ARTICLE: The Principles of Product Liability

[강석근]

ARTICLE: The Principles of Product Liability*, * Copyright © 2007 Richard W. Wright. All rights reserved. Permission is hereby granted to copy for noncommercial use as long as appropriate citation is made to this publication., 26 Rev. Litig. 1067

Symposium 2007

Reporter
26 Rev. Litig. 1067 *


Author: Richard W. Wright**

** Professor of Law, Chicago-Kent College of Law, Illinois Institute of Technology. 

Text

I. Introduction

More than any other area of tort law, the law of product liability has been the subject of continuing debate regarding the interrelated issues of its proper rationales and grounds of liability. Although the seeds of the debate go back at least 100 years, it flowered into its mature form in 1944 in Escola v. Coca Cola Bottling Co.,  1 in which the majority of the California Supreme Court used an expansive application of the res ipsa loquitur doctrine to hold the defendant liable for inferred negligence.  2 

In a concurring opinion, Justice Roger Traynor relied on four different rationales - efficient compensation, efficient deterrence, inferred negligence, and consumer expectations  3 - to argue that liability for defective products should be strict.  4 

Almost twenty years later, in 1963, Traynor's rationales and position were ratified and adopted in an opinion that he wrote for a unanimous court in Greenman v. Yuba Power Products, Inc.  5 The Greenman opinion was a catalyst for the adoption of strict product liability, based on the same rationales, in Restatement Second section 402A,  6 which in turn was rapidly adopted by most states in the United States and greatly influenced the adoption of product liability laws in other countries.  7

However, as the courts attempted to apply strict liability as articulated in section 402A, renewed debate developed on the proper extent and grounds of product liability.  8 By the time the Restatement Third was being drafted, it was generally agreed, at least among the courts, that liability for construction defects should be strict and that liability for warning defects should be based on negligence, while considerable disagreement remained about the proper grounds of liability for design defects.  9 The Restatement Third continues to invoke all four of the rationales that Traynor set forth in Escola - and a few more - to support strict liability for construction defects, but it asserts that those rationales do not support strict liability for design or warning defects, which it claims "are predicated on a different concept of responsibility"  10  and were not prominent in the cases or a subject of significant consideration at the time that Restatement Second section 402A was drafted and adopted.  11

It is true that at the time that section 402A was adopted no clear distinction was drawn between construction defects and design or warning defects, but this is not because little consideration was given to design or warning defects, but rather because the prevailing academic and judicial view was that there was no need to distinguish the types of defect since it was assumed that strict liability should apply regardless of the type of defect.

Greenman itself involved a design defect, which the court loosely described as a "defect in design and manufacture,"  12 and section 402A clearly was meant to encompass design and warning defects as well as construction defects. Their inclusion was the reason for the insertion of the mischievous phrase "unreasonably dangerous" in section 402A. This phrase, along with repetitive language in comments h through k, was intended to preclude strict liability for generic product risks if, but only if, those risks were unavoidable aspects of a useful and reasonably safe product and proper warnings were provided. 13

The Restatement Third's rote invocation of the efficiency rationales is unfortunate, not merely because they are normatively unappealing  14 but also because they fail to explain or justify the differential treatment of construction defects and design and warning defects, or indeed any of the content or structure of past or current product liability law,  15 which however can be explained and justified by Traynor's other rationales, which are based on interactive justice rather than efficiency.  16 Except for its making the existence of a reasonable alternative design an independent requirement that the plaintiff must prove to establish a design defect, rather than only a factor to be considered in its consumer-oriented risk-utility analysis of reasonableness, the Restatement Third's substantive provisions and comments present a generally accurate description of current product liability law.  17 However, the Restatement Third's reporters' strong aversion to consumer expectations language and related preference for risk-utility balancing language conceals the generally consumer-oriented and justice-based nature of its substantive provisions, which understandably has led many courts to reject its formulation of the relevant provisions.  18

In the remainder of this Article, I first summarize, in Part II, the content and structure of current product liability law and assess the extent to which the Restatement Third accurately reflects that content and structure. I then consider, in Part III, the extent to which each of the traditional rationales for strict product liability explains and justifies the actual law and the Restatement Third's provisions.

II. Product Liability Law and the Restatement Third

A. Proper Parties, Cognizable Injuries, and Defenses

A product liability action lies against a person engaged in the business of selling or otherwise distributing products, who sold or distributed a defective product if that defect actually and proximately caused physical injury to the plaintiff's person or property.  19 Whether the liability is strict or requires proof of negligence depends on the applicable definition of defect. When true strict liability exists, it is always supplemental to negligence liability; the plaintiff can always bring a negligence action.  20

If strict liability exists, it applies to sellers or distributors of products, but not services,  21 and usually only if the product is new or rebuilt rather than used, especially if the used product is sold "as is."  22 Sellers and distributors of products include not only product manufacturers (including manufacturers of component parts), but also wholesalers, retailers, lessors, bailors, and others in the chain of commercial distribution of the product. The non-manufacturer sellers and distributors are strictly liable for selling or distributing a defective product even if the relevant definition of defect requires negligence on the part of the manufacturer.  23 A few jurisdictions limit strict liability to manufacturers, and many subject non-manufacturer sellers and distributors to strict liability only if the manufacturer is unavailable, insolvent, or likely to become insolvent. 24 Although Restatement Second section 402A explicitly left the issue open,  25 the cases since have clearly established that bystanders, as well as purchasers and users of the product, are proper plaintiffs in a product liability action, whether based on negligence or strict liability.  26

Tort liability for physical injuries to person and property caused by defective products generally is not subject to contractual disclaimer, limitation, or waiver,  27 except perhaps when there has been a fully informed, freely negotiated bargain for an adequate quid pro quo by consumers with sufficient bargaining power.  28 Similarly, a product user is contributorily negligent only when she behaves unreasonably in the light of risks posed to her by a defect in the product of which she was aware or, due to its obvious or patent nature, should have been aware. She generally has no obligation to inspect products for defects but rather is entitled to assume that they are fit for the ordinary purposes for which they were made. In almost all jurisdictions, the plaintiff's contributory negligence is a partial rather than a complete defense, which often reduces rather than bars the plaintiff's recovery.  29

In the great majority of jurisdictions, there is no tort liability, under either negligence or strict product liability, for pure economic loss - injury to the plaintiff's economic expectations that does not result from actual (or perhaps threatened) physical injury to the plaintiff's person or property. In most jurisdictions, injury to the product itself is treated as pure economic loss, on the ground that such injury merely results in a non-working product and thus a failure of the plaintiff's economic expectations regarding the utility of the product itself.  30 Some jurisdictions treat injury to the product itself as being recoverable physical injury, rather than non-recoverable pure economic loss, if it occurs suddenly rather than gradually,  31 but this distinction does not seem to have any rational basis other than simplistic imagery (sudden injuries seeming to be more "tort like").

B. Construction Defects

Although no distinction among types of defect was clearly articulated in the initial cases or in Restatement Second section 402A, subsequent cases and commentary have distinguished three different types of defect: construction defects, design defects, and warning defects.  32 Construction defects are deviations from the intended design of the product, which usually occur in only some instances of the product.  33 The Restatement Third uses the term "manufacturing defect" rather than "construction defect," but the defect can arise (e.g., through mishandling) as the product passes through the chain of commercial distribution after its manufacture, and strict liability will attach to any seller or distributor in the chain who passes the product along in a defective condition.  34

Except for those few jurisdictions that have not adopted any form of strict product liability, there is general agreement that liability for construction defects is strict, rather than requiring proof of negligence.  35

C. Warning Defects

Liability for a defective warning requires proof of negligence. Section 2(c) of the Restatement Third states that a product is "defective because of inadequate instructions or warnings when the foreseeable risks of harm posed by the product could have been reduced or avoided by the provision of reasonable instructions or warnings … and the omission of the instructions or warnings renders the product not reasonably safe."  36 For both warnings and designs, the Restatement claims to "rely on a reasonableness test traditionally used in determining whether an actor has been negligent," which supposedly requires "determinations that the product could have reasonably been made safer by a better design or instruction or warning"  37 as determined through a "risk-utility balancing" tradeoff of costs and benefits to product sellers and users in order to create "incentives for manufacturers to achieve optimal levels of safety in designing and marketing products."  38

These statements are inaccurate with respect to both designs and warnings, and the inaccuracies are insufficiently remedied by language that was added after the inaccuracies were pointed out during the consideration of the relevant provisions by the American Law Institute.  39

In the first place, the definition of a defective warning in section 2(c) ignores an important distinction between "risk reduction" warnings regarding proper use, which if followed will reduce or eliminate the risk involved in using the product, and "informed choice" warnings regarding inherent, irreducible risks involved in using the product, which if warned about will not lead to any reduction in the risks involved in using the product but might cause a person to avoid those risks by deciding not to use the product. The addition of the words "or avoided" in section 2(c) and an expanded discussion of informed choice warnings in comment i to section 2 were intended to more clearly recognize and state this distinction,  40 but the attempt is undermined by the retention of the requirement in section 2(c) that, in order for an omission of a warning to be defective, the omission must render the product "not reasonably safe."  41 Although comment i asserts, as a fiat, that omission of a required informed choice warning "renders the product not reasonably safe," this is not true under any ordinary interpretation of that phrase. A reasonably designed product with reasonable risk-reduction warnings regarding use is "reasonably safe," despite the absence of an informed-choice warning regarding inherent unavoidable risks that, if given, might lead some people to decide for their own particular reasons, as is their right, not to use the product.  42 Indeed, the omission of the informed-choice warning might even reduce the overall risk of injury, if the use of the product, although involving an irreducible risk, would have eliminated or reduced a greater risk of injury from some other source.

