Opinion
This case involves an appeal and a cross-appeal after the trial court granted in part and denied in part defendant Ford Motor Company’s posttrial motions. The court denied defendant’s motion for new trial on liability and compensatory damages; denied defendant’s motion to reduce punitive damages as excessive; and denied defendant’s motion for judgment notwithstanding the verdict (JNOV). From these orders defendant appeals. The court granted defendant’s motion for new trial on punitive damages grounded on juror misconduct; from this order plaintiffs appeal.
We will affirm the judgment for compensatory damages as modified by the trial court. We will affirm the trial court’s denial of the JNOV motion. We will conclude the punitive damages award was not excessive. However, we will conclude the trial court erred in granting the new trial motion on punitive damages grounded on juror misconduct.
Facts and Procedural History
Ramon Romo bought a used 1978 Ford Bronco in 1992. On June 20,. 1993, Ramon’s son, plaintiff Juan Ramon Romo, was driving the Bronco; other members of the family—Ramon, his wife, Salustia, and his children Ramiro, Evangelina, and Maria—were passengers. Juan and Ramon were seated in the front, wearing seat belts. Salustia and Ramiro were in the back seat, also belted. Evangelina and Maria were asleep in the rear of the Bronco, unbelted.
Juan was involved in an accident and the Bronco rolled over several times. The 1978 Bronco had a steel roof over the front one-third of the passenger compartment. The rear two-thirds was made of fiberglass. As the Bronco rolled, the steel roof collapsed, killing Ramon Romo. The fiberglass roof broke loose, striking and killing Salustia and Ramiro Romo. Evangelina and Maria Romo were thrown from the car and injured. Juan remained in the car and was injured.
Juan Romo, individually and as administrator of the estates of the deceased family members, Evangelina Romo, and Maria Irene Romo through her guardian ad litem Juan Romo, sued defendant on theories of products liability and negligence; they sought compensatory and punitive damages.
*1125 Defendant did not request a bifurcated trial on the amount of punitive damages. (See Civ. Code, § 3295, subd. (d).)
The unified trial lasted from March 1999 through July 1999. At the conclusion of trial, the matter was submitted to the jury for deliberation on all issues. The jury deliberated for a week and returned a verdict for plaintiffs and against defendant. The jury awarded $6,226,793 in compensatory damages and $290 million in punitive damages. In accordance with the jury’s allocation of fault (the jury concluded plaintiff Juan Romo was 10 percent at fault, defendant was 78 percent at fault, and the driver of a second vehicle was 12 percent at fault) and based on certain technical issues concerning damages awarded to the estates, the court reduced the compensatory damages to $4,935,709.10. This reduction is not an issue on appeal. Based on its findings of juror misconduct, which we will discuss in more detail below, the court granted defendant’s motion for new trial on the issue of punitive damages.
The parties filed timely notices of appeal.
Discussion
I. Plaintiffs’ Cross-appeal of Juror Misconduct Determination
Plaintiffs have appealed, primarily, from the court’s order granting defendant’s motion for new trial on punitive damages. Second, plaintiffs contend the “jury’s punitive damage award was not excessive as a matter of California or Federal law.” The trial court did not grant the new trial based on excessiveness of damages; instead, it denied the new trial motion “on all other grounds specified” in defendant’s notice of motion for new trial. Plaintiffs, therefore, were the prevailing parties on this issue. We address the issue of excessiveness of damages only in defendant’s appeal, in part II of this opinion. In part I, we address the order for new trial on punitive damages based on the trial court’s finding of juror misconduct.
A. Standard of Review
As recently reiterated by the Supreme Court, an order granting a new trial under Code of Civil Procedure section 657 “ ‘must be sustained on appeal unless the opposing party demonstrates that no reasonable finder of fact could have found for the movant on [the trial court’s] theory.’ [Citation.]”
(Lane v. Hughes Aircraft Co.
(2000)
In the context of an order granting or denying a new trial motion on the basis the verdict is “contrary to law or evidence” (Pen. Code, § 1181, subd. 6), the trial court has broad discretion “and there is a strong presumption that it properly exercised that discretion.”
(People v. Davis
(1995)
When the trial court’s order is based on a finding of juror misconduct, however, our deference to the trial court’s order is more limited. “We accept the trial court’s credibility determinations and findings on questions of historical fact if supported by substantial evidence. [Citations.] Whether prejudice arose from juror misconduct, however, is a mixed question of law and fact subject to an appellate court’s independent determination.”
(People
v.
Nesler
(1997)
Defendant contends the standard of review of the order granting a new trial is abuse of discretion. Quoting
Malkasian v. Irwin
(1964)
Reliance by defendant upon
Malkasian
is misplaced: that case involved misconduct of counsel during the trial.
(Malkasian
v.
Irwin, supra,
61 Cal.2d at pp. 745-747.) In evaluating the misconduct of counsel committed in the trial court’s presence, that court certainly is in a better position than the appellate court to determine the net effect of such misconduct in the overall context of the trial. By contrast, in the case of juror misconduct occurring outside the presence of the trial court, the Supreme Court specifically has instructed that the reviewing court must independently apply an objective standard to determine whether the misconduct was prejudicial.
(In re Hamilton
(1999)
Likewise, defendant’s reliance on
Lane
v.
Hughes Aircraft Co., supra,
The abuse of discretion test is inapplicable in cases of juror misconduct and, in light of the more recent Supreme Court statements on this point (see also
People v. Cromer
(2001)
B. The Juror Declarations
1. The Television Program
Both parties submitted juror declarations in connection with the posttrial motions. The declarations established, and the trial court found, that one juror, Ms. S., brought to the attention of the remaining jurors a segment of the television program 60 Minutes II dealing with fires in older Ford Mustangs. Uncontested portions of the declarations established that S.’s discussion of the television program occurred during the jury’s discussion of the issue of malice, after all issues concerning liability and compensatory damages had been resolved in plaintiffs’ favor by unanimous vote of the jury.
