No. 1, January 2007
Another jury verdict, another arbitrary multimillion-dollar award. Unbounded noneconomic damages are distorting the legal system by undermining constitutional and legal caps on punitive damages and by encouraging plaintiffs’ lawyers to pursue cases of questionable merit. The problem is exacerbated by evidentiary rules that emphasize gamesmanship over truth. Noneconomic damages should be capped and objective safety standards established in order to provide a safe harbor for manufacturers from unfair liability.
It went generally unnoticed last November when the California Supreme Court refused to review an intermediate court’s decision in Buell-Wilson v. Ford Motor Co. But then again, it went generally unnoticed when a jury awarded an arbitrary $368 million in damages in that case, when the trial judge reduced that verdict to an arbitrary $150 million judgment, and when an intermediate appellate court reduced that figure to an arbitrary $82.6 million (which, with interest, works out to over $100 million). Products liability verdicts have become so run-of-the-mill that even nine-digit verdicts and their aftermath receive only local or specialty press coverage, with cursory national coverage. But Buell-Wilson demonstrates much that is wrong with the current liability regime, including the fact that the media is so jaded by litigation abuse that a $368 million verdict is barely newsworthy.
Automakers design SUVs to have high clearance so they can traverse rugged terrain. SUVs are, in many ways, safer than conventional vehicles, in part because their larger mass provides more protection when they collide with another vehicle. (In another era, Ralph Nader complained that automobiles needed to be larger to be safer.) But these features have tradeoffs: the higher clearance raises the center of gravity of the vehicle and affects the handling. The SUV is thus more likely to tip and roll over than a conventional vehicle. Products liability law permits plaintiffs’ lawyers to cherry-pick this drawback and proclaim the entire vehicle “defectively designed” even if, holistically, it is safe. Plaintiffs’ lawyers targeted an earlier Ford SUV, the Bronco II, as unsafe; the National Highway Traffic Safety Administration (NHTSA) in both the George H. W. Bush and Bill Clinton administrations refused petitions to take that vehicle off the road, noting that the Bronco II was no more prone to roll over than many other SUVs on the road. To date, politicians and regulators have elected to permit consumers to choose whether they wish to have an SUV’s off-road features, even though doing so means poorer highway performance and gas mileage. The plaintiffs’ bar, however, has turned to the courts and has sought for years to punish automakers for providing what consumers want.
Ford could be forgiven for feeling confident enough to turn down a $5 million settlement offer going into trial. There were thirteen previous jury trials over allegations that the SUV Ford Explorer’s alleged propensity to roll over was due to a defective design. Benefiting at times from unsympathetic plaintiffs, including at least one who had been driving drunk, Ford won thirteen out of thirteen cases, consistently persuading juries that their SUV was not “defective” just because the laws of physics dictated that it responded differently to road conditions than a different kind of car.
Benetta Buell-Wilson and her husband were more sympathetic plaintiffs, however. Ms. Buell-Wilson was driving her Explorer at freeway speed on Interstate 8 when a motor home in front of her lost a large piece of metal. She swerved to avoid it, and when the SUV hit the dirt shoulder off the highway at high speed, it rolled over, flipping four times before landing on the roof. The roof indented ten inches, leaving Buell-Wilson paralyzed. Ford argued that this probably occurred when centrifugal force from the rollover threw her into the roof, but the plaintiffs blamed the roof crush for her injuries.
The Explorer passed both NHTSA’s and Consumer Union’s rollover tests. But Ford was not allowed to introduce empirical evidence that it had a better safety record than other vehicles and an even better rollover rate than other SUVs Instead, plaintiffs were permitted to theorize that the vehicle’s low stability index--the simple calculation of the track width of a vehicle divided by its center-of-gravity height, disregarding important stability factors such as chassis and suspension characteristics--was a defect in and of itself, though such a stability index does not reliably predict real-world rollover performance. NHTSA rejected a minimum
stability index standard on the grounds that it fails to take important vehicle variables into consideration and would prevent utility vehicles from performing “their intended off-road occupational and recreational functions.” Nevertheless, the jury in Buell-Wilson v. Ford voted 9 to 3 against Ford. Jurors hugged the Wilson family and “joined them at a victory celebration in a nearby hotel cocktail lounge.”
