Ginny V. WHITE; Jimmie D. White, Plaintiffs-Appellees, v. FORD MOTOR COMPANY, a Delaware corporation, Defendant-Appellant, and Orscheln Company, a Missouri corporation, Defendant, 312 F.3d 998

Opinion by Judge Kleinfeld; Partial Concurrence and Partial Dissent by Judge Graber. OPINION KLEINFELD, Circuit Judge: This decision addresses several issues, the most important of which relate to punitive damages. Facts This ease went to jury trial, but on the critical factual points, there is not much dispute. Where there is, the facts are of course taken favorably to the verdict. On October 9, 1994, Jimmie White parked his company’s 1993 Ford F-350 pickup truck in Ms driveway. The driveway is sloped, not level, and the truck was parked on the slope pointing downhill. Mr.

White testified that he put the truck into first gear, set the parking brake by stepping on the brake pedal, and went inside. He did not lock the truck. The Whites’ three-year-old son Walter was playing outside, with Mrs. White checking on him through the window from time to time. While she wasn’t watching, Walter got into his father’s pickup truck. The Whites’ theory of the case was that Walter pulled or kicked it out of first gear into neutral. The gearshift lever is a long stalk sticking up from the floor.

A piggy bank turned up under the seat after the accident, so Walter may have been clambering after his lost piggy bank. The parking brake didn’t hold the truck after it was shifted from first to neutral, and it started rolling. Walter got out the passenger door, possibly falling out when the truck rolled over a bump. Tragically, the rear dual wheels of the truck rolled over the little boy’s chest and killed him. The Whites brought this products defect case against Ford and against Orscheln Company, which made the parking brake for Ford.

The Whites alleged strict product liability (defective design), negligence, failure to warn, intentional misrepresentation, and negligent infliction of emotional distress. Orscheln settled during the trial. The jury came back with a verdict against Ford for $2,305,435 in compensatory damages and $150,884,400 in punitive damages. The district court remitted the punitive damages to $69,163,037.10. The plaintiffs’ theory of the case was that the parking brake let go despite being set, and let the truck roll.

Ford knew the parking brake was prone to failure, but kept selling it without recalling it and without warning consumers of the danger. Mr. White testified that if he had been informed that Ford had a problem with the parking brake sometimes letting go, he wouldn’t have parked his truck on a slope, and if he’d been advised of a recall, he would have brought the truck in immediately to be fixed. Ford offered alternative theories of how the accident could have occurred. One was that Mr. White had put the car in first gear but not pressed the parking brake down, and the little boy pulled the gear shift into neutral.

If this is how the accident occurred, then the theory on which the Whites recovered damages, that the parking brake sometimes allowed trucks to roll despite being engaged and that Ford should have warned its customers, would be irrelevant to this accident. But taking the evidence most favorably to the plaintiff, we assume that the jury believed Mr. White’s recollection that he had engaged the parking brake. Mr.

White’s account was bolstered by testimony that the boy couldn’t have pulled the truck out of first gear into neutral, had the parking brake not been engaged, because when it was on a slope the force needed to pull it out of first was considerable. Parking brakes work with a cable pulled by the pedal, a ratchet wheel, and a pawl. A pawl is a hinged or pivoted finger that sticks into a tooth of the ratchet wheel. Parking brakes tend to get loose over time, because, the cable stretches.

In the early nineties, Orscheln designed a parking brake for Ford that was self-adjusting, so that even as the truck aged, the parking brake would still stay tight. Ford started production in 1991 of model year 1992 F-series pickup trucks with the Orscheln self-tightening parking brake. By 1990, Ford had pre-production reports of potential problems with the parking brake being designed for the 1992 F-series trucks. By the time the F-series trucks were in production, the reports had increased, from many different sources.

Sometimes, customers reported, the parking brake pressed freely right down to the floor without engaging. Ford called this the “skip-through-on-apply” or “skip out” problem. Also, some customers reported that their trucks rolled despite the parking brake being engaged. Ford called this the “rollaway” problem. Ford told Orscheln to figure out what was going wrong and fixjt, and threatened to change suppliers if Orscheln couldn’t fix it. Orscheln had a great deal of difficulty getting any parking brakes to fail in laboratory conditions, unless they subjected the brake to extraordinary abuse.

In fact, Orscheln and Ford at first believed that the brake could not fail unless subjected to severe abuse. The evidence allowed for the conclusion that the reason why they couldn’t easily replicate the failure was because it was a rare event or, alternatively, because it happened only with severe abuse. Of the roughly 785,000 1992 and 1993 F-series trucks with this brake, there were only 73 reports of skip-throughs or rollaways between 1992 and 1993, affecting about one truck in ten thousand.

In 1992, Ford established its own engineering group headed by a somewhat junior design engineer, Timothy Rakowicz, to try to figure out the problem. Following Orscheln’s tests, by November 1992, Ford and Orscheln had narrowed in on the problem. Sometimes the tip of the pawl would slip over the tops of the teeth instead of engaging in one of the gaps between the teeth of the ratchet wheel. In April 1993, Mr. Rakowicz wrote a draft report saying that a “tip-on-tip” condition could cause disengagement of the parking brake and that this problem rendered the brakes “defective.” Mr.

Rakowicz wrote that this phenomenon could cause vehicles parked on an incline to roll down a hill, and that it warranted a field campaign and owner notification. But more senior engineers at Ford disagreed with the draft and required Mr. Rakowicz to tone it down, saying instead that the Orscheln test was not valid. Meanwhile, the evidence of problems with F-series brakes piled up, and by March 1993, the National Highway Traffic Safety Administration had become involved in the investigation, having received reports of rollaways. By May of 1993, Orscheln had developed a fix.

All it took was a plastic wedge over the pawl, to make sure it pressed down between the teeth instead of skipping over them. By August 1993, enough of the plastic wedges were manufactured so that Ford could install them in all the trucks on the road. Of course, if all the trucks on the road were modified with the wedge, it would also take the expense of notifying all the customers and paying for the shop labor to disassemble the truck enough to put in the fifteen cent plastic wedge. In addition to the expense, there were other reasons not to recall all the trucks and put in the wedge.

The fix eliminated the self-adjusting feature of the parking brakes. Thus, with the fix installed, the skipovers and rollaways from this cause would be eliminated, but as the trucks aged, the parking brakes would get loose. Also, the Orscheln engineers thought the “skip out” was more an inconvenience than a safety problem, because the user would feel the brake going to the floor with no resistance instead of engaging, and had only to release it and press it down again to get it to engage.

For these reasons, instead of recalling the pickup trucks, Ford issued a Technical Service Bulletin to dealers, in November 1993. The bulletin told dealers to install the wedges if customers complained. Meanwhile, though, reports of rollaways were continuing to accumulate, and the National Highway Transportation Office of Defect Investigation was pressuring Ford to recall the trucks. These problems were reported in only a tiny percentage of trucks sold, about one in ten thousand.

But still, big pickup trucks rolling downhill without drivers are dangerous, even if only one of these had so far actually involved an injury. The White truck rolled down the driveway and killed Walter White in October 1994. It had been purchased in September 1993. By then Ford had already figured out that there was a potential rolla-way problem. At the end of August 1994, which was before the White accident, Ford decided to recall the pickup trucks to install the plastic wedges. But by the time the recall notice was formulated and mailed out to dealers, it was November, after the White accident.

