Designing the products liability system to promote efficiency is justifablebecause the injurer (seller) and victim (consumer) typically are in a contractualrelationship. Contracting will not lead to efficient outcomes when consumersundervalue the benefits of seller liability, as would occur, for example, whenconsumers underestimate product risk. Although tort liability often wouldreduce product risk in these situations, forcing sellers to pay for product-causedinjuries is likely to increase the average cost of injury compensation. Thistension between safety and insurance considerations makes it difficult to reachfirm conclusions regarding the efficiency properties of the main products liabilitydoctrines. Nevertheless, in many instances the legal rules do not depend uponthe relevant economic considerations, suggesting that the current system couldbe made more efficient.

JEL classification: D18, K13, L15

Keywords: Products Liability, Product Risk, Product Safety, Insurance

1. Introduction

Products liability--the body of law governing the allocation of losses caused byproduct use--has rapidly gained prominence over the past 50 years. Theimportance of products liability stems from the substantial social cost ofproduct-caused injuries. According to government data, product accidents inthe United States cost roughly $50 billion per year ( Keeton et al., 1989, p. 2 ).These data are crude, however ( Viscusi, 1984, pp. 48-55 ). Relying on surveyevidence, Hensler et al. (1991) estimate that accidents in the United States,excluding those resulting in latent injuries, institutionalization, or death, imposedirect and work-loss annual costs of $175.9 billion or four percent of GrossNational Product. Approximately 30 percent of these accidents involved productuse, and another 18 percent were associated with motor-vehicle use. The socialcost of nonfatal product accidents is substantial, then, and including fatalitiesand latent injuries (like those caused by exposure to toxic substances)considerably increases the total. The magnitude of these losses and the volumeof product transactions indicate that products liability rules have a significantimpact on producers, consumers, and the general economy. Consequently,products liability has become one of the most important, and politicallycontroversial, forms of civil liability.

Legal scholars who analyzed the emerging field of products liability rarelyaddressed efficiency concerns ( McKean, 1970a ; Priest, 1985 ). Similarly, courtopinions in products liability cases have paid little or no explicit attention toefficiency ( Henderson, 1991 ). But as the economic analysis of products liabilityhas developed over the past few decades, so too have legal decisionmakersbecome more concerned about the economic consequences of these liabilityrules. Today efficiency considerations often strongly influence the formulationof products liability laws, as reflected by the Restatement (Third) of Torts:Products Liability ( American Law Institute, 1997 ). This emphasis on efficiencyis defensible. Sellers include their liability costs in the product price. Consumers(potential victims) accordingly pay for and receive the benefits of tort liability,so their preference for efficient liability rules--those that maximize the net benefitof seller liability--should govern.

By analyzing products liability with an economic perspective, it becomesapparent that this body of law could be merely a specific application of contractlaw, since if unregulated market transactions were efficient, courts would onlyhave to enforce contractual allocations of product risk in order to ensureefficient outcomes. Many product-caused injuries are governed by tort law,however, making it necessary to identify the market failures that may justify tortregulation. Sections 2 through 10 accordingly develop the economic frameworkfor evaluating different liability rules. Sections 11 through 13 describe the impactthat the products liability system has had on product safety, innovation, and themarket for liability insurance. The remaining sections discuss the efficiencyproperties of the main doctrines in products liability.

2. The Basic Model for Analysing the Efficiency Properties of Contracting andTort Liability

Much of the economic analysis of products liability can be described in termsof a simple model. Shavell (1987) and Spulber (1989) provide more rigorousanalyses of many of these issues.

As the focus of the inquiry is on product-caused injuries, the model does notconsider any product characteristics unrelated to the risk of injury (such asaesthetics, functionality, and durability). Hence the "product" to be analyzed ishomogeneous in all respects except for the risk of injury posed by the productand the extent of contractual liability the seller incurs under the productwarranty. The following assumptions are also unrealistic, but most will be relaxedin the ensuing discussion. All firms have identical production technologies andsell the product, exclusive of safety and liability costs, in a perfectly competitivemarket at the unit cost of p . By making safety investments of s per unit ofproduct, a firm affects the probability or risk r(s) that the product will causeinjury. Increased safety investments reduce the risk of injury at a decreasing rate(r'(s) < 0; r"(s) > 0). All injuries caused by the product have a monetaryequivalent of L that is suffered by risk-neutral buyers who are identical andunable to influence the risk of injury.

In light of these assumptions, the total cost or "full price" P of the productis given by

P = p + s + r(s)L.

(1)

If perfectly informed consumers bear the injury cost L in the event ofaccident, they pay a purchase price of p + s for the product but recognize thatthis cost is increased by the expected accident cost r(s)L . Consequently,consumers make their purchase decisions on the basis of the full price P ratherthan the price they pay to purchase the product, so consumer demand Q ^D = Q ^D( P ) . Sellers then compete by offering the amount of safety and warrantycoverage that minimize P .

Under these conditions, it does not matter whether a perfectly informedconsumer or the seller is liable for the injury (e.g, Hamada, 1976 ). If the consumeris liable, the seller must choose the amount of safety investments to minimize P ,which from equation (1) implies that the seller chooses the amount s* defined by

1 = - r'(s*)L.

(2)

In other words, the seller invests in safety until the last dollar spent reducesexpected injury costs by one dollar. Such a product is optimally safe.

If the seller is fully liable for the consumer's injuries, it sells the product andwarranty at a price of p + s + r(s)L = P . Once again, the seller must minimize thefull price, so it chooses the optimal amount of safety investment s* . Whether theconsumer or producer is liable for the product-caused injury therefore does notaffect product safety or the full price.

3. The Significance of Imperfectly Competitive Markets

An early, influential justification for tort regulation was based on the notion thatmanufacturers can take advantage of their market power by supplying unsafeproducts ( Priest, 1985 ). However, the results obtained from the basic model areunaffected if the market is not perfectly competitive (e.g., Epple and Raviv, 1978 ).A seller's market power can be represented by the amount it can increase theproduct's full price above the competitive level. By increasing the product priceby this amount, the seller increases its profits per sale by that same amount.Alternatively, by reducing safety investments below the optimal level s* , theseller can also increase the product's full price as each $1 of reduced safetyinvestment necessarily increases expected accident costs r(s)L by more than $1.This strategy does not affect the seller's profits per sale, however, because theproduct must sell for a reduced price equal to the unit cost of p + s (any priceabove cost is equivalent to an increase in the product price). Hence a monopolistcan make higher profits by selling perfectly informed consumers an optimallysafe product at a supracompetitive price. Similar reasoning shows that if it wouldbe efficient for the seller to bear full liability under the warranty, then amonopolist would maximize profits by offering a full warranty while selling theproduct at a supracompetitive price (e.g., Heal, 1977 ).

It is possible, though, for market structure to affect product safety. The basicmodel assumes a constant marginal cost of safety investment (the term s ) perunit of product. Consequently, a manufacturer's decision regarding safetyinvestments does not depend upon its output level (as reflected by equation (2)above), implying that product safety will be unaffected by the reduced quantityof output that occurs in imperfectly competitive markets. Many product risks arelikely to depend upon the quantity of products sold or consumed by anindividual, however. As Marino (1988a , 1988b) points out, toxic chemicals maypresent a health hazard due to their cumulative effect on consumers. Conversely,consumers may develop a tolerance from cumulative exposure, thereby reducingthe risk. The higher prices, and reduced consumption, of products sold inimperfectly competitive markets would affect these kinds of product risk. Inaddition, when the cost of safety investments depends upon a manufacturer'soutput level, the amount of safety investments made by a monopolist dependson the cross-effects of safety investments and output on the monopolist's costs( Spulber, 1989, pp. 407-10 ). Whether sellers in imperfectly competitive marketssupply products that are insufficiently safe therefore is a difficult empiricalquestion. But if such market failures exist, they probably are better addressed bythe antitrust laws.

4. The Role of Consumer Information About Product Risk

The analysis so far has assumed that consumers are perfectly informed of risk,an assumption typically made by early economic analyses of products liability(e.g., McKean, 1970a ; Oi, 1973 ). But as Goldberg (1974) argued, product safetybecomes a regulatory problem only if consumers are inadequately informed.Subsequent economic analyses focused on the effects of imperfect information.

When imperfectly informed consumers are liable for their injuries, they mustestimate their expected injury costs, denoted E[r(s)L] , and hence the full price,denoted E[ P ] . Consequently, equation (1) above is changed to

E[ P ] = p + s + E[r(s)L].

(1)

In this setting, a seller must minimize E[ P ] if consumers are to buy itsproduct, so sellers choose the amount of safety investment s that minimizes E[ P ] :

1 = - E[r'( s )L].

(2)

Thus, when consumers are imperfectly informed of product risk, the sellerinvests in safety until the last dollar spent on safety reduces the consumer'sestimate of expected injury costs by one dollar ( Spence, 1977 ). If consumersunderestimate the decrease in expected injury costs, they will undervalue riskreduction and demand less than the optimal amount of safety; that is, if -E[r'(s)L] < -r'(s)L , then s < s* . A similar result occurs when consumers cannotobserve manufacturer safety investments, because consumers who cannot tellthe difference between a low-risk and high-risk product treat the differential insafety as if -E[r"(s)L] = 0 when in fact -r'(s)L > 0. Manufacturers have noincentive to incur the higher cost of producing the low-risk product, so theysupply only high-risk products, an outcome analogous to the "lemons problem"analyzed by Akerlof (1970) .

Imperfect information need not result in overly unsafe products. Ifconsumers overestimate the way in which increased safety investments reducerisk, they will attribute too great a value to safety investments and demand morethan the optimal amount of safety. Although this outcome is inefficient, it seemsunwise to construct a regulatory regime, with its attendant administrative costs,in order to reduce product safety. Hence there is a pressing need to regulatemarket transactions only if consumers undervalue safety investments.

5. Market Mechanisms that Promote Product Safety

Landes and Posner (1987, pp. 280-81) argue that it is too costly for consumersto obtain perfect information about product risks, and imperfectly informedconsumers tend to underestimate the small risks ordinarily posed by products,causing them to undervalue safety investments. In assessing this argument, wemust recognize that the cost consumers incur to get risk-related information, andtheir need for it, depends upon a variety of market mechanisms. For example,manufacturers have an incentive to provide optimally safe products if there is alarge enough proportion of well-informed "shoppers" in the market ( Schwartzand Wilde, 1983a ). The information held by some consumers therefore maybenefit others who undervalue product safety. Similarly, consumers whocommunicate among themselves by "word of mouth" advertising may increasethe amount of high-quality goods in the market ( Rogerson, 1983 ). Consumersalso can purchase product-related information from intermediaries, and suchinformation may come from sellers.

Brand names, for example, are a method sellers use to implicitly guaranteesuperior quality ( Klein and Leffler, 1981 ), because product quality must besufficiently high if the seller is to attract repeat purchases (e.g., Shapiro 1982 , 1983 ). For the same reason, sellers can convey indirect information aboutproduct quality through advertising and prices ( Milgrom and Roberts, 1986 ). Theway in which price signals quality, is highly dependent on the market context,however, as in some settings low prices signal high quality, whereas in othersettings high prices signal high quality ( Tirole, 1990, pp. 110-12 ). In addition,prices signal product quality only if consumers have at least some brand-specificinformation about quality, although this information need not be perfect( Wolinsky, 1989 ). As long as consumer experience with a product brandprovides enough information so that consumers are more likely to believe thebrand is of high quality when in fact it is, high-quality firms will attract morecustomers ( Rogerson, 1983 ).

The need to protect their reputation or brand name may force sellers toprovide more safety than is suggested by the analysis in the prior section.Nevertheless, it is unlikely that unregulated market transactions will yieldoptimally safe products when consumers are imperfectly informed of productrisk. A seller's reputation can remain intact even though its product is notoptimally safe, because consumers often have little or no ability to learn fromproduct use about the product's safety characteristics. Many risks are latent anddo not become manifest for years (like carcinogens). In addition, many safetycharacteristics are not observable during normal product use (such as whethera motor vehicle is optimally designed to minimize the risk of injury for differenttypes of accidents). Given the very low probabilities of most product-causedinjuries and the fact that optimally safe products typically pose some risk ofinjury, very little information will be conveyed by a consumer's experience of "noaccident" or "accident." For example, suppose an unsafe product doubles therisk of injury from 1 in 10,000 to 2 in 10,000. Based upon their experience, it couldtake consumers a long time (involving numerous iterations of Bayesianupdating) to discover the increased risk. Another problem is that theprice-quality relationship depicted by signalling models is based on equilibriumconditions for products that vary in one dimension of quality. Even within theconfines of such a simplified market, it is doubtful that consumers ordinarily willhave enough information about the market context to draw the correct inferencesabout product safety. And once one allows for the (realistic) possibility ofdisequilibria in markets for products that are heterogeneous in more than onedimension, it becomes even less likely that consumers will be able to obtaingood information about product safety from prices. Indeed, one empirical studyfound that price might serve as a quality signal for only one type of product --frequent but unimportant purchases ( Caves and Greene, 1996 ).

Given the limited amount of information provided by market mechanisms, itis puzzling why sellers do not voluntarily disclose risk-related information,particularly since such disclosures would be credible due to the legal prohibitionagainst fraud. Because only high-quality sellers would benefit from voluntarydisclosure, consumers could infer from the fact of nondisclosure that a seller isnot offering an optimally safe product. All sellers therefore would have todisclose, forcing them to provide optimally safe products.

It is possible that high-quality sellers do not voluntarily disclose risk-relatedinformation because consumers tend to overreact to negative information aboutproducts (see the sources cited in A. Schwartz, 1988, p. 381 ). Consequently, anyseller that discloses risk-related information could cause consumers to believethat its product is unsafe, so high-quality sellers are better off by not disclosing. Burrows (1992) provides other reasons why sellers might not voluntarilydisclose information about product risk, and Geistfeld (1997) explains why asystem of voluntary disclosure would function much like a tort regime ofnegligence.

6. Do Consumers Undervalue Product Safety ?

As the previous discussion suggests, individuals often process risk-relatedinformation in a manner that does not correspond to the standard economicmodel of decisionmaking. A substantial literature on cognitive psychology seeksto understand how individuals assess risks (e.g., Kahneman et al., 1982 ). Basedon these studies, A. Schwartz (1988 , 1992) concludes that consumers tend tooverestimate product risks, whereas Latin (1994) concludes that consumersusually underestimate risk and thus undervalue product safety. Both agree thesestudies find that individuals tend to overestimate risks that are brought to theirattention (which may explain why sellers do not voluntarily disclose risk-relatedinformation). Latin, however, persuasively argues that most product risks are notsalient because product-caused injuries are a rare occurrence for mostindividuals, leading consumers to infer (erroneously) from the more common orrepresentative experience of safe product use that risk is not present or worthworrying about. Consequently, as Landes and Posner claimed, imperfectlyinformed consumers tend to underestimate product risks.

Although consumer understanding of product risk is relevant to theregulatory problem, it should also be recognized that consumers can undervalueproduct safety even if they are perfectly informed of product risks. Supposeconsumers are risk averse and find it worthwhile to purchase a fullycompensatory health insurance policy. Suppose also that the product-causedinjury would be fully covered by this policy. Insurance companies ordinarily donot adjust premiums to reflect the riskiness of products purchased by theirpolicyholders ( Hanson and Logue, 1990 ). As the consumer's health insurancepremium is unaffected by her consumption choices, neither it nor the expectedcost of injury (which is fully insured) are relevant to the consumer's purchasedecision. The full price to the consumer consequently is given by P = p + s , andsellers minimize this full price by setting s = 0. Simply put, fully insuredconsumers have no need for risk reduction, so it does not pay for sellers toinvest in product safety. Of course, this example is extreme (because insurancepolicies rarely provide full coverage), but the conclusion is general: fullyinformed consumers will undervalue product safety when they can externalizesome of their injury costs onto an insurance company.

7. Product Warranties and the Use of Seller Liability to Promote Safety

As discussed in section 2, when the seller is fully liable for product-causedinjuries, the price at which the product sells on the market equals the full price,forcing the seller to provide the cost-effective amount of product safety. In thesecircumstances, imperfectly informed consumers only need to find the productthat sells for the lowest price in order to get the optimally safe product. Sellerliability therefore remedies the consumer's informational problem in astraightforward way, creating the possibility that imperfectly informedconsumers might be able to rely on warranties to obtain optimally safe products.