Pick a Number, Any Number: The Problem of Noneconomic Damages
The jury awarded economic damages reflecting medical expenses and lost income for the plaintiff and her husband of $4.6 million. Had Ford committed any wrongdoing, such a damages figure for a car priced less than a hundredth of that amount would have surely deterred any manufacturer even before the millions it costs to defend a products liability case in the American system. The jury awarded $118 million in noneconomic damages and another $246 million in punitive damages. The San Diego County trial judge, Kevin Enright, thought that too high, and reduced the noneconomic award to $70.4 million, with $75 million in punitives. The court of appeals decided that award was the “product of passion and prejudice” and reduced the total noneconomic damages to $23 million, with another $55 million in punitives. (The decision of liability, however, remained untouched. It is one of the strange legal fictions of modern jurisprudence that a jury adjudged to have issued damages that were the product of passion and prejudice is still given presumptive deference by judges that its reasoning in determining liability was impartial and sound.)
All of these numbers are entirely arbitrary, and their arbitrariness is highlighted by the appellate court’s reasoning that damages should be capped at $18 million because that was all that the plaintiffs’ attorney thought to ask for in his closing argument, and thus was the only number supported by “evidence”--though a closing argument is not evidence, just a request. Had the Wilsons’ attorney thought to ask for more money in his closing argument, the appellate court would have picked a different arbitrary number as its benchmark for reasonableness.
Let us concede without debate that Ms. Buell-Wilson’s injuries were catastrophic, and that placing a "fair" money value for compensating such injuries would stymie the deepest philosophical thinkers. But therein lies the problem inherent in noneconomic damages: juries and the judicial system are being asked to measure the immeasurable. Any result is essentially arbitrary; consistency is impossible. The combination of the two contradicts the fundamental principles of the rule of law that like cases are treated alike and that actions have predictable consequences.
Courts, when reviewing noneconomic damages awards, will compare them to what previous juries have awarded, but that can only lead to a one-way ratchet that will inevitably inflate results. One can already see that in this case: Mr. Wilson’s $5 million award for loss of consortium was two to twenty times as large as that in previous similar cases, but was nonetheless affirmed by the appellate court, setting a new outside range for future cases. The next case can ask for $10 million in loss of consortium and point to Buell-Wilson, and future cases can continue to play leapfrog with these numbers.
Such unbounded noneconomic damages create distortions in the legal system in two ways. First, because current jurisprudence ties the size of punitive damages awards to compensatory damages, and because noneconomic damages are considered compensatory damages, unbounded noneconomic damages work as an end around to constitutional and legislative caps on punitive damages. In recent cases, the Supreme Court has held that punitive damages must bear a rational relationship to the compensatory damages. A $2 million punitive damages award for a $4,000 problem with repainting a luxury car is constitutionally inappropriate. But that equation is disrupted when there is no predictable bound on noneconomic damages. If noneconomic damages had been capped even as high as $1 million in Buell-Wilson, then the court would not have been able to justify a $55 million punitive damages award ten times the compensatory damages under Supreme Court precedent. This ties into the second problem. The possibility of jackpot-size damages gives an incentive to plaintiffs’ attorneys to pursue cases of questionable social utility and merit. When damages are unbounded and largely random, it turns the legal system into a game of Russian roulette for defendants facing multiple trials. One can believe, as I do, that the jury system gets it right the vast majority of the time, but when defenders of the jury system make this argument, they are usually talking about an accuracy rate of 70 to 90 percent or so. When a jury can award gigantic damages, it has the power to overwhelm the decisions of other juries. A manufacturer can win thirteen cases in a row, but if the fourteenth jury votes 9 to 3 to award tens or hundreds of millions of dollars, the first thirteen juries’ decisions are effectively mooted. Ford won thirteen out of fourteen cases, but the mean case cost it $7 million in damages. Ford’s shareholders will hardly take solace to be told by reform opponents that the jury system got it right 93 percent of the time.
Defenders of the status quo often argue that there is no real problem with meritless litigation because contingent-fee plaintiffs’ lawyers have no incentive to bring cases that are not likely to succeed. Because such lawyers realize no income unless they succeed, the system is self-regulating. Even in a world with capped damages, this is not entirely true, because litigation imposes external costs on the parties that are not fully realized by the attorney, and attorneys can obtain economic rents through bringing meritless but colorable litigation that is costlier to defend against than to settle. The problem is exacerbated further when damages are essentially unbounded. Attorneys can pool resources and hold a portfolio of risky cases, waiting for the big score. It becomes economically rational to bring cases in which one is more likely than not to lose, so long as the expected variance is also large: if potential damages are large enough, plaintiffs’ attorneys can make money on meritless cases just on the mistakes that the legal system makes. Capped damages at least change that equation, permitting the judicial system to obtain the benefits of the jury and adversarial system while staunching the damage caused by inevitable judicial mistakes.