Moreover, no "risk-utility balancing" tradeoff of costs and benefits to product sellers and users, in order to create "incentives for manufacturers to achieve optimal levels of safety in designing and marketing products,"  43 is involved in deciding on the reasonableness of product designs or warnings. This is especially clear for warnings. As part of the language added to comment i states,

Warnings must be provided for inherent risks that reasonably foreseeable product users and consumers would reasonably deem material or significant in deciding whether to use or consume the product. Whether or not many persons would, when warned, nonetheless decide to use or consume the product, warnings are required to protect the interests of those reasonably foreseeable users or consumers who would, based on their own reasonable assessments of the risks and benefits [to them], decline product use or consumption. 44

As in non-product cases, the materiality or significance issue is properly analyzed from the autonomy-oriented perspective of what the foreseeable consumer would want to know rather than the paternalistic perspective of what the product seller thinks the consumer should know. 45 The economic costs to the product manufacturer or seller of providing a warning - for example, the direct costs of designing and providing an adequate warning and the indirect cost in sales (and related jobs etc.) lost due to the warning - are not taken into account. Rather, the negligence analysis is a qualitative one that focuses solely on the interests of those put at risk. As the Restatement Third recognizes, the only "balancing" that occurs takes place in eval‎uating the adequacy of the required warning, which is based entirely on its feasibility and expected effectiveness:  46

Although the liability standard is formulated in essentially identical terms [for defective designs and warnings], the defectiveness concept is more difficult to apply in the warnings context. In eval‎uating the adequacy of product warnings and instructions, courts must be sensitive to many factors. It is impossible to identify anything approaching a perfect level of detail that should be communicated in product disclosures. … In some cases, excessive detail may detract from the ability of typical users and consumers to focus on important aspects of the warnings, whereas in others reasonably full disclosure will be necessary to enable informed, efficient choices by product users… . No easy guideline exists for courts to adopt in assessing the adequacy of product warnings and instructions. In making their assessments, courts must focus on various factors, such as content and comprehensibility, intensity of expression‎, and the characteristics of expected user groups. 47

The economic costs of providing a warning are taken into account only for post-sale product warnings:

Compared with the costs of providing warnings attendant upon the original sale of a product, the costs of providing post-sale warnings are typically greater. In the post-sale context, identifying those who should receive a warning and communicating the warning to them can require large expenditures. Courts recognize these burdens and hold that a post-sale warning is required only when the risk of harm is sufficiently great to justify undertaking a post-sale warning program. 48

As initially drafted, this comment would have required a post-sale warning "only if the risk of harm outweighs the costs of providing a post-sale warning."  49 However, when an objection was made to this cost-benefit balancing language, on the ground that a warning could reasonably be required for a serious risk even if the cost of providing the warning might be thought to be greater than the expected harm, the reporters agreed and the comment was modified to conform to the blackletter of section 10,  50 which only requires that "the risk of harm is sufficiently great to justify the burden of providing a warning."  51

Courts have held that warnings need not be given of the risk of allergic or hypersensitive reactions unless the risk foreseeably affects an appreciable or substantial number of persons, but the use of this threshold requirement rather than the general (consumer-oriented) "risk-utility" analysis has been criticized, and the Restatement Third states that what counts as "substantial" should vary depending on the severity of the expected reaction.  52 As is generally true with respect to warnings or other types of information disclosure, "a product seller is not subject to liability for failing to warn or instruct regarding risks and risk-avoidance measures that should be obvious to, or generally known by, foreseeable product users."  53 On the other hand, for purposes of both design and warning, foreseeable uses are not limited to intended uses as envisioned by the product seller, and in most jurisdictions foreseeable uses include foreseeable misuse, which will not automatically undermine the prima facie case or automatically constitute contributory negligence or assumption of the risk.  54

A few courts flirted for a brief period with a strict liability version of defective warnings, which was implemented by using hindsight rather than foresight. Under the hindsight approach, a product warning (or lack thereof) is defective if, assuming that the product seller knew at the time that it sold the product what is known at the time of trial about the risks of the product, a (better) warning should have been provided about those risks.  55 This hindsight approach to warnings, which never had much support, has virtually none now,  56 although some courts shift the burden on proving the "state of the art" (the state of scientific and technical knowledge) at the time the product was sold to the defendant product seller and (incorrectly) describe this as a strict liability approach.  57

D. Design Defects

While there now is widespread agreement on the proper definition and related ground of liability for construction and warning defects, there continues to be substantial disagreement and debate about the proper definition and related ground of liability for design defects. Given recent developments, the disagreement and debate may be - as the reporters for the Restatement Third claim  58 - more about language than substance, but even so, the language is important, as others have noted.  59

The Restatement Second adopted strict liability for defective products in section 402A, without distinguishing among the various types of defects.  60 The strict liability was implemented through a consumer expectations test, filtered through confusing "unreasonably dangerous" language.  61 Comment g to section 402A states that a product is defective if it is "in a condition not contemplated by the ultimate consumer, which will be unreasonably dangerous to him,"  62and comment i states that a product is "unreasonably dangerous" if it is "dangerous to an extent beyond that which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its characteristics."  63 Comment k emphasizes that, assuming proper preparation and warning, a product is not defective if it is "unavoidably unsafe" due to "a known but apparently reasonable risk" because it is not possible "given the present state of human knowledge" to make it safe for its "intended and ordinary use." 64 The typical examples, noted in comment k, are vaccines or other drugs that have essential ingredients that pose known risks of adverse reactions in some or all users.  65 Similarly, comment i notes the danger of sugar to diabetics, liquor to alcoholics (and others), and butter to potential heart attack victims, and the risks posed by over-consumption of many food and drug products.  66

The great majority of states initially adopted the consumer expectations test in section 402A comment i as the sole test for a defective product, including design defects.

However, as problems arose in attempting to apply the test to design defects, many states supplemented or replaced it with some version of a risk-utility test.  67 

Two of the problems with the consumer expectations test were noted by the California Supreme Court in Barker v. Lull Engineering Co.  68 

First, if the risks posed by a product are "open and obvious," the consumer could not reasonably have an expectation of safety and would not be able to recover, no matter how unreasonably dangerous the product is, as some courts have held.  69 

Second, for some products ""the consumer would not know what to expect, because he would have no idea how safe the product could be made.'" 70

To avoid these problems, the Barker court supplemented the consumer expectations test with a risk-utility test and allowed the plaintiff to use either or both to establish a defective product design:

[A] product may be found defective in design even if it satisfies ordinary consumer expectations, if through hindsight the jury determines that the product's design embodies "excessive preventable danger," or, in other words, if the jury finds that the risk of danger inherent in the challenged design outweighs the benefits of such design. 71

Many courts followed California's lead in adopting this two-pronged test for a defective design. 72 Others, believing the consumer expectations test to be generally unworkable for eval‎uating product designs, and that the ordinary consumer expects nothing more nor less than reasonable testing and design in the light of the foreseeable risks and benefits to the ordinary consumer, shifted to using only a risk-utility test.  73 A substantial number of jurisdictions continue to employ only the consumer expectations test, but sometimes elaborate it using a risk-utility analysis that focuses on the risks and utilities to the ordinary consumer.  74

In order to have liability for defective designs remain strict, as previously declared, rather than being based on negligence, the Barker court and a number of other courts using the risk-utility test declare that the product's risks should be identified and assessed using hindsight rather than foresight.  75 However, most courts do not do so, and even in those jurisdictions that theoretically use hindsight regarding knowledge of the risks posed by the product, the relevant technical state of the art often is deemed to be that which was known or reasonably knowable by experts (whether or not it actually was in use) at the time the product was designed, rather than that which exists at the time of trial.  76 In some jurisdictions, the burden of proof on the state of the art or the overall risk-utility analysis is shifted to the defendant,  77 but, contrary to what is sometimes stated,  78 shifting the burden of proof does not change the ground of liability from negligence to strict liability. Few if any cases seem to have held a product seller liable based on risks that were actually unknown and unforeseeable at the time that the product was sold.

Contrary to what is sometimes assumed, the "risk-utility" test for defective product designs is not the aggregate-risk-utility test championed by efficiency theorists, which would trade off costs and benefits among product sellers, consumers, users and others in order to achieve a "socially optimal" maximization of aggregate utility or wealth. Instead, the test, as with negligence analysis generally, focuses on the risks and benefits to those put at risk - in this context generally the consumers and users of the product - and employs a qualitative rather than quantitative comparison of those risks and benefits: the foreseeable risks are unreasonable and thus negligent if they are significant unless they are not too serious and are necessary or unavoidable in order for those put at risk to obtain, directly or indirectly, desired benefits that substantially outweigh the risks.  79 Courts often reference the factors suggested by Dean John Wade, all of which except the seventh (which is sometimes mentioned but rarely relied on and never determinative) focus solely on the risks and utilities to the consumer or user of the product:  80

1. The usefulness and desirability of the product-its utility to the user and to the public as a whole.

2. The safety aspects of the product-the likelihood that it will cause injury, and the probable seriousness of the injury.