Thus the accident happened after the brake problem was discovered and figured out, after the technical bulletin to dealers had gone out, and after Ford had decided to recall the trucks to install the fix, but before the recall notices or any warnings to ultimate consumers were sent out. In response to a form entitled “Special Verdict,” the jury found that the brakes had a defective design but found that this defect was not a proximate cause of Walter White’s death.

The jury also found that the brakes were defective on account of Ford’s failure to warn and that this defect was a proximate cause of Walter White’s death. Analysis 1.Inconsistent verdicts Ford’s first argument is that the verdicts are inconsistent and, accordingly, require judgment for Ford as a matter of law. Here are the relevant verdicts, preceding the damages awards: SPECIAL VERDICT We, the jury in the above-entitled action, find as follows on particular questions of fact: 1. Was the product in question defective in design? Answer: Yes X No If you answered “no” to Question 1, do not answer Question 2.

If you answered “yes” to question[sic] 1, answer the next question. 2. Was this defect a proximate cause of the death of Walter White? Answer: Yes No X 3. Was the product in question defective for defendant Ford’s failure to warn? Answer: Yes X No If you answered “no” to Question 3, do not answer Question 4. If you answered “yes”, to Question 2[sic], answer the next question. 4. Was the defect for failure to warn a proximate cause of the death of Walter White? Answer: Yes X No 5. Was defendant Ford negligent with respect to the product in question?

Answer: Yes X No If you answered “no” to Question 5, do not answer Question 6. If you answered “yes” to Question 5, answer the next question. 6. Was Ford’s negligence a proximate cause of the death of Walter White? Answer: Yes X No .... 10. Is Ford liable to plaintiffs’'' [sic] Jimmy [sic] and Ginny White for negligent infliction of emotional distress?' Answer: Yes X No 11. Is Ford liable to plaintiffsjp[sic] Jimmy [sic] and Ginny White for ||fen-tional misrepresentation?

Answer: Yes X No Ford’s theory is that once the jury found that the product defect did not proximately cause the death, the verdict had to be for no liability. As Ford reads this verdict, “the jury accepted plaintiffs’ contention that the parking brake had a dangerous tendency to disengage spontaneously, but concluded that it did not spontaneously disengage on this occasion.” If that is what the verdict necessarily means, then Ford is correct.

A plaintiff cannot recover damages for injuries occasioned by the use of a defective product, where the product worked fine in his use and his injury was caused by something else. The Whites argue that Ford waived its objection to inconsistency of the verdicts, because it did not object before the jury was discharged. Whether failure to object before discharge of the jury constituted waiver raises a potentially complicated issue under Federal Rule of Civil Procedure 49(

b) and our decisions in Los Angeles Nut House v. Holiday Hardware Corp. and Home Indemnity Co. v. Lane Powell Moss & Miller. But we need not resolve the waiver issue because, assuming for purposes of discussion that Ford did not waive inconsistency, there was none. As we explained in Floyd v. Laws, “a court has a duty under the seventh amendment to harmonize” a jury’s “seemingly inconsistent answers” if a fair reading allows for it.

In Los Angeles Nut House, we held that “the jury verdict must be upheld unless it is impossible to harmonize the answers under a fair reading,” though we will not save the general verdict if that would “require us to torture a fair reading.” In an inconsistent verdict case, a court asks, not whether the verdict necessarily makes sense under any reading, but whether it can be read in light of the evidence to make sense.

In this case, Ford’s reading is plausible: the brakes were defective, but they didn’t fail here, and the jury was punishing Ford for making defective brakes and selling them to the general public even though the defect didn’t cause this accident. But that isn’t the only plausible reading. After reading the record, we agree with the district court that, as the case was presented to the jury, there was an alternative plausible understanding that makes sense of the verdicts.

The jury reasonably could have thought that this particular parking brake was defective and did let go, allowing the truck to run over Walter White. But the jury could nevertheless have concluded that the brake design defect wasn’t the “proximate” cause of the death. The brake had been designed years before the accident, with many intervening events.

The judge instructed the jury that “[f]or proximate causation to be established, you must find that the injury or damage to plaintiffs was the natural and probable consequence of the defendant’s negligence or wrongful act, and that the injury or damage was foreseeable in light of the attending circumstances.” The jury could have reasoned that the accident wasn’t the “natural and probable consequence” of the brake design defect, because the natural consequence of a manufacturer’s discovering a dangerous design defect would be to warn the customers and recall the product for a fix, rather than to leave the dangerous trucks on the street and leave the drivers ignorant of the hazard.

Mr. White testified that, had he known that the brake could let go despite being set, he wouldn’t have parked the truck on a slope. That is approximately how the district court, having heard all the evidence, reconciled the verdicts. On our review of the evidence and arguments, it makes sense of the verdicts to us. There is no irreconcilable conflict. 2. Evidentiary issues Ford argues that reversal is necessary because the district court erroneously admitted certain expert testimony and evidence of prior incidents.

We review a district court’s evidentiary rulings during trial, including its rulings on the admissibility of expert testimony, for abuse of discretion, and we reverse only where “such nonconstitutional error more likely than not affected the verdict.” a. Testimony of Dr. Campbell Laird The Whites called Campbell Laird, Ph. D., as an expert witness. Dr. Laird’s training and experience, mostly academic but including five years working for Ford Motor Company, was as a metallurgist. He was a professor of material science and engineering.

He testified that that meant he taught about how steel is made and molded into useful shapes, how big a rail had to be to carry the weight of a railroad and how long it would last, the crystal structure of materials, and how materials break when stressed. The great majority of his many publications dealt with how materials break. Dr. Laird was not an accident reconstruction expert. Nor did he have any training or experience to which he testified in designing or manufacturing products. He had never looked at a brake assembly before examining the Ford parking brake in this case and another related case.

He did not do any experiments with the brake parts in this case. Although he examined them, he relied largely on the Orscheln tests and the Orscheln and Ford engineering departments’ memoranda rather than on results of his own tests. Dr. Laird examined the pawl and ratchet wheel of the Whites’ truck’s brake using a scanning electron microscope. He observed wear patterns indicating repeated tip-on-tip engagement.

That meant that this particular brake had been repeatedly subject to the problem that Ford and Or-scheln had discovered, of the pawl sticking on the tip of a tooth instead of settling fully into the gap between teeth of the ratchet wheel. Ford does not dispute that this evidence was admissible. Ford appeals admission over objection of two other opinions to which Dr. Laird was allowed to testify.

He testified that the tip-on-tip engagement could produce spontaneous disengagement of the parking brake if the truck was “subject to perturbation,” a fancy way of saying that the brake could let go if someone slammed the door or shook the truck. Second, he testified that such spontaneous disengagement of the parking brake occurred in this case, causing the rollaway that killed the Whites’ child. In considering Daubert v. Merrell Dow Pharmaceuticals, Inc., the district court concluded that under our decision in McKendall v.

Crown Control Corp. the Daubert factors for “gatekeeping” could not be applied to mechanical engineers. We had held in McKendall that the Daubert factors “are relevant only to testimony bearing on ‘scientific’ knowledge” and did not apply to an expert testifying on how a product ought to have been designed. Our attempt so to limit Daubert was rejected by Kumho Tire Co., Ltd. v. Carmichael. The Supreme Court there held that Federal Rule of Evidence 702 “makes no relevant distinction between ‘scientific’ knowledge and ‘technical’ or ‘other specialized’ knowledge.” We recognized in United States v.

