across the two versions, with the modal response again shifting from “somewhat fair” to “completely fair.”

Figure 5:

Distributions of fairness assessments – Hand-rule, marginal cost of precaution.

5.5 Strict products liability – background

While the Hand rule uses economic reasoning to specify whether standards of due care have been met, economic reasoning has also been used to determine when companies should be held strictly liable for accidents that they cause. In a concurring opinion written in Escola v. Coca Cola, a 1944 case where Coca Cola was found liable for injuries caused by an exploding bottle, Justice Traynor of the Supreme Court of California notes that strict liability is the appropriate remedy in situations where manufacturers can anticipate hazards and guard against them. Moreover, Traynor argues that, as mass production increasingly came to replace individualized handicrafts, consumers’ ability to understand the soundness of a product had eroded. With manufacturers aware of the risks and in a position to guard against them, and consumers becoming increasingly reliant on branding and trademarks rather than vigilantly inspecting products’ quality, strict liability would create incentives for manufacturers to internalize the cost of accidents that they cause.

This logic implies that the use of strict liability promotes optimal levels of care when manufacturers are in a position to recognize and protect against risks; under such conditions, strict liability provides manufacturers with an incentive to minimize the total costs of accidents (i. e. the costs of precaution plus the losses from accidents). Otherwise, if producers are not in a position to recognize and avoid accidents, strict liability leads to excessive prices and frequency of disputes, and fails to provide incentives for users to take appropriate care. Vignette T3, which is presented word-for-word in Supplementary Table S6 and summarized in the third row of Table 3, explores whether the use of strict liability is considered to be fairer when it follows the economic prescription. The vignette uses a situation involving an accident caused by a lawnmower; the court finds the manufacturer liable for the harm that an accident causes. In the version that does not follow the economic prescription, the company could not have anticipated the accident; in the version that follows the economic prescription, the company could have anticipated the accident. Comparing fairness assessments of the liability ruling across the two versions can shed light on whether the use of strict liability is considered to be fairer when the nature of the information available to manufacturers generates a situation in which strict liability promotes optimal levels of care.

5.6 Strict products liability – results

Across both versions of T3, deontological attitudes toward responsibility, consumer protection and compensation for harm with respect to the use of consumer products could shape respondents’ fairness perceptions of liability. Similar to T1 and T2, attitudes toward income redistribution could also affect fairness perceptions across the two versions. Any difference in responses across the two versions will point to the variation in the manufacturer’s information exerting an additional effect on perceptions of fairness.

The third row of Table 4 shows the results for scenario T3. Figure 6 shows frequency distributions of responses across the two versions. The median and modal fairness rating of the liability ruling in the version that does not follow the economic prescription – when the manufacturer could not have anticipated the accident – is “somewhat unfair.” In the version that follows the economic prescription, when the manufacturer was in a position to anticipate the accident and protect against it, the median and modal ratings increase to “very fair.” A Mann-Whitney test, displayed in the right-most column in the third row, shows that this difference is highly significant (U = 12.58, p < 0.0001). The comparison of frequency distributions from Figure 6 provides further evidence of the shift in fairness ratings. The difference across the two versions indicates that respondents perceive the usage of strict liability to be fairer when the producer has the information needed to take an optimal level of care.

Figure 6:

Distributions of fairness assessments – strict products liability.

5.7 Punitive damages – background

The use of punitive damages is the final area of tort law that was addressed. As argued in the decision written for Sturn Ruger v. Day, and described by Friedman (2000) and Polinsky and Shavell (1998), punitive damages can promote optimal levels of care when compensatory damages do not adequately deter accidents; such is the case when an injurer causes harms that go undetected. Otherwise, if all accidents lead to legal claims, punitive damages that go beyond compensatory damages would lead to over-deterrence. The economic prescription, therefore, is that punitive damages should be greater when, all else equal, a company causes many earlier, undetected injuries compared to when a smaller number of undetected injuries occurred. Vignette V4, presented word-for-word in Supplementary Table S7 and summarized in the fourth row of Table 3, tests whether intuitions of fairness align with this prescription for the use of punitive damages. In the vignette, a toy injures a child and the child’s family is awarded both compensatory damages ($10,000) and “an additional $10 million to punish the company for its dangerous and reckless behavior.” In the version that does not follow the economic prescription, respondents are told that there is evidence that the toy company injured two other children, but the company was not held responsible for these earlier injuries because the families of these other injured children did not seek compensation. The version that follows the economic prescription is exactly the same, except that there are 900 earlier uncompensated injuries rather than only two. In the condition with 900 earlier injuries, punitive damages would be necessary for deterring accidents given the low probability that an accident would result in a legal claim. This vignette, therefore, tests whether the economic prescription for the use of punitive damages is also regarded as the fairer use of punitive damages.