3. The availability of a substitute product which would meet the same need and not be as unsafe.

4. The manufacturer's ability to eliminate the unsafe character of the product without impairing its usefulness or making it too expensive to maintain its utility.

5. The user's ability to avoid danger by the exercise of care in the use of the product.

6. The user's anticipated awareness of the dangers inherent in the product and their avoidability, because of general public knowledge of the obvious condition of the product, or of the existence of suitable warnings or instructions.

7. The feasibility, on the part of the manufacturer, of spreading the loss by setting the price of the product or carrying liability insurance. 81

While the direct benefits desired by those being put at risk and the indirect equal-freedom enhancing benefits to everyone in society are taken into account, the purely private benefits to the product seller or third parties are not taken into account.  82 Indeed, defense lawyers are generally careful to avoid making arguments that seek to justify risks imposed on the plaintiff by allegedly greater enhancements of the defendant's utility. Defendants who are thought to have knowingly made such risk-utility decisions are often deemed by juries and judges not only to have been negligent, but also to have behaved so egregiously as to justify a hefty award of punitive damages, as occurred in the Ford Pinto, asbestos, and McDonald's coffee-spill cases.  83 Indeed, such aggregate-risk-utility decisions, whereby others are knowingly put at significant risk for the private economic benefit of the defendant, provide one of the few recognized bases for an award of punitive "exemplary" damages in England.  84

In recent years, there has been a movement among the substantial number of courts that still use the consumer expectations test to limit its use to situations involving harm caused by the obvious failure of a specific product feature or function - for example, the failure of a wood lathe to hold the wood securely in place (as in Greenman)  85 or of the brakes or tires on a new car. The courts that have thus limited the consumer expectations test require the use of the consumer-oriented risk-utility test, which the Connecticut Supreme Court in Potter v. Chicago Pneumatic Tool Co. 86 calls a "modified consumer expectation test,"  87 in "instances involving complex product designs in which an ordinary consumer may not be able to form expectations of safety."  88

The Restatement Third states that a product "is defective in design when the foreseeable risks of harm posed by the product could have been reduced or avoided by the adoption of a reasonable alternative design … and the omission of the alternative design renders the product not reasonably safe."  89 The words "reasonable" and "reasonably" in this definition are elaborated through a "risk-utility balancing" test, which is described, in language remaining from the initial draft, as involving a tradeoff of costs and benefits to product sellers and users in order to create "incentives for manufacturers to achieve optimal levels of safety in designing and marketing products."  90 However, although the elaboration of the "balancing process" in the initial draft similarly described it as "reflecting a broad social perspective [that] takes a multiplicity of interests into account,"  91 the factors specifically mentioned generally focused on risks and utilities to the consumer or user:

An important consideration is whether the proposed alternative could have been implemented at acceptable cost. Other factors to be considered include the magnitude of the foreseeable risks of harm, the nature and strength of consumer expectations, the effects of the alternative [design] on product function, the relative advantages and disadvantages of proposed safety features, product longevity, maintenance and repair, esthetics, and marketability. 92

When I pointed this out at the initial meeting of the Members' Consultative Group in 1993 and noted, as discussed above, that the cases focus on the risks and utilities to the consumer or user rather than on benefits to the product seller or third parties, the reporters agreed, and no one disagreed. In the next draft the reference to "a broad social perspective" was eliminated, and the description of the relevant factors was rewritten and an illustration added to make clear the limited, consumer-oriented nature of the risk-utility test.  93 These changes were retained through successive drafts without remark or dissent and appear with only minimal editing in the final adopted version of the Restatement Third:

The [relevant] factors include, among others, the magnitude and probability of the foreseeable risks of harm, the instructions and warnings accompanying the product, and the nature and strength of consumer expectations regarding the product, including expectations arising from product portrayal and marketing… . The likely effects of the alternative design on production costs; the effects of the alternative design on product longevity, maintenance, repair, and esthetics; and the range of consumer choice among products are factors that may be taken into account… . Evidence of the magnitude and probability of foreseeable harm may be offset by evidence that the proposed alternative design would reduce the efficiency and utility of the product… . On the other hand, it is not a [relevant] factor … that the imposition of liability would have a negative effect on corporate earnings or would reduce employment in a given industry. 94

As the last sentence of this quote makes clear, the references to the "likely effects … on production costs" and the "efficiency and utility of the product" refer only to the impact on consumers' utility, rather than to aggregate social utility in the economic efficiency sense. Illustration 7, which was illustration 6 when first inserted in 1993, emphasizes this point: "Although the increase in cost to consumers is a relevant consideration, the impact of a finding of defectiveness on the general economy or on the profitability of the [product] manufacturer is not a factor to be considered in deciding whether the alternative safer design is reasonable."  95

The major controversies surrounding the Restatement Third have to do with its requirement that the plaintiff prove the availability of a reasonable alternative design that would have reduced or avoided the foreseeable risks of harm posed by the product and its rejection of any (explicit) consumer expectations test. Many courts, while treating the existence of a reasonable alternative design as a factor to be considered in the consumer-oriented risk-utility analysis of a product design, have refused to make it an absolute requirement.  96 In a recent well-known case, Potter v. Chicago Pneumatic Tool Co.,  97 the Connecticut Supreme Court stated:

Contrary to the rule promulgated in the Draft Restatement (Third), our independent review of the prevailing common law reveals that the majority of jurisdictions do not impose upon plaintiffs an absolute requirement to prove a feasible alternative design.

In our view, the feasible alternative design requirement imposes an undue burden on plaintiffs that might preclude otherwise valid claims from jury consideration. Such a rule would require plaintiffs to retain an expert witness even in cases in which lay jurors can infer a design defect from circumstantial evidence. Connecticut courts, however, have consistently stated that a jury may, under appropriate circumstances, infer a defect from the evidence without the necessity of expert testimony.

Moreover, in some instances, a product may be in a defective condition unreasonably dangerous to the user even though no feasible alternative design is available. In such instances, the manufacturer may be strictly liable for a design defect notwithstanding the fact that there are no safer alternative designs in existence. 98

The Potter court's second objection to the reasonable alternative design requirement overlooks comment e to section 2 in the Restatement Third, which provides, albeit reluctantly and too restrictively, for the possibility of liability without proof of a feasible alternative design in cases in which "the extremely high degree of danger posed by [the product's] use or consumption so substantially outweighs its negligible social utility that no rational, reasonable person, fully aware of the relevant facts, would choose to use, or to allow [others] to use, the product."  99

The court's first objection refers to a concern shared by many:  100 the potential substantial adverse effect on valid plaintiffs' claims posed by the great practical difficulty and expense involved in having to prove, as required by the Restatement Third,  101 that an alternative design was technologically feasible, practical, and reasonable. In an attempt to lessen concerns about the difficulty and costs of proving this, the Restatement Third states that (i) for some products no expert testimony would be needed, given the obvious availability of safer, reasonable alternatives, (ii) when expert testimony is required, the expert need not produce a prototype of the alternative design, and (iii) "the plaintiff is not required to establish with particularity the costs and benefits associated with adoption of the suggested design."  102

In addition, and more significantly, section 3 of the Restatement Third allows a res ipsa loquitur type of inference of a product defect, without proof of a specific construction or design defect or negligence. This inference is permitted, even when proof of a specific defect is possible under section 2, "when the incident that harmed the plaintiff: (a) was of a kind that ordinarily occurs as a result of product defect; and (b) was not, in the particular case, solely the result of causes other than product defect existing at the time of sale or distribution."  103 

Comment b to section 2 explains that such an inference is permissible when, and only when, the harm is caused by the failure of a product "to perform its manifestly intended function."  104

Section 3 was drafted to cover the situations encompassed by the narrowed version of the consumer expectations test, which applies only when there has been an obvious failure of a specific feature or function of the product.  105 In conjunction with section 2(b), it mirrors the bifurcated test for design defects that has been adopted by a growing number of courts, including the Potter court: the consumer-oriented risk-utility test as a generally applicable test, supplemented by the narrowed consumer-expectations test that applies only when there has been an obvious failure of the product to perform a "manifestly intended function."  106

Although the Restatement Third asserts at one point that the narrowed consumer-expectations test, whether denoted as such or camouflaged as in section 3, is not "in apparent conflict with the reasonable alternative design requirement in § 2(b),"  107 it otherwise acknowledges that they are distinct, alternative tests.  108 The Restatement Third ignores the fact that section 3 purposely dispenses with the reasonable alternative design requirement and related issues when it claims that "consumer expectations do not constitute an independent standard for judging the defectiveness of product designs" and that "consumer expectations do not play a determinative role in determining defectiveness [because] consumer expectations, standing alone, do not take into account whether the proposed alternative design could be implemented at reasonable cost, or whether an alternative design would provide greater overall safety."  109 In section 3, the Restatement Third employs a criterion that, by focusing on a product's "failure to perform its manifestly intended function," invokes consumers' expectations regarding intended functions that are implicit or explicit in the nature of the product and the manufacturer's or other seller's marketing of the product.