Uncontested portions of the declarations established that S. reported to the jury that in the course of the program former Ford president Lee Iaccoca stated that Ford had a policy of contesting or settling lawsuits involving the Mustangs because it was less costly than recalling the vehicles and repairing them. Uncontested portions of the declarations established that S. mentioned the program on one occasion and that her discussion of the program took less than five minutes. Uncontested portions of the declarations established that the jury foreperson, Ms. B., told S. the program was not evidence in the present case and should not be discussed further. S. told the other jurors she agreed and the matter was never raised again..
The trial court made no findings contrary to these undisputed portions of the declarations, although the trial court did find that other persons interviewed in the 60 Minutes II segment, and not Ford representatives, actually made the claims concerning the decision not to recall the Mustangs.
Conflicting portions of the declarations centered around whether S. said she had watched the program or had been told about it. Further, the declarations conflicted about the context in which S.’s comments were made. The *1128 declarations agreed the jurors were in conflict over the type of evidence necessary to show malice, some contending that documentary evidence was necessary. Most declarations stated that S. brought up Iaccoca’s purported statements as an example of the manner in which nondocumentary evidence could demonstrate malice as a general matter. The trial court interpreted two other declarations to state that S. proffered Iaccoca’s statements as evidence of malice in the present case. The trial court did not resolve the conflicts in the evidence, finding only that the declarations “alleged” that S. had made the various statements attributed to her.
The declarations also established that at some point after S. made her statements about the television program, several jurors expressed the belief that the testimony of a particular Ford witness (Mr. Darold) was highly relevant—and damaging to defendant—on the issue of malice. The jury requested a reading of a portion of that testimony. The morning of the read-back, the trial court also reinstructed the jury that it was not to consider outside evidence in connection with the present case. (The parties were concerned about widespread news reports concerning punitive damages awarded by a jury in a case against General Motors.) Only after this reinstruction and after the read-back of testimony did the jury vote, nine to three, in favor of a finding of malice.
2. The Dream
The declarations agreed that a juror, Ms. Z., recounted to the jury a dream she said she had the night before. In a scene that kept repeating during the dream, a Ford Bronco rolled over, sequentially killing each of her children and then the children of each of the other jurors. While the Bronco rolled, Z. said, Ford representatives stood by, chanting, “Where’s the proof?” She recounted the dream only one time, during deliberations on the issue of malice; the dream was not discussed at any other time. Z.’s telling of the dream occurred on the fourth day of deliberations, after the votes on liability and compensatory damages, but before the vote on malice.
The declarations conflicted about whether Z. described the dream as an “omen.” The trial court did not resolve the conflict in the evidence. Although defendant asserts the trial court “specifically credited evidence that [Juror Z.] . . . relied (and encouraged her fellow jurors to rely) on that dream in determining Ford’s liability for punitive damages,” the trial court made no such finding. Rather, the court simply set forth the evidence: “This same juror allegedly also told her fellow jurors that her dream was an ‘omen,’ thereby suggesting that it had some sort of mystical significance regarding hów to vote in the case.”
*1129 Even if we disregard the trial court’s use of the word “allegedly”—which is the antithesis of “specifically credit[ingj” the evidence 1 —there is no statement in any of the declarations that Z. claimed her dream had mystical significance. The only statement even marginally related to such a claim was this statement in B.’s first declaration: “In other words, Juror [Z.] personalized her dreams to include, not only her own children, but other jurors’ children, and as above stated, stated that [the jury’s] job was not about following the law, but ‘saving children.’ ” This statement does not constitute substantial evidence that Z. attached mystical significance to the dream and used it as the basis—in disregard of the law and the evidence—for her votes on punitive damages.
3. Prejudgment by a Juror
Although the declarations conflicted greatly on the details, most of the declarations agreed that Z. repeatedly said during deliberations that the jury must “save the babies” by finding defendant liable.
Juror declarations submitted by defendant tied this statement to statements by Z. to the effect that saving lives was too important and the jury did not need to follow the law to accomplish this goal. Of the three declarations submitted by defendant, one stated Z. made these statements “many times during the various days of jury deliberation” and “on several of the deliberations days” (Juror H.), one declaration stated Z. made the statements on the first day of deliberations only (Juror W.), and one declaration stated she made the statements on the first two days of deliberations only (Juror B.).
Juror declarations submitted by plaintiff stated that Z. expressed particular concern because the rear roof of the Bronco was made of fiberglass, affording the least protection in a rollover accident, yet people tended to put their babies in the back seat of a car to try to give them added safety. She said defendant would not do anything to remedy the situation if the jury did not punish it. Z. and other declarants specifically denied Z. said the jury did not need to follow the law. Four of the six juror declaratipns submitted by plaintiff state that early in the deliberations there was a discussion about the need to follow the law, that the foreperson (B.) reiterated that the jury had to follow the law, and that all jurors agreed they needed to and would follow the law.
Various of the declarations alleged and denied that Z. said she had made up her mind about the case during the trial and that the jury should treat “as evidence” the statements of one of plaintiffs’ attorneys.
*1130 The trial court did not expressly resolve the conflicts in the evidence, although the court did state its conclusion that “[j]uror misconduct occurred by virtue of the injection of external evidence and influences, prejudgment of the case and a refusal to follow the law. The alleged misconduct occurred on the second day of deliberations.”
C. The Trial Court’s Conclusion
In its written order granting a new trial on punitive damages, the trial court recounted in detail the contents of the three juror declarations submitted by defendant. The court then wrote:
“The Court finds that the counter-declarations submitted by plaintiffs fail to rebut the presumption of prejudice in this case. Juror misconduct occurred by virtue of the injection of external evidence and influences, prejudgment of the case and a refusal to follow the law. The alleged misconduct occurred on the second day of deliberations. The jury voted 12-0 on the issues of liability and compensatory damages. However, the jury voted 9-3 on the issues of malice and the amount of punitive damages.