For example, assume as in Grossman (1980) that the manufacturer is theleast-cost insurer and that consumers are unable to observe manufacturer safetyinvestments. In this setting, insurance costs are minimized if the manufacturerprovides a warranty that fully compensates the consumer for anyproduct-caused injuries. A manufacturer that provides full warranty coveragemust also provide an optimally safe product in order to minimize the market price(which equals the full price) of its product. A manufacturer that does not providethe optimally safe product therefore would signal this fact to consumers due tothe product's higher market price, so to avoid this outcome such a manufacturercannot offer a full warranty. Imperfectly informed consumers would infer thistype of behavior, though, and assume that products without full warrantycoverage must not be optimally safe. Manufacturers accordingly have no choicebut to offer imperfectly informed consumers optimally safe products with fullwarranty coverage.

Thus, when sellers are the least-cost insurers, imperfectly informedconsumers can use product warranties to attain efficient outcomes: by choosingwarranties that impose full liability on sellers, consumers can ensure thatproducts will be optimally safe and insurance costs will be minimized. Fullwarranties (or seller liability in tort) might not result in such equilibria, though,if sellers purchase insurance to cover their liability under the warranty. A studydirected by the U.S. Department of Commerce found that liability insurance inthe 1970s was rarely priced in a manner that reflected the degree of risk posed bythe manufacturer-policyholder's products ( Inter-Agency Task Force on ProductsLiability, 1977 ). Although such insurance reduces the manufacturer's incentiveto invest in product safety (as the increased accident costs do not increasepremiums), developments in the liability-insurance market have significantlyrestored this incentive. Based on estimates of firms' total liability costs, Priest(1991) found that self-insurance costs accounted for 4.9 percent of the total in1970 and increased to 51.7 percent in 1979. The amount of uninsured riskexposure faced by firms probably increased in the 1980s. Moreover, productsliability insurance policies now commonly rely on pricing elements that areresponsive to the level of risk posed by the policyholder's products ( G.Schwartz, 1990, pp. 320-321 ). Hence there are good reasons for expecting that theprospect of liability gives sellers an incentive to invest in safer products.

8. Are Product Sellers the Least-Cost Insurer ?

Despite the safety benefits of seller liability, warranties that make sellers fullyliable for product-caused injuries are unlikely to be efficient because sellersrarely are the least-cost insurer for all product risks. Altough, manufacturers arelikely to have a comparative advantage in insuring against some risks, like thoseinvolving repair of complicated machinery, consumers typically will have acomparative advantage in insuring against other risks ( Priest, 1981 ). In particular,risk-averse consumers ordinarily will have a comparative advantage in insuringagainst many of the risks associated with physical injury, because the costconsumers incur in making their own insurance arrangements-- "first-partyinsurance"--often is lower than the cost sellers incur in making insurancearrangements to cover product-caused injuries suffered byconsumers--"third-party insurance." In part, first-party insurance is cheaperbecause it is more capable of minimizing the costs of moral hazard and adverseselection ( Epstein, 1985 ; Priest, 1987 ). The primary reasons for the costdifferential between the two insurance mechanisms stem from the event thattriggers coverage and the scope of coverage.

Coverage under many first-party insurance policies, such as healthinsurance, is triggered by the fact of loss (like medical expenses), making thecause of injury irrelevant in most cases. The fact of injury or loss usually is easyto prove (submitting bills), so policyholders typically do not have to hire alawyer to receive insurance proceeds. By contrast, the third-party insurancesupplied by product sellers is triggered only if the product caused the injury.Often, many products are causally implicated in an accident, and a potentiallycontentious factual inquiry may be needed to resolve the liability question( Geistfeld, 1992 ). Some items of damages, particularly those pertaining topain-and-suffering damages and future economic loss, are also costly todetermine. The resultant litigation expenses increase the cost of third-partyinsurance, which probably explains why the administrative costs of third-partyinsurance per dollar of coverage exceed the administrative costs of first-partyinsurance ( Geistfeld, 1992, pp.639-642 ).

With respect to the scope of coverage, third-party insurance providescompensation for pain-and-suffering injuries whereas first-party insurancetypically does not. It might be inefficient for consumers to insure againstpain-and-suffering injuries (for reasons given in section 20 below). If so, it wouldbe more efficient for consumers to suffer these injuries without compensation (aform of first-party insurance), providing another cost advantage for first-partyinsurance.

In other respects, the scope of coverage provided by third-party insuranceis not extensive enough, as it does not cover losses unrelated to product use.To cover these contingencies (like medical expenses due to illness), individualsneed to purchase other insurance. But since first-party insurance coverage isusually triggered by the fact of loss rather than its cause, individuals who havesuch insurance might receive double compensation when injured by products:the first-party insurer is obligated to pay whenever the policyholder suffered aninsured-against loss; and the seller is obligated to pay (due to thecollateral-source rule) even though the consumer received other insuranceproceeds. Double recovery can be avoided if the first-party insurer exercises acontractual or statutory right to indemnification out of the tort recovery receivedby the policyholder, but the separate legal proceeding often is complicated andexpensive due to the need to determine which part of the tort award or settlementis covered by the policy. For this and other reasons, many insurers do notexercise this right. Insurance provided by product sellers therefore may be aninefficient form of double insurance or otherwise increase the administrative costof first-party insurance policies, providing another reason why consumers mayreduce their insurance costs if they disclaim seller liability under the warranty.

Sellers therefore will typically not be the least-cost insurer for all productrisks. Hence, imperfectly informed consumers ordinarily will not be able to relyon full warranty coverage to ensure that products are optimally safe andinsurance costs are minimized. It is still an open question, though, whether tortregulation would be efficiency-enhancing.

9. The Regulatory Problem

To account for differences in the cost faced by consumers and manufacturersin insuring against product losses, L _I will denote the consumer's cost ofcompensating the injury and L _W the seller's cost of compensating the injuryunder the product warranty. Whether the seller is liable for the injury may affectproduct safety, so the seller's safety investment will be denoted by s _I when theconsumer insures against the injury and by s _w when the seller is liable under thewarranty. Finally, we will assume that any insurance costs faced by theconsumer equal the actuarially fair amount r(s _I)L _I . (The other extreme--the casein which premiums do not depend on risk--was discussed in section 6.)

There are two possible full prices to consider:

					P _I = p + s _I + r(s _I)L _I.	(3)
					P _W = p + s _W + r(s _W)L _W.	(4)

Consumers will disclaim seller liability when doing so would reduce the fullprice (that is, when P _I < P _W ), and otherwise will purchase full warranty coverage(when P _I > P _W ).

To illustrate how the difference in insurance costs affects the analysis,suppose consumers are unable to observe manufacturer safety investments. Forreasons given in section 4, manufacturers will set s _I = 0. Consumers, however,will infer such behavior on the manufacturer's part, recognizing that the full priceis given by P _I = p + r(0)L _I . By contrast, when the manufacturer is fully liableunder the warranty, it provides an optimally safe product. Hence P _W = p + s _W*+ r(s _W*)L _W . Even though product safety increases when the manufacturer isfully liable under the warranty ( s _W* > s _I = 0), if the consumer has a comparativeadvantage in compensating the injury ( L _I < L _W ), it is possible that P _I < P _W .Consumers therefore may be better off with the less-safe products and reducedinsurance costs than with the safer products and more expensive insuranceprovided by full product warranties.

Thus, there often is a tradeoff between safety and insurance considerationswhen consumers are imperfectly informed: although increasing the amount ofseller liability can lead to safer products, it is also likely to increase the averagecost of compensating an injury. This inefficiency does not necessarily create aneed for tort regulation, however ( Geistfeld, 1995a ). As long as imperfectlyinformed consumers can accurately compare P _I and P _W , as in the example justgiven, they will choose warranties that strike the appropriate balance betweenthe costs and benefits of seller liability. At best, a tort rule could achieve asimilar balance, but more likely it will not. Inefficiencies in product marketstherefore need not create an efficiency-enhancing role for tort liability.

Imperfectly informed consumers will not choose appropriate warranties,though, when they underestimate product risk and thus underestimate theproduct's full price. (If E[r(s)] < r(s) , then E[ P _I] < P _I .) In this case, consumerssometimes choose warranties that disclaim manufacturer liability when it wouldbe inefficient do so (that is, when E[ P _I] < P _W < P _I ). A tort rule that imposes fullliability on sellers would be efficiency enhancing in this situation. It is alsopossible, however, that consumers disclaim manufacturer liability when it wouldbe efficient to do so (because E[ P _I] < P _I < P _W ). Consequently, tort regulationis not necessarily efficiency enhancing when consumers underestimate productrisk.

The type of market failure that might justify tort regulation accordinglydepends upon conditions that cause consumers to disclaim seller liability whenit would be inefficient to do so. This conclusion is not affected by extending theanalysis to include the possibility that consumers can affect the risk of injury byexercising care while using the product. As long as sellers cannot observe theamount of consumer care, full warranty coverage is likely to reduce theconsumer's incentive to take costly efforts to avoid (the fully insured) injury.Yet, the reduction in warranty coverage reduces the manufacturer's incentive tomake costly safety investments, so the warranty must balance conflicting safetyand insurance considerations ( Cooper and Ross, 1985 ; Emons, 1988 ). Holdingmanufacturers liable in tort for product-caused injuries does not solve theinformational problem, however, so this form of tort regulation cannot improveupon a warranty that efficiently allocates liability given the informationalconstraint.

An additional consideration arises if consumers have different risk profilesdue to differences in product use, abilities to reduce risk for a given level of care,or damages. Although "low risk" and "high risk" consumers may demandproducts of different qualities, manufacturer liability can force sellers to provideonly one level of quality. According to Oi (1973) , that outcome is inefficientbecause low-risk (that is, low-damage) consumers are forced to subsidizehigh-risk consumers. Absolving sellers of liability would eliminate thisinefficiency, because sellers could then provide products of different quality atdifferent prices in a manner that sorts low-risk and high-risk consumers into theappropriate product markets. However, Ordover (1979) shows that in order forsuch separating equilibria to occur, low-risk consumers must differentiatethemselves from high-risk consumers by purchasing incomplete warrantycoverage. There may be cases in which the benefits of successful differentiationare less than the benefits of mandated seller liability. Hence tort regulation is notnecessarily inefficient even though some consumers would be better off withoutsuch regulation.

10. The Choice Between Negligence and Strict Liability

We have been analyzing seller liability in terms of a rule that holds sellers strictlyliable for injuries caused by product use. Most product accidents are governedby a rule of negligence, however, which makes sellers liable for injuries causedby products that are not reasonably safe. According to the economicinterpretation of negligence, a product is not reasonably safe if it contains lessthan the optimal amount of safety s* defined by equation (2) above. Becauseeach dollar of safety investment below s* increases expected accident (and thusliability) costs by more than one dollar, sellers minimize total costs by makingtotal safety investments equal to s* . Thus, a negligence standard that isproperly defined and perfectly enforced gives sellers an incentive to supplyoptimally safe products, the same incentive created by strict liability. Negligencediffers from strict liability in that consumers under a negligence rule bear liabilityfor injuries caused by optimally safe products, giving them the opportunity toenter into insurance arrangements that minimize the cost of injury compensation.In theory, then, a negligence regime can yield optimally safe products whileenabling consumers to minimize insurance costs.

Nevertheless, negligence will not lead to efficient outcomes, whenconsumers are imperfectly informed of product risk ( Shavell 1980 ; Polinsky,1980 ). Because sellers are not liable for injuries caused by their (optimally safe)products, the product sells for p + s* . Consumers in a negligence regimetherefore need to estimate expected injury costs r(s*)L _I in order to determine theproduct's full price P . Consumers who underestimate product risk willunderestimate the full price, increasing their demand above the amount theywould choose if they were perfectly informed. Thus, even though products areoptimally safe, consumers will purchase too many products (and there will be toomany firms in the industry). This overconsumption increases the total numberof injuries above the efficient amount whenever optimally safe products pose apositive risk of injury.

Another problem with a negligence rule is that it often will be difficult (andexpensive) for the plaintiff to show that the product should have been madedifferently. Consider, for example, the complicated issues that must be resolvedin order to determine whether a product is optimally designed. The cost oflitigating these issues may undermine the safety incentives of negligenceliability. Prior to filing suit, injured consumers who are not well-informed aboutmanufacturer safety investments often will be unable to determine whether theproduct is reasonably safe. These consumers (or their contingent-fee attorneys)may be unwilling to incur the cost of proceeding with the lawsuit, enabling somemanufacturers with suboptimally safe products to escape liability. Under theseconditions, a proportion of manufacturers choose to be negligent ( Simon, 1981 ).

Another reason for expecting that the negligence standard will not beperfectly enforced stems from the possibility that judges and juries will makemistakes. The complicated issues in products liability cases (many of which arediscussed below) make court error possible. Hylton (1990) shows that anegligence standard with court error and costly litigation can lead to over- orunderdeterrence. Overdeterrence can occur because sellers of optimally safeproducts may be held liable due to court error. By increasing product safety, theseller decreases the risk of injury, thereby reducing the likelihood that it will besubjected to a lawsuit and an erroneous imposition of liability. But even thoughcourt errors can increase product safety, the increased legal uncertainty hasdeleterious effects (also discussed later). Moreover, overdeterrence may involvethe withdrawal of socially beneficial products from the marketplace.

Strict liability, by contrast, is less costly for plaintiffs and easier for courts toadminister, which increases the likelihood that it will be perfectly enforced. Inaddition, strict liability can lead to the efficient level of risk even thoughconsumers are imperfectly informed. Hence strict liability has a better potentialfor reducing product risk. Negligence, on the other hand, allows for a greaterrange of insurance arrangements and accordingly has more potential to reducethe average cost of compensating an injury. The choice between negligence andstrict liability therefore reflects the same safety-insurance tradeoff describedearlier: increased seller liability (that is, strict liability) is likely to increase safetyand per-injury insurance costs, whereas decreased seller liability (negligence) islikely to reduce safety and the average cost of compensating an injury.

11. Empirical Studies of the Effect of Seller Liability on Product Safety

Whether seller liability reduces product risk is a difficult empirical question,because the available accident data are not sufficiently refined and the injuryrate is affected by a number of other factors such as changes in technology andthe composition of products and users. Indeed, data limitations undermine theconclusions one can draw from attempts to measure the impact that sellerliability has had on product safety. For example, Priest (1988a) compares theamount of products liability litigation to death rates and the rate ofproduct-related injuries requiring emergency room treatment, concluding that theexpansion in litigation had no discernible effect on accident rates. AlthoughPriest acknowledges that the study is exploratory, Huber and Litan (1991, p. 6) assert that it raises "serious doubts that the benefits of expanded seller liabilityhave been large." But as Dewees et al. (1996, p. 203) point out, Priest's studydoes not necessarily show anything about the relationship between sellerliability and accident rates. The accidents in the study could be caused by anumber of factors unrelated to manufacturer safety investments. Moreover,increased seller liability should reduce the number of "defective" (suboptimallysafe) products, but the injury data are not segregated into accidents involvingdefective and nondefective products, making it difficult to draw usefulconclusions from the study. For example, the prior level of tort liability couldhave significantly increased the number of nondefective products on the market.Greater consumption of these nondefective products (due to increased wealth,for example) could increase the overall injury rate, even though the expansionin seller liability reduced product risk by reducing the amount of defectiveproducts on the market.

Higgins (1978) relies on accidental fatalities in the home as a proxy forproduct-caused injuries. The econometric analysis finds that producer liabilityreduces the frequency of these accidents in states with low levels of educationalattainment and increases it in states with high levels. If low educationalattainment corresponds to imperfectly informed consumers, this study partiallysupports the claim that producer liability increases safety when consumers arenot well informed of risk. However, in addition to the previously mentionedproblems of relying on such aggregated accident data, this study is problematicbecause it measures the impact of producer liability in a state by reference to theyear when its highest court expanded producer liability by eliminating of thecontractual requirement of privity. It is doubtful that this expansion in sellerliability was significant enough to produce observable results, particularly sincethe numerous exceptions to the privity doctrine meant that sellers were alreadyexposed to considerable liability for injuries suffered by victims with whom therewas no direct contractual relationship.

Graham (1991) attempts to determine the relationship between productsliability and passenger-car death rates. The regression does not detect anybeneficial impact of liability on aggregate death rates, with the extent of liabilitymeasured by an index based on the annual number of crashworthiness casesreported in the LEXIS database. Measuring liability rules by published judicialopinions is particularly problematic, however, because most lawsuits are settledprior to trial. A very effective liability rule, for example, could cause all cases tosettle, giving sellers a strong incentive to reduce risk. Yet Graham's model wouldnot impute this risk reduction to the liability rule. Moreover, MacKay (1991) argues that federal regulations of automobile design have forced allmanufacturers toward a common standard, which undermines the attempt toderive a simple causal link between products liability and traffic accidents.