The expectations regarding a product's specific features and functions, based on its nature and marketing, provide a standard for eval‎uation that, as section 3 assumes, is at least as "well-formed" as the consumer expectations that the Restatement Third finds sufficient for identifying defects in food products and used products.  110 The Restatement Third glosses over this fact when it lumps design and warning defects together in order to claim that, unlike construction defects, they "cannot be determined by reference to the manufacturer's own design or marketing standards because those standards are the very ones that plaintiffs attack as unreasonable."  111Design defects can be determined, and under both section 3 and the limited consumer-expectations test are determined, by reference to the product's "manifestly intended functions" as indicated by the nature of the product and the manufacturer's own marketing of the product.

Yet the reporters for the Restatement Third adamantly resisted any attempt to describe section 3 as a limited consumer expectations test.  112 The reasons given by the reporters and others for rejecting the consumer expectations rubric - its use to prevent liability for products with "open and obvious" yet unreasonable danger, its lack of guidance in situations in which no clear consumer expectations regarding a product's features or functions have been frustrated, and its inapplicability to bystanders  113 - do not apply to the limited version of the test as a supplement to the consumer-oriented risk-utility test, as a number of courts have observed and held.  114

In the end, if one makes the effort to take into account all of the relevant sections and comments in the Restatement Third,  115 it is generally consistent with the current content and structure of product liability law, although many jurisdictions disagree with its treating the existence of a reasonable alternative design as an independent requirement for establishing a design defect, in addition to the consumer-oriented risk-utility analysis, rather than merely a relevant but not indispensable factor to be considered in the risk-utility analysis. However, as others have noted, the particular structure and rhetoric of the Restatement Third, which emphasize the reasonable alternative design requirement and unqualified risk-utility balancing and minimize the role of consumer expectations, give it a mean-spirited appearance, at least, and perhaps a related effect, contrary to the consumer-oriented focus of section 402A that continues to be favored by many courts.  116

III. Rationales

Having examined in Part II above the content and structure of current product liability law and its reflection in the Restatement Third, we can now consider the extent to which the current law and the Restatement Third's provisions can be explained and justified by the various rationales that were relied upon by Justice Traynor in Escola, which continue to be relied upon in the Restatement Third. The discussion of each rationale begins with Traynor's exposition of the rationale in Escola.

A. Efficient Compensation (Loss Spreading)

Those who suffer injury from defective products are unprepared to meet its consequences. The cost of an injury and the loss of time or health may be an overwhelming misfortune to the person injured, and a needless one, for the risk of injury can be insured by the manufacturer and distributed among the public as a cost of doing business… . Against such a risk there should be general and constant protection and the manufacturer is best situated to afford such protection. 117

This rationale is an efficient compensation argument, using utility or happiness rather than wealth as the good to be maximized for society as a whole. The argument is based on the assumption that the more widely and thinly a loss is spread, the less aggregate unhappiness there will be, which in turn is based on the generally plausible assumption of the declining marginal utility of money (and every other good). Product sellers are presumed to be better loss spreaders than injured plaintiffs (whether purchasers, users, or bystanders), since product sellers can spread the losses to their customers (by increasing product prices), employees (by decreasing wages), and stockholders (by decreasing dividends), or even more widely by purchasing liability insurance. Although liability schemes generally are more expensive to administer and less often applicable than first-party or employer or government-provided health or disability insurance, many individuals still have little or no health or disability insurance. Shifting from negligence to strict liability will increase the injuries for which product sellers are liable, thus increasing the spreading of losses, thus enhancing aggregate utility.

The Restatement Third invokes this rationale as a reason for holding wholesalers and retailers, as well as manufacturers, strictly liable for harm caused by construction defects:

An often-cited rationale for holding wholesalers and retailers strictly liable for harm caused by manufacturing defects is that, as between them and innocent victims who suffer harm because of defective products, the product sellers as business entities are in a better position than are individual users and consumers to insure against such losses. In most instances, wholesalers and retailers will be able to pass liability costs up the chain of product distribution to the manufacturer. When joining the manufacturer in the tort action presents the plaintiff with procedural difficulties, local retailers can pay damages to the victims and then seek indemnity from manufacturers. 118

This rationale supports holding wholesalers and retailers strictly liable only if they are better loss spreaders than the injured plaintiff. This often may not be true for small, single-store retailers. Even if the retailer is a better loss spreader than the injured plaintiff, there is no reason, from the perspective of efficient compensation, to hold the retailer liable if the manufacturer is a better loss spreader and the plaintiff can recover from the manufacturer. 

In such a situation, even if the loss is eventually passed back to the manufacturer through contract or tort indemnification actions, there will be wasteful cumulative litigation costs that would be avoided by the direct action against the manufacturer.

The Restatement Third provides another argument for holding wholesalers and retailers strictly liable for construction defects, which could be interpreted as an efficient compensation rationale, an efficient deterrence rationale, or an inferred negligence rationale:

Holding retailers and wholesalers strictly liable creates incentives for them to deal only with reputable, financially responsible manufacturers and distributors, thereby helping to protect the interests of users and consumers. 119

The efficient compensation interpretation of this argument holds wholesalers and retailers (hereafter "proximate sellers") strictly liable to give them an incentive to deal with "financially responsible" manufacturers and distributors (hereafter "remote sellers") as a means of increasing the probability of efficient loss spreading through the remote sellers. It assumes that the proximate sellers will either be able to obtain indemnity agreements from the remote sellers, which may not be possible, or will expect to face a substantially lower probability of being held accountable for some or all of a plaintiff's injury if a financially responsible remote seller is available from whom the plaintiff can recover. However, it suffers from the same problems of wasteful cumulative litigation and sub-optimal loss spreading as the prior rationale.

These problems could be avoided if the availability of a sufficiently financially responsible remote seller immunized a proximate seller from liability, which is in fact the position adopted in many states and by the Restatement Third. However, from the perspective of efficient compensation, this would be undesirable if the proximate seller is a better loss spreader than the remote seller - for example, a large retail chain such as Walmart that sells goods produced by a small manufacturer.  120 In that situation, focusing on the relative loss-spreading capacity (and availability) of the retailer and the manufacturer would support a result that never occurs: holding the retailer liable and excusing the manufacturer.

A much more significant problem with the loss spreading rationale is that it is way too powerful. As the New Jersey Supreme Court recognized and held in Beshada v. Johns-Mansville Products Corp.,  121 before beating a hasty retreat two years later in Feldman v. Lederle Laboratories, 122 the loss spreading rationale justifies strict liability not only for construction defects, but also for design and warning defects defined by using a consumer expectations or hindsight risk-utility test. If these risks were unforeseeable to the manufacturer, they were even more unforeseeable to the injured plaintiff, and the manufacturer is assumed to be (and generally is) in a better position than the plaintiff to obtain insurance for (or to self-insure against) unforeseen as well as foreseen product-related risks, or to spread the losses ex post to current and future consumers, employees, and shareholders.

Indeed, the loss spreading rationale would justify pushing liability much further. If maximum spreading of losses is the goal, why allow the exclusion of liability, as occurs under both the consumer expectations test and even a hindsight risk-utility test, of the known risks posed by "unavoidably unsafe" products? Why have any defect requirement?  123 Why exclude liability in instances of unforeseeable use or misuse or allow a defense of contributory negligence? Why apply strict liability to the provision of products but not services? Why exclude liability for causing pure economic loss?  124 The efficient compensation goal, by itself, provides no limits on the liability of the best loss spreader. Yet, for that very reason, in the end it argues against rather than for strict product liability, or indeed any sort of tort liability. If the objective is to spread losses as thinly as possible, the best approach is not to expand administratively expensive and infrequently applicable tort liability, but rather to have the government provide or subsidize universal, adequate health and disability insurance. Loss spreading would be maximized by eliminating tort liability entirely and moving to nationwide social insurance funded by progressive taxation.  125

Another major problem with the loss spreading rationale, from the efficiency perspective, is that it conflicts with efficient deterrence: to the extent that a loss can be shifted or spread, there is less incentive to avoid it.  126

A different "enterprise responsibility" loss sharing rationale, which is motivated by fairness rather than utilitarian efficiency, might not be as expansive and might not displace tort liability, depending on how the "enterprise" is defined.  127 The Restatement Third includes a version of this argument, which it describes as a fairness argument, among its potpourri of rationales for strict liability for construction defects: "Many believe that consumers who benefit from products without suffering harm should share, through increases in the prices charged for those products, the burden of unavoidable injury costs that result from manufacturing defects."  128

The principle of "fairness" in this argument is not clear. Enterprise responsibility would not further interactive justice or distributive justice. Interactive justice does not seek to rectify all losses but only losses that result from unjust interactions - those involving conduct that is inconsistent with another's right to equal negative freedom (security of person and property). 129 Product users who benefit without injury from the use of a manufacturer's product do not, by that sole fact, act inconsistently with the right to equal negative freedom of other product users who unfortunately suffer injury. Similarly, contrary to what its advocates seem to think, enterprise responsibility would not be an implementation of distributive justice. Distributive justice seeks to promote, to the extent possible, each person's equal positive freedom (access to the resources needed to pursue a meaningful life). Implementation of distributive justice requires ranking all persons in society according to their relative resources and needs and then redistributing from all those who have more than their proper share of resources to all those who have too little. Such society-wide calculations and redistributions are impossible in individual tort actions.  130 More importantly, under strict product liability or any other tort action, defendants are held liable to injured plaintiffs regardless of the parties' relative or overall wealth. As Jane Stapleton notes, the loss spreading rationale for tort liability cannot be justified from a (non-utilitarian) moral perspective:

The mere portrayal of insurance as a goal of tort law … undermines any moral justification of the liability of individual defendants because it is, in effect, an argument that losses should be shifted off the victim and then redistributed back via defendants' prices to victims. While the exploitation of individual defendants in this way may have pragmatic justifications in terms of convenience for buyers, it has no moral basis… . Moreover, in practice the system works unjustly as a insurance mechanism being … open to charges of paternalism and allegedly perverse redistributional effects, in particular that it is fiscally regressive. 131

In any event, as with the efficient compensation argument, the enterprise responsibility argument, if valid, would justify strict liability not only for construction defects but also for design and warning defects,  132 non-defective products, services as well as products, and pure economic loss. The Restatement Third does not attempt to explain how either argument can be restricted and contained.