Hankey that McKendall was overruled to the extent that it held that Daubert did not apply to “non-scientifie” testimony. Thus Dau-bert and Kumho Tire did indeed require that the judge apply his gatekeeping role under Daubert to all forms of expert testimony, not just scientific testimony.

The Rule 702 inquiry under Daubert, however, “is a flexible one,” and the “factors identified in Daubert may or may not be pertinent in assessing reliability, depending on the nature of the issue, the expert’s particular expertise, and the subject of his testimony.” We are required to apply an abuse of discretion standard to review of a district court’s decision, “as much to the trial court’s decisions about how to determine reliability as to its ultimate conclusion.” The trial judge thought the two opinions to which exception is taken on appeal were admissible under Rule 702.

As to the first opinion, that the tip-on-tip engagement could produce spontaneous disengagement of the parking brake if the truck was “subject to perturbation,” the judge was within his discretion. Although Dr. Laird had not done any of the experiments himself, his training would make him competent to read the Orscheln and Ford reports, and those reports supported the opinion he gave. It’s hard to see how he was saying any more in this opinion than the Orscheln and Ford experimental results had said, which would make any error in admitting his opinion harmless.

The trial judge was also within his discretion in determining that such scientific bolstering as published articles in reference journals was not required, because there is no reason to suppose that this detail of parking brake manufacture was of general interest to the scientific community and would generate a peer-reviewed literature. The second opinion to which exception is taken is a close question. Dr. Laird testified that such spontaneous disengagement of the parking brake occurred in this case, causing the rollaway that killed the Whites’ child. The foundation for this opinion was skimpy. Dr.

Laird, well within his metallurgical expertise, identified wear on the ratchet wheel of the brake that showed repeated tip-on-tip engagement rather than the proper engagement. That justified the inference that even though the tip-on-tip engagement might be a one in ten thousand phenomenon as applied to all the Ford pickup trucks, it had indeed happened on this particular track. The way Dr. Laird got from the metal wear, showing repeated tip-on-tip engagement, to his opinion on how the accident occurred, did not rely on his metallurgy expertise at all. He just relied on simple logic.

He assumed, for purposes of his opinion, that Mr. White had parked on the sloping driveway, engaged the brake, and put the truck in first gear. Mr. White’s testimony provided a basis for those assumptions. Dr. Laird testified, based on Mr. White’s deposition testimony and an Orscheln engineer’s reports, that the brake must have been engaged, because otherwise the little boy could not have moved the shifter from first gear to neutral. The brake must have let go after the car was shifted into neutral, because otherwise the track would not have rolled.

And the boy was probably too small to have disengaged the parking brake. This opinion took advantage of Dr. Laird’s expertise in metallurgy only for his knowledge that the particular brake had repeatedly engaged in the tip-on-tip position from which spontaneous disengagement could occur. Beyond that, Dr. Laird established no more foundation than anyone trained in any kind of engineering, or even a lay person not trained in engineering, would have to venture the opinion. If Mr.

White’s testimony was correct, and we must assume that the jury so found, the truck did roll, despite being parked in first gear with the parking brake on. Dr. Laird did apply his metallurgical expertise and scientific procedures (electron microscopy) to which no objection is made on appeal, to determine that this brake had repeatedly been subject to the tip-on-tip phenomenon that, Ford and Or-scheln had determined, could lead to spontaneous disengagement and a rollaway if the truck was disturbed. For the part of Dr.

Laird’s opinion not requiring special training or experience, arguably the district court had discretion to admit the opinion under Federal Rule of Evidence 701. The question is close. A layman, which is what an expert witness is when testifying outside his area of expertise, ought not to be anointed with ersatz authority as a court-approved expert witness for what is essentially a lay opinion. We need not decide whether the district court abused its discretion in allowing the opinion into evidence in this case, because we reverse on other grounds. b.

Evidence of prior rollaways Ford argues that the district court abused its discretion in admitting evidence of other rollaways. The court allowed another Ford pickup truck owner, Tammy Bobb, to testify by deposition that her truck had rolled despite the parking brake’s being set, and allowed Dr. Laird to testify about the Bobb rollaway. The court also allowed the Whites to put an exhibit into evidence showing that Ford had received a number of customer complaints of rollaways.

Ford’s contention is that insufficient foundation to show substantial similarity was established and that the customer reports were hearsay. A “showing of substantial similarity is required when a plaintiff attempts to introduce evidence of other accidents as direct proof of negligence, a design defect, or notice of the defect.” Minor or immaterial dissimilarity does not prevent admissibility. We review a district court’s decision to allow such evidence for abuse of discretion.

Ford’s theory of dissimilarity for the Bobb rollaway is that it was caused by a manufacturing defect, not the design defect at issue in this case. But the testimony about that was somewhat ambiguous. Dr. Laird testified that the spring that helps the pawl engage properly between the teeth was not up to specifications on the Bobb brake, but he also said that the design defect was the same and that he attributed the rollaway to the design defect. The district court was within its discretion in concluding that there was sufficient similarity to allow the evidence of the Bobb rollaway to go to the jury.

As for the customer reports, they were clearly admissible' under the business records exception to the hearsay rule to show that Ford had notice of rollaways, and the district court redacted reports of dissimilar rollaways and post-White accident rollaways. Ford argues that it was error to admit the reports for the truth of the matter asserted, because even if they were not hearsay for purposes of proving that Ford had received them, they were hearsay for purposes of proving what the customers reported about rollaways. We need not resolve whether this was error, because it was harmless.

Other evidence of rollaways such as government reports and Orscheln’s experiments came in, and it is highly unlikely that the customer reports affected whether the jury believed that the tip-on-tip problem could and did cause rollaways. 3. Sufficiency of evidence for punitive damages Ford argues that it was entitled to judgment as a matter of law on punitive damages, because the evidence was insufficient to support an award.

The district court ruled that “[t]here was sufficient evidence for a reasonable jury to find that the cure for the brake problem was known and in Ford’s hands before the White vehicle was sold, and that Ford did not apply it.” Ford argues that this was “utterly at odds with the evidence.” We review de novo the district court’s denial of a Rule 50(

b) renewed motion for judgment as a matter of law. The test is whether “the evidence, construed in the light most favorable to the nonmoving party, permits only one reasonable conclusion, and that conclusion is contrary to that of the jury” Ford argues that the evidence showed that it investigated the rollaway reports carefully and promptly turned the matter over to the brake manufacturer for investigation and remedy, but that the engineers disagreed on whether there was anything wrong with the design. Out of 884,000 vehicles using the brake, only three injuries occurred.

There was evidence to support that view, and the jury could have so concluded. But it apparently didn’t. There was also evidence from which the jury could conclude that Ford knew the parking brake could engage in the tip-on-tip position and let a pickup truck roll away, didn’t fix it, didn’t recall it, and didn’t warn drivers of the trucks, all prior to the White accident. Ford concedes that the applicable legal standard is that the Whites had to prove by clear and convincing evidence that Ford acted with conscious disregard for the rights or safety of others.