Other studies have circumvented these data problems (and created others)by asking producers how their behavior has been influenced by liability. Eadsand Reuter (1983) conducted interviews with nine large manufacturers,concluding that products liability significantly influences product-designdecisions. Based on interviews with 101 senior-level corporate executives fromthe largest publicly held companies in the United States, Egon ZehnderInternational (1987) found that over half of these companies had added safetyfeatures as a result of liability concerns. About 20 percent of the companieschose not to introduce new products on account of products liability. Two otherstudies conducted by the Conference Board surveyed risk managers and CEOsof major corporations, finding that products liability concerns led to significantsafety improvements while also causing a significant number of firms todiscontinue product lines or not introduce new products ( Weber, 1987 ; McGuire,1988 ). The Egon Zehnder survey is probably the most reliable due to its excellentresponse rate; the Conference Board surveys had poor return rates and mayhave been influenced by a variety of biases ( G. Schwartz, 1994a, pp. 408-410 ).

A different approach to evaluating the effects of seller liability examines theimpact of prominent products liability lawsuits on stock prices. Viscusi andHersch (1990) find that news stories reporting on products liability suitssignificantly decrease a firm's stock value. Similarly, Jarrell and Peltzman (1985) [criticized by Hoffer et al. (1988) ] and Rubin et al. (1988) find that safety-relatedadministrative actions (product recalls) substantially reduce stock prices. In allof these studies, adverse publicity concerning product safety costs the firmmore due to the reduced stock value than does the associated liability or recallcosts. These findings suggest that firms suffer a loss of reputation when thereis an adverse event (litigation or administrative action) pertaining to the safetyof its product. As described earlier, a firm's reputation for safety is importantwhen consumers are not well-informed of product risk. These studies thereforeindirectly confirm that individuals are not perfectly informed of product risks.Moreover, the loss in stock value gives firms an additional incentive to avoidproducts liability litigation, providing another reason for believing that sellerliability increases safety.

12. The Impact of Tort Liability on Innovation

The political debate regarding products liability reform in the U.S. has ofteninvolved the claim that tort liability reduces innovation and consequentlyundermines the competitiveness of domestic products in a global economy.Although tort liability probably has reduced some types of innovation, thewelfare effects of that reduction are unclear. Moreover, tort liability has alsoinduced beneficial innovation, making it even more difficult to assess the netimpact of tort liability on innovation.

Tort liability can increase a producer's cost, relative to a rule of no liability,by forcing the firm to increase its safety investments. Tort liability also requiresthat firms make disclosures in product warnings so that imperfectly informedconsumers can better estimate accident costs (see section 18 below). Insofar astort liability increases safety investments and consumer estimates of accidentcosts, there is an increase the product's full price. Consequently, tort liability islikely to encourage safety innovations much in the same way that othercost-driven price increases, such as those stemming from labor scarcity, induceinnovation. An increase in cost enhances the profitability for the firm of anyinnovation which reduces that cost. The resultant increase in firm demand forsuch technical change should produce more innovation, a theory of technicalchange called "induced innovation." This theory has substantial analytical andempirical support for innovations unrelated to product safety ( Thirtle andRuttan, 1987 ). There is no apparent reason why the theory is not also applicableto safety innovations.

For example, an optimal research and development (R&D) program withouta fixed budget will expend resources until the marginal cost of additionalresearch equals the marginal benefit. The benefit depends on the potential costsavings from the research, savings that are increased as firms face increased tortliability. Expansions in tort liability therefore should increase R&D expendituresfor safety technologies. This conclusion is consistent with the analytical resultsobtained by Daughety and Reinganum (1995) , and the empirical study by EgonZehnder International (1987) which found that over half of the surveyedcompanies had increased their R&D expenditures as a result of liability concerns.Insofar as the increased R&D expenditures have yielded more safetyinnovations, tort liability has promoted safety innovation.

A liability rule that increases the product's price can also have a negativeeffect on innovations unrelated to product safety. Assuming that the increasedprice reduces consumer demand, both theory ( Binswanger, 1974 ) and historical

evidence ( Schmookler, 1966 ) indicate that the reduced profitability of the productline discourages innovation. But insofar as the change in demand reflectsconsumer response to a product price that more accurately reflects accidentcosts, the reduced innovation may be welfare enhancing.

Viscusi and Moore (1991a , 1991b , 1993) study the effect of liability costs oninnovation, finding that firms with new products have higher liability insurancecosts. Econometric analysis shows that increased seller liability increases safetyincentives, but at some point further increases in liability reduce innovation bymaking new products unprofitable ( Ibid., 1991b , 1993 ). One study (1993) showsthat 10 industry groups were at or near this threshold in the mid-1980s,indicating that the incentive effects of seller liability vary across industries. Thisvariable effect is confirmed by case studies of different industries regarding theimpact of tort liability on innovation ( Ashford and Stone, 1991 ; Craig, 1991 ; Graham, 1991 ; Johnson, 1991 ; Lasagna, 1991 ; Martin, 1991 ; Swazey, 1991 ).

Products liability can also affect innovation due to its influence on thestructure of business organization. If a firm suspects that a product may poserisks that are long-term and likely to result in wide-spread injury, it has anincentive to avoid paying damages by divesting production tasks that involvesuch products. To insulate itself from legal liability, the parent company mustdivest early in the R&D stage. This dynamic is consistent with an empiricalstudy of the U.S. economy which found that increased seller liability appears tohave increased the number of small corporations in hazardous sectors between1967 and 1980 ( Ringleb and Wiggins, 1990 ; see also Merolla, 1998 ). The cost ofinnovation for products involving such risks will be increased by the need todivest an operation that can more cost-effectively (absent liability concerns) beadministered within a single organization. The increased cost in turn providessome disincentive for innovation.

13. The Relationship Between Tort Liability and the Market for LiabilityInsurance

A report published by the U.S. Attorney General's Tort Policy Working Groupconcluded that increased tort liability was a major cause of the so-called "liabilityinsurance crisis" that occurred in the mid-1980s ( U.S. Department of Justice,1986 ). The liability-insurance market was in turmoil during this period: premiumrevenues tripled and the supply of coverage severely contracted ( Viscusi,1991a ). It is not evident why a contraction in the liability-insurance market wouldbe caused by legally mandated expansions in seller liability, however, asexpansions in tort liability should increase the demand for liability insurance.This conundrum has attracted much attention, leading to a number of differentexplanations for the liability-insurance crisis (surveyed in American LawInstitute, 1991a, pp. 66-97 ). For our purposes, the most interesting finding toemerge from this literature pertains to the way in which legal uncertainty affectsthe cost of liability insurance.

A standard liability-insurance policy covers a product seller's legal liabilityfor personal injury or property damage that "occurs" to third parties during thepolicy year. Often a number of years pass before legal liability is incurred by thepolicyholder (who is then indemnified by the insurer). During the periodbetween the issuance of the policy, manifestation of injury, and conclusion ofthe lawsuit, any changes in tort law may affect the costs the insurer will incurunder the policy. Thus, in order to forecast its expected costs for a group ofpolicies, a liability insurer needs to predict how tort standards, damage rules, andinsurance law (like the interpretation of an "occurrence") will change over time.In periods of legal stability, the insurer can be fairly confident about its predictedliability exposure. However, as Abraham (1987) and Trebilcock (1987) emphasize,there were various sources of legal uncertainty that liability insurers faced in the1980s, making it difficult to predict the likelihood or magnitude of covered losses.In theory, this increased uncertainty increased the variance of the insurer'sexpected loss and thus the cost of bearing that risk ( Venezian, 1975 ). Empiricalstudies also show that legal uncertainty increases the cost of liability insurance. Kunreuther et al. (1993) surveyed actuaries, underwriters, and insurers, findingthat they will add an additional cost above the expected value of loss when thereis uncertainty (or "ambiguity") regarding the probability or magnitude of theinsured-against loss. Similarly, in an econometric study involving a large numberof insurance policies issued during 1980-84, Viscusi (1993) concludes that riskambiguity tended to exert a positive influence on actual premium rates,controlling for the regulated rate. Winter (1991) provides a theoreticalexplanation for why this uncertainty can also affect the industry supply ofliability insurance. It seems likely, then, that any increased legal uncertaintycreated by the tort system contributed to the liability-insurance crisis in the1980s.

In response to this and an earlier insurance crisis in the 1970s, a number ofstates enacted legislation limiting a seller's tort liability. Most of these measuresalso reduced legal uncertainty (for example, by placing caps on the mostunpredictable elements of damages). Viscusi (1990a) finds that both theprofitability and availability of liability insurance were enhanced during 1980-84by prior legislative reforms that limited firms' liability. Viscusi et al. (1993) findthat the reforms adopted by the states between 1985 and 1988 reduced liabilitycosts and the premiums for liability insurance. This study also concludes thatits findings are consistent with the possibility that the fact of comprehensivereform is more consequential than its components. One way to explain such anoutcome is that the enactment of legislative reform reduces legal uncertainty byindicating that the legal climate is not hospitable to expanded tort liability forproduct sellers. In such a climate, liability insurers may be more confident thatthey will not be exposed to unanticipated expansions in legal liability, therebyreducing the cost of legal uncertainty that is built into premiums for liabilityinsurance.

But even if the reductions in seller liability mandated by these legislativereforms reduced liability costs and premiums, it does not follow that the reformswere efficient. Croley and Hanson (1991) argue that the rise in liability-insurancecosts reflected a more efficient level of deterrence due to the internalization ofcosts that had been externalized prior to the expansion of seller tort liability.Indeed, due to the higher cost of third-party insurance, increased seller liabilityshould have a pronounced effect on insurance costs. Moreover, becauseincreased tort liability will decrease demand when consumers underestimate risks(see sections 10 and 11 above), Manning's (1996) empirical finding that tortliability reduced consumer demand for childhood vaccines does not necessarilyestablish, as he claims, that consumers place no value on this form of tortinsurance. Instead, the relevant question for policy purposes is whether theincreased insurance costs of tort liability, and any decline in consumer demand,are justified by a reduction in product risk.

14. Introduction to the Economic Analysis of Products Liability Doctrines

Depending on the issue involved, the current products liability regime in theU.S. relies upon contracting, negligence, or strict liability to allocate liability forproduct-caused injuries. The prior analysis of the costs and benefits of thesemethods therefore can be used to analyze the efficiency properties of variousproducts liability doctrines. Consequently, the ensuing discussion will delineatethe role of contracting, negligence, and strict liability while raising a number ofpreviously undiscussed considerations relevant to the economic analysis ofproducts liability law. Epstein (1980) provides a more comprehensive overviewof products liability law and discusses the economic implications of variousdoctrines. The American Law Institute (1991b) provides more extensiveeconomic analysis of the main products liability doctrines.

Although this focus on U.S. law is limiting, the doctrines to be disscusedhave influenced the development of products liability laws in other countries,particulary the European Community and Japan.

15. The Focus of the Legal Inquiry and Its Implications for the Choice ofLiability Rules

Sellers are liable for their negligent conduct resulting in product-caused injuries.In the vast majority of states and the European Community, sellers are also liablefor injuries caused by product "defects." Although this rule is commonly called"strict products liability, " it is not the same as strict liability because liabilitydepends upon the existence of a defect.

In most states, "defect" is defined by reference to the product itself. Asdiscussed in the ensuing sections, the choice between negligence and strictliability follows from the definition of "defect" and is not based on the efficiencyproperties of these tort rules. Other states and the European Community define"defect" by reference to consumer expectations. Although it is easier to give thisapproach an economic interpretation, it too does not rely upon efficiencyconsiderations in making the choice between negligence and strict liability. Moreprecisely, the consumer expectations test can operate like a rule of strict liability,since an optimally safe product is defective if it does not conform to consumerexpectations. This outcome occurs when consumers underestimate risk, asproducts will always be more dangerous than consumers expect them to be.Conversely, when consumers are well-informed of product risk, the productalways conforms to consumer expectations and consequently absolves the sellerfrom tort liability. The consumer expectations test therefore limits tort liability tothe cases in which it has the greatest potential to be efficiency-enhancing (whenconsumers underestimate risk), but it does not rely upon an economic rationalefor its choice of strict liability over negligence.

16. Manufacturing Defects

Manufacturing defects are physical deviations from a product's intended design,thereby implicating the quality control of manufacturing and inspectionprocesses. These processes usually cannot be made perfect, so some productscontaining manufacturing defects will reach the marketplace. Whenever such adefect causes physical injury, the seller is liable even if it employed the mostefficient quality-control measures. Defining "defect" by reference to the productaccordingly results in a rule of strict liability for these cases. Jurisdictions thatrely on the consumer expectations test also employ strict liability for these casesby assuming that consumers do not expect a product to contain a manufacturingdefect.

Most agree that strict liability is the better rule for these cases. G. Schwartz(1979, pp. 459-62) , for example, argues that most manufacturing defects areattributable to negligence, but it often will be difficult for plaintiffs to prove thatthe seller or one of its agents did not use appropriate quality-control measures.Thus, even though negligence in principle would eliminate tort liabilitywhenever increased deterrence is not desirable--that is, when efficientquality-control measures already are being used--the benefit in these few casesis less than the costs that would be created for all cases if the plaintiff had toprove that the defect was caused by inadequate quality control. Strict liabilitymay also be more efficient because it gives sellers a better incentive to fosteradvances in technology that reduce the incidence of manufacturing defects( Landes and Posner, 1985 ).

17. Design Defects

A product that conforms to the manufacturer's design specifications is defectiveif the design is defective. Unlike manufacturing defects, which can be determinedby reference to deviations from product design, there is no readily availabledefinition of design defect. Consequently, courts had to develop such adefinition.

Many jurisdictions define defect by reference to consumer expectations. Thistest, however, suffers from an inherent ambiguity. The inquiry could addressconsumer expectations of product risk. As previously noted, because consumerswho underestimate risks will always find a product to be more dangerous thanthey expect, sellers are subjected to liability even if the product design satisfiesthe cost-benefit test. This logic explains the controversial result in Denny v.Ford Motor Company, 87 N.Y.2d 248 (N.Y. 1995), and is consistent with the rulethat consumer expectations justify holding sellers strictly liable formanufacturing defects. Alternatively, the inquiry could address consumerexpectations of product safety. Consumers who underestimate risk ordinarilyexpect less safety than is contained in a product. For example, consumers whoare unaware of risk expect there to be no safety investments, implying that anyamount of product safety surpasses consumer expectations, even if the productis less safe than would be efficient. That consumer expectations tend toestablish a safety standard below that of the cost-benefit test was recognized inthe influential case Barker v. Lull Engineering Company, 573 P.2d 443 (Cal. 1978).Whether consumer expectations should be defined by reference to risk or safetyis an issue that can only be resolved by determining why consumer expectationsmatter, an issue that courts have not adequately addressed. The choice betweenrisk and safety does not matter, though, for jurisdictions that definedefectiveness by reference to reasonable consumer expectations.

A reasonable consumer expects that sellers would reduce product risk in themost cost-effective manner. Hence a product design does not conform toconsumer expectations only if the seller failed to take measures that efficientlyreduce product risks. The consumer expectations test therefore can be turnedinto a negligence test for design defects. Note, though, that the logic needed tojustify a negligence rule for design defects is inconsistent with the rationale formaking sellers strictly liable for manufacturing defects, since reasonableconsumers also expect that sellers ordinarily are unable to eliminate allmanufacturing defects.

The other approach for defining a design defect is based on the risk-utilitytest. The traditional formulation of this test is not limited to the factors relevantto the issue of whether the product design efficiently minimizes product risk ( A.Schwartz, 1988 ; Viscusi, 1990b ). However, the Restatement (Third) of Torts:Products Liability states that "the test is whether a reasonable alternative designwould, at reasonable cost, have reduced the foreseeable risks of harm posed bythe product, and if so, whether the omission of the alternative design ... renderedthe product not reasonably safe" ( American Law Institute, 1997, p. 19 ). Therisk-utility test therefore has evolved into a cost-benefit test. Because this testabsolves sellers from liability (there is no defect) when the design efficientlyminimizes risk, these cases, in effect, are governed by a negligence rule.

The biggest problem with a negligence standard for design defects relatesto the court's ability to evaluate the technical engineering issues involved inproduct design ( Henderson, 1973 ; A. Schwartz, 1988 ). An erroneous finding ofdesign defect is particularly problematic, because tort liability potentiallyattaches to the entire product line. Consequently, any legal uncertainty in thisarea will have significant repercussions, suggesting that design-defect litigationhas significantly influenced developments in the market for liability insurance( Viscusi, 1991b ).