Instead, when discussing why strict liability should not apply to design and warning defects, the Restatement Third merely states that, "from a fairness perspective, requiring individual users and consumers to bear appropriate responsibility for proper product use prevents careless users and consumers from being subsidized by more careful users and consumers, when the former are paid damages out of funds to which the latter are forced to contribute through higher product prices." 133 

Once again, this argument, even if valid, does not distinguish construction defects from design and warning defects, defective products from non-defective products, products from services, or physical injury from pure economic loss. 

Moreover, it is not a valid argument for rejecting strict liability. Subsidization of careless users by careful users could easily be avoided by a contributory negligence defense to the strict liability action, without depriving careful users and bystanders of the benefit of the strict liability action.

B. Efficient Deterrence (Risk Reduction)

Even if there is no negligence … public policy demands that responsibility be fixed wherever it will most effectively reduce the hazards to life and health inherent in defective products that reach the market. It is evident that the manufacturer can anticipate some hazards and guard against the recurrence of others, as the public cannot. 134

This rationale is an efficient deterrence argument. It assumes that total injury costs and prevention costs will be minimized by making whoever is the "cheapest cost avoider" for certain types of accidents strictly liable for all injury costs resulting from such accidents, whether or not that person is already exercising reasonable care. That person will then have the constant incentive to take the most cost-effective measures to deal with such accidents. Product manufacturers are presumed to be the cheapest cost avoiders for accidents involving their products: they are in the best position to discover and eliminate or mitigate, through appropriate design or warnings, the risks of injury created by their products. They will not engage in excessive (inefficient) risk reduction, because when the costs of further risk reduction exceed the concomitant reduction in expected damages, they will choose to pay the expected damages rather than the greater costs of further risk reduction.

The Restatement Third explicitly identifies the risk-reduction rationale for strict liability as an instrumental efficiency rationale and invokes it to support strict liability for construction defects:

On the premise that tort law serves the instrumental function of creating safety incentives, imposing strict liability on manufacturers for harm caused by manufacturing defects encourages greater investment in product safety than does a regime of fault-based liability under which, as a practical matter, sellers may escape their appropriate share of responsibility. Some courts and commentators also have said that strict liability discourages the consumption of defective products by causing the purchase price of products to reflect, more than would a rule of negligence, the costs of defects. And by eliminating the issue of manufacturer fault from plaintiff's case, strict liability reduces the transaction costs involved in litigating that issue. 135

Like the efficient compensation rationale, this rationale supports strict liability for design and warning defects as well as construction defects, as the Beshada court again recognized and held, and, furthermore, supports liability in the absence of any defect based merely on a causal connection between the product and the plaintiff's injury.  136 Assuming (as is certainly true) that product manufacturers are often not held liable for failing to disclose known or suspected risks to users of their products or for failing to develop alternative designs or products in the light of known or suspected risks, and are rarely if ever held liable for failing to engage in adequate research to discover and deal with the risks posed by their products, holding defendants strictly liable under the hindsight risk-utility test or the consumer-expectations test will lead to improved risk-reduction, since defendants will have a constant liability-generated incentive to take cost-justified precautions to reduce potential injury costs, whether or not plaintiffs can prove the defendant's failure to take such precautions.  137

There also seems to be little reason under this rationale - certainly less than under the efficient compensation rationale - to exclude recovery for pure economic loss. While the plaintiff may be (better) able to insure against such loss,  138 the product manufacturer generally will be in a much better position than the plaintiff to reduce the risks created by the product, especially since the plaintiff in cases involving pure economic loss often will be a bystander rather than a user of the product.  139

Furthermore, the efficient deterrence argument would seem to apply to sellers or providers of services as well as products.  140 The seller or provider of a service is in at least as good a position to reduce the risks posed by the service as the seller or provider of a product, and often in a better position since service-related risks are usually much more controllable than product-related risks and the provider of the service remains in control of the service while the provider of the product shifts control of the product to the product user. Moreover, given this difference in control and involvement, the recipient of a service is often in a worse position to reduce the risks posed by the service than the user of a product.

The efficient deterrence rationale supports holding wholesalers and retailers strictly liable if they are the "de facto" product manufacturers who have contracted to have the actual construction done for them by someone else. It also supports holding them strictly liable as an incentive for them to deal with manufacturers who engage in efficient risk-reduction or as an indirect means of encouraging efficient risk-reduction by manufacturers through indemnity actions, especially if the injured plaintiff cannot obtain jurisdiction over the manufacturer. Although transaction costs might be lowered by immunizing the wholesaler or retailer from strict liability if the injured plaintiff can recover from the manufacturer, as provided by many states and the Restatement 

Third, there is no necessity to do so from the perspective of efficient deterrence. Even without an indemnity agreement, the manufacturer is likely to pay all or almost all the damages, and retaining strict liability for the wholesalers and retailers will provide some incentive for them to put pressure on the manufacturer to engage in efficient risk reduction, even if they have an indemnity agreement.  141

The Restatement Third makes several unsuccessful attempts to avoid the logical extension of the efficient deterrence argument for strict liability from construction defects to design defects and warning defects:

Many product-related accident costs can be eliminated only by excessively sacrificing product features that make products useful and desirable. Thus, the various trade-offs need to be considered in determining whether accident costs are more fairly and efficiently borne by accident victims, on the one hand, or, on the other hand, by consumers generally through the mechanism of higher product prices attributable to liability costs imposed by courts on product sellers… .

[The Restatement definitions of design and warning defects] achieve the same general objectives as does liability predicated on negligence. The emphasis is on creating incentives for manufacturers to achieve optimal levels of safety in designing and marketing products… . Society benefits most when the right, or optimal, amount of product safety is achieved. 142

The reporters elaborate on this argument in the reporters' note to section 2:

The rationale supporting the risk-utility-based standards for defectiveness in design and failure-to-warn cases is implicit in those standards. Liability is imposed whenever the designer or marketer of a product is in a relatively better position than are users and consumers to minimize product-related risks… . The requirement that the plaintiff establish defectiveness in design and warning cases is necessary to create appropriate incentives to cause users and consumers to engage in safe use and consumption of products. Users and consumers are relatively helpless with respect to harm caused by hidden manufacturing defects; but they are more often better risk minimizers than are product sellers with respect to generic risks involving the inherent design of products. Thus, users of knives are thought to be the appropriate actors to bear legal responsibility for the harm caused by the sharpness of such products. Liability for knife-related accidents unrelated to manufacturing defects should not be imposed on commercial sellers, who have no opportunity to design or market knives that are incapable of causing injury. 143

These arguments ignore the availability of the contributory negligence and product misuse defenses in a strict liability action and the imbalance of information and expertise between product manufacturers and users, consumers, judges, and juries regarding known or suspected risks. While consumers and users (but not bystanders) may be the cheapest cost avoiders with respect to proper use given known "unavoidable" risks - such as the sharpness of a knife - which are inherent in the desired functioning of the product, product manufacturers would seem to be the cheapest cost avoiders for unknown but potentially discoverable risks, risks of which they but not consumers are aware or should be aware, and known residual risks that exist despite safe and efficient use by consumers.

Improper, inefficient use by consumers could be averted through the contributory negligence and product misuse defenses, while retaining strict liability for the prima facie case against defendants to maintain the constant incentive for them to take full account of known risks and to engage in cost-justified research and development. However, unlike current law as restated in the Restatement Third, which employs comparative responsibility to reduce but not necessarily bar recovery by a contributorily negligent plaintiff,  144 efficiency theory states that, if there is a contributory negligence defense, it should be a complete defense, in order to minimize administrative costs.  145 Moreover, it is doubtful that a contributory negligence defense is actually needed to deter inefficient behavior by product users, who independent of any such defense already have a substantial incentive to avoid injuries to their persons and property, which given the uncertainties and cost of litigation and natural and legal restrictions on remediation will never be fully remedied and often will be totally or seriously unremedied even under a strict liability rule.  146

The Restatement Third acknowledges that "to hold a manufacturer liable for a risk that was not foreseeable when the product was marketed might foster increased manufacturer investment in safety," yet claims that "such investment by definition would be a matter of guesswork. Furthermore, manufacturers may persuasively ask to be judged by a normative behavior standard to which it is reasonably possible for manufacturers to conform."  147

The first point again ignores the problem of imperfect information by users, consumers, judges, and juries regarding risks and alternatives known to but not disclosed by manufacturers of the relevant product - as with asbestos and cigarettes for many years, and no doubt many other toxic substances. It also overstates the problem of "guesswork" in manufacturer decisions to invest in research on possible risks and new technology.