There are alternative and additional grounds for punitive damages under the Nevada statute, but we need not consider them, because the evidence was sufficient under the standard Ford concedes was applicable. Although it certainly could have concluded otherwise, the jury could permissibly conclude that Ford knew a parking brake tip-on-tip engagement could result in a rollaway, and failed to warn people driving pickup trucks about the brake in conscious disregard of their safety (that is, knowing that someone could be injured or killed and deliberately failing to warn).

The jury could have accepted Ford’s case, that the engineers were hard at work trying to replicate and cure the problem, but they could also have accepted the plaintiffs’ case, essentially that when the engineers had figured out the problem and the fix for it, Ford covered it up with a euphemistic notice to dealers, rather than consumers, instead of a plain warning and immediate recall.

The jury could have accepted Ford’s claims that “skip-through,” where a person pressed the parking brake with his foot and felt no resistance, and rollaway, where the parking brake seemed to engage but didn’t prevent the truck from rolling away, were two different things. It could also have accepted the evidence that they were both the same thing, failure of the pawl to drop properly between the teeth of the ratchet wheel. As for the small number of injuries, the jury could have concluded that it was so unlikely that an accident would occur that Ford should not be punished for it.

Or it could accept the plaintiffs’ case, that once Ford knew that pickup trucks could be rolling down hills without drivers behind the wheel, it was obvious that someone was likely to be killed if Ford didn’t do something about it. The part of Ford’s argument that gives us the most hesitancy is its citation to the Nevada authorities. The Nevada Supreme Court decision in Maduike v. Agency Rent-A-Car affirmed the dismissal of a punitive damages claim.

In Maduike, a young family driving a rental car on vacation reported a brake problem to the rental agency, but the agency refused to fix it, leaving the family stranded unless they drove the car home. They tried to do so and had an accident. The decision affirmed on the ground that no jury could have found that the car rental agency “subjected the Maduikes to ‘cruel and unjust hardship with conscious disregard of the rights of the person.’ ” Ford correctly says that the uncontradicted evidence showed that it exercised more care for the safety of its customers than did the car rental agency in Maduike.

Since Maduike, the Nevada Supreme Court has recognized in Evans v.

Dean Witter Reynolds, Inc., that “Nevada law requires clear and convincing evidence of malice before punitive damages may be recovered.” The term “malice” is a specially defined term of art in Nevada law, and means something less than the ordinary definition, “a desire to harm others or to see others suffer.” A Nevada statute broadens malice to include “implied” malice, which covers not only “conduct which is intended to injure a person” but also “despicable conduct which is engaged in with a conscious disregard of the rights or safety of others.” The Nevada statute defines “conscious disregard” as “knowledge of the probable harmful consequences of a wrongful act and a willful and deliberate failure to act to avoid those consequences.” The Nevada Supreme Court, before Ma-duike, had upheld punitive damages in Granite Construction Co. v.

Rhyne, and Maduike does not cite or purport to overrule Granite Construction. In Granite Construction, a woman hit a bull on the interstate highway and sued a construction company that was supposed to have built a fence to keep livestock off the highway. The jury could have concluded that the construction company decided not to build the fence, despite its contractual obligation to the state to do so, in order to save time and money.

The jury could also have concluded that the state had waived the fence requirement because no livestock were thought to be in the area and it wanted to avoid delay, and when the single bull was discovered, the construction company reasonably tried to find the owner and have him remove the bull. Taking the evidence most favorably to the verdict, however, the Nevada Supreme Court upheld the punitive damages on the ground that the construction company “consciously and deliberately disregarded known safety procedures” and was “guilty of malice, express or implied.” We held in Coughlin v.

Tailhook Association that, under Granite Construction, punitive damages were permissibly awarded against a hotel for failing to do anything about a rowdy party in which a woman was “attacked, groped, grabbed, and handled by a throng of men.” We held in Cough-lin that the 1995 statutory amendment to the Nevada punitive damages statute “indicates that Granite’s definition of ‘malice, express or implied,’ was correct,” and punitive damages could be awarded on the theory that “a ‘conscious disregard for the safety of others’ amountfs] to implied malice.” The Nevada Supreme Court has not stated the principle of law that reconciles the Nevada authorities.

After Maduike, the Nevada Supreme Court in Dow Chemical Co. v. Mahlum vacated an award of punitive damages following its reversal of intentional tort claims against the defendant. The Court’s opinions in Maduike and Dow Chemical do not cite or discuss Granite Construction. We are bound by our

interpretation of Nevada law in Coughlin. The

interpretation we adopted there, that “a ‘conscious disregard for the safety of others’ amount[s] to implied malice,” allows for the punitive damages in this case, taking the evidence most favorably to the prevailing party.

The White jury was asked “[d]o you find by clear and convincing evidence that Ford acted with oppression or malice in the conduct upon which you base your finding of liability for the death of Walter White?” and the jury answered “Yes.” In addition to the jury’s determination that Ford was liable on the basis of defective design, negligence, failure to warn, and negligent infliction of emotional distress, punitive damages are bolstered by the jury’s special verdict finding of “intentional misrepresentation,” which is an intentional tort and would both fall within Evans and survive the Nevada Supreme Court’s rejection of punitive damages in Dow Chemical. 4.

Excessiveness of punitive damage award and extraterritoriality Ford argues that errors in instructions require a new trial. We review a district court’s formulation of jury instructions in a civil case for abuse of discretion. Jury instructions must fairly and adequately cover the issues presented, must correctly state the law, and must not be misleading. Prejudicial error results from jury instructions that, when viewed as a whole, fail to fairly and correctly cover the substance of the applicable law.

The judge gave a Nevada pattern jury instruction to guide the jury’s judgment on the amount of punitive damages, telling the jury only to act “without passion or prejudice” and to consider reprehensibility and deterrence. The instruction said: Members of the jury, you’ve now heard the evidence of the financial condition of the defendant Ford Motor Company. Because you have answered yes to special verdict question number fifteen, you may, in your discretion, award pffljHtive or exemplary damages against jfpfen-dant Ford for sake of example by way of punishment.

Your discretion should be exercised without pasáitó or prejudice. In arriving at any awápjl of punitive damages, you are to cóésider the following: [o]ne, the reprehensibility of the conduct of the defendant; two, the amount of punitive damages which will have a deterrent effect on the defendant in light of defendant’s financial condition. That’s the complete jury instruction on punitive damages, and you’ll have this instruction with you in the jury room. Ford requested an instruction on extraterritoriality.

It would have barred the jury from punishing Ford for impact other than on Nevadans: In determining the amount of pgjjSitive damages, if any, that is necessa^jfor punishment and deterrence, yojllfnay consider only Defendant’s wrongfffl conduct that has had an impact on the citizens of Nevada. You may not award any punitive damages for the purpose of punishing Defendant relative to the sale of vehicles in other States, or for the purpose of punishing or deterring Defendant’s conduct outside the State of Nevada.

The district court refused to give this instruction, any part of this instruction, or anything equivalent. Ford’s argument for its “only Defendant’s wrongful conduct that -has had an impact on the citizens of Nevada” instruction was based on the federalism discussion in BMW of North America, Inc. v. Gore. The Court held in BMW that “the federal excessiveness inquiry appropriately begins with an identification of the states’ interests that a punitive award is designed to serve.” BMW involved deceptive trade practices and failure to disclose pre-sale repairs that affected the value of a new car.