Unintelligent Design: The Rejection of Empirical Evidence in California Courts
The Buell-Wilsons’ theory was that the Explorer’s stability index was too low and that vehicles with a low stability index tend to have a higher rate of rollover. For example, plaintiffs introduced evidence that the Ford Bronco II had a lower stability index than the Chevy S-10 Blazer and a rollover rate three times higher. Because the Explorer’s stability index was lower than the Bronco II’s, the Explorer was supposedly defective.
Bronco II statistics were admitted into evidence, but Explorer statistics were not. The trial court struck testimony that the Explorer “had one of the best rollover rates compared to other SUVs in its class” and were virtually identical to the Blazer’s. California law, according to the appellate court, looks only at “whether the product fails to perform as the ordinary consumer would expect,” and thus actual empirical performance is not just irrelevant, but admitting it into evidence would be reversible error. California jurors are thus put in the Alice-in-Wonderland scenario of determining defect by considering “the gravity of the danger posed by the challenged design [and] the likelihood that such danger would occur” without being able to consider the empirical evidence of that likelihood. And Ford was effectively prevented from rebutting the claim that the stability index predicts rollover performance.
NHTSA, however, rejects stability indexes, finding them inconsistent with national safety policy because of their failure to account for “wheelbase, kinematic and compliance characteristics of suspensions, and shock absorber characteristics.” NHTSA also found that stability index requirements would reduce the ability of SUVs to perform in off-road settings.
Because liability is divorced from objective science and even empirical evidence of a vehicle’s safety, and because courts permit attorneys to focus on individual elements of design rather than the unified whole, it is impossible for manufacturers to create a litigation-proof vehicle. In any scenario in which there are multiple ways to design a vehicle, one group of plaintiffs’ lawyers can obtain punitive damages against a manufacturer for failing to adopt a particular design paradigm, while another can obtain punitive damages against a manufacturer for adopting that same paradigm. This can even happen in the same litigation. For example, Ford was punished both for a low stability index and for not making its roof stronger, which would have raised the center of gravity, reduced its stability index, and negatively affected vehicle dynamics and rollover propensity.
This sort of whipsawing provides sound reason for federal preemption of state tort law. Where federal regulations require the balancing of competing safety interests, state tort law can frustrate the federal regulatory regime by imposing contradictory requirements. NHTSA has thus propounded regulations that seek to preempt product liability law in this area to the extent it seeks liability for designs that comply with new federal regulations. Special interest groups supporting the plaintiffs’ bar will challenge these regulations in federal court, and there is risk that state courts will simply disregard the regulatory command.
Many other states have reasonable views of the admission of empirical evidence when it comes to products liability law. Even the far-from-defendant-friendly Restatement (Third) of Torts, Products Liability acknowledges that how “the defendant’s design compares with other, competing designs in actual use is relevant to the issue of whether the defendant’s design is defective.” The irrationality of California law on this matter has national implications and demonstrates the need for a federal products liability law.
Separately, as a judicial matter, punitive damages should not be available when the questioned behavior is simply a good-faith dispute over optimal design, especially when the design at issue meets government standards. How can Ford be said to have carried on a “willful and conscious disregard of the rights or safety of others” if they designed the vehicle to be among the best performers in its class? Punitive damages are meant to be reserved for cases of acting in bad faith, and a good-faith dispute over design is not a fair means to impose punishment, even if a manufacturer turns out in hindsight to be wrong. To do otherwise potentially contradicts the “reprehensibleness” analysis that the Supreme Court asks courts to do before imposing punitive damages.
Game-Show Litigation Tactics and the Unique Disadvantage of Transacting Business in English
Every safety measure involves tradeoffs; every engineering decision involves discussions of upsides and downsides. The litigation game is played by subpoenaing records of these discussions, searching the millions of pages of e-mails and documents that express thousands of ideas floated by engineers, and using the advantage of hindsight to characterize as the “smoking gun” the one idea eventually rejected as the evidence of executive disregard for engineers’ concern over safety.