The second point is a fairness argument rather than an efficient deterrence argument, which if valid would seem to apply to manufacturers' efforts to avoid construction defects as well as design and warning defects. Moreover, it is a misconceived fairness argument. As the Beshada court pointed out, it incorrectly assumes that a manufacturer that is held strictly liable is being "judged" in a normative moral sense and being told that it should have behaved differently.  148Strict liability, by definition, is liability in the absence of moral or legal fault. Holding a defendant strictly liable does not imply that the defendant should have behaved differently. Indeed, in other tort contexts in which strict liability is employed, the defendant is believed to have been justified in behaving as it did but nevertheless is still held liable - for example, when a defendant intentionally trespasses on and injures the plaintiff's property to save lives or much more valuable property,  149 intentionally maintains a private nuisance (a substantial interference with the plaintiff's use and enjoyment of her land) and is made to pay damages but is not enjoined, 150 or maintains an ultrahazardous activity and is held liable when injury occurs without any negligence.  151 Even negligence liability does not necessarily imply moral fault or an assertion that the defendant could and should have behaved differently. Insane, young, or "hasty and awkward" defendants who are held liable for intentionally or negligently causing some injury despite clearly being unable to conform their behavior to the objective "legal fault" standard of care are held liable not because they are deemed morally at fault but because of the need to protect the persons and property of those whom they injure.  152

A similar misconception underlies the Restatement Third's attempt to portray construction defects as involving an element of intentional injury that allegedly does not exist with respect to manufacturer decisions regarding designs and warnings:

Because manufacturers invest in quality control at consciously chosen levels, their knowledge that a predictable number of flawed products will enter the market place entails an element of deliberation about the amount of injury that will result from their activity.

… The element of deliberation in setting appropriate levels of design safety is not directly analogous to the setting of levels of quality control by the manufacturer. When a manufacturer sets its quality control at a certain level, it is aware that a given number of products may leave the assembly line in a defective condition and cause injury to innocent victims who can generally do nothing to avoid injury. The implications of deliberately drawing lines with respect to product design safety are different. A reasonably designed product carries with it elements of risk that must be protected against by the user or consumer since some risks cannot be designed out of the product at reasonable cost. 153

The knowledge that some conduct or activity repeated many times over an extended period is statistically almost certain to cause some number of injuries does not constitute intent to cause such injuries in common sense or law. The "knowledge of a near certainty" type of intent exists only with respect to a discrete, concrete situation.  154 Moreover, as the Restatement Third clearly assumes,  155 there is at least as much conscious deliberation regarding types and levels of risk and hence statistically likely injuries in making decisions about designs and warnings as there is in making decisions about quality control in manufacture, and in each context the consumer or user (or bystander) must deal with the residual risk of injury. The argument does not distinguish design and warning defects from construction defects.  156

A common argument, not noted in the Restatement Third, for not extending strict liability from construction defects to design and warning defects is that, since the latter are generic defects that occur in every instance of the product, the total liability costs will be much greater than with nongeneric construction defects and may increase the price of the product so much as to put it beyond the reach of some consumers or even lead to the product's being withheld or withdrawn from the market, even though the product is useful and apparently beneficial in the light of the risks that were foreseeable at the time that it was sold. However, although the defect may be generic, the frequency with which the defect produces injury usually is quite low, often as low as the frequency of construction defects. Thus, for many design and warning as well as construction defects, the expected liability costs are such a small percentage of total production costs that they will not significantly affect production or prices. 

Even when the expected liability costs are a significant part of the total production costs, if the expected costs of injuries caused by the product (including those due to broadly but not specifically foreseeable risks)  157 exceed the utility of the product, as measured by the price that consumers are willing to pay, the efficient deterrence perspective implies that the product should not be sold. There is a problem, under the utilitarian but not the wealth-maximization version of efficiency theory, if consumers' willingness to pay is hindered due to inability to pay. However, if that is the case, as it may well be with some prescription drugs, the efficient solution would be for the government to subsidize the purchase of the product, while retaining strict liability as a constant incentive for efficient risk-reduction as well as efficient loss spreading.  158

C. Inferred Negligence

The injury from a defective product does not become a matter of indifference because the defect arises from … unknown causes that even by the device of res ipsa loquitur cannot be classified as negligence of the manufacturer. The inference of negligence [under the res ipsa loquitur doctrine] may be dispelled by an affirmative showing of proper care. If the evidence against the fact inferred is "clear, positive, uncontradicted, and of such a nature that it can not rationally be disbelieved, the court must instruct the jury that the nonexistence of the fact has been established as a matter of law." An injured person, however, is not ordinarily in a position to refute such evidence or identify the cause of the defect, for he can hardly be familiar with the manufacturing process as the manufacturer himself is. In leaving it to the jury to decide whether the inference has been dispelled, regardless of the evidence against it, the negligence rule approaches the rule of strict liability. It is needlessly circuitous to make negligence the basis of recovery and impose what is in reality liability without negligence. If public policy demands that a manufacturer of goods be responsible for their quality regardless of negligence there is no reason not to fix that responsibility openly. 159

This argument by Traynor is ambiguous. Although he discusses the great difficulty that plaintiffs face in proving that their injuries were caused by negligence in the manufacturing process, he seems more interested in setting aside concerns about proving negligence in favor of other arguments for strict product liability. Others have focused on the proof problems and have argued that those problems provide a strong reason for imposing strict liability for construction defects. 160 The others include the authors of the Restatement Third, who describe this argument as a "fairness" argument: "In many cases manufacturing defects are in fact caused by manufacturer negligence but plaintiffs have difficulty proving it. Strict liability therefore performs a function similar to the concept of res ipsa loquitur, allowing deserving plaintiffs to succeed notwithstanding what would otherwise be difficult or insuperable problems of proof."  161

This argument views strict liability for construction defects as being based on the inferred negligence rationale that underlies the traditional res ipsa loquitur doctrine.

The courts created the res ipsa loquitur doctrine to deal with situations in which the plaintiff's injury probably was negligently caused by the defendant, but it is practically impossible or extremely difficult for the plaintiff to identify the specific negligence, especially when all the relevant evidence is under the control of the defendant. Rather than deny liability in all such situations, the courts believe that justice is better served by allowing the factfinder to infer that the injury was negligently caused by the defendant if the plaintiff establishes two conditions:

(1) the injurious event that occurred ordinarily does not occur in the absence of negligence, and 

(2) other possible sources of causal negligence other than the defendant have been sufficiently eliminated or would not have been sufficient to produce the injurious event in the absence of negligence by the defendant.  162

As applied to manufacturer liability for a product defect, these conditions require that the plaintiff prove that the defect arose while the product was under the control of the manufacturer, thus satisfying the second condition, and that such a defect ordinarily would not occur in the absence of negligence, thus satisfying the first condition. The shift to strict liability retains the second condition but eliminates the first condition. What justifies the elimination of the first condition for construction defects?

One argument that is sometimes made is that construction defects that arise during the manufacturing process ordinarily are due to negligence in that process, so that the first condition is always satisfied for a construction defect and need not be proven case by case. This argument, however, is not obviously true for many products and production processes. Especially as manufacturing processes become increasingly automated, construction defects may arise most frequently as "inevitable" imperfections that occur during the automated processing in the absence of any human error or inadvertence. 

For example, in Escola, the majority, after noting that any visible defects in the bottles the defendant bottling company used should have been discovered through reasonable visual inspection, concludes that it was not likely that any of the new bottles used contained latent defects not discoverable by visual inspection because the supplier of the bottles employed pressure tests that were "almost infallible."  163 

However, the pressure tests were done on only approximately one out of every 600 bottles, and the court does not mention how many of the tested bottles have latent defects.  164 Presumably some do, and roughly the same percentage of untested bottles would be expected to have similar latent defects. One would need to know this percentage, as well as the percentage of tested and untested bottles with visible defects that pass without detection of those defects through the visual inspection process, which usually is impossible to obtain, to calculate the probability of an uncaught defect's being the result of presumed negligence (because it was visible) rather than presumed non-negligence (because it was latent).

The Escola majority's use of the res ipsa loquitur doctrine is an expansion of that doctrine, motivated by the same insuperable proof problems that motivate the traditional res ipsa loquitur doctrine. The res ipsa loquitur doctrine permits an inference of negligent causation to be drawn against a defendant who probably negligently caused the plaintiff's injury when the plaintiff, through no fault of her own but rather due to insuperable proof problems inherent in the situation, cannot identify and prove the specific negligent conduct of the defendant that caused her injury. 

The Escola majority simply takes another modest step. It permits an inference of negligent causation to be drawn against a defendant who may probably have negligently caused the plaintiff's injury when the plaintiff, through no fault of her own but rather due to insuperable proof problems inherent in the situation, cannot establish specific negligent conduct of the defendant that caused her injury or a probability that the defendant negligently caused her injury.