The Court noted that “the States need not, and in fact do not, provide such protection in a uniform manner.” In addition to noting the diverse ways states had actually protected consumers from such deception, the Court noted such hypothetical policy choices that a state might make, such as “plausibly concluding] that the administrative costs associated with full disclosure would have the effect of raising car prices to the State’s residents.” The Court in BMW imposed a territorial limitation on punitive damages in the interest of federalism.

This federalism includes the flexibility for a state to have whatever policy it chooses, subject to constitutional and congressional limits. For that flexibility to exist, no state can be permitted to impose its policies on other states.

Because no single state could “impose its own policy choice on neighboring States,” the Court held that “a State may not impose economic sanctions on violators of its laws with the intent of changing the tortfeasors’ lawful conduct in other States.” Alabama, the Court held, “does not have the power ... to punish [a defendant] for conduct that was lawful where it occurred and that had no impact on [the punishing state] or its residents,” nor may it “impose sanctions on [a defendant] in order to deter conduct that is lawful in other jurisdictions.” Evidence of extraterritorial conduct, such as sales in other states, “may be relevant to the determination of the degree of reprehensibility of the defendant’s conduct,” but admissibility of evidence of reprehensibility does not, under BMW, go the extra and substantial step of entitling a state to award enough money to deter conduct (or at least lawful conduct) in other states.

The question whether a state can impose punitive damages sufficient to punish unlawful (as opposed to lawful) conduct in other states was left open by BMW. The Court noted in BMW that because “the verdict was based in part on out-of-state conduct that was lawful where it occurred, we need not consider whether one State may properly attempt to change a tortfea-sor’s unlawful conduct in another state.” We avoided the need to consider this question in Neibel v. Trans World Assurance Co., a RICO case involving fraudulent sale of insurance and sham tax shelters.

Compensatory damages were $259,366, and we held that the $500,000 in punitive damages was “supported by [California’s] interest in protecting its own consumers and its own economy” and that, considering the evidence and the amount, “it [was] clear that only California interests [were] involved.” That means of avoiding the question is unavailable in this case. The evidence focused on the number of vehicles Ford sold nationally, and the number of parking brake failures reported nationally. The national evidence was not limited by any jury instructions or admonitions to limit its relevance to reprehensibility.

The jury was permitted, despite Ford’s request for instructions to cure this error, to award damages to vindicate the interests of all Ford pickup truck buyers everywhere. In essence, the jury was asked to measure damages by Ford’s harm to the whole country. Plaintiffs’ closing argument sought damages that would punish Ford for its conduct toward consumers in all states, not just Nevada.

Plaintiffs’ attorney emphasized that “there are 884,000 people in this country who have these vehicles that got a letter that didn’t tell them the truth as to why these vehicles were being recalled” and “your verdict for punitive damages must be loud enough so that it is on the front page of every newspaper tomorrow morning, so every person in this country knows, if they have that vehicle, they can take it into the shop and get it fixed.” Plaintiffs’ attorneys told the jury that Ford knew its actions would cause deaths of children “across the country.” Plaintiffs’ attorney repeated the “across the country” phrase several times, using a repetitive mantra for emphasis, in his “send them a message” argument.

The emotional argument wound up with plaintiffs’ attorney’s statement that he had prosecuted criminals in the district attorney’s office for twenty years, and “[his] work hasn’t changed so much after all.” Nothing in the argument addressed how many trucks were sold in Nevada, and our attention has not been directed to any evidence about how many trucks were sold in Nevada.

The entire thrust of the argument was that this Nevada jury now should vindicate the interests of all Ford truck owners everywhere with a verdict that would make the front page of every newspaper in the country, so that the chief executive officer and other top officers of Ford Motor Company would “pour[ ] out of their chairs like water and crumple[ ] on the floor.” With this evidence and argument, and the denial of the instruction, we cannot conclude, as we did in Neibel, that even without the instruction, the jury plainly was vindicating only the state’s interest in protecting its citizens or that it was addressing Ford’s reprehensibility only in the case at bar.

After the verdict, and in light of BMW, Ford moved for a remittitur based on ex-cessiveness and on the extraterritoriality rule in BMW. The district court denied the motion on extraterritoriality, saying that Ford had not shown that its conduct would be legal in other jurisdictions.

But the district court conceded that Ford had submitted materials arguing that “many states do not recognize a post-sale duty to warn and that at least five states would not permit punitive damages in this type of case.” The district court attempted to distinguish BMW on the theory that the 884,-000 pickup trucks Ford “marketed in North America was universal market conduct performed centrally; it was not peculiar to or necessarily performed in a given state.” But we cannot adopt that distinction.

In BMW, BMW marketed its cars nationally, too, and the punitive damages in that case were imposed because BMW “had adopted a national policy in 1988 concerning cars that were damaged in the course of manufacture or transportation.” In that case, a national policy affected a sale of a car in Alabama, in this case a pickup truck in Nevada. No material difference. Thus in this case, the Nevada jury was asked to impose punitive damages to protect people in other states with Ford pickup trucks, and the district court refused to reduce the punitive damages on account of the extraterritorial aspect.

The district court expressly rejected the argument that the award should be limited to punishment for the conduct that had taken place in Nevada. The court refused to limit the damages to vindication of Nevadans’ rights because Ford’s conduct was “performed centrally.” We cannot conclude that the ratio analysis by which the district court explained its remittitur limited the damages to vindicating the rights of Nevadans.

The jury was encouraged by argument to award damages for Ford’s wrongs to the entire country, and the court rejected an instruction that would have told the jury to vindicate only the wrongs done in Nevada. Possibly the jury would have chosen as large an award had it been told to vindicate only the rights of Nevadans, but possibly it would have chosen a substantially lower award. For all we know, the jury would have applied a much lower ratio than the thirty to one the court chose, or the sixty-six to one that the jury initially chose, had it been told that it should limit its scope to the interests of Nevadans.

A punitive damages award that encompasses a defendant’s extraterritorial conduct may be unconstitutional even if the size of the award itself, as compared to the compensatory damages, is not outside the bounds of due process.

The logic and language of BMW suggest that if the Court were to “consider whether one State may properly attempt to change a tortfeasor’s unlawful conduct in another state,” the answer would have to be “No.” The Tenth Circuit, the only other Circuit to have considered the question so far, reaches that conclusion. “Despite this comment we read the opinion [BMW] to prohibit reliance upon inhibiting unlawful conduct in other states.” That makes sense for two reasons. First, the core conduct in BMW, consumer fraud, would likely be wrongful to some degree in all states.

The diversity of state approaches to consumer fraud the BMW Court pointed to was in

part in how the conduct is sanctioned rather than whether it is permitted. The Court’s discussion was in part hypothetical, conceiving of potential rather than actual diversity in treatment. The Court said a state legislature “might plausibly conclude that the administrative costs associated with full disclosure would have the effect of raising car prices to the State’s residents.” That comment protected a hypothetical state legislature that might take a different approach from Alabama’s, even if no actual legislature had taken a different position.