“Because there was a bare majority on the issues of malice and the amount of punitive damages and because the standard of proof as to malice is so high, the Court finds the deliberative process on these two (2) issues was presumptively infested with passion and prejudice as to render the verdict on punitive damages fundamentally suspect. If not singly, then in combination, the juror misconduct described above was presumptively prejudicial and such presumption has not been rebutted.
“Despite admonitions from the Court, in essence, not to consider anything but the facts and the law, and despite other jurors’ admonitions not to consider the external evidence and influences discussed above, the jury was being told not to think of the proverbial pink elephant. The damage was presumptively done.”
D. The Issue of Presumptions
Jury deliberations are secret while they are occurring. No verbatim transcript or other record of the deliberations normally exists. Declarations seeking to reconstruct deliberations after the fact may be colored by the jurors’ natural inclination to protect or attack the process that resulted in the verdict, depending on whether the juror agreed or disagreed with the verdict. (See
Weathers v. Kaiser Foundation Hospitals
(1971)
Common experience requires that we recognize that while the “jury system is . . . legally fundamental ... [it is] also fundamentally human.”
(People v. Marshall
(1990)
Faced with the twin facts that jurors are allowed great freedom in their conduct of deliberations and that a court can never know exactly what influences resulted in a particular verdict, our judicial system has established certain presumptions for reviewing allegations of juror misconduct.
Jurors ordinarily are presumed to have followed the court’s instructions.
(People
v.
Sanchez
(2001)
On the other hand, “[j]uror misconduct, such as the receipt of information about a party or the case that was not part of the evidence received at trial, leads to a presumption that the defendant was prejudiced thereby and may establish juror bias.”
(People v. Nesler, supra,
The presumption of prejudice in a civil case is rebutted if the reviewing court reaches one of three conclusions: (1) the record establishes the absence
*1132
of prejudice; (2) a review of the entire record shows there is no reasonable probability of actual harm to the complaining party under the constitutional standard of
People
v.
Watson
(1956)
The interaction of the two foregoing presumptions is seen in
People v. Marshall, supra,
E. Analysis
As noted, the trial court did not expressly resolve the many factual conflicts presented by the various declarations. Implicit in the court’s conclusion, however, is a finding that the declarations submitted by plaintiffs were credible at least to the extent they described the verbal response of other jurors, the timing of the remarks by S. and Z., and the uncontradicted fact that the remarks were not the subject of discussion after the read-back of the testimony of the Ford employee, Darold. Thus, the court concluded: “Despite admonitions from the Court, in essence, not to consider anything but the facts and the law, and despite the other jurors’ admonitions not to consider the external evidence and influences discussed above, the jury was being told not to think of the proverbial pink elephant. The damage was presumptively done.” The only claims that other jurors admonished Jurors S. and Z. were contained in the declarations submitted by plaintiffs.
*1133 Despite this finding concerning jury self-admonition, it appears to us the trial court applied only the presumption that jury misconduct is prejudicial and, without any evidentiary basis for doing so, failed to consider and apply the presumption that jurors have followed the instructions they were given. In our independent review of the issues of prejudice and rebuttal of the presumption of prejudice, we conclude the record provides no basis to reject the normal presumption the jury followed the instructions. Accordingly, as we explain, we conclude the presumption of prejudice is rebutted.
The declarations establish that S. made her comments about the 60 Minutes II segment on Thursday or Friday. Also prior to the weekend, the jury asked the court for a read-back of a portion of Darold’s testimony concerning Ford’s state of knowledge. Over the course of the ensuing weekend, there was substantial news coverage of a massive punitive damages verdict rendered in a case against General Motors during the previous week. As a result of concern that the jury might be influenced by the size of the General Motors verdict, the court—without any explanation to the jury about why it was doing so—instructed the jury prior to the commencement of deliberations on the following Monday, July 12, 1999: “Ladies and gentlemen, you’re going to resume your deliberations here shortly, but just a reminder to you: During the course of the trial, the Court admonished you that you are not to read or listen to or pay attention to any news accounts concerning this case; that you are to decide this case based on the law and the facts of the evidence that was presented to you during the course of the trial. The Court wanted to repeat that admonition to you that again you are to decide this case based on the law, the facts and the evidence that are presented to you during the course of the evidence in the trial. You are not to read or listen to or consider, if you have, taken [szc] into account any newspaper articles concerning this case, any radio accounts or any television or any other media accounts concerning this case or any other case. You are to decide this case based on the facts and the law and the evidence presented during the course of this trial; okay?”
In addition, there is no conflict in the evidence showing the foreperson of the jury admonished S. that the
60 Minutes II
segment could not be used as evidence in the present case, nor is there conflict in the evidence showing S. expressly agreed with this admonition. There is no conflict in the evidence showing that the
60 Minutes II
issue was never mentioned again after this exchange between the foreperson and S. Because the jury is presumed to have followed the court’s subsequent instruction to decide the case only on the evidence and law presented, and because there is no evidence to overcome that presumption, we conclude the presumption of prejudice from mention of the
60 Minutes II
segment is rebutted by the record.
(People
v.
Marshall, supra,
*1134
Similarly, Z.’s dream was never the subject of discussion after the court’s admonition of Monday, July 12, 1999. We digress to note that defendant has not cited any cases for the proposition that having or talking about a dream while serving on a jury constitutes misconduct. Having a dream, in itself, cannot be considered the receipt of extrinsic evidence by the dreamer, since the dream is entirely an internal affair. The report of the dream—Z.’s statements to her fellow jurors—seems to us well within the realm of permissible rhetorical devices, merely another way of saying that the speaker finds the evidence very persuasive and that she fears other similar accidents if the product were not to be recalled by the manufacturer. Certainly such a statement in the latter terms would not constitute misconduct. No reason occurs to us why the use of the dream formulation to increase the rhetorical impact of the statements would be any less permissible. (Cf.