Courts could avoid these difficult issues by defining defectiveness on thebasis of relative safety, but that type of approach is unlikely to yield efficientoutcomes ( Boyd and Ingberman, 1997a ). First consider a rule which holds thata product is not defectively designed if it conforms to industry custom. Becauseconformance to custom immunizes firms from tort liability, custom reflects marketequilibria absent tort regulation. As such equilibria are often characterized byinefficiently low safety levels, adherence to custom is not ordinarily sufficientto establish that the product is properly designed (see section 9 above). Nowconsider an alternative rule that defines a product as being defectively designedwhenever a safer product is available on the market ("state of the art"). A sellerwhose product is defective under this definition is fully liable for all injuries, soit usually minimizes costs by choosing the efficient amount of safety. The seller,however, could avoid liability altogether by increasing its safety investmentsabove the efficient amount if doing so would make its product safer than otherson the market. This liability rule therefore might give sellers an incentive toprovide an inefficiently high amount of safety. Hence efficient safety levelsordinarily will not be obtained if courts determine defectiveness solely on thebasis of relative safety considerations pertaining to custom or state of the art.

The difficulty of determining whether a product is defectively designed hasled the courts to limit the scope of tort liability for design defects. Usually courtsare unwilling to consider whether a product is defective no matter how it isdesigned, recognizing that they cannot competently evaluate the total costs andbenefits of a product except in the most extreme cases ( Henderson and Twerski,1991 ). For example, courts will not consider whether a subcompact car isdefectively designed merely because larger (more expensive) cars are safer.Instead, design-defect litigation tends to involve modifications to existingproduct lines (like redesigning the gasoline tank in a subcompact car to reducerisk). Limiting the scope of tort liability in this manner allows the market todetermine the viability of product lines (subcompact cars versus larger, safercars), which enhances the likelihood that product lines can be varied to bettersatisfy consumers with different preferences. Under strict liability, by contrast,manufacturers make design choices by reference to the average consumer,thereby reducing the variety of product lines offered in the market and thelikelihood that heterogeneous consumers can find products that closely matchtheir preferences ( Oi, 1973 ).

18. Warning Defects

A product can be defective because it does not adequately warn or instruct theconsumer about product risks. As was true for design defects, the courts had todevelop a definition for this type of defect, with most choosing to definewarning defects in terms of a cost-benefit or risk-utility test. In principle, tosatisfy this test the warning must provide the minimal amount of informationnecessary for the representative consumer to estimate the product's full price,which can occur only if the warning increases the consumer's information bydescribing unavoidable material risks and cost-effective methods of use thatreduce risk ( Geistfeld, 1997 ). Courts have recognized that warnings which satisfythese criteria are not defective and accordingly absolve the seller of liability,even if the warning did not disclose a risk that injured the plaintiff. The liabilitystandard for warning defects therefore operates like a rule of negligence.

By contrast, the consumer expectations test functions like a rule of strictliability for nondisclosed risks that are not sufficiently appreciated by theordinary consumer. One implication of this approach is that the seller is liableeven if the risk was not scientifically knowable at the time of sale. In order tomake this and other cost-related considerations relevant to the liabilitydetermination, the test must adopt the expectations of a consumer whoreasonably expects sellers to disclose risks whenever it would be cost-effectiveto do so.

At present, the most problematic aspect of this form of tort liability relates tothe cost of disclosure. A warning is defective if it does not disclose, sufficientlydescribe, or properly emphasize the risk which caused injury. Even if the benefitof a proposed warning alteration is slight, courts often find the warning to bedefective on the ground that the cost of the requested disclosure is minimal ornonexistent ( Henderson and Twerski, 1990 ). This is a mistake. Empirical studieshave found that the amount and format of hazard information contained in aproduct warning affects consumers' ability to recall the information, so thatadded disclosures can reduce the effectiveness of other disclosures in thewarning (e.g., Magat and Viscusi, 1992 ). Additional disclosures also increase thetime consumers must spend to read warnings. Because these costs of disclosureare not sufficiently recognized by the courts, sellers have an incentive todisclose more than the optimal amount of risk information, thereby reducing theeffectiveness of the warning for most consumers. For example, upon readingdisclosures that offer little or no benefit, most consumers may rationally decidethat the cost of reading the entire warning is not worth the effort.

Some courts have recognized that the risk-utility test for warnings shouldaccount for information-processing costs. This position is taken by theRestatement (Third) of Torts: Products Liability ( American Law Institute, 1997,p.32 ). Indeed, ignoring the way in which information costs affect consumerbehavior is inconsistent with the various rules regarding an adequate warning( Geistfeld, 1997, p. 328 ). As virtually all jurisdictions have adopted these rules,there is ample precedent for courts to rely upon information costs whenevaluating warnings. If they do, jury instructions can be formulated that wouldsignificantly improve the likelihood that jurors will properly evaluate warnings( Ibid, pp. 329-37 ), although some argue that jurors and judges cannotcompetently evaluate information-processing costs ( Latin, 1994, p. 1284 ).

Another reason for believing that the warning doctrine is not currentlyproducing efficient outcomes pertains to the method of disclosure. The mosteffective form of risk communication probably involves symbols and commonformats ( A. Schwartz, 1992 ; Viscusi, 1993b ). The public-good nature of effectiverisk communication may require a regulatory rather than judicial solution.

For this reason, strict liability is unlikely to result in efficient warnings,contrary to the argument of Croley and Hanson (1993) . Cooter (1985) shows thatstrict liability may lead to inefficiently strong warnings because the manufactureronly considers how disclosure affects profits rather than social welfare. Thisresult is hard to understand, however. For risks that are unavoidable or inherentin the product, disclosure will not reduce risk (or liability costs) unless it inducesthe buyer not to purchase the product. In some situations, disclosure wouldinduce only high-risk buyers to opt out of the market, so the seller could reduceaverage liability costs by disclosing. But if disclosure does not reduce averageliability costs, the seller has no incentive to disclose even when disclosurewould be efficient. By contrast, when disclosure pertains to care that theconsumer must exercise in order to reduce risk, the strictly liable seller has anincentive to disclose the efficient amount of information--that which minimizesaverage liability (injury) costs.

Strict liability also gives sellers an incentive to discover the efficient amountof information ( Shavell, 1992 ). But since sellers are liable for risks that were notscientifically knowable at the time of sale, strict liability could result in inefficientoutcomes. Firms that otherwise are financially viable may be forced intobankruptcy by unanticipated liability costs that could not have been discoveredby a cost-effective R&D program ( A. Schwartz, 1985 ). Negligence avoids thisproblem, but different formulations of the negligence rule are possible and notall of them induce sellers to acquire the efficient amount of information ( Shavell,1992 ). And because plaintiffs often will have a hard time proving that a seller wasnegligent for not having discovered information, sellers apparently do not havea sufficient incentive to research product risk ( Wagner, 1997 ).

19. Defenses Based on Consumer Conduct

In most states and the European Community, recovery is reduced for plaintiffswhose misuse of the product combined with the defect in causing the injury.Whether "comparative fault" is less efficient than barring the plaintiff fromrecovery depends upon a variety of factors ( Shavell 1987, pp. 83-104 ), but itseems unlikely that comparative fault reduces the consumer's incentive to usethe product properly under ordinary circumstances. Survey evidence shows thatfor product-associated injuries that are serious but do not result in latentinjuries, long-term institutionalization, or death, only seven percent of victimsin the U.S. take action to initiate a liability claim if the injury did not occur atwork, and 16 percent take action if the injury occurred at work ( Hensler et al.,1991, p. 127 ). For the vast majority of cases, then, individuals do not expect torecover any damages from the seller, so comparative fault plays little, if any, rolein the individual's decision regarding product use. Denying recovery to thoseindividuals who misused the product for other reasons, or due to inadvertence,would reduce the seller's incentive to invest in product safety. This seems to bea large price to pay in exchange for the occasional benefit of denying recoveryto those plaintiffs who intentionally misused the product because they expectedto receive some compensation from the seller, particularly since thecompensation that such individuals receive depends upon proof of defect andis likely to be substantially reduced by comparative fault principles.

A more worrisome question is whether a product that is nondefective innormal use can become defective when misused. Many jurisdictions requiresellers to design products to account for foreseeable misuse. Landes and Posner(1987, pp. 299-301) argue that this doctrine could be efficient if properly limited.

Difficult issues also surround the defense of assumption of risk, which insome jurisdictions bars a plaintiff from recovering. The defense could be efficientif it limits seller liability to those cases in which consumers are not well-informedof risk. But merely because a consumer is aware of a risk does not imply that sheis well-informed, particularly since perfect information involves anunderstanding of how different safety configurations affect risk (and price).Moreover, availability of the defense gives sellers an incentive to make designdefects visible or to disclose the risk in the warning. Latin (1994) argues thatwarnings ordinarily are less effective at reducing risk than are design changes.

20. Nonmonetary Damages

Plaintiffs can recover monetary damages for nonmonetary injuries such as painand suffering. These damages may be inefficient. Nonmonetary injuries alter the

individual's utility function (the victim receives less utility for any given level ofwealth following the accident), which can affect the marginal utility of wealth indifferent ways. These different effects are important, because an individualmaximizes welfare by purchasing insurance (a transfer of money from thenoninjured state to the contingent, injured state) until the marginal utility ofwealth in the "injury" and "no injury" states of the world are equalized. Fornonmonetary injuries that increase the marginal utility of wealth, individualsprefer a positive amount of insurance compensation. The insurance proceedsreduce the marginal utility of wealth in the injured state so that it equals themarginal utility of wealth in the noninjured state. But for injuries that decreasethe marginal utility of wealth (like when the victim is comatose), negativeinsurance is efficient, as the individual would be better off by transferring moneyfrom the injured state (unconscious) to the noninjured state (healthy andconscious), where more utility can be derived from each dollar ( Cook andGraham, 1977 ). Because consumers (potential victims) do not prefer to pay forinsurance against these kind of injuries, fully compensatory tort awards fornonmonetary injuries may be inefficient. Survey evidence is consistent this view( Calfee and Winston, 1993 ).

Many point to pain-and-suffering damages as a primary source ofinefficiency in the current system (e.g., Danzon, 1984 ; Calfee and Rubin, 1992 ; A.Schwartz, 1988 ). One proposed remedy is to eliminate tort damages fornonmonetary injuries (thereby eliminating the insurance inefficiency) whilerequiring that firms pay a fine to the state equal to the amount needed forefficient deterrence ( Shavell, 1987 ; Polinsky and Che, 1991 ). Eliminatingpain-and-suffering damages within the current system is unlikely to be efficient,however. The absence of widespread first-party insurance for these injuries doesnot necessarily indicate a lack of consumer demand, but could stem fromsupply-side problems related to the cost of moral hazard and adverse selection( Croley and Hanson, 1995 ). Moreover, the analysis which shows thatpain-and-suffering damages are inefficient unrealistically assumes that there isno deterrence value to the tort award; that consumers are optimally insuredagainst all other tortiously caused injuries; and that sellers are forced tointernalize the cost of all tortiously caused nonmonetary injuries. Revising theanalysis to account for more realistic assumptions shows that pain-and-sufferingtort damages in the current system could be efficient if courts were to instructjuries on how to calculate the appropriate award, which is based on consumerwillingness to pay to eliminate the risk ( Geistfeld, 1995b ).

21. Punitive Damages

Punitive damages have become a focal point in the debate over products liabilityreform in the United States, even though they are awarded infrequently ( Danielsand Martin, 1990 ; Rustad, 1992 ). Punitive damages can be efficient when victimswith valid legal claims do not sue, enabling sellers to escape liability in somecases ( Cooter, 1989a ). For example, if only 50 percent of all victims sue,compensatory damages must be doubled if the seller is to internalize the full costof injury. Punittive damages can also be used to deter sellers from sendingmisleading signals of product quality ( Daughety and Reinganum, 1997 ). Theoptimal adjustment to the compensatory damages award can be positive ornegative, however, because it depends upon a variety of other factors such asthe possibility of court error ( Polinsky and Shavell, 1989 ), the impact of litigationcosts on social welfare ( Polinsky and Rubinfeld, 1988 ), insolvency ( Knoll, 1997 ),and risk aversion ( Craswell, 1996 ).

It is doubtful that punitive damage awards are set on the basis of theseeconomic considerations, as juries typically are given little or no instruction onhow to compute the appropriate award. It is also doubtful that punitive damagesare awarded when it would be efficient to do so ( American Law Institute, 1991b,pp. 243-48 ). The legal standards governing the availability of punitive damageawards have been substantially, if not wholly, influenced by intentional torts(for which punitive damages were available under the early common law). Thesestandards create problems in the products liability context, where the criticalissue is not whether the manufacturer's actions were deliberate (they usuallyare), but whether the manufacturer knew it was selling a defective product. Byfocusing on deliberate conduct rather than on the seller's awareness of defect,the inquiry can easily lead to unwarranted punitive damages. If hindsight showsthat the manufacturer erred in concluding that the cost of a safety improvementoutweighed the benefit of risk reduction, then even if the manufacturer thoughtthe product was optimally safe, the legal standard for punitive damages may besatisfied. In choosing not to decrease risk out of cost concerns, themanufacturer engaged in "wanton" or "wilful" conduct that "consciouslydisregards the rights or safety of others." Any type of cost-benefit balancinginvolving the risk of injury therefore might be subjected to punitive damages, somanufacturers in design-defect cases often are unwilling to admit that they madesafety decisions on the basis of cost considerations ( G. Schwartz, 1991a ). Thisis a perverse result given that the legal test for design defects relies oncost-benefit balancing, and indicates that the punitive damages standardundermines the accuracy of legal determinations of design defect.

22. The Enforceability of Contractual Waivers of Seller Liability

Contract terms that disclaim a seller's liability for product defects ordinarily arenot enforceable unless the disclaimer pertains to cases in which a productdamages itself, causing economic loss such as lost profits, but does not causepersonal injury or damage to any other property. Contracting probably is a moreefficient way to allocate these damages ("economic losses"), because buyershave better control over and information regarding the magnitude of loss ( Jones,1990 ). Moreover, allowing sellers to disclaim liability for economic loss isunlikely to have significant deterrence effects, as the seller remains liable for anyphysical injury or property damage caused by the product defect. In somejurisdictions, sellers can also disclaim liability for physical loss if the buyer is acommercial party. These buyers tend to be sophisticated and knowledgeableabout the consequences of risk allocation, so contracting in these situations ismore likely to be efficient.

A number of scholars argue that it would be efficient if courts were toenforce a greater variety of contractual limitations of seller liability (e.g., Epstein,1989 ; Rubin, 1993 ). But unless the contracting process is structured to giveconsumers risk-related information, these proposals raise the samesafety-insurance tradeoff presented by any proposal to limit a seller's tortliability. One way contracting can increase risk-related information is if theenforceability of a disclaimer is conditioned on the requirement that the sellerprovides a separate price quotation of its liability costs under a rule of strictliability. Such a price tells consumers something about the product's safety andenables them to compare safety across brands ( Geistfeld, 1988 ; A. Schwartz,1988 ). Nevertheless, imperfectly informed consumers are still likely to disclaimseller liability when it would be inefficient to do so ( Geistfeld, 1994 ). Givingconsumers the opportunity to sell their "unmatured tort claims" to third partiesalso has interesting possibilities ( Cooter, 1989b ; Choharis, 1995 ), although thisreform may also lead to inefficient reductions in seller liability ( A. Schwartz,1989a ). But even though these proposals do not resolve the regulatory problem,measures like them that enhance information and facilitate contracting are apromising approach to efficient reform ( A. Schwartz, 1995 ).

23. Directions for Future Research

It is commonplace to say that more empirical research is needed, butdevelopments over the past decade provide a good opportunity for studying therelationship between tort liability and injury rates. Prior empirical studies of thisissue suffer from the common problem of being unable to adequately definewhen seller liability expanded by an amount significant enough to be capturedby statistical analysis. The evolutionary nature of common-law change makessuch a definition elusive. The change in liability standards has been more abruptsince the mid-1980s, however, as the insurance crisis has spawned numerousreforms that limit seller liability. Because these widespread reforms occurred overa short period of time and were the result of legislative enactment, the timing ofthe change in liability can be readily defined, which should make it easier touncover any statistical relationship between seller liability and injury rates.