Perhaps most important, the variation in policies of punishment, even where the conduct is unlawful in all states, amounts to an important distinction in policy. For example, Nevada has no ceiling on punitive damages in a case such as this. In this case, because the jury vindicated the rights of all Ford pickup truck drivers everywhere, Nevada has effectively imposed $70 million in punitive damages in part to protect Alaskans, among others, from failure to warn of defects in pickup trucks.

But Alaska has quite a different policy on punishment by means of punitive damages: its legislature imposed a ceiling, probably $7 million in this case, with fifty cents on the dollar payable to the Alaska state treasury. By imposing ten times what Alaska would allow, with all the money going to the Whites and their attorneys, rather than half of it to the state government, Nevada has created very different incentives from Alaska for manufacturers, distributors, and plaintiffs’ attorneys.

A manufacturer of an innovative but untried product, such as the self-tightening parking brake in this case, faces much more risk selling it in Nevada than in Alaska. A national company sometimes limits its sales according to variations in risk, as when liability insurers pull out of high-verdict states or mail-order companies refuse orders from some states. The Nevada legislature has chosen an arguably more safety-oriented approach, the Alaska legislature a less risk-averse approach friendlier to innovation.

Even though both states treat distribution of defectively designed products and failure to warn of dangerous defects as tortious, the difference in how they penalize the tortious conduct expresses significantly different policy choices. The Supreme Court in BMW, speaking of BMW’s being a large corporation, said “its status as an active participant in the national economy implicates the federal interest in preventing individual States from imposing undue burdens on interstate commerce.

While each State has ample power to protect its own consumers, none may use the punitive damages deterrent as a means of imposing its regulatory policies on the entire Nation.” This federalism concern applies strongly to the case at bar. Nevada is free, in the absence of federal legislation to the contrary, to choose a policy that may sacrifice some innovation in favor of safety, and Alaska is free to choose a policy that may sacrifice some safety in favor of innovation.

Or, Alaska may think it gets more rather than less safety by limiting the penalties for innovations and economies, even though some turn out to have unanticipated risks. Neither state is entitled, in our federal republic, to impose its policy on the other.

If Nevada imposes an award based on vindicating a national interest in safety, as the jury was encouraged to do in this case and as the district court expressly permitted, then it may deter not only conduct tortious in other states, but also innovations and economies of production that other states have purposely tailored then-laws not to discourage so strongly.

Measured by the Alaska legislative policy, Nevada’s policy would be overdeterrence All of this suggests that under BMW, “a State may not impose economic sanctions on violators of its laws with the intent of changing the tortfeasors’ ... conduct in other States,” whether the extraterritorial conduct is lawful or not. Though no other circuit court has yet held one way or the other, the dictum previously quoted from the Tenth Circuit takes this position, and several district and state courts have so held.

An Eastern District of New York products liability case approves a jury instruction telling the jury it could consider “what is reasonably required to vindicate New York State’s legitimate interests in punishment and deterrence, if any,” but that it was “not authorized to impose punitive damages to protect people outside the State of New York.” We agree with the Eastern District that that instruction was consistent with BMW.

Even if the BMW territorial limitation applied only to conduct lawful in other states, as opposed to unlawful conduct differently sanctioned by other states, we would be compelled to reverse because of the extraterritorial award in this case. The failure to warn found unlawful here is probably not unlawful in all states. The jury in this case found that although the brake was defective in design, that design defect did not proximately cause Walter White’s death.

It imposed liability for Ford’s failure to warn, not for the design defect, after having been instructed that “[a] manufacturer has a responsibility to warn of a defective product after it has been manufactured and sold, if the manufacturer becomes aware of the defect.” But not all states agree with that instruction. Some impose a post-sale duty to warn, some don’t. Arkansas, Illinois, Nebraska, and Texas, to name a few, do not impose a post-sale duty to warn.

Pennsylvania recognizes a post-sale duty to warn only if the plaintiff can demonstrate that the product was defective from the date of manufacture and that the manufacturer had notice of the defect. Thus, while we assume that the court’s instruction in this case was correct as a matter of Nevada law (which has not been challenged), it is not the law everywhere. Even among the states that impose a post-sale duty to warn, there are differences in who is supposed to be warned.

The Whites, for example, were neither purchasers nor owners of the pickup truck, so they needed a rule that a warning must be calculated to go beyond purchasers and owners to actual users. Kansas, though it requires a post-sale warning, limits the post-sale duty to warn to “ultimate consumers who purchased the product,” which would leave out the Whites. Thus the jury in this case was invited to award punitive damages to vindicate the interests of Ford pickup truck drivers all over the country, but the conduct for which the jury punished Ford was actually lawful in a number of other states.

BMW reminds us, in prohibiting just such extraterritorial punishment as was imposed in this case, that “[t]o punish a person because he has done what the law plainly allows him to do is a due process violation of the most basic sort.” The district court’s refusal to limit the jury to consideration of Nevada’s interests, combined with the plaintiffs’ lawyers exhortations to let the decision resonate “across the country,” compels us to conclude that the jury here was permitted to engage in “a due process violation of the most basic sort” when it arrived at its punitive damages award.

We therefore reverse the decision of the district court as to Ford’s Rule 50(

b) motion for a new trial on punitive damages. Because we reverse on the ground.that the punitive damages award unconstitutionally allowed a Nevada jury to punish Ford for out-of-state conduct, we do not need to address de novo whether these punitive damages were unconstitutionally excessive under the BMW guideposts, as is normally required of us by the Supreme Court’s ruling in Cooper Industries, Inc. v. Leatherman Tool Group, Inc. Nor do we consider whether the award is excessive under Nevada law. On remand, the jury must decide on a punitive damages award within the territorial restraint established by BMW.

Extraterritorial conduct is admissible for its bearing on degree of reprehensibility, but the jury must be limited to punitive damages reasonably required to vindicate Nevada’s legitimate interests in punishment and deterrence, if any, and prohibited from imposing punitive damages to protect people or punish harm outside of Nevada. The jury’s award on remand may yet be constitutionally excessive, or, it may be within the limits of BMW. We do not speak on this issue, because we need not reach it.

While we are troubled by the possibility that the jury award in this case may also have been unduly influenced by the inflammatory closing argument of the plaintiffs’ attorneys, we need not reach the issue of whether' the argument requires reversal, and trust that on remand the district court will take care to ensure that the proceedings are not tainted by inflammatory argument appealing to passion or prejudice. AFFIRMED as to liability determination and compensatory damages. REVERSED AND REMANDED as to punitive damages. Each party to bear its own costs on appeal. .

This decision had to await the Supreme Court's decision in Cooper Industries, Inc. v. Leatherman Tool Group, Inc., 532 U.S. 424, 121 S.Ct. 1678, 149 L.Ed.2d 674 (2001) and our decision in Baker v. Exxon Corp (In re Exxon Valdez), 270 F.3d 1215 (9th Cir.2001). . See, e.g., United States v. Hicks, 217 F.3d 1038 (9th Cir.), cert. denied, 531 U.S. 1037, 121 S.Ct. 627, 148 L.Ed.2d 536 (2000). . Brief for Appellant at 20, White v. Ford Motor Co., 2002 WL 31687641 (emphasis in original). . 825 F.2d 1351 (9th Cir.1987). . 43 F.3d 1322 (9th Cir.1995). . 929 F.2d 1390 (9th Cir.1991). . Floyd, 929 F.2d at 1396. .