5.8 Punitive damages – results

Across both versions of vignette T4, deontological attitudes toward punishment, child welfare, consumer protection, compensation for harm and corporate/parental responsibility could shape respondents’ fairness perceptions of punitive damages. Similar to the other tort vignettes, attitudes toward income redistribution could also affect perceptions of fairness across both versions. Any difference in responses across the two versions will point to the variation in the scope of the harm exerting an additional effect on perceptions of fairness.

The results from vignette T4 show that respondents who are asked to judge the fairness of a punitive damage award do not take the frequency of earlier, undetected injuries into account. The relevant statistics are displayed in the fourth row of Table 4. Median fairness ratings across the two versions are both “somewhat fair,” and the mean ratings are statistically indistinguishable (4.30 compared to 4.26). The Mann-Whitney test, shown in the right-most column of the fourth row of Supplementary Table S2, shows that the distributions of fairness ratings are not statistically different (U = 0.25, p < 0.7274); the frequency distributions of fairness ratings, shown in Figure 7, show only subtle changes across the two versions. In contrast to the other areas of tort law that show evidence of fairness and economic prescriptions aligning, the perceived fairness of a punitive damage award is invariant to the deterrence value that the economic prescription prioritizes.

Figure 7:

Distributions of fairness assessments – punitive damages.

6.1 Efficient breach – background

Economic efficiency requires flexibility regarding a promisor’s ultimate fulfillment of the terms of a contract. Contract breach is efficient when, likely due to changing circumstances, a promisor stands to lose more by fulfilling a contract than the promissee stands to gain. In such a case, efficiency requires the promisor to breach the contract and pay expectation damages that leave the promisee indifferent between contract fulfillment and breach with damages. This economic logic was employed in Eastern S. S. Lines, Inc. v. United States, a case in which the U.S. Army leased a ship from its owner during World War II and agreed to either return the ship to its original state or pay the ship owner the cost of restoration. When the ship was returned, however, the cost of restoration was assessed to be $4 million, while the ship’s value in its restored state was assessed to be only $2 million. The ship owner sought a $4 million payment from the Army, while the Army was only willing to pay $2 million. The court ruled that the changing value of the ship was an unforeseen circumstance and that, if the parties had considered the possibility, they would have agreed that the Army should have to pay the ship owner damages equal to the value of the ship. The case’s decision notes that it would have been wasteful to compel performance and force $4 million to be spent in order to create only $2 million in value. By allowing the Army to pay damages rather than restore the ship to its original condition, the court allowed the Army to breach the contract.

Since breach of contract requires a promisor to renege on an earlier agreement, it is possible that allowing efficient breach violates people’s moral intuitions of fairness. To test the relationship between efficient breach and perceived fairness, Vignette C1, included word-for-word in Supplementary Table S8 and summarized in the first row of Table 5, describes a scenario that resembles the fact pattern in Eastern S. S. Lines, Inc. v. United States. The U.S. Army rents a ship and, despite agreeing to pay restoration costs when the contract was formed, refuses to pay these costs when returning the ship. The court sides with the Army, allowing it to breach the contract. In the version that does not follow the economic prescription, breach is inefficient; restoration is cost-effective, as the ship is worth $5 million and restoration only costs $3 million. In the version that follows the economic prescription, breach is efficient; restoration is not cost-effective, as the ship is worth $1 million and restoration costs $3 million. In both versions, the Army reneges on an agreement, but in one instance it is clearly inefficient to do so. Since any difference in the perceived fairness of the decision to let the Army breach the contract can be attributed to the difference across the two versions, the results from this vignette can shed light on whether breach is considered to be fairer when it is also efficient.

6.2 Efficient breach – results

Across both versions of vignette C1, deontological attitudes toward promise-keeping and responsibility could shape respondents’ fairness perceptions of the ruling that allows the contract breach. Additionally, since the remedy involves the payment of monetary damages, attitudes toward income redistribution could also affect perceptions of fairness across both versions of C1. Any difference in responses across the two versions will point to the variation in cost-effectiveness of specific performance exerting an additional effect on perceptions of fairness.