Iwekaogwu v. City of Los Angeles
(1999)
To return to our main point: even assuming the dream statements constituted misconduct, they were brief, the dream was not claimed to be evidence, and the dream was not discussed after the July 12, 1999, admonition. Because the jury is presumed to have followed the instruction to decide the case only on the evidence and law presented, and because there is no evidence to overcome that presumption, we conclude any presumption of prejudice from the dream statements is rebutted by the record.
(People v. Marshall, supra,
The greatest level of conflict in the declarations concerns the claims that Z. said she had prejudged the case and said the jury did not “need to follow the law” because of an overriding responsibility to “save the babies.” Those jurors who claimed Z. voiced an intention to disregard the law all said she made these statements on the first day or the first couple of days of deliberations. It was during this time that the jury deliberated the issues of liability and compensatory damages. There is no indication in any of the declarations that Z. made similar statements during the more contentious punitive damages deliberations. In fact, it is uncontradicted that the statements ceased well before the court’s July 12, 1999, admonition to decide the case on the evidence alone. Because the jury is presumed to have followed the instruction to decide the case only on the evidence and law presented, and because there is no evidence to overcome that presumption, we conclude the presumption of prejudice from Z.’s possible expressions of prejudice and refusal to follow the law is rebutted by the record.
(People v. Marshall, supra,
A similar issue arose in
People v. Harper
(1986)
*1135 interpreting the instructional definition of murder. A juror brought in a dictionary and read to the other members the dictionary definition of murder. That juror “was immediately cautioned by other members of the jury that his definition could not be considered.” (Id. at p. 1428.) Seeking further instruction from the trial court, the jury sent out a note that disclosed that the juror had consulted the dictionary. (Ibid.) The trial court then brought out the jury, explained generally why dictionary definitions are not applicable in criminal prosecutions, and admonished the jury to “ ‘put aside any discussion or consideration’ ” of the dictionary definition. (Id. at p. 1429.)
The appellate court held: “Absent evidence to the contrary, a jury is presumed to follow the instructions of the trial court. Having been given a specific admonishment in light of misconduct by one of their members, it seems probable the jurors would more closely adhere to the instructions of the trial court and thus reduce the chance for prejudice.”
(People v. Harper, supra,
186 Cal.App.3d at pp. 1429-1430.) Agreeing with the trial court’s conclusion, the reviewing court stated: “[I]t was unnecessary to rely on a presumption the jurors . . . followed the admonition when the declarations [of the jurors] made clear there was no further discussion of the dictionary definition.”
(Id.
at p. 1430.) The court concluded the presumption of prejudice arising from the juror’s misconduct had been rebutted.
(Ibid.;
see also
People
v.
Pinholster
(1992)
In the present case, of course, the jury did not inform the trial court that there had been reference by a juror to a television program and an appeal by another juror to the overriding moral necessity of saving the lives of children. Nevertheless, the court fortuitously instructed the jury to confine its deliberations and consideration of the case solely to the evidence presented at trial. The instruction did not mention the General Motors verdict that had caused the parties to request it; for all the members of this jury knew, the court somehow had ascertained that extraneous materials were being mentioned in the jury room. In any event, the record is clear that none of the misconduct was repeated in any manner after the court’s admonition. As in
People v. Harper, supra,
Finally, we consider whether the evidence establishes that, even if the record rebuts the presumption of prejudice as to the jury as a whole, there is a substantial likelihood that a juror necessary to the punitive damages verdict prejudged the case. (See
Province
v.
Center for Women’s Health & Family Birth, supra,
20 Cal.App.4th at pp. 1678-1679.) “For a juror
*1136
to prejudge the case is serious misconduct.”
(Clemens v. Regents of University of California
(1971)
In this case, the trial court impliedly found that Juror Z. stated to the other jurors that she had made up her mind about the case prior to the completion of all the testimony.
For reasons similar to those stated above, however, we find that Z.’s initial statements concerning readiness to decide the case without regard to the law does not rise to the level of prejudicial misconduct found in the foregoing cases. Rather, the record establishes that while early in the deliberations Z. may have misunderstood the nature of the jury’s function, the other jurors—particularly the foreperson, identified as a deputy district attorney—admonished her concerning the necessity to follow the law as instructed by the court. None of the juror declarations contend Z. persisted in a refusal to consider the law and the evidence, and several declarations— including Z.’s—state that she affirmatively agreed that the jury needed to follow the law. On this record, it is not substantially likely that Z. prejudged the case and failed to decide the case on the law and evidence presented at trial.
(In re Hamilton, supra,
F. Conclusion
Defendant urges that we adopt what amounts to a rigid rule that prejudicial misconduct cannot be cured either by jury self-admonition or by admonitions from the trial court. Such a contention ignores the very purpose of permitting and requiring jury deliberations: through group discussion of the law and the evidence, our common law system trusts that jurors who express wrong ideas about the evidence, the law, and their duty as jurors will be guided to a correct view of the case. In the absence of an opportunity for jurors to express such wrong conceptions and thereafter change their thinking, a jury trial might just as well conclude with the submission of ballots from the jury box at the close of the case.
Despite defendant’s consistent attempts to portray this as a case of jurors run amok, the misconduct in the present case does not begin to approach the
*1137
pervasive and insistent misconduct that resulted in reversal in such cases as
People
v.
Nesler, supra,
II. Defendant’s Appeal
Defendant appeals from the trial court’s order denying its motion for JNOV on the issue of punitive damages. Defendant also appeals from the trial court’s denial of defendant’s new trial motion on liability and compensatory damages. Finally, defendant contends the punitive damages award is the “grossly excessive product of bias, passion and prejudice.”
We will address these issues in the following order. First, we will consider whether substantial evidence supports the jury’s finding on malice, the linchpin of its punitive damages verdict. Next, we will address the issue of excessiveness of punitive damages. Because we have found there was no prejudicial jury misconduct, the claim that the trial court erred in denying a new trial on liability and compensatory damages—a claim premised on the same allegations of jury misconduct—is moot.