Los Angeles Nut House, 825 F.2d at 1354. . See General Electric Co. v. Joiner, 522 U.S. 136, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997); Old Chief v. United States, 519 U.S. 172, 174 n. 1, 117 S.Ct. 644, 136 L.Ed.2d 574 (1997); Metabolife International, Inc. v. Wornick, 264 F.3d 832 (9th Cir.2001). . United States v. Ramirez, 176 F.3d 1179, 1182 (9th Cir.1999). . 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). . 122 F.3d 803 (9th Cir.1997). . McKendall, 122 F.3d at 806. . 526 U.S. 137, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). . Kumho Tire, 526 U.S. at 147, 119 S.Ct. 1167. . 203 F.3d 1160 (9th Cir.2000). .

Hankey, 203 F.3d at 1169 n. 7. . Kumho Tire, 526 U.S. at 147, 119 S.Ct. 1167; cf. Jinro America Inc. v. Secure Investments, Inc., 266 F.3d 993 (9th Cir.2001). . Kumho Tire, 526 U.S. at 150, 119 S.Ct. 1167 (quoting Daubert, 509 U.S. at 594, 113 S.Ct. 2786). . Id. (citation and internal quotation marks omitted). . Id. at 152, 119 S.Ct. 1167; see also Hankey, 203 F.3d at 1166-67. . See, e.g., Cabrera v. Cordis Corp., 134 F.3d 1418, 1421 (9th Cir.1998). . E.g. United States v. Cordoba, 194 F.3d 1053, 1056 (9th Cir.1999). . Cf. Metabolife International, Inc. v. Wornick, 264 F.3d 832, 841 (9th Cir.2001). .

While we were considering this decision and dissent, Mukhtar v. California State University - F.3d - (9th Cir.1992), came down. Because we reverse on other grounds, we do not reach the question whether Mukh-tar would affect our analysis of the expert testimony issues. . Cooper v. Firestone Tire & Rubber Co., 945 F.2d 1103, 1105 (9th Cir.1991). . Western Recreational Vehicles, Inc. v. Swift Adhesives, Inc., 23 F.3d 1547, 1555 (9th Cir.1994). . Id. . See, e.g., United States v. Fuchs, 218 F.3d 957, 965 (9th Cir.2000). . United States v. Miller, 771 F.2d 1219, 1238 (9th Cir.1985). . See Johnson v.

Paradise Valley Unified School Dist., 251 F.3d 1222, 1226 (9th Cir.), cert. denied, 535 U.S. 1055, 122 S.Ct. 645, 151 L.Ed.2d 563 (2001). . See Forrett v. Richardson, 112 F.3d 416, 419 (9th Cir.1997) (citing Bank of the West v. Valley National Bank of Ariz., 41 F.3d 471, 477 (9th Cir.1994)). . 114 Nev. 1, 953 P.2d 24 (1998) (per curiam). . Id. at 26-27 (citation omitted). . 116 Nev. 598, 5 P.3d 1043 (2000). . Evans, 5 P.3d at 1052. . Am. Heritage Dictionary 759 (2d College Ed. 1985). . Nev.Rev.Stat. § 42.001(3) (1996). . Nev.Rev.Stat. 42.001(1) (1996). . 107 Nev. 651, 817 P.2d 711 (1991). .

Granite Construction, 817 P.2d at 713 (citation and internal quotation mark omitted). . 112 F.3d 1052 (9th Cir.1997). . Coughlin, 112 F.3d at 1054. . Id. at 1055-56. . Id. at 1056. . 114 Nev. 1468, 970 P.2d 98 (1998) (holding disfavored on other grounds in GES, Inc. v. Corbitt, 117 Nev. 265, 21 P.3d 11 (2001)). . Mahlum, 970 P.2d at 113. . Coughlin, 112 F.3d at 1056. . 5 P.3d at 1052. . Thus we do not disagree with our dissenting colleague’s position that the judgment and jury instructions conformed to Nevada law. . See Neibel v. Trans World, 108 F.3d 1123, 1129 (9th Cir.1997) (citing Fikes v.

Cleghorn, 47 F.3d 1011, 1013 (9th Cir.1995)). . See Chuman v. Wright, 76 F.3d 292, 294 (9th Cir.1996). . Swinton v. Potomac Corp., 270 F.3d 794, 802 (9th Cir.2001) (quoting Chang v. Johns-Manville Sales Corp. (In re Asbestos Cases), 847 F.2d 523, 524 (9th Cir.1988), cert. denied, 122 S.Ct. 1609 (2002)). . 517 U.S. 559, 116 S.Ct. 1589, 134 L.Ed.2d 809 (1996). . BMW, 517 U.S. at 568, 116 S.Ct. 1589. Because the Court said the inquiry “begins” with this federalism inquiry, this is where we begin.

We reject the dissent’s assumption that we ought to begin elsewhere, with^the three criteria for excessiveness that the Court sets out subsequently in its opinion. . Id. at 569, 116 S.Ct. 1589. . Id. at 570 n. 14, 116 S.Ct. 1589. . Id. at 571-72, 116 S.Ct. 1589. . Id. at 572-73, 116 S.Ct. 1589. . Id. at 574 n. 21, 116 S.Ct. 1589; see also Smith v. Ingersoll-Rand Co., 214 F.3d 1235, 1253 (10th Cir.2000). .

BMW holds that "no single State could [enact a tort law policy for the entire Nation], or even impose its own policy choice on neighboring States .... [0]ne State's power to impose burdens on the interstate market for automobiles is not only subordinate to the federal power over interstate commerce, but is also constrained by the need to respect the interests of other States.” 517 U.S. at 571, 116 S.Ct. 1589 (internal citations omitted).

The Court held squarely that "[t]he award must be analyzed in the light of [conduct that occurred within Alabama], with consideration given only to the interests of Alabama consumers, rather than those of the entire Nation.” Id. at 574, 116 S.Ct. 1589. We are not free to ignore this holding, nor to decline to follow the reasoning by which the Court reached this result. While Pacific Mutual Life Ins. Co. v. Haslip, 499 U.S. 1, 111 S.Ct. 1032, 113 L.Ed.2d 1 (1991) (intentional fraud in Alabama insurance company) and Browning-Ferris Indus. v.

Kelco Disposal, Inc., 492 U.S. 257, 109 S.Ct. 2909, 106 L.Ed.2d 219 (1989) (interference with contractual relations in limited Vermont market), the cases upon which the dissent relies, dealt with purely instate conduct engaged in by a limited number of in-state actors, BMW addresses exactly the situation in the case we have before us: tor-tious conduct on the part of a nationwide manufacturer, that reflected a nationwide policy, and that caused the same harm in all states.

BMW tells us that we may not, as a matter of federal law, permit one state to punish such nationwide manufacturers for their conduct — actual or hypothetical — in other states. . BMW, 517 U.S. at 573 n. 20, 116 S.Ct. 1589. . 108 F.3d 1123 (9th Cir.1997). . Neibel, 108 F.3d at 1131 (internal quotation marks and citation omitted) (first alteration in original); cf. Johansen v.