Regarding issues amenable to theoretical analysis, it would be useful todiscover whether prices signal product safety under market conditions that aremore realistic than those previously studied. A pressing issue concerns therelationship between liability and innovation, which relative to its importance isthe most understudied aspect of products liability. There are also a number ofproducts liability doctrines that have not been subjected to rigorous economicanalysis. For example, an issue of present concern relates to the conditionsunder which suppliers of raw materials should be liable for injuries caused by thefinal product. Those who grapple with the issue have done so largely withoutthe benefit of economic analysis, making it difficult to understand howlawmakers could place much reliance on efficiency considerations in decidinghow to resolve the issue. Absent more widespread economic analysis of therange of doctrines that comprise products liability, it seems likely that efficiencyconsiderations will continue to exert an uneven influence on the developmentof this area of the law.

Bibliography on Product Liability (5140)

Abraham, Kenneth S. (1987), 'Making Sense of the Liability Insurance Crisis', 48 Ohio StateLaw Journal , 399-411.

Abraham, Kenneth S. (1988) , 'The Causes of the Insurance Crisis', in Olson, Walter (ed.), New Directions in Liability Law , New York, The Academy of Political Science, 54-66.

Abraham, Kenneth S. (1990) , 'Products Liability Law and Insurance Profitability', 19 Journal of Legal Studies , 837-844.

Adams, Michael (1987), 'Ökonomische Analyse der Produkthaftung (Economic Analysis ofProduct Liability)', 87 Der Betriebs-Berater .

Adams, Michael (1987), 'Produkthaftung - Wohltat oder Plage - Eine ökomische Analyse(Product Liability - Benefit or Plague - An Economic Analysis)', 3124 DerBetriebs-Berater .

Adams, Michael (1989) , 'Ökonomische Begründung des AGB-Gesetzes - Verträge beiasymetrischer Information (The Economic Basis of the German Unfair Contract T)',in Der Betriebs-Berater , 781-788.

Adams, Michael (1995) , 'The Conflict of Jurisdictors - An Economic Analysis of PretrialDiscovery, Fact Gathering and Cost Allocation Rules in the United States and Germany', 3(1) European Review of Private Law , 53-94.

Akerlof, George A. (1970) , 'The Markets for "Lemons": Qualitative Uncertainty and theMarket Mechanism', 84 Quarterly Journal of Economics , 488-500. Partially reprintedin Goldberg, Victor P. (ed.), Readings in the Economics of Contract Law , New York,Cambridge University Press.

Allen, Franklin (1984) , 'Reputation and Product Quality', 15 Rand Journal of Economics ,311-327.

American Law Institute (1997) , Restatement of the Law (Third) of Torts: Products Liability ,Philadelphia, The American Law Institute.

American Law Institute (1991a) , Reporters' Study, 'Enterprise Liability for Personal Injury:Volume I, The Institutional Framework ', The American Law Institute, Philadelphia.

American Law Institute (1991b) , Reporters' Study, 'Enterprise Liability for Personal Injury:Volume II, Approaches to Legal and Institutional Change ', Philadelphia, The AmericanLaw Institute.

Arlen, Jennifer H. (1992) , 'Book Review: W. Kip Viscusi, Reforming Products Liability', 30 Journal of Economic Literature , 2170 ff.

Arlen, Jennifer H. (1993) , 'Compensation Systems and Efficient Deterrence', 52 MarylandLaw Review , 1093 ff.

Armstrong, Dwight L. (1976) , 'Products Liability, Comparative Negligence and theAllocation of Damages among Multiple Defendants', 50 Southern California LawReview , 73-108.

Arrow, Kenneth J. (1971) , Essays in the Theory of Risk Bearing , Chicago, Markham.

Arrow, Kenneth J. (1982) , 'Risk Perception in Psychology and Economics', 20 EconomicInquiry , 1-9.

Ashford, Nicholas A. and Stone, Robert F. (1991) , The Liability Maze: The Impact ofLiability Law on Safety and Innovation , Washington (DC), The Brookings Institution.

Attanasio, John B. (1988) , 'The Principle of Aggregate Autonomy and the CalabresianApproach to Products Liability', 74 Virginia Law Review , 677-750.

Ausness, Richard C. (1997) , 'An Insurance-Based Compensation System for Product-RelatedInjuries', 58 University of Pittsburg Law Review , 669-717.

Bagwell, Kyle and Riordan, Michael H. (1991) , 'High and Declining Prices Signal ProductQuality', 81 American Economic Review , 224-239.

Barker, Drucilla K. (1991) , 'An Empirical Analysis of the Effects of Product Liability Lawson Underwriting Risk', 51 Journal of Risk and Insurance , 63 ff.

Beales, J. Howard, III, Craswell, Richard and Salop, Steven C. (1981), 'The EfficientRegulation of Consumer Information', 24 Journal of Law and Economics , 491-539.

Beales, J. Howard, III, Craswell, Richard and Salop, Steven C. (1981), 'Information Remediesfor Consumer Protection', 71 American Economic Review. Papers and Proceedings ,410-413.

Becker, Gary S. and Ehrlich, Isaac (1972) , 'Market Insurance, Self-insurance, andSelf-protection', 80 Journal of Political Economy , 623-648.

Bell, John W. (1977) , 'Averting a Crisis in Product Liability : An Evolutionary Process ofSocioeconomic Justice', 65 Illinois Bar Journal , 640-648.

Bell, Peter A. (1990) , 'Analyzing Tort Law: The Flawed Promise of Neocontract', 74 Minnesota Law Review , 1177-1249.

Ben-Shahar, Omri (1997) , 'Should Products Liability Be Based On Hindsight?', WorkingPaper No. 14-97, The Foerder Institute for Economic Research, Tel Aviv University.

Benson, Bruce L. (1995) , 'Toxic Torts by Government', in , Independent Policy Report ,Oakland, Independent Institute.

Benson, Bruce L. (1996) , 'Uncertainty, the Race for Property Rights, and Rent Dissipationdue to Judicial Changes in Product Liability Tort Law', Cultural Dynamics .

Blight, Catherine (1989) , '"What if that Snail had been in a Bottle of Milk?" or ProductLiability in the UK - the Special Case of Agricultural Products', in Faure, Michael andVan Den Bergh, Roger (eds.), Essays in Law and Economics: Corporations, AccidentPrevention and Compensation for Losses , Antwerpen, Maklu, 215-232.

Bonney, Paul R. (1985) , 'Manufacturers' Strict Liability for Handgun Injuries: An EconomicAnalysis', 73 Georgetown Law Journal , 1437-1463.

Boyd, James and Ingberman, Daniel E. (1994), 'Extending Liability: Should the Sins of theProducer Be Visited Upon Others?', Business, Law, and Economics Center WorkingPaper BLE .

Boyd, James and Ingberman, Daniel E. (1994), 'Noncompensatory Damages and PotentialInsolvency', 23 Journal of Legal Studies , 895-910.

Boyd, James and Ingberman, Daniel E. (1997a) , 'Should 'Relative Safety' Be a Test ofProduct Liability?', 26 Journal of Legal Studies , 433-473.

Boyd, James and Ingberman, Daniel E. (1997b) , 'The Search for Deep Pockets: Is "ExtendedLiability" Expensive Liability?', 13 Journal of Law, Economics & Organization ,232-258.

Brown, John Prather (1974) , 'Product Liability: The Case of an Asset with Random Life', 64 American Economic Review , 149-161.

Brüggemeier, Gert (1989) , 'Die Gefährdungshaftung der Produzenten nach der EG-Richtlinie- ein Fortschritt der Rechtsentwicklung? (Strict Liability of Producers according to theEC-)', in Ott, Claus and Schäfer, Hans-Bernd (eds.), Allokationseffizienz in derRechtsordnung , Berlin, Springer-Verlag, 228-247.

Bryant, W. Keith and Gerner, Jennifer L. (1978) , 'The Price of a Warranty: The Case forRefrigerators', 12 Journal of Consumer Affairs , 30-47.

Buchanan, James M. (1970) , 'In Defense of Caveat Emptor', 38 University of Chicago LawReview , 64-73.

Burrows, Paul (1992) , 'Consumer Safety under Products Liability and Duty to Disclose', 12 International Review of Law and Economics , 457-478.

Burrows, Paul (1994) , 'Products Liability and the Control of Product Risk in the EuropeanCommunity', 10 Oxford Review of Economic Policy , 68-83.

Butters, Gerard R. (1986) , 'The Impact of Product Recalls on the Wealth of Sellers:Comments', in Ippolito, Pauline M. and Scheffman, David T. (eds.), EmpiricalApproaches to Consumer Protection Economics , Washington (DC), Federal TradeCommission, 411-413.

Calabresi, Guido (1984) , 'First Party, Third Party, and Product Liability Systems: CanEconomic Analysis of Law Tell Us Anything About Them?', 69 Iowa Law Review ,833-851.

Calabresi, Guido and Bass, Kenneth C. II (1970) , 'Right Approach, Wrong Implications: ACritique of McKean on Products Liability', 38 University of Chicago Law Review , 74-91.

Calabresi, Guido and Hirschoff, Jon T. (1972) , 'Toward a Test for Strict Liability in Tort', 81 Yale Law Journal , 1054-1085. Reprinted in Rabin, Robert L. (ed.), Perspectives onTort Law , Boston, Little Brown, 1983, 192-212.

Calfee, John E. and Rubin, Paul H. (1992) , 'Some Implications of Damage Payments forNonpecuniary Losses', 21 Journal of Legal Studies , 371-411.

Calfee, John E. and Winston, Clifford (1988) , 'Economic Aspects of Liability Rules andLiability Insurance', in Litan, Robert E. and Winston, Clifford (eds.), Liability:Perspectives and Policy , Washington (DC), The Brookings Institution, 16-41.

Calfee, John E. and Winston, Clifford (1993) , 'The Consumer Welfare Effects of Liabilityfor Pain and Suffering: An Exploratory Analysis', 1 Brookings Papers on EconomicActivity: Microeconomics , 133-196.

Camerer, Colin F. and Kunreuther, Howard (1993) , 'Making Decisions About Liability andInsurance: Editors' Comments', 7 Journal of Risk and Uncertainty , 5-15.

Cass, Ronald A. and Gillette, Clayton P. (1991) , 'The Government Contractor Defense:Contractual Allocation of Public Risk', 77 Virginia Law Review , 257-336.

Caves, R.E. and Greene, D.P. (1996) , 'Brands' Quality Levels, Prices, and AdvertisingOutlays: Empirical Evidence on Signals and Information Costs', 14 International Journalof Industrial Organization , 29-52.

Centner, Terence J. (1992) , 'The New 'Tractor Lemon Laws': An Attempt to SqueezeManufacturers Draws Sour Benefits', 14 Journal of Products Liability , 121-137.

Centner, Terence J. (1993) , 'Separating Lemons: Automobiles and Tractors under the 'MotorVehicle Warranty Rights Act' and the 'Farm Tractor Warranty Act', 30 Georgia StateBar Journal , 36-43.

Centner, Terence J. (1995) , 'Lemon Laws: Sour Options for Consumers of Farm Tractors', 18 Hamline Law Review , 347-362.

Centner, Terence J. and Wetzstein, Michael E. (1995) , 'Obligations and Penalties underLemon Laws: Automobiles Versus Tractors', 20 Journal of Agricultural and ResourceEconomics , 135-145.

Chan, Yuk-Shee and Leland, Hayne E. (1982) , 'Prices and Qualities in Markets with CostlyInformation', 49 Review of Economic Studies , 499-516.

Chapman, Kenneth and Meurer, Michael J. (1989) , 'Efficient Remedies for Breach ofWarranty', 52(1) Law and Contemporary Problems , 107-131.

Choharis, Peter Charles (1995) , 'A Comprehensive Market Strategy for Tort Reform', 12 Yale Journal on Regulation , 435-525.

Coleman, Jules L. (1992) , Risks and Wrongs , Cambridge, Cambridge University Press.

Cook, Philip J. and Graham, Daniel A. (1977) , 'The Demand for Insurance and Protection:The Case of Irreplaceable Commodities', 91 Quarterly Journal of Economics , 143-156.

Cooper, Russell and Ross, Thomas W. (1984) , 'Prices, Product Qualities and AsymmetricInformation: the Competitive Case', 51 Review of Economic Studies , 197-208.

Cooper, Russell and Ross, Thomas W. (1985a) , 'Product Warranties and Double MoralHazard', 16 Rand Journal of Economics , 103-113.

Cooper, Russell and Ross, Thomas W. (1985b) , 'Monopoly Provision of Product Qualitywith Uninformed Buyers', 3 International Journal of Industrial Organization , 433-449.

Cooter, Robert D. (1985) , 'Defective Warnings, Remote Causes, and Bankruptcy: Commenton Schwartz', 14 Journal of Legal Studies , 737-750.

Cooter, Robert D. (1989a) , 'Punitive Damages for Deterrence: When and How Much?', 40 Alabama Law Review , 1143 ff.

Cooter, Robert D. (1989b) , 'Towards A Market in Unmatured Tort Claims', 75 Universityof Virginia Law Review , 383-411.

Cooter, Robert D. (1991) , 'Economic Theories of Legal Liability (Italian translation:"Leteorie economiche della risponsibilità efficenza produttiva: alcuni contributi su notiargomenti" 379-409)', 5 Journal of Economic Perspectives , 11-30. Reprinted inMedema, Steven G. (Ed.), The Legacy of Ronald Coase in Economic Analysis, EdwardElgar Publishing, Glocestershire, forthcoming. Italian translation: "Le TeorieEconomiche Della Risponsabilit Legale"), published in Gianandrea Goisis.

Cooter, Robert D. and Sugarman, Stephen D. (1988) , 'A Regulated Market in UnmaturedTort Claims: Tort Reform by Contract', in Olson, Walter (ed.), New Directions inLiability Law , New York, The Academy of Political Science, 174-185.

Cooter, Robert D. and Ulen, Thomas S. (1988) , Law and Economics , Glenview (IL), Scott,Foresman and Company.

Cosentino, Fabrizio (1989) , 'Responsabilità da prodotto difettoso: appunti di analisieconomica del diritto (nota a U.S. Supreme Court California, 31 marzo 1988, Brown c.Abbott Laboratories) (Liability for Defective Products: An Economic Analysis of Law)', 4 Foro Italiano , 137-143.

Courville, Léon and Hausman, Warren H. (1979) , 'Warranty Scope and Reliability underImperfect Information and Alternative Market Structures', 52 Journal of Business ,361-378.

Cousy, Herman (1976) , 'Produktenaansprakelijkheid. Proeve van een Juridisch-EconomischeAnalyse (Product Liability. Attempt of a Legal-Economic Ana)', Tijdschrift voorPrivaatrecht , 995-1035.

Craig, Andrew (1991) , 'Product Liability and Safety in General Aviation', in Huber, Peter W.and Litan, Robert E. (eds.), The Liability Maze: The Impact of Liability Law on Safetyand Innovation , Washington (DC), The Brookings Institution, 456-477.

Craswell, Richard (1991) , 'Passing on the Costs of Legal Rules: Efficiency and Distributionin Buyer-Seller Relationships', 43 Stanford Law Review , 361-398.

Craswell, Richard (1996) , 'Damage Multipliers in Market Relationships', 25 Journal of LegalStudies , 463-492.

Croley, Steven P. and Hanson, Jon D. (1991) , 'What Liability Crisis? An AlternativeExplanation for Recent Events in Products Liability', 8 Yale Journal on Regulation ,1-111.

Croley, Steven P. and Hanson, Jon D. (1993) , 'Rescuing the Revolution: The Revived Casefor Enterprise Liability', 91 Michigan Law Review , 683-797.

Croley, Steven P. and Hanson, Jon D. (1995) , 'The Nonpecuniary Costs of Accidents: Painand-Suffering Damages in Tort Law', 108 Harvard Law Review , 1785-1917.

Curran, Christopher (1992) , 'The Spread of Comparative Negligence in the United States', 12 International Review of Law and Economics , 317-332.

Daniels, Stephen and Martin, Joanne (1990) , 'Myth and Reality in Punitive Damages', 75 Minnesota Law Review , 1-64.

Danzon, Patricia M. (1984) , 'Tort Reform and the Role of Governement in PrivateInsurance Markets', 13 Journal of Legal Studies , 517-549.

Danzon, Patricia M. (1985) , 'Comments on Landes and Posner: A Positive EconomicAnalysis of Products Liability', 14 Journal of Legal Studies , 569-583.

Danzon, Patricia M. (1988) , 'The Political Economy of Workers' Compensation: Lessonsfor Product Liability', 78 American Economic Review: Papers and Proceedings ,305-310.

Daughety, Andrew F. and Reinganum, Jennifer F. (1995) , 'Product Safety: Liability, R&D,and Signaling', 85 American Economic Review , 1187-1206.