Of course, if no such memo exists, one can always be generated by creative misrepresentation. Witness the following Ford memo, used as evidence in the Buell-Wilson trial, and cited in the appellate court’s opinion:
OGC [the office of general counsel] is concerned we will be the only OEM [original equipment manufacturer] with a vehicle that has a significant chance of failing the CU [Consumers Union] test. I believe that management is aware of the potential risk w/P235 tires and has accepted risk. CU test is generally unrepresentative of real world and I see no “real” risk in failing except what may result in way of spurious litigation.
Astonishingly, this e-mail between two engineers was used to justify punitive damages because it was supposedly evidence that Ford executives decided to “accept the risk” of rollovers in favor of profits. In fact, as the memo demonstrates, the engineers decided there was no real-world risk of rollovers and that the only risk was the one of “spurious litigation.” (The engineer was prescient in this regard.) The final insult: the risk identified in the memo was that of failing the artificial CU test--and the Explorer passed that test when it was performed by Ford, NHTSA, and CU. In litigation, however, judges’ abstention from interfering in adversarial tactics makes this reality irrelevant; what matters is the game show of whether attorneys can persuade a jury to adopt their preferred reading of memos like these. When evidentiary rules make truth beside the point in the adversary system, the only advantage modern justice provides over the medieval practice of selecting champions in trial by combat is its relative bloodlessness.
In any complex design decision, there will be multiple ideas for safety tested and found wanting. All such rejections can eventually be used against the manufacturer. On such thin threads, liability hangs; at such a point, products liability simply becomes a means of transferring wealth from the guilty and innocent alike to attorneys’ pockets, and it ceases to have any deterrent effect on safety.
But there is deterrence, though not desirable deterrence: what trial lawyers deter is not design defects, but design itself. And American manufacturers are at a unique disadvantage because of the state of American products liability.
The entrepreneurial trial bar is perhaps projecting when it regularly accuses industry of putting profits before safety, for that characterization is self-descriptive. Liability is driven by the projected costs and benefits: how easy is it for a litigation entrepreneur to construct a story of manufacturer disregard for safety, and what are the potential returns? It is far better to go after a model that sells millions than one that sells thousands; the former presents economies of scale in trying cases over and over until the jackpot is won. It is also far cheaper to go after a American manufacturer, whose memos and e-mails and depositions will be in English, than it is to go after a foreign manufacturer, whose Swedish memos or German e-mails or Japanese depositions will require expensive and time-consuming translations that make discovery far less feasible.
The disadvantage English-language manufacturers face in the American legal system is illustrated strikingly by the Ford Pinto. The Pinto was driven from the market through litigation and associated bad publicity. The plaintiffs’ bar enlisted the media in a successful campaign that, to this day, has given the car the reputation as a deathtrap in both popular and legal culture. In fact, the riskiness of the Ford Pinto is almost entirely an urban legend: the Pinto was actually safer than other subcompact cars. In 1975-76, the Pinto averaged 310 fatalities per year per million cars in operation. In comparison, the Toyota Corolla averaged 313, the Volkswagen Beetle averaged 374, and the Datsun 1200/210 averaged 405--a fatality rate more than 30 percent higher than that of the Pinto. While the Pinto performed poorly in rear-end collisions, fires in such collisions comprised only 0.6 percent of all automobile fatalities. Overall, the Pinto had a lower fuel-fed fire fatality rate than the average compact or subcompact car.27 But Ford had easily discoverable English-language evidence that could be manipulated against it, and when combined with the legal system’s disregard for empirical evidence, it was sufficient to put Ford on the legal system’s chopping block.
The Price Is Wrong?
Ford will appeal Buell-Wilson to the United States Supreme Court, but it is unclear whether the Roberts Court is inclined to expand its oversight over civil punitive damages or extend its punitive-damages due-process jurisprudence to noneconomics damages. Since President Clinton’s veto of a federal products liability bill in 1996, Congress has chosen not to get involved in fixing this national problem and threat to American competitiveness. It remains difficult to discern any fundamental fairness in punitive damages that can be awarded for good-faith engineering disagreements, or in wholly arbitrary noneconomic damages in which only a trial lawyer’s imagination determines the magnitudes of wealth transfers of tens or hundreds of millions of dollars. Products liability has become a means of transferring wealth from the guilty and innocent alike to attorneys’ and random plaintiffs’ pockets. This does not deter design defects--it just deters design. American competitiveness requires putting reasonable rules back in this game show: cap indeterminate noneconomic damages to reduce distortions in the system; ensure evidentiary rules that give the jury the chance to make a judgment based on the best scientific evidence available (including use of neutral court-appointed experts to assist jurors in the battle of the experts); and create objective safety standards for manufacturers that provide a safe harbor from unfair liability determinations. Only with these fixes can America focus on manufacturing products rather than manufacturing litigation tales.