Combustion Engineering, Inc., 170 F.3d 1320, 1333 (11th Cir.1999) (embarking on BMW excessiveness analysis only after holding that the conduct being punished, acidic seepage into waterways, "occurred] in a single state and that [the state’s statutes] expressed a strong interest in deterring environmental pollution”); Deters v. Equifax Credit Info. Servs., 981 F.Supp. 1381, 1389-90 (D.Kan.1997) (noting that in action under federal statute, federalism concerns, which must otherwise be addressed, are not implicated in assessing constitutionality of punitive damages award). . Emphasis added. . Cf.

Ingersoll-Rand, 214 F.3d at 1253. . BMW, 517 U.S. at 563, 116 S.Ct. 1589. . The error was not one of formulation of jury instructions, but of substance. The plaintiff has produced extensive evidence of extraterritorial conduct, and in argument urged the jury to punish Ford for the extraterritorial conduct, not just the Nevada conduct. The substantive error was not one of Nevada law, but of federal constitutional law. The pre-Gore cases discussed by the dissent do not touch upon extraterritoriality, because the issue was not before the court in those cases. Gore does, in

section II of the Court’s opinion, which is prior to and separate from

section III, which lays out the three criteria for exces-siveness, reprehensibility, ratio, and sanctions for comparable misconduct. An award is unconstitutional if it violates the principles laid out in either section. The dissent concludes that the award violates the

section III standards. We do not reach

section III, because we conclude that it violates the

section II standards. In some cases, such as Neibel, no extraterritoriality instruction is needed, because it was "clear that only California interests are involved.” 108 F.3d at 1131. Ford's proposed instruction was less than complete as it stood, because, as we have explained, the jury may consider extraterritoriality insofar as it bears on reprehensibility of the conduct. "[S]uch evidence may be relevant to the determination of the degree of reprehensibility of the defendant’s conduct.” BMW, 517 U.S. at 574, n. 21, 116 S.Ct. 1589.

Where we differ with the dissent is that there is no getting around the proposition that, one way or another, whether by the nature of the evidence that comes in, the arguments of counsel made to the jury, the instructions, or otherwise, under

section II of BMW, a state cannot impose punitive sanctions for conduct that affected other states but had no impact on the plaintiff’s state or its residents. "Alabama does not have the power, however, to punish BMW for conduct that was lawful where it occurred and that had no impact on Alabama or its residents.” Id. at 572-73, 116 S.Ct. 1589. In some cases the distinction between using the evidence as it bears on reprehensibility but not as a measure of damages might be so gossamer as to be difficult for a jury to apply, but in others the significance of the distinction will be quite clear.

For example, where the jury wants to punish a national manufacturer for something it did nationally, it might well be tempted to multiply the appropriate amount of punitive damages for one plaintiff by the number of hypothetical plaintiffs the jury thinks likely to be harmed. That can't be done, under BMW, where the award would thereby include amounts based on out-of-state victims. . Id. at 573 n. 20, 116 S.Ct. 1589. . Continental Trend Res., Inc. v. OXY USA Inc., 101 F.3d 634, 636-37 (10th Cir.1996); but see Owens-Corning Fiberglas Corp. v.

Ballard, 739 So.2d 603, 606 (Fla.App. 4th Dist. 1998) (disagreeing with Continental Trend’s

interpretation of BMW on grounds that “where the defendant's conduct is considered tortious in all 50 states ... the same due process concerns implicated in BMW do not arise”). . Continental Trend, 101 F.3d at 637. . BMW, 517 U.S. at 570 n. 14, 116 S.Ct. 1589. . Nev.Rev.Stat. § 42.005(2) (1996). . Alaska Stat. § 09.17.020(g) (Lexis 2000). . Alaska Stat. § 09.17.020(j) (Lexis 2000). . BMW, 517 U.S. at 585, 116 S.Ct. 1589. . See, e.g., Baker v. Exxon Corp. (In re Exxon Valdez), 270 F.3d 1215, 1244 (9th Cir.2001). .

See, e.g., BMW, 517 U.S. at 593, 116 S.Ct. 1589 (Breyer, J. concurring); In re Exxon Valdez, 270 F.3d at 1244. . BMW, 517 U.S. at 572, 116 S.Ct. 1589 (elision is of word "lawful”). . See Continental Trend, 101 F.3d at 637. . See, e.g., Ace v. Aetna Life Ins. Co., 40 F.Supp.2d 1125, 1133(D.Alaska 1999) (reading "BMW case broadly enough to suggest that Alaska must leave some room within which the other states can exercise their own interests in defining the precise extent of and in deterring wrongful conduct”); Hampton v.

Dillard Dep’t Stores, Inc., 18 F.Supp.2d 1256, 1276 (D.Kan.1998) ("the punitive damage award must relate to conduct occurring within the state”); Geressy v. Digital Equip. Corp., 950 F.Supp. 519, 521-22 (E.D.N.Y.1997) (noting that punitive damage awards are limited by "principle of our federal system that state legislation, state policy, and judicial development of state law can only be directed at activity within the state.”); Ford Motor Co. v.

Ammerman, 705 N.E.2d 539, 561 (1999) (assuming no jurisdiction condones sale of defective products, “[n]onetheless it is up to each jurisdiction to make that determination for itself.... Thus any punitive damages award should be limited to protecting this State's consumers.”). . Geressy, 950 F.Supp. at 521. . Id. at 524. . Boatmen's Trust Co. v. St. Paul Fire & Marine Ins. Co., 995 F.Supp. 956, 962 (E.D.Ark.1998). . Birchler v. Gehl Co., 88 F.3d 518, 521 (7th Cir.1996). . Anderson v. Nissan Motor Co., 139 F.3d 599, 602 (8th Cir.1998). . McLennan v. Am. Eurocopter Corp., 245 F.3d 403, 430 (5th Cir.2001). .

DeSantis v. Frick Co., 745 A.2d 624, 627-32 (2000) (defect must have existed when product left manufacturer's hands); Sullivan v. Modern Group Ltd., 46 Pa. D. & C.4th 524, 530-31 (2000). . Hiner v. Deere & Co., 161 F.Supp.2d 1279, 1290 (D.Kan.2001). . 517 U.S. at 573 n. 19, 116 S.Ct. 1589 (citation and internal quotation marks omitted). . 532 U.S. 424, 121 S.Ct. 1678, 149 L.Ed.2d 674 (2001). . Cooper v. Firestone Tire & Rubber Co., 945 F.2d 1103, 1107 (9th Cir.1991); see also Bird v. Glacier Elec. Coop., Inc., 255 F.3d 1136, 1145 & n. 16 (9th Cir.2001) (citing Standard Oil Co. v.

Perkins, 347 F.2d 379, 388 (9th Cir.1965)). . Those standards are: (

a) whether there is a reasonable relationship between the punitive damages award and the harm likely to result from the defendant's conduct as well as the harm that actually has occurred; (

b) the degree of reprehensibility of the defendant’s conduct, the duration of that conduct, the defendant’s awareness, any concealment, and the existence and frequency of similar past conduct; (

c) the profitability to the defendant of the wrongful conduct and the desirability of removing that profit and of having the defendant also sustain a loss; (

d) the "financial position” of the defendant; (

e) all the costs of litigation; (

f) the imposition of criminal sanctions on the defendant for its conduct, these to be taken in mitigation; and (

g) the existence of other civil awards against the defendant for the same conduct, these also to be taken in mitigation. Haslip, 499 U.S. at 21-22, 111 S.Ct. 1032.