A. Substantial Evidence
Defendant contends “plaintiffs failed as a matter of law to prove that Ford can be held liable for punitive damages.” “This argument calls for our determining whether substantial evidence supported the jury’s verdict.”
(Hardison v. Bushnell
(1993)
1. Standard of Review
“ ‘Where findings of fact are challenged on a civil appeal, we are bound by the “elementary, but often overlooked principle of law, that . . . the power of an appellate court begins and ends with a determination as to whether there is any substantial evidence, contradicted or uncontradicted,” to support the findings below.
(Crawford
v.
Southern Pacific Co.
(1935)
On appeal from an order denying judgment notwithstanding a jury’s verdict, we independently apply the same standard as the trial court, that is, the substantial evidence test stated above.
(Osborn v. Irwin Memorial Blood Bank
(1992)
Our standard of review is no different where, as here, the challenge to the sufficiency of the evidence arises in the context of a punitive damages verdict. Defendant seeks support for a heightened standard of review by relying upon the maxim that California law “disfavors punitive damages.” That reliance is unavailing.
Defendant’s premise is correct, in a limited way: When the issue is whether to permit administrative agencies to award such damages in the absence of a legislative mandate (see
Dyna-Med, Inc.
v.
Fair Employment & Housing Com.
(1987)
Where the Legislature has provided for such damages under specified conditions, however, it is the function of the judiciary to review a jury verdict pursuant to the law as written—not to favor or disfavor, approve or disapprove, the Legislature’s choice in providing for such damages. “The judiciary, in reviewing statutes enacted by the Legislature, may not undertake to evaluate the wisdom of the policies embodied in such legislation; absent a constitutional prohibition, the choice among competing policy considerations in enacting laws is a legislative function.”
(Superior Court v. County of Mendocino
(1996)
Defendant also suggests our review of the verdict and of the order denying the JNOV motion must be elevated beyond ordinary substantial evidence *1139 review because the punitive damages statute (Civ. Code, § 3294) requires proof of malice by clear and convincing evidence. This contention is wrong if, by it, defendant means to imply that we give less deference, to the facts and inferences found by the trier of fact.
Defendant’s contention, understood in this way, ignores the fact that courts routinely apply the substantial evidence standard in criminal cases, where the burden of proof is beyond a reasonable doubt. (See, e.g.,
People v. Lewis
(2001)
We reiterate, then, that in reviewing a punitive damages verdict, just as in reviewing a criminal verdict or an ordinary civil verdict, we give full deference to the trier of fact. Thus, even though the Legislature added a “clear and convincing evidence” requirement to the punitive damages statute, it did not alter “the standard of our review (all we are required to find is substantial evidence to support a determination by clear and convincing evidence [citation]) . . . .”
(Tomaselli v. Transamerica Ins. Co.
(1994)
2. Corporate State of Mind
(11a) Defendant contends the evidence was insufficient to prove it acted with malice in designing and manufacturing the 1978 Bronco or in failing to warn consumers about the absence of rollover- protection. Underlying this argument is a fundamental misconception of the required proof, stated in defendant’s reply brief as follows: “Section 3294 [authorizing punitive damages] makes absolutely clear that plaintiffs were required to introduce clear and convincing proof that at least one particular Ford employee, officer, director or managing agent had the requisite malicious state of mind in 1978.” (Italics in original.)
Since 1980, Civil Code section 3294, subdivision (b), has provided, in relevant part: “With respect to a corporate employer, the advance knowledge and conscious disregard, authorization, ratification or act of oppression, fraud, or malice must be on the part of an officer, director, or managing agent of the corporation.” The purpose of the amendment was to clarify the existing common law requirement that corporate malice occur at the level of “employees [who] exercised substantial discretionary authority over decisions that resulted in an ‘ad hoc formulation of policy,’ . . .”
(White
v.
Ultramar, Inc.
(1999)
*1140 In adopting the amendment, the Legislature sought language that would express the requirement that punitive damages liability for employers requires more direct involvement by the employer than ordinary respondeat superior liability, but also sought to make clear that such liability did not attach only to the acts of those “employed in a managerial capacity.” (See White v. Ultramar, Inc., supra, 21 Cal.4th at pp. 571, 576.) The adopted language reflects a legislative intent “that principal liability for punitive damages not depend on employees’ managerial level, but on the extent to which they exercise substantial discretionary authority over decisions that ultimately determine corporate policy. Thus, supervisors who have broad discretionary powers and exercise substantial discretionary authority in the corporation could be managing agents.” (Id. at pp. 576-577.)
In
White v. Ultramar, Inc.,
the Supreme Court concluded that the present version of Civil Code section 3294, subdivision (b), sought to preserve the pragmatic stance adopted in
Egan
v.
Mutual of Omaha Ins. Co.
(1979)
In most of the cases in which the “managing agent” issue has resulted in reversal of a punitive damages award, initial liability arises from a particular tortious act of an employee of the corporation. (See, e.g.,
Cruz
v.
HomeBase
(2000)
There is no requirement that the evidence establish that a particular committee or officer of the corporation acted on a particular date with “malice.” A corporate defendant cannot shield itself from liability through layers of management committees and the sheer size of the management *1141 structure. It is enough if the evidence permits a clear and convincing inference that within the corporate hierarchy authorized persons acted despicably in “willful and conscious disregard of the rights or safety of others.” (See Civ. Code, § 3294, subd. (c)(1).)
A plaintiff may satisfy the “managing agent” requirement of Civil Code section 3294, subdivision (b), through evidence showing the information in the possession of the corporation and the structure of management decision-making that permits an inference that the information in fact moved upward to a point where corporate policy was formulated. These inferences cannot be based merely on speculation, but they may be established by circumstantial evidence, in accordance with ordinary standards of proof.
Defendant contends, in addition, that plaintiffs failed to prove defendant engaged in “despicable” conduct within the meaning of Civil Code section 3294, subdivision (c)(1). This argument focuses on individual acts and decisions of defendant’s employees, while complaining that plaintiffs attempt to “create a composite picture of alleged ‘maliciousness’ by imputing to some fictional monolith bits and pieces of supposed knowledge of many different people on many different topics through many different decades.”