Daughety, Andrew F. and Reinganum, Jennifer F. (1997) , 'Everybody Out of the Pool:Products Liability, Punitive Damages, and Competition', 13 Journal of Law, Economics,and Organization , 410-432.

De Alessi, Louis (1994) , 'Reputation and the Efficiency of Legal Rules', 14 Cato Journal ,11-21.

De Alessi, Louis and Staaf, Robert J. (1987) , 'Liability, Control and the Organization ofEconomic Activity', 7 International Review of Law and Economics , 5-20.

Dewees, Donald N., Duff, David and Trebilcock, Michael J. (1996) , Exploring the Domainof Accident Law: Taking the Facts Seriously , New York, Oxford University Press.

Dorfman, Robert (1970) , 'The Economics of Product Liability: A Reaction to McKean', 38 University of Chicago Law Review , 92-102.

Dungworth, Terence (1988) , Product Liability and the Business Sector: Litigation Trendsin Federal Courts , Santa Monica (CA), Institute for Civil Justice, Rand Corporation.

Eads, George C. and Reuter, Peter (1983) , Designing Safer Products: Corporate Reponsesto Product Liability Law and Regulation , Santa Monica (CA), Institute for Civil Justice,Rand Corporation.

Egon Zehnder International (1987) , 'The Litigious Society: Is It Hampering Creativity,Innovation, and Our Ability to Compete?', 2(3) Corporate Issues Monitor , 1 ff.

Eisman, Deborah E. (1983) , 'Product Liability: Who Should Bear the Burden?', 27 AmericanEconomist , 54-57.

Emons, Winand (1988) , 'Warranties, Moral Hazard, and the Lemons Problem', 46 Journalof Economic Theory , 16-33.

Emons, Winand (1990) , 'Some Recent Developments in the Economic Analysis of LiabilityLaw: An Introduction', 146 Journal of Institutional and Theoretical Economics ,237-248.

Epple, Dennis and Raviv, Artur (1978) , 'Product Safety: Liability Rules, Market Structure,and Imperfect Information', 68 American Economic Review , 80-95.

Epstein, Richard A. (1980) , Modern Product Liability Law , Westport, Quorum Books, 211p.

Epstein, Richard A. (1984) , 'The Legal and Insurance Dynamics of Mass Tort Litigation', 13 Journal of Legal Studies , 475-506.

Epstein, Richard A. (1985) , 'Products Liability as an Insurance Market', 14 Journal of LegalStudies , 645-669.

Epstein, Richard A. (1986) , 'The Temporal Dimension in Tort Law', 53 University ofChicago Law Review , 1175-1218.

Epstein, Richard A. (1987) , 'The Risks of Risk/Utility', 48 Ohio State Law Journal ,469-477.

Epstein, Richard A. (1988) , 'The Political Economy of Product Liability Reform', 78 American Economic Review: Papers and Proceedings , 311-315.

Epstein, Richard A. (1989) , 'The Unintended Revolution in Product Liability Law', 10 Cardozo Law Review , 2193-2222.

Faure, Michael G. (1986) , 'Commentaar bij de paper van Dr.Jan C. Bongaerts: Coase enProduktenaansprakelijkheid (Comment on the paper of Dr. Jan C. Bongaerts: Coas)', inVan Den Bergh, Roger (ed.), Verslagboek Eerste Werkvergadering Recht en Economie ,Antwerpen, Handelshogeschool, 31-37.

Faure, Michael G. and Van Buggenhout, Willy (1987) , 'Produktenaan sprakelijkheid. DeEuropese Richtlijn: Harmonisatie en Consumentenbescherming? (The EuropeanDirective Concerning Product Liability, Harmonization and Consumer Protection?)', 51 Rechtskundig Weekblad , 33-88.

Finsinger, Jörg (1991) , 'The Choice of Risky Technologies and Liability', 11 InternationalReview of Law and Economics , 11-22.

Finsinger, Jörg, Hoehn, T. and Pototschnig, A. (1991) , 'The Enforcement of ProductLiability Rules: A Two-Country Analysis of Court Cases', 11 International Review ofLaw and Economics , 133-148.

Finsinger, Jörg and Simon, Jürgen (1988) , Eine ökonomische Bewertung derEG-Produkthaftu (An Economic Appreciation of the EC- Product Li), Lüneburg,Arbeitsbericht Hochschule Lüneburg, Fachbereich Wirtschafts- und Sozialwissenschaften,44 ff.

Finsinger, Jörg and Simon, Jürgen (1989) , 'An Economic Assessment of the EC ProductLiability Directive and the Product Liability Law of the Federal Republic of Germany',in Faure, Michael and Van Den Bergh, Roger (eds.), Essays in Law and Economics:Corporations, Accident Prevention and Compensation for Losses , Antwerpen, Maklu,185-214.

Fischoff, Baruch and Merz, Jon F. (1994) , 'The Inconvenient Public: Behavioral ResearchApproaches to Reducing Product Liability Risks', in Hunziker, Janet R. and Jones,Trevor O. (eds.), Product Liability and Innovation: Managing Risk in an UncertainEnvironment .

Gal-Or, Esther (1989) , 'Warranties as a Signal of Quality', 22 Canadian Journal ofEconomics , 50-61.

Galitsky, S. (1979) , 'Manufacturers' Liability: An Examination of the Policy and Social Costof a New Regime', 3 University of New South Wales Law Journal , 145-174.

Geistfeld, Mark (1988) , 'Note: Imperfect Information, the Pricing Mechanism, and ProductsLiability', 88 Columbia Law Review , 1057-1068.

Geistfeld, Mark (1992) , 'Implementing Enterprise Liability: A Comment on Henderson andTwerski', 67 New York University Law Review , 1157 ff.

Geistfeld, Mark (1994) , 'The Political Economy of Neocontractual Proposals for ProductsLiability Reform', 72 Texas Law Review , 803-847.

Geistfeld, Mark (1995a) , 'Manufacturer Moral Hazard and the Tort Contract Issue inProducts Liability', 15 International Review of Law and Economics , 241-257.

Geistfeld, Mark (1995b) , 'Placing a Price on Pain and Suffering: A Method for Helping JuriesDetermine Tort Damages for Nonmonetary Injuries', 83 California Law Review ,773-852.

Geistfeld, Mark (1997) , 'Inadequate Product Warnings and Causation', 30 University ofMichigan Journal of Law Reform , 309-351.

Gerner, Jennifer L. and Bryant, W. Keith (1981) , 'Appliance Warranties as a MarketSignal?', 15 Journal of Consumer Affairs , 75-86.

Gilmore, Grant (1970) , 'Products Liability: A Commentary', 38 University of Chicago LawReview , 103-116.

Goldberg, Victor P. (1974) , 'The Economics of Product Safety and Imperfect Information', 5 Bell Journal of Economics and Management Science , 683-688.

Goldberg, Victor P. (1988) , 'Accountable Accountants: Is Third- Party Liability Necessary?', 17 Journal of Legal Studies , 295-312.

Graham, Daniel A. and Price, Ellen R. (1984) , 'Contingent Damages for Products Liability', 13 Journal of Legal Studies , 441-468.

Graham, John D. (1991) , 'Product Liability and Motor Vehicle Safety', in Huber, Peter W.and Litan, Robert E. (eds.), The Liability Maze: The Impact of Liability Law on Safetyand Innovation , Washington (DC), The Brookings Institution, 120-190.

Grether, David M., Schwartz, Alan and Wilde, Louis L. (1986) , 'The Irrelevance ofInformation Overload: An Analysis of Search and Disclosure', 59 Southern CaliforniaLaw Review , 277-303.

Grether, David M., Schwartz, Alan and Wilde, Louis L. (1988) , 'Uncertainty and ShoppingBehavior: An Experimental Analysis', 55 Review of Economic Studies , 323-342.

Grether, David M., Schwartz, Alan and Wilde, Louis L. (1992) , 'Price, Quality and Timingof Moves in Markets with Incomplete Information: An Experimental Analysis', 102 Economic Journal , 754-771.

Grossman, Sanford J. (1981) , 'The Informational Role of Warranties and Private Disclosureabout Product Quality', 24 Journal of Law and Economics , 461-483.

Gupta, Pola and Ratchford, Brian T. (1992) , 'Estimating the Efficiency of ConsumerChoices of New Automobiles', 13 Journal of Economic Psychology , 375-397.

Hamada, Koichi (1976) , 'Liability Rules and Income Distribution in Products Liability', 66 American Economic Review , 228-234.

Hamada, Koichi (1995) , 'Product Liability Rules: A Consideration of Law and Economicsin Japan', 46 Japanese Economic Review , 2-22.

Hamada, Koichi, Ishida, Hidetoh and Murakami, Masahiro (1986) , 'The Evolution andEconomic Consequences of Product Liability Rules in Japan', in Saxonhouse, Gary R. andYamamura, Kozo (eds.), Law and Trade Issues of the Japanese Economy: Amer , Seattle,University of Washington Press, 83-106.

Hammitt, James K., Carroll, Stephen J. and Relles, Daniel A. (1985) , 'Tort Standards andJury Decisions', 14 Journal of Legal Studies , 751-762.

Hanson, Jon D. and Logue, Kyle D. (1990) , 'The First-Party Insurance Externality: AnEconomic Justification for Enterprise Liability', 76 Cornell Law Review , 129-196.

Harrington, Scott E. (1988) , 'Prices and Profits in the Liability Insurance Market', in Litan,Robert E. and Winston, Clifford (eds.), Liability: Perspectives and Policy , Washington(DC), The Brookings Institution, 42-100.

Harrington, Scott E. (1991) , 'Liability Insurance: Volatility in Prices and in the Availabilityof Coverage', in Schuck, Peter H. (ed.), Tort Law and the Public Interest: Competition,Innovation, and Consumer Welfare , New York, W.W. Norton & Co., 47-79.

Hay, Bruce L. (1992) , 'Conflicts of Law and State Competition in the Product LiabilitySystem', 80 Georgetown Law Journal , 617-652.

Heal, Geoffrey M. (1977) , 'Guarantees and Risk Sharing', 44 Review of Economic Studies ,549-560.

Heinkel, R. (1981) , 'Uncertain Product Quality: the Market for Lemons with an ImperfectTesting Technology', 12 Bell Journal of Economics , 625-636.

Henderson, James A. (1973) , 'Judicial Review of Manufacturer's Conscious Design Choices:The Limits of Adjudication', 73 Columbia Law Review , 1531 ff.

Henderson, James A. (1980) , 'Extending the Boundaries of Strict Products Liability:Implications of the Theory of Second Best', 128 University of Pennsylvania Law Review ,1036-1093.

Henderson, James A. (1983) , 'Product Liability and the Passage of Time: The Imprisonmentof Corporate Rationality', 58 New York University Law Review , 765-795.

Henderson, James A. (1991) , 'Judicial Reliance on Public Policy: An Empirical Analysis', 59 George Washington Law Review , 1570 ff.

Henderson, James A. and Twerski, Aaron D. (1990) , 'Doctrinal Collapse in ProductsLiability: The Empty Shell of Failure to Warn', 65 New York University Law Review ,265-327.

Henderson, James A. and Twerski, Aaron D. (1991) , 'Closing the American ProductsLiability Frontier: The Rejection of Liability Without Fault', 66 New York UniversityLaw Review , 1263-1331.

Hensler, Deborah R. , Marquis, M. Susan, Abrahamse, Allan F., Berry, Sandra H., Ebener,Patricia A., Lewis, Elizabeth G., Lind, E. Allan, Maccoun, Robert J., Manning, WillardG., Rogowski, Jeannette A. and Vaiana, Mary E. (1991), Compensation for AccidentalInjuries in the United States , Santa Monica (CA), Institute for Civil Justice, RandCorporation.

Higgins, Richard S. (1978) , 'Producers' Liability and Product Related Accidents', 7 Journalof Legal Studies , 299-321.

Higgins, Richard S. (1981) , 'Products Liability Insurance, Moral Hazard, and ContributoryNegligence', 10 Journal of Legal Studies , 111-130.

Hoffer, George E., Pruitt, Stephen W. and Reilly, Robert J. (1988) , 'The Impact of ProductRecalls on the Wealth of Sellers: A Reexamination', 96 Journal of Political Economy ,663-670.

Holmström, Bengt (1979) , 'Moral Hazard and Observability', 10 Bell Journal of Economics ,74-91.

Horvitz, Sigmund A. and Stern, Louis H. (1987) , 'Liability Rules and the Selection of aSocially Optimal Production Technology', 7 International Review of Law andEconomics , 121-126.

Huber, Peter W. (1986) , 'Safety and the Second Best: The Hazards of Public RiskManagement in the Courts', 85 Columbia Law Review , 277-337.

Huber, Peter W. (1988) , Liability: The Legal Revolution and Its Consequences , New York,Basic Books, 260 p.

Huber, Peter W. (1992) , 'Liability and Insurance Problems in the Commercialization of NewProducts: A Perspective from the United States and England', in Rosenbergh, Nathan,Landau, Ralph, and Mowery, David C. (eds.), Technology and the Wealth of Nations ,Stanford, Stanford University Press, 207-221.

Huber, Peter W. and Litan, Robert E. (1991) , 'Overview', in Huber, Peter W. and Litan,Robert E. (eds.), The Liability Maze: The Impact of Liability Law on Safety andInnovation , Washington (DC), The Brookings Institution, 1-27.

Hylton, Keith (1990) , 'Costly Litigation and Legal Error Under Negligence', 6 Journal ofLaw, Economics and Organization , 433 ff.

Inter-Agency Task Force On Products Liability (1977) , Washington, D.C., United StatesDepartment of Commerce , Washington (DC), United States Department of Commerce.

Jarrell, Gregg A. and Peltzman, Sam (1985) , 'The Impact of Product Recalls on the Wealthof Sellers', 93 Journal of Political Economy , 512-536. Reprinted in Ippolito, PaulineM. and Scheffman, David T. (eds.), Empirical Approaches to Consumer ProtectionEconomics , Washington (DC), Federal Trade Commission, 377-409.

Johnson, Rollin B. (1991) , 'The Impact of Liability on Innovation in the ChemicalIndustry', in Huber, Peter W. and Litan, Robert E. (eds.), The Liability Maze: The Impactof Liability Law on Safety and Innovation , Washington (DC), The BrookingsInstitution, 428-455.

Jones, P. and Hudson, J. (1996) , 'Signalling Product Quality: When Is Price Relevant?', 30 Journal of Economic Behavior and Organization , 257-66.

Jones, William K. (1990) , 'Product Defects Causing Commercial Loss: The Ascendency ofContract Over Tort', 44 University of Miami Law Review , 731-806.

Kahneman, Daniel, Slovic, Paul and Tversky, Amos (1982) , Judgment and Uncertainty:Heuristics and Biases , Cambridge, Cambridge University Press.

Kahneman, Daniel and Tversky, Amos (1979) , 'Prospect Theory: An Analysis of DecisionUnder Risk', 47 Econometrica , 160-173.

Kambhu, John (1982) , 'Optimal Product Quality under Asymmetric Information and MoralHazard.', Bell Journal of Economics , 483-492.

Kaprelian, Mark A. (1985) , 'Privity Revisited: Tort Recovery by a Commercial Buyer fora Defective Product's Self-Inflicted Damage', 84 Michigan Law Review , 517-540.

Keenan, Donald C. and Rubin, Paul H. (1988) , 'Shadow Interest Groups and SafetyRegulation', 8 International Review of Law and Economics , 21-36.

Keeton, W. Page, Owen, David, Montgomery, John E. and Green, Michael D. (1989) , Products Liability and Safety: Cases and Materials (2d ed.) .

Kinkaid, Peter M. and Stuntz, William J. (1983) , 'Note: Enforcing Waivers in ProductsLiability', 69 Virginia Law Review , 1111-1152.

Kirchner, Christian (1989) , 'Kommentar (Comment On Brüggemeier, DieGefährdungshaftung der)', in Ott, Claus and Schäfer, Hans-Bernd (eds.), Allekationseffizienz in der Rechtsordnung , Berlin, Springer-Verlag, 248-253.

Kitch, Edmund W. (1988) , 'American Law and Preventive Vaccination Programs', inPlotkin, Stanley A. and Mortimer, Edward A. (eds.), Vaccines , Philadelphia, W.B.Saunders Company, 612 ff.

Kitch, Edmund W. and Mortimer, Edward A., Jr. (1994) , 'American Law, PreventiveVaccine Programs, and the National Vaccine Injury Compensation Program', in Plotkin,Stanley A. and Mortimer, Edward A. (eds.), Vaccines , Philadelphia, W.B. SaundersCompany, 93 ff.