Ted Frank (email@example.com) is a resident fellow and director of the Liability Project at AEI. As an attorney working in private practice, he represented Ford in unrelated litigation in 2003-05 and General Motors in products liability litigation in 1995-97. The opinions in this piece are his, and he does not speak for Ford or General Motors or their attorneys.
Research assistant Philip Wallach and AEI editorial associate Nicole Passan worked with Mr. Frank to edit and produce this Liability Outlook.
1. Buell-Wilson v. Ford Motor Co., 141 Cal. App. 4th 525 (2006). Facts are drawn from the public court opinions and briefs and from press coverage: Nora Lockwood Tooher, “Explorer Rollover Yields $368.6 Million Verdict,” Lawyers Weekly USA, December 30, 2004; Ray Huard, “$123 Million Awarded in SUV Rollover,” San Diego Union-Tribune, June 3, 2004; Myron Levin, “Jury Orders Ford to Pay $122.6 Million,” Los Angeles Times, June 3, 2004; “Verdict Ends Ford Streak,” Detroit News, June 3, 2004; Ray Huard, “Jury Raises Compensation in SUV Crash,” San Diego Union-Tribune, June 4, 2004; and Michelle Morgante, “Judge Lowers Ford Rollover Victim’s Award,” Associated Press, August 19, 2004.
2. Ralph Nader, Unsafe at Any Speed: The Designed-In Dangers of the American Automobile (New York: Simon & Schuster, 1966).
3. Compare Gerry S. Bahling, R. Thomas Bundorf, G. S. Kaspzyk, Edward A. Moffatt, Kenneth F. Orlowski, and James E. Stocke, “Rollover and Drop Tests--The Influence of Roof Strength on Injury Mechanics Using Belted Dummies” (Technical Paper No. 902314, Society of Automotive Engineers, Warrendale, PA, 1990), and Ford Motor Company, “Rollover Crashes and Roof Strength,” available at www.ford.com/en/company/about/sustainability/2005-06/safVehicleHowCrashes.htm; with Martha Bidez, John E. Cochran, and Dottie King, “Roof Crush as a Source of Injury in Rollover Crashes,” response to request for comments, NHTSA-1999-5572-8 Notice 2, issued October 22, 2001, available at www.citizen.org/documents/ACF6545.pdf (accessed on December 15, 2006) (disputing Bahling et al. and noting that Ford and GM add roll bars to test vehicles). But see Ford Motor Company, “Statement: Roof Strength,” news release, 2004, available at http://media.ford.com/article_display.cfm?article_id=20563 (criticizing Bidez et al. as “seriously flawed, unscientific, and [misinterpreting] the data she is relying on”). NHTSA “believes that there is a relationship between the amount of roof intrusion and the risk of injury to belted occupants in rollover events.” See NHTSA, “Notice of Proposed Rulemaking for Federal Motor Vehicle Safety Standard 216, Docket Number NHTSA-2005-22143" (“NPRM FMVSS 216”), Federal Register 70, no. 162 (August 2005): 49223. NHTSA, however, based its conclusion on a study that measured correlation, rather than causation.
4. NHTSA, “Denial of Petition for Rulemaking: Vehicle Rollover Resistance,” Federal Register 52 (December 1987): 49033, 49035, and 49037.
5. Ray Huard, “Jury Raises Compensation in SUV Crash.”
6. Buell-Wilson, 141 Cal. App. 4th at 532.
7. Ibid., at 555 n. 9.
8. Ibid., at 552-53.
9. BMW of North America v. Gore, 517 U.S. 559 (1996).
10. See generally Neil Vidmar, “The Performance of the American Civil Jury: An Empirical Perspective,” Arizona Law Review 40 (1998): 849; see also David M. Studdert et al., “Claims, Errors, and Compensation Payments in Medical Malpractice Litigation,” New England Journal of Medicine 354 (2006): 2024-33 (finding medical malpractice juries had accuracy rates ranging from 72-84 percent).