Stripped of pejorative innuendo, defendant clearly is a “fictional monolith”: corporations are legal constructs—legal fictions
(Francis T. v. Village Green Owners Assn.
(1986)
Defendant also asserts that plaintiffs are “unable to present evidence of any ‘malicious or despicable’ conduct with respect to the design and production of the 1978 Bronco.” This ignores the fact that the design and production of the vehicle was the despicable conduct: we think it obvious that putting on the market a motor vehicle with a known propensity to roll over and, while giving the vehicle the appearance of sturdiness, consciously deciding not to provide adequate crush protection to properly belted passengers (in the words of a corporate memo introduced in evidence, “penalizing” passengers for wearing a seatbelt) constitutes despicable conduct. Such conduct could kill people. The question is not whether the conduct, if it occurred, was despicable, the question is whether there is sufficient evidence from which a rational trier of fact could find that the knowing conduct occurred.
Defendant’s attack on the sufficiency of the evidence includes discussions of the evidence of a manufacturing defect and of defendant’s failure to *1142 include rollover warnings with the 1978 Bronco. We will not examine this evidence, since plaintiffs did not rely on it to prove malice. Instead, we will turn directly to the third component of plaintiffs’ case, proof of a design defect.
Attacking the sufficiency of the evidence of design defect, defendant contends plaintiffs’ own evidence showed “conclusively” that negligent manufacture of the Bronco, not negligent design, caused the deaths and injuries involved in this case. Had it not been for defective, welds along the roof pillar, “the metal portion of the roof would not have crushed,” according to defendant; “if the roof had not crushed in the entire Romo family . . . would have ‘walked away’ from the accident.” This claim wholly ignores the requirement that we view the evidence in the light most favorable to the verdict, as we will explain.
Plaintiffs did present evidence that the Bronco was not only poorly designed but that it also was negligently manufactured. Plaintiffs presented testimony from a metallurgy/materials-failure expert, Dean Jacobson, to the effect that the front roof pillar was negligently welded. And he testified that those welds failed and caused the passenger side of the Bronco’s roof to “splay out” as it collapsed. But he also testified that if the welds had been good, only the passenger side of the steel roof “would have stayed up. It would have mitigated the amount of intrusion on the right side [of the steel roof] and I believe basically the right side would have stayed up.”
Jacobson did not testify that, in the absence of faulty welds, the entire Romo family would have walked away from the accident. To the contrary, defendant objected to Jacobson’s expressing any opinions at all about the cause of the passengers’ injuries, since he was not a biomechanics expert. However, Jacobson testified that the rear roof—the molded fiberglass roof— “would have been gone” even if the welds had been in accordance with the design specification. Of equal importance, Jacobson testified that if defendant had installed the type of integrated roll bar it initially planned for the 1978 Bronco and had installed the type of steel roll bar at the rear of the fiberglass roof that defendant did include on the 1980 Bronco, the roof would not have crushed despite the negligent welds. Thus, in essence, Jacobson testified that the design defects were a sufficient independent cause of the steel roof’s collapse during the rollover.
Defendant’s quotation from the record concerning the “entire Romo family” walking away from the accident if the roof did not collapse came from a different witness, plaintiffs’ biomedical engineering expert, Martha Bidez. Her testimony had nothing to do with the causes for the collapse of the roof. *1143 Instead, she took the sequence of events established by the accident reconstruction expert and opined concerning the manner in which the various victims’ injuries occurred during the course of the accident. For example, after the witness described the mechanism by which the minor victims were injured, she was asked, “Why didn’t they sustain skull fractures like other members of the family?” Bidez answered, “The children in the back did not sustain skull fractures because the roof did not crush in on top of them and the roof did not strike them in the head. ftO The roof struck their mother and their brother, and [the girls] were able to simply be deposited out on the highway. There was no roof crush; that is what caused the death of their father.”
To provide a summarization of Bidez’s testimony, counsel asked, “If the roof doesn’t crush in, what happens to the members of this family?” Bidez answered, “If the roof doesn’t crush in . . . the Romo family stays intact. Every one of the family members . . . walk away from this collision.”
We have engaged in this rather extended exposition of the evidence to demonstrate a point defendant repeatedly ignores in its briefs: substantial evidence review requires examination of all the evidence, not just selective portions of the evidence favorable to the appellant, and it requires evaluation of that evidence in the light most favorable to the verdict, not an interpretation that ignores the inferences drawn by the trier of fact.
(Jessup Farms v. Baldwin
(1983)
Further attacking the sufficiency of the evidence of design defect, defendant contends the metal portion of the 1978 Bronco roof was “ahead of its time regarding roof strength.” It contends the evidence established that there was no federal standard for roof strength when the Bronco was manufactured, but that the vehicle far exceeded the passenger car standard that existed then and exceeded the current light truck standard. “These facts, as a matter of law, preclude the imposition of punitive damages,” according to defendant.
Viewing the evidence in the light most favorable to the verdict, however, a rational trier of fact could conclude that it was merely coincidental that the Bronco met the later federal standards, and that this coincidence—occurring years after the design and production of the 1978 Bronco—did not reflect the
*1144
corporate state of mind prevailing at the time in question. The standard for award of damages under Civil Code section 3294 is malice in fact
(Taylor v. Superior Court
(1979)
Instead, substantial evidence permits an inference that defendant’s policymakers were strongly motivated to bring the 1978 Bronco to market and knew they could not do so if they followed the routine practice of preproduction testing and development. Substantial evidence supports an inference that, in order to bring the product to market, defendant’s policymakers decided to disregard established policies by which defendant had determined that crush resistance needed to be higher (the twice-the-curb-weight standard) and that fiberglass or molded plastic should not be used for components comprising the passenger compartment unless reinforced by a steel skeleton. Given the existence of these corporate standards and the total absence of preproduction testing, a rational finder of fact could infer that the failure to design and test the 1978 Bronco to meet these standards reflected “a willful and conscious disregard of the . . . safety” (Civ. Code, § 3294, subd. (c)(1)) of those who would travel in the Bronco.