Klein, Benjamin and Leffler, Keith B. (1981) , 'The Role of Market Forces in AssuringContractual Performance', 89 Journal of Political Economy , 615-641.

Knoll, Michael S. (1997) , 'Products Liability and Legal Leverage: The Perverse Effect ofStiff Penalties', 45 UCLA Law Review , 99-141.

Kolstad, Charles D., Ulen, Thomas S. and Johnson, Gary V. (1990) , 'Ex Post Liability forHarm vs. Ex Ante Safety Regulation: Substitutes or Complements?', 80 AmericanEconomic Review , 888-901.

Komesar, Neil K. (1990) , 'Injuries and Institutions: Tort Reform, Tort Theory, and Beyond', 65 New York University Law Review , 23-77.

Kunreuther, Howard, Hogarth, Robin M. and Meszaros, Jacqueline (1993) , 'Insurer Ambiguityand Market Failure', 7 Journal of Risk and Uncertainty , 71-88.

Lacey, N.J. (1988) , 'Recent Evidence on the Liability Crisis', 55 Journal of Risk andInsurance , 499-508.

Lamken, Jeffrey A. (1989) , 'Note: Efficient Accident Prevention as a ContinuingObligation: The Duty to Recall Defective Products', 42 Stanford Law Review , 103 ff.

Landes, William M. and Posner, Richard A. (1984) , 'Tort Law as a Regulatory Regime forCatastrophic Personal Injuries', 13 Journal of Legal Studies , 417-434.

Landes, William M. and Posner, Richard A. (1985) , 'A Positive Economic Analysis ofProducts Liability', 14 Journal of Legal Studies , 535-567.

Landes, William M. and Posner, Richard A. (1986) , 'New Light on Punitive Damages', 10(1) Regulation , 33-36.

Landes, William M. and Posner, Richard A. (1987) , The Economic Structure of Tort Law ,Cambridge (MA), Harvard University Press, 330 p.

Larsen, Kim D. (1984) , 'Note: Strict Products Liability and the Risk-Utility Test for DesignDefect: An Economic Analysis', 84 Columbia Law Review , 2045-2067.

Lasagna, Louis (1991) , 'The Chilling Effect of Product Liability on New Drug Development',in Huber, Peter W. and Litan, Robert E. (eds.), The Liability Maze: The Impact ofLiability Law on Safety and Innovation , Washington (DC), The Brookings Institution,334-359.

Latin, Howard (1994) , '"Good" Warnings, Bad Products, and Cognitive Limitations', 41 UCLA Law Review , 1193-1295.

Lawrence, William H. and Minon, John H. (1984) , 'The Effect of Abrogating theHolder-in-Due-Course Doctrine on the Commercialization of Innovative ConsumerProducts', 64 Boston University Law Review , 325-374.

Letsou, Peter V. (1986) , 'A Time-Dependent Model of Products Liability', 53 University ofChicago Law Review , 209-231.

Litan, Robert E. (1991), 'The Safety and Innovation Effects of U.S. Liability Law: TheEvidence', 81 American Economic Review. Papers and Proceedings , 59-64.

Litan, Robert E. (1991), 'The Liability Explosion and American Trade Performance: Mythsand Realities', in Schuck, Peter H. (ed.), Tort Law and the Public Interest: Competition,Innovation, and Consumer Welfare , New York, W.W. Norton & Co., 127-150.

Litan, Robert E. and Winston, Clifford (1988) , 'Policy Options', in Litan, Robert E. andWinston, Clifford (eds.), Liability: Perspectives and Policy , Washington (DC), TheBrookings Institution, 223-241.

Lutz, Nancy A. (1989) , 'Warranties as Signals Under Consumer Moral Hazard', 20 RandJournal of Economics , 239-254.

Lyndon, Mary L. (1995) , 'Tort Law and Technology', 12 Yale Journal on Regulation ,137-176.

MacKay, Murray (1991) , 'Liability, Safety, and Innovation in the Automotive Industry', inHuber, Peter W. and Litan, Robert E. (eds.), The Liability Maze: The Impact of LiabilityLaw on Safety and Innovation , Washington (DC), The Brookings Institution, 1 ff.

Magat, Wesley A. and Viscusi, W. Kip (1992) , Informational Approaches to Regulation ,Cambridge (MA), The MIT Press.

Magat, Wesley A., Viscusi, W. Kip and Huber, Joel (1988) , 'Consumer Processing of HazardWarning Information', 1 Journal of Risk and Uncertainty , 201-232.

Manne, Henry G. (1970) , 'Edited Transcript of AALS-AEA Conference on ProductsLiability', 38 University of Chicago Law Review , 117-141.

Manning, Richard L. (1994) , 'Changing Rules in Tort Law and the Market for ChildhoodVaccines', 37 Journal of Law and Economics , 247-275.

Manning, Richard L. (1996) , 'Is the Insurance Aspect of Producer Liability Valued byConsumers? Liability Changes and Childhood Vaccine Consumption', 13 Journal of Riskand Uncertainty , 37-51.

Mantell, Edmund H. (1984) , 'Allocative and Distributive Efficiency of Products LiabilityLaw in a Monopolistic Market', 7 Journal of Products Liability , 143-152.

Marino, Anthony M. (1988a) , 'Products Liability and Scale Effects in a Long-RunCompetitive Equilibrium', 8 International Review of Law and Economics , 97-107.

Marino, Anthony M. (1988b) , 'Monopoly, Liability and Regulation', 54 Southern EconomicJournal , 913-927.

Marino, Anthony M. (1991) , 'Market Share Liability and Economic Efficiency', 57 Southern Economic Journal , 667-675.

Martin, Robert (1991) , 'General Aviation Manufacturing: An Industry Under Siege', in Huber,Peter W. and Litan, Robert E. (eds.), The Liability Maze: The Impact of Liability Lawon Safety and Innovation , Washington (DC), The Brookings Institution, 478-499.

Matthews, Steven and Moore, John (1987) , 'Monopoly Provision of Quality and Warranties:An Exploration in the Theory of Multidimensional Screening', 55 Econometrica ,441-467.

Matthews, Steven and Postlewaite, Andrew (1985) , 'Quality Testing and Disclosure', 16 Rand Journal of Economics , 328-340.

Maxwell, John W. (1998) , ' Minimum quality standards as a barrier to innovation', 58 Economics Letters , 355-60.

McFadden, Daniel L. and Train, Kenneth E. (1996) , 'Consumers' Evaluation of NewProducts: Learning from Self and Others', 104 Journal of Political Economy , 683-703.

McGuire, E. Patrick (1988) , The Impact of Product Liability , New York, Conference Board.

McKean, Roland N. (1970a) , 'Products Liability: Trends and Implications', 38 University ofChicago Law Review , 3-63.

McKean, Roland N. (1970b) , 'Products Liability: Implications of Some Changing PropertyRights', 84 Quarterly Journal of Economics , 611-626.

Merolla, A. Todd (1998) , 'The Effect of Latent Hazards on Firm Exit in ManufacturingIndustries', 18 International Review of Law and Economics , 13-24.

Milgrom, Paul R. and Roberts, John (1986) , 'Price and Advertising Signals of ProductQuality', 94 Journal of Political Economy , 796-821.

Moore, Michael J. and Viscusi, W. Kip (1990) , Compensation Mechanisms for Job Risks:Wages, Workers' Compensation, and Product Liability , Princeton, Princeton UniversityPress.

Moorhouse, John C. (1992) , 'Joint Torts and Interfirm Contracting', 20 Atlantic EconomicJournal , 10-20.

Nakao, Takeo (1982) , 'Product Quality and Market Structure', 13 Bell Journal ofEconomics , 133-142.

Neely, Richard (1988) , The Product Liability Mess: How Business Can Be Rescued from thePolitics of the State Courts , New York, Free Press/MacMillan, 181 p.

Nelson, Phillip (1970) , 'Information and Consumer Behavior', 78 Journal of PoliticalEconomy , 311-329.

Nelson, Phillip (1974) , 'Advertising as Information', 82 Journal of Political Economy ,729-754.

O'Connell, Jeffrey (1988) , 'Neo-No-Fault: A Fair Exchange Proposal for Tort Reform', inOlson, Walter (ed.), New Directions in Liability Law , New York, The Academy ofPolitical Science, 186-195.

O'Reilly, James T. (1987) , 'The Risks of Assumptions: Impacts of Regulatory Label Warningupon Industrial Products Liability', 37 Catholic University Law Review , 85-117.

Oi, Walter Y. (1973) , 'The Economics of Product Safety', 4 Bell Journal of Economics ,3-28.

Oi, Walter Y. (1974) , 'The Economics of Product Safety: a Rejoinder', 5 Bell Journal ofEconomics , 3-28.

Olson, Walter (1988) , 'Overdeterrence and the Problem of Comparative Risk', in Olson,Walter (ed.), New Directions in Liability Law , New York, The Academy of PoliticalScience, 42-53.

Ordover, Janusz A. (1979) , 'Products Liability in Markets with Heterogeneous Consumers', 8 Journal of Legal Studies , 505-525.

Ordover, Janusz A. and Weiss, Andrew (1981) , 'Information and the Law: Evaluating LegalRestrictions on Competitive Contracts', 71 American Economic Review. Papers andProceedings , 399-404.

Owen, David G. (1985) , 'The Intellectual Development of Modern Products Liability Law:A Comment on Priest's View of the Cathedral's Foundations', 14 Journal of LegalStudies , 529-533.

Page, Joseph A. (1990) , 'Deforming Tort Reform, (book review of Peter Huber, 'Liability:The Legal Revolution and Its Consequences)', 78 Georgetown Law Journal , 649-697.

Palfrey, Thomas R. and Romer, Thomas (1983) , 'Warranties, Performance, and theResolution of Buyer-Seller Disputes', 14 Bell Journal of Economics , 97-117.

Png, Ivan Paak-Liang and Reitman, David (1995) , 'Why Are Some Products Branded andOthers Not?', 38 Journal of Law and Economics , 207-224.

Polinsky, A. Mitchell (1980) , 'Strict Liability Versus Negligence in a Market Setting', 70 American Economic Review , 363-367.

Polinsky, Mitchell A. and Rogerson, William P. (1983) , 'Products Liability, ConsumerMisperceptions, and Market Power', 14 Bell Journal of Economics , 581-589.

Polinsky, A. Mitchell and Rubinfeld, Daniel L. (1988) , 'The Welfare Implications of CostlyLitigation for the Level of Liability', 17 Journal of Legal Studies , 151-164.

Polinsky, A. Mitchell and Shavell, Steven (1989) , 'Legal Error, Litigation, and the Incentiveto Obey the Law', 5 Journal of Law, Economics, and Organization , 99-108.

Polinsky, A. Mitchell and Yeon-Koo Che (1991) , 'Decoupling Liability: Optimal Incentivesfor Care and Litigation', 22 Rand Journal of Economics , 562-570.

Prasad, Ravi (1992) , 'Amendments to the New Australian Products Liability Laws', Officeof Regulation Review .

Priest, George L. (1981) , 'A Theory of the Consumer Product Warranty', 90 Yale LawJournal , 1297-1352. Partially reprinted in Goldberg, Victor P. (ed.), Readings in theEconomics of Contract Law , Cambridge, Cambridge University Press, 1989, 174-184.

Priest, George L. (1982) , 'The Best Evidence of the Effect of Products Liability Law on theAccident Rate: Reply', 91 Yale Law Journal , 1386-1401.

Priest, George L. (1985) , 'The Invention of Enterprise Liability: A Critical History of theIntellectual Foundations of Modern Tort Law', 14 Journal of Legal Studies , 461-527.

Priest, George L. (1987) , 'The Current Insurance Crisis and Modern Tort Law', 96 Yale LawJournal , 1521-1590.

Priest, George L. (1988a) , 'Products Liability Law and the Accident Rate in Liability:Perspectives and Policy', in Litan, Robert E. and Winston, Clifford (eds.), Liability:Perspectives and Policy , Washington (DC), The Brookings Institution, 184-222.

Priest, George L. (1988b) , 'The Disappearance of the Consumer from Modern ProductsLiability Law', in Maynes, E. Scott and ACCI Research Committee (eds.), The Frontierof Research in the Consumer Interest: Proceedings of the International Conference onResearch in the Consumer Interest , Columbia, American Council on Consumer Interests,771-791.

Priest, George L. (1988c) , 'Understanding the Liability Crisis', in Olson, Walter (ed.), NewDirections in Liability Law , New York, The Academy of Political Science, 196-211.

Priest, George L. (1989) , 'Strict Products Liability: The Original Intent', 10 Cardozo LawReview , 2301-2327.

Priest, George L. (1991) , 'The Modern Expansion of Tort Liability: Its Sources, Its Effects,and Its Reform', 5(3) Journal of Economic Perspectives , 31-50.

Priest, George L. (1992) , 'Can Absolute Manufacturer Liability Be Defended?', 9(1) YaleJournal on Regulation , 237-263.

Priest, George L. (1993) , 'Economic Problems of Accidents and Compensation', 15 University of Hawaii Law Review , 544-552.

Ramseyer, J. Mark (1996) , 'Products Liability Through Private Ordering: Notes on AJapanese Experiment', 144 University of Pennsylvania Law Review .

Rea, Samuel A., Jr. (1982) , 'Nonpecuniary Loss and Breach of Contract', 11 Journal of LegalStudies , 35-53.

Rea, Samuel A., Jr. (1984) , 'Contingent Damages, Negligence and Absolute Liability: AComment on Graham and Peirce', 13 Journal of Legal Studies , 469-474.

Rea, Samuel A., Jr. (1985) , 'Comments on Epstein', 14 Journal of Legal Studies , 671-674.

Rice, David A. (1985) , 'Product Quality Laws and the Economics of Federalism', 65 BostonUniversity Law Review , 1-64.

Ringleb, Al H. and Wiggins, Steven N. (1990) , 'Liability and Large-Scale, Long-TermHazards', 98 Journal of Political Economy , 574-595.

Riordan, Michael H. (1986) , 'Monopolistic Competition with Experience Goods', 101 Quarterly Journal of Economics , 265-279.

Risa, Alf Erling (1994) , 'Preference Revelation in Strict Liability Product Safety Markets', 14 International Review of Law and Economics , 41-52.

Rogerson, Carol and Trebilcock, Michael J. (1986) , 'Products Liability and the AllergicConsumer: A Study in the Problems of Framing an Efficient Liability Regime', 36 University of Toronto Law Journal , 52-103.

Rogerson, William P. (1983) , 'Reputation and Product Quality', 14(2) Bell Journal ofEconomics , 508-516.

Rose-Ackerman, Susan (1990) , 'Market-Share Allocations in Tort Law: Strengths andWeaknesses', 19 Journal of Legal Studies , 739-746.

Rose-Ackerman, Susan (1991), 'Tort Law in the Regulatory State', in Schuck, Peter H. (ed.), Tort Law and the Public Interest: Competition, Innovation, and Consumer Welfare , NewYork, W.W. Norton & Co., 80-104.

Rose-Ackerman, Susan (1991), 'Regulation and the Law of Torts', 81 American EconomicReview: Papers and Proceedings , 54-58.

Rose-Ackerman, Susan (1994) , 'Product Safety Regulation and the Law of Torts', inHunziker, Janet R. and Jones, Trevor O. (eds.), Product Liability and Innovation:Managing Risk in an Uncertain Environment , Washington (DC), National AcademyPress, 151 ff.

Rothschild, Michael and Stiglitz, Joseph E. (1976) , 'Equilibrium in Competitive InsuranceMarkets: An Essay on the Economics of Imperfect Information', 90 Quarterly Journalof Economics , 629-649.

Rubin, Paul H. (1993) , Tort Reform by Contract , Washington (DC), American EnterpriseInstitute for Public Policy Research.

Rubin, Paul H. and Bailey, Martin J. (1994) , 'The Role of Lawyers in Changing the Law', 23 Journal of Legal Studies , 807-831.

Rubin, Paul H., Murphy, Dennis and Jarrell, Gregg A. (1988) , 'Risky Products, Risky Stocks', 1 Regulation , 35-39.

Rubinfeld, Daniel L. (1984) , 'On Determining the Optimal Magnitude and Length ofLiability in Torts', 13 Journal of Legal Studies , 551-563.

Rustad, Michael (1992) , 'In Defense of Punitive Damages in Products Liability', 78 IowaLaw Review , 1-88.

Sage, William M. (1988) , 'Drug Product Liability and Health Care Delivery Systems', 40 Stanford Law Review , 989-1026.