Some might object that the extension of the res ipsa loquitur doctrine in Escola is a radical departure from, rather than a modest extension of, preexisting law and policy, on the ground that the traditional res ipsa loquitur doctrine, unlike its extension in Escola, does not involve any relaxation of the usual proof requirements but rather is a straightforward implementation of the "preponderance of the evidence" burden of proof, which merely requires establishing a greater than fifty percent probability of the fact at issue. However, this interpretation of the preponderance standard, although commonly stated, is incorrect, as the courts generally realize when the argument is made in a naked statistical manner. The statistical odds of something's happening (e.g. a horse's winning a race) can never establish what actually happened, no matter how high the odds are. Proof of what actually happened depends on concrete evidence from the particular occasion that instantiates one story of what happened and discredits the instantiation of competing stories, to a degree sufficient to induce a minimal belief in the truth of the fact at issue.  165

By permitting an inference of negligent causation based on non-particularistic, statistical evidence that the injury or accident ordinarily does not occur in the absence of negligence, the traditional res ipsa loquitur doctrine already constitutes a major departure from the usual requirement that the plaintiff prove that on the particular occasion the defendant actually was negligent and that the defendant's negligence caused the plaintiff's injury. The fact that it is a major departure is implicitly acknowledged by most courts when they treat satisfaction of the conditions for the doctrine's being applicable as only giving rise to a permissible inference of negligent causation rather than a rebuttable presumption, even when the defendant does not introduce any contrary evidence. 

Apparently, an unease about allowing an inference that the defendant was negligent in the particular case, based merely on the statistical fact that most such injuries are due to negligence, leads the courts to allow but not require the factfinder to make the inference, despite the adverse effect such a do-as-you-feel position has on equal treatment of similarly situated defendants and plaintiffs and the rule of law. If it is applied, the res ipsa loquitur doctrine should result in a rebuttable presumption rather than a mere permissible inference, under both the traditional version of the doctrine and the expanded version employed in Escola.

One more step in the argument is needed to justify a non-rebuttable presumption of negligence - in effect, strict liability - for construction defects. The step is suggested although not articulated by Traynor in Escola. A manufacturer's rebuttal evidence on the cause of a construction defect almost always will be limited, as it was in Escola, to evidence of the manufacturer's general quality control practices, which, as Traynor noted, merely reintroduces the insuperable proof problem that the plaintiff faces given the impossibility of obtaining specific proof of what caused the particular defect.  166 

To avoid making the res ipsa loquitur doctrine - and negligence liability for construction defects - generally vacuous, the defendant should only be allowed to rebut the inference or presumption of negligence through specific proof of non-negligent causation of the defect in the particular instance,  167 or at least through concrete evidence of the specific causes of a statistically relevant and valid sample of construction defects that establishes that most such defects are not due to negligence. Rarely if ever will such evidence be obtainable, by the defendant or the plaintiff. Even videotaping the actual production of each instance of the product and x-raying it at one or more crucial stages in the manufacturing process - which almost never occurs - is unlikely to document all or even most instances of negligence, and the videotapes and x-rays usually are not preserved.  168 Doubts about the soundness and credibility of such evidence might support its rejection even in the rare instances when it might be offered.

This argument parallels the arguments underlying other tort doctrines that shift the burden of proof or allow proportional recovery in situations in which the defendant possibly tortiously caused the plaintiff's injury but insuperable proof problems inherent in the type of situation prevent the plaintiff (and usually also the defendant) from establishing the tortious cause of the injury.  169 Being unable due to the insuperable proof problems to resolve the interactive-justice issue between the parties under the usual, first-best liability rules, the courts have developed second-best liability rules to do justice the best they can rather than giving up on doing justice.

In addition to the argument based on insuperable problems in proving specific negligence, there is a more direct negligence-based argument for strict liability for construction defects, which focuses on the objective nature of the negligence standard and the rationale for that objective nature. As previously noted, the negligence standard for those putting others at risk is an objective "legal fault" standard rather than a subjective moral fault standard. Interactive justice requires that the standard be set at the level that is necessary and appropriate to secure others' equal freedom and their related rights in their persons and property, and a person is held liable for failing to meet that standard regardless of the person's ability to achieve it.  170

Although it is impossible for any actual person always to achieve the relevant standard of care, the "reasonable person" of negligence law is the ideal person who always does so, and a failure to do so is negligent: "The actor is required to do what this ideal individual would do in his place. The reasonable man is a fictitious person, who is never negligent, and whose conduct is always up to standard."  171

The proper level of safety for a product is set by the reasonable intended design. A product that deviates from the reasonable intended design fails to achieve that proper level of safety. If the failure is due to a human deviation from proper procedures in the manufacturing process, that human deviation, no matter how inadvertent or "innocent," is negligent. Otherwise, the failure must be due to some flaw arising in the nonhuman portion of the manufacturing process. A manufacturer should not be allowed, by substituting machines for humans, to avoid liability for such deviations from the ideal process, just as an employer is not allowed, by substituting others for herself, to avoid liability for inevitable human error and inadvertence. The negligence of the employee is imputed to the employer, who is held strictly liable as a result of this vicarious liability. Similarly, the flaw in the production process, which results in the product's failure to achieve the level of care set by the reasonable intended design, is attributed to the manufacturer, who is held strictly liable as a result of this vicarious liability.  172

Each of these negligence-based rationales not only is consistent with, but depends heavily on, the defect requirement and the prevailing definition of a construction defect. Consideration of each also explains the lack of general strict liability for design or warning defects.

The second rationale relates a product's deviation from the intended design to the objective nature of the negligence standard and the need for an objective standard to secure persons' equal freedom and their rights in their persons and property. While construction defects, as deviations from the intended design, result in the product's failure to achieve the objective standard of safety set by the reasonable intended design, the issue for design and warning defects is precisely what the objective standard of safety should be, which is resolved through use of the interactive-justice-based negligence tests: the consumer-oriented risk-utility test for product designs and the foreseeable material risk test (with consideration of information overload issues) for product warnings.  173

The strict liability rationale based on insuperable problems in proving negligence rests on two assumptions: 

(1) there is a significant possibility that there was negligence and 

(2) it is practically impossible for the plaintiff to prove such negligence.

Both of these assumptions are much weaker in the design and warning context than they are in the manufacturing context. Regarding the second element, it generally will be much less difficult for the plaintiff to prove negligence in the overall design or in warning about the foreseeable risks, since both of these issues are eval‎uated in the light of (usually generally available) abstract technical and economic information, than to prove negligence in the manufacture of the particular instance of the product, which requires concrete evidence of what caused the flaw in that particular instance of the product. There is an even more significant problem with the first element. Unlike construction defects, which being deviations from the intended design often result from negligence, we do not ordinarily assume, merely because a product causes injury, that there is a significant possibility that it was negligently designed or that there was a negligent failure to warn. Even when the design or warning seems unreasonable now in the light of current knowledge (the strict liability hindsight approach), that is not a sufficient basis for inferring that it must have been unreasonable in the light of what was reasonably foreseeable at the earlier time when the product was designed, manufactured, and sold (the negligence foreseeable-risk approach).

Nevertheless, there clearly is reason to be concerned about product manufacturers' demonstrated ability to hide evidence of what they knew or should have known about the risks posed by a product (e.g., asbestos and cigarettes), and it is very difficult to establish negligent failure to engage in reasonable testing and research. Manufacturers generally have much better information than plaintiffs about the "state of the art" with respect to possible technology and product risks. Although it is not generally impossible, as with construction defects, for plaintiffs to obtain the necessary information to prove negligence, it can be very burdensome for them to have to do so. Requiring them to do so may result in their failure to prosecute, at all or sufficiently well, many valid claims of negligent design or failure to warn. These concerns justify shifting the burden of proof, or at least the burden of producing evidence, to the manufacturer on the technical state of the art and knowledge regarding product risks, and perhaps also related economic feasibility issues, when there is a significant question about these matters.  174 

This has been done by a few states and partially by the European Directive on Product Liability.  175 A larger group of states has gone further and uses hindsight risk-utility analysis, as the initial draft of the Restatement Third would have done, which theoretically results in true strict liability.  176 The concerns about the difficulty and expense of proof also justify, as stated in comment f to section 2 of the Restatement Third, foregoing the need to engage in a detailed risk-utility analysis when the availability of a reasonable, safer alternative design is obvious.  177 They justify, as stated in section 3 of the Restatement Third and held by many courts, dispensing with the need to prove negligence when there was an obvious failure of a specific intended or advertised feature or function of the product,  178 which almost always will involve either a construction defect (for which there is strict liability based on inferred negligence), a negligent design, or a negligent failure to warn.

The inferred-negligence rationales discussed so far do not apply to wholesalers or retailers or others in the chain of distribution of the product, who generally are not involved in the manufacture or design of the product or the formulation of the product warnings but rather are mere conduits through which products pass. 

These non-manufacturers can be held liable under normal negligence rules for a construction defect caused by their own negligent mishandling of the product or for a negligent failure to warn of product risks of which they are aware or should be aware. 