In a similar manner, defendant attacks the sufficiency of the evidence of design defect of the fiberglass portion of the roof by pointing out plaintiffs’ expert testimony that having a removable roof on the Bronco was reasonable 2 and that defendant’s designated “most knowledgeable” representative testified that, like a convertible’s cloth top, the fiberglass roof was not designed to survive a rollover accident.
However, other evidence permitted a reasonable trier of fact to conclude that defendant’s decision to put the unreinforced fiberglass roof was despicable because defendant knew that:
—unlike convertible passenger cars, the high center of gravity of truck-based utility vehicles made them more likely than passenger cars to roll over;
—defendant’s safety engineers previously had concluded unreinforced fiberglass should never be used for a part of a vehicle intended to enclose the *1145 passenger compartment and had concluded no utility vehicle should be produced without a roll bar—and that the original design for the 1978 Bronco, in fact, included a roll bar as standard equipment;
—the competitor’s removable fiberglass roof did contain metal reinforcement;
—defendant’s own testing conducted after production and delivery of the 1978 Bronco (testing defendant consciously decided not to undertake in order to bring the product to market more quickly, despite such testing as a routine part of product development for other vehicles) showed the roof failed to meet defendant’s safety standards and led defendant to include steel reinforcement in the 1980 Bronco.
In light of this credible evidence, it is frivolous to contend, based on defendant’s excerpts from the record, that “Ford’s decision to incorporate a fiberglass roof into certain vehicles—even if retrospectively deemed incorrect two decades after the fact—was not, as a matter of law, a ‘vile’ or ‘loathsome’ act that can support an award of punitive damages,” as defendant says in its opening brief.
Finally, defendant contends that the decision to use an unreinforced fiberglass top on the 1978 Bronco was never disclosed to “top management,” citing the testimony of a former Ford executive who testified on behalf of plaintiff. The testimony of the former Ford executive, Thomas Feaheny, was somewhat different than represented by defendant. Feaheny testified: “The decision to put [the fiberglass top] on that vehicle was not a top management decision. It was made in the truck operations and the truck department. And I guess I’m fairly confident that they never disclosed to higher management that there was this risk” that the fiberglass roof “almost inevitably” would fail in a rollover accident.
Nevertheless, Feaheny testified that in the ordinary course of business, the responsible executives within the light truck division would have known about the safety engineers’ conclusions concerning the use of unreinforced fiberglass for passenger compartment panels. These executives clearly had the power within the corporation to enforce or not enforce as company policy the safety engineers’ findings.
As stated previously, the test for whether malicious conduct is attributable to a corporation is whether the acts were done by a person or persons with authority to make ad hoc policy decisions on behalf of the corporation.
(White
v.
Ultramar, Inc., supra,
The statutory standard for corporate malice does not require, as defendant contends, that a “Ford officer, director, or managing agent believed that the company was compromising consumer safety by reducing costs or expediting production.” It will suffice if the evidence, in the light most favorable to the verdict, establishes that corporate policymakers
in fact
compromised consumer safety by reducing costs or expediting production
in willful and conscious disregard
of consumer safety. “[Mjalice does not require actual intent to harm. . . . Conscious disregard for the safety of another may be sufficient where the defendant is aware of the probable dangerous consequences of his or her conduct and he or she willfully fails to avoid such consequences.”
(Angie M.
v.
Superior Court
(1995)
In the present case, the direct and circumstantial evidence, together with the inferences therefrom, is sufficient to permit a rational trier of fact to clearly and convincingly conclude that defendant acted with malice in its design and introduction into the market of the 1978 Bronco.
B. Claim of Excessiveness of Punitive Damages
The trial court denied defendant’s motion for a new trial on “all other grounds” stated in defendant’s motion, apart from the jury misconduct basis upon which it granted the motion. Among those grounds upon which the motion was denied was the claim that the amount of punitive damages was excessive. On appeal, this claim has three components: (1) The jury improperly was permitted to consider defendant’s overall size and worth in setting the amount of the award; (2) the amount of the award is factually insupportable; and (3) the “unprecedented” size of the award violates defendant’s due process right to notice of the penalty that it might incur at the time it acts maliciously.
1. Review Under California Law
“The amount of punitive damages is determined in the discretion of the jury. An appellate court will not reverse the jury’s determination unless the award as a matter of law is excessive or appears so grossly disproportionate to the relevant factors that it raises a presumption it was the
*1147
result of passion or prejudice. ... In reviewing the verdict the appellate court is guided by three main factors: the reprehensibility of the defendant’s conduct, the actual harm suffered by the victims, and the wealth of the defendant.”
(Rufo
v.
Simpson
(2001)
The second of the three considerations as set out in
Rufo
v.
Simpson, supra,
While we clearly must consider the degree of injury to the victims as part of our review of the punitive damages award for excessiveness, neither the Legislature nor the judiciary has established a fixed ratio between the compensatory and punitive damages that provides a standard for determining excessiveness. (See
Lane v. Hughes Aircraft Co., supra,
We also note that the trial court, sitting as an independent trier of fact (see
Sierra Club Foundation v. Graham
(1999)
We turn now to the factors prescribed by the Supreme Court in
Neal
v.
Farmers Ins. Exchange, supra,
Rather incredibly, defendant contends its conduct—even viewed in the light most favorable to the verdicts—“was at the very lowest end of the reprehensibility spectrum.” We disagree.