The results from vignette C1 are presented in the first row of Table 6. Figure 8 shows frequency distributions of responses across the two versions. The median and modal fairness ratings of the ruling that allowed the Army’s breach when performance was efficient was “somewhat unfair”; when breach was efficient, the same ruling received median and modal fairness ratings of “somewhat fair.” A Mann-Whitney shows that the difference between the distributions across inefficient and efficient breach is highly significant (U = 5.653; p < 0.0001), while the comparison of distributions from Figure 8 shows how the perceived fairness of allowing a breach of contract shifts depending on whether performance or breach was efficient.

Figure 8:

Distributions of fairness assessments – efficient breach.

6.3 Foreseeability – background

The final common law domain to be examined is the doctrine of foreseeability, which provides a prescription for how to allocate risk within a contract. Every contract that is formed is subject to potential unforeseen consequences, and either the parties themselves (ex ante) or the courts (ex post) must allocate these risks. A risk is allocated efficiently when it is placed on the party that is better able to bear it. Hadley v. Baxendale is frequently cited as a case that set one especially important precedent for the efficient allocation of risk within a contract. The case involves a dispute following the late delivery of a milling business’s replacement part. Because the milling business (Hadley) could not function without the part, it suffered losses from the late delivery that equaled all lost profits from the two days that the delivery was late. As a result, the business sought damages equal to these lost profits. The court held that Baxendale (the party contracted to make the repair and delivery) did not have to pay these damages, however, since it was not possible for him, given the information provided by Hadley and given the common practices at the time, to foresee Hadley’s extreme dependence on the timely delivery. In coming to this conclusion, the court created incentives for parties in Hadley’s position to disclose risks that would otherwise not be recognized by the counterparty. With a duty to disclose being placed on the party with the ability to recognize the risk and bring it to light during contract formation, there is a greater likelihood that both parties appreciate the relevant risks and, as a result, a greater likelihood that agreed upon contracts are mutually beneficial.

Vignette C2, which is presented word-for-word in Supplementary Table S9 and summarized in the second row of Table 5, is based on the fact pattern in Hadley v. Baxendale. It tests whether the foreseeability doctrine aligns with perceptions of fairness. The vignette involves a bakery ordering a replacement part and losing five days’ worth of profits when the delivery is later than the agreed upon date. Like the ruling in Hadley v. Baxendale, the judge in the vignette rules that the deliverer owes some compensation for the late delivery, but an amount that is significantly less than the five days of lost profits sought by the bakery. In the version that does not follow the economic prescription, the scope of the harm caused by the late delivery was foreseeable to the deliverer; respondents are told (1) that the bakery told the deliverer about its dependence on the part and (2) that it was known, common practice for bakeries to not have a backup part on its premises. In the version that follows the economic prescription, in contrast, the damages were not foreseeable to the deliverer; the bakery did not tell the deliverer about its dependence on timely delivery and the deliverer could not be expected to recognize the risk because it was known, common practice to keep a backup on the premises. Full expectation damages that compensate the bakery for all of its losses represent the efficient remedy when the deliverer agrees to the contract with knowledge of what the consequences of late delivery would be. But damages this high are an inefficient remedy, which prohibit the formation of mutually beneficial contracts, when the deliverer does not understand the risks to which he is exposed when agreeing to the contract. A comparison of fairness ratings across the two versions sheds light on whether the ruling that follows from the foreseeability doctrine is regarded as being fairer.

6.4 Foreseeability – results

Across both versions of vignette C2, deontological attitudes toward promise-keeping and responsibility could shape respondents’ fairness perceptions of a ruling that fails to fully compensate for harm caused by a contract breach. Similar to vignette C1, attitudes toward income redistribution could also affect perceptions of fairness across both versions of vignette C2. Any difference in responses across the two versions will point to the variation in the foreseeability of severe harm exerting an additional effect on perceptions of fairness.

The results from vignette C2 are presented in the second row of Table 6. Figure 9 shows frequency distributions of responses across the two versions. While the median and modal fairness ratings are both “somewhat fair” across the two versions, a Mann-Whitney test does point to a statistically significant difference (U = 1.98, p = 0.047), with fairness ratings being significantly higher in version that follows the economic prescription. The frequency distributions from Figure 9 reveal an increased frequency of “very fair” and “completely fair” responses in the version that follows the economic prescription. Despite the evidence that points to a difference across the two versions of this vignette, the alignment between the economic prescription and intuitions of fairness with regards to the foreseeability doctrine is substantially weaker than the alignment seen in other areas of the common law.

Figure 9:

Distributions of fairness assessments – foreseeability.