The only reason to hold them liable for defects attributable to the manufacturer is if they failed to exercise reasonable care to deal in products from manufacturers who could reasonably be expected to take proper care in designing and manufacturing those products and warning about the related risks and who would be available to compensate plaintiffs for any injuries nevertheless caused by construction defects (which involve inferred negligence) or negligent design or warning. While obvious failures of non-manufacturers to exercise such care could be proven through normal negligence analysis, non-obvious failures might be very difficult to prove, which could justify using the inferred negligence rationale to hold the non-manufacturer strictly liable unless the manufacturer is subject to suit by the plaintiff and solvent, especially since the non-manufacturer can avoid the strict liability by only dealing in products of manufacturers who are available and solvent. This rule has been adopted by many jurisdictions, but others hold non-manufacturers vicariously strictly liable for defects attributable to the manufacturer even when the manufacturer is available and solvent.  179

The inferred-negligence rationales support the distinction between products and services, since proving negligent service, although sometimes difficult, will generally not involve the insuperable proof problems that almost always exist when attempting to prove the specific negligent cause of a construction defect, nor be as difficult or burdensome as trying to establish a negligent design. 180 Non-liability for pure economic loss is also easily justifiable, as a simple application of the preexisting rule disallowing recovery for pure economic loss in ordinary negligence actions, on the grounds that no interference with the plaintiff's rights has taken place in the absence of physical interference with the plaintiff's person or property, and that physical injury to the product itself merely results in a non-working product and thus a failure of the plaintiff's economic expectations regarding the utility of the product, which is properly handled through contract law rather than tort law.  181 Similarly, the defenses of contributory negligence and assumption of the risk apply, as usual in negligence law, for reasons of justice rather than efficient deterrence.  182 Finally, since the inferred-negligence rationales are an extension of traditional negligence liability, they easily explain recovery by remote users and bystanders  183 and the inapplicability of contractual privity requirements, disclaimers, and limitations.

D. Consumer Expectations

The retailer, even though not equipped to test a product, is under an absolute liability to his customer, for the implied warranties of fitness for proposed use and merchantable quality include a warranty of safety of the product. This warranty is not necessarily a contractual one, for public policy requires that the buyer be insured at the seller's expense against injury.

The courts recognize, however, that the retailer cannot bear the burden of this warranty, and allow him to recoup any losses by means of the warranty of safety attending the wholesaler's or manufacturer's sale to him. Such a procedure, however, is needlessly circuitous and engenders wasteful litigation. Much would be gained if the injured person could base his action directly on the manufacturer's warranty.  184

As handicrafts have been replaced by mass production with its great markets and transportation facilities, the close relationship between the producer and consumer of a product has been altered. Manufacturing processes, frequently valuable secrets, are ordinarily either inaccessible to or beyond the ken of the general public. The consumer no longer has means or skill enough to investigate for himself the soundness of a product, even when it is not contained in a sealed package, and his erstwhile vigilance has been lulled by the steady efforts of manufacturers to build up confidence by advertising and marketing devices such as trade-marks. Consumers no longer approach products warily but accept them on faith, relying on the reputation of the manufacturer or the trade mark. 

Manufacturers have sought to justify that faith by increasingly high standards of inspection and a readiness to make good on defective products by way of replacements and refunds. The manufacturer's obligation to the consumer must keep pace with the changing relationship between them; it cannot be escaped because the marketing of a product has become so complicated as to require one or more intermediaries. Certainly there is greater reason to impose liability on the manufacturer than on the retailer who is but a conduit of a product that he is not himself able to test. 185

The first quotation reflects the sales law background of this rationale. Sales law developed an implied warranty of merchantability regarding the quality of a product - its fitness for the ordinary purposes for which it was sold - to replace the earlier regime of "caveat emptor" (let the buyer beware). As with contract law generally, liability for breach of the warranty is strict. Liability does not depend on proof of negligence, but it does depend on proof that the product contains a defect or deviation from the norm that renders it unmerchantable. Initially, again as with contract law generally, the implied warranty only encompassed the expected economic value or utility of the product. However, when tort law took a wrong turn from which it took a long time to recover, by not allowing a person injured by a product to hold the manufacturer (or any remote seller) of the product liable for negligently causing that injury unless he was in privity of contract with the manufacturer, sales law was modified to allow recovery against the immediate seller under the implied warranty of merchantability, who could then sue her immediate seller on the same basis, and so on back to the manufacturer. However, not only was this backtracking "needlessly circuitous and … wasteful," ultimate passage of liability back to the manufacturer would not occur if a link in the chain was broken by the unavailability or insolvency of, or a disclaimer or limitation of liability by, the seller occupying that link. Moreover, recovery under the warranty was - and to some extent continues to be - burdened by inspection, reliance, and notice of injury requirements.  186

Traynor and others argued, ultimately successfully,  187 that the injured plaintiff should be allowed a direct action in tort against the manufacturer, free of all the limitations imposed by sales law but retaining sales law's strict liability, based on the same reasonable consumer expectation of product quality regarding safety that had been imported into sales law to remedy the plaintiff's inability, at the time that the importation occurred, to sue the manufacturer in tort.  188 The basis for the consumer's reasonable expectation of safety, which sounds in tort rather than contract, is laid out in the second quote above.  189 Once the tort action against the manufacturer became available, the distortion of sales law to include expectations regarding the safety of a product in addition to expectations regarding its economic value and utility no longer is necessary or (perhaps) appropriate, but rather than being eliminated it has been retained, along with further distortions such as the elimination, for claims involving personal injury only, of the privity requirement and the ability to disclaim or limit warranties.  190

As with all the other rationales, the Restatement Third asserts that the consumer expectatons rationale supports strict liability for construction defects, but not design and warning defects:

Products that malfunction due to manufacturing defects disappoint reasonable consumer expectations of product performance… . Consumer expectations as to proper product design or warning are typically more difficult to discern than in the case of a manufacturing defect… . Such defects cannot be determined by reference to the manufacturer's own design or marketing standards because those standards are the very ones that plaintiffs attack as unreasonable. 191

Once again, the asserted distinction does not hold up. Just as construction defects can be defined as a failure of the product to conform to the intended design, section 3 of the Restatement Third acknowledges that design defects can in part be defined as an obvious failure of the product to perform a clearly intended function. The product's clearly intended functions, which are indicated by the nature of the product and the manufacturer's marketing, are viewed from the perspective of the reasonable consumer.  192 Consumer expectations also play a decisive role in the other test for a defective design, the consumer-oriented risk-utility test, under which risks are reasonable only if they are acceptable from the standpoint of the ordinary consumer.  193

Moreover, contrary to the Restatement Third's assertion, reasonable consumer expectations do not explain strict liability for construction defects, nor indeed, unlike with design defects, play any role in the definition of construction defects. In the absence of a disclaimer or warning, product purchasers do have a general expectation that the products they purchase will be fit and safe for their intended purposes, but they are also aware that sometimes products are defective ("lemons") and that product warranties (if they exist) are limited to fairly brief periods and often exclude liability for consequential damages. A consumer expectation of no defects and complete safety is unrealistic, and moreover would justify strict liability not only for construction defects but also for design and warning defects, and indeed in the absence of any defect.

In the end, all that consumers can reasonably expect is that product manufacturers will take reasonable care in designing and manufacturing products and warning about related risks, which is the negligence standard.  194 

The shift from negligence to strict liability for any type of defect cannot be explained by the consumer expectations rationale, but rather only by the inferred negligence rationales.  195

There are other problems with the consumer expectations rationale as an independently sufficient explanation and justification of current product liability law. It cannot explain liability to bystanders. It would allow liability to be excluded or limited by explicit disclaimers and limitations, or if the danger posed by the product, no matter how unreasonable, is "open and obvious." It would preclude liability for any unintended use of the product. It does not provide a reason for distinguishing between products and services or for precluding liability for pure economic loss when it is employed outside of sales law.

IV. Conclusion

Current product liability law is a negligence-based liability regime. 

Its content and structure, including its strict liability features, can all be explained in terms of negligence law.

The strict liability that exists for construction defects, for clear failures of products to perform their intended functions, and for non-manufacturers in the chain of distribution of products (if they are allowed to avoid liability when the manufacturer can be held liable) is based on and supportable only by inferred negligence.

Consumer expectations play a vital role in negligence law in general and product liability law in particular. 

The risk-utility test in product liability law is not an efficiency-oriented aggregate-cost-benefit test, according to which risks and harms to some are justified by benefits to others, but rather a consumer-oriented test which conforms to the general test in negligence law for those putting others at risk.

According to this general test, the creation of significant foreseeable risks to others is unreasonable unless the risks are substantially outweighed by direct or indirect desired benefits to those put at risk, cannot be reduced further without loss of those desired benefits, are not too serious, and are made known to those put at risk through proper warnings.

The Restatement Third's consumer-oriented risk-utility test for design defects and consumer-oriented material significance test for required warnings are proper elaborations of this general test in the product liability context.

Unfortunately, however, the Restatement's overall structure and rhetoric, which denigrates consumer expectations and emphasizes optimal risk reduction and unqualified risk-utility balancing for product designs and warnings, conveys a quite different, consumer-unfriendly impression.

This impression is reinforced by the Restatement Third's invocation of and primary reliance on (albeit only superficially) the normatively unattractive efficiency rationales, especially the efficient deterrence rationale, which is doubly unfortunate because, as the Beshada opinion accurately explains, the efficiency rationales lead to the unprincipled, unbounded liability that the Restatement Third is eager to avoid.

The reporters for the Restatement Third defended the inclusion of the efficiency rationales on the ground that they often appear, or at least used to appear, in court opinions.  196 The fact that they are mentioned in court opinions makes it even more important to emphasize that these rationales do not support and cannot explain a properly limited and just law of product liability, in order to avoid excesses such as the Beshadaopinion and to provide a more principled and coherent ground for judicial decision-making.

The Restatement Third's adoption of a supposedly neutral "functional" approach, layered in efficiency rhetoric but with only a superficial and transparently incoherent discussion of the efficiency rationales and other proffered rationales, left it without an appealing, coherent normative base and thereby undermined, and continues to undermine, both its acceptance by and its helpfulness to courts and others seeking a principled law of product liability.