Involuntary manslaughter is the unlawful killing of a human being, inter alia, “in the commission of a lawful act which might produce death . . . without due caution and circumspection.” (Pen. Code, § 192.) Involuntary manslaughter is a felony. As we have concluded above, the evidence supports the jury’s conclusion that defendant willfully and consciously ignored the dangers to human life inherent in the 1978 Bronco as designed, resulting in the deaths of three persons. Such corporate conduct can constitute involuntary manslaughter (see
Granite Construction Co. v. Superior Court
(1983)
Further, defendant ignored its own internal safety standards while leaving the Bronco with the appearance of incorporating a roll bar into the vehicle’s roof structure. Defendant declined to test the strength of the Bronco’s roof before placing the vehicle in production, and when it did test the roof strength in conjunction with the redesign of the Bronco for the 1980 product line, defendant concluded the roof in fact failed to meet its own standards. Defendant modified the roof for the 1980 Bronco, but it did nothing to remedy the structural inadequacies in the tens of thousands of 1978 and 1979 Broncos that were already on the road. In this context—in contrast to the issue of defendant’s underlying negligence, not contested on this appeal— it avails defendant nothing that other auto manufacturers also used fiberglass roofs on contemporaneous fiberglass utility vehicles.
The harm to the victims is obvious and severe. Defendant does not contend otherwise.
*1149
The final issue is whether the punitive damages award is excessive in light of defendant’s financial condition. There is no doubt that $290 million is a lot of money. It is also true, however, that defendant is a huge company with enormous monetary resources. At the time of trial, defendant had daily after-tax profits of $20 million and its net worth was $25 billion, of which $14 billion was in cash on hand. Defendant earned well over $100 million in profit from the 1978 and 1979 Bronco. As far as the present record shows (see
Stevens
v.
Owens-Corning Fiberglas Corp.
(1996)
There is no doubt that $290 million would be a huge windfall to the Romo family. (See
College Hospital Inc.
v.
Superior Court
(1994)
It is empirically impossible, of course, to know exactly the level of punitive damages that would have a sufficient punitive and deterrent effect without unduly stifling innovation and competition in the marketplace. Any determination of an amount of punitive damages by a jury or by this court on appeal is, in this sense, somewhat arbitrary. The jury and the trial court each concluded $290 million provided sufficient punishment and deterrence. That damages award is 1.2 percent of defendant’s net worth and nine days of its profits at the time of trial. Although these amounts are slightly higher than the award approved in
Neal
v.
Farmers Ins. Exchange, supra,
2. Review Under Federal Constitution
In the recent case of
Cooper Industries, Inc.
v.
Leatherman Tool Group, Inc.
(2001)
This due process limitation, apparently arising from both the excessive fines clause of the Eighth Amendment and from the due process clause of the Fourteenth Amendment (see
Cooper Industries, Inc.
v.
Leatherman Tool Group, Inc., supra,
In
BMW,
the Supreme Court noted certain “guideposts” to assist its determination whether the defendant had received adequate notice of the severity of the penalty that might be imposed on it. Those were “the degree of reprehensibility of the [conduct]; the disparity between the harm or potential harm suffered by [plaintiff] and his punitive damages award; and the difference between [the punitive damages award] and the civil penalties authorized or imposed in comparable cases.”
(BMW, supra,
As we have already discussed, defendant’s conduct was grossly reprehensible. While defendant did not intend the death of the victims, the award here cannot be compared to cases “involving a business fraud resulting only in economic harm.”
(Rufo
v.
Simpson, supra,
There is a great disparity between the compensatory and punitive damages award. While this factor weighs in favor of a finding of excessiveness, the importance of the factor is somewhat reduced, as we have explained above, because wrongful death compensatory damages do not fully reflect the harm to the victims.
Finally, the nature of the conduct involved in the present case makes inapplicable the third
BMW
factor, comparison with civil fines for similar conduct. Such a comparison is appropriate when the conduct to be punished cannot result in imprisonment if committed by an individual, but is inappropriate when the malicious conduct could result in substantial imprisonment if committed by an individual instead of a corporation. (See
BMW, supra,
517 U.S. at pp. 583-584 [116 S.Ct. at pp. 1602-1603], distinguishing
Pacific Mutual Life Insurance Co. v. Haslip
(1991)
Defendant contends that no California case has awarded anywhere near $290 million in punitive damages and, accordingly, defendant lacked notice of “the severity of the penalty that a State may impose,” as required by
BMW. (BMW, supra,
We note that neither the United States Supreme Court nor any California court has held that a jury’s consideration of the size and resources of a defendant invalidates a punitive damages award. Rather, the courts recognize that a defendant’s wealth cannot be the sole criterion for the size of a punitive damages award: ability to pay an award cannot justify an
otherwise excessive
punitive damages award. (See
BMW, supra,
Here, the punitive damages verdict was not “otherwise excessive”; instead, consideration of the size and net worth of a corporate defendant was necessary if the jury was to appropriately vindicate the state’s interest in protecting the lives of its citizens.
Disposition
The order granting defendant’s motion for new trial is reversed. In all other respects, the judgment is affirmed. Plaintiffs are awarded costs on appeal.
Buckley, J., and Cornell, J., concurred.
A petition for a rehearing was denied July 23, 2002, and the petition of appellant Ford Motor Company for review by the Supreme Court was denied October 23, 2002. Kennard, J., Baxter, J., and Brown, J., were of the opinion that the petition should be granted.
Notes
For example, in defendant’s response brief filed in the cross-appeal, counsel write (with our italics added): “It is also undisputed that during deliberations, Ms. [S.] misattributed the extremely prejudicial quote about settling cases rather than recalling allegedly defective vehicles to Mr. Iacocca.”
The expert did not testify that an unreinforced fiberglass roof was reasonable. He merely testified that a removable roof was reasonable. The earlier version of the Bronco, for example, had a removable steel roof. He was highly critical of the use of unreinforced fiberglass, combined with the absence of a roll bar, for the 1978 Bronco. In other words, contrary to the implication of defendant’s excerpt, plaintiffs’ expert did not express approval of the fiberglass roof on this vehicle.
Defendant also contends that a California plaintiff seeking a punitive damages award based, in part, on a defendant’s financial condition must prove the defendant’s profits from California only and in some manner allocate the defendant’s net worth to its California operations. In
BMW, supra,