Satterthwaite, Mark A. (1979) , 'Consumer Information, Equilibrium Industry Price, and theNumber of Sellers', 10 Bell Journal of Economics , 483-502.

Saving, Thomas R. (1995) , 'Comment: Legal Implications of Imperfect Information inConsumer Markets', 151 Journal of Institutional and Theoretical Economics , 52-57.

Schlee, Edward E. (1996) , 'The value of information about product quality', 27 RANDJournal of Economics , 803-815.

Schwartz, Alan (1985) , 'Products Liability, Corporate Structure, and Bankruptcy: ToxicSubstances and the Remote Risk Relationship', 14 Journal of Legal Studies , 689-736.

Schwartz, Alan (1988) , 'Proposals for Products Liability Reform: A Theoretical Synthesis', 97 Yale Law Journal , 353-419.

Schwartz, Alan (1989a) , 'Commentary on "Towards a Market in Unmatured Tort Claims":A Long Way Yet to Go', 75 Virginia Law Review , 423-430.

Schwartz, Alan (1989b) , 'Views of Addiction and Duty to Warn', 75 Virginia Law Review ,509-560.

Schwartz, Alan (1990) , 'The Myth That Promisees Prefer Supracompensatory Remedies: AnAnalysis of Contracting for Damage Measures', 100 Yale Law Journal , 369-407.

Schwartz, Alan (1992) , 'The Case Against Strict Liability', 60 Fordham Law Review ,819-842.

Schwartz, Alan (1995) , 'Legal Implications of Imperfect Information in Consumer Markets', 151 Journal of Institutional and Theoretical Economics , 31-48.

Schwartz, Alan and Wilde, Louis L. (1979), 'Intervening in Markets on the Basis ofImperfect Information: A Legal and Economic Analysis?', 127 University ofPennsylvania Law Review , 630-682.

Schwartz, Alan and Wilde, Louis L. (1979), 'Equilibrium Comparison Shopping', 46 Reviewof Economic Studies , 543-553.

Schwartz, Alan and Wilde, Louis L. (1982) , 'Competitive equilibria in markets forheterogeneous goods under imperfect information: a theoretical analysis with policyimplications', 13 Bell Journal of Economics , 181-193.

Schwartz, Alan and Wilde, Louis L. (1983a) , 'Imperfect Information in Markets forContract Terms: The Examples of Warranties and Security Interests', 69 Virginia LawReview , 1387-1485.

Schwartz, Alan and Wilde, Louis L. (1983b) , 'Warranty Markets and Public Policy', 1 Information Economics and Policy , 55-67.

Schwartz, Alan and Wilde, Louis L. (1985) , 'Product Quality and Imperfect Information', 52 Review of Economic Studies , 251-262.

Schwartz, Gary T. (1979) , 'Foreword: Understanding Products Liability', 67 California LawReview , 435-496.

Schwartz, Gary T. (1983) , 'New Products, Old Products, Evolving Law, Retroactive Law', 58 New York University Law Review , 796-852.

Schwartz, Gary T. (1985) , 'Directions in Contemporary Products Liability Scholarship', 14 Journal of Legal Studies , 763-778.

Schwartz, Gary T. (1986) , 'Economic Loss in American Tort Law: The Examples of J'Aireand of Products Liability', 23 San Diego Law Review , 37-78.

Schwartz, Gary T. (1990) , 'The Ethics and the Economics of Tort Liability Insurance', 75 Cornell Law Review , 313-365.

Schwartz, Gary T. (1991a) , 'The Myth of the Ford Pinto Case', 43 Rutgers Law Review ,1013-1068.

Schwartz, Gary T. (1991b) , 'Products Liability and Medical Malpractice in ComparativeContext', in Huber, Peter and Litan, Robert (eds.), Liability Maze: The Impact ofLiability Law on Innovation and Safety , 28-80.

Schwartz, Gary T. (1993) , 'The A.L.I. Reporters' Study', 15 University of Hawaii LawReview , 529-543.

Schwartz, Gary T. (1994a) , 'Reality in the Economic Analysis of Tort Law: Does Tort LawReally Deter?', 42 University of California Los Angeles Law Review , 377-444.

Schwartz, Gary T. (1994b) , 'A National Health Program: What Its Effect Would Be onAmerican Tort Law and Malpractice Law', 79 Cornell Law Review , 1339-1381.

Schwartz, Victor (1983) , 'The Post-Sale Duty to Warn: Two Unfortunate Forks in the Roadto a Reasonable Doctrine', 58 New York University Law Review , 892-905.

Schwartz, Victor E. (1992) , 'Innovation and Our Product Liability System: Let Us End theConflict on Incentives', 27(4) Business Economics , 15-18.

Shapiro, Carl (1982) , 'Consumer Information, Product Quality, and Seller Reputation', 13 Bell Journal of Economics , 20-35.

Shapiro, Carl (1983) , 'Consumer Information, Product Quality, and Seller Reputation', 98 Quarterly Journal of Economics , 659-679.

Shavell, Steven (1979) , 'Risk Sharing and Incentives in the Principal and AgentRelationship', 10 Bell Journal of Economics , 55-73.

Shavell, Steven (1980) , 'Strict Liability versus Negligence', 9 Journal of Legal Studies , 1-25.

Shavell, Steven (1982) , 'On Liability and Insurance', 13 Bell Journal of Economics ,120-132.

Shavell, Steven (1984), 'A Model of the Optimal Use of Liability and Safety Regulation', 15 Rand Journal of Economics , 271-280.

Shavell, Steven (1984), 'Liability for Harm versus Regulation of Safety', 13 Journal of LegalStudies , 357-374.

Shavell, Steven (1987) , Economic Analysis of Accident Law , Cambridge (MA), HarvardUniversity Press, 312 p.

Shavell, Steven (1992) , 'Liability and the Incentives to Obtain Information about Risk', 21 Journal of Legal Studies , 259-270.

Shieh, Shiou (1993) , 'Incentives for Cost-Reducing Investment in a Signaling Model ofProduct Quality', 24 Rand Journal of Economics , 466-477.

Siliciano, John A. (1987) , 'Corporate Behavior and the Social Efficiency of Tort Law', 85 Michigan Law Review , 1820-1864.

Simon, Marilyn J. (1981) , 'Imperfect Information, Costly Litigation, and Product Quality', 12 Bell Journal of Economics , 171-184.

Simon, Marilyn J., Wolf, Robert G. and Perloff, Jeffrey M. (1985) , 'Product Safety, LiabilityRules and Retailer Bankruptcy', 51 Southern Economic Journal , 1130-1141.

Smallwood, Dennis E. and Conlisk, John , 'Product Quality in Markets Where Consumers areImperfectly Informed', 93 Quarterly Journal of Economics , 1-23.

Smithson, Charles W. and Thomas, Christopher R. (1988) , 'Measuring the Cost toConsumers of Product Defects: The Value of "Lemon Insurance"', 31 Journal of Lawand Economics , 485-502.

Sowle, Kathryn Dix (1991) , 'Toward a Synthesis of Product Liability Principles: Schwartz'sModel and the Cost-Minimization Alternative', 46 University of Miami Law Review ,1-110.

Speir, John P. (1990) , 'Efficiency Implications of the Sindell-Rexall Rule', 10 Cato Journal ,603-607.

Spence, A. Michael (1975) , 'Monopoly, Quality and Regulation', 6 Bell Journal ofEconomics , 417-429.

Spence, A. Michael (1978) , 'Consumer Misperception, Product Failure and Product Liability', 44 Review of Economic Studies , 561-572.

Spulber, Daniel F. (1989) , Regulation and Markets , Cambridge (MA), MIT Press.

Swazey, Judith P. (1991) , 'Prescription Drug Safety and Product Liability', in Huber, PeterW. and Litan, Robert E. (eds.), The Liability Maze: The Impact of Liability Law onSafety and Innovation , Washington (DC), The Brookings Institution, 291-333.

Sykes, Alan O. (1989) , 'Reformulating Tort Reform, (book review of Peter Huber, 'Liability:The Legal Revolution and Its Consequences)', 56 University of Chicago Law Review ,1153 ff.

Symposium (1970) , 'Products Liability: Economic Analysis and the Law', 38 University ofChicago Law Review , 1-141.

Thoman, Lynda (1994) , 'Strict liability and negligence rules when the product isinformation', 44 Economic Letters , 205-213.

Tietz, Gerald F. (1993) , 'Strict Products Liability, Design Defects and CorporateDecisionmaking: Greater Deterrence Through Stricter Process', 38 Villanova LawReview , 1361-1460.

Tirole, Jean (1990) , The Theory of Industrial Organization , Cambridge (MA), The MITPress.

Trebilcock, Michael J. (1987) , 'The Social-Insurance-Deterrence Dilemma of Modern NorthAmerican Tort Law: A Canadian Perspective on the Liability Insurance Crisis', 24 SanDiego Law Review , 929-1002.

Trebilcock, Michael J. (1990) , 'Comment (on Viscusi, "The Performance of LiabilityInsurance in States with Different Products-Liability Statutes")', 19 Journal of LegalStudies , 845-849.

Twerski, Aaron D. (1983) , 'The Role of the Judge in Tort Law: From Risk-Utility toConsumer Expectations: Enhancing the Role of Judicial Screening in Product LiabilityLitigation', 11 Hofstra Law Review , 861 ff.

U.S. Department Of Justice Tort Policy Working Group (1986) , Report on the Causes,Extent, and Policy Implications of the Current Crisis in Insurance Affordability andAvailability ,

Washington (DC), Government Printing Office.

Venezian E.C. (1975) , 'Insurer Capital Needs Under Parameter Uncertainty', 42 Journal ofRisk and Insurance , 19 ff.

Viscusi, W. Kip (1984) , Regulating Consumer Product Safety , Washington (DC), AmericanEnterprise Institute for Public Policy Research.

Viscusi, W. Kip (1985) , 'Consumer Behavior and the Safety Effects of Product SafetyRegulation', 28 Journal of Law and Economics , 527-553.

Viscusi, W. Kip (1986) , 'The Determinants of the Disposition of Product Liability Claimsand Compensation for Bodily Injury', 15 Journal of Legal Studies , 321-346.

Viscusi, W. Kip (1988), 'Product Liability and Regulation: Establishing the AppropriateInstitutional Division of Labor', 78 American Economic Review. Papers andProceedings , 300-304.

Viscusi, W. Kip (1988), 'Product Liability Litigation with Risk Aversion', 17 Journal ofLegal Studies , 101-121.

Viscusi, W. Kip (1988), 'Pain and Suffering in Product Liability Cases: SystematicCompensation or Capricious Awards?', 8 International Review of Law and Economics ,203-220.

Viscusi, W. Kip (1989), 'Toward a Diminished Role for Tort Liability: Social Insurance,Government Regulation, and Contemporary Risks to Health and Safety', 6 Yale Journalon Regulation , 65-107.

Viscusi, W. Kip (1989), 'The Interaction between Product Liability and Workers'Compensation as Ex Post Remedies for Workplace Injuries', 5 Journal of Law,Economics & Organization , 185-210.

Viscusi, W. Kip (1989), 'Prospective Reference Theory: Toward an Explanation of theParadoxes', 2 Journal of Risk and Uncertainty , 235-264.

Viscusi, W. Kip (1990a) , 'The Performance of Liability Insurance in States with DifferentProducts-Liability Statutes', 19 Journal of Legal Studies , 809-836.

Viscusi, W. Kip (1990b) , 'Wading Through the Muddle of Risk-Utility Analysis', 39 American University Law Review , 573-614.

Viscusi, W. Kip (1991a) , 'The Dimensions of the Product Liability Crisis', 20 Journal ofLegal Studies , 147-177.

Viscusi, W. Kip (1991b) , Reforming Products Liability , Cambridge (MA), Harvard UniversityPress.

Viscusi, W. Kip (1991c) , 'Product and Occupational Liability', 5(3) Journal of EconomicPerspectives , 71-91.

Viscusi, W. Kip (1993a) , 'The Risky Business of Insurance Pricing', 7 Journal of Risk andUncertainty , 117-139.

Viscusi, W. Kip (1993b) , Product-Risk Labelling: A Federal Responsibility , Washington(DC), American Enterprise Institute for Public Policy Research.

Viscusi, W. Kip and Hersch, Joni (1990) , 'The Market Response to Product SafetyLitigation', 2 Journal of Regulatory Economics , 215-230.

Viscusi, W. Kip and Magat, Wesley A. (1987) , Learning About Risk: Consumer and WorkerResponses to Hazard Information , Cambridge (MA), Harvard University Press.

Viscusi, W. Kip, Magat, Wesley A. and Huber, Joel (1987) , 'An Investigation of theRationality of Consumer Valuations of Multiple Health Risks', 18 Rand Journal ofEconomics , 465-479.

Viscusi, W. Kip and Moore, Michael J. (1991a) , 'Rationalizing the Relationship betweenProduct Liability and Innovation', in Schuck, Peter H. (ed.), Tort Law and the PublicInterest: Competition, Innovation, and Consumer Welfare , New York, W.W. Norton &Co., 105-126.

Viscusi, W. Kip and Moore, Michael J. (1991b) , 'An Industrial Profile of the Links betweenProduct Liability and Innovation', in Huber, Peter W. and Litan, Robert E. (eds.), TheLiability Maze: The Impact of Liability Law on Safety and Innovation , Washington(DC), The Brookings Institution, 81-119.

Viscusi, W. Kip and Moore, Michael J. (1993) , 'Product Liability, Research andDevelopment, and Innovation', 101 Journal of Political Economy , 161 ff.

Viscusi, W. Kip, Zeckhauser, Richard J., Born, Patricia and Blackmon, Glenn (1993) , 'TheEffect of 1980s Tort Reform Legislation on General Liability and Medical MalpracticeInsurance', 6 Journal of Risk and Uncertainty , 165-186.

Wade, John W. (1983) , 'On the Effect in Product Liability of Knowledge Unavailable Priorto Marketing', 58 New York University Law Review , 734-764.

Wagner, Wendy E. (1997) , 'Choosing Ignorance in the Manufacture of Toxic Products', 82 Cornell Law Review , 773-855.

Ward, John O. (1988) , 'Origins of the Tort Reform Movement', 6(3) Contemporary PolicyIssues , 97-107.

Weber, Nathan (1987) , Product Liability: The Corporate Response , New York, ConferenceBoard.

Weicher, John C. (1981) , 'Product Quality and Value in the New Home Market: Implicationsfor Consumer Protection Regulation', 24 Journal of Law and Economics , 365-397.

Weinrib, Ernest J. (1985) , 'The Insurance Justification and Private Law', 14 Journal of LegalStudies , 681-687.

Welling, Linda (1991) , 'A Theory of Voluntary Recalls and Product Liability', 57 SouthernEconomic Journal , 1092-1111.

Wheeler, Malcolm E. (1984) , 'The Use of Criminal Statutes to Regulate Product Safety', 13 Journal of Legal Studies , 593-618.

Wilde, Louis L. (1992) , 'Comparison Shopping as a Simultaneous Move Game', 102 Economic Journal , 562-569.

Williams, Stephen F. (1993) , 'Second Best: The Soft Underbelly of Deterrence Theory inTort', 106 Harvard International Law Journal , 932 ff.

Williamson, Oliver E. (1995) , 'Legal Implications of Imperfect Information in ConsumerMarkets: Comment', 151 Journal of Institutional and Theoretical Economics , 49-51.

Winter, Ralph A. (1988) , 'The Liability Crisis and the Dynamics of Competitive InsuranceMarkets', 5 Yale Journal on Regulation , 455-499.

Winter, Ralph A. (1991) , 'The Liability Insurance Market', 5(3) Journal of EconomicPerspectives , 115-136.

Wolinsky, Asher (1983) , 'Prices as Signals of Product Quality', 50 Review of EconomicStudies , 647-658.

X (1987) , 'Note: Designer Genes That Don't Fit: A Tort Regime for Commercial Releasesof Genetic Engineering Products', 100 Harvard Law Review , 1086-1105.

X (P.M.K. and W.J.S.) (1983), 'Note: Enforcing Waivers in Products Liability', 69 VirginiaLaw Review , 1111-1152.

Other References

Binswanger, Hans P. (1974) , 'A Microeconomic Approach to Induced Innovation', 84 Economic Journal , 940-58.

Schmookler, Jacob (1966) , Invention and Economic Growth , Cambridge (MA), HarvardUniversity Press.

Thirtle, Colin G. and Ruttan, Vernon W. (1987) , The Role of Demand and Supply in theGeneration and Diffusion of Technical Change , Chur, Switzerland, Harwood AcademicPublishers, 173 p.