11. See, e.g., Association of Trial Lawyers of America, “Contingent Fees Are the Key to the Courtroom,” available at www.atla.org/pressroom/FACTS/tortreform/contingentfees.aspx (accessed on December 15, 2006).
12. Buell-Wilson, 141 Cal. App. 4th at 543.
13. Ibid., at 545.
14. Barker v. Lull Engineering Co., 20 Cal. 3d 413, 431 (1978).
15. NHTSA, “Denial of Petition for Rulemaking: Vehicle Rollover Resistance”, 49035.
16. NPRM FMVSS 216.
17. American Law Institute, Restatement of the Law Third, Torts: Products Liability (Philadelphia: American Law Institute, 1998): §2, comment (d), 19-20, available at http://www.ali.org/ali/promo6081.htm (accessed on December 15, 2006).
18. California Civil Code §3294 (standard for punitive damages).
19. See also David Owen, “Problems in Assessing Punitive Damages against Manufacturers of Defective Products,” University of Chicago Law Review 49, no. 1 (1982): 37-38.
20. See, e.g., State Farm Mutual Automobile Insurance Company v. Campbell, 538 U.S. 408 (2003).
21. The opinion disingenuously omitted the last sentence of the quoted text. Compare Buell-Wilson, 141 Cal. App. 4th at 537 with Buell-Wilson v. Ford Motor Co. Petition for Rehearing at 4.
22. Buell-Wilson, 141 Cal. App. 4th at 537.
23. One example of the plaintiffs’ bar’s disregard for auto safety in bringing products liability cases can be found in the seatback cases. Most manufacturers have replaced rigid seatbacks with yielding seatbacks as a safety measure. See, e.g., Charles Y. Warner, Charles E. Stother, Michael B. James, and Robin L. Decker, Occupant Protection in Rear-End Collisions: II, The Role of Seat Back Deformation in Injury Reduction, SAE 912914, Proceedings of the 35th Stapp Car Crash Conference, 1991. Doing so substantially reduces the number of head and neck injuries in low-speed collisions, albeit with the disadvantage that the seat may fail entirely in the smaller number of high-speed collisions if its occupant is overweight. Courts have refused to recognize that this is a legitimate tradeoff to improve safety and have upheld jury verdicts finding liability for failing to provide a rigid seatback. For example, see Potter v. Ford Motor Co., 2006 WL 1698971, Tenn. Ct. App. (2006) (affirming judgment holding Ford 70 percent liable for $10 million in damages when the plaintiff was driving an Escort with bald tires in the rain and lost control of the car, spinning backward into tree at 30 miles per hour, and seatback failed); Mikolajczyk v. Ford Motor Co., 2006 WL 3392219, Ill. App. 1 Dist., (2006) (No. 1-05-3133) (affirming judgment holding Ford 40 percent responsible and 100 percent liable when a drunk driver plowed into a stopped car at 60 miles per hour and the seatback failed); and Carillo v. Ford Motor Co., 325 Ill. App. 3d 955 (2001) (affirming judgment holding Ford 30 percent responsible and 100 percent liable for $14.5 million in damages when a cocaine- and PCP-impaired driver rear-ended and paralyzed a plaintiff whose seatback failed). In Carillo, as in Buell-Wilson, Ford was not permitted to introduce statistics of how the seatbacks performed in rear-impact collisions. The Carillo plaintiffs, however, were allowed to introduce anecdotal evidence from other paraplegics.
24. For the finding that public foreign corporate defend-ants perform worse than American corporate defendants in American courts, though acknowledging possibility of selection bias, see Utpal Bhattacharya, Neal Galpin and Bruce Haslem, “The Home Court Advantage in International Corporate Litigation,” Journal of Law and Economics (2006, forthcoming), available at http://ssrn.com/abstract=932690.
25. Compare Mark Dowie, “Pinto Madness,” Mother Jones 18 (September/October 1977) (alleging 500 to 900 deaths), with Gary T. Schwartz, “The Myth of the Ford Pinto,” Rutgers Law Review 43 (1991): 1013-68 at 1028-32 (NHTSA investigation found only twenty-seven deaths from rear-collision fuel-fed fire, not all of which were result of product design). See also Superman II (Warner Bros., 1980) and Top Secret! (Paramount, 1984), both of which show an exploding Ford Pinto as a humorous gag.
26. Gary T. Schwartz, “The Myth of the Ford Pinto” at 1028 n. 62.
27. Ibid., 1028-32.